Perioperative complications in children withpulmonary hypertension undergoing generalanesthesia with ketamine

GLYN D. WILLIAMS M B CM B C hh BB* , HARJOT MAAN B SB S†,

CHANDRA RAMAMOORTHY M DM D*, KOMAL KAMRA M DM D*,

SUSAN L. BRATTON M D M P HM D M P H‡, ELLEN BAIR N PN P* , CALVIN

C. KUAN M DM D*, GREGORY B. HAMMER M DM D* AND JEFFREY

A. FEINSTEIN M D M P HM D M P H§

*Division of Pediatric Cardiology, Department of Anesthesia, Lucile Packard Children’sHospital, Stanford University, Stanford, CA, †Medical Student, Dartmouth Medical Center,Hanover, NH, ‡Pediatric Critical Care Medicine, University of Utah, Salt Lake City, UT and§Division of Pediatric Cardiology, Department of Pediatrics, Lucile Packard Children’s Hospital,Stanford University, Stanford, CA, USA

SummaryBackground: Pulmonary arterial hypertension (PAH) is associated

with significant perioperative risk for major complications in children,

including pulmonary hypertensive crisis and cardiac arrest. Uncer-

tainty remains about the safety of ketamine anesthesia in this patient

population.

Aim: Retrospectively review the medical records of children with

PAH to ascertain the nature and frequency of peri-procedural

complications and to determine whether ketamine administration was

associated with peri-procedural complications.

Methods: Children with PAH (mean pulmonary artery pressure

‡25 mmHg and pulmonary vascular resistance index ‡3 Wood units)

who underwent general anesthesia for procedures during a 6-year

period (2002–2008) were enrolled. Details about the patient, PAH,

procedure, anesthetic and postprocedural course were noted,

including adverse events during or within 48 h of the procedure.

Complication rates were reported per procedure. Association between

ketamine and peri-procedural complications was tested.

Results: Sixty-eight children (median age 7.3 year, median weight

22 kg) underwent 192 procedures. Severity of PAH was mild (23%),

moderate (37%), and severe (40%). Procedures undertaken were

major surgery (n = 20), minor surgery (n = 27), cardiac catheterization

(n = 128) and nonsurgical procedures (n = 17). Ketamine was

administered during 149 procedures. Twenty minor and nine major

complications were noted. Incidence of cardiac arrest was 0.78% for

cardiac catheterization procedures, 10% for major surgical procedures

and 1.6% for all procedures. There was no procedure-related

mortality. Ketamine administration was not associated with increased

complications.

Correspondence to: G.D. Williams, 300 Pasteur Drive, M ⁄ C 5640, Stanford, CA 94305, USA (email: jumbo@stanford.edu).

Pediatric Anesthesia 2010 20: 28–37 doi:10.1111/j.1460-9592.2009.03166.x

28 � 2009 Blackwell Publishing Ltd

Conclusions: Ketamine appears to be a safe anesthetic option for

children with PAH. We report rates for cardiopulmonary resuscitation

and mortality that are more favorable than those previously reported.

Keywords: children; complications; general anesthesia; hypertension;

ketamine; perioperative; pulmonary

Introduction

Pulmonary arterial hypertension (PAH) is defined as

a mean pulmonary arterial blood pressure that

exceeds 25 mmHg at rest or 30 mmHg during

exercise in association with variable degrees of

pulmonary vascular remodeling, vasoconstriction,

and in situ thrombosis (1,2).

Children with PAH typically have an increased

requirement for medical resources (3) and may

receive multiple general anesthetics for procedures

related to PAH assessment and management. The

pathophysiology of PAH (4–6) and the associated

anesthetic considerations for adults and children

have been reviewed (4,5,7,8). PAH contributes to

perioperative morbidity and mortality (7,9–15).

Although ketamine has been used successfully in

the anesthetic management of patients with PAH

(16), its use remains controversial because it has

been associated with increased pulmonary arterial

pressure in adult patients (17). Studies evaluating

the effect of ketamine on pulmonary artery pressure

in children with PAH have reported conflicting

results (18–21).

We previously studied the hemodynamic

responses to ketamine in children with PAH and

found that ketamine in the presence of sevoflurane

did not cause increased pulmonary vascular

resistance (PVR) (22). The effects on PVR of keta-

mine, either as the sole agent or when administered

along with anesthetic agents other than sevoflurane,

were not investigated. Consequently, uncertainty

remains whether ketamine without concomitant

sevoflurane is an appropriate anesthetic option for

children with PAH. To address this ambiguity, we

conducted a 6-year retrospective review of children

with PAH who underwent diagnostic and

therapeutic procedures to ascertain the nature and

frequency of peri-procedural complications.

We hypothesized that ketamine administration dur-

ing anesthesia was not associated with peri-proce-

dural complications.

Methods

Institutional Review Board approval with waiver of

informed consent was obtained for this retrospective

cohort of children with PAH who underwent gen-

eral anesthesia for a procedure during the period

from October 1, 2002, to October 31, 2008. Potential

study participants were identified by searching the

institution’s Heart Center database for patients

with the diagnosis of pulmonary hypertension who

underwent general anesthesia for cardiac catheteri-

zation. The medical records of these patients were

then reviewed to ascertain whether criteria for study

enrollment were met.

Inclusion criteria were as follows: (i) a mean

pulmonary artery pressure ‡25 mmHg and a calcu-

lated PVR (indexed) ‡3 Wood units; (ii) all patients

receiving pulmonary vasodilator therapy for PAH

(diagnosis confirmed by cardiac catheterization);

and (iii) age <18 years.

Exclusion criteria were as follows: (i) elevation

of pulmonary artery pressure because of main or

proximal branch pulmonary artery obstructions and

(ii) the diagnosis of tetralogy of Fallot, pulmonary

atresia, and major aorto-pulmonary collaterals.

The medical records of eligible patients were

examined to identify all procedures performed

under general anesthesia during the study period.

