INDOMETHACIN: CAPILLARY GAS CHROMATOGRAPHIC - ELECTRON CAPTURE DETECTION ANALYSIS AND PHARMACOKINETIC STUDIES IN

THE FETAL LAMB

by

RAJESH KRISHNA

B. Pharm., University of Kerala, Trivandrum, India, 1990

M . Pharm., Banaras Hindu University, Varanasi, India, 1992

A THESIS SUBMITTED FN PARTIAL FULFILMENT OF

THE REQUIREMENTS FOR THE DEGREE OF

M A S T E R OF SCIENCE

(Biopharmaceutics and Pharmacokinetics)

in

THE F A C U L T Y OF G R A D U A T E STUDIES

Faculty of Pharmaceutical Sciences

(Division of Clinical Pharmacy)

We accept this thesis as conforming

to the required standard

THE UNIVERSITY OF BRITISH C O L U M B I A

February 1995

© Rajesh Krishna, 1995

In presenting this thesis in partial fulfilment of the requirements for an advanced

degree, at the University of British Columbia, I agree that the Library shall make it

freely available for reference and study. I further agree that permission for extensive

copying of this thesis for scholarly purposes may be granted by the head of my

department or by his or her representatives. It is understood that copying or

publication of this thesis for financial gain shall not be allowed without my written

permission.

Department of ^ ^ ^ ^ ' ( 4 S c u ^ c ^ A

The University of British Columbia Vancouver, Canada

Date

DE-6 (2/88)

A B S T R A C T

11

Indomethacin is currently used in the treatment of patent ductus arteriosus,

preterm labour, and for polyhydramnios. In order to investigate its clearance and

disposition in the chronically instrumented ovine fetus, a sensitive assay was

required.

The developed GC-ECD method involves the extraction of indomethacin and

a-methyl indomethacin, the internal standard, from acidified plasma, urine,

amniotic and tracheal fluids using a simple one-step liquid-liquid extraction

procedure with ethyl acetate. The organic extract is evaporated and the residue

derivatised with N-methyl-N-(tert-butyldimethylsilyl)-trifluoroacetamide

(MTBSTFA). The derivatised samples were reconstituted with toluene and aliquots

of 2 jxL injected into the GC. Calibration curves were linear over the working range

of 1-32 ng of indomethacin/mL with a coefficient of determination (r ) > 0.99,

following extraction of 0.1 mL of plasma, and 0.1-1.0 mL of urine, amniotic and

tracheal fluids. Mean recoveries from plasma and urine were 96.99 ± 9.25 % and

94.74 ± 6.75 %, respectively. The limit of quantitation is 1 ng/mL (< 10% C.V.,

S/N > 10).Inter- and intra-day variabilities at each concentration point were < 11 %

for concentrations between 2-32 ng/mL and < 20 % at the LOQ of 1 ng/mL.

Ill

The mean steady state fetal femoral arterial concentration of indomethacin

was 181.67 ± 45.05 ng/mL in the high dose group (n=5), and 41.29 ± 3.14 ng/mL in

the low dose group (n=2). The arterio-venous concentration difference for

indomethacin at apparent steady state was minimal but statistically significant,

suggesting a low placental permeability to indomethacin in sheep. The fetal total

body clearance was estimated to be 44.55 + 11.14 mL/min (n=7). The placental

clearance of indomethacin, as determined by the Fick principle, was calculated to be

8.85 ± 0.81 mL/min/kg. The renal clearance was estimated to be 0.015 + 0.009

mL/min (n=5). Indomethacin appeared in very low concentrations in amniotic fluid

and fetal urine in high dose group fetuses. It was not observed to accumulate in

tracheal fluid. The drug appeared to induce lactic acidosis and a reduction in fetal

urinary output in some fetuses of the high dose group.

K. Wayne Riggs, Ph.D.

Research Supervisor

iv

TABLE OF CONTENTS Page

ABSTRACT ii

TABLE OF CONTENTS iv LIST OF FIGURES ix LIST OF TABLES xiii LIST OF SCHEMES xv LIST OF ABBREVIATIONS xvi ACKNOWLEDGEMENTS xxi DEDICATION xxii

1. INTRODUCTION 1 1.1. Indomethacin 1

1.1.1. Pharmacology of indomethacin 1 A. Mechanism of action 1 B. Clinical uses 6 C. Renal effects of indomethacin 11 D. Fetal effects of indomethacin 14

1.1.2. Pharmacokinetics of indomethacin 15

1.1.2.1. Absorption 15

1.1.2.2. Protein binding 17 1.1.2.3. Distribution 18 1.1.2.4. Metabolism and fate 19 1.1.2.5. Elimination kinetics 24

1.1.3. Placental transfer 27

1.1.3.1. Pharmacokinetic characterisation 28

1.1.3.2. The chronically instrumented pregnant sheep model 28

1.1.3.3. Placental transfer of indomethacin

1.1.4. Analytical methods

V

30 32

1.2. Rationale and specific aims 35

1.2.1. Rationale 35 1.2.2. Specific aims 36

2. EXPERIMENTAL 37 2.1. Materials and supplies 37

2.1.1. Chemicals 37 2.1.2. Reagents 38 2.1.3. Solvents 38 2.1.4. Gases 39 2.1.5. Supplies 39

2.2. Instrumentation 40 2.2.1. Gas chromatography 40 2.2.2. Gas chromatography-Mass spectrometry 41 2.2.3. Columns 41 2.2.4. Liquid chromatography-mass spectrometry/

mass spectrometry 42 2.2.5. Differential scanning calorimetry 42 2.2.6. Nuclear magnetic resonance spectrometry 43 2.2.7. General equipment 43 2.2.8. Physiological monitoring 43

2.3. Preparation of stock solutions 44 2.3.1. Preparation of drug stock solution 44

2.3.1.1. Indomethacin 44

2.3.2. Preparation of internal standard solutions 45 2.3.2.1. Zomepirac sodium 45

VI

2.3.2.2. Sulindac 45

2.3.2.3. a-Methyl indomethacin 46

2.3.2.4. Other experimental internal standards 46 2.3.3. Preparation of reagent solutions 46

2.4. Sample preparation 48 2.4.1. Analysis of intact indomethacin levels 48 2.4.2. Analysis of glucuronide conjugates 50

2.5. Preparation of calibration curves 51

2.5.1. Calibration curve with a-methyl indomethacin 51

2.5.2. Calibration curve with zomepirac sodium 51 2.6. Chemistry 52

2.6.1. Derivatisation procedures for indomethacin 52 2.7. Optimisation of assay parameters 54

2.7.1. Derivatisation of indomethacin 54 2.7.2. Chromatography 55 2.7.3. Extraction 55 2.7.4. Validation 56

2.7.4.1. Extraction recovery studies 56 2.7.4.2. Inter- and intra-day variability 57 2.7.4.3. Linearity 57 2.7.4.4. Stability 57

2.8. Animal experiments 58 2.8.1. Animal handling and surgical preparation 59 2.8.2. Recording and monitoring procedures 60 2.8.3. Determination of organ blood flows 61

2.8.4. Experimental protocol for fetal infusion studies 62

2.9. Data analysis 63

2.9.1. Pharmacokinetic calculations 63 2.9.2. Statistical calculations 64

3. RESULTS 66 3.1. Development of a capillary GC-ECD method for

indomethacin 66

3.1.1. Purity assessment of indomethacin 66 3.1.2. Optimisation of derivatisation 71 3.1.3. Selection of internal standards 78 3.1.4. Optimisation of GC conditions " 86 3.1.5. Optimisation of drug extraction 101 3.1.6. Gas chromatography-mass spectrometry 108

3.2. Validation of the GC-ECD method 113 3.2.1. GC operating conditions 113 3.2.2. Drug extraction from biological fluids 118 3.2.3. Recovery studies 123 3.2.4. Stability studies 123 3.2.5. Inter-day and intra-day variability 129

3.2.6. Calibration curves 129 3.3. Fetal infusion studies 137

3.3.1. Analysis of plasma samples 137 3.3.2. Analysis of urine samples 150

3.3.3. Drug disposition in amniotic and tracheal fluid

compartments 151

4. DISCUSSION 156 4.1. Development of a capillary GC-ECD method 156 4.2. Disposition of indomethacin in sheep 179

4.2.1. Placental transfer of indomethacin in sheep 180

4.2.2. Placental clearance 182 4.2.3. Elimination of indomethacin by the fetal kidney 186

4 .3. Drug disposition in amniotic and tracheal fluids

4.3.1. Amniotic fluid

4 .3.2. Tracheal fluid

4.4. Effect of indomethacin on physiological factors

S U M M A R Y & C O N C L U S I O N S

R E F E R E N C E S

IX

LIST OF FIGURES page

1. Chemical structure of indomethacin 1

2. Inhibition of PG synthesis by indomethacin (Adapted from 3 Shen and Winter, 1977).

3. Proposed role of PG's in labour (Adapted from Miller, 5 1983)

4. Metabolism of indomethacin in the adult human (Adapted 21 from Vree etal, 1993).

5. NMR spectra of indomethacin 67-68

6. DSC scan of indomethacin: (a) open pan, and (b) closed 70 pan methods.

7. LC/MS/MS spectrum of indomethacin 72

8. GC-ECD chromatograms of indomethacin derivatised by 73 (a) BSA, and (b) MTBSTFA.

9. GC-ECD chromatogram of indomethacin derivatised by 74 PFBBr.

10. Determination of optimum derivatising temperature for 76 indomethacin with MTBSTFA (data are mean values + 1 S.D., n=4).

11. Determination of optimum reaction time between 77 indomethacin with MTBSTFA (data are mean values ± 1 S.D., n=4).

12. Chemical structures of tested internal standards. 79

13A Derivatisation of 5-fluoro DL tryptophan with PFBBr for 80 use as potential internal standard for indomethacin.

13B GC-ECD chromatograms of indomethacin analogs, 5- 81 fluoroindomethacin, and oc-methyl indomethacin, derivatised by MTBSTFA.

14. Illustration of a decomposition peak "X" associated with 83 sulindac. Chromatograms indicate: (1) indomethacin and sulindac, (2) sulindac only, and (3) blank.

15. DSC scan of sulindac and recrystallised sulindac. 84

16. GC-ECD chromatogram of indomethacin and zomepirac in 85 fetal plasma.

17. Chemical structures of indomethacin and a-methyl 87 indomethacin.

18. Effect of injection port temperature on the chromatography 88-89 of (A) 20 ng/mL, and (B) 100 ng/mL indomethacin concentrations (n=4).

19. Effect of various column temperatures on peak retention 91 times.

20. Chromatograms of indomethacin and zomepirac using a 92 narrow bore fused silica Ultra-2 column.

21. Chromatogram of indomethacin and zomepirac using a 93 narrow bore DB-1701 column.

22. Effect of purge delay times on the peak area counts of 95 indomethacin tBDMS derivative (n=4).

23. Effect of various detector temperatures on the peak area 96 counts of indomethacin tBDMS derivative.

24. Determination of optimum column head pressure: Effects 97-98 of (1) peak indomethacin area counts, (2) peak area ratio, (3) retention times, and (4) peak width.

25. Determination of optimum column head pressure with 99-100 reference to (A) peak retention time, and (B) peak width.

26. Extraction efficiency of various organic solvents (n=4), in 102 a volume of 2 mL, using tracheal fluid as the model biological fluid.

27. Effect of solvent volume on the extraction efficiency of 105

ethyl acetate.

28. Determination of optimum pH on the extraction of 106 indomethacin (n=4).

29 Determination of optimum extraction time for 107 indomethacin from sheep biological fluid matrix using ethyl acetate at pH 5.0 (data are mean values ± 1 S.D., n=4)

30. GC-MS spectra of indomethacin tBDMS in the EI (A), 109-111 NCI (B), and PCI (C) modes.

31. GC-MS spectra of indomethacin-TMS 114-115

32. GC-MS spectra of indomethacin-PFB 116-117

33. Representative chromatograms of indomethacin and a- 119 methyl indomethacin spiked in fetal plasma (the inset shows the magnified baseline).

34. Representative chromatograms of indomethacin and a- 120 methyl indomethacin spiked in fetal urine.

35. Representative chromatograms of indomethacin and a- 121 methyl indomethacin spiked in amniotic fluid.

36. Representative chromatograms of indomethacin and 122 zomepirac in amniotic fluid.

37. Stability study of indomethacin infusate formulation at 128 room temperature, and at standard refridgeration conditions.

38. Representative calibration curve for indomethacin in fetal 134 plasma.

39. Representative calibration curve of indomethacin in fetal 135 urine.

40. Representative calibration curve of indomethacin in 136 amniotic fluid.

xii

41. Effect of low dose (E 1137, E 1115), and high dose (E 139-141 2160, E 1221) indomethacin on fetal arterial blood gas parameters (A & B), glucuose and lactate concentrations (C & D), pH and base excess (E & F).

42. Mean concentration-time profiles of indomethacin in 143 plasma, urine, and amniotic fluid for high dose ewes (n=5).

43. Mean concentration-time profiles of indomethacin in FA 144 and UV plasma for low dose ewes (n=2).

44. Effect of indomethacin on the fetal urinary flow rate. 153

45. Representative plot for the determination of renal 154 clearance.

LIST O F T A B L E S

xiii

page 1. Metabolites of indomethacin. 22

2. Pharmacokinetics of indomethacin in neonates. 26

3. Chemical shift values of indomethacin and their respective 69 group assignment.

4. Extraction efficiency (%) of organic solvents used to extract 104 indomethacin at different concentrations using tracheal fluid as the model biological fluid.

5a. Extraction recoveries (n=4), inter-day variability (n=4), and 124 intra-day variability (n=3) for indomethacin in (A) fetal plasma, and (B) fetal urine.

5b. Inter- and intra-day variability parameters for indomethacin 125 in amniotic fluid of fetal sheep, with zomepirac as the internal standard.

5b. (Continued) Extraction recoveries of indomethacin spiked in 126 amniotic fluid using zomepirac as the internal standard.

6. Stability of indomethacin in acidified sample matrix, during 127 freeze-thaw cycles (with a-methyl indomethacin as the I.S.), and on the GC autosampler tray (with zomepirac as the I.S.).

7. Variability parameters for the limit of quantitation (LOQ) of 130 1 ng/mL.

8. Calibration curve data for indomethacin in fetal plasma 131 (n=4).

9. Calibration curve data for indomethacin in fetal urine (n=4). 132

10. Calibration curve data for indomethacin in amniotic fluid 133 (n=8).

11 A. Animal experiment sampling and dosing protocol. 138

11B. Determination of apparent steady state plasma concentration 138

xiv

12. Mean concentration-time data for indomethacin in fetal 142 femoral arterial (FA) plasma, umbilical venous (UV) plasma, and fetal urine samples in high dose ewes (n=5).

13. Arterio-venous concentration difference for indomethacin at 146 apparent steady state.

14. Arterio-venous concentration difference in high dose ewes: 147 statistical evaluation.

15. Determination of fetal total body clearance. 148

16. Calculation of placental and non-placental clearances for 149 indomethacin.

17. Concentration of indomethacin in fetal urine. 152

18. Determination of renal clearance for indomethacin in high 155 dose group fetuses.

19. Stability of indomethacin in aqueous solution (Adapted from 175 Shen and Winter, 1977).

20. Stability of indomethacin in surfactant solubilised systems 178 (Adapted from Ghanem et al., 1979).

XV

LIST OF SCHEMES

Page 1. Sample preparation protocol for indomethacin. 49

2. Resonance forms of indomethacin-tBDMS ester. 161

3. Degradation of indomethacin in alkali. 175

ABBREVIATIONS

xvi

ACS American Chemical Society

AMN amniotic fluid

AVP arginine vasopressin

[A-V]indo(sS) aterio-venous indomethacin concentration difference at apparent steady state

BE base excess

BSA bis (trimethylsilyl) acetamide

BSTFA N, O-Bis (trimethylsilyl) trifluoroacetamide

°C degree Celsius

C.V. coefficient of variation

cAMP cyclic adenosine monophosphate

CBF cerebral blood flow

Cfa concentration of drug in fetal femoral arterial blood at steady

state

CLf 0 fetal non-placental clearance

CL P placental clearance

CL r renal clearance

CL t b total body clearance

C s s steady state plasma concentration

d m concentration of drug in umbilical venous blood at steady

state

DA ductus arteriosus

DBI N-deschlorobenzoyl indomethacin

DCM dichloromethane

DMF dimethyl formamide

DMI O-desmethyl indomethacin

DPMA diphenylmethoxyacetic acid

DSC differential scanning calorimetry

EA ethyl acetate

ECD electron capture detection

EI electron-impact ionisation

eV electron volts

F fetal

F-T freeze-thaw

F/M fetal-to-maternal ratio

FA fetal femoral arterial

[FA] i n d o ( S S) fetal arterial plasma indomethacin concentration at apparent steady state

FDA Food and Drug Administration

F E N a + fractional excretion of sodium ion

F s s fetal steady-state concentration

g acceleration due to gravity

g gram

GC gas chromatography, gas chromatograph

GC-MS gas chromatography-mass spectrometry

GFR glomerular filtration rate

h hour

H 2 hydrogen gas

HCO"3 bicarbonate ion

XV111

HETP

HP

HPLC

i.d.

I.U.

i.v.

IND

IPA

kg

ko

kPa

L

LC/MS/MS

LOQ

M

m/z

MA

MAO

meq

mg

min

mL

MLCK

MS

M.,

height equivalent to a theoretical plate

Hewlett-Packard

high performance liquid chromatography

internal diameter

international units

intravenous

indomethacin

isopropyl alcohol

kilogram

infusion rate

kilopascal

liter

liquid chromatography/mass spectrometry/mass spectrometry

limit of quantitation

maternal

mass to charge ratio

maternal arterial

monoamine oxidase

milliequivalents

milligram

minute

milliliter

myosin light chain kinase

mass spectrometry

maternal steady-state concentration

MSTFA N-methyl-N-trimethylsilyl trifluoroacetamide

MTBSTFA N-methyl-N-(tert-butyldimethylsilyl) trifluoroacetamide

M« microgram

|lL microliter

N normal

n sample size

NCI negative ion chemical ionisation

ND not detectable

ng nanogram

NMR nuclear magnetic resonance spectrometry

NSAID non-steroidal anti-inflammatory drug

o 2 oxygen

PCI positive ion chemical ionisation

pC0 2 partial pressure of carbon dioxide

PDA patent ductus arteriosus

PFB pentafluorobenzyl

PFBBr pentafluorobenzyl bromide

PFP pentafluoropropanol

PG prostaglandin

pH negative logarithm (base 10) of the hydrogen ion concentration

p0 2 partial pressure of oxygen

psi pounds per square inch

PTFE polytetrafluoroethylene

Qum umbilical blood flow

r correlation coefficient

coefficient of determination

RBF

RIA

rpm

Rt

S/N

SD

sec

SEM

T1/2

TBDM-SIM

TBDMCS

tBDMS

TEA

TMS

TR

UHP

UR

UTV

UV

[UV ] i n d o ( S S )

VP y=mx + c

renal blood flow

radioimmunoassay

revolutions per minute

retention time

signal-to-noise ratio

standard deviation

second

standard error of the mean

half life

N-t-butyldimethylsilylimidazole

t-butyldimethylchlorosilane

tert-butyldimethylsilyl

triethylamine

trimethylsilyl

tracheal fluid

ultra-high purity

fetal urine

uterine vein

umbilical venous

umbilical venous plasma indomethacin concentration at apparent steady state

vasopressin

linear regression statistics of a straight line

ACKNOWLEDGEMENTS

xxi

I would like to thank my research supervisor, Dr. Wayne Riggs, for his guidance, constructive criticism, and friendship. His constant encouragement is sincerely appreciated. Thanks also to my committee members, Dr. Frank Abbott, Dr. James E. Axelson, Dr. Dan W. Rurak, Dr. Martin P.R. Walker, and Dr. Jack Diamond, for their expert scientific input and constructive criticism.

I would like to thank Dr. Andras Szeitz for initial help during method development and friendship throughout the program. Mr. Eddie Kwan is thanked for his assistance with the experimental studies. Mr. Roland Burton is thanked for his help during mass spectrometric studies.

I am thankful to Dr. Rob Thies for his help in the preparation and

development of slides for the oral presentation.

Many thanks also to my colleagues, Mr. John Gordon, Dr. George Tonn, Ms.

Judit Orbay, Dr. Wei Tang, Dr. Anthony Borel, Dr. Weiping Tan, Mr. Sanjeev

Kumar, Mr. John Kim, Ms. Eva Law, Mr. Nanda Kumar, and Dr. Sarvajna Dwivedi,

for their constructive suggestions and friendship.

I gratefully acknowledge the award of Stanley Pharmaceuticals (Novopharm group) Scholarship by Stanley Pharmaceuticals, North Vancouver and UBC, and the Graduate Student Travel Award of the Faculty of Graduate Studies for poster presentation at the 9th annual meeting and exposition of AAPS at San Diego. I am thankful to Ms. Jennie Campbell and Mr. Frank Wang of International House, UBC, for their support. Award of a Graduate Teaching Assistantship and a Graduate Research Assistantship by the Faculty is acknowledged. Financial support for the research project was from the BC Health Research Foundation.

DEDICATION

'This thesis

is

dedicated

to my

parents

for their

Cove, support, and patience.

1. INTRODUCTION 1

1.1 Indomethacin

Indomethacin (l-[p-chlorobenzoyl]-5-methoxy-2-methyl-indole-3-acetic

acid), introduced as a non-steroidal antiinflammatory drug (NSAID), is a potent

prostaglandin synthetase inhibitor used in the treatment of gouty arthritis (O'Brien et

al., 1968), ankylosing spondylitis (Kass, 1965), osteoarthritis (Bilka et al, 1964),

rheumatoid arthritis (O'Brien et al, 1968), patent ductus arteriosus (Friedman et al.,

1976), and preterm labour (Zuckerman et al., 1974).

CH2COOH

Figure 1: Chemical structure of indomethacin.

1.1.1 Pharmacology Of Indomethacin

A. Mechanism Of Action

2 General Pharmacological Activity

Since its introduction in 1963, a number of mechanisms have been proposed

®

to account for indomethacin's (and other Aspirin like drugs) mode of action.

These include uncoupling of oxidative phosphorylation, mucopolysaccharide

biosynthesis inhibition, inhibition of platelet function, stabilisation of cellular and

lysosomal membranes, and effects on leukocyte migration (Morsdorf, 1964;

Schonhofer, 1967; Whitehouse, 1968; Bryant et al., 1963). Structural-activity

relationship studies at that time indicated the presence of a hypothetical

antiinflammatory receptor site consisting of a cationic core and two noncoplanar

hydrophobic regions (Shen, 1965; Shen, 1967). In the early 70s, the inhibition of

®

prostaglandin (PG) synthetase by indomethacin and Aspirin at therapeutically

effective concentrations was established (Vane, 1971; Smith and Willis, 1971;

Ferreira et al., 1971). Following this discovery, Ham et al., 1972, screened a series

of substituted aryl acid compounds in a PG synthetase system. However, the

elucidation of the exact mechanism of action was hindered largely due to the fact

that PG synthesis systems in different tissues possess markedly dissimilar

pharmacological and biochemical profiles, and that NSAIDs may inhibit the enzyme

system at different sites in the biosynthetic pathway. At that time, it was also

determined that the inactivation of PG synthetase was reversible and time-dependent

(Smith and Lands, 1971; Raz et al., 1973). Indomethacin was observed to be a

competitive inhibitor of PG synthetase in sheep seminal vesicles with a kj of 6.5 x

10"6 M (Ku and Wasvary, 1973). The mechanism of action is illustrated in Figure 2.

Figure 2: Inhibition of PG synthesis by indomethacin (Adapted from Shen and

Winter, 1977)

Patent Ductus Arteriosus

Prostaglandin E 2 (PGE2) is primarily responsible for maintaining the patency

of the ductus arteriosus (DA) in the fetus (Friedman, 1978; Coceani and Olley,

1973; Coceani et al, 1975; Coceani et al., 1978a and 1978b; Clyman al, 1980).

Indomethacin treatment leads to constriction of the DA mediated via cyclo-

oxygenase inhibition.

4 Premature Labour

Prostaglandins, PGE l 5 PGE2, and PGF2a, all stimulate uterine contractions in

pregnant subjects and are hence suitable candidates for the induction of labour

(Embrey, 1970; Beazley and Gillespie, 1971; Roth-Brandel et al, 1970). Of these,

prostaglandins, PGF2rx and PGE 2 are potent compounds used therapeutically for the

induction of labour (Thiery and Amy, 1975). The two common formulations are

PGE 2 gel for intracervical application and PGF 2 a for slow i.v. infusion. If the gel

cannot be employed or if the patient experiences primary dysfunctional labour

(Andersson et al., 1983), then the i.v. PGF 2 a formulation may be used. The exact

mechanism by which these PGs stimulate uterine contractions is unclear.

It has been shown that during labour there is a rise in PG concentrations in

blood (Karim, 1968) and amniotic fluid (Karim and Devlin, 1967; MacDonald and

Casey, 1993). A subsequent paper reported that exogenously administered PG's can

induce labour (Karim et al., 1969). These reports suggested a physiological role of

the PG's in uterine motility. The critical step in the initiation of labour is the local

production of PG's, i.e., production of PGE 2 in the amnion, which then reaches the

decidua. In the decidua, PGE 2 can either undergo conversion to PGF 2 a or initiate

the synthesis of further PGE 2 and PGF 2 a (Nakala et al., 1986; McCoshen et al.,

1990; Casey and McDonald, 1988; McDonald et al, 1991). Although the exact

mechanism by which PGs stimulate uterine contractions is unclear, there is some

5 evidence to suggest that they affect transmembrane calcium flow (Huszar, 1989). It

is speculated that PCs may activate smooth muscle cell contractile protein and that

the process involves an increase in calcium transport through the plasma membrane.

PGF2a stimulates Ca + + influx into the smooth muscle cell. This influx of Ca + +

results in the activation of myosin light-chain kinase (MLCK) initiating

phosphorylation. Shortening of smooth muscle cells is facilitated by actin-myosin

interactions (Figure 3). Indomethacin inhibits PGF2a production and thereby

interferes with its modulatory action on transmembrane calcium flow. This results

in an inhibition of uterine muscle activity. Therefore, by inhibiting uterine activity,

PG synthetase inhibitors like indomethacin are indicated for tocolysis (Zuckerman

etal., 1974; Wiquist et al, 1975; Niebylera/., 1980).

SARCOPLASMIC RETICULUM colcium pool

ff-udrimgw; agonists noprotere»«4

calmodulin

0 phosphorylation

| cAMP| '

(5 • acetylcholine^ | theophylline a - adrenergic^ papaverine

colcium

phosphodiesterase

/ \ MYOSIN LIGHT-CHAIN

KINASE CAMP

ACT1N • MYOSIN-P ACTOMYOSIN-P

Figure 3: Proposed role of PG's in labour (Adapted from Miller, 1983).

6 B. Clinical Uses

Anti-inflammatory, Analgesic, and Antipyretic Uses

As previously mentioned, indomethacin is effective in rheumatic disorders

(ankylosing spondylitis, osteoarthritis, rheumatoid arthritis) as well as for moderate

pain such as dysmenorrhoea. The anti-inflammatory, analgesic, and antipyretic

effects of indomethacin are the subject of several research papers: uses in Bartter's

syndrome (Zancan et al., 1976; Haluska a/., 1977; Donker et al., 1977; Rado et

al., 1978), in fever (Silberman et al., 1965; Kieley et al, 1969; Engerrall et al,

1986), in hypercalcaemia (Elliot and McKenzie, 1983; Dindogru et al., 1975;

Blum, 1975), in osteoarthritis (Zachariae, 1966; Harth and Bondy, 1969; Woolf et

al, 1978; Gordin et al, 1985), in pain (Holmlund and Slodin, 1978; Elder and

Kapadia, 1979; Mattila et al, 1983; Ebbehoj et al, 1985; Jonsson et al, 1987),

and in rheumatoid arthritis (Donnelly et al, 1967; Wright et al, 1969; Baber et al,

1979; Ekstrand et al, 1980).

Patent Ductus Arteriosus

The ductus arteriosus (DA) is a vascular shunt in the fetal circulation that

diverts about 90% of the right ventricular output past the lungs into the descending

aorta. PGE 2 is one of the most important determinants in the patency of the DA.

7 Closure of the ductus is imperative in the neonate as failure results in the

development of a left-to-right shunt from the aorta to pulmonary artery. Other

complications include hepatomegaly, hyperactive precordium, and congestive

cardiac failure. The incidence of patent ductus arteriosus (PDA) is about 80% in

neonates weighing less than lOOOg and 10-15% in those weighing between 1500-

2000g (Douider et al, 1988). PDA closure has been postulated to be age related

(McCarthy etal, 1978).

Surgical ligation (Gay et al., 1973) and treatment with a prostaglandin

synthetase inhibitor, such as indomethacin (Friedman et al., 1976), are two

commonly used methods of correcting PDA. Increased oxygen tension and

decreased prostaglandin concentrations are two common determinants of ductal

closure at birth (Noerr, 1991). Intravenous indomethacin, in the sodium trihydrate

form, is the drug of choice for neonatal PDA. However, renal dysfunction is

claimed to be a prominent side effect of such a therapy. Decreased urine output and

free water clearance, increased serum creatinine and blood urea nitrogen have all

been reported during indomethacin use (Betkerur et al., 1981; John et al., 1984). A

mean reduction in cerebral blood flow (CBF) by 24% has also been reported in a

study involving pharmacological closure of the DA at an i.v dose of 0.2 mg/kg in

human premature infants (Pryds et al, 1988). Changes in regional blood flow

velocity have also been observed following indomethacin infusion in preterm

infants with symptomatic PDA. These changes include a reduction in time-

8 averaged mean velocity, peak systolic velocity, and end-diastolic velocity in the

anterior and middle cerebral artery (Austin et al, 1992). Reductions in CBF

velocity to the extent of 50% in adult humans and 30% in newborn pigs have been

observed (Wennmalm et al, 1981; Leffler et al., 1985). The apparent volume of

distribution of indomethacin appears to be a useful marker of permanent PDA

closure since it varies significantly pre- and post-PDA closure (Gal et al., 1991).

Polyhydramnios

Polyhydramnios is a complication arising from excess amniotic fluid and is

determined by several methods including the maximum-vertical pocket technique

(Chamberlain et al., 1984), amniotic fluid index (Phelan et al., 1987), and

gestational age adjusted amniotic fluid index (Moore and Cayle, 1990). A 30% fetal

mortality rate due to premature rupture of the membranes as a result of

polyhydramnios has been reported (Cabrol et al., 1987). Conventionally, the

management of polyhydramnios incorporates repeated transabdominal

amniocentesis under tocolytic therapy and fluid withdrawal. Puncture

complications include infection and membrane rupture. Prostaglandin synthetase

inhibitors reduce amniotic fluid volume as well as inhibit premature labour (Cabrol

et al., 1987). The reduction of amniotic fluid volume is thought to be due to a

decrease in urinary flow (Cifuentes et al., 1979; Kirshon et al., 1988; Rosen et al.,

1991; Kirshon et al, 1991; Walker et al., 1992a). One risk of such therapy,

9 however, is the possibility of oligohydramnios (Nordstorm et al., 1992; Hendricks

etal, 1990; Krishon^a/., 1991; Goldenberg et al, 1989).

Indomethacin is recommended in the treatment of symptomatic

polyhydramnios (Krishon et al., 1990; Smith et al., 1990). The mechanism by

which indomethacin's effects are thought to be mediated in polyhydramnios is

through a reduction in fetal urinary output. The underlying mechanism by which

indomethacin reduces amniotic fluid volume by a fall in urinary output, however,

remains unclear. Suppression of vasodilatory renal PG's by indomethacin leading

to a fall in renal blood flow (Matson et al., 1981; Kover, 1980) is one proposed

mechanism for the reduction in fetal urinary output. However, recent results

indicate that indomethacin may interact with arginine vasopressin (AVP) at the V2

receptor level and augment the action of AVP; an increase in fetal AVP

concentration was observed when indomethacin reduced fetal urinary output at a

dose of 0.0025 mg/kg/min five hour infusion. (Walker et al., 1992a). When

indomethacin (0.05 mg/kg, 5 h) was co-administered with the AVP-V2 receptor

antagonist (d [CH2]5-D-Phe-Ile4, Arg8)-VP, the fetal oliguria that was observed with

indomethacin alone was inhibited by the antagonist, suggesting that V2 receptor

played a significant role (Walker et al., 1994). Indomethacin may increase

circulating AVP levels and enhance the peripheral effects of AVP in the fetus. It is

interesting to note that PGE 2 inhibits cAMP production (Levenson et al., 1982), and

this effect seems to be mediated via the V2 receptor of AVP (Levenson et al., 1982;

10 Somenburg et al, 1990). Considering the fact that PGE 2 is synthesised in the

kidney in response to AVP and oxytocin leads one to speculate that there exists a

negative feedback role for the PGs (Walker et al, 1993). Inhibition of PGE 2

synthesis by indomethacin will potentiate AVP action through release and

peripheral action of AVP (Walker etal., 1993; Levenson et al., 1982).

Studies by Walker et al., 1992b, have shown that short-term infusion of

indomethacin at a dose of 0.017 mg/kg/min for 5 hours leads to an increase in fetal

urinary output, urinary osmolality, sodium and chloride. However, when the dose

infused was reduced to 0.0025 mg/kg/min for 5 hours, there was a reduction in

urinary output, suggesting fetal antidiuresis at low levels of indomethacin (Walker

etal, 1992a).

Preterm Labour

One of obstetrics' major challenges lies in the therapy of preterm labour. The

only tocolytic drug approved to date by the U.S. FDA is the -sympathomimetic,

ritodrine hydrochloride. A potential therapeutic alternative to ritodrine are the

prostaglandin inhibitors as it is known that prostaglandins PGE,, PGE2, and PGF 2 c x

stimulate uterine contractions and induce labour (Embrey, 1970; Roth-Brandel et

al., 1970). In 1974, Zuckerman et al., demonstrated a marked inhibitory effect of

indomethacin on uterine contractions in premature labour. In a randomised

11 comparative trial of indomethacin and ritodrine, Besinger et al (1991) reported that

the two were equally effective in delaying preterm birth. In 1992, Carlan et al.,

compared two PG synthesis inhibitors, namely, indomethacin and sulindac for

efficacy in refractory preterm labour. Sulindac was equally effective as

indomethacin although the group that responded to indomethacin had a longer time

to delivery, higher birth weight, and a lower number of days in the intensive care

unit. The sulindac group, however, demonstrated a renal sparing effect speculated

to be due to minimal placental transfer. Several fetal risks have been identified

following indomethacin administration including premature ductal closure, renal

dysfunction, oligohydramnios, and pulmonary hypertension (Arcilla et al., 1969;

Manchester et al., 1976; Rudolph, 1977; Friedman et al., 1978; Cifuentes et al.,

1979). Other complications include tricuspid regurgitation, right ventricular

dysfunction, and pericardial effusion (Hallak et al., 1991). Maternal side effects

included perspiration, hypertension, gastritis, dizziness, and nausea (Grella and

Zanor, 1978).

C. Renal Eff ects Of Indomethacin

Production of large quantities of hypotonic urine (amniotic fluid), excretion

of sodium in large amounts, low renal blood flow (RBF), and low glomerular

filtration rate (GFR) with greater fractional blood flow to mature inner nephrons are

some of the characteristics of fetal renal function (Kleinman, 1978). As gestation

12 advances, there is a progressive increase in urinary flow rate, RBF, and GFR, and a

reduction in fractional excretion of Na + (FE N a+). Postnatal renal adaptation includes

increases in GFR and RBF, redistribution of intrarenal blood flow from inside to the

periphery of the cortex, reduction in FEjsf a +, and a greater ability to concentrate

urine. AVP levels are high at birth and during hypoxia and hemorrhage (Kleinman,

1978; Gleason et al, 1988; Robillard et al, 1982). Optimal urine concentrating

ability depends upon the GFR, the AVP level, and the ability of the renal collecting

tubule to respond to AVP. AVP enhances water resorption from the collecting

tubule leading to anti-diuresis. The exact role by which renal prostaglandins modify

A VP's effects is still unclear. In 1968, Grantham and Orloff proposed that AVP was

controlled by a negative feedback loop wherein it stimulates intrarenal biosynthesis

and production of PGs, and these PGs inhibit AVP's effects at the collecting tubule.

