c

ccaE11aSc

ctmoa

Pharmacokinetic Study of Vitreous and SerumConcentrations of Triamcinolone Acetonide After

Posterior Sub-Tenon’s Injection

KYLE KOVACS, SUSHANT WAGLEY, MATTHEW T. QUIRK, OLGA M. CERON, PAOLO A. SILVA,

RAVINDER J. SINGH, HOVHANNES J. GUKASYAN, AND JORGE G. ARROYO

as1r

Icri

● PURPOSE: To compare a theoretical pharmacokineticmodel of triamcinolone acetonide after posterior sub-Tenon’s injection with experimental serum and undi-luted vitreous triamcinolone acetonide concentrationsobtained during pars plana vitrectomy.● DESIGN: Clinical-practice, prospective, interventionalase series study.

● METHODS: This study compared computer-modeledtriamcinolone acetonide diffusion after posterior sub-Tenon’s injection with triamcinolone acetonide levels inexperimental undiluted vitreous and serum samples from57 patients undergoing vitrectomy assessed via massspectrometry and high-pressure liquid chromatography.At least 5 pairs of samples were collected at each of 7time points (1 day, 3 days, and 1, 2, 3, 4, and 8 weeks)after triamcinolone acetonide injection, with 6 controlswithout injection. Cortisol levels were measured in 31sets of samples.● RESULTS: The theoretical model predicted that triam-inolone acetonide levels in systemic blood, vitreous, andhoroidal extracellular matrix would plateau after 3 dayst 15 ng/mL, 227 ng/mL and 2230 ng/mL, respectively.xperimental vitreous levels of triamcinolone peaked at11 ng/mL at day 1, then reached a plateau in the range5 to 25 ng/mL, while serum triamcinolone levels peakedt day 3 near 35 ng/mL and plateaued near 2 to 8 ng/mL.erum triamcinolone and cortisol levels were inverselyorrelated (Spearman �0.42, P � .02).

● CONCLUSIONS: The theoretical model predicts effi-ient delivery of triamcinolone acetonide from the pos-erior sub-Tenon’s space to the extracellular choroidalatrix. The experimental findings demonstrate low levelsf serum triamcinolone that alter systemic cortisol levelsnd higher vitreous levels lasting at least 1 month. Both

Accepted for publication Oct 19, 2011.From Beth Israel Deaconess Medical Center, Division of Ophthalmol-

ogy, Boston, Massachusetts (K.K., S.W., M.T.Q., O.M.C., J.G.A.); JoslinDiabetes Center, Beetham Eye Institue, Boston, Massachusetts (P.A.S.);Harvard Medical School, Department of Ophthalmology, Boston, Mas-sachusetts (P.A.S., J.G.A.); Mayo Clinic, Department of LaboratoryMedicine and Pathology, Rochester, Minnesota (R.J.S.); and PfizerGlobal Research and Development, La Jolla Laboratories, San Diego,California (H.J.G.).

Inquiries to Jorge G. Arroyo, Retina Service, Director, Division ofOphthalmology, Beth Israel Deaconess Medical Center, 330 BrooklineAve, Shapiro 5th floor, Boston, MA 02215; e-mail: jarroyo@bidmc.

harvard.edu

© 2012 BY ELSEVIER INC. A0002-9394/$36.00doi:10.1016/j.ajo.2011.10.021

ssessments support trans-scleral delivery of posteriorub-Tenon’s triamcinolone. (Am J Ophthalmol 2012;53:939–948. © 2012 by Elsevier Inc. All rightseserved.)

T RIAMCINOLONE ACETONIDE IS A LONG-ACTING

corticosteroid depot that has been used extensivelyin ophthalmology. Traditionally, it has been deliv-

ered by periocular injection into the subconjunctival orposterior sub-Tenon’s space to treat a variety of conditionsranging from uveitis to cystoid macular edema. Morerecently, triamcinolone has been injected directly into thevitreous in the treatment of such conditions as diabeticmacular edema, retinal vein occlusions, neovascular age-related macular degeneration (AMD), and reducing refrac-tory cystoid macular edema in the short term, particularlyin patients with uveitis.1–5 However, intravitreal-deliveredsteroid carries significant complication risks such as theelevation of intraocular pressure (IOP) and progression ofcataract.6 Thirty percent of patients experienced elevatedOP requiring antiglaucoma medications and reportedataract progression in 13% to 30% of cases that oftenequired surgery within 1 year of triamcinolone acetonidenjection.4,7–9 In disease processes that primarily involve

the choroid, such as macular degeneration, posterior sub-Tenon’s injection of triamcinolone may be a lower-riskmethod for long-acting steroid delivery compared to intra-vitreal injection. Therefore, a better understanding of theocular and systemic pharmacodynamics of posterior sub-Tenon’s triamcinolone may be useful in devising noveltreatment strategies for inflammatory, oxidative, and isch-emic choroidal diseases such as AMD.