Data were recorded describing patient demograph-

ics, preprocedure diagnoses, procedure performed,

anesthesia management, cardiac catheterization

findings, postprocedure course until discharge from

hospital, and any adverse events that occurred

during or within 48 h of the procedure. Duration

of follow-up and mortality were also recorded.

PULMONARY HYPERTENSION: IS KETAMINE SAFE? 29

� 2009 Blackwell Publishing Ltd, Pediatric Anesthesia, 20, 28–37

Definitions of complications were consistent with

previous reports (23). An incident was an observed

change in monitored values that was transient, had

no effect on the patient’s condition and required

minimal or no treatment. A minor complication was a

transient event that had no long-term ill effect on the

patient and resolved with specific treatment. A major

complication was a potentially life-threatening event

requiring immediate treatment. For example, occur-

rence of an arrhythmia would be (i) excluded if there

was no hemodynamic disturbance and no interven-

tion required; (ii) an incident if the arrhythmia

caused minor hemodynamic disturbance and

resolved spontaneously; (iii) a minor complication

if there was minor hemodynamic disturbance and

a therapeutic intervention was required; and (iv) a

major complication if there was severe hemodynamic

compromise that required prompt intervention.

For each procedure, patients were classified based

on the severity of their PAH. For those undergoing

cardiac catheterization, the PAH categories were

mild [pulmonary vascular resistance index (PVRI) >

3–7 Wood units], moderate (PVRI > 7–12 Wood

units), or severe (>12 Wood units). Patients under-

going procedures other than cardiac catheterization

were classified on the basis of the preprocedure

echocardiogram as follows: mild (systolic pulmo-

nary artery blood pressure (PAP) £70% of systolic

systemic blood pressure), moderate (systolic

PAP > 70–100% of systolic systemic blood pressure)

or severe (systolic PAP > 100% of systolic systemic

blood pressure) (13). The most recent cardiac cath-

eterization data were used if PAP could not be

estimated from the preprocedure echocardiogram.

Patients were categorized as reactive to pulmonary

vasodilator therapy if the following criteria were

met at cardiac catheterization: (i) a decrease of

‡10 mmHg in mean PAP to values <40 mmHg if

baseline mean PAP was >40 mmHg; (ii) ‡20%

decrease in mean PAP and PVRI if baseline mean

PAP was <40 mmHg (24).

All medical services were provided at a tertiary

children’s hospital. Anesthesia care was provided by

pediatric anesthesiologists or by pediatric cardiac

intensivists. The anesthetic technique employed was

determined by the physician of record.

Procedures were classified as (a) nonsurgical,

which includes cardiac catheterization and other

nonsurgical procedures, and (b) surgical, which

may be minor or major. Major surgery was defined

as an extensive surgical procedure involving the

cranium, vertebral column, chest, abdomen, or

pelvic cavity.

Descriptive statistics were used, and they

included median and interquartile ranges. Categor-

ical data were compared using the chi-squared test,

while continuous data were compared using the

Mann–Whitney U test. The relative risk ratio (RR)

with 95% Confidence Intervals (CI) was calculated.

A multivariable logistic regression model to evaluate

factors associated with either major or minor peri-

procedure complications was developed. Factors

associated with complication in the bivariate analy-

sis were considered using a forward selection

procedure with entry of variables into the model

set at P £ 0.05 and removal defined as P £ 0.10 odds

ratio (OR) with 95% CI were calculated. SPSSSPSS version

15.0 software (SPSS Inc, Chicago, IL, USA) was used

for the analysis. Complication rates were reported

per procedure rather than per patient, as some

children had more than one procedure. Complica-

tions were compared for all study subjects. Then the

subgroup that had nonsurgical procedures was

evaluated separately.

Results

During the study period, 68 children (35 boys, 33

girls) underwent 192 procedures. Median (range)

patient age was 7.3 year (14 days–18.4 year) and

weight was 22 kg (2.4–105 kg). Congenital heart

defects were present in 43 (61.8%) patients; 11

children were diagnosed with genetic disorders

(trisomy 21: n = 7, trisomy 18: n = 1, Stickler

syndrome: n = 1, Williams syndrome: n = 1, multi-

ple dysmorphic features of nonspecific pattern:

n = 1). Clinical types of pulmonary hypertension

are detailed in Table 1. The classification is sub-

jective, because some patients had multiple factors

associated with PAH.

Selected details about the procedures, the patients’

preanesthesia status, and the anesthetic techniques

employed are provided in Table 2 and Figure 1.

Prior to their procedure, 17 (8.9%) patients had an

endotracheal tube or tracheostomy tube in situ, 14

(7.3%) had an arterial monitoring catheter, 21

(10.9%) had a gastrostomy tube, and 6 (3.1%) were

on intravenous inotropic support.

30 G.D. WILLIAMS ET AL.

� 2009 Blackwell Publishing Ltd, Pediatric Anesthesia, 20, 28–37

Anesthesia care for the procedures was provided

by pediatric cardiac anesthesiologists [184 (96%)],

pediatric general anesthesiologists [6 (3%)], and

pediatric cardiac intensivists [2 (1%)]. An anesthe-

sia trainee (resident or fellow) was present for 161

(83.9%) procedures. Drugs utilized for induction of

anesthesia included ketamine (52.1%), sevoflurane

(47.9%), propofol (10.9%), and etomidate (3.6%),

(data given as percentage of total procedures).

Anesthetic agents for maintenance included keta-

mine (66.7%), propofol (63%), midazolam (29.1%),

isoflurane (18.2%), and sevoflurane (15.1%). The

percentage of cases per year that received ketamine

averaged about 50 during the first 2 years and

increased to about 90 during the last 2 years of the

study period. Neuromuscular blocking drugs were

administered during 57 (29.7%) and opioids in 71

(37%) procedures. The airway of many patients

[122 (63.5%)] was managed by providing nasal

cannulae oxygen or by application of a facemask.

