Editors Frank G. Holz
`Richard F. Spaide
`
`Medical Retina
`
`With 91 Figures, Mostly in Colour
`and 13 Tables
`
`123
`
`Samsung Bioepis Exhibit 1053
`Page 1
`
`

`

`Series Editors
`
`Günter K. Krieglstein, MD
`Professor and Chairman
`Department of Ophthalmology
`University of Cologne
`Kerpener Straße 62
`50924 Cologne
`Germany
`
`Robert N. Weinreb, MD
`Professor and Director
`Hamilton Glaucoma Center
`Department of Ophthalmology
`University of California at San Diego
`9500 Gilman Drive
`La Jolla, CA 92093-0946
`USA
`
`Volume Editors
`
`Frank G. Holz, MD
`Professor and Chairman
`Department of Ophthalmology
`University of Bonn
`Ernst-Abbe-Straße 2
`53127 Bonn
`Germany
`
`Richard F. Spaide, MD
`Assistant Clinical Professor
`Vitreous, Retina, and Macula Consultants
`of New York, and
`LuEsther T. Mertz Retinal Research Center
`Manhattan Eye, Ear, and Throat Hospital
`460 Park Avenue
`New York, NY 10022
`USA
`
`ISBN 978-3-540-33671-6
`Springer Berlin Heidelberg NewYork
`
`ISSN 1612-3212
`
`Library of Congress Control Number: 2007927503
`This work is subject to copyright. All rights are reserved,
`whether the whole or part of the material is concerned,
`specifically the rights of translation, reprinting, reuse of
`illustrations, recitation, broadcasting, reproduction on
`microfilms or in any other way, and storage in data banks.
`Duplication of this publication or parts thereof is permit-
`ted only under the provisions of the German Copyright
`Law of September 9, 1965, in its current version, and per-
`mission for use must always be obtained from Springer-
`Verlag. Violations are liable for prosecution under the
`German Copyright Law.
`Springer is a part of Springer Science + Business Media
`springer.com
`© Springer-Verlag Berlin Heidelberg 2007
`
`The use of general descriptive names, registered names,
`trademarks, etc. in this publication does not imply, even
`in the absence of a specific statement, that such names are
`exempt from the relevant protective laws and regulations
`and therefore free for general use.
`Product liability: The publishers cannot guarantee the ac-
`curacy of any information about dosage and application
`contained in this book. In every individual case the user
`must check such information by consulting the relevant
`literature.
`Editor: Marion Philipp, Heidelberg, Germany
`Desk Editor: Martina Himberger, Heidelberg, Germany
`Production: LE-TeX Jelonek, Schmidt & Vöckler GbR,
`Leipzig, Germany
`Cover Design: WMXDesign GmbH, Heidelberg,
`Germany
`Printed on acid-free paper
`24/3180Wa
` 5 4 3 2 1 0
`
`Samsung Bioepis Exhibit 1053
`Page 2
`
`

`

`Chapter 5
`Intravitreal Injections:
`Techniques and Sequelae
`Heinrich Heimann
`
`5
`
`Core Messages
`■ Through the introduction of new treat-
`ment strategies for exudative age-related
`macular degeneration, the number of
`intravitreal injections has increased dra-
`matically over the past few years. It is
`likely that this form of therapy will be-
`come the most common surgical inter-
`vention in ophthalmology within a short
`period of time.
`■ Although severe ocular adverse events
`associated with intraocular injections are
`rare, the rate can increase significantly if
`certain standards for intraocular inter-
`ventions are not followed. Several guide-
`lines on the technique for intravitreal
`injections have been published in recent
`years. Strict adherence to these guide-
`lines is advisable.
`
`■ Endophthalmitis is the most feared
`complication of intravitreal injections.
`It is usually caused by bacterial contami-
`nation during or immediately after the
`injection and occurs in about 0.1% of in-
`jections and in about 1% of patients with
`repeated injections.
`■ Ocular hypertension and cataract devel-
`opment are not typically seen after anti-
`VEGF therapy.
`■ Based on the studies and data currently
`available, no major difference in the risk
`profile of the anti-VEGF drugs used at
`present can be seen.
`■ Triamcinolone is associated with a high-
`er rate of secondary ocular hypertension
`(40%) and need for glaucoma surgery
`(1%) than anti-VEGF agents. It also has
`a higher rate of cataract progression in
`about 40% of patients within the first
`year of treatment.
`
`5.1 Introduction
`The introduction of new drugs for the treatment
`of age-related macular degeneration (AMD) has
`led to a significant change in ophthalmological
`practice. Until a few years ago, intravitreal injec-
`tions were reserved for a small number of rare
`diseases (e.g., endophthalmitis, viral retinitis).
`Within a short period of time, the numbers of
`injections have increased and are now second
`only to cataract surgery as the most common
`treatment in most tertiary centers across Europe
`and the United States. It is likely that intravit-
`real injections with the anti-vascular endothelial
`
`growth factor (VEGF) type of action will soon
`become the most common intraocular procedure
`performed worldwide. With intravitreal injection
`we can obtain a high intraocular concentration
`of a drug, with minimal systemic exposure.
`This tremendous and rapid change indicates a
`significant challenge for ophthalmological units;
`a tidal wave of intraocular injections, re-injec-
`tions, and follow-up examinations has to be inte-
`grated into daily routine without compromising
`patients’ safety. It is therefore mandatory to main-
`tain essential safety standards for all injections
`whilst avoiding unnecessary and costly examina-
`tions and safety measures. In order to maximize
`
`Samsung Bioepis Exhibit 1053
`Page 3
`
`

