Blank spaces indicate no reported events.

^aPrevalence per eye (%).

^bPrevalence per injection (%).

IVT, intravitreal; EO, endophthalmitis; RD, retinal detachment; I/U, iritis/uveitis/vitritis/anterior chamber inflammation; HE, hemorrhage; OH, ocular hypertension; CT, cataract; HP, hypotony; CS, case series; RCT, randomized controlled trial; POS, prospective open study; RS, retrospective study; CMV, cytomegalovirus; CME, cystoids macular edema; RAO, retinal artery occlusion; RVO, retinal vein occlusion.

The overall prevalence of endophthalmitis after IVT injection is low. Retrospective reports indicate a per-injection endophthalmitis risk between 0% and 0.87% (29–32). Jager et al. (23) 2004 reported an overall prevalence of endophthalmitis, excluding cases reported as noninfectious endophthalmitis or pseudoendophthalmitis, of 0.5% per eye and 0.2% per injection. Endophthalmitis was reported very rarely after IVT injection of gas or other compounds other than antiviral agents and triamcinolone acetonide, with a combined prevalence of 0.1% per eye and 0.1% per injection for all other compounds (10,23,33,34). The trials for ranibizumab, ranibizumab versus verteporfin for neovascular AMD (ANCHOR) (35) and ranibizumab for neovascular AMD (MARINA) (36), also demonstrated a low rate of endophthalmitis. At 1 year, ANCHOR reported a 0.07% per-injection endophthalmitis rate, and by 2 year, there were only three cases of endophthalmitis out of 5,921 injections (0.05%) (35).

Singerman et al. (VISION Trial: VEGF Inhibition Study in Ocular Neovascularization) (37) reported a low rate of serious ocular events during 3 years of pegaptanib sodium for treatment of neovascular AMD with a rate of endophthalmitis from 0.16% per injection (12 of 7,545 injections) in the first year to 0.10% per injection (4/4091) in the 2nd year and to 0.06% per injection (2/3227) in the 3rd year (P < 0.0001). This decrease in the rate of endophthalmitis may result at least partially from a protocol emphasizing the importance of adherence to aseptic technique when performing IVT injections. Other reasons for lower rates of endophthalmitis may also be the widespread adoption of and increasing familiarity with IVT injections by retina specialists over the 3 years of this study.

The Pan American Collaborative Retina Study Group (PACORES, 2008) mentioned seven cases of bacterial endophthalmitis of a total of 4,303 IVT injections of bevacizumab during 12 months of safety monitoring (28). The cultures yielded five cases of coagulase-negative staphylococci and one case each of Staphylococcus aureus and Streptococcus.

Noninfectious (sterile) endophthalmitis and pseudoendophthalmitis may also occur after IVT triamcinolone injections. Some studies (Table 23.2) have reported incidents of 0.5% to 0.9% per IVT triamcinolone injections (32,37,38).

RETINAL DETACHMENT

Retinal detachment is uncommon (23,27,28,32,36,37) after IVT injection. The VISION trial (27) described an incidence of rhegmatogenous retinal detachment of 0.03% per injection (1/3227) (27). ANCHOR and MARINA studies demonstrated a 0.01% per-injection detachment rate (35,36).

Jager et al. (23) mentioned an overall prevalence of retinal detachment of 3.9% per eye and 0.9% per injection. The majority of cases identified in this review appeared to be related to patients underlying condition, principally CMV retinitis and vitreous hemorrhage in the setting of diabetic retinopathy (23).

IRITIS/UVEITIS

Secondary uveitis is an uncommon complication of IVT injection. Ocular inflammation has been observed in 26.9% of eyes with CMV retinitis treated with IVT cidofovir (22,39), in 38.2% of eyes treated with IVT fomivirsen (9), and in 48.4% of eyes treated with IVT ovine hyaluronidase (13) for the treatment of vitreous hemorrhage. Taskintuna et al. (22) reported a high prevalence of intraocular inflammatory reactions in the fellow (untreated) eyes of some patient, many of whom previously had been treated with systemic antiviral agents. Moreover, acute intraocular inflammation has been reported to occur after intravenous administration of cidofovir. These findings suggest that the occurrence of noninfectious intraocular inflammation in patients receiving IVT cidofovir is a consequence of the therapeutic agent itself and not the IVT injection. Another confounding factor in identifying the cause of intraocular inflammation in patients receiving cidofovir is the condition known as immune recovery uveitis in which patients with CMV retinitis develop inflammation when their CD4 count increases after beginning treatment with HAART (39,40).

OCULAR HYPERTENSION

Increases in IOP after IVT injection may be transient or sustained. Although transient IOP responses to IVT injection were described inconsistently in many reports, sustained ocular hypertension was associated primarily with the administration of triamcinolone acetonide (23,29,30,41). Prevalence of ocular hypertension after IVT injection of triamcinolone varies between studies (38%–52%) (23–25,29,30). Excluding IVT injection of triamcinolone acetonide, Jager et al. (23) described a prevalence of ocular hypertension of 2.4% per eye and 0.4% per injection, with most these events having been reported in association with fomivirsen (9,23).