While the exact mechanism(s) remain to be elucidated, several studies have

demonstrated that the prostaglandin inhibitors enhance the anti-diuretic effect of

AVP (Anderson et al, 1975; Berl et al, 1977; Fejes-Toth et al., 1977; Gullner et

al, 1980; Lum etal:, 1977).

When indomethacin was administered to close the PDA, the observed

neonatal renal dysfunction was characterised by reduced urine output, decreased

free water clearance, and increased blood urea nitrogen and serum creatinine. Early

studies related these events to reduced renal PG production (Betkerur et al., 1981;

Cifuentes et al., 1979; Levenson et al, 1982; Terragno et al., 1977).

13

Maternal indomethacin administration often leads to adverse renal effects in

the human fetus and neonate (Mogilner et al, 1982; Veerema et al., 1983;

Vanhaesebrouk et al., 1988; VanDer Heijden et al., 1988). A reduction in fetal

urinary output has been demonstrated to be the underlying mechanism by which

indomethacin acts in polyhydramniotic states (Hickok et al., 1989; Krishon et al.,

1988; Krishon et al., 1991). These effects of indomethacin seem to be dose

dependent as fetal urinary output increased when indomethacin was administered at

a dose of 0.017 mg/kg/min, but decreased when the dose was reduced to 0.0025

mg/kg/min in the pregnant sheep model, suggesting fetal antidiuresis at lower

plasma indomethacin concentrations (Walker et al., 1992a, 1992b). This incidence

of fetal oliguria has been attributed to reduced RBF following inhibition of renal

prostaglandins (Matson et al, 1981; Millard et al, 1979). Walker et al (1994)

proposed, however, that this fetal oliguria is a result of an effect of AVP on the fetal

kidney and not due to RBF. Co-administration of indomethacin and the AVP-V2

receptor antagonist {(d[CH2]5-D-Phe-Ile4,-Arg8)-Vassopressin} resulted in

inhibition of oliguria due to indomethacin alone (Walker et al, 1993, 1994),

suggesting that oliguria was due to activation of the AVP-V2 receptor and not to a

reduction in RBF.

14 D . Fetal Effects Of Indomethacin

Indomethacin has been reported to exert marked fetal side effects following

either fetal or maternal drug administration. The latter scenario is widely perceived

as a consequence of easy and high degree of placental transfer. Indomethacin

increased fetal pulmonary arterial pressure at a dose of 0.05 mg/kg i.v in pregnant

ewes (Ohara et al, 1991). The study reported that ductal constriction by

indomethacin caused increases in the systolic pulmonary artery-aortic pressure

difference as well as fetal heart rate. A marked fall in placental blood flow was

noted after indomethacin was given in pregnant rabbits (Katz et al, 1981).

Indomethacin reduced CBF in the human fetus by about 40% (Vanbel et al, 1990).

A drastic decline in fetal urine output was noted when short term oral indomethacin

was used for therapy of preterm labour (Krishon et al, 1988). Indomethacin also

stimulates sleep state independent fetal breathing which is abolished by PGE 2

infusion in fetal lambs (Jansen et al, 1984). A similar study demonstrated that PG

synthesis inhibitors stimulated breathing movements in fetal sheep (Kitterman et al,

1979). Stevenson et al, 1992 evaluated the effects of indomethacin on the ovine

fetus. They administered indomethacin at a dose of 10 mg/kg i.v to the ewe and 12

mg/kg i.v to the fetus. Increases in fetal arterial P02, placental blood flow, and

urine osmolality, and decreases in lung liquid flow and free water clearance were

observed. Walker et al, 1992b, administered indomethacin as a 0.35 mg/kg bolus

followed by a 0.017 mg/kg infusion over 5 h in chronically catheterised ovine

15 fetuses and observed a marked increase in fetal urinary output (about 84%) and

urinary osmolality as well as sodium and chloride concentrations. Fetal heart rate

was also increased.

Though there are a number of studies illustrating fetal indomethacin effects

there does not exist a comprehensive database on the drug's dose-response

characteristics which would rationalise the continued use of the drug to a

therapeutic advantage.

1.1.2 Pharmacokinetics Of Indomethacin

1.1.2.1. Absorption

Following oral administration, indomethacin is readily and almost completely

absorbed from the gastrointestinal tract with peak plasma concentrations appearing

within 120 minutes (Duggan et al., 1972; Alvan et al., 1975). Dissolution of

indomethacin capsules was observed to be greater in a low acidity condition

compared to a more acidic environment (Aoyagi et al., 1985). Bioavailability of

indomethacin is markedly altered by various diets; absorption being faster with a

high protein and a high lipid diet compared to a high carbohydrate diet (Wallusch et

al., 1978). Bioequivalence studies have indicated that the bioavailability of many

indomethacin formulations are far from being equal. A 30% variation in

16 bioavailability was observed from formulations marketed in Finland (Turakka and

Airakinsen, 1974) and in India; two out of ten products studied were found to be

bioinequivalent (Chaudhari et al., 1984). However, this bioavailability information

should take into account the extensive enterohepatic recirculation indomethacin

undergoes as approximately half an intravenous dose of indomethacin is subjected

to biliary recycling (Kwan et al., 1976). Bile facilitates the dissolution of poorly

soluble indomethacin, due to micellar solubilisation, and thereby increases its rate of

absorption (Miyazaki et al, 1980; Miyazaki et al., 1981).

Significant species differences exist in the pharmacokinetics of intravenously

(bolus) administered indomethacin. Plasma levels of 14C-indomethacin were

markedly higher in dogs and rats than in guinea pigs and rhesus monkeys, and tissue

levels of 1 4C were greater in the guinea pig than in the rat (Baer et al., 1974).

Indomethacin can easily penetrate the skin of humans (Snyder, 1975; Eckard,

1982), rabbits (Naito and Tsai, 1981), rats (Shima et al, 1981; Wada et al, 1982),

and guinea pigs (Inagi et al.,-1981), from various formulations (gels, ointments,

etc.) indicating the potential for transdermal delivery. Indomethacin is detected in

the aqueous fluid, iris, and cornea following ocular absorption in the rabbit (Hanna

and Sharp, 1972; Green et al, 1983). Circadian changes, as a determinant of drug

administration time, have been observed with indomethacin; its

chronopharmacokinetics being well studied in humans (Clench et al., 1981) and in

rats (Guissou et al, 1987). The former study indicated that indomethacin

17 administered in the morning had the highest plasma concentration, and shortest time

to peak in addition to greater elimination (Clench et al, 1981).

1.1.2.2. Protein Binding

Indomethacin, a weakly acidic organic compound, binds strongly with

plasma proteins, the binding being in the order of 90% as determined by equilibrium

dialysis (Hucker et al., 1966; Hvidberg et al., 1972). An association constant of

0.86 x 10 L/mol with a total number of binding sites of 15 has been determined for

human serum albumin (Hvidberg et al., 1972). An ultracentrifugation method

yielded a considerably higher value of about 99%, with an association constant of 3

x 105 L/mol and an indication of a single primary binding site and seven secondary

binding sites with association constants in the order of 1.4 x 104 L/mol (Mason and

McQueen, 1974). Binding of indomethacin was reduced by aspirin and cinmetacin,

and increased by phenylbutazone (Mason and McQueen, 1974; Montero et al.,

1986). In another study, using indomethacin as a model drug, Zona et al., 1986,

suggested that the conformation of the protein in plasma played a key role in drug-

protein interaction, and that binding was favoured when human serum albumin

adopted a "loose" structure. Binding of indomethacin to human serum albumin,

studied by the dynamic dialysis method, yielded an association constant of 1.4 x

106/M with 2.71 as the number of primary binding sites (Montero et al., 1986).

Indomethacin was bound to proteins in the human cerebrospinal fluid in the order of

18 40%, as determined by equilibrium dialysis, with albumin being the protein of

importance (Muller et al., 1991).

Protein binding studies using radiolabeled indomethacin in maternal and fetal

plasma ultrafiltrates in sheep have revealed that 97.6% of the fetal and 98.5% of the

maternal drug is bound to plasma proteins (Anderson et al., 1980c). In another

study involving unlabeled indomethacin and an equilibrium dialysis procedure, the

binding in maternal, fetal, and nonpregnant control was determined to be 85.6%,

43.8%, and 92.1%, respectively (Harris and Van Petten, 1981). Binding of

indomethacin in the postnatal stage is speculated to be related to bilirubin levels

while levels of albumin in the mother and fetus appear to be responsible for the

maternal-to-fetal binding gradient (Nau et al., 1983).

1.1.2.3. Distribution

Indomethacin appears in synovial fluid, peak concentrations ocurring about a

hour later than in serum in patients with rheumatoid arthritis (Emori et al., 1973).

Similar results have been obtained with intraarticularly administered indomethacin

(Neander et al., 1992). Following intramuscular administration, the levels of

indomethacin in synovial fluid and blood were identical after an hour (Dittrich et

al., 1984). Indomethacin was not detected in spinal fluid in the same study, while

minimal amounts were detected in another study (Hucker et al., 1966).

19 Indomethacin was detected in saliva following intravenous administration to dogs,

the levels in parotid saliva and mandibular-sublingual saliva being 7.4 and 4.4% of

the levels in plasma respectively (Watanabe et al, 1981). Binding of indomethacin

to salivary proteins was greater in parotid saliva. Indomethacin appears in breast

milk; the milk:plasma ratio being 0.37 (Lebedevs et al., 1991). Indomethacin

readily and rapidly crosses the placental membrane in humans (Moise et al., 1990),

rabbits (Parks et al., 1977; Harris and Van Petten, 1981), and rats (Klein et al.,

1981). Transfer in sheep, however, appears to be more limited (Harris and Van

Petten, 1981; Anderson etal, 1980c).

1.1.2.4. Metabolism and Fate

Metabolism of Indomethacin in The Adult

The metabolism of indomethacin in several species (monkey, guinea pig,

dog, rats, and humans) has been the subject of numerous reports (Harman et al.,

1964; Duggan et al, 1972; Hucker et al, 1966; Yesair et al, 1970; Baer, 1972;

Vree et al., 1993). Two major metabolites of indomethacin are N-

deschlorobenzoyl-indomethacin (DBI) and O-desmethyl-indomethacin (DMI)

(Harman et al, 1964). Both are excreted in unconjugated as well as glucuronide

conjugated forms. Therefore, the metabolic pathways in man appear to be O-

desmethylation and N-deacylation as well as conjugation with glucuronic acid. The

20 major pathway, however, appears to be desmethylation mediated via the hepatic

microsomal enzyme system which is followed by a deacylation step that is

extramicrosomal in nature (Duggan et al., 1972). These metabolites are found in

high concentrations in plasma. Metabolites of indomethacin are devoid of

prostaglandin synthesis inhibiting activity (Shen, 1965).

Significant species variation has been observed with the metabolism of

indomethacin. The glucuronide of N-deschlorobenzoyl-indomethacin is primarily

eliminated in the urine of monkeys and guinea pigs. The primary product of

excretion in rabbit urine was indomethacin glucuronide. O-desmethyl-indomethacin

was not detected. Neither free nor conjugated indomethacin was detected in rat

urine (Harman et al., 1964). There was also minimal urinary excretion of

indomethacin or its metabolites in the dog. In another study (Yesair et al., 1970), it

was found that dogs excreted most of the drug given i.v. unchanged in the feces. In

monkeys indomethacin was extensively metabolised to DBI and excreted in the

urine, while in rats DMI was equally excreted in urine and feces.

In human subjects, less than 33% of the administered dose is excreted as

indomethacin with 50% of that as the conjugate. Unconjugated desmethyl-

indomethacin and desmethyl-desbenzoyl indomethacin contributed to total fecal

excretion which was in the order of 21-42% of the dose (Duggan et al., 1972).

Following the oral administration of indomethacin (100 mg) in a single human

21

subject (Vree et al, 1993), the percentage of the dose excreted can be individualised

as follows: indomethacin (0.76%), DMI (0.58%), indomethacin acyl glucuronide

(32.1%), DMI acyl glucuronide (7.3%), and DMI ether glucuronide (29.8%) (Figure

4).

NKJ«schkvobmzcr/<HrKkMTwUucin -«cy( g lucuronKte

r s o g l u c u r o n t d e

Indomethacin Indomethacin-acyl glucuronide

etfior glucuronide

O^esmethylHndoniethacin (J^esraettiyWrKtomettwcin-acyt fltucufonide

Figure 4: Metabolism of indomethacin in the adult human (Adapted from

Vree et al., 1993)

22

Hucker et al., 1966, reported that less than 5% of an oral dose of

indomethacin was excreted unchanged in the urine. In other studies (Duggan et al.,

1972; Kwan et al., 1976), the amount of indomethacin recovered in urine was in the

order of 27%, half of which was in the form of glucuronide conjugate. Urinary free

metabolites recovered were about 18% and 13% of the dose administered

respectively, with the remainder as conjugates. Indomethacin, desmethyl-

indomethacin, and deschlorobenzoyl-indomethacin are found in human plasma and

urine in high quantities while desmethyl-N-deschlorobenzoyl-indomethacin appears

primarily in the feces (Shen and Winter, 1977; Duggan et al., 1972). Indomethacin

and its metabolites undergo enterohepatic circulation in humans, rats, dogs, and

monkeys (Kwan et al., 1976; Yesair et al, 1970). The metabolic pathway for

indomethacin in humans is summarised in Figure 4, while the routes of formation

are presented in Table 1.

Table 1: Metabolites of Indomethacin

Pathway Product of Metabolism Location of Metabolic Enzymes

O-Demethylation DMI Microsomal oxidation involving cytochrome

P450 Deacylation DBI Cytosol

Glucuronidation Glucuronides (acyl, ether) Endoplasmic reticulum

23 Maternal/Fetal/Neonatal Indomethacin Metabolism

In rabbits, there is a significant increase in the production of DMI after 2

weeks of life (Evans et al, 1981). In the same study, indomethacin was found to be

mainly metabolised to DBI in neonatal rabbit hepatocytes by non-microsomal

deacylation. The major components of eliminated drug in adults were the

glucuronides of indomethacin and its metabolites, however, less than 8% of the total

metabolites were conjugated in hepatocytes from 25 day old rabbits indicating that

glucuronidation was not the major metabolic pathway in the neonate (Evans et al,

1981). The rate of production of DBI and DMI in fetal rabbits is in the order of

0.23 and 0.13 nmoles/106/min, respectively. A very high maturational increase in

DBI production was determined on day 12 of postnatal life with a value of 1.3

nmole/106/min (Evans et al, 1981). In newborn rats the metabolism of

indomethacin is minimal and has been attributed to reduced affinity between

enzyme and substrate as well as to the presence of competitive inhibitors. There is,

however, a progressive increase in the formation of DMI from day 1 of postnatal

life to 60 days of age (Clozel et al, 1986). Endogenous cytosolic factors appear to

selectively inhibit microsomal oxidation of indomethacin in neonatal rabbit liver

(Evans et al., 1981). This finding is consistent with the changes in K m observed in

the neonatal rat study (Clozel et al, 1986).

24 1.1.2.5. Elimination Kinetics

Pharmacokinetic reports on indomethacin have been extensive ever since its

introduction in 1962. However, early studies depended heavily on

spectrophotometric and spectrofluorimetric methods of analysis and reported

elimination half-lives were in the range of 1.5-2.3 hours (Duggan et al., 1972;

Hucker et al, 1966; Hvidberg et al, 1972; Traeger et al, 1972). Given the

sensitivity limitations of these assay methods it seems highly probable that the true

elimination phase was not accurately determined. Using a more sensitive mass

fragmentographic method, however, the mean elimination half-life of indomethacin

was found to be in the range of 6-7.6 hours following either oral, i.v., or rectal

administration (Alvan et al., 1975).

While differences in assay methods and sensitivity account for some of the

inter- and intra-individual variability in elimination kinetics of indomethacin, the

majority of this is considered to be due to extensive enterohepatic recycling and the

irregularity of biliary discharge (Helleberg, 1981).

Plasma concentration-time data obtained from single dose studies were fit to

a two compartment open model (Alvan et al., 1975). Decay of indomethacin from

plasma was characterised by a rapid initial phase (t 1 / 2 ~ 90 min) and a slower

elimination phase (t 1 / 2 ~ 10 hours). The apparent volume of distribution of

25 indomethacin was in the order of 0.34 L/Kg - 1.57 L/Kg, and plasma clearance

varied from 0.044 - 0.109 L/Kg (Alvan et al., 1975). Steady state concentrations in

plasma ranged from 0.3-0.6 |ig/mL. The long elimination phase was speculated to

be due to enterohepatic recycling of the drug. There is no indication of dose-

dependent elimination (Alvan et al., 1975; Duggan et al., 1972).

Using fluorimetric methods, the half-life of indomethacin in adults and

newborns was found to be 2.2 h and 14.7 h respectively (Traeger et al., 1973).

Using a gas chromatographic (GC) method, Friedman et al., 1978, determined the

half-lives of indomethacin in two premature infants with patent ductus arteriosus to

be 21 and 24 h respectively, with a volume of distribution of 4.4 L/Kg and plasma

clearance of 0.145 L/Kg/h in one of the infants. While drug elimination in urine

was not determined, it was suggested that the prolonged half-life could be due to

renal immaturity. In another study involving nine preterm infants, the half-life

following oral administration was found to be 11-20 hours (Bhat et al., 1979).

These same authors report that protein binding of indomethacin was in the order of

95%. The kinetic parameters obtained for indomethacin in neonates from various

studies are summarised in Table 2.

26

Table 2: Pharmacokinetics of Indomethacin in Neonates

Postnatal Age (days)

Birth Weight (g)

hn (h) Varea (L/Kg) CI (mUKg/h)

Reference

7-14 1301+72 14.5 + 3 0.80 38 Thalji et al, 1979

10.8 + 6.0 1253 ± 389 20.7 ± 8 0.35+0.21 13+9.5 Thalji et al, 1980

9.5 + 2.2 1300 + 200 33.9 + 11.7 0.35 + 0.16 7.6 + 3.0 Vert et al, 1980

8.8 ±6.8 1213 + 309 32 0.19-0.61 7.0 Brash et al, 1981

10.2 + 2.2 1200 + 400 21.6 0.3 9.2 Yeh et al, 1985

5.7+4.2 1125 + 399 - 0.280 2.63 Wiest et al, 1991

9.5 + 1.3 1455 + 119 16.2 + 2.9 0.36 + 0.01 17.1+2.8 Bhat et al, 1980

The elimination half-life is markedly prolonged in neonates when compared

to adults and is related inversely to gestational age (Vert et al, 1980; Evans et al,

1979). Data, however, is severely lacking for indomethacin in-utero

pharmacokinetics in the fetus.

Following a single oral dose of indomethacin, the total clearance in the

elderly (79.5 ± 1.3 y) was 0.8 mL/min/Kg compared to 1.4 mL/min/Kg in younger

subjects (36.9 ± 3 y) with an elimination rate constant of 0.23 h"1 and 0.32 h"1 ,

respectively (Oberbaur et al, 1993). The fall in clearance in the elderly has been

attributed to a decline in renal function and a reduced renal excretion of

indomethacin since about 27% of the dose is eliminated through the kidneys

27 (Duggan et al, 1972; Kwan et al., 1976; Traeger et al., 1973; Oberbaur et al.,

1993).

1.1.3. Placental Transfer

The maternal-placental-fetal unit is a complex pharmacokinetic system

comprised of two independent circulations. The lipoprotein nature of the placenta

is, however, very similar to other biological membranes in the body and plays an

important role in controlling drug transfer. The unbound lipophilic drug in the

unionised form can easily cross the placenta by passive diffusion while other

mechanisms such as facilitated diffusion, active transport, and pinocytosis are more

prevalent for placental passage of nutrients and other endogenous substances. The

mechanisms of placental transfer and subsequent degree of fetal drug exposure have

been the subjects of numerous reviews over the years (Moya and Thorndike, 1962;

Green et al, 1979; Krauer et al, 1980; Levy, 1981; Szeto et al, 1982b, 1982d;

Mihaly and Morgan, 1984; Reynolds and Knott, 1989; Szeto, 1993).

Several factors may potentially affect maternal-fetal drug exchange and fetal

exposure during the progression of normal pregnancy including the following: (i)

physico-chemical properties such as lipophilicity, molecular weight, and ionisation,

(ii) placental permeability, stage of gestation, and type of placentation, (iii)

orientation of maternal and fetal placental blood flows, (iv) placental metabolism of

28 drugs, (v) fetal circulation, (vi) pH, (vii) uterine and umbilical blood flows, and

(viii) protein binding.

1.1.3.1. Pharmacokinetic Characterisation

Pharmacokinetic evaluation of maternal/fetal drug disposition generally

involves consideration of the mother and fetus each as a single compartment. Such

a system assumes quick equilibration of the drug. If the drug enters fetal tissues

slowly with concurrent elimination from the maternal compartment, a deep

compartment is presumed to exist. The pharmacokinetic data of meperidine was

treated using the mother and fetus each as a single compartment (Szeto et al., 1978).

Such a system assumes fetal and maternal drug concentrations to be at steady state.

If a biexponential decay of the drug is observed in the fetus, then the fetus should be

considered as a two compartmental model to account for fetal distribution kinetics

(Szeto, 1982). The placenta (Krauer and Krauer, 1977) and the amniotic fluid

(Brien and Clarke, 1988) have also been considered as distinct kinetic

compartments.

1.1.3.2. The Chronically Instrumented Pregnant Sheep Model

Several experimental animal models, such as rats, guinea pigs, baboons,

goats, pigs, sheep, and monkeys, have been used in placental transfer studies due to

29 ethical and technical constraints precluding such studies in humans. The commonly

used animal models in fetal medicine have been reviewed (Nathanielsz, 1980).

Small animal models such as rats and guinea pigs are often inadequate when serial

sampling is required due to low blood volume. Of the available models, the

anatomy and physiology of the monkey closely parallels the human situation and

theoretically would seem to be the best to conduct studies on placental transfer.

However, several complications such as the cost of the preparation, ease of

handling, and size limitation of the fetus restrict the use of this model.

Sheep, on the other hand, are docile, have large fetuses conducive for chronic

preparation and similar physiology and biochemistry (Comline and Silver, 1974)

compared to humans. Both sheep and primate models demonstrate similar fetal

behavioural states (viz. fetal breathing movements, eye movements, etc) in the last

trimester of gestation (de Vries et al., 1982; de Vries et al., 1985; Nijhuis et al.,

1982). They have been used to study in-utero fetal physiology and placental

transfer of a number of endogenous compounds (Szeto et al., 1978; Van Petten et

al., 1978; Rurak et al., 1991). Sheep do, however, possess a different type of

placentation, and care must be excercised when extrapolating data to the human

situation, as is necessary with any animal model.

1.1.3.3. Placental Transfer of Indomethacin 30

Indomethacin has been reported to traverse the placental membrane in rabbits

(Parks et al, 1977; Harris and Van Petten, 1981), rats. (Klein et al, 1981), sheep

(Harris and Van Petten, 1981; Anderson et al, 1980c), and humans (Moise et al,

1990), however, detailed information is lacking.

When indomethacin was given orally at a dose of 2 mg/kg in pregnant

rabbits, in a study period of two hours, the drug was detected in maternal (M) and

fetal (F) blood and amniotic fluid within 15 minutes; the F/M ratio at the end of the

study being 0.568. Although indomethacin did appear in the amniotic fluid, its

concentration relative to maternal and fetal blood levels was much lower. At the

end of the study period detectable concentrations of free drug were seen in maternal

urine, i.e., 19.2 |i.g/mL, and bile, i.e., 0.35 |ig/mL (Parks et al, 1977). In another

study involving pregnant rabbits, indomethacin was administered subcutaneously in

a dose of 10 mg/kg for a study period of six hours (Harris and Van Petten, 1981). In

contrast to the former study where fetal plasma levels remained lower than maternal

levels throughout the experiment, fetal plasma indomethacin concentrations

exceeded maternal levels and remained elevated over a period of time before falling.

At the end of the 1st hour, the F/M ratio was 0.795. Indomethacin was detected in

the amniotic fluid 2 hours after dosing and its concentration at the end of six hours

was comparable to that of maternal plasma. A major concern as to the reliability of

31 these data is the analytical methodology involved, namely, spectrophotometric and

spectrofluorimetric methods of analysis.

In a study with pregnant rats, indomethacin was administered at a dose of 4

mg/kg by gastric intubation. At the end of the 3rd hour, maternal and fetal plasma

indomethacin concentrations were 16.8 ± 1.6 |ig/mL and 4.6 ± 0.3 |Lig/mL,

respectively, with a F/M ratio of 0.274 (Klein et al, 1981).

In 1980c, Anderson et al, were the first group to attempt description of

placental transfer in quantitative terms. Their study involved simultaneous

administration of 14C-indomethacin as bolus in the fetus and 3H-indomethacin as an

infusion in the ewe. Maternal and fetal blood were analysed for radioactivity in a

scintillation counter and placental blood flows measured using radiolabeled

microspheres. Placental clearance (Clp) was determined by the formula, C l p = Ctot x

Fss/Mss, where Ctot is the total clearance, and F s s and M s s are fetal and maternal

steady state concentrations. Fetal and placental tissue clearances were estimated to

be 3.63 ± 0.58 ml/min/kg and 2.10 + 0.32 ml/min/kg, respectively, with a F/M ratio

of 0.28. Major criticisms of this method include: (a) the requirement that steady

state conditions prevail, (b) it only provides single point estimation, (c) the model

is limited to cases where fetal clearance is zero and simultaneous solving of the

equations is impossible (Szeto, 1982b), (d) it is applicable only in instances where

drug clearance by the placenta from mother and fetus are equal, (e) the issue of an

: 32

isotope effects not addressed, and (f) the method cannot differentiate intact drug

from its metabolites.

When indomethacin was administered as an intravenous infusion at a dose of

10 mg/kg into pregnant ewes, maternal and fetal plasma levels at the end of infusion

were 13.5 ± 0.7 Lig/mL and 0.6 ± 0.1 pig/ml with a F/M ratio of 0.04 (Harris and

Van Petten, 1981). Indomethacin concentrations in amniotic fluid were markedly

higher than fetal and maternal plasma levels at the end of the study period. A major

concern is again the non-specific analytical methodology involved.

Moise et al., 1990, administered 50 mg of indomethacin orally to pregnant

human patients. Maternal and fetal indomethacin levels (cordocenteses) were 218 +

21 ng/ml and 219 ± 13 ng/ml, respectively, with a F/M ratio of 1.004. The fetal

plasma to amniotic fluid concentration ratio was 10.0 ± 1.2.

1.1.4. Analytical Methods

Indomethacin has been analysed in biological fluids, for a variety of study

requirements, by several methods including liquid chromatography (HPLC), gas

chromatography (GC), mass fragmentography, radioimmunoassay (RIA), and

spectrophotometric methods. The detection limits of most HPLC assays fall in the

range of 10 - 100 ng/mL, GC assays in the range of 1-50 ng/mL, and RIA methods

33 in the range of 50 ng/mL. Spectrophotometric methods of analysis are sensitive in

the |ig range with poor selectivity.

Major disadvantages of available methods include:

1. Radioimmunoassay is highly cross reactive to glucuronide conjugates of

indomethacin and its metabolites (Hare et al., 1977).

2. HPLC assays with fluoresence detection require post-column in-line hydrolysis

necessitating auxiliary pump setups (Bernstein and Evans, 1982; Mawatari et

al., 1989), and those with UV detection are not sensitive in the lower nanogram

range.

3. GC methods are plagued by long retention times (Nishioka et al., 1990), multi-

step derivatisation, lack of internal standards (Helleberg et al., 1976; Ferry et

al., 1974), long derivatisation reaction times (Nishioka et al., 1990), use of

hazardous and corrosive derivatising agents (Evans, 1980; Helleberg et al.,

1976), and inadequate internal standards (Evans, 1980; Guissou et al., 1983).

Other complications include large sample size (>1 mL), inadequate recovery,

multiple extractions, and high variability at lower concentrations.

Of the available methods, GC with electron-capture detection would appear

to provide the best sensitivity and selectivity; only one GC method has applied

34 fused silica capillary column technology to indomethacin measurement to date

(Nishioka et al, 1990).

Several derivatising reagents have been investigated for GC assays, viz.,

diazomethane (Guissou et al, 1983; Giachetti et al, 1983; Plazonnet and Van den

Heuvel, 1977), diazopropane (Arbin, 1977), l-ethyl-3-p-tolyltriazine (Nishioka et

al, 1990), pentafluorobenzyl bromide (Sibeon et al, 1978), hexafluoroisopropanol

and trifluoroacetic anhydride (Matsuki et al, 1983), bis (trimethylsilyl) acetamide

(Plazonnet and Van den Heuvel, 1977), ethyl iodide and extractive alkylation

(Jensen, 1978), pentafluoropropanol and pentafluoropropionic anhydride (Evans,

1980), and diazoethane (Helleberg, 1976). The simplest of the derivatisation

reactions is with bis (trimethylsilyl) acetamide, a silylating agent producing

trimethylsilyl esters of indomethacin.

Selection of internal standards for indomethacin, by either HPLC or GC, is

yet another problem encountered in the literature. Several reports incorporate

internal standards with no structural similarity to indomethacin including

phenylbutazone (Mehta and Calvert, 1983), phenacetin (Avgerinos and

Malamataris, 1989), testosterone (Kwong et al, 1982), itraconazole (Al-Angary et

al, 1990), and penfluridol (Guissou et al, 1983). Others use inappropriate internal

standards such as indomethacin propyl ester (Nishioka et al, 1990) and

indomethacin methyl ester (Arbin, 1977), and still others do not use any at all

35 (Helleberg, 1976; Ferry et al, 1974). Some suitable internal standards include di­

methyl indomethacin (Stubbs et al., 1986; Bayne et al, 1981), and 5-fluoro

indomethacin (Sibeon et al., 1978).

1.2. Rationale and Specific Aims

1.2.1. Rationale

Available analytical methods for indomethacin were impractical for use in

the current study design with respect to sensitivity, selectivity, speed, sample

volumes, and variability at lower concentrations. Therefore, there is a need to

develop a sensitive, selective, and rapid analytical procedure.

Available data on placental transfer were limited and may be unreliable

because of the analytical methodology used and inadequate experimental design.

The chronically instrumented pregnant sheep modelis a versatile animal model for

studying placental drug transfer and maternal/fetal drug pharmacokinetics. The

model closely simulates the human system with respect to fetal renal,

cardiorespiratory, and behavioural states (de Vries et al., 1982; de Vries et al.,

1985) and fetal biochemistry and physiology (Van Petten et al., 1978). The model

also allows for serial sampling to an extent which is impractical in small animal

models such as rats, rabbits, and guinea pigs.

36

Reports on the limited placental transfer of indomethacin in sheep preclude

the use of the Szeto model (Szeto et al., 1982b) where measurable steady-state

concentrations of drug must be obtained in both the ewe and fetus following

individual maternal and fetal infusion. This coupled with the disadvantages of the

Anderson method (Anderson et al., 1980c) of clearance estimation illustrate the

need to determine placental clearance values by an alternate technique. The Fick

diffusion model, employing direct fetal administration to provide plasma

concentrations corresponding to those observed in studies of human tocolysis

(Moise et al., 1990), appeared then, to be a more suitable and reliable method for

our studies.

1.2.2. Specific Aims

1. To develop and optimise a sensitive and selective gas chromatographic method

with electron-capture detection for indomethacin in sheep biological fluids using

fused-silica capillary column technology.

2. To apply the developed method for the determination of plasma, urine, amniotic

and tracheal fluid indomethacin concentrations following long term infusion of

the drug in chronically instrumented sheep fetuses, and to calculate placental,

non-placental, and renal clearances.

37

2. EXPERIMENTAL

2.1 Materials and Supplies

2.1.1 Chemicals

Phenylbutazone (4-butyl-l,2-diphenyl-3,5-pyrazolidinedione, lot 110H0632),

acemetacin (1 -[p-chlorobenzoyl]-5-methoxy-2-methylindole-3-acetic acid

carboxymethyl ester, lot 38F0022), sulindac ([Z]-5-fluoro-2-methyl-l-[p-

(methylsulfinyl) benzylidene] indene-3-acetic acid, lot 101H0643), indomethacin

(l-[p-chlorobenzoyl]-5-methoxy-2-methylindole 3-acetic acid, lot 60H0448),

carprofen (lot 28F0440), zomepirac sodium (5-[p-chlorobenzoyl]-l,4-dimethyl-

pyrrole-2-acetic acid [sodium salt], lot 13H0228) and flufenamic acid (lot 61H3562)

were obtained from Sigma Chemicals Co., St. Louis, MO, USA. 5-Fluoroindole-2-

carboxylic acid (lot 10130DF) was purchased from Aldrich Chemicals Co.,

Milwaukee, WI, USA. 5-Fluoroindomethacin (Merck Research Laboratories, West

Point, PA, USA) and a-methylindomethacin (Merck-Frosst Inc., Kirkland, Quebec)

were generous gifts. Injectable ampicillin (PenbritinR), injectable gentamicin

sulphate (GaramycinR), injectable atropine sulphate, halothane (FluothaneR), and

injectable heparin (HepaleanR), were obtained from the Pharmacy Department of

Women's Hospital, Vancouver, BC. Sodium chloride for injection USP was

38 obtained from Abbott Laboratories, Montreal, Quebec. The enzyme, p-

glucuronidase (Bovine liver, type B-l), lot 117F7256, was purchased from Sigma

Chemicals Co., St. Louis, MO, USA.

2.1.2 Reagents

Sodium acetate (anhydrous, analytical reagent grade) was obtained from

BDH (Toronto, Ontario) and glacial acetic acid (American Chemical Society, ACS

reagent) from Allied Chemical (Pointe Claire, Quebec). Triethylamine was

purchased from Pierce Chemical Co., Rockford, Illinois, USA.

Pentafluorobenzylbromide (PFBBr, lot 930413082), N-methyl-N-(tert-

butyldimethylsilyl) trifluoroacetamide (MTBSTFA, lot 930817150), and N,0-Bis-

(trimethylsilyl) acetamide (BSA), were purchased from Pierce Chemical Co.

(Rockford, Illinois, USA). 2,2,3,3,3-Pentafluoro-l-propanol (97%) and

pentafluoropropionic anhydride (99%) were obtained from Aldrich Chemical Co.,

Milwaukee, WI, USA.

2.1.3 Solvents

39

The following (distilled in glass) were purchased from Caledon Laboratories

Ltd., Georgetown, Ontario: ethyl acetate, ethyl ether, hexane, toluene,

dichloromethane, and acetonitrile. Deionised, high-purity water (hereafter refered

to as water) was produced on-site by reverse osmosis using a Milli-Q® water system

(Millipore, Mississauga, Ontario).

2.1.4. Gases

Hydrogen, carrier gas for G C - E C D , and argon-methane (95:5), make-up gas

for E C D , and ultra-pure helium, carrier gas for G C - M S , were obtained from

Matheson Gas Products, Edmonton, Alberta. Nitrogen N F was obtained from

Praxair Canada Inc., Mississauga, Ontario.

2.1.5. Supplies

Membrane filters (0.45) (Millipore, Mississauga, Ontario); disposable plastic

pipet tips (National Scientific, San Rafael, C A , USA); needles and disposable

plastic syringes (Luer-LokR) for drug administration and sample collection

(Beckton-Dickinson Canada, Mississauga, Ontario); borosilicate glass pasteur

pipets (John Scientific, Toronto, Ontario); heparinised blood gas syringes

(Marquest Medical Products Inc., Englewood, C O , USA); heparinised VacutainerR

40 tubes (Vacutainer Systems, Rutherford, NJ, USA); PyrexR 15 mL disposable

culture tubes (Corning Glass Works, Corning, NY, USA); polytetrafluoroethylene

(PTFE) lined screw caps (Canlab, Vancouver, British Columbia); polystyrene tubes

(Evergreen Scientific International Inc., Los Angeles, CA, USA); polyvinyl tubing

for catheterisation (Dow Corning, Midland, MI, USA).