`

`68
`
`Intravitreal Injections: Techniques and Sequelae
`
`the number of patients that can be treated, the
`workload associated with each patient has to be
`minimized within the treating unit; therefore, it is
`likely that more and more pre- and postoperative
`examinations will be shifted to places outside the
`centers where the injections are performed. In a
`relatively short period of time, ophthalmologists
`and optometrists who are currently not perform-
`ing intraocular injections will be confronted with
`a larger number of patients following intravitreal
`injections. In this chapter, the techniques, com-
`plications, and guidelines for intravitreal injec-
`tions are reviewed. Because of the recent shift in
`
`Table 5.1 Complications of intravitreal injections
`
`the application of intravitreal drugs, the chapter
`focuses on studies of anti-VEGF substances and
`triamcinolone.
`
`5.2 Complications
`of Intravitreal Injections
`With an appropriate technique, the intravitreal
`injection of a drug is a straightforward surgi-
`cal procedure that carries a low rate of serious
`complications (Table 5.1). Yet, the first published
`multicenter study showed that the disregard
`
`Peri-
`operative
`X
`
`Early
`(>7 days)
`X
`
`Late
`(>7 days)
`
`Incidence
`
`~20–40%
`
`Related to in-
`jected substance
`–
`
`X
`X
`X
`
`X
`
`X
`X
`X
`X
`
`X
`
`X
`
`X
`X
`
`X
`
`X
`X
`
`X
`X
`
`X
`
`X
`X
`
`~30%
`~30%
`<1%
`First year:
`~15%
`~40% (triamcinolone)
`<1%
`
` ~20%
`<1%
`~30%
`~20%
`
`~1%
`
`~0.15% per injection
`~1% per patient with
`multiple injections
`<1%
`Triamcinolone
`~40%
`
`–
`
`–
`X
`
`–
`
`–
`–
`X
`X
`
`X
`
`X
`
`–
`X
`
`Complication
`
`Conjunctival
`hemorrhage
`Conjunctival scarring
`Punctate keratitis
`Pain
`Traumatic cataract
`Cataract progression
`
`Central retinal
`artery occlusion
`Vitreous reflux
`Vitreous hemorrhage
`Vitreous floaters
`Intraocular
`inflammation
`Uveitis/pseudo-
`endophthalmitis
`Endophthalmitis
`
`X
`X
`X
`
`X
`
`X
`X
`X
`X
`
`Retinal detachment
`Ocular hypertension
`
`X
`
`Samsung Bioepis Exhibit 1053
`Page 4
`
`

