Endophthalmitis rate low after intravitreal injections

August 1, 2008

The development of endophthalmitis after intravitreal injection of anti-vascular endothelial growth factor drugs is a rare event, with an incidence of 0.03% for treated cases and 0.02% for culture-positive cases.

Key Points

Fort Lauderdale, FL-Monitoring of the incidence of culture-proven endophthalmitis following anti-vascular endothelial growth factor (VEGF) intravitreal injections has indicated that the rate of clinically suspected infectious endophthalmitis is low after intravitreal injections of anti-VEGF agents over 2.5 years, Andrew A. Moshfeghi, MD, reported at the annual meeting of the Association for Research in Vision and Ophthalmology.

He described a retrospective study that was designed both to identify the rate of clinically suspected/infectious endophthalmitis after intravitreal injections during the anti-VEGF era and to characterize the identified cases to gain a better understanding of the types of organisms involved and the timing of the infection and its course. Dr. Moshfeghi is a medical director at the Palm Beach Gardens, FL, campus of the Bascom Palmer Eye Institute and an assistant professor of ophthalmology at the Miller School of Medicine, University of Miami.

He defined the anti-VEGF era as beginning when the FDA approved pegaptanib sodium (Macugen, [OSI]Eyetech/Pfizer) in January 2005 until the time of the presentation. This period also includes the FDA approval of ranibizumab (Lucentis, Genentech) in June 2006. Bevacizumab (Avastin, Genentech) has been administered intravitreally since May 2005.

A total of 24,865 injections of the anti-VEGF drugs bevacizumab, ranibizumab, and pegaptanib were administered off-label at all campuses of the Bascom Palmer Eye Institute, Dr. Moshfeghi reported. Most of the injections were given to treat wet age-related macular degeneration. Other cases included, however, involved treatment for central retinal vein occlusion (CRVO), branch retinal vein occlusion, diabetic macular edema, proliferative diabetic retinopathy, cystoid macular edema (CME), and neovascular glaucoma.

Dr. Moshfeghi and colleagues reviewed the medical records of patients with suspected infectious endophthalmitis following an intravitreal injection. They found that seven cases of clinically suspected endophthalmitis developed; five of these cases were culture-positive, and two were culture-negative.

"This translated to a rate of endophthalmitis almost 0.03% for 'treated cases' and 0.02% for culture-positive cases," he said.

Representative cases

A representative case was that of an 81-year-old man treated with pegaptanib, three injections of bevacizumab, and one injection of ranibizumab for wet AMD. The patient presented the day after the injection of ranibizumab with a red, painful eye; visual acuity (VA) of hand motions; hypopyon; and vitritis. Results of the vitreous cultures were negative. VA recovered to the baseline level. One year later, VA improved to 20/80.

Another case was that of a 59-year-old man with CRVO and a VA of 20/400. The patient was treated with off-label bevacizumab and was re-treated for persistent CME with a VA of 20/80. Although the hypopyon and anterior chamber fibrin appeared to contract and improved on day 4 after treatment, the patient's condition continued to worsen. Cultures identified Streptococcus sanguinis/gordoni that was refractory to treatment with erythromycin but responded to penicillin, vancomycin, ceftriaxone, and levofloxacin. VA rapidly decreased to no light perception with 360° ciliochoroidal detachment and a retinal detachment.

Further analysis indicated that most of the cases were caused by gram-positive Streptococcus. Most patients returned to the baseline VA level, but two patients developed no light perception vision.

"Most patients presented within 24 to 48 hours after the intravitreal injection with pain and decreased vision and were primarily found to have gram-positive organisms. Although good outcomes were seen in a few cases, certainly dismal outcomes were also encountered," Dr. Moshfeghi recounted.

Tom Harper, MD, and Harry W. Flynn Jr., MD, who collaborated in the study, are carrying out a subgroup analysis of the cases presenting with streptococcal infection, because they tend to have a unique presentation and can have a very poor treatment course, he added.