Recent clinical investigations of the relative therapeuticmerits of intravitreal and sub-Tenon’s triamcinolone ace-tonide in patients with uveitic macular edema have showninconclusive results, with Choudhry and Ghosh finding nosignificant difference in final visual acuity outcomes.10

Understanding the mechanism at work in both modes ofdelivery might help to evaluate their clinical relevance. Ingeneral, vitreous steroid concentrations have been lowerand more variable following sub-Tenon’s delivery.11–13

However, the heterogeneity of triamcinolone pharmacoki-netic studies has made these comparative assessments

tenuous. Studies have varied greatly in follow-up duration,

LL RIGHTS RESERVED. 939

fFibdtTmsatfiw

study size, study design, the amount of corticosteroidinitially delivered, and the differing methods of steroidconcentration analysis. Similarly designed studies by Ma-son and associates14 and Beer and associates15 have dif-ered by numerous orders of magnitude in their findings.urthermore, the systemic absorption of the steroid follow-ng sub-Tenon’s injection has also become a major concernecause of its possible influence on patients with metaboliciseases such as diabetes. Recent studies have indicatedhat systemic absorption of triamcinolone following sub-enon’s injection occurs at levels sufficient to affectetabolic control in some patients, which would pose a

ignificant risk factor in their use.16 Degenring and associ-tes warn of the systemic effects of triamcinolone ace-onide following their intravitreal injections, despitending less systemic triamcinolone absorption comparedith sub-Tenon’s assessments.17

In this study, we sought to answer the following ques-tions: 1) What are the pharmacokinetics of posteriorsub-Tenon’s triamcinolone acetonide in the serum andvitreous? 2) What is the route of entry of posteriorsub-Tenon’s triamcinolone acetonide into the eye? 3) Arethere systemic side effects of posterior sub-Tenon’s triam-cinolone acetonide delivery? To answer these questions,we developed an ocular simulation model that could beused to predict the serum, choroidal, and vitreous concen-trations of triamcinolone acetonide after sub-Tenon’s in-jection. We then compared these predictions with thetriamcinolone acetonide levels in both serum and undi-luted vitreous samples obtained at the time of vitrectomyin patients who had previously received sub-Tenon’s ste-

FIGURE 1. Schematic showing the multiple compartments ubidirectional or unidirectional permeability constant is associatean infinite sink. ECM � extracellular matrix.

roid injection. In order to study the effects of systemic

AMERICAN JOURNAL OF940

triamcinolone on the pituitary adrenal axis we also mea-sured serum cortisol levels. To the best of our knowledgethis study is the most comprehensive study to date on theserum and vitreous pharmacokinetics after posterior sub-Tenon’s injection for the time points 1 day to 56 days. Inaddition, this is the first study to assess the effects ofsystemic triamcinolone acetonide on cortisol levels. Weanticipate that this study will to help clarify much of thedebate surrounding the mechanism of action and thepervasive variability of the literature.

METHODS

IN SILICO MODELING OF POSTERIOR SUB-TENON’S TRIAM-

cinolone acetonide was carried out using an ocular simu-lation (OCUSIM; Bend Research, Inc, Bend, Oregon,USA) computer program. The governing principles andassumptions used were: 1) ocular bioavailability deter-mines overall systemic absorption and bioactivity; 2) siteof deposition and physicochemical properties of triamcin-olone acetonide may dictate rates of dissolution,absorption, metabolism, and elimination; 3) posterior sub-Tenon’s triamcinolone acetonide has high hepatic first-pass inactivation, whereas there is no first-pass metabolismvia ocular route; and 4) no standardized methods exist fortesting dissolution rates from sub-Tenon’s injection. Themultiple compartments used in this model are shown inFigure 1. Physiological and anatomic parameters for theocular simulation were customized to model the human eyeby entering appropriate volumes of distribution and surface

in the Ocular Simulation Program and their interactions. Ah each arrow. The compartment marked “Excretion” serves as

sedd wit

areas of posterior compartment tissues.18–20 The physico-

OPHTHALMOLOGY MAY 2012

fvu

at1aoetfi

psdmd

tvvuOitfbmqw

oi

2pitsffTtifrintwwscoqt

11gcilnd

chemical and physiological parameters of triamcinoloneacetonide used in the ocular simulation were its logP(estimated partition coefficient in an n-octanol/water sys-tem: value 2.53),21 ocular pigment binding (estimated,ollowing the method by Koeberle and associates,22 to beery low from equilibrium sepia melanin association: val-es with capacity of 7 � 104 (ng/g)(mL), Kon of 3 � 10�6

(mL/ng)(s), and Ka of 7 � 10�9 mL/ng), and systemicclearance (37 L/hr and t½ of 2 hr) following intravenousadministration.23,24 Formulation characteristics weredapted from Kenalog-40 (Bristol-Myers Squibb, Prince-on, New Jersey, USA): 1) saturation concentration of.3 � 104 ng/mL,25 2) spherical suspension particles withn average radius of 1 � 104 nm,26–28 3) total dose potencyf 40 mg/mL, 4) particle density of 1.2 g/mL, and 5)stimated dissolution rate of 1 � 10�6 cm2/s.28 Based onhese parameters, theoretical calculations were carried outor a 40-mg posterior sub-Tenon’s triamcinolone acetonidenjection.