2.2 Instrumentation

2.2.1 Gas Chromatography

A HP (Hewlett-Packard, Avondale, PA, USA) model 5890 series II gas

chromatograph equipped with a HP model 7673 autosampler, split-splitless capillary

inlet system, model HP 3365 chemstation, and a 6 3Ni electron-capture detector, was

used for all analyses. A PyrexR glass inlet liner (78 mm x 4 mm, i.d.) and

ThermogreenR LB-2 silicone rubber septum (Supelco, Bellefonte, PA, USA) were

used for all analyses.

The optimised operating conditions for routine analyses were: column,

Ultra-2® fused silica capillary column cross-linked with 5% phenylmethylsilicone,

25m x 0.31 mm x 0.52p:m film thickness, (Hewlett-Packard, Avondale, PA, USA);

injection mode, splitless; injection port temperature, 200 °C; initial column

41 temperature, 210 °C (1 min); oven programming rate, 40 °C/min; final oven

temperature, 300°C (6 min); detector temperature (ECD), 330 °C; carrier gas, ultra

high purity hydrogen; column head pressure, 10 psi (i.e., 70 kPa, flow rate ~1

ml/min); purge delay time, 0.75 min; make-up gas, argon-methane, 95:5 (flow rate,

65 mL/min).

2.2.2. Gas Chromatography-Mass Spectrometry

A HP model 5890 series II gas chromatograph (Hewlett-Packard, Avondale,

PA, USA) equipped with a HP 7673 automatic injector system, and an HP model

5989A mass spectrometer, HP 59827A vacuum gauge controller, and an HP

98789A on-line computer, with a HP Ultra-2® fused silica capillary column (25 m x

0.31 mm x 0.52|im film thickness) was used for mass spectrometric studies. The

carrier gas, helium, was operated at a column head pressure of 3 psi. In the

chemical ionisation mode, methane was used as the reagent gas. The electron-

impact-MS ionisation energy was 200 eV, emission current 250 |iA, and source

pressure, 1.2 torr.

2.2.3. Columns

42 Ultra-2® fused silica capillary column cross-linked with 5%

phenylmethylsilicone (25 m x 0.31 mm, i.d., x 0.52 |um film thickness) and Ultra-2®

(25 m x 0.2 mm x 0.33 \im film thickness) narrow-bore columns were obtained

from Hewlett-Packard, Avondale, PA, USA; a DB-1701 column (30 m x 0.25 mm,

i.d., x 0.25 \im film thickness) was obtained from J & W Scientific, Folsom, CA,

USA.

2.2.4. Liquid Chromatography-Mass Spectrometry/Mass Spectrometry

A Varian VG Biotech LC/MS/MS system (Fisons Instruments, VG Biotech,

Altrincham, UK) equipped with an electrospray interface, VG Quattro triple

quadrupole mass spectrometer (two high performance quadrupole mass analysers,

and a hexapole collision cell between the two mass analysers), a Dynolite™ detector

system, with a PC-compatible computer to run the VG MassLynx software system

(Mass Spectrometry Data System, VG Biotech, Fisons Instruments, Pointe Claire,

Quebec, Canada).

2.2.5. Differential Scanning Calorimeter

43 A 910 Differential Scanning Calorimeter module interfaced to a Du Pont

Series 99 Thermal Analyzer (Du Pont, Wilmington, DE, USA) was used for

indomethacin melting point and water content determinations.

2.2.6. Nuclear Magnetic Resonance Spectrometry

Proton NMR spectra were obtained on a Briiker WH-200 spectrometer (200

Mhz) in the Department of Chemistry, University of British Columbia, Vancouver,

BC. D6-dimethylsulfoxide was used as the dissolution solvent.

2.2.7. General Equipment

Vortex-type mixer (Vortex-GenieR), incubation oven (IsotempR 500 series),

AccumetR pH meter 915, Fisher Scientific Industries, Springfield, MA, USA; IEC

Model HN-SII centrifuge, Damon/IEC Division, Needham Hts. MA, USA;

rotating-type tube mixer (LabquakeR Tube Shaker, model 415-110), Labindustries,

Berkeley, CA, USA; infusion pump (Harward model 944), Harvard Apparatus,

Millis, MA, USA.

2.2.8. Physiological Monitoring

44 Polygraph recorder to monitor fetal arterial pressure, venous pressure,

amniotic fluid pressure, heart rate, and breathing rate (Sensormedics Inc., Anaheim,

CA, USA); Apple He computer and computer data acquisition system consisting of

Interactive Systems (Daisy Electronics, Newton Square, PA, USA), analog to digital

converter and clock card (Mountain Software, Scott's Valley, CA, USA);

Instrument Laboratory System 1306 pH/blood gas analyser (Instrument Laboratory

System, Lexington, MA, USA); Advanced Digimatic Osmometer 3D2 to measure

urine osmolality, Needham Heights, MA, USA; American Optical hand-held

refractometer to measure plasma protein concentration (American Optical); YSI

Glucose/Lactate Analyser (Yellow Springs Instruments, Yellow Springs, CO, USA)

to measure glucose and lactate concentrations; Nova 5 Electrolyte Analyser

(Newton, MA, USA) to measure urine electrolyte concentrations.

2.3. Preparation of Stock Solutions

2.3.1. Preparation of Drug Stock Solution

2.3.1.1 Indomethacin

Approximately 10 mg of indomethacin was accurately weighed, transfered to

a 50 mL volumetric flask and made up to volume with methanol. A 0.5 mL aliquot

of solution was diluted with water to volume in a 50 mL volumetric flask. One mL

45 of the latter solution was further diluted to volume with water in a 50 mL volumetric

flask to give a final working stock solution with a concentration of 0.04 |ig/mL.

Volumes of 25, 50, 200, 400, and 800 |iL were used to prepare the calibration curve.

2.3.2 Preparation of Internal Standard Solutions

2.3.2.1. Zomepirac Sodium

Approximately 27.04 mg of zomepirac sodium (equivalent to ~ 25 mg of

zomepirac free acid) was accurately weighed and transfered to a 250 mL volumetric

flask and made up to volume with water. A 1 mL aliquot of this solution was

transfered to a 100 mL volumetric flask and made up to volume with water to

produce a working stock solution with a concentration of 1000 ng/mL. A 100 |iL

aliquot of this solution was used in the assay.

2.3.2.2. Sulindac

Approximately 10 mg of sulindac was accurately weighed and transfered to a

50 mL volumetric flask and made up to volume with methanol. A 0.5 mL aliquot of

solution was diluted with water to volume in a 50 mL volumetric flask producing a

46 concentration of 2 jig/mL. A volume of either 100 |iL (200 ng) or 50 |iL (100 ng)

was used.

2.3.2.3 a-Methyl Indomethacin

Approximately 5 mg of alpha-methyl indomethacin was accurately weighed

and made up to 1 mL with methanol. An aliquot of 100 p:L was transfered to a 10

mL volumetric flask and made up to volume with methanol. A 100 uX aliquot of

the latter solution was transfered to another 10 mL volumetric flask and made to

volume with water to produce a working solution with a concentration of 500

ng/mL. A volume of 50 \ih (25 ng) was used in the assay.

2.3.2.4 Other Experimental Internal Standards

Working solutions of 2 |ig/mL were separately prepared for: 5-fluoro indole-

2-carboxylic acid, 5-fluoro DL tryptophan, 5-fluoro indole-3-acetic acid, flufenamic

acid, carprofen, acemetacin, phenylbutazone, and 5-fluoro-indomethacin.

2.3.3. Preparation of Reagent Solutions

47 Acetate buffer pH 5.0 was prepared according to British Pharmacopoeia

standards by dissolving 13.6 g of sodium acetate and 6 mL of glacial acetic acid in

sufficient water to produce 1000 mL. An aliquot of 2 mL (excess) was used in the

assay to produce an aqueous phase pH of ~ 5.00.

Acetate buffer pH 2.45, B.P., was prepared by mixing 200 mL of 1M

hydrochloric acid with 200 mL of 1 M sodium acetate and diluting to 1000 mL with

water. The pH of the final solution was adjusted to 2.45 with 1 M hydrochloric

acid.

Acetate buffer pH 3.5 was prepared by dissolving 25 g of ammonium acetate

in 25 mL of water and adding 38 mL of 7 N hydrochloric acid.

Acetate buffer pH 4.6 was prepared by dissolving 5.4 g of sodium acetate in

50 mL of water, adjusting to pH 4.6 with glacial acetic acid and diluting to 100 mL

with water.

Acetate buffer pH 6.0 was prepared by dissolving 100 g of ammonium

acetate in 300 mL of water, adjusting to pH 6.0 with 4.1 mL of glacial acetic acid

and diluting to 500 mL with water.

48 2.4. Sample Preparation

2.4.1. Analysis of Intact Indomethacin Levels

Ovine fetal fluids (plasma, urine, etc.), typically 0.1 mL, were pipeted into

clean 15 mL borosihcate Kimax culture tubes with polytetrafluoroethylene

(PTFE)-lined screw caps. To the biological fluid were added 50 fiL of internal

standard, cx-methyl indomethacin, and 2 mL of acetate buffer pH 5.0 and the

mixture adjusted to a final volume of 3.0 mL with water (final pH of the aqueous

phase ~ 5.00). The mixture was gently vortex-mixed and 5 mL of ethyl acetate was

added. The aqueous phase was extracted for 20 minutes on a rotary shaker. The

samples were then placed in a freezer at -5 °C for 10 minutes to facilitate breakage

of any emulsion formed in the extraction step. This was followed by centrifugation

for 10 minutes at 3000 g. The upper organic layer was transfered to clean 15 mL

tubes and evaporated to dryness in a water bath maintained at 37 °C under a gentle

stream of nitrogen gas. To the residue was added 100 |iL of toluene containing 8

|iL of MTBSTFA. This was vortex-mixed and placed in an oven at 60 °C for 50

minutes. The samples were allowed to cool to room temperature after which 200

|iL of toluene was further added, vortex-mixed, and transfered to automatic sampler

injection vials. Aliquots of 2 |iL were injected into the gas chromatograph.

Scheme-I summarises the sample preparation protocol.

49 Biological Sample

(fetal plasma, urine, amniotic, or tracheal fluid) 0.1-1.0 mL

Internal Standard (a-methyl indomethacin, 50 pL, 25 ng)

2.0 mL pH 5.0 Acetate Buffer +

Distilled Water, q.s. to 3.0 mL total aqueous phase +

5.0 mL Organic Solvent (Ethyl Acetate)

Mix 20 minutes on rotary shaker

Freeze at -5 C for 10 minutes

Centrifuge at 3000 g for 10 minutes

Aqueous Organic (Waste)

T Evaporate to dryness under

nitrogen gas at 37 C

i

Derivatise residue with MTBSTFA (8 pL) and toluene (92 uL) at 60°C for 50 minutes

i Cool to room temperature and add toluene

to obtain a final reconstitution volume of 300 uL

I 2.0 uL for injection

Scheme 1: Sample Preparation Procedure For Indomethacin

50 2.4.2. Analysis of Glucuronide Conjugates

The glucuronide conjugates in urine and amniotic fluid samples were

analysed by enzyme hydrolysis using ( -glucuronidase. The method for preparation

of samples is similar to that described above. About 200 mg of (3-glucuronidase

was accurately weighed and made to volume in a 25 mL volumetric flask with pH

5.00 acetate buffer, resulting in a concentration of 5000-Units/mL. Prior to

extraction with ethyl acetate, 0.5 mL of the enzyme solution was added to the

buffered urine samples, and incubated at 37°C for 3 hours. Due to limited volume

of urine samples, it was not possible to conduct a rigorous study to determine

optimum hydrolytic conditions (incubation time and amount of enzyme) with a

satisfactory "n" value. However, a study was carried out with these limitations

along with input from Bayne et al, 1981, and Vree et al, 1993 (Vree et al., 1993,

determined optimum conditions for indomethacin conjugates in urine to be 10,000

Units of enzyme activity and an incubation time of 2 hours, while the conditions

reported by Bayne et al., were 1500 Units and 30 minutes, respectively). The

concentration of glucuronide-conjugated indomethacin was calculated by

subtracting the non-conjugated indomethacin concentration (as determined by the

method described in section 2.4.1 above) from the samples measured following

hydrolysis.

51

2.5. Preparation of Calibration Curve

2.5.1. Calibration Curve with ex-Methyl Indomethacin

Serial quantities of the working indomethacin stock solution (1, 2, 8, 16, and

32 ng) were added to 0.1 mL of blank fetal sheep biological fluid (plasma, urine,

amniotic, or tracheal fluid). A 50 p:L aliquot of the internal standard (a-methyl

indomethacin) was added, and the solution made up to a fixed volume of 1 mL. The

samples were then extracted and derivatised as described previously. Determination

of indomethacin concentration was made by plotting the peak area ratios of the tert-

butyldimethylsilyl derivatives of indomethacin and a-methyl indomethacin against

the known amount of indomethacin added to each sample.

2.5.2. Calibration Curve with Zomepirac Sodium

Serial quantities of the working indomethacin stock solution (1, 2, 8, 16, and

32 ng) were added to 0.1 mL of blank fetal sheep biological fluid (plasma, urine,

amniotic or tracheal fluid). A 100 jiL aliquot of the internal standard (zomepirac

sodium) was added, and the solution made up to a fixed volume of 1 mL. The

samples were then extracted and derivatised as described previously. Determination

52 of indomethacin concentrations was made by plotting the peak area ratios of the

terf-butyldimethylsilyl derivatives of indomethacin and zomepirac against the

known amount of indomethacin added to each sample.

2.6. Chemistry

2.6.1. Derivatisation Procedures For Indomethacin

[1] Reaction with bis (trimethylsilyl') acetamide (BSA): Method of Plazonnet

and vanden Heuvel, 1977.

A specified amount of indomethacin (1 (J-g/mL) was evaporated to dryness

under oxygen-free nitrogen in a water bath maintained at 40 °C. The residue was

dissolved in 90 |iL of ethyl acetate and 10 |iL of BSA added. The contents were

derivatised for 15 min at 60 °C in a hot air oven, cooled to room temperature and

diluted to 1 mL with ethyl acetate before injecting 1 jiL into the GC.

[2] Reaction with N-methvl-N-(tert-butyldimethvlsilvl)-trifluoroacetamide

(MTBSTFA)

53

A methanolic solution of indomethacin (0.5 (Xg/mL) was evaporated to

dryness under oxygen-free nitrogen gas in a water bath maintained at 40 °C. The

residue was derivatised with 100 | i L of toluene containing 8 \iL of M T B S T F A at 60

°C for 30 minutes. The solution was cooled to room temperature and 1 or 2 | i L

injected into the G C .

[3] Reaction with pentafluorobenzyl bromide (PFBBr): Method of Sibeon et al.,

1978)

A methanolic solution of indomethacin (0.5 |i.g/mL) was evaporated to

dryness under oxygen-free nitrogen gas in a water bath maintained at 40 °C. To

each tube about 25 mg of potassium carbonate (catalyst) and 0.5 m L of a stock

solution of PFBBr (10 |J,L of supplied solution is made up to 10 mL with acetone)

were added. The reaction was carried out at 60 °C for 30 minutes. Excess

derivatising agent was removed using a stream of nitrogen gas at room temperature.

To each tube was then added 1 mL of distilled water and 1 mL of hexane. The

tubes were mixed on a shaker for 10 minutes and centrifuged at low speed for a

further 10 minutes. About 2 | i L of the upper hexane layer was injected into the G C .

[4] Reaction with a mixture of pentafluoropropanol in pentafluoropropionic

anhydride (PFP): Method of Evans. 1980.

54

A methanolic solution of indomethacin was evaporated to dryness under

oxygen-free nitrogen in a water bath maintained at 40 °C. To the residue was added

10 p:L of a mixture of 2,2,3,3,3,-pentafluoro-l-propanol in pentafluoropropionic

anhydride (1:4, v/v) and derivatised at 75 °C for 20 min. The solution was allowed

to cool to room temperature, evaporated to dryness under nitrogen, and reconstituted

in 10 (J.L of ethyl acetate and 1 or 2 \xL injected into the GC.

2.7. Optimisation of Assay Parameters

2.7.1. Derivatisation of Indomethacin

The following derivatisation parameters were examined:

1. Selection of a suitable derivatising agent: reaction with BSA, MTBSTFA,

PFBBr, or PFP (see section 2.6.1.).

2. Volume of derivatising agent-2 |iL to 100 (J,L.

3. Temperature of derivatisation-27 to 90 °C.

4. Time course of derivatisaton-10 min-1 hour.

5. Selection of derivatisation solvent (toluene, acetonitrile, etc.)

6. Reconstitution volume (0.1-1.0 mL)

55

2.7.2. Chromatography

1. Selection of a suitable internal standard (sulindac, indole derivatives, zomepirac,

carprofen, acemetacin, phenylbutazone, 5-fluoroindomethacin, a-methyl

indomethacin).

2. Injection mode (split vs splitless).

3. Injection port temperature (140-300 °C).

4. Oven temperature programming (various initial temperatures, ramp rates, and

final temperatures and times were examined).

5. Column (nonpolar vs polar, wide bore narrow bore).

6. Column head pressure (5-15 psi).

7. Injector port liner configuration (packed vs unpacked inlet liner).

8. Purge activation time (10-120 seconds).

9. Detector temperature (300-360 °C).

2.7.3. Extraction

1. Solvent for extraction (ethyl acetate, hexane, hexane-ethyl acetate mixtures,

toluene, toluene-triethylamine (0.0125M) mixtures, hexane-isopropylalcohol

(10%) mixtures, diethyl ether, dichloromethane).

5 6

2. Volume of solvent (1-5 mL).

3. pH of extraction medium (pH 2-6.00, acetate buffers).

4. Time of extraction (5-60 minutes).

2.7.4. Validation

2.7.4.1. Extraction Recovery Studies

Extraction recoveries of indomethacin were determined at five concentration

points, i.e., 1, 2, 8, 16, and 32 ng/ml. Two groups of samples were used in this

study, namely the test and the control group. Samples in both groups contained

blank ovine fetal biological matrix (plasma, urine, amniotic or tracheal fluid), and

the internal standard, a-methyl indomethacin. Only samples from the test group

were spiked with IND to produce final concentrations of 1, 2, 8, 16, and 32 ng/ml.

Both test and control group samples were subjected to the liquid-liquid extraction

procedure (see section 2.4.1.) after which, aliquots of indomethacin, prepared in

methanol, were added to the control samples to produce concentrations of 1, 2, 8,

16, and 32 ng/ml. Control and test group samples were dried under nitrogen,

derivatised with MTBSTFA, reconstituted with toluene, and chromatographed as

described earlier (section 2.4.1.). Extraction recovery at each concentration point

57 was calculated as the ratio of peak area counts observed with test (extracted) to that

in control (unextracted) and expressed in percentage.

2.7.4.2. Inter-day & Intra-day Variability

Intra-day and inter-day variability were determined at five concentration

points, namely, 1, 2, 8, 16, and 32 ng/ml. For evaluating intra-day variability, three

calibration curves were prepared on the same day, and for inter-day variability,

calibration curves were prepared on three separate days (values are reported as %

C.V.).

2.7.4.3. Linearity

The linearity of the calibration curves was investigated over the

concentration range of 1-500 ng/mL for plasma and urine samples.

2.7.4.4. Stability Studies

Studies were performed periodically to assess the stability of the samples

during storage and analysis. Freezer storage stability was determined at one

concentration point (8 ng/ml): a batch of biological matrices (plasma and urine)

58 were spiked with a known amount of indomethacin, stored at - 2 0 ° C , removed

periodically (over 14 months) and analysed. In order to assess the stability of

indomethacin during freeze-thaw cycles, blank biological samples (plasma) were

spiked with indomethacin (8 ng/ml), frozen at standard freezer temperatures (-

20°C) , and thawed at room temperature on the bench-top for four consecutive

cycles, and analysed. A bench-top stability study (ambient conditions) was

performed which involved spiking plasma and urine samples with a known amount

of indomethacin, and allowing them to stand at room temperature for 0, 12, and 24

h. At the end of each study, the internal standard, oc-methyl indomethacin, was

added and the samples extracted and analysed as described above. The stability of

indomethacin in acidified biological matrix (pH 5.00) was investigated by adding an

excess of pH 5.00 acetate buffer to the biological fluid (plasma/urine), spiking with

a known amount of indomethacin and allowing the samples to remain on the bench-

top for 0, 6, 12, and 24 h. The internal standard, a-methyl indomethacin, was then

added and the samples processed as before. Finally, the stability of processed

samples on the G C autosampler tray was investigated by repeatedly injecting the

sample vials at 24 h intervals for ten days. Stability studies were also performed on

the indomethacin formulation used in the fetal infusion studies, at room temperature

and at 4 °C.

2.8. Animal Experiments

59

2.8.1. Animal Handling & Surgical Preparation.

Ten time dated pregnant sheep of Dorset/Suffolk breeds were used. These

pregnant ewes were kept in the animal unit at the Children's Variety Research

Center in groups of two or more in different but adjacent pens in full view of one

another. The animals were placed on a standard diet and water. Ethical approval

for these studies was obtained from the UBC Animal Care Committee and the

guidelines of the Canadian Council on Animal Care were followed.

The sheep were operated on between 124 and 125 days of gestation (term =

145 days). Atropine (6 mg) was administered as an intravenous bolus to control

salivation followed by (-15 min later) induction of anaesthesia with sodium

pentothal (1 g). The ewes were intubated with an endotracheal tube and anaesthesia

maintained using a mixture of halothane (1-2%), nitrous oxide (60%), and oxygen.

The abdominal area was shaved and the surgical area swabbed with a 10%

povidone-iodine topical solution following which the animal was draped with sterile

sheets. Following a midline abdominal incision, access to the fetus was gained

through an incision of the uterine wall free of placental cotyledons and major blood

vessels. The fetal hind limbs were extracted and polyvinyl catheters were placed in

the inferior vena cava and descending aorta through the femoral and tarsal vessels,

60

respectively. The fetal bladder was cannulated through a suprapubic incision,

followed by non-occlusive catheterisation of the common umbilical vein at the

umbilicus. Two catheters were placed in the amniotic cavity and anchored to the

abdominal skin of the fetus. A carotid artery was also catheterised, via a separate

uterine incision, and a non-occlusive catheter was inserted into the trachea, via an

incision below the larynx. A catheter was also placed in the amniotic cavity, and

anchored to the skin on the neck. The hysterotomy and laparotomy incisions were

then closed. The maternal inferior vena cava and descending aorta were cannulated

via maternal femoral vessels. All catheters were passed subcutaneously to a small

incision in the maternal abdominal wall on the left flank where they exited. They

were stored in a denim pouch attached to the flank. Postoperatively, ampicillin (500

mg) was adminsitered as an intramuscular prophylactic antibiotic for three days to

the ewe, and ampicillin (500 mg) intra-amniotically to the fetus at the time of

surgery and daily into the amniotic fluid for the duration of the preparation.

2.8.2. Recording and Monitoring Procedures

The analog signals were also processed by an on-line computerised data

acquisition system, which stored one minute averages of the variables on floppy

diskettes. Fetal arterial pressure, amniotic fluid pressure, heart rate, and fetal

tracheal pressure were continuously monitored using a polygraph recorder

(Sensormedics Inc.). Fetal arterial and venous pressures were adjusted to zero

61

reference by subtracting amniotic fluid pressure continuously using the computer.

Initial studies involved determination of urinary flow rate twice daily using an

intermittant measurement system. In later studies, fetal urine flow rate was

measured continuously by the computer by means of a bladder catheter which

drained by gravity into a sterile reservoir, the hydrostatic pressure of which was

monitored by the computer. When the preset pressure was exceeded, a calibrated

peristaltic pump was activated to pump urine from the reservoir into the amniotic

cavity via an amniotic catheter. The volume pumped each minute was stored on

diskette and provided a measurement of urine flow. Fetal arterial blood (0.7 mL)

samples were collected for measuring pH, pC02, p02, HC0 3, and base excess at

39.5 °C (Instrument Laboratory System 1306 pH/Blood Gas Analyser, Lexington,

Mass., USA). Glucose and lactate concentrations in fetal blood samples were also

determined using a YSI 2300 Glucose/Lactate Analyser. Hematocrit was estimated

using a micro-capillary centrifuge. Blood 0 2 saturation and hemoglobin

concentrations were determined using an OSM-2 Hemoximeter. Plasma protein

concentration was determined using an American Optical hand-held refractometer.

2.8.3. Determination of Organ Blood Flows

Fetal organ blood flows were determined by injecting 15+3 \im diameter

microspheres containing either 153Gd, 85Sr, 5 1Cr, 46Sc, or 9 5Nb radioisotopes

62 incorporated in their structure and suspended in 10% dextran. A different isotope

was used on each day of the 5-day protocol. Before injection, the microspheres

were mixed and placed in an ultrasonic bath for 5 min to minimize aggregation of

the spheres. Immediately after collecting the second (pm) sample of the day, 0.5

mL of the microsphere solution containing 1.2 x 106 microspheres was injected via

the fetal tarsal venous catheter and the catheter flushed with 3 mL of heparinised

saline. Lower and upper body reference blood samples via the fetal aortic and

carotid catheters were obtained during three minutes, begining about 30 sec prior to

injecting microspheres, at a 2.2 mL/min withdrawal rate using a Harvard pump.

Animals were euthanised after each experiment with pentobarbital sodium (100

mg/Kg) and the operated fetus removed and weighed. Fetal tissues and organs of

interest were obtained at autopsy, weighed and carbonised in an oven at 300°C for

3-5 days. The gamma emission generated by the radioactive microspheres was

measured from the fetal organs (kidneys, brain, adrenals, bone, liver, placenta,

amnion, chorion, gut, atria, ventricles, spleen, fetal aortic and carotid reference

samples) using a multi-channel gamma counter, isotope separation being achieved

using computer software.

2.8.4. Experiment Protocol For Fetal Infusion Studies

63

At least 24 h before each experiment, the ewe was placed in an experimental

pen in full view of companion ewes with access to standard diet and water. Each

experiment was composed of five days with the first day being the control day when

no drug was given. Indomethacin in a solution of 1.1% ethanol and 0.75% sodium

bicarbonate in normal saline was infused into the fetal vein at a dose of 0.0025 (high

dose) or 0.0006 (low dose) mg/Kg/min (2 mL/h) or vehicle alone (control) on days

2 through 4. The infusion was terminated on day 5 and the fetus also monitored

during this 24 h recovery period. Samples (2 mL) of fetal arterial blood, umbilical

venous blood, amniotic and tracheal fluids, and fetal urine were collected twice

daily at six hour intervals and stored at -20 °C pending assay. Blood samples were

collected in heparinised VacutainerR collection tubes, centrifuged at 3000 g for 10

minutes at -10 °C and the plasma transfered into PyrexR 15 mL culture glass tubes

for storage and analysis. Withdrawn fetal blood was promptly replaced with an

equal volume of heparinised drug-free maternal blood via the tarsal vein catheter.

2.9. Data Analysis

2.9.1. Pharmacokinetic Calculations

[1] Fetal total body clearance (Cltb) of indomethacin was estimated as:

Cltb = V C ss

64 where k<j was the infusion rate and C s s the apparent fetal arterial steady-state

indomethacin concentration.

[2] Placental clearance (Clp) was estimated as:

Clp — Qum X (Cf a-C u m)/Cf a

where Q u m is the umbilical blood flow rate, C f a the fetal femoral arterial

indomethacin concentration, and Cm the umbilical venous indomethacin

concentration at apparent steady state.

[3] Fetal non-placental indomethacin clearance (Clfo) was then calculated as:

Clfo = Cltb-Clp

[4] Renal clearance (Clr) was estimated by plotting the average urinary excretion

rate versus the drug concentration at the midpoint of the collection interval and

fitting the least squares straight line forced through the origin. The slope of the line

provided the renal clearance value in mL/min.

2.9.2. Statistical Calculations

65 Statistical evaluations were performed on analytical, physiological, and

pharmacokinetic parameters using established statistical tests as outlined in each

section. Data are expressed as the mean ± S.E.M. or 1 S.D as indicated. Coefficient

of variation is reported in percentage. Linear regression statistics are presented for

analytical data, and the paired Mest was used to determine whether fetal umbilical

venous and femoral arterial indomethacin concentrations were significantly

different. A significance level of P<0.05 was used.

66

3. RESULTS

3.1. Development of a capillary GC-ECD method for indomethacin.

3.1.1. Purity assessment of indomethacin

A proton NMR spectrum of indomethacin in deuterated dimethylsulfoxide

was taken at 200 Mhz (Figure 5). The spectrum was compared to a spectrum

obtained with an authentic (Merck standard) sample (O'Brien et al., 1984). The

chemical shift values of the test indomethacin sample (Table 3) compare well with

Merck standard indomethacin sample.

A differential scanning calorimetry (DSC) analysis of indomethacin is

illustrated in Figure 6. Indomethacin was observed to melt at a temperature of

162°C. The melting point of indomethacin has been reported to be 153-154 °C

(Shen et al., 1963). However, official monographs indicate the melting point to be

155 and 162 °C for two polymorphic forms of indomethacin (NF XIII), while the

BP (1980) specifies a m.p. of 158-162°C. There are two crystalline forms of

indomethacin, Form I which melts at 160-162°C, and Form II which melts at 153-

155°C, the former being more stable than the latter (O'Brien et al., 1984).

67

PRA12 HI FT HUN IN DMSO-06 WITH EXPANS R. KRISHNA

HA060S.11S AU PR06: XOO.AU

DATE 6-5-93 SF SY 01 SI TD SW HZ/PT PH RD AO RG NS TE

200.133 80.0 4390.000

32768 32768 4000.000

.244 0.0 0.0 4.096 10 32

298 FH 5000 02 4080.000 DP 63L PO LB GB CX CY F l F2

.300

. 100 25.00 15.00 U.500P -l.OOOP

HZ/CM 100.063 PPH/CH .500 SR 3290.15

' I ' 9.0 11.0 10.0 6.0 5.0

PPM

Figure 5a: NMR spectrum of indomethacin in DMSO

68

Figure 5b

69

Table 3: Chemical shift values of indomethacin and their respective group

assignment

Chemical Shift (ppm)

Indomethacin Sigma Chemical Co.

Indomethacin Merck Standard*

Assignment

2.2 (s) 2.22 (s) 2-CH3

3.62 (s) 3.67 (s) 3-CH2CO2H

3.78 (s) 3.74 (s) OCH3

6.7-6.9 (m) 6.55-7.10 (m) H4, H 6 , H 7

7.6-7.8 (m) 7.45-7.80 (m) c i - ( yi

+ Proton NMR spectrum of Indomethacin Merck Standard 6375-66-1 (Adapted

from O'Brien et al, 1984).

70

u o so

(a)

(b)

Temperature (°C) ->

Figure 6: DSC scan of indomethacin: (a) open pan, and (b) closed pan

methods

71

The LC/MS/MS spectra of indomethacin is illustrated in Figure 7. The

salient features of the spectrum are: m/z 139-141 due to the /?-chlorobenzoyl group,

and molecular ion at m/z 358.

3.1.2. Optimisation of Derivatisation

Derivatising Agents

Indomethacin was derivatised with four different derivatising reagents to

determine the most suitable agent with respect to reaction time, chemical safety of

the agent, completeness of the reaction, chromatographic behaviour of the resulting

derivative, as well as peak geometry and shape. The type of reactions selected were

(I) silylation (with BSA, MTBSTFA), and (ii) fluorination (PFBBr, PFP).

The chromatographic behaviour of the resulting derivatives following

derivatisation with BSA, MTBSTFA, and PFBBr is illustrated in Figures 8 and 9.

Derivatisation with PFP was incomplete and was, therefore, not pursued further.

The silylation reactions were much easier to perform. Peak response was best with

MTBSTFA and PFBBr. Derivatisation with PFBBr afforded compounds with high

volatility and short retention times (Figure 9). The response was slightly greater

than that obtained using MTBSTFA, however, the procedure was comparatively

more cumbersome and time consuming involving the use of a catalyst. Overall, the

72

312 357

CH 3 0

139

H

201 I 293 311

358

357J

J. 100 120 140 160 ISO ZOO 220 240 260 280 300 320 340 360 380 400 420 440 460 480 500

Figure 7: LC/MS/MS spectrum of indomethacin

73

Figure 8: GC-ECD chromatograms of indomethacin derivatised by (a) BSA,

and (b) MTBSTFA.

74

Retention time (minutes)

Figure 9: GC-ECD chromatogram of indomethacin derivatised by PFBBr.

75 reaction with MTBSTFA was quick, safe, complete and produced peaks of superior

geometrical shape and integrity.

Reaction Conditions

The effect of temperature on the derivatisation reaction was examined, from

room temperature (27°C) to 90°C and is shown in Figure 10. Maximum variability

was noted with lower temperatures (27°C, 45°C). Area counts were maximal at

60°C and did not increase following any further increase in temperature, and thus

this was chosen as the optimal temperature for derivatisation.

Different reaction times were also examined, from 10 min to 60 min, at a

constant temperature of 60°C (Figure 11). Mean area counts increased as

derivatisation time was prolonged from 10 min to an hour, with a plateau appearing

at 50 minutes. Keeping sample preparation time in mind as well as noting the least

variability at 50 minutes, the reaction time was set at 50 min.

Various concentrations of MTBSTFA were also examined, i.e., from 2% to

100% solution, the solvent being toluene (from 2 \\L in 100 |iL toluene to 100 |iL

MTBSTFA). The area counts did not differ significantly over the volume range,

however, an effect on peak height was observed. The optimal volume was

76

60

20 40 60 80 100

Temperature of Derivatisation (degree C)

Figure 10: Determination of optimum derivatising temperature for

indomethacin with MTBSTFA (data are mean values ± 1 S.D., n=4).

77

Reaction Time (Minutes)

Figure 11: Determination of optimum reaction time between indomethacin

and MTBSTFA (data are mean values ± 1 S.D., n=4).

78

determined as 8 fiL MTBSTFA in 100 |0,L toluene (8%), since this afforded the

greatest peak response in terms of peak height.

3.1.3. Selection of Internal Standards

Ideally, the internal standard should possess close structural resemblance to

the analyte, have similar chromatographic behaviour, and elute in a region free from

endogenous interference. Several compounds were screened, viz., sulindac, 5-

fluoro indole 2 carboxylic acid, 5-fluoro dl tryptophan, 5-fluoro indole 3 acetic acid,

zomepirac, acemetacin, carprofen, 5-fluoroindomethacin, and oc-methyl

indomethacin (Figure 12).

The indole analogs (Figure 12) are chemicals of fairly low molecular weight

and did not appear on the chromatogram when derivatised with MTBSTFA. When

these same compounds were derivatised with PFBBr, retention times of the

resulting derivatives were very short with interfering bridged peaks as shown in

Figure 13a. The indomethacin analogs, namely, 5-Fluoroindomethacin and a-

methyl indomethacin produced excellent peaks with MTBSTFA as shown in Figure

13b. Sulindac produced a satisfactory peak when derivatised with either MTBSTFA

or PFBBr with good retention times as well as pilot calibration curves yielding an

r>0.999. However, an unidentified peak of significant magnitude eluted with

79

CH2C00H

5-Fluoro Indole 3-acetic acid

1 COOH

5-Fluoro Indole 2 Carboxylic Acid

• CH2-CH-COOH I

NH2

5-Fluoro DL Tryptophan

Carprofen

C H 3

< N

CH2COOH

\ C H 3

C=0

[6 CL

Zomepirac

•CH2COOH

I

C=0

CL

5-Fluoroindomethacin

CH3O

9H3 CH-COOH

I c = o

CL

Alpha-Methyl Indomethacin

CH

SO I

C H 3

Sulindac

Figure 12: Chemical structures of tested internal standards

80

Blank chromatogram

Vi tt © a Vi

<u 04 u o

u

Q

?2 H

*0 L v ^l

10

10H

72

J 40-

c O-

o 4-> C L >>

s_

o s-o 3

Retention Time (Minutes)

10

Figure 13: (A) Derivatisation of 5-fluoro DL tryptophan with PFBBr for use

as a potential internal standard for indomethacin.