An endotracheal tube was used in 63 (32.8%)

cases, a laryngeal mask in 4 (2.1%), and a trache-

ostomy tube was in situ in 3 (1.6%) patients.

Intravenous inotropes were administered as

a bolus or by infusion during 28 (14.6%) proce-

dures. Anesthesia records were incomplete in 2

(1%) cases.

Four incidents occurred, all during cardiac cathe-

terization procedures. Vital signs were minimally

affected and no treatment was required. Two were

pulmonary in nature (small emesis on emergence,

cough with contrast injection) and two were transient

catheter-induced cardiac events (atrial arrhythmia,

heart block). Minor (n = 20) and major (n = 9)

Table 1

Clinical classification of pulmonary hypertension in 68 patients

Category n % of total

1. Pulmonary arterial hypertension (PAH)1.1. Idiopathic 12 17.61.3. Associated with:

1.3.1. Collagen vascular disease 2 2.9Juvenile rheumatoid arthritis: n = 1Fibromuscular dysplasia: n = 1

1.3.2. Congenital systemic-to-pulmonary shunts 32 47.1Cardiac malformations: n = 29Noncardiac vascular malformations: n = 3

1.3.3. Portal hypertension (congenital hepatic fibrosis: n = 1) 1 1.51.3.6. Other 4 5.9

Polycystic kidney disease: n = 2Myeloproliferative disorders: n = 2

1.4. Associated with significant venous or capillary involvement1.4.1. Pulmonary veno-occlusive disease 8 11.8

Congenital pulmonary vein stenosis: n = 81.4.2. Pulmonary capillary hemangiomatosis 1 1.5

1.5. Persistent pulmonary hypertension of the newborn 3 4.4Distal pulmonary artery stenoses: n = 3

2. Pulmonary hypertension with left heart disease2.1. Left-sided atrial or ventricular heart disease (cardiomyopathy: n = 6) 6 8.82.2. Left-sided valvular heart disease (congenital valvular defects: n = 3) 3 4.43. Pulmonary hypertension associated with lung diseases and ⁄ or hypoxemia3.1. Chronic obstructive pulmonary disease 4 5.9

Lung disease of prematurity: n = 43.2. Interstitial lung disease 1 1.53.3. Sleep-disordered breathing (obstructive sleep apnea: n = 6) 6 8.83.4. Alveolar hypoventilation disorders (scoliosis: n = 2) 2 2.93.6. Developmental abnormalities (diaphragmatic hernia: n = 2) 3 4.44. Pulmonary hypertension because of chronic thrombotic and ⁄ or embolic disease4.2. Thromboembolic obstruction of distal pulmonary arteries 1 1.55. Miscellaneous 0 0

A total of 89 factors associated with pulmonary hypertension were identified in 68 patients. Source: Simonneau G, Galie N, Rubin LJ et al.Clinical classification of pulmonary hypertension. J Am Coll Cardiol 2004; 43: 5S–12S.

PULMONARY HYPERTENSION: IS KETAMINE SAFE? 31

� 2009 Blackwell Publishing Ltd, Pediatric Anesthesia, 20, 28–37

complications of all patients are detailed in Table 3.

The nature of events that initiated complications were

ascribed as follows: cardiac (n = 14), pulmonary

(n = 10), technical (n = 4), and allergic (n = 1).

There were no procedure-related deaths. One

patient undergoing cardiac catheterization and two

patients undergoing major surgery required cardio-

pulmonary resuscitation during their procedure.

The first of these three cases involved a 15- year-

old girl undergoing heart lung transplantation for

severe idiopathic PAH who probably had a pulmo-

nary hypertensive crisis during tracheal intubation

(developed bradycardia, hypotension, and decreas-

ing oxygen saturations) after receiving midazolam,

ketamine, and rocuronium. The second case was

a 10- year-old boy with an atrial septal defect and

severe PAH who was receiving epoprostenol and

nitric oxide and underwent scoliosis surgery via

a posterior approach. Pulmonary artery pressures

were equal to systemic arterial pressures at the start

of the procedure but over the next 5 h of surgery

twice became markedly suprasystemic for periods

lasting >15 min. The second suprasystemic episode

Table 2

Types of procedures performed (n = 192) and preproceduremedical status of the patients

N (%)

Procedure typesNonsurgical procedures

Cardiac catheterization 128 (66.7%)Other proceduresa 17 (8.9%)

Surgical proceduresMinorb 27 (14.1%)Majorc 20 (10.3%)

Preprocedure medical statusBaseline PAH classificationd

Mild 45 (23.4%)Moderate 70 (36.5%)Severe 77 (40.1%)

Preprocedure medicationsPulmonary vasodilator therapye 113 (58.9%)Heart failure therapyf 58 (30.2%)Diuretic therapy 73 (38%)Anticoagulant therapyg 77 (40.1%)FiO2 > 0.21 57 (29.7%)

Preprocedure hospital statusIn-patient 58 (30.2%)

ASA physical statusII 4 (2.1%)III 113 (58.9%)IV 75 (39.0%)

ASA, American Society of Anesthesiologists. aIncluded comput-erized tomography (n = 11), radionucleotide lung perfusion scans(n = 5), magnetic resonance imaging (n = 4) and pulmonaryfunction tests (n = 1). Four patients underwent two imagingprocedures during the same anesthetic. bIncluded central venouscatheter insertion or removal (n = 18), airway examination orsurgery (n = 4), lumbar puncture and ⁄ or bone marrow aspiration(n = 2), and other procedures (n = 3). cIncluded heart and ⁄ or lungsurgery with cardiopulmonary bypass (n = 14), spine surgery(n = 3), cerebral embolization procedures (n = 2) and omphalocelerepair (n = 1). dPulmonary arterial hypertension (PAH) wasclassified as follows (see text for details): Mild: Pulmonaryvascular resistance index (PVRI) >3–7 Wood units or estimatedsystolic pulmonary artery pressure (PAP) £70% of systolicsystemic arterial pressure (SAP); Moderate: PVRI >7–12 Woodunits or estimated systolic PAP >70–100% of SAP; Severe: PVRI>12 Wood units or estimated systolic PAP >100% of systolic SAP.ePulmonary vasodilator therapy included sildenafil [69 (35.9%)procedures], treprostinil sodium or epoprostenol sodium [63(32.8%) procedures], bosentan [49 (25.5%) procedures], and nitricoxide [9 (4.7%) procedures]. fHeart failure drugs includedcaptopril, carvedilol, digoxin, dobutamine, dopamine, enalapril,epinephrine, esmolol, milrinone, nifedipine. gAnticoagulant drugsincluded coumadin and aspirin.

