Endophthalmitis Rare in Intravitreal Injections

Laird Harrison

April 29, 2016

Vigilance and quick treatment can save the vision of most patients who develop endophthalmitis as a rare result of intravitreal injections, researchers say.

Vision returned to baseline in eight of 13 eyes with endophthalmitis in a case series of 17,359 injections published in the May issue of Ophthalmology.

The case series yielded a 5-year incidence of seven in 10,000 injections, of which five in 10,000 were microbiologically proven. The investigators note that this is higher than the incidence reported in previous studies, which only included microbiologically proven cases, and most of which did not use polymerase chain reaction.

"Although intravitreal injections are considered a great advancement in the management of retinal diseases because they have potential to improve vision, endophthalmitis is a concern to a treating ophthalmologist," write Rajiv Raman, MD, DNB, from Shri Bhagawan Mahavir Vitreo Retinal Services, Sankara Nethralaya, Chennai, Tamil Nadu, India, and colleagues.

Previous studies have shown a prevalence of postinjection endophthalmitis of between 0% and 0.87%, the researchers write.

To understand the risk better, they retrospectively reviewed records of 17,359 intravitreal injections administered between January 1, 2010, and December 31, 2014, at two centers of a tertiary eye care institute in India.

Physicians administered 9932 injections of bevacizumab (Roche), 4108 of ranibizumab (Lucentis, Novartis), and 3319 of triamcinolone acetate (Kenalog, Bristol-Myers Squibb).

Physicians withdrew bevacizumab directly from vials containing 4 mL and loaded it into syringes just before injection. They grouped patients together for treatment to provide multiple doses from each vial. Five of the patients who developed endophthalmitis received injections from the same vial of bevacizumab.

For the ranibizumab and triamcinolone, the physicians prepared injections directly from a single-use vial.

At the start of the study period, all patients received at least 1 day of ciprofloxacin as prophylaxis. Beginning in February 2013, only patients whose fellow eye was blind or whose eyes had signs of blepharitis received ciprofloxacin. There was no statistically significant difference in the incidence of endophthalmitis between those who received the prophylaxis and those who did not.

The patients with endophthalmitis presented symptoms, on average, 4 days after the injection, typically complaining of decreased visual acuity, pain, and redness. Nine had hypopyon, three had corneal edema, one had increased intraocular pressure, and all 13 had vitritis.

Six of the patients with endophthalmitis were being treated for diabetic maculopathy, four for choroid neovascular membranes, one for proliferative diabetic retinopathy, and two for macular edema. Eleven of them had diabetes.

Eight of the patients with endophthalmitis received bevacizumab, one ranibizumab, and four triamcinolone. Gram staining was positive in six of the patients, three of whom showed Gram-positive cocci and three of whom showed Gram-negative bacilli.

In five of the patients, polymerase chain reaction screening was positive: four for eubacterial genome and one for both Eubacterium genome and Propionibacterium acnes.

Culture-positive results showed staphylococcus in three patients and Acinetobacter calcoaceticus in one.

The investigators found Burkholderia species in the vial of bevacizumab used in the cluster of five patients who were infected after being injected from it, but the organism grown from the vial could not be isolated from the patients' ocular fluids. In addition, these patients were negative on culture and Gram staining. One was positive for eubacterial genome on polymerase chain reaction.

Because patients treated from other vials of the same batch did not have postinjection problems, the investigators speculate that only that vial was contaminated.

The physicians treated five of the patients with endophthalmitis by injecting antibiotics alone, and the other eight with pars plana vitrectomy combined with antibiotics injection.

Presenting visual acuity ranged from counting fingers to perception of light. At 3 months' follow-up, final visual acuity ranged from counting fingers to 6/9. The improvement was statistically significant (P = .001). One patient developed hypotony and cataract.

The researchers have disclosed no relevant financial relationships.

Ophthalmology. 2016;123:1162-1164. Abstract

For more news, join us on Facebook and Twitter


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.