81

Figure 13: (B) GC-ECD chromatograms of indomethacin analogs, 5-

fluoroindomethacin, and a-methyl indomethacin, derivatised with MTBSTFA.

82

sulindac. Association of this peak with sulindac was confirmed by injecting

indomethacin alone, indomethacin and sulindac, sulindac alone, and blank plasma

(Figure 14). Oven temperature programming changes did not affect the presence of

peak "X", nor did recrystallising sulindac help. Figure 15 illustrates the DSC scan

of sulindac and its recrystallised form. It is speculated that sulindac may be oxidised

to a sulphone in the injection port, and this might have eluted as peak 'X' on the

chromatogram. Because of this additional peak we were unable to use this

compound as an internal standard for this assay.

Zomepirac is closely related to indomethacin and provided a fairly good

response with MTBSTFA (Figure 16). However, an extraneous peak in

experimental fetal plasma samples co-eluted with it, precluding the use of this

compound as the internal standard. This interference was not observed in fetal or

maternal plasma samples stored beyond two years (blank samples), amniotic and

tracheal fluids, or in fetal or maternal urine samples. Following analysis of samples

collected from the present infusion studies, however, the interference was detected,

leading to a re-examination of chromatographic conditions (oven temperature

programming, column stationary phase [polar/nonpolar], column diameter

[wide/narrow bore]), use of PFBBr as a potential derivatising agent, as well as

screening of additional internal standards.

83

Figure 14: Illustration of a decomposition peak "X" associated with sulindac.

Chromatograms indicate: (1) indomethacin and sulindac, (2) sulindac only,

and (3) blank.

84

Sulindac

recrystallised sulindai

(open pan)

! ; . ;L I. recrystallised sulindac

*P~^ (closed pan)

Figure 15: DSC scan of sulindac and recrystallised sulindac.

85

3 . O <= <A

•SI

o CE 00

1 .

N

Q Indo.

Blank

RETENTION TIME IN MINUTES

Figure 16: Chromatogram of indomethacin and zomepirac spiked in fetal

plasma.

86

a-methyl indomethacin (Figure 17), a generous gift of Merck Frosst Inc,

provided all the qualities required for an ideal internal standard, i.e., it was a very

close structural analog of indomethacin, eluted close to the analyte in a region free

from endogenous interference (Figure 13B), and produced a response factor of

-1.00 at a concentration equal to that of indomethacin. On this basis it was chosen

as the internal standard for all further studies.

3.1.4. Optimisation of G C conditions

In order to obtain maximal sensitivity, the splitless mode of injection

(suitable for trace level analyses) was used throughout. Injection port temperature is

another critical parameter since it influences greatly the thermal stability of the

derivatised compound as well as the degree of drug transfer onto the head of the

column. Hence, the effect of various temperatures (140-300 °C) on indomethacin's

response were examined. This evaluation was conducted using a low (20 ng/mL)

and a high (100 ng/mL) sample concentration of indomethacin. As illustrated in

Figure 18, maximal absolute area counts were obtained at injection port

temperatures in the range 180-200 °C. A drastic decline in area counts was

observed when the temperature was increased from 200-300 °C. The profile was

similar at both concentrations likely indicating thermal decomposition at higher

87

C H 3 0 ^ ^ C H 2 C O O H

B

c=o

Indomethacin

CI

C H 3

^ C H - C O O H

a-methyl indomethacin (Internal Standard)

Figure 17: Chemical structures of indomethacin and a-methyl indomethacin.

88

1500

0) 375 h

100 175 250 325 400

Injection Port Temperature (degree C)

Figure 18A: Effect of injection port temperature on the peak area counts of

(A) 20 ng/mL, and (B) 100 ng/mL indomethacin concentrations (n=4).

89

8000

Injection Port Temperature (degree C)

Figure 18B

90

temperatures. While a temperature of 180 °C resulted in the highest response the

chromatogram contained many peaks from endogenous compounds. The

chromatograms, however, were much cleaner at 200 °C and therefore, this

temperature was selected for use in the assay. Figure 19 illustrates the effect of

various initial column temperatures on the retention time of indomethacin at a

constant programming rate of 8 °C/min. An initial column temperature of 210 °C

and a final temperature of 300 °C with a programming rate of 40 °C/min provided

optimal resolution, peak geometry, run times, and response and were therefore,

chosen for the final assay.

Various column types were also investigated, viz., an Ultra-2 wide bore non-

polar column [stationary phase: 5% phenylmethylsilicone], Ultra-2 narrow bore

column, and a DB-1701 polar column [stationary phase: methyl (7% phenyl, 7%

cyanopropyl) silicone]. Use of the DB 1701 column was observed to be more

appropriate with PFBBr derivatised samples, and did not appear to be useful for

MTBSTFA derivatised samples as illustrated in Figure 21. Representative

chromatograms obtained for the narrow bore Ultra-2 column are illustrated in

Figure 20 and for the wide bore column in Figures 13B and 16. While both Ultra-2

non-polar columns provided rapid run times and excellent chromatography, the

wide-bore column allows for a greater volume of sample to be injected (Figures

13B, 16) and was therefore, selected for use in the assay.

91

Figure 19: Effect of various initial column temperatures on peak retention

times.

92

Figure 20: GC-ECD chromatograms of indomethacin and zomepirac using a

narrow bore fused silica Ultra-2 column.

Figure 21: GC-ECD chromatogram of indomethacin and zomepirac using

narrow bore DB-1701 column.

94

The time delay following injection for inlet purging to occur is critical in

determining the degree of sample transfer onto the column and, therefore, needs to

be optimised. To examine this, purge delay times ranging from 10-120 seconds

were examined (Figure 22). A plateau was reached at 40-120 seconds. Therefore, a

purge delay time of 0.75 min was chosen.

Various detector temperatures were also examined, ranging from 300-360 °C

(Figure 23). Detector temperature had no effect on overall indomethacin peak area

counts over the range studied so 330 °C was selected, since at lower temperatures

the ability to maintain an uncontaminated detector was in question. There was no

significant effect of make-up gas (argon-methane; 95/5) flow rate on the detector

response, therefore a flow rate of 65 mL/min was used.

Various column head pressures (5-15 psi) were also investigated (Figures 24

and 25). Maximal peak area counts and area ratios were noted at a column head

pressure of 10 psi. Furthermore, as column head pressure was increased from 5 to

15 psi, there was a proportionate fall in retention time and peak width. Hence, a

column head pressure of 10 psi was chosen since this afforded maximum peak area

counts and intermediate peak width and peak retention time.

95

Figure 22: Effect of purge delay times on the peak area counts of indomethacin

tBDMS derivative (n=4).

96

5000

4000

3000 h

2000

1000 h

250 300 350 400

Detector Temperature (degree C)

Figure 23: Effect of various detector temperatures on the peak area counts of

indomethacin tBDMS derivative (n=4).

Figure 24A: Determination of optimum column head pressure: Effects on (A)

peak indomethacin area counts and (B) peak area ratio.

Figure 24B

99

Figure 25: Determination of optimal column head pressure with reference to

peak retention time (A), and peak width (B).

100'

1.00

0.80 h

0.20 h

0 4 8 12 16 20

Column Head Pressure (PSI)

Figure 25B

101

3.1.5. Optimisation of drug extraction

The extraction efficiency of various organic solvents was examined by

comparing the ratio of extracted samples/control samples following extraction from

a model biological fluid matrix (tracheal fluid) using acetate buffer pH 5.00 (Figure

26 and Table 4). Of the solvents examined, only ethyl acetate and dichloromethane

were able to extract 1 ng indomethacin effectively. Solvent mixtures were also

examined, namely, (a) toluene-TEA (0.0125M), (b) hexane-IPA (10%), and (c)

hexane-ethyl acetate (1:4, 2:3, 3:2, 4:1). As expected, the addition of triethylamine

(TEA) to toluene reduced extraction efficiency drastically. This is contrast to the

fact that addition of TEA to toluene markedly improves the recovery of basic

compounds (Wright, 1992). While the addition of isopropyl alcohol (IPA, 10%) to

hexane improved the overall extractability of indomethacin recoveries were still not

satisfactory with this solvent system. The addition of hexane to ethyl acetate was

observed to reduce the recovery of indomethacin (Table 4). For ethyl acetate (EA),

extraction efficiency improved with increasing volume (Figure 27), but increasing

solvent volume had no effect with dichloromethane. The recoveries following

extraction with 2 mL EA, 4 mL EA, and 5 mL EA, were 53.66 ± 4.32%, 73.01 ±

2.76%, and 84.59 ± 5.63%, respectively (Table 4). Based on these observations,

ethyl acetate, in a volume of 5 mL, was selected for use in the assay since maximal

recovery and extraction efficiency were noted at these conditions.

102

Figure 26: Extraction efficiency of various organic solvents (n=4) using

tracheal fluid as the model biological fluid at an indomethacin concentration of

12 ng/mL.

103

Helleberg (1976), and Ou and Frawley (1984) suggested that attainment and

maintenance of an aqueous phase pH of 5.00 was essential for optimal recovery of

indomethacin. To examine this, a series of acetate buffers were prepared in the pH

range 2-6 (BP, 1988). As shown in Figure 28, attainment of pH 5.00 was essential

for optimal extraction; a lower pH reduced extraction efficiency considerably. This

pH aggrees well with the pKa of 4.5-4.7 for indomethacin.

The effect of extraction times ranging from 5-60 min on indomethacin

response were also examined (Figure 29). The highest variability was observed at 5

min and the lowest at 20 min. Maximal area counts were also observed at 20 min

and hence an extraction time of 20 minutes was chosen.

104

Table 4: Extraction efficiency (%) of organic solvents used to extract

indomethacin at different concentrations using tracheal fluid as the

model biological fluid.

Solvent Volume Indomethacin Concentration [ng/mL] [mL] 1 6 12

Ethyl Acetate 2 58.43 52.56 50.00 Toluene 2 - 52.56 55.25 Hexane 2 - 20.26 23.57

Ethyl Acetate 4 75.90 70.38 72.75 Diethyl Ether 2 - 49.66 39.05

Toluene-TEA* 2 - 16.90 13.30 Hexane-IPA** 2 - 36.25 35.66

Dichloromethane 2 80.72 55.79 61.11 Ethyl Acetate 5 91.04 80.60 82.13

Dichloromethane 4 - 55.94 61.58 * Triethylamine (0:0125 M) ** Isopropyl alcohol (10%)

105

2 4 5

Volume of Ethyl Acetate (mL)

Figure 27: Effect of solvent volume on the extraction efficiency of ethyl

acetate.

106

pH of Extraction Medium

Figure 28: Determination of optimum pH on the extraction of indomethacin

(n=4).

107

Figure 29: Determination of optimum extraction time for indomethacin from

tracheal fluid matrix using ethyl acetate at pH 5.00 (data are mean ± 1 S.D.,

n=4).

108

3.1.6. Gas chromatography-Mass spectrometry (GC-MS)

GC-MS was performed on the tBDMS, TMS, and PFB derivatives of

indomethacin in the EI, NCI, and the PCI modes (Figures 30-32) to confirm

chemical structure. The GC-MS of indomethacin-TMS has been previously

reported (Plazonnet and Vandenheuvel, 1977), but no data exits for indomethacin-

tBDMS.

Following derivatisation of indomethacin with MTBSTFA and PFBBr, both

GC-ECD and GC-MS demonstrate the formation of a single species as evidenced by

a single, symmetrical sharp peak in the respective GC-ECD and total ion

chromatogram (GC-MS), Figures 8 and 9. Derivatisation with BSA produced a

single peak, however, the peak geometry was poor compared to the tBDMS and

PFBBr derivatives (Figure 8).

The tBDMS derivative of indomethacin was subjected to detailed GC-MS

analysis to obtain the EI, NCI, and PCI spectrums (Figure 30). Following EI, the

most intense peak occurs at m/z 139 due to the p-chlorobenzoyl fragment. The

characteristic [M-57]+ peak at m/z 414 represents the loss of the C(CH 3) 3+ fragment.

The EI scan also shows a molecular ion peak at m/z 471. The spectrum following

chemical ionisation shows the molecular ion peak at m/z 471 (NCI, M+), and m/z

472 (PCI, MFT).

109

Abundance

1 B0OOO0 -|

]

1139

1

312

CH 30.

CH3

^ j S ^ _ _ ^ - C H ^ C O O — Si --(j; — C H 3

W^~-NT^-CH3 C H 3 C H 3

139 c=o

CH3

111 9 CI

370 /

312

216 275 \ 1_ L

333

414 /

L i —r 100 200 300

[— 400

MassVCharge

Figure 30A: GC-MS spectra of the indomethacin tBDMS derivative in the EI

(A), NCI (B), and PCI (C) modes.

110

ce 357

C H 3 C H 3

, M 0 0 <H C H 3 0 ~ ^ = ^ _ _ ^ C H 2 C O O - S i — C — C H 3

^ ^ - N T ^ - C H s C H 3 C H 3

I

c=o

100000-! a

20000-

357 J12 81 192 231 299 332 /

l 1 1 1 1 1 1 1 1 1 1 1 1 r- -i 1 1 ,— 100 200 300 400

Mass/Charge

Figure 30B

I l l

456

139

CH3O

XXX CH 3

312

• C H j C O O - S i - -

139 + c=o

6 CI

C H 3

C H 3 |

r CH3I

C H 3

357

M + H: 472

174

312

207

-I**1—<-

231

/ 300

\ L A 360

300

Mass/Charge

400 500

Figure 30C

112

The GC-MS of indomethacin-TMS derivative (Figure 31) is similar to that of

the tBDMS derivative. There is extensive fragmentation under EI conditions, and

the base peak at m/z 139 is again due to the /?-chlorobenzoyl fragment. The

molecular ion was observed, with significant abundance, at m/z 429. The spectrum

following chemical ionisation shows a molecular ion peak at m/z 429 (NCI, M+) and

m/z 430 (PCI, MFT).

In EI-MS, indomethacin-PFB ester produces a base peak of m/z 139

corresponding to the p-chlorobenzoyl fragment. Pentafluorobenzyl esters undergo

dissociative electron capture under NCI conditions as a part of the electron-

molecule interaction. The characteristic [M-l81]" is observed at m/z 356 which is

the base peak in the NCI spectrum (Figure 32). The molecular ion peak under PCI

conditions(MH+, PCI) is observed at m/z 538. A mass spectrometric assay operated

in the negative-ion chemical ionisation mode would appear to provide the best

sensitivity for indomethacin derivatised by PFBBr.

113

3.2. Validation of the GC-ECD method

3.2.1. GC Operating Conditions

The optimised operating conditions for routine analyses were: column,

Ultra-2 fused silica capillary column cross-linked with 5% phenylmethylsilicone (25

m x 0.31 mm x 0.52 \im film thickness); injection mode, splitless; injection port

temperature, 200 °C; initial column temperature, 210 °C (1 min); oven

programming rate, 40 °C/min; final oven temperature, 300 °C (6 min); detector

temperature (ECD), 330 °C; carrier gas, ultra-high purity hydrogen at a column

head pressure of 70 kPa (column flow rate, ~1 mL/min); purge delay time, 0.75

min; and detector make-up gas, argon-methane (95/5, 65 mL/min).

114

EI

141

/

73

• A . . I 327

/

370

— 200

i — i — 300 Masi/Charge

NCI

i 100

111 148

— i 1 (-218 265

299

200 300

Mi i l /Chi fgt

443

/ —I—

500

Figure 31A: GC-MS spectra of the indomethacin-TMS derivative.

115

Abundance

Average of 15.207 to 15.437 min. from USepOlOlOOt d ( _ K H , I K / * - T M S )

PCI

B0O0- |

iU.ii.-ii

207 223

200

312

300

Mass/Charge

340

J r - A -

370

Figure 31B

A b u n d a n c e

IGOOOO- EI

75 181 207 398

/ — r

400 300

Miss/Charge

—I— 500

Abundance Average of213fi9 to 21.810 min. from lxSep02010O2.d £_PC', l-do-?^)

279

\ 400

PCI

538

\

300

Mass/Charge

Figure 32A: GC-MS spectra of the indomethacin-PFB derivative.

117

Avenge of21 J2S lo 21.741 min. from lxSepO2010O4.d(k*:/ -real

111 148 246 322

->—i—'—r 300

Mass/Charge

•V-384

NCI

493

479/

—I— 200 500

Figure 32B

118

3.2.2. Drug extraction from biological fluids

The extraction and derivatisation procedure for indomethacin is summarised

in Scheme 1. Figures 33-36 illustrate the representative chromatograms of extracts

of biological fluids from the chronically instrumented ovine fetus. Both the tBDMS

derivatives of indomethacin and a-methyl indomethacin elute in regions free from

endogenous interference (Figures 33-35). The retention time of the internal

standard a-methyl indomethacin (Figure 33, peak a) is -4.8 min and that of

indomethacin (Figure 33, peak b) is -5.0 min. Figure 36 is a representative

chromatogram of indomethacin and zomepirac (the retention time of zomepirac

being -3.5 min and that of indomethacin -5.1 min) spiked in amniotic fluid

obtained during initial testing of zomepirac as a possible internal standard.

119

w c o a. w

o OJ O Q

4 - . O e - a - J

3 . O <= 4 - 1

1 . O .

a 4 , e s

Retention Time [minutes]

Figure 33: Representative chromatograms of the tBDMS derivative of

indomethacin and alpha-methyl indomethacin spiked in fetal plasma (the inset

shows the magnified baseline). The superimposed blank indicates no

interference from endogenous components.

Figure 34: Representative chromatograms of the tBDMS derivatives of

indomethacin and a-methyl indomethacin spiked in fetal urine.

121

C o a <M <U L.

U © w

Q

. O e 4-;

3 . O eS—

2.0e4H

1 . O <s 4-!

S .Oe4i

I i

U CQ JS

E o •a >>

S •

i 1

OS 5 s o -a s

Blank

Retention time (minutes)

Figure 35: Representative chromatograms of the tBDMS derivative of

indomethacin and a-methyl indomethacin spiked in amniotic fluid along with

the superimposed control.

122

1 . 4 e 5 H

1 : 2 e S J

1 .Oe£H

C 8.0e4-J o a

£ 6 . 0 e 4 j u 4)

4.0e4H

2 . 0 e 4 H

es a E o N

"3 CQ

S o •a fi

.... jv...i£i!Ai~. _.._JV.

Blank

Retention Time (Minutes)

Figure 36: Representative chromatograms of indomethacin and zomepirac

spiked in amniotic fluid.

123

3.2.3. Recovery studies

The extraction recoveries for plasma and urine of the fetal lamb, determined

by the matrix-compensation method, are presented in Table 5a. Mean recoveries

from plasma and urine were 96.99 ± 9.25% and 94.74 ± 6.75% respectively. This is

greater than the solid-phase extraction procedure of Nishioka et al., 1990 (about

87%), and the liquid-liquid extraction procedure, with ethylene dichloride, of

Sibeon et al., 1978 (about 85%), reported for plasma. However, recoveries

comparable to the present study have also been reported: viz 95% from serum using

dichloromethane (Jensen, 1978), and ~ 92% from plasma with ethyl acetate (Evans,

1980). In amniotic fluid, the recovery of indomethacin averaged 78.42 ± 2.57%

(Table 5b).

3.2.4. Stability studies

Indomethacin was found to be stable for at least four freeze/thaw cycles, at

least 24 h on bench-top, at least 24 h in acidified plasma and urine, and at least 1

week on the autosampler tray (Table 6). Indomethacin is also stable in plasma and

urine samples stored in the freezer following sampling in the time periods examined

to date (14 months). Indomethacin is also stable in the formulation prepared for the

infusion studies (pH 7.8) for at least 96 h as illustrated in Figure 37.

124

Table 5a: Extraction recoveries (n=4), inter-day variability (n=4) and intra-day variability (n=3) for indomethacin in (A) fetal plasma, and (B) fetal urine,

using cc-methyl indomethacin as the internal standard.

(A) Fetal Plasma

Concentration Inter-day Intra-day Recovery % (ng/mL) variability variability (S.D.2)

% C V 1 % C V 2 7.54 4.10 101.77

(2.05) 8 1.49 3.66 104.82

(3.24) 16 5.44 1.94 97.52

(3.33) 32 4.18 1.03 83.85

(1.29) slope 4.00 1.13 -

1 Coefficient of Variation; 2 One Standard Deviation

(B) Fetal Urine

Concentration Inter-day Intra-day Recovery % (ng/mL) variability variability (S.D.2)

% C V 1 % C V 2 9.05 11.14 100.71

(9.41) 8 7.75 1.09 100.35

(12.51) 16 7.01 0.37 87.83

(3.09) 32 3.92 3.56 90.07

(3.44) slope 3.11 4.11 -

1 Coefficient of Variation; 2 One Standard Deviation

125

Table 5b: Inter- and intra-day variability parameters for indomethacin in amniotic fluid of fetal sheep, with zomepirac as the internal standard.

Intra-day variability

Cone. ng/mL

Calibration curve 1

Calibration curve 2

Calibration curve 3

Mean % c . v .

1 0.0172 0.0169 0.0165 0.01686 2.08 2 0.0315 0.0334 0.0330 0.03263 3.09 . 8 0.125 0.1228 0.1250 0.1242 1.02 16 0.2238 0.2139 0.2223 0.2200 2.42 32 0.4324 0.4168 0.4247 0.4247 1.84 r 0.999 0.999 0.999 0.999 -Intercept 0.0086 0.0099 0.0100 0.0095 8.22 Slope 0.0133 0.0127 0.01307 0.01304 2.04

Inter-day variability

Cone. ng/mL

D A Y 1 DAY 2 DAY 3 DAY 4 D A Y 5 % C.V.

1 0.0174 0.0184 0.0207 0.0164 0.0164 10.02 2 0.0321 0.0333 0.0324 0.0323 0.0297 4.18 8 - 0.1120 0.1157 0.1167 0.1057 4.43 16 0.1998 0.2019 0.2065 0.2081 0.1836 4.87 32 0.3778 0.4100 0.3751 0.3995 0.3280 8.36 r 0.999 0.999 0.998 0.999 0.997 -slope 0.0116 0.0125 0.0115 0.0123 0.0100 8.48

126

Table 5b (Continued): Extraction recoveries of indomethacin spiked in amniotic fluid using zomepirac as the internal standard.

Concentration (ng/mL) % Recovery 2 80.24 8 80.87 16 75.34 32 77.42 Mean 78.42 Standard Deviation 2.57

127

Table 6: Stability of indomethacin in acidified sample matrix, during freeze-thaw cycles (with a-methyl indomethacin as the I.S.), and on the G C

autosampler tray (with zomepirac as the I.S.).

(A) Acidified biological sample (urine) matrix at pH 5.0 (n=2)

Time -> Area Ratio Mean

(% C V . ) Time -> Control 6 hours 12 hours 24 hours Mean (% C V . )

8 ng/mL 0.2187 0.2371 0.2468 0.2183 0.2302 (6.13)

(B) Freeze-thaw (F-T) cycle stability of indomethacin in plasma (n=2)

Cycles Control 1st F-T 2nd F-T 3rd F-T 4th F-T Mean (% C V . )

8 ng/mL 0.2307 0.2774 0.2491 0.2742 0.2618 0.2586 (7.43)

(C) GC autosampler tray stability of indomethacin (n=6)

Cone. ng/mL

A R E A RATIO Mean % C V .

Cone. ng/mL t=0h t=24h t=60h t=6d t=10d Mean % C V . 1 0.0114 0.0146 0.0153 0.0163 0.0129 0.0140 14.53 2 0.0230 0.030 0.0289 0.0304 0.0250 0.0274 11.94 4 0.044 0.051 0.0541 0.0567 0.0478 0.0507 9.92 6 0.076 0.079 0.0781 0.0803 0.0774 0.0781 2.09 8 0.0902 0.1065 0.1013 0.1006 0.0929 0.0983 6.75 10 0.119 0.134 0.1248 0.1228 0.1214 0.1244 4.64 20 0.249 0.2694 0.2402 0.2250 0.2240 0.2415 7.78 30 0.313 0.3607 0.3389 0.3061 0.3044 0.3246 7.53 40 0.485 0.5529 0.4769 0.3833 0.4392 0.4674 13.35 r 0.999 0.996 0.999 0.997 0.998 - -slope 0.0122 0.0132 0.0115 0.0099 0.0105 0.0115 11.26

128

+ — room temp 4 Deg. C

1000

100 t

100

Figure 37: Stability study of indomethacin infusate formulation at room

temperature, and at standard refridgeration conditions.

129

3.2.5. Inter-day and intra-day variability

Inter-day and intra-day variabilities (as % C.V.) for plasma were less than

8% (less than 4% for slope), and 5% (less than 1.5% for slope) at all concentration

points studied, respectively (Table 5a). For urine (Table 5a), the values were below

10% and 11% for inter- and intra-day variability, respectively. In amniotic fluid,

respective CV's for intra- and inter-day variability were < 4% and 11% (Table 5b).

The variability parameters for the LOQ are presented in Table 7: In urine and

plasma, respective CV's for inter- and intra-day variability parameters were below

10% and 20%, respectively, for both fluids.

3.2.6. Calibration Curves

The calibration curve parameters of indomethacin from plasma, urine, and

amniotic fluid of fetal sheep are presented in Tables 8-10 and Figures 38-40.

Working calibration curves of 1-32 ng/mL were linear with coefficients of

determination (r2) > 0.99 [linear regression statistics were: Y=0.05042X + 0.00543

for plasma, Y=0.03574X + 0.05947 for urine and Y=0.02637X + 0.03694 for

amniotic fluid]. The coefficient of variation at each concentration point as well as

for the slope, were below 11% for all three fluids. The limit of quantitation in both

fluids was 1 ng/mL with a signal-to-noise ratio of >10.

130

Table 7: Variability parameters for the limit of quantitation (LOQ) of 1 ng/mL

(A) Inter-day variability (n=10)

Biological Fluid*

Day 1 Day 2 Day 3 Mean % C . V . #

Urine 0.03260 (10.435%)

0.03415 (7.58%)

0.03338 (7.99%)

0.03337 2.32%

Plasma 0.05732 (13.269%)

0.08356 (15.51%)

0.07175 (11.744%)

0.07087 18.54%

* blank fluid samples are from the fetal lamb. NB: Numbers in paranthesis indicate the coefficient of variation in %. # Coefficient of variation for days 1, 2, and 3.

(B) Intra-day variability (n=3)

Biological Fluid*

Set l Set 2 Set 3 Mean % c . v . #

Urine 0.03486 (15.47%)

0.03160 (1.574%)

0.03048 (6.137%)

0.03231 7.04%

Plasma 0.05928 (5.081%)

0.05991 (14.41%)

0.05739 (14.291%)

0.05886 2.23%

* blank fluid samples are from the fetal lamb. NB: Numbers in paranthesis indicate the coefficient of variation in %. # Coefficient of variation for sets 1,2, and 3.

131

Table 8: Calibration curve data for indomethacin in fetal plasma (n=4).

Concentration Area Ratio ± S.D.1 % C V . 2

(ng/ml)

1 0.0579 ± 0.0048 8.41

2 0.1043 ± 0.0088 8.45

10 0.5087 ± 0.0342 6.74

16 0.7831 + 0.0682 8.71

32 1.6247 ± 0.0864 5.32

[r2=0.999, int.=0.00543, slope=0.05042] 1: values are mean ± one standard deviation 2: coefficient of variation

132

Table 9: Calibration curve data for indomethacin in fetal urine (n=4).

Concentration Area Ratio ± S.D1. % c . v 2 .

(ng/ml)

1 0.0672 ± 0.0004 0.69

2 0.0974 ± 0.0029 3.01

8 0.3882 ± 0.0347 8.95

16 0.6877 ± 0.0252 3.67

32 1.1683 ± 0.0446 3.82

[r2=0.990, int.=0.05947, slope=0.03574] 1: values are mean ± one standard deviation 2: coefficient of variation

133

Table 10: Calibration curve data for indomethacin in amniotic fluid (n=8).

Concentration Area Ratio ± S.D.1 % C V . 2

(ng/mL)

1 0.05429 ±0.00338 6.24

2 0.08985 ± 0.00945 10.52

8 0.24974 ± 0.00456 1.82

16 0.47369 ±0.01367 2.88

32 0.87350 ±0.01983 2.27

1 values are mean ± one standard deviation 2 Coefficient of variation [r = 0.9996, r 2 = 0.999, int. = 0.03694, slope = 0.02637]

134

CO

rr CO <D

3.00

2.40

1.80

1.20

0.60 h

0.00 10 20 30 40

Concentration (ng/mL)

Figure 38: Representative calibration curve for indomethacin in fetal plasma.

135

Figure 39: Representative calibration curve of indomethacin in fetal urine.

136

0.00 ' 1

0 12 24 36

Amount of Indomethacin spiked (ng/mL)

Figure 40: Representative calibration curve of indomethacin in amniotic fluid.

i

137

3.3. Fetal Infusion Studies

Table 11A lists the ewes on which experiments were performed, samples

collected, and the dose of indomethacin administered. In two animals (E 1137 and E

1115), the dose of indomethacin infused was one-third of that normally administered

(i.e., low dose ewes). In all of the animals studied, hemodynamic and metabolic

parameters were within normal limits at the start of experimentation (Figures 41 A-F).

Indomethacin was observed to induce lactic acidosis in the high dose group of fetuses

(E 1221, E 2160), as evidenced by markedly high blood lactate levels (Figure 4ID),

and a reduction in pH (Figure 41E) and base excess (Figure 41F), while blood glucuose

levels remained constant (Figure 41C). This was not seen in low dose group (E 1137,

E 1115).

3.3.1. Analysis of Plasma Samples

The concentration of indomethacin was determined in FA and UV plasma

samples (Table 12) obtained following administration of indomethacin in seven

chronically instrumented ovine fetuses and the mean profiles are illustrated in Figures

42 and 43. Table 11B summarises the attainment of steady state concentrations, i.e.,

time achieved, samples used, the average concentrations and the variability. The

indomethacin concentrations in the fetus, in the high dose group, are similar to those

observed during human tocolysis (Moise et al., 1990) with mean values ranging

138

Table 11 A: Animal experiment sampling and dosing protocol.

Ewe No. Samples Collected Dose of Indomethacin (mg/min)

E796 UV, FA, MA, UTV, 0.0075 AMN, TR, UR

E984Y UV, FA, UR, AMN, TR 0.0075 E2160 UV, FA, UR, AMN, TR 0.0075 E 1221 UV, FA, UR, AMN, TR 0.0075 E 1211 UV, FA, UR, AMN, TR 0.0075 E 1137 UV, FA, UR, AMN, TR 0.0019 E 1115 UV, FA, UR, AMN, TR 0.0019 E562V UV, FA, UR, AMN, TR Vehicle only (Control) E2182 UV, FA, UR, AMN, TR Vehicle only (Control) E2221 UV, FA, UR, AMN, TR Vehicle only (Control)

Abbreviation: UV, umbilical venous; FA, fetal femoral arterial; UR, fetal urine; MA, maternal arterial; AMN, amniotic fluid; TR, tracheal fluid; UTV, uterine vein.

Table 11B: Determination of apparent steady state plasma concentration.

Ewe No. Sample Duration ofSS1

attained (h)

Mean2

(ng/mL) S.D.3 % C.V4.

E796 FA 5 23-69 195.07 16.83 8.63 E984Y FA 5-69 113.12 8.40 7.43 E2160 FA 5-69 222.25 13.05 5.87 E 1221 FA 23-69 161.69 16.58 10.25 E 1211 FA 23-69 216.21 45.02 20.82 E1137 6 FA 5-69 39.07 5.69 14.58 E1115 7 FA 5-69 43.51 7.93 18.22

1: steady state 2: mean steady state plasma indomethacin concentration 3: standard deviation 4: coefficient of variation 5: fetal femoral arterial plasma sample 6, 7: fetuses of the low dose group

CM O CL

CD c <

100

80

60

40

20

- A •

E 1137

E984

A E1115

E796

E2160

E 1211

E1221 1 3 9

-40 -20 20 40 60 80

Time (Hours)

CM o o Q.

0) r <

- + — E 1137 A E 1115 - • - E2160 • E 1221

— A — E984 E796 E 1211

100

80

60 A = A — — •

•

— f JL M-qp

• •< ... 40 t /

+ 20 _

i i

-40 -20 20 40 60 80

Time (Hours)

Figure 41A: Effect of indomethacin on fetal arterial blood gas parameters.

+ — E 1137

A - E 984

140 A E 1115

E796

E2160

E 1211

E 1221

4.00

3.50

3.00

2.50

2.00

1.50

1.00

0.50

0.00

I

/ N

+ *: A

-40 -20 0 20 40 60 80

Time (Hours)

• + - E1137 A E 1115 - • - • E2160 • E1221

A - E984 E796 E 1211

25

20 h

15

10

5 h

-40 -20 20

Time (Hours)

40 60 80

Figure 41B: Effect of indomethacin on fetal arterial glucose and lactate concentrations.

141 + •

• •

E 1137

E984

E 1115

E796

• - E2160

E 1211

E 1221

- + - E1137 * E 1115 E2160 • E1221

~ A - E 984 E 796 E 1211

10

6

2

-2

-6

-10

-14

-18

-22

-26

-30

- i * * * * J * . . 4 •

l . * Vv . \ \ \ \ •

\ A \ * \

\ \

\ \ •

I I I

-40 -20 20 40 60 80

Time (Hours)

Figure 41C: Effect of indomethacin on fetal arterial pH and base excess.

142

3 O

a

.a s s c

CQ S W3 II « s

fa a>

•2 «

CQ - s CQ

O

« CQ

.4>

fi S3

- ' C

.S 3

CQ w

S e 5 «• fi -a

CQ O

§. £ fi eo

c o CQ

CQ

£ Vi CQ

fi o at ^ e & o u c CQ

< N 1-1

CQ

H

Q

fa U "5k fi O

fa

"5k

< fa

N O N O

vol co i n N O

C O oo

00 I

N O

d CN co

N O C N

N O

in o C N

C N in I C M

O N C O

»n C N

C N

N O I co co

O N

C N C N

oo oo

C O

C N

1—I N O

d

N O in in o

oo O N N O

N O C O

| C N

C N O O

wo

co C O

o in oo N O

in in

N O C N

in

C O C N

C O O N

C N

O O

O N

d C N

O O C O

O N

C N

O N O N I in C O

C O

O O C O

CN N O I C O

N O N O

in in

r--N O

O N O N

O O I> N O co C O

C O

O N

O N

00 o C N

C N O

N O C O

co O N oo

oo

1—1 CN

fa t/5

in C N C N

C N

CN

O N

O O

in O N

O N I

in

C N

O N C O I oo CN

N O N O

N O O N N O

C O

oo

i n q oo I N O

O N

in

co p col O N

oo

O CN

O CN

O N O

00 N O

T t

N O

4>

& H

2 © in

co C N

O N C N

N O C N

i n O N N O

C N

N O O O

* o

1 _ L

C O

in O N

O N C N

C N C N

* N O I 1 -

O N

C N

C/3 QJ

cd

a o

* O N

O

Mean (S.E.M.) Concentration-Time Profile

— + — FA - O - U V - A - urine - • - A M N

1000 F ~

0 20 40 60 80 100

Time (Hours) ^ Post-Infusion

;ure 42: Mean concentration-time profiles of indomethacin in plasma, urine,

and amniotic fluid for high dose ewes (n=5).

144

Mean (S.E.M.) Concentration-Time Profile

0 20 40 60 80

Time (Hours)

Figure 43: Mean concentration-time profiles of indomethacin in FA and UV

plasma for low dose ewes (n=2).

145

from —111-208 ng/mL during the infusion period. MA and UTV samples from one

ewe (E796) were analysed but no drug was detected in them, i.e., the concentration

was below the detection limit of the assay. The arterio-venous concentration

difference for indomethacin at apparent steady state in the high dose ewes averaged

11.72 ± 1.53 ng/mL (mean ± 1 S.D.) (Table 13), and while small, was significantly

different (a=0.05, t-test, Table 14). These results suggest a low placental

permeability to indomethacin in sheep.