To reduce postoperative inflammation and cellular re-roliferation, patients are routinely given preoperativeub-Tenon’s triamcinolone acetonide between 1 and 56ays prior to vitrectomy surgery for epiretinal membrane oracular hole. In our study, this procedure was performed as

escribed by Nozik.29 Briefly, a cotton tip soaked inproparacaine 0.5% is placed under the lid in the supero-temporal quadrant of the eye for approximately 5 minutes.One drop of a fourth-generation floroquinolone is giventopically every minute times 3. A ¾-inch 25-gauge needleon a 1-cc syringe is placed in the fornix superotemporallywith the bevel towards the globe. The tip of the needle ismoved side to side as it is moved posteriorly along thesclera, ending in a position immediately behind the mac-ula. During the introduction of the needle, the limbalvessels are carefully observed for synchronous lateral move-ments that would indicate engagement of the needle withthe sclera. Forty milligrams of triamcinolone acetonide(Kenalog-40; Bristol-Myers Squibb), in 1 cc, are theninjected slowly into the posterior sub-Tenon’s space.

Undiluted vitreous and serum samples were collectedbased on a previously established technique.30 Prior tourning on the infusion to start the vitrectomy, theitreous cutter was used to obtain 0.5 to 1 cc of undiluteditreous specimen. A 3-cc syringe on a 3-way stopcock wassed to provide manual aspiration of the specimen.nce a sufficient volume of vitreous was obtained, the

nfusion was turned on and the stopcock was rotated andhe vitrectomy was continued. Patients were monitoredor complications, specifically severe increases in IOPetween posterior sub-Tenon’s administration and im-ediately postoperatively. The number of patients re-

uiring pharmacologic intervention for elevated IOPas recorded.At least 5 pairs of samples (serum and undiluted vitre-

us) were collected as close as possible to each of 7

ntended time points after injection: 1 day, 3 days, 1 week, n

POSTERIOR SUB-TENON TRIAMCINOLONVOL. 153, NO. 5

weeks, 3 weeks, 4 weeks, and 8 weeks. Six additionalairs of samples from patients who did not receive a steroidnjection were obtained and served as controls, yielding aotal of 57 samples in all. Following collection, serumamples were immediately spun in a centrifuge at 1600 Gor 10 minutes. The supernatant was then extracted androzen at �80 C along with the collected vitreous samples.he frozen samples were later shipped overnight on dry ice

o the Endocrine Medical Laboratory of the Mayo Clinicn Rochester, Minnesota. There the samples were analyzedor triamcinolone acetonide levels using liquid chromatog-aphy – tandem mass spectrometry. Briefly, the stablesotope of triamcinolone-d1 acetonide-d6 228 pmol (100g) was added to 50 �L of the fluid. Acetonitrile was addedo each sample to precipitate proteins. The supernatantsere then extracted with 4.0 mL of methylene chloride,ashed, and finally dried. Fifteen microliters of the recon-

tituted extract was then injected onto a reversed-phaseolumn and analyzed using a tandem mass spectrometerperating in the positive mode (API 3000 Sciex triple-uadripole mass spectrometer, ABI-Sciex, Toronto, On-ario, Canada).31 The minimum level of detection for this

system was 0.05 ng/mL. Cortisol levels were also assessedusing liquid chromatography – tandem mass spectrometryfor the 31 samples sent in the final shipment to theEndocrine Medical Laboratory in Minnesota.32

Preliminary statistical analyses were performed using Stat-View 5 for Macintosh and Windows (SAS Institute, Inc,Cary, North Carolina, USA), and final analyses were run onSPSS software version 12.0 for Windows (SPSS Inc, Chi-cago, Illinois, USA). Spearman correlation tests were used toassess possible relationships between serum triamcinoloneconcentrations, vitreous triamcinolone concentrations, serumcortisol concentrations, and vitreous cortisol concentrations.A t test was used to see if cortisol levels were significantlydepressed from physiological levels.

RESULTS

THE EXPERIMENTAL RESULTS FOR THE SERUM AND VITRE-

ous samples are shown in the Table. Vitreous and serumsamples were collected during vitrectomies at the followingtime intervals: 1, 3, 7, 14, 21, 28, and 56 days afterinjection. The mean observed time intervals betweeninjection and sample collection for each group were 1 � 0days (1-day group), 3.88 � 0.64 days (3-day group), 6.44 �.01 days (7-day group), 12.83 � 0.98 days (14-day group),9.29 � 2.21 days (21-day group), 32 � 4.38 days (28-dayroup), and 61 � 3.74 days (56-day group). Vitreousoncentrations of triamcinolone peaked 1 day followingnjection at 111.43 � 138.93 ng/mL. Vitreous triamcino-one concentration then plateaued at around 15 to 25g/mL, reaching a minimum of 7.99 � 7.0 ng/mL at 56ays. Serum concentrations reached a maximum of 35

g/mL at day 3, with a large standard deviation of 88.21

E ACETONIDE PHARMACOKINETICS 941

0iii

8

to.dw.ast

mpp

ng/mL caused by 1 sample at 235 ng/mL. Serum concen-trations plateaued around 5 ng/mL from day 3 onwards.Experimental results are plotted in Figure 2, with thestandard error of the mean highlighting the large datavariability.

Thirty-one sets of samples were also analyzed for theircortisol levels. Vitreous concentrations were found to dipfrom their initial mean of 0.31 � 0.15 �g/dL to the 0.18 to.25 �g/dL range over the first week following sub-Tenon’snjection, before gradually rising over time (Table). Phys-ologic serum cortisol concentrations, indicated by pre-njection controls, were recorded at 15.54 � 8.35 �g/dL.