`

`5.2 Complications of Intravitreal Injections
`
`69
`
`of basic standards for intraocular surgery can
`quickly increase the rate of complications to un-
`acceptable levels. In the initial stages of the VI-
`SION trial, the rates of bacterial endophthalmitis
`were more than 4-fold higher than those after
`routine cataract surgery [8, 14]. Serious adverse
`events occurred in 19% of patients (169 out of
`892), although serious adverse events were also
`noted in 15% of patients (45 out of 298) with
`sham treatment [8]. These figures underline that
`intravitreal injections should be treated as intra-
`ocular surgery and conducted according to the
`standards applied to all intraocular procedures,
`e.g., asepsis of the operating field and a sterile
`technique throughout the process. Even with
`the greatest care, complications associated with
`intravitreal injections will never be avoided com-
`pletely; however, as demonstrated in the recent
`large prospective trials, in only very few cases
`will these complications lead to a long-term re-
`duction in visual acuity or discontinuation of the
`treatment if dealt with in a timely and appropri-
`ate fashion [4, 8, 14, 29].
`
`5.2.1 Methodology
`In this chapter, the complications associated with
`intraocular injections are reviewed. A methodo-
`logical review is flawed by several problems:
`1. The most serious side effects of intravitreal in-
`jections (e.g., endophthalmitis, retinal detach-
`ment, glaucoma, cataract) are rare following
`anti-VEGF treatment and occur in less than
`1% of injections and in about 1–2% of patients
`undergoing repeated injections. Only ran-
`domized prospective trials with large patient
`numbers are able to reflect the true rate of
`complications associated with the drug exam-
`ined and the treatment protocol applied. Such
`trials have been published for pegaptanib and
`ranibizumab (Table 5.2) [4, 8, 14, 29].
`2. For the two other currently most commonly
`used drugs, bevacizumab and triamcinolone,
`such studies do not exist and it is unlikely that
`this will be the case in the near future.
`3. Other important, potentially sight-threaten-
`ing complications of injections are intraocular
`inflammation, cataract development, and a
`rise in intraocular pressure (IOP). The exami-
`
`nation methods, time points, and definitions
`for the detection of these complications vary
`from study to study.
`4. Different protocols for the performance of in-
`travitreal injections have been used. The dif-
`ferences in the injection technique (e.g., sub-
`conjunctival anesthesia, perioperative drug
`treatment, etc.) can have a significant influ-
`ence on the complication rate.
`
`5.2.2 Perioperative Complications
`Complications following intravitreal injections
`can be divided according to their occurrence
`into perioperative, early postoperative, and late
`(Table 5.1). Furthermore, complications from
`the intraocular injection procedure itself have to
`be distinguished from possible biological side-ef-
`fects of the injected substance.
`
`5.2.2.1 Conjunctival Hemorrhage
`Conjunctival hemorrhage is related to the trauma
`caused by manipulations during the injection,
`e.g., forceps or needle injuries. Obviously, when
`subconjunctival anesthesia is used, the need for
`two injections increases the risk of this complica-
`tion. In one series, subconjunctival hemorrhages
`could be seen in 18% of patients with intravitreal
`injections and topical anesthesia versus 40% in
`patients with subconjunctival anesthesia [22].
`Nevertheless, even with topical anesthesia, hem-
`orrhage can be seen in up to 37% of patients [7].
`In the vast majority of cases, conjunctival
`hemorrhages is more cosmetically disturbing
`than harmful. They clear spontaneously within
`7–14 days and do not require any therapy. Very
`rarely, they can progress to cause significant an-
`terior segment problems that require surgical
`intervention. The risk of significant conjunctival
`hemorrhage seems to be increased in patients
`with anticoagulant therapy, e.g., warfarin [15].
`In contrast to their relative insignificance in the
`majority of cases, conjunctival hemorrhage is of-
`ten perceived as a serious side-effect from the pa-
`tients’ point of view. Particularly in an outpatient
`setting without a scheduled short-term follow-up
`examination, patients have to be instructed about
`
`Samsung Bioepis Exhibit 1053
`Page 5
`
`

