Anatomic and Visual Outcomes of Noninfectious
`Endophthalmitis after Intravitreal Triamcinolone
`
`STEVEN J. YOON, DAVID Y. RHEE, JEFFREY L. MARX, GREGORY R. BLAHA, ADAM H. ROGERS,
`CAROLINE R. BAUMAL, ELIAS REICHEL, AND JAY S. DUKER
`
`● PURPOSE: To describe the anatomic and visual outcomes
`of patients in whom noninfectious endophthalmitis devel-
`oped after injection of intravitreal triamcinolone acetonide.
`● DESIGN: Retrospective case series.
`● METHODS: Ophthalmologic evaluations of patients in
`whom noninfectious endophthalmitis developed after intra-
`vitreal triamcinolone took place on the day of injection, at
`the time of presentation of noninfectious endophthalmitis,
`at the time of clearance of inflammation, and on follow-up
`examination. Seventeen eyes of 17 patients were identified
`from 2 institutions. Noninfectious endophthalmitis was
`identified based on history of visual loss immediately or
`soon after injection, lack of ocular pain, hypopyon, anterior
`or vitreous inflammation, and triamcinolone crystals
`present in the anterior or posterior chambers. Main out-
`come measures were Snellen visual acuity (VA) and mean
`foveal thickness by optical coherence tomography.
`● RESULTS: Mean VA and mean foveal thickness on the
`day of injection of intravitreal triamcinolone were 20/
`132 (logarithm of the minimum angle of resolution
`[logMAR], 0.82 ⴞ 0.45) and 432 ⴞ 118 ␮m, respec-
`tively. Mean VA at time of noninfectious endoph-
`thalmitis (mean, 1.9 days after injection) was 20/4444
`(logMAR, 2.35 ⴞ 0.98). At last follow-up (mean, 57.6
`days), VA and mean foveal thickness were 20/56 (log-
`MAR, 0.44 ⴞ 0.30) and 301 ⴞ 71 ␮m, respectively.
`● CONCLUSIONS: VA and mean foveal thickness in all
`patients with noninfectious endophthalmitis after intra-
`vitreal triamcinolone improved to better than preinjec-
`tion levels in this series. At last follow-up, no patient had
`sustained visual loss from noninfectious endophthalmi-
`tis. Noninfectious endophthalmitis after intravitreal tri-
`amcinolone may not exclude good visual and anatomic
`prognoses.
`(Am J Ophthalmol 2009;147:1031–1036.
`© 2009 by Elsevier Inc. All rights reserved.)
`
`S INCE THE INTRODUCTION OF INTRAVITREAL TRIAM-
`
`cinolone in 2002 by Martidis and associates and
`Greenberg and associates, its use has become wide-
`spread by retinal specialists for a variety of intraocular
`inflammatory, neovascular, and edematous diseases, such as
`
`Accepted for publication Dec 29, 2008.
`From the New England Eye Center (S.J.Y., D.Y.R., A.H.R., C.R.B.,
`E.R., J.S.D.), Boston, Massachusetts; and the Department of Ophthal-
`mology, Lahey Clinic (J.L.M., G.R.B.), Burlington, Massachusetts.
`Inquiries to Steven J. Yoon, 750 Washington Street, Box 450, Boston,
`MA 02111; e-mail: steveyoon13@hotmail.com
`
`chronic uveitis, refractory diabetic macular edema (DME),
`proliferative vitreoretinopathy, branch retinal vein occlusion,
`and pseudophakic cystoid macular edema (CME).1– 4 Corti-
`costeroids inhibit leukotriene and prostaglandin synthesis
`creating an anti-inflammatory effect and also have been
`shown to stabilize the blood-retinal barrier and reverse cap-
`illary permeability.5,6 The therapeutic spectrum of intravit-
`real triamcinolone continues to widen as clinical studies
`validate its efficacy.