On postinjection day 1, serum cortisol concentrationwas 10.15 � 1.94 �g/dL, with concentrations dipping to

FIGURE 2. Vitreous and serum triamcinolone acetonide conc

TABLE. Vitreous and Serum TriamcinoloneTime With Standard Deviations and t

Time (Days)

Vitreous

Triamcinolone

Acetonide (ng/mL)

Tria

Aceto

(Control) 0

n � 6

1 111.43 � 138.93 7.9

n � 6

3 20.20 � 27.45 35.0

n � 8

7 17.38 � 12.43 2.1

n � 9

14 30.62 � 32.98 4.4

n � 6

21 19.97 � 24.17 1.4

n � 7

28 22.42 � 36.63 7.7

n � 8

56 7.99 � 7.01 5.3

n � 5

aThere are fewer cortisol values since cortis

shipment of 31 samples.

.33 � 4.59 �g/dL on day 7. w

AMERICAN JOURNAL OF942

Serum concentrations of triamcinolone were also foundo be significantly correlated with vitreous concentrationsf triamcinolone (Spearman coefficient � 0.65, P �

0001). Cortisol levels in serum samples were found toecrease with an increase in systemic corticosteroid levels,ith a Spearman correlation coefficient of �0.42 (P �

02). When systemic levels of triamcinolone were highestt day 3, levels of cortisol were found to have variedignificantly from their physiologic baseline as measured inhe control samples (P � .01).

Two cases were found in which patients required phar-acologic intervention for elevated IOP following both

osterior sub-Tenon’s triamcinolone injection and parslana vitrectomy. In both cases the IOP was controlled

tions at various time points.

tonide and Cortisol Concentrationsa Overumber of Samples in Each Group (n)

lone

g/mL)

Vitreous Cortisol

(�g/dL)

Serum Cortisol

(�g/dL)

0.31 � 0.15 15.54 � 8.35

n � 5 n � 6

.82 0.20 � 0.04 10.15 � 1.94

n � 3 n � 3

8.21 0.19 � 0.22 8.61 � 2.65

n � 5 n � 5

.51 0.24 � 0.16 8.33 � 4.59

n � 4 n � 5

.50 0.41 13.79 � 2.02

n � 1 n � 2

.84 0.27 � 0.26 11.18 � 3.50

n � 4 n � 5

9.83 0.47 � 0.50 12.59 � 4.24

n � 3 n � 3

.84 1.11 31.60

n � 1 n � 1

ncentrations were only assessed in the final

entra

Acehe N

Serum

mcino

nide (n

0

n � 6

0 � 8

n � 6

0 � 8

n � 7

7 � 2

n � 7

3 � 4

n � 6

8 � 1

n � 7

6 � 1

n � 9

2 � 8

n � 5

ol co

ith drops alone. Although there were a few other cases

OPHTHALMOLOGY MAY 2012

with slightly elevated IOP on postoperative day 1, thesewere thought to be surgically induced since they resolved

FIGURE 3. Theoretical systemic and intraocular levels of triampredicted using the ocular simulation. (Top) Graph includes learound 2230 ng/mL. (Bottom) Graph has removed the choroidlevels plateau at 15 ng/mL.

without medical intervention and were mild in severity.

POSTERIOR SUB-TENON TRIAMCINOLONVOL. 153, NO. 5

The theoretical model developed through the ocularsimulation program predicted that triamcinolone levels in

lone acetonide after 40 mg posterior sub-Tenon’s injection asfor the choroid extracellular matrix (ECM), which plateau at. The vitreous levels plateau at 227 ng/mL, while the serum

cinovelsECM

the serum and eye would reach their maximum level

E ACETONIDE PHARMACOKINETICS 943

tm0tBl

wsldrsassgdtTmhdmwfia

within 36 hours after posterior sub-Tenon’s injection andremain at that level for 1 month. The highest levels oftriamcinolone were predicted in the choroidal extracellu-lar matrix (ECM) with a plateau at 2230 ng/mL (Figure 3,Top). The vitreous levels were predicted to reach 227ng/mL while the systemic blood levels were predicted toreach 15 ng/mL (Figure 3, Bottom).

DISCUSSION

INTRAOCULAR PHARMACOKINETICS OF POSTERIOR SUB-

Tenon’s triamcinolone acetonide injection have not beenwell documented, and the present study was undertakenwith a large number of subjects in order to provide moreconsistent results among the heterogeneous findings todate.33 Additionally, this is the only study of its kind tocompare experimental results and theoretical simulationoutputs of predicted steroid concentrations in variousintraocular compartments. For higher accuracy and preci-sion, the current study measured vitreous and serumtriamcinolone levels using liquid chromatography – tan-dem mass spectrometry. Our experimental results demon-strated that serum levels remained between 1.5 and 7.8ng/mL throughout the entire 2-month study period, with apeak level of 35.00 � 88.21 ng/mL at day 3. After a peakvalue of 111.43 � 138.95 ng/mL at day 1, the averagevitreous levels were found to remain between 17 and 31ng/mL at all other points.