0

10

20

30

40

50

60

70

Ket

amin

e

Sev

ofl

ura

ne

Ket

amin

e+

sevo

flu

ran

e

Mid

azo

lam

+o

pio

id

Pro

po

fol

Ket

amin

e+

pro

po

fol

Eto

mid

ate

Sev

ofl

ura

ne

+p

rop

ofo

l

n

Drugs used for induction of anesthesia

0102030405060708090

Ket

amin

e+

pro

po

fol

Vo

lati

le

Ket

amin

e+

pro

po

fol+

vola

tile

Ket

amin

e

Pro

po

fol

Mid

azo

lam

+o

pio

id

Ket

amin

e+

vola

tile

Pro

po

fol+

vola

tile

n

Drugs used for maintenance of anesthesia

Figure 1Drugs used for induction and maintenance of anesthesia during192 procedures.

32 G.D. WILLIAMS ET AL.

� 2009 Blackwell Publishing Ltd, Pediatric Anesthesia, 20, 28–37

Ta

ble

3

Su

mm

ary

of

maj

or

and

min

or

per

i-p

roce

du

reco

mp

lica

tio

ns

du

rin

g19

2p

roce

du

res

Typ

eE

ven

tM

anag

emen

tP

roce

dure

Age

(yea

r)P

VR

Min

or

(n=

20)

Car

dia

cS

up

rav

entr

icu

lar

tach

yca

rdia

Ad

eno

sin

eC

ard

iac

cath

eter

izat

ion

8.1

Sev

ere

Car

dia

cS

up

rav

entr

icu

lar

tach

yca

rdia

Res

olv

edw

ith

cath

eter

man

ipu

lati

on

Car

dia

cca

thet

eriz

atio

n9.

2S

ever

eC

ard

iac

Atr

ial

flu

tter

Ex

tern

alca

rdio

ver

sio

nC

ard

iac

cath

eter

izat

ion

10.7

Mo

der

ate

Car

dia

cM

ult

iple

epis

od

essu

pra

ven

tric

ula

rta

chy

card

iaA

den

osi

ne,

exte

rnal

card

iov

ersi

on

Car

dia

cca

thet

eriz

atio

n1.

0S

ever

e

Car

dia

cS

up

rav

entr

icu

lar

tach

yca

rdia

bef

ore

CP

BE

xte

rnal

card

iov

ersi

on

AV

SD

rep

air

(CP

B)

0.3

Mo

der

ate

Car

dia

cS

up

rav

enti

cula

rta

chy

card

iaA

den

osi

ne

Car

dia

cca

thet

eriz

atio

n0.

04S

ever

eC

ard

iac

Pu

lmo

nar

yh

yp

erte

nsi

on

Inh

aled

nit

ric

ox

ide,

epin

eph

rin

eb

olu

sP

AP

VR

rep

air

(CB

P)

3.9

Mo

der

ate

Car

dia

cH

yp

ote

nsi

on

Ep

hed

rin

eb

olu

sC

ard

iac

cath

eter

izat

ion

13.2

Sev

ere

Car

dia

cH

yp

ote

nsi

on

Ph

eny

lep

hri

ne

bo

luse

sC

entr

alli

ne

inse

rtio

n5.

4S

ever

eC

ard

iac

Hy

po

ten

sio

nE

ph

edri

ne

bo

lus

CT

ches

t,lu

ng

scan

6.4

Mo

der

ate

Pu

lmo

nar

yL

ary

ng

osp

asm

wit

hd

esat

ura

tio

nA

irw

ayse

cure

dw

ith

end

otr

ach

eal

tub

eC

ard

iac

cath

eter

izat

ion

18.4

Mo

der

ate

Pu

lmo

nar

yP

ost

op

erat

ive

som

nu

len

ce(c

hro

nic

hy

po

ven

tila

tor)

Nal

ox

on

eto

rev

erse

op

ioid

Car

dia

cca

thet

eriz

atio

n14

.8M

od

erat

e

Pu

lmo

nar

yIn

term

itte

nt

nat

ive

airw

ayo

bst

ruct

ion

Air

way

secu

red

wit

hen

do

trac

hea

ltu

be

Car

dia

cca

thet

eriz

atio

n9.

0M

ild

Pu

lmo

nar

yS

ever

eb

ron

cho

spas

mB

ron

cho

dil

ato

rd

rug

s,ad

mit

ICU

Car

dia

cca

thet

eriz

atio

n4.

1M

ild

Pu

lmo

nar

yC

ou

gh

ing

com

pro

mis

ing

nat

ive

airw

ayA

irw

ayse

cure

dw

ith

end

otr

ach

eal

tub

eC

ard

iac

cath

eter

izat

ion

1.2

Sev

ere

Pu

lmo

nar

yL

ary

ng

osp

asm

com

pro

mis

ing

nat

ive

airw

ayB

ag-m

ask

ven

tila

tio

nC

ard

iac

cath

eter

izat

ion

4.0

Mo

der

ate

Pu

lmo

nar

yIn

term

itte

nt

ob

stru

ctio

no

fn

ativ

eai

rway

Air

way

secu

red

wit

hla

ryn

gea

lm

ask

Car

dia

cca

thet

eriz

atio

n9.