`Well-known complications associated with intraocular
`corticosteroids include ocular hypertension and the pro-
`gression of cataracts.7 Infectious endophthalmitis is a rare
`but potentially devastating outcome of any intravitreal
`injection, and the added immunosuppressive effects of
`intraocular corticosteroids warrants increased concern.
`Acute infectious endophthalmitis after intravitreal triam-
`cinolone injection has been reported with an incidence of
`0.10% to 0.87%8,9 per injection. An endophthalmitis-like
`noninfectious intraocular inflammatory reaction has also
`been reported and has been termed sterile endophthalmitis.
`This entity is thought to represent a sterile inflammatory
`response against a component of the drug formulation.10,11
`Pseudoendophthalmitis is a term that indicates the dispersion
`of triamcinolone crystals in the anterior chamber (AC),
`and pseudohypopyon is a term used to describe the layering
`of triamcinolone crystals in the inferior AC angle. In this
`study, we grouped these entities together as noninfectious
`endophthalmitis. Herein, we describe the short-term ana-
`tomic and visual outcomes after the clearance of nonin-
`fectious endophthalmitis after intravitreal triamcinolone.
`
`METHODS
`
`THE RECORDS OF PATIENTS RECEIVING INTRAVITREAL TRI-
`amcinolone acetonide (Kenalog; Bristol-Myers Squibb,
`Princeton, New Jersey, USA) injections from 6 retinal
`specialists at the New England Eye Center, Boston, Mas-
`sachusetts, and Lahey Clinic, Burlington, Massachusetts,
`were reviewed. From January 2002 through June 2007
`approximately 1,600 intravitreal triamcinolone injections
`were administered between the 2 institutions. The risks
`and benefits of the procedure were discussed, and written
`informed consent was obtained from each patient before
`injection. Standard sterile technique was used with a
`sterile lid speculum, calipers, sterile gloves, 5% povidone–
`
`0002-9394/09/$36.00
`doi:10.1016/j.ajo.2008.12.034
`
`© 2009 BY ELSEVIER INC. ALL RIGHTS RESERVED.
`
`1031
`
`Novartis Exhibit 2300.001
`Regeneron v. Novartis, IPR2021-00816
`
`

`

`TABLE 1. Cause of Cystoid Macular Edema in Patients in
`Whom Noninfectious Endophthalmitis Developed after
`Intravitreal Triamcinolone Administration
`
`TABLE 2. Prior Ocular History in Patients in Whom
`Noninfectious Endophthalmitis Developed after Intravitreal
`Triamcinolone Administration
`
`Cause of Macular Edema
`
`% of Patients
`
`Ocular History
`
`% of Patients
`
`Diabetes
`Pseudophakia
`Epiretinal membrane
`Branch retinal vein occlusion
`Uveitis
`Cancer-associated retinopathy
`
`41% (7)
`18% (3)
`18% (3)
`12% (2)
`6% (1)
`6% (1)
`
`Prior intraocular surgery
`Pseudophakia
`Pars plana vitrectomy
`Open posterior capsule
`Prior intravitreal triamcinolone injection
`
`82% (14)
`71% (12)
`53% (9)
`17% (3)
`53% (9)
`
`iodine, topical tobramycin or ofloxacin, and topical tetra-
`caine for anesthesia. A pars plana injection was performed
`with previously unopened bottles of intravitreal triamcin-
`olone acetonide (4 mg/0.1 ml) using 27- or 30-gauge
`needles. Topical antibiotics then were used 4 times daily for
`3 to 5 days. Follow-up appointments were made 4 to 6 weeks
`after the injection. Patients were asked to call if decreased
`vision developed after the injection, and a nurse called each
`patient 1 to 2 days after the injection to verify that no
`difficulties were encountered after the intravitreal injection.