Silva and associates12 found levels of triamcinolone inhe vitreous comparable to our findings using a similarethod of triamcinolone measurement with a sensitivity of

.5 ng/mL (Figure 4). Silva and associates also looked athe distance of the steroid depot from the sclera using-scan ultrasonography and found a strong inverse corre-

FIGURE 4. Vitreous concentrations of triamcinolone acetonidefrom our study (in black) with those found by Silva and associatbetween the 2 studies, with both showing early peaks followed

ation between depot distance and vitreous concentration,

AMERICAN JOURNAL OF944

hich could explain the high degree of variability amongtudies.12 Thomas and associates13 reported intravitrealevels of triamcinolone in 20 patients measured 1 to 29ays after posterior sub-Tenon’s injection, finding a largeange of vitreous concentrations from 0 to 4.95 �g/mL,ignificantly higher than those in our study or that of Silvand associates. While Thomas and associates do notpecify the diagnosis of the patients in their study, wepecifically chose to work with patients who were under-oing surgery for epiretinal membrane or macular hole andid not have any evidence of vasculopathy or retinopathyhat might result in breakdown of the blood-retinal barrier.he discrepancy could also be attributable to differingeasurement techniques. Thomas and associates use aigh-pressure liquid chromatography with a minimumetection level of 5 ng/mL, whereas we used high-perfor-ance liquid chromatography – tandem mass spectrometryith a minimum level of detection of 0.05 ng/mL.13 Ourndings are further confirmed by results from Shen andssociates,33 who used detection levels of 0.05 ng/mL

following 40-mg sub-Tenon’s injection and found a 1-daymaximum vitreous concentration of 92 ng/mL with a meanvitreous concentration of 40.97 ng/mL. Both values fit inthe range described in our experimental findings. Otherpossible explanations for the discrepancy in vitreous con-centrations between studies could include: differences intrans-scleral diffusion rates between patients, samplingerror within the vitreous cavity, the differences in thevolume of distribution between patients, variable rate ofsteroid dissolution, and phakic vs pseudophakic status.While these issues were not directly addressed in ourstudy, some of these problems could be controlled infuture studies by performing a fluid/air exchange toincrease the undiluted vitreous sample volume and by

t compares vitreous concentrations of triamcinolone acetonide(in gray). Of note is the similarity in the relative concentrations

general plateau.

. Ploes12

by a

using a B-scan ultrasound to identify the exact location

OPHTHALMOLOGY MAY 2012

h�

tfcc

cdamatbo

of the depot in relation to the sclera as performed bySilva and associates.12

As expected, our experimental results for the vitreousconcentration of triamcinolone following posterior sub-Tenon’s injection were also lower than most of theexperimental results of similar studies investigating intra-vitreal delivery. Jonas and associates delivered 20 to 25 mgof triamcinolone intravitreally and measured aqueous con-centrations of the steroid up to 6 months after injection,finding values as high as 436 ng/mL.34 Mason and associ-ates injected only 4 mg of triamcinolone intravitreally andfound a range from 0 to 380 ng/mL over the following 1.25to 5 months.14 Inoue and associates found significantly

igher values after intravitreal delivery, averaging 1.22g/mL over a time span of 3 to 19 days.35 Finally, Beer and

associates measured aqueous concentrations of triamcino-lone following intravitreal delivery and found values rang-ing from 0.0088 to 0.541 ng/mL for the month followinginjection.15

Our results also showed lower concentrations of triam-cinolone in the serum compared with the vitreous; how-ever, given the relative net volumes of vitreous and serumin the body (5 liters vs 4 cc), the net systemic absorptionwas much greater than the amount of corticosteroidentering the vitreous. Zaka-ur-Rab and associates16 foundsimilar levels of systemic triamcinolone, though with anearlier spike in serum concentration before our study’s firsttime point of 1 day, as graphed in Figure 5. Our study findsa strong Spearman correlation (0.65; P � .001) betweenvitreous and serum concentrations. This indicates a trans-scleral path of absorption into the choroid/serum and intothe vitreous, further supported by the strong inversecorrelation between depot distance and triamcinoloneconcentration recorded by Silva and associates.12 Thefinding that the concentration of serum cortisol decreasedwith an increase in triamcinolone concentration, as well as

FIGURE 5. Serum concentrations of triamcinolone acetonidefrom our study (black) with those found by Zaka-ur-Rab and aconcentration followed by a taper. The concentrations at timelower than the detection limit of 5 ng/mL used in their study.

the change from physiologic baseline at day 3, suggests that

POSTERIOR SUB-TENON TRIAMCINOLONVOL. 153, NO. 5

the small amount of systemic absorption of the steroidresulting from sub-Tenon’s delivery may be enough todisrupt patient metabolic feedback mechanisms. Otherstudies have suggested that this large net systemic absorp-tion could possibly affect metabolic equilibrium, and con-cordantly that patients with disrupted metabolic processessuch as diabetics should not receive posterior sub-Tenon’ssteroid treatment or should be monitored while receivingsuch treatment.16,33 Our experimental findings supporthese studies’ suggestions that the systemic absorptionollowing sub-Tenon’s delivery is in fact clinically signifi-ant, though further research will be needed to morelearly elucidate this effect.

To compare in vivo results, to predict triamcinoloneoncentrations in compartments that are experimentallyifficult to sample, and to understand the triamcinolonebsorption mechanism in various intraocular compart-ents, we developed an ocular simulation model based on

natomic and physiological parameters of the eye. Litera-ure values were programmed to model the human systemy changing the volumes of distribution and surface areasf the posterior compartment tissues.18–20 Predictions from

our ocular simulation model are graphed in Figure 3.Though the model did not accurately predict the long-term distribution and clearance, the similarities in themaximal values obtained experimentally and in the theo-retical model may make the values for the other compart-ments useful in gauging their respective maximal values.Out of all compartments measured, the highest concentra-tion of triamcinolone was found in the choroidal extracel-lular matrix (2230 ng/mL).