`

`70
`
`Intravitreal Injections: Techniques and Sequelae
`
`Table 5.2 Ocular adverse events in studies of pegaptanib and ranibizumab
`
`Follow-up
`(months)
`
`Number
`of patients
`
`Number
`of injections
`
`Endophthal-
`mitis (n)
`
`Per patient
`(%)
`
`Per injection
`(%)
`
`Pegaptanib
`VISION [14]
`0.3 mg
`
`1.0 mg
`3.0 mg
`First year
`Second year
`
`Ranibizumab
`MARINA [29]
`0.3 mg
`
`0.5 mg
`Total
`
`ANCHOR [4]
`0.3 mg
`
`0.5 mg
`Total
`
`Control groups
`VISION
`(sham injection)
`ANCHOR
`(PDT)
`MARINA
`(sham injection)
`
`12
`
`24
`
`24
`
`12
`12
`
`12
`
`24
`
`12
`
`24
`
`295
`
`301
`296
`892
`374
`
`238
`
`240
`478
`
`137
`
`140
`277
`
`298
`
`140
`
`238
`
`7,545
`4,091
`
`12
`4
`
`NA
`
`NA
`10,443
`
`1,507
`
`1,568
`3,075
`
`2
`
`3
`5
`
`0
`
`2
`2
`
`0.16
`0.1
`
`0.05
`
`0
`
`0.006
`
`1.3
`1.0
`
`0.8
`
`1.2
`1.0
`
`0
`
`0.14
`0.07
`
`0
`
`0
`
`Samsung Bioepis Exhibit 1053
`Page 6
`
`

`

`5.2 Complications of Intravitreal Injections
`
`71
`
`Inflammation
`(%)
`
`Uveitis
`(%)
`
`Cataract
`(%)
`
`Cataract
`surgery (%)
`
`Glaucoma
`(%)
`
`Retinal
`detachment (%)
`
`Split in different
`grades, high-
`est grade 17%
`18
`22
`
`>35 mmHg
`9
`
`9
`15
`
`16
`
`14
`13
`
`17
`
`21
`19
`
`12
`
`8
`
`16
`
`3
`
`12
`
`<1
`
`5.1
`
`7.2
`6.1
`
`5.3
`
`1.2
`
`18
`
`0
`
`1.2
`
`1.2
`1.2
`
`0
`
`15
`
`15
`
`11
`
`12.9
`
`26
`
`7.0
`
`15.7
`
`1
`0.3
`
`1.6
`
`–
`
`–
`–
`
`0.7
`
`– 0
`
`.3
`
`>30 mmHg
`13.0
`17.6
`
`>40 mmHg
`2.3
`2.3
`
`>30 mmHg
`8.8
`8.6
`
`>40 mmHg
`2.9
`2.9
`
`>30 mmHg
`4.2
`>30 mmHg
`3.4
`
`>40 mmHg
`0.7
`>40 mmHg
`0
`
`0.7
`
`0.4
`
`Samsung Bioepis Exhibit 1053
`Page 7
`
`

`

`72
`
`Intravitreal Injections: Techniques and Sequelae
`
`their appearance and natural course. Strategies to
`minimize their incidence include the use of topi-
`cal versus subconjunctival anesthesia and the use
`of a cotton bud for conjunctival manipulation in-
`stead of forceps.
`
`5.2.2.2 Conjunctival Scarring
`Repeat injections, especially when combined with
`subconjunctival anesthesia, will lead to conjunc-
`tival scarring. Significant conjunctival scarring
`or ocular surface disorders as a result of multiple
`intravitreal injections have not been reported in
`larger studies to date. However, if repeated injec-
`tions are necessary, it is recommended to record
`the position of the injection and to avoid repeated
`injection in the same location if possible.
`
`5.2.2.3 Pain
`Pain can be related to the anesthesia, the needle
`entry through the conjunctiva and sclera, and the
`rise in IOP associated with the injection. When
`subconjunctival anesthesia is used, patients ex-
`perience pain associated with the subconjuncti-
`val injection, but less pain when the needle en-
`ters the posterior segment. In the VISION trial,
`34% of patients (subconjunctival anesthesia and
`intravitreal injection) versus 28% of the controls
`(subconjunctival anesthesia and sham injection)
`experienced pain [8]. When topical anesthesia is
`used, the pain associated with the subconjunc-
`tival injection is avoided; however, the intravit-
`real injection is more painful [22]. Overall, the
`patients’ expected levels of pain and discomfort
`are usually greater than those from the actual ex-
`perience [30].
`
`5.2.2.4 Punctate Keratitis
`and Corneal Edema
`Punctate keratitis was detected in 32% of patients
`in the VISION trial versus 27% of controls [8].
`It has not been noted as a complication in other
`trials. The most likely cause is a combination of
`the anesthesia of the ocular surface, a reduced
`blink reflex after corneal anesthesia, flushing of
`the corneal surface with antiseptic solution, the
`
`prolonged exposure of the cornea for the dura-
`tion of the injection, and irregularities of the
`ocular surface after the injection. A combination
`of these factors and the intermittent rise in IOP
`can also cause corneal edema; this was noted in
`9% of patients in the VISION trial [8]. However,
`to date there are no suggestions that the corneal
`surface or the corneal endothelium can be af-
`fected through a direct toxic effect of the injected
`substances. To avoid damage to the corneal
`epithelium, the exposure time following topical
`anesthesia should be minimized. Alternatively,
`ocular lubricants can be applied following the
`procedure.
`
`5.2.2.5 Vitreous Reflux
`The intravitreal injection of 0.1 ml of fluid is asso-
`ciated with a rise in IOP to around 45 mmHg [3].
`During removal of the needle, vitreous, liquefied
`vitreous or fluid can exit the posterior segment
`through the needle path into the subconjunctival
`space or transconjunctivally to the ocular surface.
`This phenomenon can be observed in about 20%
`of injections [3]. Two problems can be associated
`with a vitreous prolapse:
`1. A substantial amount of the injected drug can
`be misplaced.
`2. A “vitreous wick” can serve as an entry site
`for bacteria from the ocular surface and may
`significantly increase the risk of postoperative
`endophthalmitis [5].
`
`The rate of vitreous reflux is dependent on
`the needle, the injected volume, the consequen-
`tial rise in IOP, and the injection technique used.
`Vitreous reflux can be avoided using sharp small-
`gauge needles (e.g., 27-, 30- or 31-gauge) and/or
`a short scleral tunnel for injection by pulling
`the conjunctiva over the injection site and using
`a slightly angled scleral path with the injection
`needle.
`
`5.2.2.6 Traumatic Cataract
`With the appropriate training of surgeons carry-
`ing out the injection and the injection technique,
`the rate of traumatic cataract should be minimal.
`Yet, sudden head movements of the patient or ac-
`
`Samsung Bioepis Exhibit 1053
`Page 8
`
`