Fetal total body clearance was estimated to be 44.55 + 11.14 mL/min (n=7,

Table 15). It was only possible to calculate placental and non-placental clearances

in two animals because of incomplete umbilical blood flow data. The placental

clearance in E 796 was determined to be about 32.00 mL/min, and the non-placental

clearance to be 6.44 mL/min (Table 16). In a low dose ewe (E 1115), the values

were 26.47 mL/min and 22.16 mL/min, respectively. These values are much lower

than those reported for either diphenhydramine or metoclopramide (Rurak et al.,

1991). Ritodrine, on the other hand, had a fetal placental clearance of about 5

mL/min/kg and a markedly higher fetal non-placental clearance of 63 mL/min/kg

(Rurak et al, 1991).

146

Table 13: Arterio-venous concentration difference for indomethacin at

apparent steady state.

Ewe No. [ F A ] t a d o (ss) [UVJindo (SS) [A-VJindo (SS) Extraction

ng/mL 1 ng/mL 2 ng/mL 3 Ratio4

796 195.07 181.71 13.37 0.068

984Y 113.12 101.67 11.45 0.101

2160 222.25 - - -

1221 161.69 151.35 10.34 0.064

1211 216.21 - - -

1137* 39.07 36.67 2.39 0.061

1115* 43.51 - - -

* Low dose group 1: Mean apparent steady-state fetal arterial (FA) indomethacin concentration. 2: Mean apparent steady-state umbilical venous (UV) indomethacin concentration. 3: Mean arterio-venous (A-V) indomethacin concentration difference. 4: Extraction ratio = C F A - C U V / C F A , where C is the concentration of indomethacin at

apparent steady-state.

147

Table 14: Arterio-venous concentration difference: statistical evaluation

To determine whether fetal umbilical venous and femoral arterial indomethacin concentration differences are statistically significant.

Two-tailed paired sample t-test

Ewe FA Indo. steady UV Indo. steady [A-V] Indo. at state cone. state cone. steady state [ng/mL] [ng/mL] [ng/mL]

796 195.078 181.71 13.368 984 113.123 101.673 11.450 1221 161.694 151.356 10.338

n=3, v=n-l=2 d=l 1.718 ng/mL S d = 0.8849

t = d7Sa= 13.241

t0.05(2).2 = 4.303

Reject H 0 (i.e., p,d=0), at oc=0.05

148

Table 15: Determination of fetal total body clearance.

Ewe No. Infusion Rate

mg/min

[FA]i„d0 (ss) ng/mL*

CLtb mL/min**

796 0.0075 195.07 38.45

984Y 0.0075 113.12 66.29

2160 0.0075 222.25 33.75

1221 0.0075 161.69 46.38

1211 0.0075 216.21 34.69

1137 0.0019 39.07 48.63

1115 0.0019 43.51 43.67

* Mean steady-state fetal arterial (FA) concentration of indomethacin

** Total body clearance.

Mean total body clearance: 44.55±11.14 mL/min

Clfo

[mL

/min

] 6.445

22.159

Cl p

[mL/

min

] 32.001

26.467

S 38.446

48.626

O

* [m

L/m

in]

as as

<r>

O

* [m

L/m

in]

so so t-

i — i

co O

* [m

L/m

in]

[A-V]

[ng/

mL]

13.368

2.398

I S

195.078

39.073

ight

ta

l Wei

[kgl

3.393

3.194

Fel

Ewe *o

as i>

e'­en

1/5

T3 O .—I U H

<D &

4 C / 3

CD 4-* fli > 4 o <D OH fi fli

X !

fli > O

T3 fli

a o ~3 o

5 S » 6 E * «« « u

^ ^ s o

T3

-a o > ~J3

£ E J

150

3.3.2. Analysis of Urine Samples

The results from the analysis of indomethacin in urine samples are

summarised in Table 17 and Figure 42. As Table 17 indicates, the levels of intact

indomethacin in fetal urine are, in general, very low. The urine samples were also

subjected to enzyme ((^-glucuronidase) hydrolysis to determine if there were

conjugates present. While there is considerable variability, the results indicate that

the fetal lamb does have some ability to conjugate indomethacin. Figure 44

illustrates the urine flow rate (mL/min) as a function of time during indomethacin

infusion. In three of the fetuses (E 2160, E 984Y, and E 796), there was a marked

reduction in urine flow rate within 23 hours (-50% reduction) of initiating the

indomethacin infusion. However, in the other two fetuses, urine flow was observed

to increase.

The estimated fetal renal clearance of indomethacin, in the high dose group,

is about 0.015 ± 0.009 mL/min (Table 18) (Figure 45), and is not significantly

different from zero. This finding is consistent with observations for other weak

organic acids in the fetal lamb, viz., diphenylmethoxyacetic acid, a metabolite of

diphenhydramine (Tonn et al., 1994), and valproic acid (Gordon et al., 1994).

Cumulative amounts of drug excreted could not be calculated since urine samples

151

were only collected over a one-half to one hour period at each blood sampling

interval.

3.3.3. Drug Disposition in Amniotic and Tracheal Fluid Compartments

The concentration of indomethacin in amniotic fluid, in the high dose

experiments, was significantly lower than the plasma concentrations and slightly

higher than fetal urine levels as illustrated in Figure 42. Amniotic fluid

indomethacin concentrations were below detection limits in E 1137 and E 1115, the

two low dose ewes. These findings are in sharp contrast to basic drugs such as

metoclopramide, labetalol, and ritodrine which have been shown to accumulate

considerably in amniotic fluid (Rurak et al., 1991). Conjugates were not measured

in amniotic fluid due to limitations in sample volume.

Indomethacin levels in fetal tracheal fluid samples were below the detection

limit of the assay. Again, this is in sharp contrast to basic drugs, such as

metoclopramide, ritodrine, and labetalol, wherein the levels in tracheal fluid far

exceeded those in fetal plasma, the extent of accumulation being the greatest for

metoclopramide (Rurak et al., 1991).

152

a *E s Is 4-t

cu OJ

cu s o TS a

CM O

e cs M

c CU CJ a ©

—

H

CN

o

1—I

T t 00 O S

p

V© ON f -

cu P

TJ J. S i 1

P «T

P TJ S.

P

p TJ

sis p

TJ J,

p TS S.

* I

TS P

P siT

cu

5 CN

CO

o *o

<u cd G O

o

c/3 c/5 cs-

5 00 <+H ' O

cu "3 S I

o fi o > <o ^ s

CU

o c

e l l — <u — O •—i G

CU S T3 « T j U - O fl cn CU

O ^ °

P Q

CU

CU

t rt

T3 (U H—» o CU H—» CU C O

rt 00

n3 cu rt 00

O o

o rt

O G

OH

o t-t 00 cu C/3 o

T3 OH £

(U . - I

cu

-2 P

153

Figure 44: Effect of indomethacin on the fetal urinary flow rate.

154

Figure 45: Representative plot for the determination of renal clearance.

155

Table 18: Determination of renal clearance of indomethacin in high dose

group fetuses

Ewe No. Renal Clearance (CL r )

mL/min

E 7 9 6 0.00169

E 9 8 4 Y 0.00544

E2160 0.00817

E 1221 0.05200

E 1211 0.01200

156

4. DISCUSSION

4.1. Development of a capillary G C - E C D method

Purity assessment of indomethacin

According to O'Brien et ah, 1984, indomethacin is defined as the crystalline,

non-solvated free acid entity of the compound (Form I; see section 3.1.1.). The

indomethacin sample we procured from Sigma Chemical Co., St. Louis, MO, was

subjected to a variety of physical tests to authenticate its purity for our experimental

studies. NMR studies indicated that the chemical shift values obtained with our

sample compared well with those obtained with Merck reference standard (Table 3)

(O'Brien et al, 1984). DSC studies indicated that the melting point of our

indomethacin sample was at 162°C (Figure 6). The Form I (crystalline)

indomethacin has a melting point of 162°C (NF XIII). Since no other peak was

observed in the DSC scan, other than that which melted at 162°C, we can attribute

the melting temperature as observed to Form I. This is in full aggreement with that

observed with the Merck standard; the recorded temperature being 162°C (O'Brien

et al., 1984). The mass spectrum obtained in an LC/MS/MS system was

characterised by a molecular ion at m/z 358 (the molecular weight of indomethacin

is 357.81) and a characteristic base peak at m/z 139 corresponding to the p-

157

chlorobenzoyl group (Figure 7). These findings confirmed that indomethacin, as

obtained from Sigma, is comparable in its physical assessment to the Merck

standard. Indomethacin was therefore used in our experimental studies without

further purification.

Assay development

Several packed column GC assay procedures have been described for the

determination of indomethacin in human plasma/serum and urine (Guissou et ah,

1983; Sibeon et al, 1978; Matsuki et al, 1983; Plazonnet and Vandenheuvel,

1977; Jensen, 1978; Arbin, 1977; Evans, 1980; Ferry et al, 1974; Helleberg,

1976). These packed column methods suffer from serious limitations of poor

resolution and chromatography, since they are prone to interference from

endogenous peaks due to fewer theoretical plates compared to capillary columns.

There is also one report on the use of fused silica capillary column technology for

indomethacin (Nishioka et al, 1990) determination in plasma. However, this

method has a number of disadvantages including: high sample volume (1 mL),

lengthy derivatisation step with l-ethyl-3-p-tolyltriazine (i.e., 6 hours), lengthy

retention times (-20 min), a moderate recovery of 87% using a solid-phase

extraction procedure, and a detection limit of 2 ng/mL.

158

For our detailed pharmacokinetic studies with indomethacin there is the

potential for interference from components in biological matrices (e.g., tracheal and

amniotic fluids, plasma, and urine) as well as limitations on fetal fluid sampling

volume, in particular blood (-250-500 |iL of plasma). The existing analytical

methods were, therefore, not suitable and it was necessary to develop an analytical

method with high sensitivity, specificity, simplicity, reproducibility and robustness.

Selection of derivatising agent

Although the conjugated /?-chlorobenzoyl group in indomethacin affords the

inherent electron-capture property in the molecule, derivatisation is necessary to

increase volatility and thereby improve chromatography. This subsequently results

in enhanced sensitivity. A number of derivatising agents have been used for

indomethacin including the explosive diazoalkanes (Arbin, 1977; Giachetti et al.,

1983; Guissou et al, 1983; Helleberg, 1976; Plazonnet and Vandenheuvel, 1977),

and poisonous fluorine containing reagents (Matsuki et al, 1983; Evans, 1980;

Sibeon et al, 1978). There have been only two instances where the derivatising

agent is reportedly non-toxic, i.e., bis (trimethylsilyl) acetamide (Plazonnet and

Vandenheuvel, 1977), and l-ethyl-3-p-tolyltriazine (Nishioka et al, 1990).

However, bis (trimethylsilyl) acetamide (BSA) derivatisation produces

trimethylsilyl esters that are unstable to hydrolysis, and thus require protection from

159

moisture, and the use of freshly derivatised samples for GC runs. The use of ethyl

tolyltriazine is also not desirable since the derivatisation step is very lengthy, i.e., in

the order of 6 hours. Since BSA as a silylating agent provides the advantages of

safety and simplicity of the derivatisation step, further newer silylating agents were

explored.

Silylation of organic compounds improves volatility greatly by blocking

protic sites thereby reducing dipole-dipole interactions, as illustrated below:

H-Y + — I

Si I

6+ Y I H

V -S i -

8-X Y-SI- + HX

There is a nucleophilic reaction between the electronegative heteroatom and the

silicon atom of the silyl donor resulting in a bimolecular transition state. The

derivatisation proceeds to completion if the basicity of the leaving group (X) is

greater than that of Y. Silylation with MTBSTFA confers several advantages over

other silylating agents (BSA, MSTFA, BSTFA, TBDM-SIM, and TBDMCS) such

as:

(a) increased reactivity, especially its ability to silylate carboxyls,

(b) short derivatisation reaction times, and

(c) superior hydrolytic stability (while TMS derivatives need to be freshly prepared

and injected into the GC, tBDMS derivatives are stable to hydrolysis as was

160

observed in this study where the processed samples were stable on the GC

autosampler tray for at least one week)

Based on our studies, MTBSTFA derivatisation was rapid and produced

tBDMS derivatives with superior hydrolytic stability in agreement with previous

studies by Mawhinney and Madson, 1982, for other compounds with reactive

carboxyl groups.

Derivatising with PFBBr produced comparable sensitivity to MTBSTFA.

Because sample preparation was more cumbersome and time consuming with

PFBBr we chose to use MTBSTFA for our assay. This is the first instance wherein

a silylating agent like MTBSTFA has been used for indomethacin with high

sensitivity electron-capture detection.

The electron-capture properties of tBDMS-indomethacin derivatives can be

attributed to the conjugated p-chlorobenzoyl group, more specifically the carbonyl

group in resonance with an imide structure, i.e., the two aromatic ring systems

enhancing charge delocalisation (Plazonnet and Vandenheuvel, 1977; Matin and

Rowland, 1972). The carbonyl carbon acts as the first site of electron acceptance

while the final electron-capture response is dependent upon resonance stabilisation

as illustrated in Scheme 2.

161

Although the GC-MS spectra of indomethacin derivatised by BSA has been

reported (Plazonnet and Vandenheuvel, 1977), there are no detailed reports in the

literature of mass spectra of tBDMS derivatised indomethacin. The tBDMS

derivative of indomethacin was subjected to GC-MS analysis to obtain EI, NCI, and

PCI spectrums (Figure 30). Following EI, the most intense peak occurs at m/z 139

due to the />chlorobenzoyl fragment (Figure 30 A). The characteristic [M-57]+

peak at m/z 414 represents the loss of the C(CH 3) 3+ fragment. The EI scan also

shows a molecular ion peak at m/z 471. The spectrum following chemical

162

ionisation shows the molecular ion peak at m/z 471 (NCI, M+) and m/z 412 (PCI,

MH +) (Figure 30, B and C).

During optimisation of the extraction conditions for this assay, a number of

solvents were examined (viz., ethyl acetate, toluene, hexane, diethyl ether, toluene-

0.0125M TEA, hexane-10% IPA, hexane-ethyl acetate, and methylene chloride),

with ethyl acetate providing the highest extraction efficiency (Table 4). In order to

improve the selectivity of extraction, various mixtures of hexane-ethyl acetate were

examined. However, the addition of hexane markedly reduced the extraction

efficiency of ethyl acetate and provided only a marginal improvement in selectivity.

Of all of the solvents tested ethyl acetate alone was found to provide the best

recovery of indomethacin from sheep biological fluids and the cleanest

chromatograms.

A number of solvents have been used in existing assay procedures for

indomethacin, the most common being dichloroethane (Sibeon et al, 1978; Bayne

et al, 1981; Helleberg, 1976), ethyl acetate (Matsuki et al, 1983; Evans, 1980),

hexane-ethyl acetate (4:1), (Plazonnet and Vandenheuvel, 1977), dichloromethane

(Jensen, 1978), and hexane-amyl alcohol 10% (Ferry et al, 1974). Nishioka et al,

1990, are the only group who have attempted a solid-phase extraction procedure for

indomethacin, however, this method does not offer any advantages over the

163

conventional liquid-liquid extraction, either in extraction recovery (87%), or in

terms of cleanliness of the chromatogram.

The extractability of indomethacin is highly pH-dependent. Our results

indicated that attainment and maintenance of pH 5.00 was required for optimal

recovery. These results aggree well with similar observations of Helleberg (1976),

and Ou and Frawley (1984). Recovery is drastically reduced when pH is less than 4

or higher than 6. The pKa of indomethacin is 4.5-4.7 (O'Brien et al, 1984), and

this likely explains the optimal recovery around pH 5.00. Varying times of

extraction have been reported in the literature for indomethacin including 5 min

(Giachetti et al, 1983), 15 min (Ferry et al, 191 A; Helleberg, 1976), and as high as

40 min (Jensen, 1978). In the present study, however, extraction time was

optimised as 20 minutes.

Mean recoveries from plasma and urine in the present study were 96.99 ±

9.25% and 94.74 ± 6.75%, respectively. These values are comparable with those

reported in the literature: 95% from serum using dichloromethane (Jensen, 1978),

92% from plasma using ethyl acetate (Evans, 1980), 95% from serum with

dichloroethane (Helleberg, 1976), and 95% from plasma with a 50:50 mixture of

petroleum ether-dichloromethane (Johnson and Ray, 1992). The values are greater

than the solid-phase extraction procedure of Nishioka et al., 1990, i.e., about 87% in

164

plasma and the liquid-liquid extraction procedure with dichloromethane of Sibeon et

al., 1978, i.e., about 85% from plasma.

The column used in this assay is a fused silica wide bore capillary column

cross-linked with 5% phenylmethylsilicone (25 m x 0.32 mm x 0.52 pjn film

thickness). Chromatographic'behaviour of indomethacin and the internal standard,

in terms of resolution, retention times, sensitivity, and peak geometry, was optimal

with this relatively non-polar column stationary phase (Rtind0=~4.8 min). When a

narrow bore column with the same stationary phase was used (25 m x 0.2 mm x

0.33 |im film thickness), retention times (Rt) were longer and sensitivity was lower.

The decrease in sensitivity is likely in part due to the lower sample capacity of the

narrow column as only a 1 p i volume can be injected compared to 2 |iL for the

wide bore column. With a narrow bore column of the same phase, there is both a

reduction in internal diameter was well as in stationary phase thickness. Equation 1

illustrates the relationship between phase ratio, p\ internal radius, r, of a wall coated

open tubular column, and stationary phase film thickness, df.

P=r/2df (1)

From this equation, it can be observed that a decrease in df (if r is constant) will

result in a reduction in partition ratio and a decrease in component retention and

resolution with reduction in run times (Hyver and Sandra, 1989). However, with the

narrow-bore column, the internal diameter is also smaller, with run times being

165

dramatically longer. When a polar DB-1701 column was used (narrow-bore, methyl

[7% phenyl, 7% cyanopropyl] silicone) of dimensions, 30 m x 0.25 mm x 0.25 |im,

however, the elution behaviour of the analytes were reversed. Zomepirac, being the

more polar compound was retained longer in the column than was indomethacin.

The splitless mode of injection was used for all analyses since this method of

sample introduction is generally very suitable for trace level analysis. This is

because 100% of the injected sample is introduced into the column (Hyver and

Sandra, 1989).

The time delay following inlet purging to occur is critical for the following

reasons and was therefore optimised. The purge delay time is dependent upon the

solvent and solute characteristics, capacity of the vaporising chamber, sample

volume, speed of injection, and the velocity of the carrier gas. Short purge delay

times will prevent complete transfer of solute into the column while long delay

times will prevent venting of residual waste vapour resulting in their transfer onto

the column and solvent tailing. A purge delay time of 0.75 min was chosen from the

tested range of 10-120 seconds as this provided optimal peak response (Figure 22).

The overall peak characteristics (peak shape, geometry, etc.) of the analyte

are dependent upon the injection port temperature. Stability of the derivatised

166

compound is also important as it may decompose at higher temperatures resulting in

a loss of sensitivity. A very low injection port temperature may result in peak

broadening as a result of slower sample vapourisation with the time for transfer into

the column being longer. In this study, it was determined that maximal peak area

counts of indomethacin were observed over the 180-200°C range and that any

further increase in temperature resulted in a drastic decline in absolute indomethacin

area counts. Usually the injection port temperature is higher than the column

temperature. However, in the present study an injection port temperature of 200°C

and an initial column temperature of 210°C were chosen as these provided optimal

response.

The column temperature has a direct effect on run time and peak geometry.

It also has a marked effect on peak separation. The log of partition ratio (k) is

inversely related to the temperature. The lower the column temperature, the larger

the partition ratio, thereby increasing the time the analyte spends in the stationary

phase. This increased retention of the analyte in the liquid phase confers greater

selectivity. Various column temperatures were examined and an initial column

temperature of 210°C was selected with quick ramping at 40°C/min to 300°C, the

final column temperature. Maximal absolute indomethacin area counts were

observed at this final column temperature. These column temperatures also resulted

167

in short analysis times (~9 min) and are within the maximum 400 °C temperature

rating recommended for these columns.

The carrier gases commonly used in GC are hydrogen, helium, and nitrogen.

Nitrogen is a denser gas compared to either hydrogen or helium. Of the three gases,

the major advantage of using hydrogen is that the value for the theoretical plate

height increases very slowly when one increases the average linear gas velocity

through the column. This results in a in faster analysis with little loss in resolution.

In the current study, hydrogen was used as the carrier gas for these reasons. A

variety of carrier gases have, however, been used for the GC analysis of

indomethacin including nitrogen (Sibeon et al., 1978; Arbin, 1977; Evans, 1980;

Ferry et al., 197'4; Helleberg, 1976); argon-methane, 90:10 (Plazonnet and

Vandenheuvel, 1977; Jensen, 1978; Guissou et al., 1983); and helium (Nishioka

al., 1990). Except for the method of Nishioka et al., 1990, these assays involve the

application of packed column technology.

Make-up gases are frequently used in many capillary GC methods since the

low column flow rates (1-5 mL/min) employed do not sufficiently maintain cell

volume or effectively sweep the electron capture detector cell (ECD). The ECD is

a concentration sensitive detector, the sensitivity being inversely related to flow.

Flow through the cell then, must be sufficient to maintain ECD cell volume in order

168

to accomodate the high surface area of the 6 3Ni source as well as to efficiently

sweep the cell so that solute mixing does not occur. Use of the make-up gas also

assures an equlibrium concentrations of thermal electrons in the ECD. Argon^

methane (95:5) was chosen as a make-up gas at a flow rate of 65 mL/min, since this

provided the greatest sensitivity, the lowest S/N ratio, and minimal band spreading.

The length of the capillary column entering the detector cell appeared to

influence absolute indomethacin area counts markedly. In spite of the

recommended depth of 72 mm into the ECD cell (Hewlett-Packard support line), a

depth of about 40 mm provided maximal area counts and was therefore chosen for

the assay. This could be due to effective sweeping of the make-up gas at the

detector base.

In summary, the optimised operating conditions for routine analyses were:

column, Ultra-2 fused silica capillary column cross-linked with 5%

phenylmethylsilicone (25 m x 0.31 mm x 0.52 urn film thickness); injection mode,

splitless; injection port temperature, 200°C; initial column temperature, 210°C (1

min); oven programming rate, 40°C/min; final oven temperature, 300°C (6 min);

detector temperature (ECD), 330°C; carrier gas, ultra high purity hydrogen gas at a

column head pressure of 70 kPa (column flow rate ~1 mL/min); purge delay time,

0.75 min; make-up gas, argon-methane, 95:5, at a flow rate of 65 mL/min.

169

Internal Standards

Several compounds were screened as potential internal standards in this assay

(Figure 12 illustrates some of these). A variety of internal standards have been

reported in the literature including compounds that are not structurally related to

indomethacin, such as phenylbutasone (Mehta and Calvert, 1983), phenacetin

(Avgerinos and Malamataris, 1989), testosterone (Kwong et al., 1982), itraconazole

(Al-Angaary et al, 1990), penfluridol (Guissou et al., 1983), esters of indomethacin,

such as indomethacin propyl ester (Nishioka et al., 1990), and indomethacin methyl

ester (Arbin, 1977), and close structural analogs of indomethacin, such as a-methyl

indomethacin (Stubbs et al., 1986; Bayne et al., 1981), and 5-fluoroindomethacin

(Sibeon et al, 1978).

Ideally the internal standard should:

1. Be a close structural analog of indomethacin, so that it exhibits similar extraction

and chromatographic behaviour.

2. Possess an active carboxylic acid site for derivatisation like indomethacin.

3. Elute in a region free from endogenous interference and close to the analyte of

interest.

170

4. Produce a peak response equivalent to the analyte's, resulting in area ratio of

unity at equivalent concentrations.

5. Not exhibit any chemical interactions with indomethacin.

6. Possess good chemical stability following derivatisation.

With these requirements in mind several compounds were evaluated as potential

candidates. Sulindac, a close structural analog of indomethacin, was one of the first

compounds to be tested (Figure 12). Chemically, it had an ideal halogen (i.e.,

fluorine) function on the 5 th position of the heterocyclic ring and an acetic acid

function on the 3rd. It also produced a satisfactory peak when derivatised with

either MTBSTFA or PFBBr, and calibration curves when derivatised with the

former, were linear with r>0.999. However, it was speculated that sulindac was

oxidised to a sulphone in the injection port, due to the appearance of a coeluting

peak in the chromatogram (Figure 14). Use of lower injection port temperatures did

not affect peak "X" as it continued to elute. Recrystallising sulindac from

chloroform also did not help, although one polymorph was selectively isolated as

depicted in the DSC scan in Figure 15. The fact that there is no reported GC

method for sulindac in the literature would also appear to indicate that this

compound is not stable under these conditions. Based upon these observations this

compound was not considered further.

171

Zomepirac, a structurally similar compound to indomethacin, provided a

fairly good chromatographic response when derivatised with MTBSTFA and

appeared to satisfy most of the requirements mentioned above. Figure 16 illustrates

a representative chromatogram of indomethacin and zomepirac spiked in fetal

plasma. When spiked in stored plasma, amniotic, tracheal fluids, and urine (fetal

and maternal) as well as freshly collected amniotic and tracheal fluids and fetal

urine, both indomethacin (Rt = 5.9 min) and zomepirac (Rt = 3.9 min) eluted in

regions free from any significant interference. However, the analysis of

experimental samples and use of freshly collected blank fetal plasma samples

resulted in a co-eluting endogenous peak that interfered with that of zomepirac. It

was, therefore, not possible to use zomepirac as an internal standard for the analysis

of experimental samples. Zomepirac was used, however, as the internal standard

during initial validation studies with amniotic fluid (Tables 5b and 6c). Figure 36

provides a representative chromatogram of indomethacin and zomepirac spiked in

amniotic fluid and illustrates that no interference was present for either compound.

A further re-examination of chromatographic conditions, columns,

derivatising agents, and screening of other internal standards (carprofen,

acemetacin, 5-fluoroindomethacin) was then conducted. However, none of these

proved to be satisfactory, a-methyl indomethacin, on the other hand, met all the

requirements of an ideal internal standard outlined above. It is a very close

172

structural analog of indomethacin (Figure 17), elutes in a region free from any

endogenous interference (Figure 33) and close to the analyte, and yields an area

ratio of about unity at equivalent concentrations, a-methyl indomethacin was

chosen, therefore, as the internal standard for the final assay procedure.

Using a-methyl indomethacin, calibration curves were linear over the

working range of 1-32 ng of indomethacin. mL"1 with a coefficient of determination

(r2) > 0.99, following extraction of 0.1 mL of plasma, and 0.1-1.0 mL of urine,

amniotic and tracheal fluids. Linearity was also observed over a concentration

range of 1-500 ng/mL (r2 > 0.99). Inter- and intra-day variabilities at each

concentration point were < 11 % for concentrations between 2-32 ng/mL and < 20

% at the L O Q of 1 ng/mL. The limit of quantitation is 1 ng/mL (< 10% C . V . , S/N >

10). The developed method is therefore reproducible and robust for accurate

determination of indomethacin concentrations in biological fluids.

Derivatisation Conditions

Derivatisation of indomethacin and a-methyl indomethacin with 100 u L of

toluene containing 8 uL of M T B S T F A (8% v/v) at 6 0 ° C for 50 minutes was found

to be rapid, constant, and reproducible with maximal indomethacin area counts. No

significant differences in peak response were observed between the temperatures of

173

60-90°C (Figure 10) and between reaction times of 50-60 minutes (Figure 11). A

reaction time of 50 minutes and temperature of 60 °C was chosen to ensure

complete derivatisation. This is the first reported use of MTBSTFA as a

derivatising agent for indomethacin. Plazonnet and Vandenheuvel, 1977, used BSA

to silylate indomethacin producing the trimethylsilyl ester. They derivatised

indomethacin with 10 |iL of BSA and 90 uL of ethyl acetate (10%), and

reconstituted the final samples with 3% BSA in ethyl acetate since the resulting

TMS esters were prone to hydrolysis. However, this problem is not evident for

MTBSTFA since this molecule contains sterically crowded alkyl substituents on the

silicon atom that confers hydrolytic stability. In this assay, 100 |iL of a 8% v/v

solution of MTBSTFA (8 |iL) in toluene (92 fiL) produced optimal reaction

conditions.

A solvent is usually used in the reaction medium for derivatisation and for

reconstitution, i.e., dilution prior to injection into GC. Some of the recommended

solvents in silylation reactions include pyridine, dimethylformamide (DMF),

dimethyl sulphoxide, tetrahydrofuran, and acetonitrile (Blau and Halket, 1993).

However, toluene appeared to be a better medium for MTBSTFA derivatisation and

for sample dilution than any of these solvents (DMF, Acetonitrile, or ethyl acetate),

and on this basis was selected for the assay.

174

It has been suggested that addition of t-butyldimethylchlorosilane

(TBDMCS, 1% v/v) as a catalyst improves the silylation power of MTBSTFA

(Mawhinney and Madson, 1982; Ballard et al, 1990; Lewis and Yudkoff, 1987).

However, addition of this agent did not have any effect on indomethacin area

counts.

MTBSTFA has been popularly used over the recent years to derivatise

various chemicals and drugs for GC determinations. The carboxy lie acid

derivatisations reported in the literature include: carboxylic and hydroxyl groups in

tartaric acid and y-resorcylic acid (Kim et al, 1989), and valproic acid and its

metabolites (Abbott et al, 1986).

Stability studies on indomethacin

Indomethacin is a pale yellow crystalline powder. It may exist in amorphous

or crystalline forms [Form I (mp 160-162 °C) and Form II (mp 153-155°C), with

Form I being more stable]. Form II converts to the more stable Form I on storage.

Indomethacin is sparingly soluble in water and early pharmaceutical formulations

incorporated a small amount of mild alkali (sodium carbonate or bicarbonate) to

improve solubility. However, in the presence of strong alkali such as sodium

175

hydroxide, indomethacin (I) decomposes to 2-methyl-5-methoxy-indole-3-acetic

acid (II) and p-chlorobenzoic acid (III), as shown below in Scheme-3.

Scheme-3: Degradation of indomethacin in alkali

The following Table illustrates some of the stability data on indomethacin in

aqueous solution:

Table 19: Stability of indomethacin in aqueous solution (Adapted from Shen and Winter, 1977)

pH Temperature (°C) Degradation Half-life

(Tw) 12.00 RT* 1 min

10.00 RT 1-1.5 h

9.40 50 0.38 h

8.14 50 3.13 h

8.00 RT 195 h

* room temperature

176

In our study, indomethacin was administered in a vehicle of 1.1% ethanol

and 0.75% sodium bicarbonate in normal saline. The pH of this infusate is about

7.8-7.9. Stability studies on indomethacin in this vehicle indicated that the drug was

stable for at least 96 hours at both room temperature and standard regfrigerator

temperature (4°C) (Figure 37). This observation is consistent with that of Shen and

Winter, 1977, who determined the T i / 2 of indomethacin in an aqueous solution, at

pH 8.00, to be about 195 h at room temperature.

Several pharmaceutical formulations have been designed for indomethacin

administration in animal experimentation from the combined standpoints of stability

and solubility. It is strongly recommended that these solutions be freshly prepared

before each study. These indomethacin formulations usually contain sodium

carbonate, mannitol, and Emulphor-620, with the pH of the resulting solution being

between 8.4-9.1 (Shen and Winter, 1977).

Indomethacin powder solubilised in a vehicle of simple syrup with 10%

alcohol (pH 5.2) for pediatric administration appeared to be stable for at least 224

days at 24°C (Das Gupta et al, 1978). Curry et al, 1982, reported that a solution of

indomethacin in 0.1 M sodium carbonate (pH 10.7) lost 75% of its content in 80

minutes: Adjustment of the pH to 7.4 resulted in stability for at least 24 h. These

authors also found that a solution of indomethacin in polyethylene glycol, PEG 400,

1 7 7

was stable. However, Curry et al., 1982, also indicated that inclusion of sodium

carbonate in a vehicle can be potentially dangerous, and recommended that while

preparing a solution containing Na2C03, the sodium carbonate must be added very

slowly to the indomethacin slurry to avoid rapid degradation in a high alkalinity

environment. It appears that vehicles containing PEG or alcohol would be highly

desirable from the standpoint of stability, but not quite so from the pharmacological

perspective.

The maximal stability of indomethacin in aqueous solution at 25 °C is at pH

4.9, and solutions at this pH are predicted to be stable for at least 2.0 years (Kahns et

al., 1989). A 3% aqueous indomethacin solution was prepared in 30% each of ethyl

carbamate and ethylurea and the mixture boiled for 30 sees (pH 6.95) resulting in

stability for 25 months at 20°C (Pawelczyk and Knitter, 1978).

Indomethacin was observed to be most stable in surfactant systems

solubilised in Tween 80 or Brij 99. The T 1 / 2 in these solutions being more than 6

months (Ghanem et al., 1979). The following Table summarises some of these

vehicles:

178

Table 20: Stability of indomethacin in surfactant solubilised systems (Adapted from Ghanem et al., 1979)

Vehicle Condition T 5 0 (months) Tween 80 RT* 9.8 Tween 80 5°C 12.6

Brij 99 RT 10.7 Brij 99 5°C 15.5

Tween 80, 40% syrup, sodium benzoate

RT 6.8

Tween 80, 60% syrup, sodium benzoate

RT 12.4

Tween 80, 20% glycerol, 40% propylene glycol

RT 11.9

Tween 80, 10% ethanol, 10% glycerol

RT 11.9

Brij 99, chlorocresol 5°C 17.7 Brij 99, sodium thiosulphate 5°C 17.7 * room temperature

It appears then, that the following parameters are important to ensure drug

stability: p H of the infusate (it is preferable to keep the p H from exceeding 8.0

units); mode and order of ingredient mixing (as determined by Curry et al., 1982);

and use of surfactants (these wi l l aid against degradation). I also strongly

recommended that freshly prepared solutions be used. Use of the sodium trihydrate

salt of indomethacin, which is presently being employed in the treatment of patent

ductus arteriosus, however, may be preferable as it would be completely soluble in

aqueous solutions.

The stability of indomethacin in biological fluids and extracted samples was

also assessed. The drug was found to be stable for at least four freeze-thaw cycles,

179

at least 24 h on the bench-top, at least 24 h on the acidified sample matrix, and at

least one week on the autosampler tray (Table 6). Indomethacin is also stable in

plasma and urine samples stored in the freezer in the time periods studied to date

(over 14 months).

Data on the stability of indomethacin during analytical development in

existing assays have been sparse. Al-Angary et al., 1990, found that indomethacin

was stable for at least 1 month in serum at -20°C. Indomethacin was found to be

stable for at least 30 minutes in various extraction solvents including ethyl acetate,

hexane 2 propanol, chloroform, benzene-cyclohexane, and hexane-diethyl ether, all

acidified to pH 2.0 with HC1 (Catrula and Cusido, 1992).

4.2. Disposition of indomethacin in sheep

Following assay development and validation the method was applied to the

study of indomethacin disposition in the chronically instrumented pregnant sheep

model. The indomethacin concentrations in the fetus, obtained in the high dose

group, were similar to those observed during human tocolysis , i.e., ~ 200 ng/mL

(Moise et al., 1990). Steady state levels in fetal plasma were achieved over the 5-69

hour period following indomethacin infusion, while in some fetuses this was

delayed and was observed over the 23-69 hour period. Since samples were only

180

collected twice daily, six hours apart, it was not possible to determine the exact time

and/or duration of the steady state achieved. The mean steady state fetal femoral

arterial concentration of indomethacin was 181.67 + 45.05 ng/mL in the high dose

ewes (n=5), and 41.29 ± 3.14 ng/mL in the low dose ewes (n=2). Fetal total body

clearance was estimated to be 44.55 ± 11.14 mL/min (n=7). Indomethacin was

detected in very low levels in urine (Section 4.2.3.). The urine samples were also

subjected to enzyme hydrolysis with [ -glucuronidase, and the results indicate the

presence of glucuronide metabolites of indomethacin, suggesting that the fetal lamb

does have some ability to conjugate indomethacin (Table 17). This aspect of fetal

drug metabolism (i.e., conjugation), however, requires further investigation.