2M

od

erat

eP

ulm

on

ary

Hy

po

ven

tila

tio

nA

ssis

tb

ag-m

ask

ven

tila

tio

nC

ard

iac

cath

eter

izat

ion

13.4

Sev

ere

All

erg

yR

ash

,u

nk

no

wn

etio

log

yS

tero

id,

anti

his

tam

ine

VS

Dre

pai

r(C

PB

)7.

3M

ild

Tec

hn

ical

Scr

ota

lh

emat

om

aC

on

serv

ativ

em

anag

emen

tC

ard

iac

cath

eter

izat

ion

12.4

Sev

ere

Maj

or

(n=

9)C

ard

iac

Acu

teo

nse

tp

uls

eles

sel

ectr

ical

acti

vit

yC

ard

iop

ulm

on

ary

resu

scit

atio

nC

ard

iac

cath

eter

izat

ion

1.8

Sev

ere

Car

dia

cV

entr

icu

lar

tach

yca

rdia

cfr

om

pu

lmo

nar

yh

yp

erte

nsi

ve

cris

iso

nin

du

ctio

no

fan

esth

esia

Car

dio

pu

lmo

nar

yre

susc

itat

ion

Hea

rt-l

un

gtr

ansp

lan

t(C

PB

)15

.2S

ever

e

Car

dia

cV

entr

icu

lar

tach

yca

rdia

fro

mp

ulm

on

ary

hy

per

ten

siv

ecr

isis

du

rin

gsu

rger

yC

ard

iop

ulm

on

ary

resu

scit

atio

n,

rep

osi

tio

nsu

pin

eP

ost

erio

rsc

oli

osi

sre

pai

r9.

8M

od

erat

e

Car

dia

cH

yp

ote

nsi

on

wit

hp

osi

tiv

ep

ress

ure

ven

tila

tio

nP

roce

du

reab

ort

edM

RI

ches

t15

.7S

ever

e

Pu

lmo

nar

yR

igh

tu

pp

erlo

be

atel

ecta

sis

1d

ayp

ost

pro

ced

ure

Re-

adm

itto

ICU

,re

spir

ato

ryth

erap

yC

ard

iac

cath

eter

izat

ion

16.5

Sev

ere

Pu

lmo

nar

yR

eper

fusi

on

lun

gin

jury

afte

rb

allo

on

ang

iop

last

yo

fb

ran

chp

ulm

on

ary

arte

ries

Re-

adm

itto

ICU

,re

spir

ato

ryth

erap

yC

ard

iac

cath

eter

izat

ion

17.2

Mo

der

ate

Tec

hn

ical

Ep

idu

ral

PC

Am

alfu

nct

ion

cau

sin

gh

yp

ov

enti

lati

on

Re-

adm

itto

ICU

,re

spir

ato

ryth

erap

yL

un

gtr

ansp

lan

t(C

PB

)12

.7M

od

erat

e

Tec

hn

ical

Re-

op

erat

ion

for

po

sto

per

ativ

eb

leed

ing

Flu

idre

susc

itat

ion

,su

rgic

alh

emo

stas

isH

eart

-lu

ng

tran

spla

nt

(CP

B)

10.8

Sev

ere

Tec

hn

ical

Fal

sean

eury

smo

ffe

mo

ral

arte

ryS

urg

ical

rep

air

Car

dia

cca

thet

eriz

atio

n1.

2S

ever

e

PA

PV

R,p

arti

alan

om

alo

us

pu

lmo

nar

yv

eno

us

retu

rn;C

PB

,car

dio

pu

lmo

nar

yb

yp

ass;

ICU

,in

ten

siv

eca

reu

nit

;AV

SD

,atr

iov

entr

icu

lar

sep

tal

def

ect;

VS

D,v

entr

icu

lar

sep

tal

def

ect;

CT

,co

mp

ute

rize

dto

mo

gra

ph

yim

agin

g;

PC

A,

pat

ien

tco

ntr

oll

edan

alg

esia

;M

RI,

mag

net

icre

son

ance

imag

ing

.P

ulm

on

ary

hy

per

ten

sio

nca

teg

ori

zed

asm

ild

,m

od

erat

ean

dse

ver

e.M

ild

:P

ulm

on

ary

vas

cula

rre

sist

ance

ind

ex(P

VR

I)>

3–7

Wo

od

un

its

or

esti

mat

edsy

sto

lic

pu

lmo

nar

yar

tery

pre

ssu

re(P

AP

)£7

0%o

fsy

sto

lic

syst

emic

arte

rial

pre

ssu

re(S

AP

);M

od

erat

e:P

VR

I>

7–12

Wo

od

un

its

or

esti

mat

edsy

sto

lic

PA

P>

70–1

00%

of

SA

P;

Sev

ere:

PV

RI

>12

Wo

od

un

its

or

esti

mat

edsy

sto

lic

PA

P>

100%

of

syst

oli

cS

AP

.

PULMONARY HYPERTENSION: IS KETAMINE SAFE? 33

� 2009 Blackwell Publishing Ltd, Pediatric Anesthesia, 20, 28–37

appeared refractory, and epinephrine was adminis-

tered to augment left ventricular output and coro-

nary flow. Ventricular tachycardia with hypotension

ensued shortly thereafter. The third case was a

1.8- year-old girl with mild PAH and congenital

heart disease who for uncertain reasons suddenly

developed pulseless electrical activity during cardiac

catheterization. All three patients survived to dis-

charge from hospital and were alive 5, 6, and

13 months later at last follow-up. The incidence of

hemodynamic instability requiring chest compres-

sions was 0.78% for cardiac catheterization proce-

dures, 10% for major surgical procedures and 1.6%

for all procedures.

Three infants did not survive to discharge from

hospital. One was a neonate with severe PAH

unresponsive to drug therapy who underwent

cardiac catheterization and was found to have

inoperable pulmonary vein stenoses. The following

day, the parents requested that medical support be

withdrawn, and the patient died soon thereafter.