`

`5.2 Complications of Intravitreal Injections
`
`73
`
`cidental moving of the needle tip toward the lens
`can never be fully avoided, and even intralenticu-
`lar injections have been reported [19].
`In the VISION trial, the rate of traumatic cata-
`ract was 0.07% (5 out of 892) [8]. The importance
`of the appropriate injection technique is under-
`lined by the fact that, in this trial, 2 of the 5 lens
`injuries occurred on the same day in one center
`[8]. In the MARINA trial, lens damage was noted
`in 0.2% (1 out of 477) [29]. In the ANCHOR trial,
`no lens damage was noted during the first year in
`277 patients with multiple injections [4].
`
`5.2.2.7 Cataract Progression
`Any intraocular manipulation is likely to be asso-
`ciated with an increase in cataract formation over
`time. The analysis of this complication is made
`difficult by several factors:
`1. Different systems for grading and diverse def-
`initions of “cataract progression” have been
`used in studies on intravitreal injections.
`2. The systems for cataract grading currently
`used are still biased by subjective differences
`in the grading of lens opacities by different ex-
`aminers.
`3. The majority of published studies are retro-
`spective in nature and a precise recording
`of cataract progression does not seem to be
`achievable with such study designs.
`4. A difference in cataract progression may only
`be seen after several years of follow-up. To
`date, most studies cannot provide the neces-
`sary data because follow-ups are limited to 1
`or 2 years.
`5. A significant amount of cataract progression
`is likely to occur in the age group of patients
`with AMD unrelated to any intervention.
`
`The latter is underlined by the rates of cata-
`ract progression that were noted in the control
`groups of prospective trials. These were 16% in
`the MARINA trial (sham injection), 7% in the
`ANCHOR trial (photodynamic therapy + sham
`injection), and 26% in the VISION trial (sham
`injection) [4, 8, 14, 29]. In comparison, similar or
`slightly higher rates of cataract progression could
`be noted in the treatment groups (MARINA 15%,
`ANCHOR 11–13%, and VISION 17–22%) [4, 8,
`14, 29]. Very few patients underwent cataract
`
`surgery during the trial periods in these studies.
`Overall, there seems to be little or no cataracto-
`genicity of repeated intraocular injections with
`anti-VEGF substances compared with their re-
`spective control groups within the first 2 years.
`In contrast to the intravitreal application of
`anti-VEGF substances, intravitreal triamcino-
`lone is associated with a higher rate of cataract
`development. By and large, cataract surgery has
`been performed in 20–45% of patients within the
`first year of the initial injection [21, 33]. The rate
`of cataract formation and consecutive surgery
`is time-dependent. In a prospective 2-year trial,
`cataract surgery was performed in 54% of pa-
`tients (15 out of 28) with diabetic macular edema
`treated with intravitreal triamcinolone compared
`with 0% (0 out of 21) in the control group [12].
`The cataractogenicity of local and systemic
`steroids has long been established; yet, the pre-
`cise mechanisms are not clearly understood [20].
`In the majority of cases, steroid use is associated
`with posterior subcapsular cataract formation.
`In addition, cortical cataracts and, to a lesser
`extent, nuclear cataract progression can also be
`seen following intravitreal triamcinolone injec-
`tion [11, 33]. Interestingly, there seems to be a
`highly significant association of the second ma-
`jor complication of intravitreal triamcinolone,
`the rise in IOP, with the progression of subcapsu-
`lar posterior and cortical cataracts [11]. A higher
`rate of cataract progression should therefore be
`expected in patients who are classified as steroid
`responders according to a rise in IOP following
`the injection.
`
`5.2.2.8 Retinal Perforation
`Entering the posterior segment with a sharp in-
`strument carries the potential risk of retinal in-
`jury. This can occur when the entry site is too
`posterior (>5 mm), the needle is too long and ad-
`vanced too far, the needle is pointed in the wrong
`direction or detached retina or choroid are in the
`way of the needle. No such retinal injuries have
`been reported in recent multicenter trials [4, 8,
`14, 29]. Inspection of the retinal periphery before
`and after the injection should be included in the
`routine injection procedure. With appropriate
`training of surgeons, retinal injuries should be
`avoidable.
`
`Samsung Bioepis Exhibit 1053
`Page 9
`
`