4.2.1 Placental transfer of indomethacin in sheep

Indomethacin is a polar and fairly medium molecular weight (MW 357.8)

organic acid that crosses the placental membrane easily in humans (Moise et al,

1990) and rabbits (Parks et al, 1977; Harris and VanPetten, 1981). In the present

study with the pregnant sheep model, however, there is a limited placental

permeability to indomethacin. The sheep placenta has been shown to be relatively

impermeable to polar compounds (Boyd et al., 1976). There is a low but

statistically significant arterio-venous indomethacin concentration difference at

apparent steady state indicating placental uptake of the drug (Figures 42, and 43;

181

Tables 13 and 14). Indomethacin levels were below detection limits in maternal

arterial and uterine vein plasma samples collected in one ewe suggesting little or

quantitatively insignificant placental transfer of indomethacin. In other studies, the

fetal-to-maternal (F/M) drug ratio has been reported to be 0.28 (Anderson et al,

1980c) and 0.04 (Harris and Vanpetten, 1981) in sheep, indicating limited placental

transfer of indomethacin in this species. The epitheliochorial placenta in sheep has

a placental pore radius of 4-5 A, while the hemochorial placenta in rabbits and

humans behaves as if it has a placental pore radius of -300 A. This difference could

explain, in part, the limited placental permeability in sheep. In general, the sheep

placenta has been reported to be relatively impermeable to polar compounds (Boyd

et al., 1976). The partition coefficient of the nonionised form of indomethacin and

that of the ionised form of the drug, between 1-dodecanol and water at 32°C, has

been determined to be 2400 and 1.1, respectively (De Vos et al., 1994). Ionised

indomethacin is also reported to form ion-pairs with such ions as Na+, K+, and NH 4+

(De Vos et al., 1994). It is likely that indomethacin partitions into a lipophilic phase

predominantly in the un-ionised form. Indomethacin could form ion-pairs with

cations in the physiological milieu, and thus its implication in absorption through

biological membranes may be underestimated (De Vos et al., 1994).

A relatively high degree of plasma protein binding has been reported for

indomethacin in maternal (98.5%) and fetal (97.6%) sheep (Anderson et al., 1980c).

182

This is markedly higher than that observed in pregnant rabbits, where the maternal

binding of indomethacin (85.6%) is significantly higher than fetal binding (43.8%)

(Harris and Vanpetten, 1981). However, these differences cannot be attributed

solely to species differences because: (a) the methods of protein binding used are

different; ultracentrifugation (Anderson et al., 1980c), and equilibrium dialysis

(Harris and Vanpetten, 1981), and (b) use of radiolabeled indomethacin (14C) by

Anderson et al., 1980c, and unlabeled indomethacin by Harris and Vanpetten, 1981.

4.2.2. Placental clearance

To date, three major methods have been reported in the literature for

determining placental clearance: the Fick principle (Meschia et al., 1967), the Szeto

(Szeto et al, 1982b) and the Anderson methods (Anderson et al., 1980c). Placental

clearance is defined as the ratio of flux in the concentration difference between the

maternal artery and fetal umbilical artery (Meschia et al., 1967), while fetal

clearance may be defined as the number of mL of plasma from which a drug is

completely removed per unit of time (Anderson et al., 1980c).

The Fick principle involves sampling of maternal arterial and uterine venous

blood or fetal femoral arterial and umbilical venous blood to determine drug

concentrations in these samples following a respective maternal or fetal steady state

infusion. It also requires quantitative data on uterine or umbilical blood flow,

183

respectively. Placental clearance is then calculated as a product of the blood flow

rate and the ratio of arterio-venous concentration difference to the arterial drug

concentration at apparent steady state. The Anderson method involves the

simultaneous administration of a fetal intravenous bolus dose of labeled drug and

maternal i.v. infusion of unlabeled drug. The total fetal body clearance (Ctot = C p +

Cf) can be determined from the fetal bolus. Placental clearance, then, is determined

as the product of the total body clearance (Ctot) and the ratio of steady state fetal to

maternal concentrations of the drug (F/M). However, a major limitation is the

inability of the method to differentiate between the intact drug and its metabolites.

Other limitations include the difficulty of procuring a stable isotope or radiolabeled

form of the drug and possibility of an "isotope effect" occuring, i.e., selectivity of

the placenta between the isotope and its natural element. The Szeto method

involves the representation of the maternal-fetal unit as a two compartment

pharmacokinetic system that requires paired drug infusions to steady state to both

the mother and the fetus. A potential limitation again, is the availiability of a stable

isotope labeled drug and an analytical method that will selectively quantitate the

unlabeled compound and the stable isotope. If the stable isotope labeled drug is not

available, then the experiments need to be performed on different days with a

suitable washout period between them. This may result in considerable variablility

due to time-dependent changes in both maternal and fetal physiology. The fetus, for

example, grows at the rate of about 5% per day (Kong et al., 1975). Apart from

these limitations, the requirement that detectable drug concentrations should be seen

184

in the maternal circulation across the placenta during fetal infusion studies, and vice

versa, must be met. In the current fetal indomethacin infusion studies, drug

concentrations were below detection limits in MA and UTV samples precluding

therefore, the application of the Szeto method to determine placental clearance.

The Fick principle is also not without problems. It requires more extensive

surgical catheterisation of the umbilical venous or uterine blood vessel apart from

either MA or FA vessels. It also requires determination of either the umbilical or

uterine blood flow. Another problem lies in the complexities of the fetal circulation

as there may be a difference between observed and actual concentrations leaving the

sites of exchange, e.g., uterine venous blood may be a mix of myometrical and

placental venous blood.

In spite of these potential problems the Fick principle may be used to

determine placental clearance. However, it must be borne in mind that the method

allows for the estimation of either the fetal to maternal placental clearance or vice

versa, at any given time. Examples of its use include the estimation of placental

clearance of glucose analogs (Stacey et al., 1978), sodium and chloride (Armentrout

et al, 1977), and ritodrine (Wright, 1992) in the fetal lamb.

In this study, there is a low but statistically significant difference between the

indomethacin concentrations in FA and UV plasma at apparent steady state (mean

185

difference: 11.718 + 0.8849 ng/mL; mean ± s.e.m., n=3), indicating placental

uptake of indomethacin (Tables 13 and 14). Placental clearance values were

determined only for two ewes, one at the high dose, and the other at the low dose.

This was due to surgical failures of cannulation, catheter failures during

experimentation (especially UV sampling), and the unavailability of umbilical blood

flow data. The placental clearance of indomethacin as determined by the Fick

method was calculated to be 9.43 mL/min/kg in E 796 and 8.28 mL/min/kg in E

1137. These values are only somewhat higher than the value of 2.10 ± 0.32

mL/min/kg reported by Anderson et al, 1980c. The mean placental clearance value

for indomethacin in this study, i.e., 8.85 mL/min/kg, is similar to that determined for

ritodrine, i.e., 9.2 + 2.7 mL/min/kg (Wright, 1992), but markedly lower than the

values reported for other drugs in the fetal lamb, namely, metaclopramide - 103 ± 13

mL/min/kg (Riggs et al., 1989), diphenhydramine - 124 ± 22 mL/min/kg (Rurak et

al, 1991), and methadone - 168 ± 29 mL/min/kg (Szeto et al, 1982a). The

observed placental clearance of indomethacin in E 796 (high dose ewe) accounts for

more than 80% of the total fetal body clearance while in the low dose ewe (E 1137)

it accounts for about 55%.

186

4.2.3. Elimination of indomethacin by the fetal kidney

It was evident from the present study that the pathway for elimination of

indomethacin, a weak acid, by the fetal kidney are not matured. Very low levels of

indomethacin were observed in fetal urine samples collected from the high-dose

ewes, and indomethacin concentrations were below detection limits in urine samples

from the low dose ewes. Similar observations have been reported for other weakly

acidic agents such as valproic acid (Gordon et al., 1994), diphenylmethoxy acetic

acid or DPMA (Tonn et al., 1994), and p-aminohippuric acid (Elbourne et al., 1990)

in the ovine fetus. It has been speculated that the pathways for the excretion of

organic acids by the fetal kidney mature after birth in contrast to that for organic

bases which appear to mature much earlier (Elbourne et al., 1990). The estimated

fetal renal indomethacin clearance value in this study (Table 18) was very low

(0.015 ± 0.009 mL/min) and not significantly different from zero. This is again,

consistent with findings for valproic acid (Gordon et al., 1994) and

diphenylmethoxy acetic acid, an acid metabolite of diphenhydramine (Tonn et al.,

1994), in the fetal lamb.

187

4.3. Drug disposition in amniotic and tracheal fluids

4.3.1. Amniotic fluid

Indomethacin concentrations in amniotic fluid were significantly lower than

FA plasma levels in the present study. Amniotic fluid indomethacin concentrations

were below detection limits in E 1137 and E 1115, the two low dose ewes. These

findings are also consistent with those observed for valproic acid and DPMA

(Gordon et al., 199»4; Tonn et al., 1994). However, they are in sharp contrast to

observations for basic drugs (metoclopramide, dyphenhydramine, labetalol, and

ritodrine) which have been shown to accumulate considerably in amniotic fluid

(Rurak et al., 1991). Renal excretion of drug and subsequent appearance in

amniotic fluid has been observed for meperidine (Szeto et al., 1978), lidocaine

(Morishima et al., 1979), and ethanol (Clarke et al., 1987) in the fetal lamb. It has

been suggested that basic drugs accumulate in amniotic fluid due to the mechanism

of "ion-trapping" due to the lower pH of amniotic fluid as compared to fetal plasma

(Szeto et al., 1978). The sources of drug appearance in the amniotic fluid are two­

fold, i.e., fetal renal drug elimination (time delay in the appearance of drug in the

amniotic fluid following dosage is characteristic for this type), and diffusion across

chorio-allantoic membranes. The latter route was suggested to be a predominant

pathway for the appearance of meperidine in amniotic fluid (Szeto et al., 1978).

Renal elimination, on the other hand, has been suggested to be important for

1 8 8

compounds such as mannitol (Basso et al, 1977). Metabolites have also been

detected in amniotic fluid with the glucuronide conjugates of ritodrine serving as an

example (Wright et al., 1991). The presence of glucuronide conjugates in amniotic

fluid for indomethacin in the present study, however, could not be confirmed due to

limited sample volume.

4.3.2. Tracheal fluid

Indomethacin levels in fetal tracheal fluid samples were below assay

detection limits. While above detectable limits, the concentrations of valproic acid

(Gordon et al., 1994) and DPMA (Tonn et al., 1994), two other acidic compounds,

in tracheal fluid have been reported to be very low. This is in sharp contrast to basic

drugs, such as metaclopramide, ritodrine, labetalol, and dyphenhydramine, wherein

the drug levels in tracheal fluid were equivalent or even higher than those in fetal

plasma, the extent of accumulation being greatest for metoclopramide (Riggs, 1989;

Wright etal, 1991; Yoo, 1989; Yeleswaram, 1992).

Basic drugs are known to accumulate in the adult lung. Amphiphilic amines

are preferentially taken up by the lung which appears to be a major organ of

clearance for these agents. It has been suggested that molecules containing the

amino group and a lipophilic alkyl function are necessary for lung specificity and

uptake (Fowler et al., 1976) and that predictions can be made on the basis of drug

189

basicity and lipid solubility. Okumura et al., 1978, have implied the existence of

active transport system or a specialised binding system as a mechanism for

accumulation of basic drugs in the lung. There is also evidence to suggest that

mitochondrial monoamine oxidase (MAO) is a binding site for tertiary basic drugs

in the lung (Yoshida et al., 1989) and may hence act as a reservoir. Carrier-

mediated transport processess for the uptake of organic anions (Schanker, 1978) and

amino acids (Lin and Schanker, 1981) have been reported in the rat lung. However,

there is no reported accumulation of weakly acidic drugs in the tracheal fluid,

suggesting that the lung plays a minor role in the elimination of these compounds.

4.4. Effect of indomethacin on physiological factors

Indomethacin appears not to influence arterial p02 and arterial pC02 in both

high and low dose ewes, as they remain constant throughout the study period in

relation to control (Figures 41 A, B). Indomethacin also did not appear to affect

fetal arterial blood lactate levels in low dose ewes. However, in three high dose

ewes lactate levels were markedly higher than respective controls (Figure 41 D). It

was interesting to note that the drug did not influence arterial glucose levels in

either of the two dose groups (Figure 41 C). Significant decreases in arterial pH as

well as base excess were observed when indomethacin was given in higher doses,

whereas the values remained more or less constant in low dose ewes compared to

the respective controls (Figures 41 E, F). It can be observed that indomethacin

190

caused lactic acidosis in some high dose ewes (E 1221, E 984, and E 2160) and that

the effect might be dose related as the same was not observed in low dose ewes.

Further, a severe metabolic acidosis was observed when indomethacin was

administered as a 6 hour intravenous infusion at a dose of 17 jig/min in hypoxic

ovine fetuses (Hooper et al., 1992). The severe lactic acidosis observed in E 1221,

E 2160, and E 984 (high dose group fetuses) were characterised by markedly

reduced fetal pH. All the three fetuses died following cessation of drug infusion.

That the deaths of these fetuses was attributable solely to drug effect is

questionable, as death due to infectious etiology cannot be ruled out. Further, high

dose group fetuses that survived the experimental protocol demonstrate that the drug

does not cause significant lactic acidemia. Also, no significant reduction in arterial

pH, p02, pC02, or base excess was observed. Further investigation is underway to

elucidate the relationship between indomethacin and lactic acidemia.

In three of the high dose ewes (E 2160, E 984Y, and E 796), there was a

significant reduction in fetal urine flow rate, in the order of 50%, within 23 hours of

initiating indomethacin infusion (Figure 44). A similar reduction in urinary flow

rate, in the order of 55%, was observed within 2 hours of a 5 h indomethacin

infusion at an identical dose (Walker et al., 1992a). This effect on urinary flow rate

appears to be dose related. When indomethacin was infused at a rate of 0.017

mg/kg/min for 5 h, fetal urinary output significantly increased by an order of -85%

within 3 hours (Walker et al., 1992b). Further studies are being carried out to

elucidate the mechanism of oliguria and its physiological implications.

192

5 S U M M A R Y AND CONCLUSIONS

The following points summarise the salient features of the developed assay

method:

1. Simple one step liquid-liquid extraction procedure.

2. High efficiency fused silica capillary column technology.

3. Rapid run times (9.25 min)

4. Simple and rapid derivatisation with the relatively safer MTBSTFA.

5. Small sample volumes (0.1-1.0 mL).

6. Sensitivity of 1 ng/mL (Limit of quantitation, < 10% C.V., S/N >10).

7. Excellent chromatography with both the analyte, indomethacin, and the internal

standard, a-methyl indomethacin, eluting in regions free from endogenous

interference.

Overall the developed method offers several advantages over existing

analytical techniques for indomethacin in biological fluids, the most important

being: improved sensitivity, simpler sample preparation, improved selectivity, and a

smaller biological fluid sample requirement.

193

Application of the method to indomethacin measurement in plasma, urine,

tracheal and amniotic fluid samples was demonstrated by investigating its

disposition in the ovine fetus. The following observations were made:

1. The fetal plasma concentrations of indomethacin in the high dose animals were

similar to those seen during human tocolysis (Moise et al., 1990).

2. The low levels of indomethacin in amniotic fluid and fetal urine are consistent

with our observations for other organic acids (e.g., valproic acid,

diphenylmethoxyacetic acid) in the fetal lamb.

3. Unlike our observations with basic amine compounds (e.g., metoclopramide,

diphenhydramine), indomethacin was not observed to accumulate in tracheal

fluid.

4. The mean steady-state fetal femoral arterial concentration of indomethacin was

181.67 ± 45.05 ng/mL in the high dose group (n=5) and 41.29 ± 3.14 ng/mL in

the low dose group (n=2).

5. The arterio-venous concentration difference for indomethacin, at apparent steady

state, was minimal but statistically significant suggesting a low placental

permeability to indomethacin in sheep. This finding is consistent with

observations in other studies (Anderson et al., 1980; Harris and Vanpetten,

1981).

194

6. The fetal total body clearance of indomethacin was estimated to be 44.55 ±

11.14 mL/min (n=7).

7. The placental clearance of indomethacin as determined by the Fick principle was

calculated to be 8.85 ± 0.81 mL/min/kg.

1 9 5

6 REFERENCES

Abbott, F.S., Kassam, J., Acheampong, A., Ferguson, S., Panesar, S., Burton, R., Farrell, K., and Orr, J.: Capillary gas chromatographic-mass spectrometry of valproic acid metabolites in serum and urine using tert-butyldimethylsilyl derivatives. / Chromatogr. 375 (2): 285-298, 1986.

Ahuja, S.: Derivatisation in gas chromatography. JPharmSci. 65: 163-182,1976.

Al-Angary, A.A., El-Sayed, Y.M., Al-Meshal, M.A., Lutfi, K.M.: High-performance liquid chromatographic analysis of indomethacin in serum. / Clin Pharm Ther. 15: 257-265, 1990.

Alvan, G., Orme, M., Bertilsson, L., Ekstrand, R., and Palmer, L.: Pharmacokinetics of indomethacin. Clin Pharmacol Ther. 18: 364-373, 1975.

Alvan, G.: Individual differences in the disposition of drugs metabolised in the body•. Clin Pharmacokinet. 3: 155-175,1978.

Anderson, D.F., Phernetton, T.M., and Rankin, J.H.G.: Prediction of fetal drug concentrations. Am J Obstet Gynecol. 137: 735-738, 1980a.

Anderson, D.F., Phernetton, T.M., and Rankin, J.H.G.: The placental transfer of acetylsalicylie acid in near-term ewes. Am J Obstet Gynecol. 136: 814-818, 1980b.

Anderson, D.F., Phernetton, T.M., Rankin, J.H.G.: The measurement of placental drug clearance in near-term sheep: indomethacin. J Pharmacol Exp Ther. 213: 100-104, 1980c.

Anderson, D.F., Stock, M.K., and- Rankin, J.H.G.: Placental transfer of dexamethasone in near-term sheep. J Dev Physiol. 1: 431-436,1979.

Anderson, R.J., Berl, T., McDonald, K.M., Schrier, R.W., McCool, A., and Gilbert, L.: Evidence for an in-vivo antagonism between vasopressin and prostaglandin in the mammalian kidney. J Clin Invest. 56: 420-426, 1975.

Andersson, K. -E., Forman, A., and Ulmsten, U.: Pharmacology of labour. Clin Obstet Gynecol. 26(1): 56-77,1983.

Aoyagi, N., Ogata, H., Kaniwa, N., and Ejima, A.: Bioavailability of indomethacin capsules in humans. (II): Correlation with dissoluation rate. Int J Clin Pharmacol Ther Toxicol. 23: 529-534,1985.

196

Aoyagi, N., Ogata, H., Kaniwa, N., and Ejima, A.: Bioavailability of indomethacin capsules in humans. (I): Bioavailability and effects of gastric acidity. Int J Clin Pharmacol Ther Toxicol. 23: 469-47'4, 1985.

Aranda, J.V., Turmen, T., and Sasyniuk, B.I.: Pharmacokinetics of diuretics and methylxanthines in the neonate. Eur J Clin Pharmacol. 18: 55-63,1980.

Arbin, A.: Three alkylation methods for the determination of indomethacin in plasma by electron-capture gas chromatography. J Chromatogr. 144: 85-92, 1977.

Arcilla, R.A., Thilenius, O.G., and Ranniger, K.: Congestive heart failure from suspected ductal closure in utero. J Pediatr. 75: 74, 1969.

Armentrout, T., Katz, S., Thornburg, K.L., and Faber, J.J.: Osmotic flow through the placental barrier of chronically prepared sheep. Am J Physiol. 233: H 466-H 474, 1977.

Astier, A., and Renat, B.: Sensitive high-performance liquid chromatographic determination of indomethacin in human plasma, pharmacokinetic studies after single oral dose. J Chromatogr. (Biomed Appl.) 233: 279-288, 1982.

Austin, N.C., Pairandeau, P.W., Hames, T.K., and Hall, M.A.: Regional cerebral blood flow velocity changes after indomethacin infusion in preterm infants. Arch Disease Child. 67: 851-854, 1992.

Avgerinos, A and Malamataris, S: High-performance liquid chromatographic determination of indomethacin in human plasma and urine. J Chromatogr. (Biomed Appl.) 495: 309-313, 1989.

Baber, N., Halliday, L.D.C., van den Heuvel, W.J.A., Walker, R.W., Sibeon, R., Keenan, J.P., Littler, T., and Orme, M.L'e.: Indomethacin in rheumatoid arthritis: Clinical effects, pharmacokinetics, and platelet studies in responders and ononresponders. Annals Rheum Dis. 38: 128-137, 1979.

Baer, J.E., Hucker, H.B., and Duggan, D.E.: Bioavailability of indomethacin in man. Annals Clin Res. 6 (supp.ll): 44-46, 1974.

Baer, J.E.: Role of drug metabolism in drug research and development: basic considerations. J Pharm Sci. 61: 1674-1677,1972.

Bakke, O.M., and Haram, K.: Time-course of transplacental passage of diazepam: influence of injection-delivery interval on neonatal drug concentrations. Clin Pharmacokin. 7: 353-362, 1982.

Ballard, K.D., Knapp, D.R., Oatis, J.E.Jr., and Walle, T.: Resolution of the seven isomeric ring-hydroxylated propranolols as tert-butyldimethylsilyl derivatives by

197

capillary gas chromatography-mass spectrometry. J Chromatogr. 277: 333-339, 1983.

Bass, L.: Flow dependence of first-order uptake of substances by heterogenous perfused organs. J Theoret Biol. 86: 365-376,1980.

Basso, A., Fernandez, A., Althabe, O., Sabini, S., Piriz, H., and Belitzky, R.: Passage of mannitol from mother to amniotic fluid and fetus. Obstet Gynecol. 49: 628-631, 1977.

Bayne, W.F., East, T., Dye, D.: High-pressure liquid chromatographic method with postcolumn, in-line hydrolysis and fluorometric detection for indomethacin in biological fluids. JPharmSci. 70: 458-459, 1981.

Beaven, M.A., and Bayer, B.M.: Factors influencing the uptake and disposition of indomethacin-[14C] in cell cultures. Biochem Pharmacol. 29: 2055-2061,1980.

Beazley, J.M., and Gillespie, A.: Double-blind trial of prostaglandin E2 and oxytocin in the induction of labour. Lancet. 1: 152,1971.

Berl, T., Raz, A., Wald, H., Horowitz, J., and Czazkes, W.S.: Prostaglandin synthesis inhibition and the action of vasopressin: Studies in man and rat. Am J Physiol. 232: F529- F 537, 1977.

Berman, W., Friedman, Z., and Vidyasagar, D.: Pharmacokinetics of inhibitors of prostaglandin synthesis in the prenatal period. Sem Perinatol. 4: 67-72, 1980.

Bernstein, M.S., and Evans, M.A.: High-performance liquid chromatography-fluoresence analysis for indomethacin and metabolites in biological fluids. J Chromatogr. (Biomed Appl.) 229: 179-187,1982.

Besinger, R.E., Niebyl, J.R., Keyes, W.G., and Johnson, T.R.B.: Randomised comparative trial of indomethacin and ritodrine for the long-term treatment of preterm labour. Am J Obstet Gynecol: 164: 981-988,1991.

Betkerur, M.V., Yeh, T.F., Miller, K., Glasser, R.J., and Pildes, R.S.: Indomethacin and its effect on renal function and urinary kallikrein excretion in premature infants with patent ductus arteriosus. Pediatr. 68: 99-102, 1981.

Bhat, R., Vidyasagar, D., Fisher, E., Hastreiter, A., Ramirez, J.L., Burns, L., and Evans, M.: Pharmacokinetics of oral and intravenous indomethacin in preterm infants. Dev Pharmacol Ther. 1: 101-110,1980.

198

Bhat, R., Vidyasagar, D., Vadapalli, M., Whalley, C, Fisher, E., Hastreiter, A., Evans, M.: Disposition of indomethacin in preterm infants. J Pediatr. 95: 313-316, 1979.

Bianchetti, G., Monin, P., Marchal, F., Dubruc, C., Boutroy, M.J., Morselli, P.L., and Vert, P.: Pharmacokinetics of indomethacin in the premature infant. Dev Pharmacol Ther. 1: 111-124,1980.

Bilka, P.J., Wollheim, F., and Williams, R.C.Jr.: Indomethacin: a new antirheumatic agent. Minn Med. 47: 777-781,1964.

Blau, K., and Halket, J.: Handbook of derivatives for chromatography. John Wiley, New York, 1993.

Blum, I.: Indomethacin in hypercalcaemia. Lancet. 1: 866, 1975.

Bond, J.T., Bailie, M.D., and Hook, J.B.: Maturation of renal organic acid transport in vivo: substrate stimulation by penicillin. J Pharmacol Exp Ther. 199: 25-31, 1976.

Bonds, D.R., Anderson, S., and Meschia, G.: Transplacental diffusion of ethanol under steady state conditions. J Dev Physiol. 2: 409-416,1980.

Boyd, R.D.H., Haworth, C., Stacey, T.E., and Ward, R.H.T.: Permeability of the sheep placenta to unmetabolised polar non-electrolytes. J Physiol. 256: 617-634, 1976.

Brash, A.R., Hickey, D.E., Graham, T.P., Stahlman, M.T., Oates, J.A., and Cotton, R.B.: Pharmacokinetics of indomethacin in the neonate. Relation of plasma indomethacin levels to response of the ductud arteriosus. N Engl J Med. 305: 67-72, 1981.

Brash, A.R., Hickey, D.E., Graham, T.P., Stahlman, M.T., Oates, J.A., and Cotton, R.B.: Pharmacokinetics of indomethacin in the neonate. Relation of plasma indomethacin levels to response of the ductus arteriosus. ./V Engl J Med. 305: 67-72, 1981.

Brien, J.F., and Clarke, D.W.: Disposition and fetal effects of ethanol during pregnancy. In "Toxicologic and pharmacologic principles in pediatrics", Kacew, S., and Lock, S. (Eds.), Hemisphere, 1988.

British Pharmacopoeia, London, 1980.

British Pharmacopoeia, London, 1988.

199

Brown, Y.L., Kandrotas, R.J., Douglas, J.B., Gal, P.: High-performance liquid chromatographic determination of indomethacin serum concentrations. / Chromatogr. 459: 275-279, 1988.

Bryant, C, Smith, M.J.H., and Hines, W.J.W.: Effects of salicylate and y-resorcylate on the metabolism of radioactive succinate and fumarate by rat-liver mitochondria and on dehydrogenase enzymes. BiochemJ. 86: 391-396, 1963.

Buddingh, F., Parker, H.R., Ishizaki, G., and Tyler, W.S.: Long-term studies of the functional development of the fetal kidney in sheep. Am J Vet Res. 32: 1993-1998, 1971.

Buick, A.R., Doig, M.V., Jeal, S.C., Land, G.S., McDowall, R.D.: Method validation in the bioanalytical laboratory. J Pharm Biomed Anal. 8: 629-637, 1990.

Burrows, J.L.: Evaluation of calibration routines in pharmaceutical and bioanalytical separations, with reference to Monte Carlo simulations. / Pharm Biomed Anal. 11: 523-531, 1993.

Cabrol, D., Landesman, R., Muller, J., Uzan, M., Surean, C, and Saxena, B.B.: Treatment of polyhydramnios with prostaglandin synthetase inhibitor (indomethacin). Am J Obstet Gynecol. 157: 422-426, 1987.

Campbell, D.B.: The use of kinetic-dynamic interactions in the evaluation of drugs. Psychopharmacoldgy. 100: 433-450, 1990.

Cardone, M.J., Willavize, S.A., Lacy, M.E.: Method validation revisited: a chemometric approach. Pharm Res. 7: 154-160,1990.

Carlan, S.J., O'Brien, W.F., O'Leary, T.D., and Mastrogiannis, D.: Randomised comparative trial of indomethacin and sulindac for treatment of refractory preterm labour. Obstet Gynecol. 79 (2): 223-238, 1992.

Carmichael, M.C., and Kumar, R.: Molecular biology of vasopressin receptors. Sem Nephrol. 14: 341-348, 1994.

Carr, G.P., and Wahlich, J.C.: A practical approach to method validation in pharmaceutical analysis. J Pharm Biomed Anal. 8: 613-618,1990.

Casey, M.L., and MacDonald, P.C.: Bimolecular processes in the initiation of parturition: Decidual activation. Clin Obstet Gynecol. 31: 533, 1988.

Caturla, M.C., and Cusido, E.: Solid-phase extraction for the high-performance liquid chromatographic determination of indomethacin, suxibuzone,

200 phenylbutazone, and oxyphenbutazone in plasma, avoiding degradation of compounds. J Chromatogr (Biomed Appl). 581: 101-107,1992.

Causey, A.G., Hill, H.M., and Phillips, L.J.: Evaluation of criteria for the acceptance of bioanalytical data. J Pharm Biomed Anal. 8: 625-628, 1990.

Cerretani, D., Runci, F.M., Segre, G.: High-pressure liquid chromatographic (HPLC) determination of indomethacin in plasma. Eur Rev Med Pharmacol Sci. 10: 409-413, 1988.

Chamberlain, P.F., Manning, F.A., Morrison, I., et al.: Ultrasound evaluation of amniotic fluid volume. Am J Obstet Gynecol. 150: 250-254, 1984.

Chaudhari, G.N., Tipnis, H.P., Doshi, B.S., and Kulkarni, R.D.: Bioequivalence studies in humans of indomethacin capsules marketed in India. Eur J Clin Pharmacol. 26: 261-264, 1984.

Chiang, C.N., and Lee, C.C. (Eds.): Prenatal Drug Exposure: Kinetics and Dynamics. NIDA Research Monograph. 60: 1-147,1985.

Ching, M.S., Mihaly, G.W., Morgan, D.J., Date, N.M., Hardy, K.J., and Smallwood, R.A.: Low clearance of cimetidine across the human placenta. / Pharmacol Exp Ther. 241: 1006-1009, 1987.

Chiou, W.L., and Lam, G.: The significance of the arterial-venous plasma concentration difference in clearance studies. Int J Clin Pharmacol Ther Toxicol. 20: 197-203, 1982.

Chiou, W.L., Lam, G., Chen, M.-L., and Lee, M.G.: Instantaneous input hypothesis in pharmacokinetic studies. JPharmSci. 70: 1037-1039,1981.

Chremos, A.N., Shen, D., Gibaldi, M., Proctor, J.D., and Newman, J.H.: Time-dependent change in renal clearance of bethanidine in humans. J Pharm Sci. 65: 140-142, 1976.

Cifuentes, R.F., Olley, P.M., Balfe, J.W., Raddle, I.C., and Soldin, S.J.: Indomethacin and renal function in premature infants with persistent patent ductus arteriosus. J Pediatr. 95: 583-587,1979.

Clarke, D.W., Patrick, J., Wlodek, M.E., Smith, G.N., Richardson, B., and Brien, J.F.: The role of fetal urinary excretion in the transfer of ethanol into amniotic fluid after maternal adminstration of ethanol to the near-term pregnant ewe. Can J Physiol Pharmacol. 65: 1120-1124,1987.

201

Clarke, G.S.: The validation of analytical methods for drug substances and drug products in UK pharmaceutical laboratories. / Pharm Biomed Anal. 12: 643-652, 1994.

Clench, J., Reinberg, A., Dziewanowska, Z., Ghata, J., and Smolensky, M.: Circardian changes in the bioavailability and effects of indomethacin in healthy subjects. Eur J Clin Pharmacol. 20: 359-369,1981.

Clozel, M., Behary, K., and Aranda, J.V.: Indomethacin metabolism in liver microsomes during postnatal development in the rat. Biol Neonate. 50: 83-90, 1986.

Clyman, R.I., Mauray, F., Rudolph, A.M., and Heymann, M.A.: Age-dependent sensitivity of the lamb ductus arteriosus to indomethacin and prostaglandins. J Pediatr. 96: 94-98, 1980.

Coceani, F., and Olley, P.M.: The response of the ductus arteriosus to prostaglandins. Can J Physiol Pharmacol. 51: 220-225, 1973.

Coceani, F., Bishai, I., White, E., Bodach, E., and Olley, P.M.: Action of prostglandins, endoperoxides, and thromboxanes on the lamb ductus arteriosus. Am J Physiol. 234: H 117-H 122, 1978a.

Coceani, F., Olley, P.M., and Bodach, E.: Lamb ductus arteriosus: Effect of prostaglandin synthesis inhibitors on the muscle tone and the response to prostaglandin E2. Prostaglandins. 9: 299-308, 1975.

Coceani, F., Olley, P.M., and Lock, J.E.: Prostaglandins, ductus arteriosus, pulmonary circulation: current concepts and clinical potential. Eur J Clin Pharmacol. 18: 75-81, 1980.

Coceani, F., Olley, P.M., Bishai, I., Bodach, E., and White, E.P.: Significance of the prostaglandin system to the control of muscle tone of the ductus arteriosus. In: Coceani, F., and Olley, P.M. (Eds.). Prostaglandins and Perinatal Medicine, Advances in Prostaglandin and Thromboxane Research, Volume 4, Raven Press, New York, pp. 325-333, 1978b.

Colditz, P., Murphy, D., Rolfe, P., and Wilkinson, A.R.: Effect of infusion rate of indomethacin on cerebrovascular responses in preterm neonates. Arch Disease Child. 64: 8-12, 1989.

Collins, J.M., and Dedrick, R.L.: Contribution of lungs to total body clearance: linear and non-linear effects. J Pharm sci. 71: 66-70, 1982.

202 Comline, R.S., and Silver, M.: Recent observations on the undisturbed fetus in utero and its delivery. In "Recent advances in physiology", Linden, R.J. (Ed.). Churchill Livingstone, Edinburgh, pp. 406, 1974.

Conn, R.B., Sabo, A.J., Landes, D., and Ho, J.Y.L.: Inconstancy of renal clearance values with changing plasma concentrations. Nature. 203: 143-146,1964.

Cook, J.A., and Smith, D.E.: Theoretical limits of changes in plasma protein binding or renal clearance. J Pharm Sci. 74: 108-109,1985.

Cooper, J.K., McKay, G., Hawes, E.M., Midha, K.K.: High-performance liquid chromatographic assay of indomethacin and its application in pharmacokinetics in healthy volunteers. J Chromatogr. (Biomed Appl.) 233: 289-296, 1982.

Cowan, F.: Indomethacin, patent ductus arteriosus, and cerebral blood flow. J Pediatr. 109: 341-344, 1986.

Curry, S.H., Brown, E.A., Kuck, H., and Cassin, S.: Preparation and stability of indomethacin solutions. Can J Physiol Pharmacol. 60: 988-992,1982.

Czuba, M.A., Morgan, D.J., Ching, M.S., Mihaly, G.W., Ghabrial, H., Hardy, K.J., and Smallwood, R.A.: Disposition of the diastereoisomers quinine and quinidine in the ovine fetus. J Pharm Sci. 80: 445-448, 1991.

Das Gupta, V., Gibbs, C.W.Jr., and Ghanekar, A.G.: Stability of pediatric liquid dosage forms of ethacrynic acid, indomethacin, methyldopate hydrochloride, prednisone, and spironolactone. Am J Hosp Pharm. 35: 1382-1385,1978.

Dawes, G.S.: Foetal and Neonatal Physiology. Year Book Medical Publishers, Chicago, 1968.

De Vos, A., Vervoort, L., and Kinget, R.: Release of indomethacin from transparent oil-waters gels. J Pharm Sci. 83: 641-643,1994.

de Vries, J.I.P., Visser, G.H.A., and Prechtl, H.F.B.: The emergence of fetal behaviour. II Quantitative aspects. Early Hum Dev. 12: 99-120,1985.

de Vries, J.I.P., Visser, G.H.A., and Prechtl, H.F.R.: The emergence of fetal behaviour. I Qualitative aspects. Early Hum Dev. 7: 301-322,1982.