Nan and associates measured triamcinolone concen-trations in rabbit retinal pigment epithelium (RPE)/choroid, vitreous, and other compartments at varioustime frames after a 40-mg posterior sub-Tenon’s injec-tion.36 Similar to our findings, both Nan and associ-

t compares the serum triamcinolone acetonide concentrationsates16 (gray). Both studies show a similar early peak in serumts of Zaka-ur-Rab and associates at 14 days and 42 days were

. Plossocipoin

ates36 and Shen and associates33 reported highest

E ACETONIDE PHARMACOKINETICS 945

Aaritv

pibc

mcts

triamcinolone concentration in the choroid, suggestingthat penetration and diffusion into the vitreous couldfollow the pathway through the sclera and into thechoroid/RPE driven by a concentration gradient, aspredicted by the results of our theoretical model. Simi-larly, Lee and associates37 studied the pharmacokineticsof posterior sub-Tenon’s injection of Oregon Green488 –labeled triamcinolone acetonide (OGTA) in rab-bit eyes, calculating a permeability constant for thetrans-scleral diffusion of 1.12 � 10�7 cm/s for OGTA.

fter 1 cc sub-Tenon’s injection in a live rabbit, Leend associates found a maximum concentration in theetina/choroid of 25.77 � 10.26 ng/mL 3 hours after thenjection, further suggesting the trans-scleral entry ofriamcinolone acetonide into the choroid and theitreous.In the present study’s experimental component the

rolonged presence and plateauing of corticosteroid levelsn the choroidal ECM could have resulted from theinding of triamcinolone to pigment, which is abundant inhoroid/RPE.36 This binding of lipophilic triamcinolone

acetonide, as described by Nan and associates,36 to pig-ment could further serve as a secondary depot in additionto the primary episcleral depot provided by the posteriorsub-Tenon’s injection. The pigment binding may alsoexplain the experimental plateau occurring after the initialpeak in concentrations. After the quick drop from the peakfollowing systemic clearance, some drug could remainbound to pigment in the RPE (or in another secondarydepot systemically) that would release over time, leadingto a plateau effect. These regional mechanisms of triam-cinolone exposure and clearance, and the route of triam-cinolone entry into the vitreous, could help in derivingtherapeutic regimens for clinical practice as it especiallyrelates to the treatment of chorioretinal inflammation andproliferation.33,36

Intravitreal injection of triamcinolone acetonide hasbeen found to be associated with such complications asretinal detachment, vitreous hemorrhage, and injury to thelens, as well as an increased risk of cataract formation,

increased intraocular pressure, and endophthalmitis.6,38

ogy 2002;109(5):920–927.

AMERICAN JOURNAL OF946

Posterior sub-Tenon’s injection may be associated withsuch complications as ptosis, increased intraocular pres-sure, cataract formation, and potential ocular penetra-tion.39 Hirano and associates found that intravitrealdelivery more frequently required antiglaucoma medica-tions compared with a sub-Tenon’s approach.40 Further-

ore, some of the inactive ingredients found inommercially available triamcinolone acetonide formula-ions may have better tolerance when administered viaub-Tenon’s route vs intravitreally.27,41 One problem with

the design of the present study was the influence ofvitrectomy on assessments of the long-term effects ofsub-Tenon’s delivery of triamcinolone. Comparisonsof steroid-induced glaucoma rates between sub-Tenon’sand intravitreal delivery are difficult because of the lack ofpatient follow-up visits between the steroid injection andtheir vitrectomy to assess change in IOP, as well as theconfounding effect of the vitrectomy itself on IOP values.However, only 2 patients required pharmacologic inter-vention for postsurgically elevated IOP, which is signifi-cantly lower than the reported rates of steroid-inducedglaucoma when using intravitreal delivery. It seems thatthe posterior sub-Tenon’s route offers fewer ocular com-plications than its intravitreal counterpart, with this dif-ference in ocular safety profile probably attributable to themuch higher levels of vitreous steroid concentrationachieved with intravitreal injection compared to posteriorsub-Tenon’s injection; however, further research is neededin this area.

Posterior sub-Tenon’s injection of triamcinolone ace-tonide may provide an excellent method of delivering along-acting steroid to the choroid, with significantly fewerocular complications as compared to intravitreal injection.However, our results also indicate that the small amount ofsystemic triamcinolone acetonide resulting from posteriorsub-Tenon’s injection may be enough to disrupt patientmetabolic equilibrium. While our experimental and theo-retical findings provide valuable insight into determiningthe pharmacokinetics of triamcinolone acetonide, furtherresearch on the ocular mechanisms and the systemic side

effects of ocular administration is necessary.

