Laird Harrison

August 15, 2016

SAN FRANCISCO — Vancomycin administered during cataract procedures can sometimes cause blindness, researchers reported here at the American Society of Retina Specialists (ASRS) 2016 Annual Meeting.

Final visual acuity was 20/200 or worse in patients who developed hemorrhagic occlusive retinal vasculitis, probably as a hypersensitivity reaction to the antibiotic, said Andre Witkin, MD, from the New England Eye Center at Tufts University in Boston.

Although "we think it's very rare," he told Medscape Medical News, "in most patients, it caused severe vision loss."

The ASRS and the American Society of Cataract and Refractive Surgeons formed a joint task force to investigate reports of the syndrome.

After the research team searched the literature and contacted retina specialists from across the United States, they identified 35 eyes in 22 patients that met their criteria for hemorrhagic occlusive retinal vasculitis.

The characteristic findings were a normal exam the first day after the cataract procedure followed by delayed-onset vision loss, which was sometimes asymptomatic. Other findings included mild to moderate anterior chamber and vitreous inflammation, sectoral hemorrhage in a nonperfused retina (often along venules), a retinal hemorrhage (often large or confluent), and peripheral ischemia with macular ischemia in the advanced disease.

In many cases, these findings were associated with a similar reaction in the fellow eye, often with a faster onset and a more severe course, no significant increase in venous dilation or tortuosity, and rapid progression to neovascular glaucoma.

The patients ranged in age from 51 to 84 years, and 15 of the 22 patients were women. Thirty-two eyes received a bolus intracameral injection of vancomycin during surgery, two received vancomycin through the irrigation bottle, and one received it through an intravitreal injection.

Extensive work-ups were negative in all 22 patients, and systemic corticosteroids were administered to 15 patients. In addition, 20 of the 35 eyes received anti-vascular endothelial growth-factor (VEGF) treatment, 15 underwent panretinal photocoagulation, and three received intravitreal corticosteroids.

Vision Loss

Despite these treatments, 22 eyes ended up with 20/200 vision or worse, and eight of these had no light perception. Nineteen eyes developed neovascular glaucoma.

Seven of the 35 eyes were treated with intravitreal vancomycin because clinicians suspected endophthalmitis. Five of these ended up with no light perception and for the two remaining eyes, visual acuity was 20/400 and 20/800.

Hemorrhagic occlusive retinal vasculitis is probably related to type 3 hypersensitivity to vancomycin, Dr Witkin explained. As evidence of this, he pointed out that similar skin reactions to the drug have been documented and that the condition affects the venules.

If this is the case, a skin test might not help identify at-risk patients because skin testing only works for type 1 and some type 4 sensitivities, he reported.

In addition, hemorrhagic occlusive retinal vasculitis seems to be dose-dependent; outcomes were worse in eyes that received a second dose of vancomycin and better in those where the original dose was smaller.

The condition appears to be rare, but the task force could have overlooked milder cases that were never reported, he said.

And the frequency of the condition might be increasing because more clinicians are injecting intracameral vancomycin. A set of surveys showed that in 2007, 4.2% of surgeons were injecting the antibiotic; this rose to 9.2% in 2014, he reported.

Given these findings, Dr Witkin said he recommends that cataract surgeons weigh the risk for hemorrhagic occlusive retinal vasculitis against the risk for endophthalmitis and consider avoiding vancomycin when doing bilateral surgery.

He also suggested that they conduct a dilated exam before operating on the second eye and consider using cefuroxime or moxifloxacin instead of vancomycin.

For retina specialists, he recommended a work-up to rule out other syndromes, such as viral retinitis. And if they suspect hemorrhagic occlusive retinal vasculitis, they should avoid intravitreal vancomycin.

Instead, Dr Witkin said he recommends the aggressive use of steroids, possibly intraocular, as well as systemic and topical, and early use of anti-VEGFs and panretinal photocoagulation.

He asked that physicians submit new cases to a registry on the ASRS website.

After his presentation, a member of the audience asked if the cases could be related to excessive doses of vancomycin. Dr Witkin explained that all the patients in this series received 1 mg of vancomycin except one, who received 2 mg.

Vancomycin 1 mg is considered to be a safe dose, and the delayed onset suggests that the syndrome is not the result of toxicity, he pointed out.

In response to another question from the audience, Dr Witkin reported that the task force has not ruled out other possible causes, and did not include the use of the drug in the definition of hemorrhagic occlusive retinal vasculitis.

"I don't think this is a new phenomenon," said session moderator Damien Rodger, MD, PhD, from the Roski Eye Institute at the University of Southern California in Los Angeles.

He explained that he has observed obliterated retinal blood vessels in some patients he has treated with vitrectomies for this type of infection, and speculated that vancomycin can cause harm in some patients with endophthalmitis.

"We thought it was the infectious problem, but it could be the vancomycin that we originally treated them with," he told Medscape Medical News.

Still, it is too early to ban the use of vancomycin in all cataract procedures, he said. "Until we know exactly who is at risk, making a blanket statement to avoid it would be a bit premature."

Dr Witkin has disclosed no relevant financial relationships. Dr Rodger reports financial relationships to Allergan and Second Sight.

American Society of Retina Specialists (ASRS) 2016 Annual Meeting. Presented August 14, 2016.


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.