De Zeeuw, D., Brater, D.C., Jacobson, H.R.: Micro-HPLC and transfer techniques for directly measuring drug (indomethacin) transport across the isolated perfused renal proximal straight tubule. J Pharmacol Meth. 19: 275-282,1988.

De Zeeuw, D., Leinfelder, J.L., Bruter, D.C.: Highly sensitive measurement of indomethacin using a high-performance liquid chromatographic technique

203 combined with post-column in-line hydrolysis. J Chromatogr. (Biomed Appl.) 380: 157-162, 1986.

De Zeeuw, D., Jacobson, H.R., and Brater, D.C.: Indomethacin secretion in the isolated perfused proximal straight rabbit tubule. Evidence for two parallel transport mechanisms. J Clin Invest. 81: 1585-1592,1988.

Dindogru, A., Gailani, S., Henderson, E.S., Wallace, H.J.Jr., and Fitzpatrick, J.: Indomethacin in hypercalcaemia. Lancet. 2: 365, 1975.

Dittrich, P., Kohler, D.G., Primbs, P., and Kukovetz, W.R.: Pharmacokinetics of indomethacin i.m. in blood, synovial fluid, synovial membrane, muscle, fat, bone, and spinal cord. Eur J Rheum Inflamm. 7: 45-50, 1984.

Donker, A.J., de Jong, P.E., van Eps, L.W., Brentjens, J.R., Bakker, K., and Doorenbos, H.: Indomethacin in Bartter's syndrome: Does the syndrome represent a state of hyperprostaglandinism? Nephron. 19(4): 200-213,1977.

Donnelly, P., Lloyd, K., and Campbell, H.: Indomethacin in rheumatoid arthritis: An evaluation of its anti-inflammatory and side effects. BrMedJ. 1: 69-75,1967.

Douidar, S.M., Richardson, J., and Snodgrass, W.R.: Role of indomethacin in ductus closure: An update evaluation. Dev Pharmacol Ther. 11(4): 196-212, 1988.

Dousa, T.P.: Cyclic-3', 5'-nucleotide phosphodiesterases in the cyclic adenosine monophosphate (cAMP)-mediated actions of vasopressin. Sem Nephrol. 14: 333-340, 1994.

Dudley, D.K.L., and Hardie, M.J.: Fetal and neonatal effects of indomethacin used as a tocolytic agent. Am J Obstet Gynecol. 151: 181-184, 1985.

Duggan, D.E., and Kwan, K.C.: Enterohepatic recirculation of drugs as a determinant of therapeutic ratio. Drug Metab Rev, 9: 21-41,1979.

Duggan, D.E., Hogans, A.F., Kwan, K.C., and McMahon, F.G.: The metabolism of indomethacin in man. J Pharmacol Exp Ther. 181: 563-575,1972.

Duggan, D.E., Hooke, K.F., Noll, R.M., and Kwan, K.C.: Enterohepatic circulation of indomethacin and its role in intestinal irritation. Biochem Pharmacol. 25: 1749-1754, 1975.

Dutton, G.J.: Developmental aspects of drug conjugation with special reference to glucuronidation. Ann Rev Pharmacol Toxicol. 18: 17-35, 1978.

204 Dutton, G.J.: Glucuronide synthesis in fetal liver and other tissues. Biochem J. 71: 141-148, 1959.

Dvorchik, B.H.: Drug disposition during pregnancy. Biol Res Pregnancy. 3: 129-137, 1982.

Ebbehoj, N., Friis, J., Svendsen, L.B., Bulow, S., and Madsen, P.: Indomethacin treatment of acute pancreatitis. A controlled double-blind trial. Scand J Gastroenterol. 20 (7): 798-800, 1985.

Eckard, R.: Amuno S: Klinisch-pharmakologische grundlagen. Akt Rheumatol. 7: 20, 1982.

Edwards, A.D., Wyatt, J.S., Richardson, C, Potter, A., Cope, M., Delpy, D.T., and Reynolds, E.O.R.: Effects of indomethacin on cerebral hemodynamics in very preterm infants. Lancet. 335: 1491-1495,1990.

Ekstrand, R., Alvan, G., L'e Orme, M., Lewander, R., Palmer, L., and Sarby, B.: Double-blind dose-response study of indomethacin in rheumatoid arthritis. Eur J Clin Pharmacol. 17 (6): 437-442, 1980.

Elbourne, I., Lumbers, E.R., and Hill, K.J.: The secretion of organic acids and bases by the ovine fetal kidney. Experimental Physiol. 75: 211-221,1990.

Elbourne, I., Lumbers, E.R., and Hill, K.J.: The secretion of organic acids and bases by the ovine fetal kidney. Exp Physiol. 75: 211-221,1990.

Elder, M.G., and Kapadia, L.: Indomethacin in the treatment of primary dysmenorrhoea. Br J Obstet Gynecol. 86: 645-647,1979.

Elliott, G.T., and McKenzie, M.W.: Treatment of hypercalcemia. Drug Intel Clin Pharm. 17 (1): 12-22, 1983.

Embrey, M.P.: Induction of abortion by prostaglandins El, E2. Br Med J. 2: 256, 1970.

Emori, H.W., Champion, G.D., Bluestone, R., and Paulus, H.E.: Simultaneous pharmacokinetics of indomethacin in serum and synovial fluid. Ann Rheum Dis. 32: 433-435, 1973.

Engervall, P., Bjorkholm, M., Grimfors, G., and Holm, G.: Antipyretic effect of indomethacin in malignant lymphoma. Acta Med Scand. 219: 501-505, 1986.

Eronen, M., Pesonen, E., Kurki, T., Ylikorkala, O., and Hallman, M.: The effects of indomethacin and a -sympathomimetic agent on the fetal ductus arteriosus during

205 treatment of premature labour: a randomised double-blind study. Am J Obstet Gynecol. 164: 141-146, 1991.

Eronen, M.: The hemodynamic effects of antenatal indomethacin and (3-sympathomimetic agent on the fetus and the newborn. A randomised study. Pediatr Res. 33: 615-619, 1993.

Evans, D.H., Levene, M.I., and Archer, L.N.J.: The effect of indomethacin on cerebral blood-flow velocity in premature infants. Dev Med Child Neurol. 29: 776-782, 1987.

Evans, M.A., Bhat, R., Vidyasagar, D., Vadapalli, M., Fisher, E., and Hastreiter, A.: Gestational age and indomethacin elimination in the neonate. Clin Pharmacol Ther. 26: 746-751, 1979.

Evans, M.A., Papazafiratou, C., Bhat, R., Vidyasagar, D.: Indomethacin metabolism in isolated neonatal and fetal rabbit hepatocytes. Pediatr Res. 15: 1406-1410, 1981.

Evans, M.A.: GLC microdetermination of indomethacin in plasma. J Pharm Sci. 69: 19-220, 1980.

Evershed, R.P.: Advances in silylation. In "Handbook of derivatives for chromatography", Blau, K., and Halket, J (Eds.). John Wiley, New York, pp. 51-108, 1993.

Feely, J., and Wood, A.J.J.: Effect of inhibitors of prostaglandin synthesis on hepatic drug clearance. Br J Clin Pharmacol. 15: 109-111,1983.

Feigen, L.P., Klainer, E., Chapnick, B.M., and Kadowitz, P.J.: The effect of indomethacin on renal function in pentobarbital-anesthetised dogs. J Pharmacol Exp Ther. 198: 457-463, 1976.

Fejes-To'th, G., Magyar, A., and Walter, J.: Renal response to vasopressin after inhibition of prostaglandin synthesis. Am J Physiol. 232: F416-F423, 1977.

Ferreira, S.H., and Vane, J.R.: New aspects of the mode of action of nonsteroidal antiinflammatory drugs. Ann Rev Pharmacol. 14: 57-73, 1974.

Ferreira, S.H., Moncada, S., and Vane, J.R.: Indomethacin and aspirin abolish prostaglandin release from the spleen. Nature New Biol. 231: 237-239, 1971.

Ferry, D.G., Ferry, D.M., Moller, P.w., McQueen, E.G.: Indomethacin estimation in plasma and serum by electron capture gas chromatography. J Chromatogr. 89: 110-112,1974.

206

Flower, R.J.: Drugs which inhibit prostaglandin biosynthesis. Pharmacol Rev. 26: 33-67, 1974.

Fowler, J., Gallagher, B.M., MacGregor, R.R., and Wolf, A.P.: Carbon-11 labeled aliphatic amines in lung uptake and metabolism studies. J Pharmacol Exp Ther. 198: 133-145, 1976.

Friedman, N.F., Heymann, M.A., and Rudolph, A.M.: Commentary: New thoughts on an old problem-patent ductus arteriosus in the preterm infant. / Pediatr. 90: 338, 1977.

Friedman, W.F., Hirschklau, M.J., Printz, M.J., Pitlick, P.T., and Kirkpatrick, S.E.: Pharmacological closure of patent ductus arteriosus in the premature infant. New Engl J Med. 295: 526-529,1976.

Friedman, W.F.: Studies of responses of the ductus arteriosus in intact animals. In "The ductus arteriosus", Heymann, M.A., and Rudolph, A.M. (Eds.). Report of the seventy-fifth Ross conference on pediatric research. Ross Laboratories, Columbus, Ohio, pp. 35-43, 1978b.

Friedman, Z., and Berman, W.: Hematologic effects of prostaglandins and thromboxanes and inhibitors of their synthesis in the perinatal period. Sem Perinatol. 4: 73, 1980.

Friedman, Z., Whitman, V., Maisels, M.J., Berman, W., Marks, K.H., and Vesell, E.S.: Indomethacin disposition and indomethacin induced platelet dysfunction in premature infants. J Clin Pharmacol. 18: 272-279, 1978.

Gal, P., Ransom, J.L., Weaver, R.L., Schall, S., Wyble, L.E., Carlos, R.Q., and Brown, Y.: Indomethacin pharmacokinetics in neonates: the value of volume of distribution as a marker of permanent patent ductus arteriosus closure. Ther Drug Monit. 13: 42-45, 1991.

Garrett, E.R.: Pharmacokinetics and clearances related to renal processes. Int J Clin Pharmacol. 16: 155-172,1978.

Gay, J.H., Daily, W.J.R., Meyer, B.H.P., Trump, D.S., Cloud, D.T., and Moltham, M.E.: Ligation of the patent ductus arteriosus in premature infants: Report of 45 cases. J Pediatr Surg. 8(5): 677-683,1973.

Gerlowski, L.E., and Jain, R.K.: Physiologically based pharmacokinetic modelling: principles and applications. J Pharm Sci. 72: 1103-1127,1983.

Ghanem, A-H., El-Sabbagh, H., Abdel-Alim, H.: Stability of indomethacin solubilised system. Pharmazie. 34: 406-407, 1979.

207 Giachetti, C., Canali, S., Zanolo, G.: Separation of non-steroidal anti-inflammatory agents by high-resolution gas chromatography. Preliminary trials to perform pharmacokinetic studies. J Chromatogr. 279: 587-592, 1983.

Gibaldi, M., and Koup, J.R.: Pharmacokinetic concepts-drug binding, apparent volume of distribution, and clearance. Eur J Clin Pharmacol. 20: 299-305, 1981.

Gibaldi, M., Levy, G., and McNamara, P.J.: Effect of plasma protein and tissue binding on the biologic half-life of drugs. Clin Pharmacol Ther. 24: 1-4,1978.

Gillespie, W.R., and Veng-Pedersen, P.: Theorems and implications of model independent elimination/distribution function decomposition of linear and some non-linear drug dispositions. II Clearance concepts applied to the evaluation of distribution kinetics. J Pharmacokinet Biopharm. 13: 441-451, 1985.

Gines, P., Abraham, W.T., and Schrier, R.W.: Vasopressin in pathophysiological states. Sem Nephrol. 14: 384-397, 1994.

Gleason, C.A., Clyman, R.I., Heymann, M.A., Mauray, F., Leake, R., and Roman, C: Indomethacin and patent ductus arteriosus: Effect on renal function in preterm lambs. Am J Physiol. 254(23): F38-F44, 1988.

Goldenberg, R.L., Davis, R.D., and Baker, R.C.: Indomethacin-induced oligohydramnios. Am J Obstet Gynecol. 160: 1196-1197,1989.

Gordin, A., Sotka, S., and Nuutila, J.: Comparison of a slow-release indomethacin tablet and naproxen in osteoarthritis. Curr Med Res Opinion. 9(7): 500-504, 1985.

Gordon, J.G.: personal communication, 1994.

Grantham, J.J., and Orloff, J.: Effect of prostaglandin E l on the permeability response of the isolated collecting tubule to vasopressin, adenosine 3', 5'-monophosphate and theophylline. J Clin Invest. 47: 1154-1161,1968.

Green, K., Bowman, K., Luxenberg, M.N., and Friberg, T.R.: Penetration of topical indomethacin into phakic and aphakic rabbit eyes. Arch Ophthal. 101: 284-288, 1983.

Green, T.P., O'Dea, R.F., and Mirkin, B.L.: Determinants of drug disposition and effect in the fetus. Ann Rev Pharmacol Toxicol. 19: 285-322, 1979.

Grella, P., and Zanor, P.: Premature labor and indomethacin. Prostaglandins. 16: 1007-1017, 1978.

208

Guissou, P., Cuisinaud, G., Legheand, J., and Sassard, J.: Chronopharmacokinetics of indomethacin in rats. Arzneim Forsch. 37: 1034-1037, .1987.

Guissou, P., Cuisinaud, G., Sassard, J: Gas chromatographic determination of indomethacin and its O-desmethylated metabolite in human plasma and urine. / Chromatogr. (Biomed Appl.) 277: 368-373,1983.

Gullner, H.G., Gill, J.R., Bartler, F.C., and Dusing, R.: The role of the prostglandin system in the regulation of renal function in normal women. Am J Med. 69: 718-724, 1980.

Gund, P., and Shen, T.Y.: A model for the prostaglandin synthetase cyclooxygenation site and its inhibition by antiinflammatory arylacetic acids. J Med Chem. 20: 1146-1152, 1977.

Hall, S., and Rowland, M.: Influence of fraction unbound upon the renal clearance of furosemide in the isolated perfused rat kidney. / Pharmacol Exp Ther. 232: 263-268, 1985.

Hall, S., and Rowland, M.: Relationship between renal clearance, protein binding, and urine flow for digitoxin, a compound of low clearance in the isolated perfused rat kidney. J Pharmacol Exp Ther. 228: 174-179,1983.

Hallak, M., Reiter, A.A., Ayres, N.A., and Moise, K.J.Jr.: Indomethacin for preterm labor: fetal toxicity in a dizygotic twin gestation. Obstet Gynecol. 78: 911-913, 1991.

Halushka, P.V., Wohltmann, H., Privitera, P.J., Hurwitz, G., and Margolius, H.S.: Bartter's syndrome: Urinary prostaglandin E-like material and kallikrein: indomethacin effects. Annals Intern Med. 87 (3): 281-286,1977.

Ham, E.A., Cirillo, K.J., Zanetti, M., Shen, T.Y., and Kuehl, F.A.: Studies on the mode of action of non-steroidal, anti-inflammatory agents. In Prostaglandins in Cellular Biology, Ramwell, P.W., and Pharriss, B.B. (Eds), Plenum Press, New York, pp. 345-352, 1972.

Hanna, C, and Sharp, J.D.: Ocular absorption of indomethacin by the rabbit. Arch Ophthal. 88: 196-198, 1972.

Hare, L.E., Ditzler, C.A., Duggan, D.E.: Radioimmunoassay of indomethacin in biological fluids. J Pharm Sci. 66: 486-489,1977.

Harman, R.E., Meisinger, M.A.P., Davis, G.E., and Kuehl, F.A.: The metabolites of indomethacin, a new antiinflammatory drug. J Pharmacol Exp Ther. 143: 215-220, 1964.

209 Harris, W.H., and Van Petten, G.R.: Placental transfer of indomethacin in the rabbit and sheep. Can J Physiol Pharmacol. 59: 342-346,1981.

Harth, M., and Bondy, D.C.: Indomethacin and acetylsalicylic acid in the treatment of osteoarthritis of the hips and knees. Can Med Ass J. 101 (6): 311-316, 1969.

Heijden, A.J.v.d., Provoost, A.P., Nauta, J., Grose, W., Oranje, W.A., Wolff, E.D., and Sauer, P.J.J.: Renal functional impairment in preterm neonates related to intrauterine indomethacin exposure. Pediatr Res. 24: 644-648, 1988.

Hekman, P., and Van Ginneken, C.A.M.: Kinetic modelling of the renal excretion of isopyracet in the dog. J Pharmacokinet Biopharm. 10: 77-92, 1982.

Helleberg, L.: Clinical pharmacokinetics of indomethacin. Clin Pharmacokin. 6: 245-258, 1981.

Helleberg, L.: Determination of indomethacin in serum and urine by electron-capture gas-liquid chromatography. J Chromatogr. 117: 167-173,1976.

Hendricks, S.K., Smith, J.R., Moore, D.E., and Brown, Z.A.: Oligohydramnios associated with prostaglandin synthetase inhibitors in preterm labour. Br J Obstet Gynecol. 97: 312-316, 1990.

Hickok, D.E., Hollenbach, K.A., Reilley, S.F., and Nyberg, D.A.: The association between decreased amniotic fluid volume and treatment with non-steroidal anti­inflammatory agents for preterm labour. Am J Obstet Gynecol. 160: 1523-1531, 1989.

Hill, K.J., and Lumbers, E.R.: Renal function in adult and fetal sheep. / Dev Physiol. 10: 149-159, 1988.

Holmlund, D., and Sjodin, J-G.: Treatment of ureteral colic with intravenous indomethacin. / Urol. 120: 676-677, 1978.

Honore, B., Brodersen, R., and Robertson, A.: Interaction of indomethacin with adult human albumin and neonatal serum. Dev Pharmacol Ther. 6: 347-355,1983.

Hooper, S.B., Harding, R., Deayton, J., and Thorburn, G.D.: Role of prostaglandins in the metabolic responses of the fetus to hypoxia. Am J Obstet Gynecol. 166: 1568-1575, 1992.

Hori, R., Sunayashiki, K., and Kamiya, A.: Pharmacokinetic analysis of renal handling of sulphamethizole. J Pharm Sci. 65: 463-465, 1976.

210

Hubert, Ph., Renson, M., Crommen, J: A fully automated high-performance liquid chromatographic method for the determination of indomethacin in plasma. J Pharm Biomed Anal. 7: 1819-1827, 1989.

Hucker, H.B., Stauffer, S.C., White, S.D., Rhodes, R.E., Arison, B.H., Umbenhauer, E.R., Bower, R.J., McMahon, F.G.: Physiologic disposition and metabolic fate of a new anti-inflammatory agent, cis-5-fluoro-2-methyl-l-[p-(methylsulfinyl)-benzylidenyl]-indene-3-acetic acid in the rat, dog, rhesus monkey, and man. Drug Metab Disp. 1: 721-736, 1973.

Hucker, H.B., Zacchei, A.G., Cox, S.V., Brodie, D.A., and Cantwell, N.H.R.: Studies on the absorption, distribution, and excretion of indomethacin in various species. J Pharmacol Exp Ther. 153: 237-249,1966.

Huszar, G.: Physiology of the myometrium, Chapter 7. In: Maternal-fetal medicine: Principles and practice. Creasy, R.K., and Resnik, R., pp. 141, 1989.

Hvidberg, E., Lausen, H.H., and Jensen, J.A.: Indomethacin. Plasma concentrations and protein binding in man. Eur J Clin Pharmacol. 4: 119-124, 1972.

Hyver, K.J., and Sandra, P.: High resolution gas chromatography. Hewlett-Packard, Avondale, 1989.

Inagi, T., Muramatsu, T., Nagai, H., et al.: Influence of vehicle composition on the penetration of indomethacin through guinea pig skin. Pharm Bull. 29: 1708-1714, 1981.

Jansen, A.H., De Boeck, C, Ioffe, S., and Chernick, V.: Indomethacin-induced fetal breathing: Mechanism and site of action. J Appl Physiol. 57: 360-365,1984.

Jensen, K.M.: Determination of indomethacin in serum by an extractive alkylation technique and gas-liquid chromatography. / Chromatogr. 153: 195-202, 1978.

John, E.G., Vasan, U., Hastreiter, A.R., Bhat, R., and Evans, M.A.: Intravenous indomethacin and changes in renal function in premature infants with patent ductus arteriosus. Pediatr Pharmacol. 4: 11-19, 1984.

Johnson, A.G., and Ray, J.E.: Improved high-performance liquid chromatographic method for the determination of indomethacin in plasma. Ther Drug Monit. 14: 61-65, 1992.

Juchau, M.R., Chao, S.T., and Omiecinski, C.J.: Drug metabolism by the human fetus. Clin Pharmacokinet. 5: 320-339, 1980.

211 Kahns, A.H., Jensen, P.B., Mork, N., and Bundgaard, H.: Kinetics of hydrolysis of indomethacin and indomethacin ester prodrugs in aqueous solutions. Acta Pharm Nord. 1 (6): 327-336, 1989.

Kamiya, A., Okumura, K., and Hori, R.: Quantitative investigation on renal handling of drugs in rabbits, dogs, and humans. J Pharm Sci. 72: 440-443, 1983.

Kanto, J.H.: Use of benzodiazepines during pregnancy, labour, and lactation with particular reference to pharmacokinetic considerations. Drugs. 23: 354-380, 1982.

Karim, S.M.M., and Derlin, J.: Prostaglandin content of amniotic fluid during pregnancy and labour. J Obstet Gynecol Br Commonw. 74: 230,1967.

Karim, S.M.M., Trussel, R.R., Hiller, K., and Patel, R.C.: Induction of labour with prostaglandin F2a- J Obstet Gynecol Br Commonw. 76: 769, 1969.

Karim, S.M.M.: Appearance of prostaglandin F 2 a in human blood during labour. BrMedJ. 4: 618, 1968. Karnes, H.T., and March, C: Calibration and validation of linearity in chromatographic biopharmaceutical analysis. J Pharm Biomed Anal. 9: 911-918, 1991.

Kass, E: Indomethacin in ankylosing spondylitis. Acta Rheum Scand. 11: 205-211, 1965.

Katz, M., and Creasy, R.K.: Uterine blood flow distribution after indomethacin infusion in the pregnant rabbit. Am J Obstet Gynecol. 140: 430-434, 1981.

Kazemifard, A.G., and Moore, D.E.: Liquid chromatography with amperometric detection for the determination of non-steroidal anti-inflammatory drugs in plasma. J Chromatogr. {Biomed Appl.) 533: 125-132, 1990.

Kennedy, R.L., Bell, J.U., Miller, R.P., Doshi, D., de Sousa, H., Kennedy, M.J., Heald, D.L., Bettinger, R., and David, Y.: Uptake and distribution of lidocaine in fetal lambs. Anesthesiol. 72: 483-489, 1990.

Kiely, J.M.: Symptomatic control of fever in lymphoma and leukemia. Mayo Clin Proc. 44: 272-280, 1969.

Kier, L.B., and Whitehouse, M.W.: Similarities in the interatomic distances of some anti-inflammatory agents and inflammagenic amines: A possible insight into their common receptors. J Pharm Pharmacol. 20: 793-795,1968.

212 Kill, F.: Dynamics of renal proximal tubular secretion. Nature. 189: 927-928, 1961.

Kim, K.R., Hahn, M.K., Zlatkis, A., Horning, E.C., and Middleditch, B.S.: Simultaneous gas chromatography of volatile and non-volatile carboxylic acids as tert-butyldimethylsilyl derivatives. J Chromatogr. 468: 289-301,1989.

Kirshon, B., Mari, G., and Moise, K.J.Jr.: Indomethacin therapy in the treatment of symtomatic polyhydramnios. Obstet Gynecol. 75: 202-205, 1990.

Kirshon, B., Moise, K.J.Jr., Mari, G., and Willis, R.: Long-term indomethacin therapy decreases fetal urine output and results in oligohydramnios. Am J Perinatol. 8: 86-88, 1991.

Kirshon, B., Moise, K.J.Jr., Wasserstum, N., Ou, C.-N., and Huhta, J.C.: Influence of short-term indomethacin therapy on fetal urine output. Obstet Gynecol. 72: 51-53, 1988.

Kirshon, B.: Fetal urine output in hydramnios. Obstet Gynecol. 73: 240-242, 1989.

Kitterman, J.A., Liggins, G.C., Clements, J.A., and Tooley, W.H.: Stimulation of breathing movements in fetal sheep by inhibitors of prostaglandin synthesis. / Dev Physiol. 1: 453-466, 1979.

Kitterman, J.A., Liggins, G.C., Fewell, J.E., and Tooley, W.H.: Inhibition of breathing movements in fetal sheep by prostaglandins. J Appl Physiol. 54: 687-692, 1983.

Klein, K.L., Scott, W.J., Clark, K.E., and Wilson, J.G.: Indomethacin-placental transfer, cytotoxicity, and teratology in the rat. Am J Obstet Gynecol. 141: 448-452, 1981.

Kleinman, L.I.: The Kidney. In "Perinatal physiology", Stave, U (Ed.). Plenum, New York, chapter 27, pp. 589-616, 1978.

Knepper, M.A., Nielsen, S., Chiou, C.-L., and DiGiovanni, S.R.: Mechanism of vasopressin action in the renal collecting duct. Sem Nephrol. 14: 302-321, 1994.

Kong, L.J., Garrett, W.N., and Rattray, P.V.: A description of the dynamics of fetal growth in sheep. J Animal Sci. 41: 1065-1243,1975.

Koppel, R., and Rabinovitch, M.: Regulation of fetal lamb ductus arteriosus smooth muscle migration by indomethacin and dexamethasone. Pediatr Res. 33: 352-358, 1993.

213

Kover, G.: The effect of indomethacin on renal function. Acta Physiol Acad Sci Hung. 55: 283-298, 1980.

Koyima, I.: Urine flow dependence of renal clearance and inter-relation of renal reabsorption and physicochemical properties of drugs. Drug Metab Disp. 14: 239-245, 1986.

Krauer, B., and Krauer, F.: Drug kinetics in pregnancy. Clin Pharmacokin. 2: 167-181, 1977.

Krauer, B., Krauer, F., and Hytten, F.E.: Drug disposition and pharmacokinetics in the maternal-placental-fetal unit. Pharmacol Ther. 10: 301-328, 1980.

Ku, E.C., and Wasvary, J.M.: Inhibition of prostaglandin synthetase by Su-21524. Fed Proc. 32: 3302, 1973.

Kuehl, F.A.Jr.: Prostaglandin research: clinical implication. Sem Arthritis Rheumat. 12 (supp.l): 119-125,1982.

Kullama, L.K., Ross, M.G., Lam, R., Leake, R.D., Ervin, M.G., and Fisher, D.A.: Ovine maternal and fetal renal vassopressin receptor response to maternal dehydration. Am J Obstet Gynecol. 167: 1717-1722,1992.

Kulmacz, R.J., Ren, Y., Tsai, A.L., and Palmer, G.: Prostaglandin H synthetase: Spectroscopic studies of the interaction with hydroperoxides and with indomethacin. Biochemistry. 29: 8760-8771, 1990.

Kumar, R.: Endocytosis of the vasopressin receptor. Sem Nephrol. 14: 357-367, 1994.

Kwan, K.C., Breault, G.O., Umbenhauer, E.R., McMahon, F.G., and Duggan, D.E.: Kinetics of indomethacin absorption, elimination, and enterohepatic circulation in man. J Pharmacokin Biopharm. 4: 255-280,1976.

Kwong, E., Pillai, G.K., McErlane, K.M.: HPLC analysis of indomethacin and its impurities in capsule and suppository formulations. J Pharm Sci. 71: 828-830, 1982.

Lam, G., and Chiou, W.L.: Determination of renal clearances using arterial and venous plasma: procainamide in rabbits. J Pharm Sci. 70: 1373-1375,1981.

Lang, J.R., and Bolton, S.: A comprehensive method validation strategy for bioanalytical applications in the pharmaceutical industry-1. experimental considerations. J Pharm Biomed Anal. 9: 357-361,1991.

214

Lang, J.R., and Bolton, S.: A comprehensive method validation strategy for bioanalytical applications in the pharmaceutical industry-2. statistical analyses. / Pharm Biomed Anal. 9: 435-442, 1991.

Lawrence, J.R., Bryson, S.M., Sumner, D.J., Campbell, B.C., and Whitby, B.: The renal clearance of disopyramide after bolus intravenous injection. Biopharm Drug Dispos. 1: 51-57, 1980.

Lebedevs, T.H., Wojnar-Horton, R.E., Yapp, P., Roberts, M.J., Dusci, L.J., Hackett, L.P., and Ilett, K.F.: Excretion of indomethacin in breast milk. Br J Clin Pharmac. 32:751-754,1991.

Leffler, C.W., Busija, D.W., Fletcher, A.M., et al.: Effects of indomethacin upon cerebral hemodynamics in newborn piglets. Pediatr Res. 19: 1160-1164, 1985.

Leffler, C.W., Mirro, R., Shibata, M., Parfenova, H., Armstead, W.M., and Zuckerman, S.: Effects of indomethacin on cerebral vasodilator responses to arachidonic acid and hypercapnia in newborn pigs. Pediatr Res. 33: 609-614, 1993.

Leichtweiss, H.-P.: Carrier-mediated placental transfer. Placenta (supp.l): 115-124, 1981.

Levenson, D.J., Simmons, C.E.Jr., and Brenner, B.M.: Arachidonic acid metabolism, prostaglandins, and the kidney. Am J Med. 72: 354-374,1982.

Levy, G.: Effect of plasma portein binding on renal clearance of drugs. J Pharm Sci. 69: 482-483, 1980.

Levy, G.: Pharmacokinetics of fetal and neonatal exposure to drugs. Obstet Gynecol. 58: 9S-16S, 1981.

Lewis, S., and Yudkoff, M.: Gas chromatographic-mass spectrometric determination of isotopic enrichment of 6-15NH2 in adenine nucleotides. Anal Biochem. 145 (2): 354-361, 1985.

Lin, J.H., Cocchetto, D.M., and Duggan, D.E.: Protein binding as a primary determinant of the clinical pharmacokinetic properties of non-steroidal anti­inflammatory drugs. Clin Pharmacokin. 12: 402-432, 1987.

Lin, Y-J., and Schanker, L.S.: Active transport of foreign amino acids by rat lung slices. Biochem Pharmacol. 30(21): 2937-2943,1981.

215

Lum, G.M., Aisenbrey, G.A., Dunn, M.J., Berl, T., Schrier, R.W., and McDonald, K.M.: In vivo effect of indomethacin to potentiate the renal medullary cyclic AMP response to vasopressin. J Clin Invest. 59: 8-13, 1977.

MacDonald, P.C., and Casey, M.L.: The accumulation of prostaglandins (PG) in amniotic fluid is an aftereffect of labor and not indicative of a role for PGE 2 or PGF 2 a in the initiation of human parturition. J Clin Endocrinol Metab. 76: 1332-1339, 1993.

Machin, G.A.: Fetal indomethacin exposure and renal dysgenesis. Am J Med Genetics. 44: 112-113, 1992.

MacTaggart, D.L., and Farwell, S.O.: Analytical use of linear regression. Part I: regression procedures for calibration and quantitation. J AO AC Int. 75: 594-614, 1992.

Manchester, D., Margolis, H.S., and Sheldon, R.E.: Possible association between maternal indomethacin therapy and primary pulmonary hypertension of the newborn. Am J Obstet Gynecol. 126: 467, 1976.

Manning, M., Chan, W.Y., and Sawyer, W.H.: Design of cyclic and linear peptide antagonists of vasopressin and oxytocin: Current status and future directions. Regulatory Peptides. 45: 279-283, 1993.

Mari, G., Moise, K.J.Jr., Deter, R.L., Kirshon, B., and Carpenter, R.J.: Doppler assessment of the renal blood flow velocity waveform during indomethacin therapy for preterm labour and polyhydramnios. Obstet Gynecol. 75: 199-201,1990.

Martin, B.K.: Drug urinary excretion data-some aspects concerning the interpretation. Br J Pharmacol. 29: 181-193, 1967.

Mason, R.W., and McQueen, E.G.: Protein binding of indomethacin: binding of indomethacin to human plasma albumin and its displacement from binding by ibuprofen, phenylbutasone and salicylate in vitro. Pharmacol. 12: 12-19,1974.

Matin, S.B., and Rowland, M.: Electron-capture sensitivity comparison of various derivatives of primary and secondary amines. J Pharm Sci. 61: 1235-1240,1972.

Matson, J.R., Stokes, J.B., and Robillard, J.E.: Effects of inhibition of prostaglandin synthesis on fetal renal function. Kidney Int. 20: 621-627, 1981.

Matsuki, Y., Ito, T., Kojima, M., Katsumura, H., Ono, H., Nambara, T.: A gas chromatographic-mass spectrometric method for the evaluation of bioconversion of indomethacin prodrugs into the parent drug. Chem Pharm Bull. 31: 2033-2038, 1983.

216

Mattila, M.A.K., Ahlstrom-Bengs, E., and Pekkola, P.: Intravenous indomethacin or oxycodone in prevention of post-operative pain. Br Med J. 287: 1026,1983.

Mawatari, K.-L, Iinuma, F., Watanabe, M: Fluorimetric determination of indomethacin in human serum by high-performance liquid chromatography coupled with post-column photochemical reaction with hydrogen peroxide. J Chromatogr. (Biomed Appl.) 491: 389-396,1989.

Mawhinney, T.P., and Madson, M.A.: N-methyl-N-(tert-butyldimethylsilyl) trifluoroacetamide and related N-tert-butyldimethylsilyl amides as protective silyl donors. J Org Chem. 47(17): 3336-3339,1982.

Mazzoni, M.R., and Lucacchini, A.: Evidence for the existence of a specific binding site for indomethacin on bovine vesicular gland microsomes. Biotech Appl Biochem. 9: 339-345, 1987.

McCarthy, J.S., Zies, L.G., and Gelband, H.: Age-dependent closure of the patent ductus arteriosus by indomethacin. Pediatr. 62: 706-712, 1978.

McCormick, D.C., Edwards, A.D., Brown, G.C., Wyatt, J.S., Potter, A., Cope, M., Delpy, D.T., and Reynolds, E.O.R.: Effect of indomethacin on cerebral oxidised cytochrome oxidase in preterm infants. Pediatr Res. 33: 603-608,1993.

McCoshen, J.A., Hoffman, D.R., Kredentser, J.V., et al.: The role of fetal membranes in regulating production, transport, and metabolism of prostaglandin E2 during labour. Am J Obstet Gynecol. 163: 1632,1990.

McCray, P.B., and Bettencourt, J.D.: Prostaglandins stimulate fluid secretion in human fetal lung. J Dev Physiol. 19: 29-36,1993.

McDonald, P.C., Koga, S., and Casey, M.L.: Decidual activation in parturition: examination of amniotic fluid for mediators of the inflammatory response. Ann NY Acad Sci. 622: 315, 1991.

McElnay, J.C., Passmore, A.P., Crawford, V.L.S., McConnell, J.G., Taylor, I.C., and Walker, F.S.: Steady state pharmacokinetic profile of indomethacin in elderly patients and young volunteers. Eur J Clin Pharmacol. 43: 77-80, 1992.

Mehta, A.C., and Calvert, R.T.: An improved high-performance liquid chromatographic procedure for monitoring indomethacin in neonates. Ther Drug Monit. 5: 143-145, 1983.

Meschia, G., Battaglia, F.C., and Bruns, P.D.: Theoretical and experimental study of transplacental diffusion. J Appl Physiol. 22: 1171-1178,1967.

217

Meschia, G.: Physiology of transplacental diffusion. Obstet Gynecol Ann. 5: 21-38, 1976.

Mihaly, G.W., and Morgan, D.J.: Placental drug transfer: effects of gestational age and species. Pharmacol Ther. 23: 253-266, 1984.

Millard, R.W., Baig, H., and Vatner, S.R.: Prostaglandin control of the renal circulation in response to hypoxemia in the fetal lamb in-utero. Circ Res. 45: 172-179, 1979.

Miller, F.C.: Uterine motility in spontaneous labour. Clin Obstet Gynecol. 26 (1): 78-85, 1983.