`

`74
`
`Intravitreal Injections: Techniques and Sequelae
`
`5.2.2.9 Vitreous Floaters
`Vitreous floaters are commonly noticed by pa-
`tients following intravitreal injections. In the
`VISION trial, they were perceived in 33% of
`patients in the treatment group versus 8% of pa-
`tients with sham injections [14]. Due to the crys-
`talline structure of the injected substance, float-
`ers are more prevalent following triamcinolone
`injections. Floaters usually subside within the
`first week without the need for additional thera-
`peutical measures.
`
`5.2.2.10 Vitreous Hemorrhage
`Vitreous hemorrhage is another potential compli-
`cation of intravitreal injections. It can be caused
`by the injection itself during which the needle
`penetrates the choroids to reach the posterior
`chamber as well as by post-injection changes
`within the vitreous, leading to vitreous traction
`with preretinal hemorrhage. Its incidence, how-
`ever, is relatively low; furthermore, the diseases
`most commonly treated with intravitreal injec-
`tions, e.g., exudative AMD or diabetic retinopa-
`thy, are frequently associated with vitreous hem-
`orrhage not related to surgical interventions.
`In the VISION trial, the rate of vitreous hem-
`orrhage was 0.21% (16 out of 7,545) [8]. How-
`ever, only 9 of the 16 incidents were judged to
`be associated with the injection procedure. In
`the MARINA trial, the rates of vitreous hemor-
`rhage were 0.4% (2 out of 477) in the treatment
`group versus 0.8% (2 out of 236) in the group
`with sham treatment [29]. Usually, no treatment
`is required for vitreous hemorrhage secondary to
`intravitreal injections. Pars plana vitrectomy can
`be considered in advanced cases or those without
`spontaneous regression.
`
`5.2.2.11 Retinal Toxicity
`Retinal toxicity is a potential complication of in-
`travitreally applied drugs. Such toxicity, can lead
`to changes in the electroretinogram, may cause
`visible changes through alterations of the fundus
`pigmentation and can lead to gross photoreceptor
`and macular malfunction. Cases of retinal toxic-
`
`ity or presumed toxicity have been documented
`following intravitreal application of tissue plas-
`minogen activator, methotrexate, amikacin and
`hyaluronidase [6, 17, 35].
`So far, there have been no reports of presumed
`toxicity of anti-VEGF drugs, including bevaci-
`zumab, in vivo or in vitro with the currently used
`dose regimen [4, 8, 10, 29]. However, some con-
`cerns remain that constant VEGF suppression
`might increase the long-term risks of toxicities
`for retinal tissue [25].
`The question of possible toxicity of triamcino-
`lone has been raised in in vitro studies on rabbits
`and on cultured human retinal pigment epithelial
`(RPE) cells [32, 36]. This toxicity has been linked
`to the concentration of the drug, the preservative
`in commercial preparations or the direct contact
`of the triamcinolone crystals with the retina [21,
`32, 36]. In clinical series, no evidence of triam-
`cinolone toxicity has been reported so far [21].
`
`5.2.2.12 Intraocular Inflammation
`Anti-VEGF substances are humanized antibod-
`ies, or aptamers, that potentially can cause an
`intraocular immune reaction. Clinical studies
`have therefore specifically looked at intraocular
`inflammation following their intravitreal appli-
`cation. Intraocular inflammatory cells, however,
`can also be seen after any manipulation of the eye
`or the ocular adnexa, including intravitreal injec-
`tions or sham injections.
`A dose-related increase in intraocular inflam-
`mation has been documented for two drugs after
`intravitreal application, ranibizumab and hy-
`aluronidase. Early phase I/II studies of ranibi-
`zumab were associated with a relatively high rate
`of low-grade to moderate intraocular inflamma-
`tion in up to 78% of patients [16]. This might
`have been due to the lyophilized preparation of
`the antibody. In a further dose-escalating study,
`Rosenfeld et al. found a dose-dependent re-
`sponse; injections up to 500 µg of ranibizumab
`seem to be well tolerated, but higher doses were
`accompanied by more intraocular inflammation
`[28]. In the large prospective clinical trials of
`ranibizumab (up to 500 µg) and pegaptanib (up
`to 3 mg), the rates of (mostly low-grade) ante-
`rior and posterior chamber inflammation were
`
`Samsung Bioepis Exhibit 1053
`Page 10
`
`