`Snellen visual acuity (VA) and mean foveal thickness
`with optical coherence tomography were measured on the
`day of injection, on presentation of noninfectious endoph-
`thalmitis, at a time when clearance of inflammation was
`observed, and on follow-up examination. Eyes were iden-
`tified as having noninfectious endophthalmitis based on
`history of decreased vision within 24 to 48 hours of
`injection without pain, and characteristic findings such as
`hypopyon, AC and vitreous cell, vitreous haze, and triam-
`cinolone crystals identified by slit-lamp biomicroscopy and
`indirect ophthalmoscopy. B-scan ultrasonography was per-
`formed if no view to the posterior segment was appreci-
`ated. Patients underwent vitreous culture and intravitreal
`injection of antibiotics at the discretion of the attending
`physician. Any patient whose gram-stain, aqueous, or
`vitreous culture results were positive was excluded from
`this study. Patients then were followed up closely at regular
`intervals at the physician’s discretion.
`
`RESULTS
`
`SEVENTEEN EYES OF 17 PATIENTS WERE IDENTIFIED AS HAV-
`ing noninfectious endophthalmitis after intravitreal triamcin-
`olone injection. Clinical characteristics of involved eyes are
`summarized in Tables 1 and 2. Intravitreal triamcinolone was
`used to treat DME in 7 patients (41%), pseudophakic CME
`in 3 patients (18%), CME associated with an epiretinal
`membrane in 3 patients (18%), CME associated with branch
`retinal vein occlusion in 2 patients (12%), CME associated
`with uveitis in 1 patient (6%), and CME associated with
`cancer-associated retinopathy in 1 patient (6%).
`
`Nine (53%) of the patients had received at least 1 prior
`intravitreal triamcinolone injection before presentation of
`noninfectious endophthalmitis. None of these 9 eyes were
`diagnosed with any untoward inflammation previously. Nine
`patients (53%) had prior pars plana vitrectomies performed in
`the affected eye. Twelve (71%) of the patients were pseu-
`dophakic, and only 3 of these patients had an open posterior
`capsule. Fourteen (82%) of the 17 patients had some type of
`prior intraocular surgery performed.
`The average time to presentation after injection was 1.9
`days (range, 1 to 3 days). All patients had decreased vision
`in the affected eye at presentation. No patients reported
`ocular pain. However, 5 patients reported periorbital
`soreness, and 1 patient reported photophobia.
`All patients demonstrated AC and vitreous cellular reac-
`tion at the time of presentation of noninfectious endoph-
`thalmitis. Eleven patients had a visible pseudohypopyon
`(Figure 1, Top left and right), and all patients demonstrated
`various degrees of vitreous haze. Triamcinolone crystals were
`visible in the AC and vitreous in 9 (53%) of the 17 patients,
`and it was identified on B-scan ultrasonography when no
`view to the posterior segment was appreciated (Figure 2,
`right). The presumed triamcinolone particles seemed to be
`larger and more crystalline in character than typical inflam-
`matory white blood cells (Figure 1, Bottom left; Figure 2,
`right). Fibrin was identified in the AC in 3 of the patients
`(18%). Eight patients underwent vitreous cultures and injec-
`tions of intravitreal antibiotics. One patient with light per-
`ception vision at
`the time of presentation underwent
`vitrectomy with intravitreal antibiotics. No growth of any
`organisms was identified on gram-stain, aqueous, or vitreous
`cultures of these patients.