A second infant with atrioventricular septal defect

had five procedures during an 80-day hospitaliza-

tion and died 14 days after the last procedure from

complications of PAH. The third infant had trisomy

21 syndrome and lung disease of prematurity (born

at 31 weeks gestational age) and died 17 days after

cardiac catheterization from respiratory failure with

sepsis. Postdischarge information was available for

65 patients. Median duration of follow-up was

27.7 months (range 0–66.3 months). Three deaths

occurred at 50, 121, and 612 days after hospital

discharge.

Factors associated with increased risk of compli-

cations (major + minor) when all 192 procedures

were included in the bivariate analysis were type of

procedure (major surgery) (P = 0.017; RR 3.03 (95%

CI 1.32–6.94), airway instrumentation with laryngeal

mask or endotracheal tube (P = 0.004; RR 1.53, 95%

CI 1.02–2.31), and opioid administration during

anesthesia (P < 0.001; RR 1.77, 95% CI 1.22–2.58).

Preprocedure vasodilator therapy (P = 0.012; RR

0.43, 95% CI 0.22–0.81) and propofol administration

during anesthesia (P = 0.01: RR0.82, 95% CI 0.58–

1.15) were associated with decreased risk of compli-

cations. Ketamine administration was not associated

with risk of complications (P = 0.963). Because the

patient’s airway was instrumented during all major

surgeries (the procedure group associated with

significantly increased risk) and propofol was only

administered during 1 (5%) major surgery, the

bivariate analysis was repeated with nonsurgery

cases only (n = 145). Factors associated with the

risk of complications (major + minor) were airway

instrumentation (P = 0.022; RR 1.55, 95% CI 0.92–

2.62) and opioid administration (P = 0.026; RR 1.97,

95% CI 1.14–3.40). Preprocedure vasodilator ther-

apy decreased the risk of complications (P = 0.004;

RR 0.46, 95% CI 0.23–0.92). Ketamine, propofol,

and volatile agents were not associated with

complications.

The multivariable logistic regression model dem-

onstrated the following factors to be independently

associated with complications (major + minor):

preprocedure vasodilator therapy (OR 0.31, 95% CI

0.13–0.70), major surgery (OR 3.1, CI 1.1–9.0). Keta-

mine was not significantly associated with compli-

cations (P = 0.36), neither were opioids, propofol nor

airway instrumentation.

Pulmonary vascular reactivity testing was per-

formed during 40 of the cardiac catheterizations, and

in these cases, the occurrence of complications (6, all

minor) did not differ between vasodilator respond-

ers and nonresponders.

Patients received ketamine and ⁄ or volatile anes-

thesia during 180 (93.8%) of all procedures.

To examine the hypothesis that volatile anesthesia

may mitigate the possible effect of ketamine to

increase PVR and the risk of adverse events, we

grouped these cases as follows: ketamine and no

volatile agents (Group K, n = 57); volatile agents and

no ketamine (Group V, n = 31); ketamine and vol-

atile agents (Group K + V, n = 92). The occurrence

of complications were Group K: minor 6 (10.5%),

major 0 (0%); Group V: minor 4 (12.9%), major 2

(6.5%); Group K + V: minor 10 (10.9%), major 7

(7.6%). The incidence of complications did not differ

between groups (P = 0.341). Again, to reduce the

influence of procedure type on complications, non-

surgery procedures were examined to test whether

the occurrence of complications (major + minor)

was influenced by the type of anesthetic agents

administered (Table 4). Occurrence of complications

was similar irrespective of the anesthetic drug

combination employed.

The peri-procedural care of PAH patients under-

going nonsurgical procedures was relatively

uniform. Therefore, these cases (n = 145) were

34 G.D. WILLIAMS ET AL.

� 2009 Blackwell Publishing Ltd, Pediatric Anesthesia, 20, 28–37

regarded as one group in order to ascertain whether

there were patient characteristics or aspects of peri-

procedure management that differed between

patients that did and did not receive ketamine

(Table 5). Patients receiving ketamine were not

different than patients that did not receive ketamine

with regard to age, weight, diagnosis, and proce-

dure. More patients in the ketamine group tended to

have severe PAH (44% vs 28%, P = 0.062). Fewer

patients receiving ketamine had their airway instru-

mented (16% vs 69%, P < 0.001), more received

antisialogogue medications (22% vs 0%, P = 0.005)

and fewer were administered opioids (25% vs 74%,

P < 0.0001). Airway instrumentation was more

likely in younger patients (median 3.0 year vs

9.0 year, P < 0.001) and less likely in patients with

severe PAH (27% vs 48%, P = 0.035).

Discussion

The new findings of this retrospective study of

children with PAH who underwent anesthesia for

diagnostic and therapeutic procedures are as

follows: (i) ketamine administration is not associated

with increased complications, either when adminis-

tered as the sole anesthetic agent or when combined

with propofol or volatile anesthesia; (ii) rates for

cardiopulmonary resuscitation and mortality were

Table 4

Peri-procedural complications that occurred during nonsurgeryprocedures (n = 145), grouped according to the anesthetic drugsadministered

Anesthetic agents n

Complications

Minor Major All (%)

Ketamine only 10 1 0 1 (10)Ketamine + propofol 37 5 0 5 (13.5)Ketamine + volatile 17 1 1 2 (11.8)Propofol onlya 3 0 0 0 (0)Propofol + volatilea 14 2 0 2 (14.3)Volatile onlya 7 1 2 3 (42.9)Ketamine + propofol+ volatile

52 6 2 8 (15.4)

Other drugsb 5 0 0 0 (0)Totals 145 16 5 21 (14.5)

The occurrence of all complications did not between groups (chi-square test, P > 0.05). aData grouped together to permit statisticalanalysis. bUnable to include in the statistical analysis.