Transiently elevated IOP, in contrast, is common and may occur after IVT administration of any agent. Although most studies make no mention of transient ocular hypertension, some studies have reported a transient increase in IOP when injecting simultaneous expansile gas and recombinant tPA (tissue plasminogen activator) (4.7% of eyes) (33) and pegaptanib sodium (4.8% of eyes) (11). This transient increase in IOP was likely related to the amount of volume injected in the eye, which was 0.09 mL of pegaptanib in the VISION study rather than the standard 0.5 mL, and the use of 0.05 mL of 100% sulfur hexafluoride expansile gas plus 0.05 mL of tPA in the former study.

Gillies et al. (41) reported that 21 of 75 eyes (28%) treated with 4 mg triamcinolone required glaucoma drops at some point during a 3-year period. Over a mean period of 8 months, 71% of eyes discontinued treatment as the IOP normalized, but six eyes (8%) required continued IOP-lowering therapy. In the Diabetic Retinopathy Clinical Research Network (DRCR.net) study comparing IVT triamcinolone acetonide and focal–grid photocoagulation for diabetic macular edema, 102 of 254 eyes (40%) in the 4-mg triamcinolone group, 51 of 256 eyes (20%) in the 1-mg triamcinolone group, and 34 of 330 eyes (10%) in the laser group developed some adverse IOP-related event (defined as an IOP elevation greater than 10 mmHg from baseline, IOP greater than 30 mmHg, initiation of IOP-lowering medications, or a diagnosis of glaucoma) in a 2-year follow-up (24). The Standard of Care versus Corticosteroid for Retinal Vein Occlusion (SCORE) study compared IVT triamcinolone with standard of care for both branch retinal vein occlusion (BRVO) and central retinal vein occlusion (CRVO). In the first year of the BRVO study, IOP-lowering medications were initiated in 57 of 138 (41%) eyes in the 4-mg triamcinolone group, 11 of 136 (8%) eyes in the 1-mg group, and 3 of 137 (2%) eyes in the standard care group (26). In the CRVO study, IOP-lowering therapy started in 32 of 91 (35%) eyes in the 4-mg triamcinolone group, 18 of 92 (20%) eyes in the 1-mg group, and 7 of 88 (8%) eyes in the observation group (25).

INTRAOCULAR HEMORRHAGE

Intraocular hemorrhage (i.e., retinal hemorrhage, vitreous hemorrhage, or hyphema) was reported occasionally after IVT injection, and although the etiology was unclear, in most instances, it was probably attributable to the underlying pathologic condition for which the injection was administered and not the IVT injection procedure per se. Jager et al. reported an overall prevalence of intraocular hemorrhage after IVT injections of 6% (176/2,940) per eye and 1.3% (176/13,400) per injection in an extensive review of the literature from 1966 to 2004.

More recent studies are consistent with a low prevalence of intraocular hemorrhage after IVT injection (25,26). Mason et al. (42) reported no hemorrhagic complications (choroidal hemorrhage, vitreous hemorrhage, or increased submacular hemorrhage) in a single-center retrospective case series review of 520 patients (675 eyes) receiving anti-VEGF injections in patient on systemic anticoagulants.

CATARACT

Cataract development and/or progression has been reported rarely after IVT injection, with exception of eyes treated with triamcinolone acetonide, fomivirsen, cidofovir, and methotrexate (9,23,43). The studies that reported the greatest number of cataract development were those evaluating the off-label use of triamcinolone acetonide. Early reports demonstrated cataract formation, most notably posterior subcapsular cataract, in 7% to 50% of patients (24,31,44). These early reports are limited by short follow-up periods that underestimated long-term cataract progression (44,45). By year 2 of the DRCR.net Study comparing IVT triamcinolone acetonide and focal–grid photocoagulation, 74% of eyes in the steroid injection arm, 23% (47 of 256 eyes) receiving 1 mg and 51% (101 of 254 eyes) receiving 4 mg, underwent cataract surgery (24). By year 3, the cumulative probability of cataract surgery was 31% in the laser group, 46% in the 1-mg triamcinolone group, and 83% in the 4-mg triamcinolone group (P < 0.001) (46). In the SCORE BRVO study, 60% of patients injected with triamcinolone, 25% (27 of 136 eyes) receiving 1 mg and 35% (38 of 138 eyes) receiving 4 mg, versus 13% (15 of 137 eyes) of laser-treated patients had onset or progression of lens opacity. Similar progression of lens opacity was seen in the SCORE CRVO study, where 59% of patients injected with triamcinolone, 26% (20 of 92 eyes) receiving 1 mg and 33% (25 of 91 eyes) receiving 4 mg, versus 18% (12 of 88 eyes) of observation patients had onset or progression of lens opacity (25,26).

Cataract formation was also common in eyes treated with antiviral agents, with a prevalence of 4.8% (44/909) per eye and 0.4% (44/10,839) per injection. Most of these cases were reported after administration of IVT fomivirsen (9).

IVT methotrexate administration was associated with an increased prevalence of cataract. A case series by Smith et al. (43), involving 494 IVT injections of methotrexate in 26 eyes, described cataract progression in 19 eyes (73%) when followed for 180 to 1,050 days. It was not described whether these were newly formed or existing cataracts; however, 12 of these eyes had undergone vitrectomy before initial IVT methotrexate therapy, and four additional eyes underwent vitrectomy before an additional course of IVT injection was administered.

HYPOTONY

Hypotony occurred rarely after IVT injection. Some studies described hypotony in 3.1% of eyes treated with IVT antiviral therapy, including 22 cases reporting low IOP after administration of cidofovir (21,22) and 6 cases after injecting fomivirsen (9

Stay updated, free articles. Join our Telegram channel

Join