`The mean preinjection Snellen VA was 20/132 (logarithm
`of the minimum angle of resolution [logMAR], 0.82 ⫾ 0.45)
`and mean central foveal thickness of 432 ⫾ 118 ␮m. At the
`time of presentation of noninfectious endophthalmitis, mean
`VA was 20/4444 (logMAR, 2.35 ⫾ 0.98). At the time of
`clearance of inflammation (mean, 15.1 days), mean VA was
`20/99 (logMAR, 0.70 ⫾ 0.41). Of the 11 of 17 patients
`undergoing optical coherence tomography examination at
`that time, the mean central foveal thickness measured 30 2 ⫾
`66 ␮m, and the decrease in foveal thickness averaged 109
`␮m. On short-term follow-up (mean, 57.6 days), mean VA
`was 20/56 (logMAR, 0.44 ⫾ 0.30), and mean central foveal
`
`1032
`
`AMERICAN JOURNAL OF OPHTHALMOLOGY
`
`JUNE 2009
`
`Novartis Exhibit 2300.002
`Regeneron v. Novartis, IPR2021-00816
`
`

`

`FIGURE 1. Slit-lamp photographs showing noninfectious en-
`dophthalmitis after intravitreal triamcinolone. (Top left) Eleven
`patients demonstrated layering of steroid particles in the inferior
`anterior chamber (AC) or pseudohypopyon, as demonstrated in
`this photograph. (Top right) Discrete layering clouds of triamcin-
`olone crystals are visible in the inferior angle of a patient. (Bottom
`left) AC cellular reaction and steroid particles are visible in the
`AC of a patient in a slit-lamp photograph.
`
`thickness was 301 ⫾ 71 ␮m, with a mean change of 131 ␮m.
`The data is summarized in Table 3 and Figures 3 and 4.
`
`DISCUSSION
`
`NONINFECTIOUS ENDOPHTHALMITIS AFTER INTRAVITREAL
`triamcinolone has become a recognized entity with the
`increased use of
`intravitreal triamcinolone for refractory
`macular edema. The incidence of noninfectious endoph-
`thalmitis after intravitreal triamcinolone has been reported to
`range from 0.2% to 1.6% in various series,10 –13 and our
`approximate incidence was 1.06% of injections at The New
`England Eye Center and Lahey Clinic. Previously reported
`case reports of noninfectious endophthalmitis describe pain-
`less visual loss immediately or soon after the injection. Other
`clinical characteristics in reports include a layering of steroid
`particles in the inferior AC angle simulating a hypopyon,
`vitreous haze, and white crystalline opacities in the aqueous
`humor and vitreous. This entity may represent the dispersion
`
`of triamcinolone crystals in the AC and a distinct sterile
`inflammatory response against a component of the drug.
`These patients typically have no ocular pain and are reported
`to recover rapidly within days to preinjection levels of VA
`without intervention.
`It has been suggested that noninfectious endophthalmi-
`tis may be caused by an acute toxic reaction to the
`triamcinolone acetonide particle or the vehicle of the drug
`suspension.10,12 Experimental studies found that the vehi-
`cle used in intravitreal triamcinolone was nontoxic to the
`rabbit retina14 and also nontoxic to the human retina
`based on electrophysiologic analysis.15 The vehicle of the
`commercial formulation of triamcinolone includes 6.9 mg
`sodium chloride, 15 mg benzyl alcohol, 7.5 mg carmellose
`sodium, and 0.4 mg polysorbate 80. The benzyl alcohol
`preservative in the preparation has been speculated to be a
`stimulus for the inflammatory reaction. Studies with fil-
`tered or preservative-free triamcinolone acetonide to im-
`prove the potential toxicity of the suspension thus far have
`been inconclusive.13
`
`VOL. 147, NO. 6
`
`NONINFECTIOUS ENDOPHTHALMITIS
`
`1033
`
`Novartis Exhibit 2300.003
`Regeneron v. Novartis, IPR2021-00816
`
`

`

`FIGURE 2. Images showing noninfectious endophthalmitis after intravitreal triamcinolone. (Left) Fundus photograph showing
`vitreous haze and triamcinolone crystals in a patient. (Right) B-scan ultrasonography image showing intravitreal triamcinolone
`crystals of a patient when no view to the posterior segment was appreciated.
`
`TABLE 3. Snellen Visual Acuity and Central Foveal Thickness Measurements on the Day of Injection, Presentation, Clearance,
`and Follow-up for Patients in Whom Noninfectious Endophthalmitis Developed after Intravitreal Triamcinolone Administration
`
`Day of Injection
`
`Presentation
`
`Clearance
`
`Follow-up
`
`Case No.