Table 5

A comparison between patients that did and did not receive ketamine anesthesia for nonsurgical procedures (n = 145)

Characteristic Ketamine (n = 116) No ketamine (n = 29) P value

Age at procedure (median, year) 8.0 6.4 nsWeight at procedure (median, kg) 23.2 19.0 nsCardiac catheterization 104 (89.7%) 24 (82.8%) nsOther nonsurgical procedure 12 (10.3%) 5 (17.2%) nsDiagnosis of congenital heart disease 33 (28.4%) 8 (27.6%) nsInvasive monitoring of systemic arterial blood pressure 18 (15.5%) 2 (6.9%) nsAirway instrumented (endotracheal tube or laryngeal mask) 25 (21.6%) 20 (69.0%) <0.001Premedication with midazolam 77 (66.4%) 17 (58.6%) nsInduction or maintenance with propofol 89 (76.7%) 17 (58.6%) nsInduction or maintenance with volatile agents 69 (59.5%) 21 (72.4%) nsOpiate administered 22 (19.0%) 18 (62.1%) <0.001Antisialogogue administered 26 (22.4%) 0 (0%) 0.005Intravenous introp administered 9 (7.8%) 0 (0%) nsComplication (minor + major) 16 (13.8%) 5 (17.2%) nsComplication requiring external cardiac compressions 1 (0.9%) 0 (0%) nsEmesis in the postanesthesia care unita 11 (13.6%) 3 (15.0%) nsAnti-emetic administered in postanesthesia care unitb 10 (12.3%) 4 (20.0%) nsPVR group: mild 27 (23.3%) 10 (34.5%) nsPVR group: moderate 37 (31.9%) 11 (37.9%) nsPVR group: severe 52 (44.8%) 8 (27.6%) nsSame day discharged home postprocedure 33 (28.4%) 7 (24.1%) nsSame day admission to intensive care unit postprocedure 26 (22.4%) 6 (20.7%) nsHospital stay (median, h) 24.0 36.8 ns

Data reported as n (%) unless otherwise stated. ns: No significant difference between groups (chi-squared test, P > 0.05). aInformationabout postprocedure nausea and emesis during the patient’s stay in the postanesthesia care unit was available for 101 of 145 cases of which81 received ketamine and 20 did not receive ketamine.

PULMONARY HYPERTENSION: IS KETAMINE SAFE? 35

� 2009 Blackwell Publishing Ltd, Pediatric Anesthesia, 20, 28–37

lower than those previously reported for children

with PAH; (iii) peri-procedure risk factors included

major surgery, airway instrumentation and opioid

administration; (iv) during nonsurgical procedures,

exposure to airway instrumentation and opioid

administration was less likely when patients

received ketamine.

PAH was a predictor of perioperative death in

adult patients undergoing cardiac (9) and noncar-

diac surgery (10). In a retrospective study of

pediatric and adult patients with congenital heart

disease undergoing noncardiac surgery, PAH was a

predictor of perioperative morbidity (12). Similarly,

preoperative PAH was a significant risk factor for

postoperative in-hospital death in infants and chil-

dren undergoing open-heart surgery (14). Taylor

and colleagues studied 70 children with PAH who

underwent anesthesia and cardiac catheterization

and found a 5.7% risk of cardiac massage and

a mortality of 1.4% (15). Carmosino et al. (13)

reviewed 246 cardiac catheterization and noncardiac

surgical procedures in children with PAH and

noted an incidence of cardiac arrest of 2.1% and

a mortality rate of 1.4% for cardiac catheterization.

Our data for children undergoing cardiac catheter-

ization (cardiac arrest 0.8%, mortality 0%) were

encouraging and better than previous reports

(13,15), but they remain inferior to the 0.5% cardiac

arrest risk reported for all children with heart

disease who underwent cardiac catheterization

(23). A perioperative 30-day mortality of 8.5% after

cardiac surgery was reported for children at risk of

PAH (14), whereas 7.1% of our study patients

required cardiopulmonary resuscitation following

open-heart surgery and none died. Complications

occurred at all time phases during and after the

procedure. Exceptional vigilance throughout the

peri-procedural period is warranted for patients of

this high-risk group.

Baseline suprasystemic PAH has been reported to

be a significant predictor of major complications

(13). Our findings suggested pulmonary vasodilator

therapy ameliorated risk and perhaps this effect

explains why we found no association between

complications and severity of PAH. Airway instru-

mentation has been associated with life-threatening

or fatal escalation of pulmonary artery pressures

(25), and we found airway management was pre-

dictive of complications. Opioids can favorably

modify the adverse hemodynamic responses to

airway instrumentation, but they may also increase

PVR secondary to respiratory depression. In our

study, patients with PAH who underwent major

surgery were particularly challenging to manage.

Perturbations of homeostasis are quite frequent

during major surgery; they resulted in pulmonary

hypertensive crises in two study patients. Long

duration of anesthesia was predictive of morbidity

in adults with PAH (12).

The increased risk of perioperative complications

in patients with PAH is attributed largely to

the adverse consequences of PAH (8). Therefore,

the finding that ketamine was not associated with

increased risk of peri-procedural complications was

consistent with our previous report that ketamine (in

the presence of sevoflurane) did not increase PVR

in children with PAH (22). Additionally, the admin-

istration of volatile anesthetic agents and ⁄ or propo-

fol concomitantly with ketamine provided no

advantage over ketamine anesthesia alone with

regard to the occurrence of complications. In terms

of safety, ketamine appears to be an acceptable

option for children with PAH.

Patients who received ketamine during nonsur-

gical procedures were less likely to have their

airways instrumented. Our earlier work showed

that ketamine maintained PaO2 and did not alter

cardiac index, mean systemic arterial pressure,

arterial pH or PaCO2 in patients breathing sponta-

neously via their native airway. Respiratory effort

seemed relatively preserved (22). While this

anesthetic technique was safe and perhaps advan-

tageous with regard to airway manipulation and

PAH, we recommend meticulous care to avoid

hypoxia and hypercapnia, because these conditions

increase PVR. The opioid-sparing effect of ketamine

was anticipated because the drug has analgesic

properties.