`

`5.2 Complications of Intravitreal Injections
`
`75
`
`significantly lower and were seen in up to 20%
`within the treatment groups. There was no cor-
`relation between the level of intraocular inflam-
`mation and the injected drug dose and a compa-
`rable rate and severity of events could be seen in
`the respective control groups [4, 8, 14, 29]. Typi-
`cally, intraocular inflammations following ra-
`nibizumab and pegaptanib can be seen on day 1
`after the injection and usually subside without
`treatment within 14 days. Anti-inflammatory
`treatment is not necessary in most cases. There
`does not seem to be an increased risk of more se-
`vere inflammation with repeated injections that
`would indicate an amplification of the immune
`response as a reaction to the injected proteins.
`In contrast, lower intraocular inflammation
`scores were seen following repeated injections in
`a dose-escalating study with more frequent and
`higher dosed injections compared with the cur-
`rently recommended doses of ranibizumab [28].
`Another drug that was documented to cause
`a biological reaction following intravitreal injec-
`tion is bovine hyaluronidase. Increased intra-
`ocular inflammatory responses with a dose re-
`lationship could be seen. The rates of moderate
`or severe iritis in the study by Kupperman et al.
`were 0% (observation) and 8% (saline injection)
`in the control groups versus 20% (7.5 IU hyal-
`uronidase), 37% (55 IU), and 40% (75 IU) [24]
`in the treatment groups.
`Intraocular inflammatory reactions (with the
`exception of pseudo-endophthalmitis or endo-
`phthalmitis, see Sect. 5.2.2.13) do not seem to
`be a particular problem following intravitreal
`triamcinolone [21]. When injected into the ante-
`rior chamber and capsular bag following routine
`cataract surgery, there is a dose-related reduction
`of the anterior chamber inflammatory response
`[13]. In one series of 759 consecutive intravitreal
`injections, no significant anterior chamber reac-
`tion could be noted [23]. This is probably due to
`the immunosuppressive effect of the drug.
`
`5.2.2.13 Uveitis and Pseudo-
`endophthalmitis
`The vast majority of inflammatory reactions
`in the anterior and posterior chamber follow-
`ing intravitreal injections are of low grade and
`
`do not require any treatment. They are usually
`thought to be associated with the injection pro-
`cedure. Occasionally, severe intraocular inflam-
`matory reactions can be seen. In these rare cases
`it is important, although sometimes extremely
`difficult, to differentiate a sterile inflammatory
`process from an inflammatory reaction as-
`sociated with infectious endophthalmitis. To
`complicate things further, the term “pseudo-
`endophthalmitis” has been introduced. This
`has almost exclusively been used for severe in-
`traocular inflammations following intravitreal
`triamcinolone injections.
`Low-grade inflammatory reactions following
`intravitreal injections are thought to be a reac-
`tion to the injection procedure, whereas severe,
`noninfectious reactions might be caused by the
`drug itself (or its solvent) [29]. To distinguish
`low-grade inflammation from clinically signifi-
`cant inflammatory reactions, the latter are often
`classified as uveitis. In a phase I/II study of ra-
`nibizumab, one case of a severe recurrent uveitis
`was noted [16]; the uveitis recurred after each
`injection of the drug, underlining the likelihood
`of a biological reaction to the injected antibody.
`In the larger prospective studies of ranibizumab,
`uveitis could be seen in 1 of the 277 cases in the
`ANCHOR trial and in 1.2% of cases (6 out of
`478) in the MARINA trial. In the Bevacizumab
`Safety Survey, 10 cases of uveitis out of a total of
`5,228 patients (0.19%) were reported [10]. No
`severe intraocular inflammations other than en-
`dophthalmitis were reported to occur following
`pegaptanib injection in the VISION trial [14].
`Pseudo-endophthalmitis (synonyms: sterile,
`noninfectious or toxic endophthalmitis) is char-
`acterized by intraocular changes or a severe in-
`flammatory reaction mimicking the clinical pic-
`ture of infectious endophthalmitis. This reaction
`is almost exclusively seen following intravitreal
`injection of unaltered, commercially available
`triamcinolone. Three different paths can lead to
`its manifestation:
`1. Immune reaction to the vehicle of the injected
`drug, leading to severe intraocular uveitis.
`2. Collection of triamcinolone crystals in the
`anterior chamber, mimicking the picture of a
`hypopyon. This can particularly seen in eyes
`with previous vitrectomy and defects in the
`lens capsule or zonular fibers, or when part
`
`Samsung Bioepis Exhibit 1053
`Page 11
`
`