`
`Cause
`
`VA
`
`CFT (␮m) Day
`
`VA
`
`1
`2
`3
`4
`5
`6
`7
`8
`9
`10
`11
`12
`13
`14
`15
`16
`17
`Average values
`Standard
`deviations
`
`Diabetes
`Diabetes
`BRVO
`CAR
`ERM
`Diabetes
`Pseudophakia
`Pseudophakia
`Diabetes
`ERM
`Diabetes
`Uveitis
`Diabetes
`ERM
`BRVO
`Pseudophakia
`Diabetes
`
`20/60
`20/100
`20/400
`20/40
`20/2000
`20/100
`20/50
`20/160
`20/60
`20/400
`20/400
`20/200
`20/100
`20/60
`20/50
`20/80
`20/80
`20/132
`0.82 ⫾ 0.45
`(logMAR)
`
`403
`276
`552
`453
`546
`312
`506
`593
`590
`366
`390
`376
`362
`322
`214
`550
`530
`432
`⫾ 118
`
`3
`1
`3
`2
`2
`1
`3
`1
`2
`3
`2
`1
`2
`1
`2
`1
`3
`1.9
`
`20/20000
`20/2000
`20/2000
`20/60
`20/20000
`20/2000
`20/2000
`20/20000
`20/200
`20/20000
`20/20000
`20/20000
`20/20000
`20/20000
`20/60
`20/200000
`20/2000
`20/4444
`2.35 ⫾ 0.98
`(logMAR)
`
`Day
`
`30
`15
`14
`23
`21
`19
`8
`9
`5
`21
`12
`7
`13
`14
`14
`22
`10
`15.1
`
`VA
`
`CFT (␮m)a
`
`Day
`
`VA
`
`CFT (␮m)
`
`20/50
`20/80
`20/200
`20/60
`20/2000
`20/80
`20/60
`20/160
`20/50
`20/40
`20/100
`20/100
`20/200
`20/60
`20/30
`20/200
`20/80
`20/99
`0.70 ⫾ 0.41
`(logMAR)
`
`413
`—
`368
`237
`364
`237
`—
`302
`—
`347
`308
`222
`—
`308
`221
`—
`—
`302
`⫾ 66
`
`44
`36
`35
`63
`39
`61
`28
`44
`20
`40
`33
`189
`91
`117
`57
`52
`31
`57.6
`
`20/30
`20/60
`20/125
`20/30
`20/400
`20/60
`20/50
`20/60
`20/30
`20/30
`20/70
`20/100
`20/60
`20/50
`20/20
`20/40
`20/60
`20/56
`0.44 ⫾ 0.30
`(logMAR)
`
`334
`203
`277
`198
`413
`255
`380
`256
`338
`360
`262
`275
`266
`310
`201
`386
`404
`301
`⫾ 71
`
`BRVO ⫽ branch retinal vein occlusion; CAR ⫽ cancer-associated retinopathy; CFT ⫽ central foveal thickness; ERM ⫽ epiretinal membrane;
`logMAR ⫽ logarithm of the minimum angle of resolution; VA ⫽ visual acuity.
`aNot all patients underwent optical coherence tomography at the clearance visit. The mean decrease in foveal thickness of the 11 of 17
`patients measured 109 ␮m less than that of the initial scan.
`
`A recent report compared the incidence of noninfectious
`endophthalmitis after injection with commercially available
`triamcinolone acetonide vs preservative-free triamcinolone
`
`(sodium chloride, monobasic, and dibasic sodium phosphate,
`0.04% polysorbate 80, water).16 The incidence of noninfec-
`tious endophthalmitis in the preservative-free group was
`
`1034
`
`AMERICAN JOURNAL OF OPHTHALMOLOGY
`
`JUNE 2009
`
`Novartis Exhibit 2300.004
`Regeneron v. Novartis, IPR2021-00816
`
`

`

`Best Corrected Visual Acuity
`
`20,56
`
`201132
`
`2014444
`
`0., 000
`
`O.JSOO
`
`0.3000
`
`0.2SOO
`
`0.2000
`
`0.1500
`
`0.1000
`
`0.0500
`
`0.0000
`
`lritial
`Pc,et,t,j ~.9 d)
`Folcmllp (IS.1 cl)
`Follc,v.(Jp (57.6d)
`FIGURE 3. Bar graph showing mean best-corrected visual acuity
`(Snellen acuity and logarithm of the minimum angle of resolution
`units) on the day of injection, presentation, clearance, and fol,
`low-up for patients in whom noninfectious endophthalmitis devel,
`oped after intravitreal triamcinolone administration.