This investigation has the limitations of a retro-

spective study; data may be incomplete or missing.

Fortunately, complications were entered real-time

into a database by the attending physicians provid-

ing care, thus reducing concern that important

complications were not included. Another limitation

was the relatively small number of procedures.

There may be inadequate statistical power to iden-

tify a significant association between some variables

and complications (Type II error).

36 G.D. WILLIAMS ET AL.

� 2009 Blackwell Publishing Ltd, Pediatric Anesthesia, 20, 28–37

References1 Galie N, Rubin LJ. Introduction: new insights into a chal-

lenging disease. A review of the Third World Symposium onPulmonary Arterial Hypertension. J Am Coll Cardiol 2003; 43:

1S.2 Simonneau G, Galie N, Rubin LJ et al. Clinical classification of

pulmonary hypertension. J Am Coll Cardiol 2004; 43: 5S–12S.3 Tulloh R. Management and therapeutic options in pediatric

pulmonary hypertension. Expert Rev Cardiovasc Ther 2006; 4:

361–374.4 Fischer LG, Van Aken H, Burkle H. Management of pulmo-

nary hypertension: physiological and pharmacologicalconsiderations for anesthesiologists. Anesth Analg 2003; 96:

1603–1616.5 Blaise G, Langleben D, Hubert B. Pulmonary arterial hyper-

tension: pathophysiology and anesthetic approach. Anesthesi-ology 2003; 99: 1415–1432.

6 Rashid A, Ivy D. Severe paediatric pulmonary hyperten-sion: new management strategies. Arch Dis Child 2005; 90:

92–98.7 Friesen RH, Williams GD. Anesthetic management of children

with pulmonary arterial hypertension. Pediatr Anesth 2008; 18:

208–216.8 MacKnight B, Martinez EA, Simon BA. Anesthetic manage-

ment of patients with pulmonary hypertension. Semin Cardio-thorac Vasc Anesth 2008; 12: 91–96.

9 Reich DL, Bodian CA, Krol M et al. Intraoperative hemo-dynamic predictors of mortality, stroke, and myocardialinfarction after coronary artery bypass surgery. Anesth Analg1999; 89: 814–822.

10 Ramakrishna G, Sprung J, Ravi BS et al. Impact of pulmonaryhypertension on the outcomes of noncardiac surgery. J Am CollCardiol 2005; 45: 1691–1699.

11 Robitaille A, Denault AY, Couture P et al. Importance of rela-tive pulmonary hypertension in cardiac surgery: the meansystemic-to-pulmonary artery pressure ratio. J CardiothoracVasc Anesth 2006; 20: 331–339.

12 Warner MA, Lunn RJ, O’Leary PW et al. Outcomes of non-cardiac surgical procedures in children and adults withcongenital heart disease. Mayo Clin Proc 1998; 73: 728–734.

13 Carmosino MJ, Friesen RH, Doran A et al. Perioperativecomplications in children with pulmonary hypertensionundergoing noncardiac surgery or cardiac catheterization.Anesth Analg 2007; 104: 521–527.

14 Bando K, Turrentine MW, Sharp TG et al. Pulmonary hyper-tension after operations for congenital heart disease: analysisof risk factors and management. J Thorac Cardiovasc Surg 1996;112: 1600–1609.

15 Taylor CJ, Derrick G, McEwan A et al. Risk of cardiac cathe-terization under anaesthesia in children with pulmonaryhypertension. Br J Anaesth 2007; 98: 657–661.

16 Heller AR, Litz RJ, Koch T. A fine balance, one-lung ventilationin a patient with Eisenmenger syndrome. Br J Anaesth 2004; 92:587–590.

17 Spotoft H, Horshin JD, Sorensen MB et al. The cardiovasculareffects of ketamine for induction of anesthesia in patients withvalvular heart disease. Can Anaesth Soc J 1979; 26: 463–467.

18 Berman W, Fripp RR, Rubler M et al. Hemodynamic effects ofketamine in children undergoing cardiac catheterization.Pediatr Cardiol 1990; 11: 72–76.

19 Wolfe RR, Loehr JP, Schaffer MS et al. Hemodynamic effects ofketamine, hypoxia and hyperoxia in children with surgicallytreated congenital heart disease residing greater than or equalto 1,200 meters above sea level. Am J Cardiol 1991; 67: 84–87.

20 Morray JP, Lynn AM, Herndon PS et al. Hemodynamic effectsof ketamine in children with congenital heart disease. AnesthAnalg 1984; 63: 895–899.

21 Hickey PR, Hansen DD, Cramolini GM et al. Pulmonary andsystemic hemodynamic responses to ketamine in infants withnormal and elevated pulmonary vascular resistance. Anesthe-siology 1985; 62: 287–293.

22 Williams GD, Philip BM, Chu LF et al. Ketamine does notincrease pulmonary vascular resistance in children with pul-monary hypertension undergoing sevoflurane anesthesia andspontaneous ventilation. Anesth Analg 2007; 105: 1578–1584.

23 Bennett D, Marcus R, Stokes M. Incidents and complicationsduring pediatric cardiac catheterization. Pediatr Anesth 2005;15: 1083–1088.

24 Galie N, Torbicki A, Barst R et al. Guidelines on diagnosis andtreatment of pulmonary arterial hypertension. The Task Forceon Diagnosis and Treatment of Pulmonary Arterial Hyper-tension of the European Society of Cardiology. Eur Heart J2004; 25: 2243–2278.

25 Burrows FA, Klinck JR, Rabinovitch M et al. Pulmonaryhypertension in children: perioperative management. CanAnaesth Soc J 1986; 33: 606–628.

Accepted 27 August 2009

PULMONARY HYPERTENSION: IS KETAMINE SAFE? 37

� 2009 Blackwell Publishing Ltd, Pediatric Anesthesia, 20, 28–37