`

`76
`
`Intravitreal Injections: Techniques and Sequelae
`
`of the injected triamcinolone is misdirected
`alongside the anterior vitreous surface.
`3. The crushing of triamcinolone crystals in
`small-gauge cannulae (30- or 31-gauge), lead-
`ing to an almost immediate and dense vitre-
`ous haze following the injection.
`
`There is no unambiguous clinical sign to dif-
`ferentiate pseudo-endophthalmitis from infec-
`tious endophthalmitis. However, some clinical
`features are more commonly associated with
`pseudo-endophthalmitis:
`1. Clinical signs develop within the first few
`hours of the injection. This does not exclude
`infectious endophthalmitis, but newly diag-
`nosed intraocular inflammation days after
`the injection should is more indicative of
`infectious endophthalmitis and rather than
`pseudo-endophthalmitis.
`2. Pain, periorbital swelling and photophobia
`are usually absent in pseudo-endophthalmi-
`tis. However, it should be kept in mind that
`pain is also absent in about 20% of infectious
`endophthalmitis.
`3. Dense vitreous infiltrates and pseudo-hypo-
`pyon seen in pseudo-endophthalmitis are
`accompanied by a relatively low-grade ante-
`rior chamber reaction (e.g., very few anterior
`chamber cells above the crystals).
`4. Anterior chamber taps or vitreous biopsies
`fail to isolate bacterial or fungal organisms.
`
`Some authors advocate purification of com-
`mercially available triamcinolone before intra-
`vitreal injection in order to eliminate the vehicle
`and to lower the occurrence of toxic reactions.
`With this technique, only one case of pseudo-en-
`dophthalmitis could be seen after 759 injections
`in a series by Kreissig et al. [23]. In contrast, in
`a series of 922 injections without removal of the
`vehicle, 8 cases of pseudo-endophthalmitis were
`seen (0.87%) [26].
`subsides
`Pseudo-endophthalmitis usually
`without any specific therapy over 7–14 days. In
`contrast, infectious endophthalmitis can prog-
`ress extremely quickly without appropriate treat-
`ment with potentially devastating consequences.
`Weighing the potential risks of treating a case of
`pseudo-endophthalmitis with intraocular anti-
`biotics against delaying treatment of infectious
`endophthalmitis