`
`Mean Foveal Thickness
`
`lritid
`Post ~(1,9d)
`FollOW<Jp(tS.1 d)
`Follow..,(57.6d)
`FIGURE 4. Bar graph showing mean central foveal thickness
`(in micrometers) on the day of injection, presentation, clear,
`ance, and follow-up for patients in whom noninfectious en,
`dophthalmitis developed after intravitreal triamcinolone.
`
`1.9% (n = 646 injections) vs 7.3% (n = 69 injections) in the
`commercially available triamcinolone. The incidence of non(cid:173)
`infectious endophthalmitis occurring with preservative-free
`triamcinolone in this study demonstrates that this entity still
`occurs despite the removal of the benzyl alcohol, and at a rate
`similar to that of previously published reports with commer(cid:173)
`cially available triamcinolone. The very high incidence of
`noninfectious endophthalmitis in the commercially available
`triamcinolone group in this retrospective series is curious and
`may be related to the disparity of the number of injections
`between the 2 groups (n = 646 vs n = 69). Prospective
`studies comparing preservative-free triamcinolone vs com-
`
`mercially available triamcinolone would elucidate further the
`effect of benzyl alcohol on noninfectious endophthalmitis.
`In 4 of the 17 patients during the 5.5-year span of the
`study, noninfectious endophthalmitis developed within a
`period of 7 weeks. These 4 patients were administered
`injections of intravitreal triamcinolone from the same lot
`number. The cluster of cases is suspicious, and there have
`been previous reports with similar experiences; Nelson and
`associates reported 7 cases in a span of 5 weeks.12 This
`leads us to speculate that the triamcinolone formulation in
`these particular lots created an inflammatory reaction,
`potentially related to an unknown additive in the solution
`or an unidentified bacterial endotoxin remaining despite
`sterilization. 12
`One patient in this series was reinjected with intravit(cid:173)
`real triamcinolone on 3 separate occasions to the same eye
`after the episode of noninfectious endophthalmitis. No
`inflammatory reaction or dispersion of triamcinolone crys(cid:173)
`tals was observed in the subsequent injections, which may
`lead us to believe the triamcinolone formulation in that
`bottle or lot, may be more responsible for the inflammatory
`reaction than the individual's immune response.
`Another suggested mechanism of noninfectious en(cid:173)
`dophthalmitis is thought to be the result of the impaction
`of the drug into the barrel of a 30-gauge needle, resulting
`in high-velocity dispersion into a vitreous suspension at
`time of injection into the eye.10 Microscopy of vitreous
`samples in reports reveal a dense collection of triamcino(cid:173)
`lone crystals with no cells, suggesting that clinical signs
`17
`were a result of dispersed triamcinolone particles. 13•
`However, in this and other studies, noninfectious endoph(cid:173)
`thalmitis occurred in patients with the use of both 2 7- and
`30-gauge needles for injection.