ALL AUTHORS HAVE COMPLETED AND SUBMITTED THE ICMJE FORM FOR DISCLOSURE OF POTENTIAL CONFLICTS OFInterest. Publication of this article was supported by the Mills and Margaret Cox Endowment Funds, by the Grimshaw-Gudewicz Charitable Foundation(Fall River, Massachusetts), and by the Macula Society Award from the Retina Research Foundation (Cleveland, Ohio). The ocular simulation softwarewas developed by Bend Research Inc (Bend, Oregon, USA) under contract with Pfizer. The authors report no proprietary interests in the results of thestudy. Involved in design and conduct of study (K.K., M.Q., H.G., P.S., J.A.); collection, management, analysis, and interpretation of data (K.K., S.W.,M.Q., O.C., H.G., R.S., J.A.); and preparation, review, or approval of manuscript (K.K., S.W., P.S., J.A.). Prospective Institutional Review Boardapproval was obtained from the Beth Israel Deaconess Medical Center and an informed consent form was approved by the IRB and signed by all studyparticipants before involvement in the study. Patient information was maintained in a locked room on a password-protected computer with all subjectsde-identified to ensure HIPAA compliance. There is no Clinical Trials registration number for this study.

REFERENCES

1. Martidis A, Duker J, Greenberg P, et al. Intravitreal triam-cinolone for refractory diabetic macular edema. Ophthalmol-

2. Yilmaz T, Weaver C, Gallagher M, et al. Intravitreal triamcinoloneacetonide injection for treatment of refractory diabetic macularedema: a systematic review. Ophthalmology 2009;116(5):902–911.

3. Isaac D, Abud M, Frantz K, et al. Comparing intravitreal

triamcinolone acetonide and bevacizumab injections for the

OPHTHALMOLOGY MAY 2012

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

3

3

3

3

3

3

3

treatment of diabetic macular edema: a randomized double-blind study. Acta Ophthalmol doi: 10.1111/j.1755�3768.2009.01817.x.

4. Hogewind B, Zijlstra C, Klevering B, Hoyng C. Intravitrealtriamcinolone for the treatment of refractory macular edemain idiopathic intermediate or posterior uveitis. Eur J Oph-thalmol 2008;18(3):429–434.

5. Kok H, Lau C, Maycock N, et al. Outcome of intravitrealtriamcinolone in uveitis. Ophthalmology 2005;112(11):1916–1917.

6. Gillies M, Simpson J, Billson F, et al. Safety of an intravitrealinjection of triamcinolone: results from a randomized clinicaltrial. Arch Ophthalmol 2004;122(3):336–340.

7. Maca S, Abela-Formanek C, Kiss C, et al. Intravitrealtriamcinolone for persistent cystoid macular oedema in eyeswith quiescent uveitis. Clin Experiment Ophthalmol 2009;37(4):389–396.

8. Ozkiris A. Intravitreal triamcinolone acetonide injection forthe treatment of posterior uveitis. Ocul Immunol Inflamm2006;14(4):233–238.

9. Couch S, Bakri SJ. Intravitreal triamcinolone for intraocularinflammation and associated macular edema. Clin Ophthal-mol 2009;3:41–47.

0. Choudhry S, Ghosh S. Intravitreal and posterior subtenontriamcinolone acetonide in idiopathic bilateral uveitic mac-ular oedema. Clin Experiment Ophthalmol 2007;35(8):713–718.

1. Jonas J. Intraocular availability of triamcinolone acetonideafter intravitreal injection. Am J Ophthalmol 2004;137(3):560–562.

2. Silva P, Singh R, Bakri S, et al. Vitreous concentration oftriamcinolone acetonide after a single transseptal depotinjection. Ocul Immunol Inflamm 2009;17(3):216–220.

3. Thomas E, Wang J, Ege E, et al. Intravitreal triamcinoloneacetonide concentration after subtenon injection. Am JOphthalmol 2006;142(5):860–861.

4. Mason J, Somaiya M, Singh R. Intravitreal concentrationand clearance of triamcinolone acetonide in nonvitrecto-mized human eyes. Retina 2004;24(6):900–904.

5. Beer P, Bakri S, Singh R, et al. Intraocular concentration andpharmacokinetics of triamcinolone acetonide after a singleintravitreal injection. Ophthalmology 2003;110(4):681–686.

6. Zaka-ur-Rab S, Mahmood S, Shukla M, et al. Systemicabsorption of triamcinolone acetonide after posterior sub-Tenon injection. Am J Ophthalmol 2009;148(3):414–419.

7. Degenring R, Jonas J. Serum levels of triamcinolone ace-tonide after intravitreal injection. Am J Ophthalmol 2004;137(6):1142–1143.

8. Friedrich S, Cheng Y, Saville B. Drug distribution in thevitreous humor of the human eye: the effects of intravitrealinjection position and volume. Curr Eye Res 1997;16(7):663–669.

9. Olsen T, Aaberg S, Geroski D, Edelhauser H. Human sclera:thickness and surface area. Am J Ophthalmol 1998;125(2):237–241.

0. Swan K, Wilkins J. Extraocular muscle surgery in earlyinfancy—anatomical factors. J Pediatr Ophthalmol Strabis-

mus 1984;21(2):44–49.

POSTERIOR SUB-TENON TRIAMCINOLONVOL. 153, NO. 5

1. Phillips C, Michniak B. Transdermal delivery of drugs withdiffering lipophilicities using azone analogs as dermal pene-tration enhancers. J Pharm Sci 1995;84(12):1427–1433.

2. Koeberle M, Hughes P, Skellern G, Wilson C. Binding ofmemantine to melanin: influence of type of melanin andcharacteristics. Pharm Res 2003;20(10):1702–1709.