Mirkin, B.L.: Maternal and fetal distribution of drugs in pregnancy. Clin Pharmacol Ther. 14: 643-647, 1973.

Miyazaki, S., Yamahira, T., and Nadai, T.: Effect of bile flow on indomethacin absorption in rats. Acta Pharm Suec. 18: 135-138,1981.

Miyazaki, S., Yamahira, T., Inoue, H., and Nadai, T.: Interaction of drugs with bile components II Effect of bile on the absorption of indomethacin and phenylbutazone in rats. Chem Pharm Bull. 28: 323-326, 1980.

Mogilner, B.M., Ashkenazy, M., Borenstein, R., and Lancet, M.: Hydrops fetals caused by maternal indomethacin treatment. Acta Obstet Gynecol Scand. 61: 183-185, 1982.

Moise, K.J.Jr., Huhta, J.C., Sharif, D.S., Ou, C.-N., Kirshon, B., Wasserstrum, N., and Cano, L.: Indomethacin in the treatment of premature labor. Effects on the fetal ductus arteriosus. N Eng J Med. 319: 327-331, 1988:

Moise, K.J.Jr., Ou, C.-N., Krishon, B., Cano, L.E., Rognerud, C, and Carpenter, R.J.: Placental transfer of indomethacin in the human pregnancy. Am J Obstet Gynecol. 162: 549-554, 1990.

Montero, M.T., Pouplana, R., Vails, O., and Garcia, S.: On the binding of cinmetacin and indomethacin to human serum albumin. / Pharm Pharmacol. 38: 925-927, 1986.

Moore, T.R., and Cayle, J.E.: The amniotic fluid index in normal human pregnancy. Am J Obstet Gynecol. 162: 1168-1173, 1990.

Morel, A., Lolait, S.J., and Brownstein, M.J.: Molecular cloning and expression of rat V i a and V2 arginine vasopressin receptors. Regulatory Peptides. 45: 53-59, 1993.

218 Morishima, H.O., Finster, M., Pedersen, H., Fukunaga, A., Ronfeld, R.A., Vassallo, H.G., and Covino, B.G.: Pharmacokinetics of lidocaine in fetal and neonatal lambs and adult sheep. Anesthesiol. 50: 431-436, 1979.

Morsdorf, K.: In "Non-steroidal antiinflammatory drugs", Garattini, S., and Dukes, M.N.G., (Eds.), Excerpta Medica Foundation, Amsterdam, pp. 85-97, 1964.

Morselli, P.L., Franco-Morselli, R., and Bossi, L.: Clinical pharmacokinetics in newborns and infants. Age related differences and therapeutic implications. Clin Pharmacokinet. 5: 485-527, 1980.

Moya, F., and Thorndike, V.: Passage of drugs across the placenta. Am J Obstet Gynecol. 84: 1778-1798, 1962.

Muller, N., Lapicque, F., Monot, C, Payan, E., Gillet, P., Bannwarth, B., and Netter, P.: Protein binding of indomethacin in human cerebrospinal fluid. Biochem Pharmacol. 42: 799-804, 1991.

Naito, S-I., and Tsai, Y-H.: Percutaneous absorption of indomethacin from ointment bases in rabbits. Int J Pharm. 8: 263-276,1981.

Nakala, S., Skimmer, K., Mitchell, B.F., et al.: Changes of prostaglandin transfer across fetal membranes obtained after spontaneous labour. Am J Obstet Gynecol. 155: 1337, 1986.

Nathanielsz, P.W.: Monographs in fetal physiology 2: Animal models in fetal medicine (I). Elsevier/North-Holland Biomedical Press, Amsterdam, 1980.

Natochin, Y.V.: Renal Pharmacology: comparative, developmental, and cellular aspects. Ann Rev Pharmacol. 14: 75-90, 1974.

Nau, H., Luck, W., Kuhnz, W., and Wegener, S.: Serum protein binding of diazepam, desmethyldiazepam, furosemide, indomethacin, warfarin, and phenobarbital in human fetus, mother, and newborn infant. Pediatr Pharmacol. 3: 219-227, 1983.

Nau, H.: Clinical pharmacokinetics in pregnancy and perinatology. I Placental transfer and fetal side effects of local anaesthetic agents. Dev Pharmacol Ther. 8: 149-181, 1985.

National formulary, XIII.

Neander, G., Eriksson, L.O., Wallin-Boll, E., Ersmark, H., and Grahnen, A.: Pharmacokinetics of intraarticular indomethacin in patients with osteoarthritis. Eur J Clin Pharmacol. 42: 301-305,1992.

219

Niebyl, J.R., Blake, D.A., White, R.D., Kumor, K.M., Dubin, N.H., Robinson, J.C., and Egner, P.G.: The inhibition of premature labor with indomethacin. Am J Obstet Gynecol. 136: 1014-1019, 1980.

Nijhuis, J.G., Prechtl, H.F.R., Martin, C.B., and Bots, R.S.G.M.: Are there behavioural states in the human fetus? Early Hum Dev. 6: 177-195, 1982.

Nishioka, R., Harimoto, T., Umeda, I., Yamamoto, S., Oi, N.: Improved procedure for determination of indomethacin in plasma by capillary gas chromatography after solid-phase extraction. J Chromatogr. (Biomed Appl.) 526: 210-214,1990.

Nix, A.B.J., and Wilson, D.W.: Assay detection limits: concept, definition, and estimation. Eur J Clin Pharmacol. 39: 203-206,1990.

Noerr, B.: Indomethacin. Neonatal Network. 10: 81-83, 1991.

Nordstrom, L., and Westgren, M.: Indomethacin treatment for polyhydramnios-effective but potentially dangerous? Acta Obstet Gynecol Scand. 71: 239-241, 1992.

O'Brien, M., McCauley, J., Cohen, E.: Indomethacin. In Analytical Profiles of Drug Substances. Florey, K. (Ed.). Academic Press. 13: 211-238,1984.

O'Brien, WM: Indomethacin: a survey of clinical trials. Clin Pharmacol Ther. 9: 94-107, 1968.

Oberbaur, R., Krivanek, P., and Turnheim, K.: Pharmacokinetics of indomethacin in the elderly. Clin Pharmacokinet. 24: 428-434,1993.

Ohara, T., Ogata, H., and Tezuka, F.: Effects of indomethacin in-utero on the pulmonary vasculature of the newborn lambs. Tohoku J Exp Med. 164: 67-79, 1991.

Okumura, K., Yoshida, H., and Hori, R.: Tissue distribution and metabolism of drugs. Ill Accumulation of drugs by the isolated perfused rat lung. J Pharm Dyn. 1: 230-237, 1978.

Olkkala, K.T., Maunuksela, E.L., and Korpela, R.: Pharmacokinetics of postoperative intravenous indomethacin in children. Pharmacol Toxicol. 65: 157-160, 1989.

Ou, C.-N., and Frawley, V.L.: Liquid-chromatographic determination of indomethacin in blood from newborns with patent ductus arteriosus. Clin Chem. 30: 898-901, 1984.

220 Pacifici, G.M., Sawe, J., Kager, L., and Rane, A.: Morphine glucuronidation in human fetal and adult liver. Eur J Clin Pharmacol. 22: 553-558, 1982.

Parks, B.R., Jordan, R.L., Rawson, J.E., and Douglas, B.H.: Indomethacin: studies on absorption and placental transfer. Am J Obstet Gynecol. 129: 464-465,1977.

Paulick, R.P., Meyers, R.L., Rudolph, CD., and Rudolph, A.M.: Venous and hepatic vascular responses to indomethacin and prostaglandin Ei in the fetal lamb. Am J Obstet Gynecol. 163: 1357-1363,1990.

Pawelczyk, E., and Knitter, B.: Kinetics of drug degradation. Part 58: A method of preparation and the stability of 3% aqueous indomethacin solution. Pharmazie. 33: 586-588, 1978.

Pelkonen, O.: Xenobiotic metabolism in the maternal-placental-fetal unit: implications for fetal toxicity. Dev Pharmacol Ther. 7 (supp.l): 11-17, 1984.

Peng, G.W., and Chiou, W.L.: Analysis of drugs and other toxic substances in biological samples for pharmacokinetic studies. J Chromatogr. (Biomed Appl.) 531: 3-50, 1990.

Perucca, E., and Crema, A.: Plasma protein binding of drugs in pregnancy. Clin Pharmacokin. 7: 336-352, 1982.

Phelan, J.P., Smith, C.V., Broussard, P., et al.: Amniotic fluid volume assessment with the four-quadrant technique at 36-42 weeks' gestation. J Reprod Med. 32: 540-542, 1987.

Plazonnet, B., and Van den Heuvel, W.J.A.: Preparation, gas chromatography and mass spectrometry of methyl and trimethylsilyl esters of indomethacin, J Chromatogr. 142: 587-596, 1977.

Plotz, CM.: Clinical application of indomethacin: efficacy and tolerability. Sem Arthritis Rheumat. 12 (supp.l): 100-102,1982.

Pratzel, H., Dittrich, P., and Kukovetz, W.: Spontaneous and forced cutaneous absorption of indomethacin in pigs and humans. J Rheumatol. 13: 1122-1125, 1986.

Prietsch, V., Maier, R., Schmitz, L., and Obladen, M.: Long-term variability of heart rate increases with successful closure of patent ductus arteriosus in preterm infants, Biol Neonate. 61: 142-149,1992.

221 Pryds, O., Greisen, G., and Johansen, K.H.: Indomethacin and cerebral blood flow in premature infants treated for patent ductus arteriosus. Eur J Pediatr. 147: 315-316, 1988.

Rado, J.P., Simatupang, T., Boer, P., and Dorhout Mees, E.J.: Pharmacologic studies in Bartter's syndrome: Effect of DDAVP and indomethacin on renal concentrating operation. Part II. Int J Clin Pharmacol Biopharm. 16 (1): 22-26, 1978.

Rajadurai, V.S., and Yu, V.Y.H.: Intravenous indomethacin therapy in preterm neonates with patent ductus arteriosus. J Pediatr Child Health. 27: 370-375, 1991.

Raz, A., Stern, H., and Kenig-Wakshall, R.: Indomethacin and aspirin inhibition of PGE2 synthesis by sheep seminal vesicles microsome powder and seminal vesicle slices. Prostaglandins. 3: 337-352, 1973.

Reynolds, F., and Knott, C: Pharmacokinetics in pregnancy and placental drug transfer. Oxford Rev Reprod Biol. 11: 389-449,1989.

Riggs, K.W., Rurak, D.W., Yoo, S.D., McErlane, B.A., Taylor, S.M., McMorland, G.H., and Axelson, J.E.: Drug accumulation in lung fluid of the fetal lamb after maternal or fetal administration. Am J Obstet Gynecol. 157: 1286-1291, 1987.

Riggs, K.W.: Metoclopramide kinetics in sheep: Maternal-fetal disposition, fetal pharmacodynamics and a comparison between pregnant and non-pregnant ewes. PhD thesis, University of British Columbia, Vancouver, 1989.

Robertson, G.L.: The use of vasopressin assays in physiology and pathophysiology. Sem Nephrol. 14: 368-383, 1994.

Robillard, J.E., Weitzman, R.E., Fisher, D.A., and Smith, F.G.: Developmental aspects of renal tubular reabsorption of water and fetal renal response to arginine vasopressin. In "The kidney during development: Morphology and function." Spitzer, A (Ed.), Masson, New York, chapter 27, pp. 205-213, 1982.

Rosen, D.J.D., Fejgin, M.D., Rabinowitz, R., Regev, R.H., and Beyth, Y.: Indomethacin therapy and fetal urine production in twins with polyhydramnios. J Perinat Med. 19: 173-176,1991.

Roth-Brandel, U., Bygdeman, M., and Wiquist, N.: Effect of intravenous administration of prostaglandin E! and F 2 a on the contractility of the nonpregnant uterus in vivo. Acta Obstet Gynecol Scand. (Suppl.) 49: 1,1970.

222 Rowe, J.S., and Carless, J.E.: Investigations on the in vitro dose-related metabolism of indomethacin in four laboratory species. Eur J Drug Metab Pharmacokin. 7: 159-163, 1982.

Rowland, M., Benet, L.Z., and Graham, G.G.: Clearance concepts in pharmacokinetics. J Pharmacokin Biopharm. 1: 123-136,1973.

Rubaltelli, F.F., Chiozza, M.L., Zanardo, V., and Cantarutti, F.: Effect on neonate of maternal treatment with indomethacin. J Pediatr. 94: 161,1979.

Rudolph, A.M., and Heymann, M.A.: Hemodynamic changes induced by blockers of prostaglandin synthesis in the fetal lamb in-utero. Adv Prostaglandin Thromboxane Leukotriene Res. 4: 231,1978.

Rudolph, A.M.: The effects of nonsteroidal antiinflammatory compounds on fetal circulation and pulmonary function. Obstet Gynecol. 58: 63S-67S, 1981.

Rudolph, A.M.: The fetal circulation and its response to stress. / Dev Physiol. 6: 11-19, 1984.

Rurak, D.W., Wright, M.R., and Axelson, J.E.: Drug disposition and effects in the fetus. J Dev Physiol. 15: 33-44,1991.

Rurak, D.W.: Fetal behavioural states: pathological alterations with drug/alcohol abuse. Sem Perinatol. 16: 239-251,1992.

Sabik, J.F., Assad, R.S., and Hanley, F.L.: Prostaglandin synthesis inhibition prevents placental dysfunction after fetal cardiac bypass. J Thorac Cardiovasc Surg. 103: 733-742, 1992.

Saliba, E., Autret, E., Nasr, C, Sue, A.L., and Laugier, J.: Perinatal pharmacology and cerebral blood flow. Biol Neonate. 62: 252-257, 1992.

Schanker, L.S.: Drug absorption from the lung. Biochem Pharmacol. 27: 381-385, 1978.

Schoenfeld, A., Bar, Y., Merlob, P., Ovadia, Y.: NSAIDs: maternal and fetal considerations. Am J Reprod Immunol. 28: 141-147,1992.

Schonhofer, P., and Anspach, K.F.: Die aktivitat der L-glutamin-D-fructose-6-phosphat-aminotransferase im verlauf einer durch carrageenin induzierten entzundung und ihre hemmbarkeit durch phenylbutazon. Arch Int Pharmacodyn. 166 (2): 382-389, 1967.

223 Schooley, D.L., Kubiak, F.M., and Evans, J.V.: Capillary gas chromatographic analysis of volatile and non-volatile organic acids from biological samples as the tert-butyldimethylsilyl derivatives. J Chromatogr Sci. 23(9): 385-390, 1985.

Schwertschlag, U., Gerber, J.G., Barnes, J.S., and Nies, A.S.: A dual effect of nonsteroidal anti-inflammatory drugs on renal water handling in humans. / Pharmacol Exp Ther. 238: 653-658, 1986.

Shah, V.P., Midha, K.K., Dighe, S., McGilveray, I.J., Skelly, J.P., Yacobi, A., Layoff, T., Vishwanathan, C.T., Cook, C.E., McDowall, R.D., Pittman, K.A., Spector, S.: Analytical methods validation: bioavailability, bioequivalence, and pharmacokinetic studies. Int J Pharm. 82: 1-7,1992.

Shen, T.Y., and Winter, C.A.: Chemical and biological studies on indomethacin, sulindac, and their analogs. Adv Drug Res. 12: 89-245, 1977.

Shen, T.Y., Windholz, T.B., Rosegoy, A., Witzel, B.E., Wilson, A.N., Willet, J.D., Holtz, W.J., Ellis, R.L., Mattzuk, A.R., Lucas, S., Stammer, C.H., Holly, F.W., Sarett, L.H., Risley, E.A., Nuss, G.W., and Winter, C.A.: Non-steroidal anti­inflammatory agents. J Am Chem Soc. 85: 488-489, 1963.

Shen, T.Y.: Synthesis and biological activity of some indomethacin analogs. In "International symposium of non-steroidal anti-inflammatory drugs", Garattini, S., and Dukes, M.N.G., (Eds.), International congress series No. 82, pp. 13-20, 1965.

Shen, T.Y.: The discovery of indomethacin and the proliferation of NSAIDs. Sem Arthritis Rheumat. 12 (supp.l): 89-93,1982.

Shima, K., Matsusaka, C, Hirose, M., et al.: Biopharmaceutical characteristics of indomethacin gel ointment. Chem Pharm Bull. 29: 2338-2344, 1981.

Sibeon, R.G., Baty, J.D., Baber, N., Chan, K., L'e Orme, M.: Quantitative gas-liquid chromatographic method for the determination of indomethacin in biological fluids. J Chromatogr. 153: 189-194,1978.

Silberman, H.R., McGinn, T.G., and Kremer, W.B.: Control of fever in Hodgkin's disease by indomethacin. J Am Med Ass. 194(6): 597-600,1965.

Simmons, M.A., Battaglia, F.C., and Meschia, G.: Placental transfer of glucose. / Dev Physiol. 1: 227-243, 1979.

Singh, A.K., Jang, Y., Mishra, U., Granley, K.: Simultaneous analysis of flunixin, naproxen, ethacrynic acid, indomethacin, phenylbutazone, mefenamic acid and thiosalicylic acid in plasma and urine by high-performance liquid chromatography

224 and gas chromatography-mass spectrometry. J Chromatogr. (Biomed Appl.) 568: 351-361, 1991.

Skeith, M.D., Simkin, P.A., and Healey, L.A.: The renal excretion of indomethacin and its inhibition by probenecid. Clin Pharmacol Ther. 9: 89-93, 1967.

Skellern, G. G., and Salole, E.G.: A high-speed liquid chromatographic analysis of indomethacin in plasma. J Chromatogr. 114: 483-485, 1975.

Smith, J.B., and Willis, A.L.: Aspirin selectively inhibits prostaglandin production in human platelets. Nature New Biol. 231: 235-237, 1971.

Smith, L.G., Kirshon, B., and Cotton, D.B.: Indomethacin treatment for polyhydramnios and subsequent infantile nephrogenic diabetes insipidus. Am J Obstet Gynecol. 163: 98-99, 1990.

Smith, P.C., and Benet, L.Z.: High-performance liquid chromatographic method for the determination of indomethacin and its two primary metabolites in urine. J Chromatogr. (Biomed Appl.) 306: 315-321,1984.

Smith, W.L., and Lands, W.E.M.: Stimulation and blockade of prostaglandin biosynthesis. JBiolChem. 21: 6700-6702,1971.

Snyder, D.S.: Cutaneous effects of topical indomethacin an inhibitor of prostaglandin synthesis on UV-damaged skin. J Invest Dermatol. 64: 322-325, 1975.

Sonnenburg, W.K., Zhu, J.H., and Smith, W.L.: A prostaglandin E receptor coupled to a pertussis toxin-sensitive guanine nucleotide regulatory protein in rabbit cortical collecting tubule cells. JBiolChem. 265: 8479-8483,1990.

Srivastava, M., Collins, M.D., Scott, W.J.Jr., Wittfoht, W., and Nau, H.: Transplacental distribution of weak acids in mice: accumulation in compartments of highpH. Teratology. 43: 325-329,1991.

Stacey, T.E., Weedon, A.P., Haworth, C, Ward, R.H.T., and Boyd, R.D.H.: Fetomaternal transfer of glucose analogs by sheep placenta. Am J Physiol. 234: E 32-E 37, 1978.

Stanford, N., Roth, G.J., Shen, T.Y., and Majerus, P.W.: Lack of covalent modification of prostaglandin synthetase (cyclo-oxygenase) by indomethacin. Prostaglandins. 13: 669-675, 1977.

225 Steele, A., and Ayromlooi, J.: Effects of indomethacin on the acid-base balance and hemodynamics of chronically instrumented pregnant sheep. Dev Pharmacol Ther. 11: 96-101, 1988.

Stein, G., Kunze, M., Zaumseil, J., and Traeger, A.: Zur pharmakokinetik von indomethazin und indomethazin-metaboliten bei wiederholter applikation an nierengesunden und nierengeschadigten patienten. Int J Clin Pharmacol. 15: 470-473, 1977.

Stevenson, K.M., and Lumbers, E.R.: Effects of indomethacin on fetal renal function, renal and umbilicoplacental blood flow and lung liquid production. / Dev Physiol. 17: 257-264, 1992.

Stubbs, R.J., Schwartz, M.S., Chiou, R., Entwistle, L.A., Bayne, W.F.: Improved method for the determination of indomethacin in plasma and urine by reversed-phase high-performance liquid chromatography. J Chromatogr. (Biomed Appl.) 383: 432-437, 1986.

Surborg, V.K.H.: Vergleichende metabolische studien von [14C]-markiertem indometacin und acemetacin bei rattess. Arzneim Forsch. 30: 1384-1391,1980.

Szeto, H.H., Mann, L.I., Bhaktharathsalan, A., Liu, M., and Inturisz, C.E.: Meperidine pharmacokinetics in the maternal-fetal unit. / Pharmacol Exp Ther. 206: 448, 1978.

Szeto, H.H., Umans, J.G., and McFarland, J.: A comparison of morphine and methadone disposition in the maternal-fetal unit. Am J Obstet Gynecol. 143: 700-706, 1982a.

Szeto, H.H., Umans, J.G., and Rubinow, S.I.: The contribution of transplacental clearances and fetal clearance to drug disposition in the ovine maternal-fetal unit. Drug Metab Disp. 10: 382-386, 1982b.

Szeto, H.H., Umans, J.G., Umans, H.R., and McFarland, J.W.: The relationship between maternal and fetal plasma protein binding of methadone in the ewe during the third trimester. Life Sci. 30: 1271-1279, 1982c.

Szeto, H.H.: Behavioural states and their ontogeny: animal studies. Sem Perinatol. 16: 211-216, 1992.

Szeto, H.H.: Kinetics of drug transfer to the fetus. Clin Obstet Gynecol. 36: 246-254, 1993.

Szeto, H.H.: Maternal-fetal pharmacokinetics and fetal dose-response relationships. Ann NY Acad Sci. 562: 42-55, 1989.

226 Szeto, H.H.: Pharmacokinetics in the ovine maternal-fetal unit. Ann Rev Pharmacol Toxicol. 22: 221-243, 1982d.

Tang-Li, D.D.-S., Tozer, T.N., Riegelman, S.: Dependence of renal clearance on urine flow: A mathematical model and its application. J Pharm Sci. 72: 154-158, 1983.

Terhaag, B., and Hermann, U.: Biliary elimination of indomethacin in man. Eur J Clin Pharmacol. 29: 691-695,1986.

Terragno, N.A., Terragno, D.A., and McGiff, J.C.: Contribution of prostaglandin to the renal circulation in conscious, anesthetised, and laparotomised dogs. Circ Res. 40: 590-595, 1977.

Terweij-Groen, G.P., Heemstra, S., Kraak, J.C.: Rapid determination of indomethacin and salicylic acid in serum by means of reversed-phase liquid chromatography. J Chromatogr. (Biomed Appl.) 181: 385-397,1980.

Thalji, A., Yeh, T.F., Raval, D., et al.: Pharmacokinetics of intravenous indomethacin in premature infants. Pediatr Res. 13: 374,1979.

Thalji, A.A., Carr, I., Yeh, T.F., Raval, D., Luken, J.A., and Pildes, R.S.: Pharmacokinetics of intravenously administered indomethacin in premature infants. J Pediatr. 97: 995-1000, 1980.

Thiery, M., and Amy, J-J.: Induction of labour with prostaglandins. In Karim, S.M.M. (Ed.), Prostaglandins and reproduction. Press Limited, pp. 149, 1975.

Tomson, G., Lunell, N.-O., Sundwall, A., and Rane, A.: Placental passage of oxazepam and its metabolism in mother and newborn. Clin Pharmacol Ther. 25: 74-81, 1979.

Tonn, G.R.: personal communication, 1994.

Traeger, A., Joschel, N., and Zamumseil, J.: Pharmacokinetics of indomethacin in pregnant and parturient women and in their newborn infants. Zentralbl Gynaekol. 95: 635-641, 1973.

Traeger, A., Stein, G., Kuntze, M., and Zaumseil, J.: Zur pharmakokinetik von indomethazin bei nierengeschadigten patienten. Int J Clin Pharmacol. 6: 237-242, 1972.

Tucker, G.T.: Measurement of the renal clearances of drugs. Br J Clin Pharmacol. 12: 761-770, 1981.

227 Turakka, H., and Airaksinen, M.M.: Biopharmaceutical assessment of phenylbutasone and indomethacin preparations. Annals Clin Res. 6 (Suppl. 11): 34-41, 1974.

van Bel, F., Van de Bor, M., Stijnen, T., Baan, J., and Ruys, J.H.: Cerebral blood flow velocity changes in preterm infants after a single dose of indomethacin: duration of its effect. Pediatrics. 84: 802-807,1989.

van Bel, F., van Zwieten, P.H.T., Lya, L., and Ouden, D.: Contribution of colour doppler flow imaging to the evaluation of the effect of indomethacin on neonatal cerebral hemodynamics. J Ultrasound Med. 9: 107-109,1990.

van der Heijden, A.J., Provoost, A.P., Nauta, J., et al.: Renal functional impairment in preterm neonates related to intrauterine indomethacin exposure. Pediatr Res. 24: 644-648, 1988b.

van der Heijden, A.J., Provoost, A.P., Nauta, J., Wolff, E.D., and Sauer, P.J.J.: Indomethacin as an inhibitor of preterm labour. Effect on postnatal renal function. Contrib Nephrol. 67: 152-154, 1988a.

van der Meer, M.J., and Hundt, H.K.L.: Estimation of serum indomethacin at therapeutic levels by means of thin-layer chromatography and spectrophotometry. J Chromatogr. 181: 282-285, 1980.

van Kreel, B.K.: A mathematical approach to the mechanisms of placental transfer. Placenta (supp.l): 81-100,1981.

van Petten, G.R., Mathison, H.J., Harris, W.H., and Mears, G.J.: Chronic preparation of the pregnant ewe and fetus for pharmacological research: The placental transfer and fetal effects of bunitrolol. / Pharmacol Meth. 1: 45-60, 1978.

van der Veyver, I.B., and Moise, K.J.: Prostaglandin synthetase inhibitors in pregnancy. Obstet Gynecol Survey. 48: 493-502, 1993.

Vane, J.R.: Inhibition of prostaglandin synthesis as a mechanism of action for aspirin-like drugs. Nature New Biol. 231: 232-239,1971.

Vanhaesebrouck, P., Thiery, M., Leroy, J.G., et al.: Oligohydramnios, renal insufficiency and ileal perforation in preterm infants after intrauterine exposure to indomethacin. J Pediatr. 113: 738-743, 1988.

Veersema, D., de Johng, P.A., van Wijck, J.A.M.: Indomethacin and the fetal renal nonfunctional syndrome. Eur J Obstet Gynecol Reprod Biol. 16: 113-121,1983.

228

Vert, P., Bianchetti, G., Marchal, F., Monin, P., and Morselli, P.L.: Effectiveness and pharmacokinetics of indomethacin in premature newborns with patent ductus arteriosus. Eur J Clin Pharmacol. 18: 83-88,1980.

Vree, T. B., Van den Biggelaar-Martea, M., Verwey-van Wissen, C. P. W. G. M.: Determination of indomethacin, its metabolites and their glucuronides in human plasma and urine by means of direct gradient high-performance liquid chromatographic analysis. Preliminary pharmacokinetics and effect of probenecid. J Chromatogr. (Biomed Appl.) 616: 271-282, 1993.

Wada, Y., Etoh, Y., Ohira, A., Kimata, H., Koide, T., Ishihama, H., and Mizushima, Y.: Percutaneous absorption and anti-inflammatory activity of indomethacin in ointment. J Pharm Pharmacol. 34: 467-468, 1982.

Waddell, W.J., and Marlowe, C.: Transfer of drugs across the placenta. Pharmacol Ther. 14: 375-390, 1981.

Waddell, W.J.: Localisation and metabolism of drugs in the fetus. Federation Proceedings. 31: 52-62, 1972.

Wade, J.B.: Role of membrane traffic in the water and Na + responses to vasopressin. Sem Nephrol. 14: 322-332, 1994.

Wahlich, J.C., and Carr, G.P.: Chromatographic system suitability tests-what should we be using? J Pharm Biomed Anal. 8: 619-623,1990.

Walker, M.P.R., Cheung, C.Y., and Brace, R.A.: Plasma atrial natriuretic factor and arginine vasopressin responses to indomethacin in the ovine fetus [Abstract]. Am J Obstet Gynecol. 166: 379,1992c.

Walker, M.P.R., Rurak, D.W., and Riggs, K.W.: Fetal consequences of prostaglandin synthesis blockade. BC Health Research Foundation grant, 1993.

Walker, M.P.R., Moore, T.R., and Brace, R.A.: Indomethacin and arginine vasopressin interaction in the fetal kidney: A mechanism of oliguria. Am J Obstet Gynecol. Ill: 1234-1241, 1994.

Walker, M.P.R., Moore, T.R., and Brace, R.A.: Urinary and cardiovascular responses to indomethacin infusion in the ovine fetus. Am J Obstet Gynecol. 167: 834-843, 1992b.

Walker, M.P.R., Moore, T.R., Cheung, C.Y., and Brace, R.A.: Indomethacin-induced urinary flow rate reduction in the ovine fetus is associated with reduced free water clearance and elevated plasma arginine vasopressin levels. Am J Obstet Gynecol. 167: 1723-1731, 1992.

229 Walker, R.W., Gruber, V.F., Rosenberg, A., Wolf, F.J., Vanden Heuvel, W.J.A.: computerised mass spectrometric assay for sulindac and its metabolites in human plasma. Anal Biochem. 95: 579-586, 1979.

Wallusch, W.W., Nowak, H., Leopold, G., Netter, K.J.: Comparative bioavailability: Influence of various diets on the bioavailability of indomethacin. Int J Clin Pharmacol. 16: 40-44, 1978.

Wang, L.H., Rudolph, A.M., and Benet, L.Z.: Comparative study of acetaminophen disposition in sheep at three developmental stages: the fetal, neonatal, and adult periods. Dev Pharmacol Ther. 14: 161-179,1990.

Wang, L.H., Rudolph, A.M., and Benet, L.Z.: Distribution and fate of acetaminophen conjugates in the fetal lambs in-utero. / Pharmacol Exp Ther. 235: 302-307, 1985.

Ward, R.M.: Maternal drug therapy for fetal disorders. Sem Perinatol. 16: 12-20, 1992.

Watanabe, J., Urasaki, Y., Nakase, Y., Ueda, H., Iwamoto, K., and Ozeki, S.: Excretion of indomethacin into saliva following intravenous administration to dogs. J Pharm Dyn. 4: 336-344,1981.

Weiner, I.M., and Mudge, G.H.: Renal tubular mechanisms for excretion of organic acids and bases. Am J Med. 36: 743-762.

Wells, P.G.: Analgesics: direct embryopathic effects and indirect biochemical effects modulating teratogenecity of other drugs and chemicals. In Toxicologic and Pharmacologic Principles in Pediatrics. Kacew, S., and Lock, S. (Eds.) Hemisphere Publishing Corporation, pp. 127-166, 1988.

Wennwalm, A., Eriksson, S., and Wahren, J.: Effect of indomethacin on basal and carbon dioxide stimulated cerebral blood flow in man. Clin Physiol. 1: 227-234, 1981.

Whitehouse, M.W.: The molecular pharmacology of anti-inflammatory drugs: Some possible mechanisms of action at the biochemical level. Biochem Pharmacol. Suppl.: 293-307, 1968.

Wiest, D.B., Pinson, J.B., Gal, P.S., Brundage, R.C., Schall, S., Ransom, J.L., Weaver, R.L., Purohit, D., and Brown, Y.: Population pharmacokinetics of intravenous indomethacin in neonates with symtomatic patent ductus arteriosus. Clin Pharmacol Ther. 49: 550-557,1991.

230 Wilkening, R.B., and Meschia, G.: Current topic: comparative physiology of placental oxygen transport. Placenta. 13: 1-15,1992.

Wilkening, R.B., Anderson, S., and Meschia, G.: Non steady state placental transfer of highly diffusible molecules. J Dev Physiol. 6: 121-129,1984.

Wilkening, R.B., Anderson, S., Martensson, L., and Meschia, G.: Placental transfer as a function of uterine blood flow. Am J Physiol. 242: H429-H436, 1982.

Wilkinson, G.R.: Clearance approaches in pharmacology. Pharmacol Rev. 39: 1-47,1987.

Winther, J.B., Hoskins, E., Printz, M.P., Mendoza, S.A., Kirkpatrick, S.E., and Friedman, W.F.: Influence of indomethacin on renal function in conscious newborn lambs. Biol Neonate. 38: 76-84,1980.

Wood, J.H., and Leonard, T.W.: Kinetic implications of drug resorption from the bladder. Drug Metab Rev. 14: 407-423,1983.

Woolf, D., Huskinsson, E.C., and Nicol, C.G.: Indomethacin and benorylate in osteoarthritis of the hip. Practitioner. 221 (1325): 791-792, 1978.

Wright, M.R., Rurak, D.W., van der Weyde, M.P., Taylor, S.M., and Axelson, J.E.: Clearance and disposition of ritodrine in the fluid compartments of the fetal lamb during and after constant rate fetal intravenous infusion. J Pharmacol Exp Ther. 258: 897-902, 1991.

Wright, M.R.: Ritodrine: Analysis, comparative maternal and fetal pharmacokinetics and pharmacodynamics in sheep, PhD thesis, University of British Columbia, Vancouver, 1992.

Wright, V., Walker, W.C., and McGuire, R.J.: Indomethacin in the treatment of rheumatoid arthritis. Ann Rheum Dis. 28: 157, 1969.

Wurtzel, D.: Prenatal administration of indomethacin as a tocolytic agent: effect on neonatal renal function. Obstet Gynecol. 76: 689-692, 1990.

Yaffe, S.J., and Juchau, M.R.: Perinatal Pharmacology. Ann Rev Pharmacol. 14: 219-238, 1974.

Yaffe, S.J., Friedman, W.F., Rogers, D., Lang, P., Ragni, M., and Saccar, C: The disposition of indomethacin in preterm babies. J Pediatr. 97: 1001-1006,1980.

Yeh, K.C.: Pharmacokinetic overview of indomethacin and sustained-release indomethacin. Am J Med. 79 (Supp. 4C): 3-12,1985.

231 Yeleswaram, K.: Analysis, pharmacokinetics, metabolism, and pharmacodynamics of labetalol in pregnant and non-pregnant sheep. PhD thesis, University of British Columbia, Vancouver, 1992.

Yesair, D.W., Callahan, M . , Remington, L . , and Kensler, C.J.: Role of the entero­hepatic cycle of indomethacin on its metabolism, distribution in tissues and its excretion by rats, dogs, and monkeys. Biochem Pharmacol. 19: 1579-1590,1970.

Yoo, S.D.: Maternal-fetal disposition, fetal pharmacodynamics, and comparative pharmacokinetics of diphenhydramine in pregnant and non-pregnant sheep. PhD thesis, University of British Columbia, Vancouver, 1989.

Yoshida, H . , Kamiya, A . , Okumura, K. , and Hori, R.: Contribution of monoamine oxidase (MAO) to the binding of tertiary basic drugs in lung mitochondria. Pharm Res. 6 (10): 877-882, 1989.

Zachariae, E . : Osteoarthritis treated with indomethacin. Nordisk Medicin. 75 (14): 384-385, 1966.

Zancan, L . , Zacchello, F., and Mantero, F.: Indomethacin for Bartter's syndrome. Lancet. 2: 1354, 1976.

Zinic, M . , Kuftinec, J., Hofman, H . , Kajfez, F., Meic', Z.: Zomepirac sodium. In Analytical Profiles of Drug Substances. Florey, K . (Ed.). Academic Press. 15: 673-698, 1986.

Zona, C , Roscetti, G. , Venturelli, F., and Roda, L . G . : Effect of different parameters on the binding of two anti-inflammatory drugs to human serum albumin. Pharmacol Biochem Behav. 24: 1031-1037,1986.

Zuckerman, H . , Reiss, U . , and Rubinstein, I.: Inhibition of human premature labour by indomethacin. Obstet Gynecol. 44: 787-792, 1974.