`The diagnosis of noninfectious endophthalmitis is a
`critical judgment by the clinician that may avoid unnec(cid:173)
`essary invasive treatments. Nine patients in this study
`received treatment for presumed infectious endophthalmi(cid:173)
`tis in this study based on the severity of clinical findings
`at presentation. The gram stains and cultures of these 9
`patients had negative results, and these were presumed to
`be noninfectious cases in hindsight after rapid recovery of
`vision and inflammation. The lack of growth in these
`cultures does not exclude the possibility of infectious
`endophthalmitis, because 18% of vitreous samples from
`the Endophthalmitis Vitrectomy Study had culture-nega(cid:173)
`tive results. 18 The treatment with intravitreal antibiotics
`in our series occurred earlier in the study, and if these cases
`presented today, very close observation may have been
`determined based on our clinical experience with nonin(cid:173)
`fectious endophthalmitis. It is notable that 2 patients
`(0.13%) of the approximately 1,600 injections of intra(cid:173)
`vitreal triamcinolone were reported by the physicians to have
`developed infectious endophthalmitis after intravitreal triam(cid:173)
`cinolone. These patients had a distinct presentation and
`clinical course, with a later onset, presence of ocular pain, and
`a slow visual recovery. However, if a case is questionable, it is
`
`VOL. 147, No. 6
`
`NONIN FECTIOUS ENDOPHTHALM ITIS
`
`1035
`
`Novartis Exhibit 2300.005
`Regeneron v. Novartis, IPR2021-00816
`
`

`

`important not to overlook a potential case of infectious
`endophthalmitis and to treat appropriately.
`All patients in this series of noninfectious endoph-
`thalmitis after intravitreal triamcinolone had painless vi-
`sual
`loss within 3 days of
`injection. Snellen VA on
`follow-up examinations in all patients improved to better
`than VA levels before the injection of intravitreal triam-
`cinolone. Mean foveal thickness by ocular coherence
`tomography in all patients on follow-up examination
`improved to better than mean foveal thickness measure-
`ments before the injection of intravitreal triamcinolone.
`Although the data for these patients in this study are
`reported over several months to differentiate the effect of
`the triamcinolone injection vs the natural course of the
`
`disease, no patient experienced long-term visual loss from
`noninfectious endophthalmitis. Noninfectious endoph-
`thalmitis after intravitreal triamcinolone may be managed
`with close observation without intervention. However, it
`is important to differentiate this entity from progressive
`inflammation from an early presentation of an acute
`infectious endophthalmitis.
`Despite the poor VA at the time of noninfectious endoph-
`thalmitis after intravitreal triamcinolone, this event may not
`exclude good visual and anatomic outcomes. Although it is
`clear further research and experience is needed to understand
`further the mechanism of noninfectious endophthalmitis, this
`entity may not necessarily limit our use of intravitreal triam-
`cinolone based on anatomic and visual outcomes.
`
`THIS STUDY WAS SUPPORTED IN PART BY A RESEARCH TO PREVENT BLINDNESS, NEW YORK, NEW YORK CHALLENGE GRANT
`to the New England Eye Center/Department of Ophthalmology, Tufts University School of Medicine, and also supported in part by a Departmental
`Grant from Massachusetts Lions, New Bradford, Massachusetts. The authors indicate no financial conflict of interest. Involved conception and design
`(S.J.Y., D.Y.R., J.L.M., G.R.B., A.H.R., C.R.B., E.R., J.S.D.); analysis and interpretation (S.J.Y., D.Y.R., J.S.D.); writing the article; (S.J.Y.); critical
`revision (D.Y.R., J.L.M., G.R.B., A.H.R., C.R.B., E.R., J.S.D.); final approval (J.L.M., J.S.D.); data collection (S.J.Y.); provision of patients and resources
`(J.L.M., G.R.B., A.H.R., C.R.B., E.R., J.S.D.); statistical expertise (S.J.Y., D.Y.R., J.S.D.); obtaining funding (J.L.M., J.S.D.); literature search (S.J.Y.);
`and administrative and logistical support (D.Y.R., J.S.D.). This study received Institutional Review Board (IRB) approval from the Tufts University IRB
`and the Lahey Clinic IRB.
`
`REFERENCES
`
`1. Young S, Larkin G, Branley M, Lightman S. Safety and efficacy
`of intravitreal triamcinolone for cystoid macular oedema in
`uveitis. Clin Experiment Ophthalmol 2001;29:2–6.