3. Rohatagi S, Hochhaus G, Mollmann H, et al. Pharmacoki-netic and pharmacodynamic evaluation of triamcinoloneacetonide after intravenous, oral, and inhaled administra-tion. J Clin Pharmacol 1995;35(12):1187–1193.

4. Derendorf H, Hochhaus G, Rohatagi S, et al. Pharmacoki-netics of triamcinolone acetonide after intravenous, oral, andinhaled administration. J Clin Pharmacol 1995;35(3):302–305.

5. Behl C, Block L, Borke M. Aqueous solubility of 14C-triamcinolone acetonide. J Pharm Sci 1976;65(3):429–430.

6. Oishi M, Maeda S, Hashida N, Ohguro N, Tano Y, Kuro-kawa N. Pharmacokinetic behavior of intravitreal triamcin-olone acetonide prepared by a hospital pharmacy. Jpn JOphthalmol 2008;52(6):489–492.

7. Bitter C, Suter K, Figueiredo V, Pruente C, Hatz K, Surber C.Preservative-free triamcinolone acetonide suspension devel-oped for intravitreal injection. J Ocul Pharmacol Ther2008;24(1):62–69.

8. Missel P, Horner M, Muralikrishnan R. Simulating dissolu-tion of intravitreal triamcinolone acetonide suspensions inan anatomically accurate rabbit eye model. Pharm Res2010;27(8):1530–1546.

9. Nozik R. Periocular injection of steroids. Trans Am AcadOphthalmol Otolaryngol 1972;76(3):695–705.

0. Arroyo J, Bula D, Yang L, Chen D. Retinal biopsy techniquesfor the removal of retinal tissue fragments. Ophthalmic SurgLasers Imaging 2005;36(1):76–78.

1. Singh R, Machacek D. Measurement of triamcinolone ace-tonide by liquid chromatography tandem mass spectrometry.Clin Chem 2002;48(6 Suppl):A122.

2. Taylor R, Grebe S, Singh R. Quantitative, highly sensitiveliquid chromatography-tandem mass spectrometry methodfor detection of synthetic corticosteroids. Clin Chem 2004;50(12):2345–2352.

3. Shen L, You Y, Sun S, et al. Intraocular and systemicpharmacokinetics of triamcinolone acetonide after a single40-mg posterior subtenon application. Ophthalmology 2010;117(12):2365–2371.

4. Jonas J, Kreissig I, Degenring R. Retinal complications ofintravitreal injections of triamcinolone acetonide. GraefesArch Clin Exp Ophthalmol 2004;242(2):184–185.

5. Inoue M, Takeda K, Morita K, et al. Vitreous concentrationsof triamcinolone acetonide in human eyes after intravitrealor subtenon injection. Am J Ophthalmol 2004;138:1046–1048.

6. Nan K, Sun S, Li Y, et al. Characterisation of systemic andocular drug level of triamcinolone acetonide following asingle sub-Tenon injection. Br J Ophthalmol 2010;94:654–658.

7. Lee S, Kim E, Geroski D, et al. Pharmacokinetics ofintraocular drug delivery of Oregon green 488-labeled triam-cinolone by subtenon injection using ocular fluorophotom-etry in rabbit eyes. Invest Ophthalmol Vis Sci 2008;49(10):

4506–4514.

E ACETONIDE PHARMACOKINETICS 947

4

4

38. Jonas J, Kreissig I, Degenring R. Retinal complications ofintravitreal injections of triamcinolone acetonide. GraefesArch Clin Exp Ophthalmol 2004;242(2):184–185.

39. Lafranco Dafflon M, Tran V, Guex-Crosier Y, Herbort C.Posterior sub-Tenon’s steroid injections for the treatment ofposterior ocular inflammation: indications, efficacy and sideeffects. Graefes Arch Clin Exp Ophthalmol 1999;237(4):289–295.

AJO History of OJules Gonin an

The search for an effective treatment for retinal detach-ment involved many false hypotheses and miscon-ceived surgical techniques, even after De Wecker and

Leber in 1875 and 1882 proposed that retinal tears precededand caused the detachment. When Jules Gonin (1870-1935)insisted that their closure was essential, and thereforesearched carefully for them, he began to have unprecedentedsuccess in reattaching the retina from 1916 onwards. In 1930

Submitted by Baldur Gloor from the Cogan Ophthalmic Histor

AMERICAN JOURNAL OF948

0. Hirano Y, Ito T, Nozaki M, et al. Intraocular pressureelevation following triamcinolone acetonide administrationas related to administration routes. Jpn J Ophthalmol 2009;53(5):519–522.

1. Younis H, Shawer M, Palacio K, Gukasyan H, Stevens G,Evering W. An assessment of the ocular safety of inactiveexcipients following sub-tenon injection in rabbits. J OculPharmacol Ther 2008;24(2):206–216.

halmology Seriese Nobel Prize

onin’s name was presented to the Nobel Prize committee,hich then sought the opinion of the prominent Alfred Vogt

1879-1943). Vogt quite wrongly and groundlessly cast doubtn Gonin’s priority in discovering this first consistently suc-essful operation for retinal detachment. It was enough for theobel committee to defer judgment and eventually pass overonin entirely. This has become one of the more glaring

missions in the history of the Nobel Prize.

phtd th

,

,

Gw(ocNGo

y Society.

OPHTHALMOLOGY MAY 2012