`2. Martidis A, Duker JS, Greenberg PB, et al. Intravitreal
`triamcinolone for refractory diabetic macular edema. Oph-
`thalmology 2002;109:920 –927.
`3. Greenberg PB, Martidis A, Rogers AH, et al. Intravitreal
`triamcinolone acetonide for macular oedema due to central
`retinal vein occlusion. Br J Ophthalmol 2002;86:247–248.
`4. Heier JS, Topping TM, Baumann W, et al. Ketorolac versus
`prednisolone versus combination therapy in the treatment of
`acute pseudophakic cystoid macular edema. Ophthalmology
`2000;107:2034 –2038.
`5. Wilson CA, Berkowitz BA, Sato Y, et al. Treatment with
`intravitreal steroid reduces blood-retinal barrier breakdown
`due to retinal photocoagulation. Arch Ophthalmol 1992;
`110:1155–1159.
`6. Ozaki NK, Beharry KD, Nishihara KC, et al. Regulation of
`retinal vascular endothelial growth factor and receptors in
`rabbits exposed to hyperoxia. Invest Ophthalmol Vis Sci
`2002;43:1546 –1557.
`7. Fracs R, Fracs P, Wingate W. Intravitreal triamcinolone and
`elevated intraocular pressure. Aust N Z JOphthalmol 1999;
`27:431– 432.
`8. Westfall AC, Osborn A, Kuhl D, et al. Acute endophthalmi-
`tis incidence: intravitreal triamcinolone. Arch Ophthalmol
`2005;123:1075–1077.
`9. Moshfeghi DM, Kaiser PK, Scott IU, et al. Acute endoph-
`thalmitis following intravitreal triamcinolone acetonide in-
`jection. Am J Ophthalmol 2003;136:791–796.
`
`10. Sutter FK, GIllies MC. Pseudo-endophthalmitis after intra-
`vitreal injection of triamcinolone. Br J Ophthalmol 2003;87:
`972–974.
`11. Nelson ML, Tennant MT, Sivalingam A, et al. Infectious
`and presumed noninfectious endophthalmitis after intravit-
`real triamcinolone acetonide injection. Retina 2003;23:686 –
`691.
`12. Roth DB, Chieh J, Spirn MJ, et al. Noninfectious endoph-
`thalmitis associated with intravitreal triamcinolone injec-
`tion. Arch Ophthalmol 2003;121:1279 –1282.
`13. Jonas JB, Kreissig I, Degenring RF. Endophthalmitis after
`intravitreal injection of triamcinolone acetonide. Arch Oph-
`thalmol 2003;121:1663–1664.
`14. Kim H, Csaky KG, Gravlin L, et al. Safety and pharmaco-
`kinetics of a preservative-free triamcinolone acetonide for-
`mulation for intravitreal administration. Retina 2006;26:
`523–530.
`15. Lang Y, Leibu R, Shoham N, et al. Evaluation of intravitreal
`Kenalog toxicity in humans. Ophthalmology 2007;114:724 –
`731.
`16. Maia M, Farah ME, Belfort RN, et al. Effects of intravitreal
`triamcinolone acetonide injection with and without preser-
`vative. Br J Ophthalmol 2007;91:1122–1124.
`17. Chen SD, Lochhead J, Patel CK. Diffuse intraocular disper-
`sion of triamcinolone particles as a cause of sterile endoph-
`thalmitis. Arch Ophthalmol 2004;122:1733.
`18. Endophthalmitis Vitrectomy Study Group. Results of the
`Endophthalmitis Vitrectomy Study: a randomized trial of
`immediate vitrectomy and of intravenous antibiotics for the
`treatment of postoperative bacterial endophthalmitis. Arch
`Ophthalmol 1995;113:1479 –1496.
`
`1036
`
`AMERICAN JOURNAL OF OPHTHALMOLOGY
`
`JUNE 2009
`
`Novartis Exhibit 2300.006
`Regeneron v. Novartis, IPR2021-00816
`
`