Retrospective Study Finds Neovascular Glaucoma Etiology May Affect Surgical Outcomes

One-year outcomes varied among patients with differing neovascular glaucoma etiologies, but many achieved meaningful vision after surgical treatment that included glaucoma drainage devices.

Vision loss from neovascular glaucoma (NVG) is a challenging condition to manage, as it can manifest abruptly and often results in poor visual prognosis even after timely surgical treatment. NVG has various root causes, and a recent study in BMC Ophthalmology assessed surgical outcomes across etiologies in patients who were treated with glaucoma drainage devices (GDD).

Common causes of NVG include proliferative diabetic retinopathy (PDR) and ischemic central retinal vein occlusion (RVO), but ocular ischemic syndrome and inflammatory disorders can also be at the root of the condition. Urgent surgical management is often needed to control elevated intraocular pressure (IOP).

The single-center study retrospectively reviewed the records of patients with NVG to provide insight into visual acuity (VA) outcomes 1 year after glaucoma surgery in patients with PDR, RVO, and other etiologies. “No previous reports, that we are aware of, specifically address the differences in surgical outcome between etiologies of NVG, such as the differences in surgical outcomes between PDR and RVO patients,” the authors wrote.

Patients with diagnosed NVG and treated with GDD surgery at the University of Miami Clinical and Translational Science Institute Bascom Palmer Eye Institute or the Anne Bates Leach Eye Hospital between September 2011 and May 2019 were identified via electronic medical record review. Additional inclusion criteria were neovascularization of the iris or angle documented in the clinical notes and IOP greater than 21 mm Hg at baseline.

Patients who were younger than 18 years, had a history of the previous glaucoma procedures (cyclodestructive surgery, trabeculectomy, glaucoma drainage implant) in the affected eye, or had no 3-month follow-up visit after surgery were excluded.

Overall, 1850 eyes with NVG were identified in the initial search, and data from 120 eyes were included in the final analysis. Data points were presented at 3 and 12 months, with 3-month data used as surrogate final VA and IOP for 11 patients who were missing 1-year follow-up data.

Of the 120 eyes in the final analysis, 61.7% had NVG secondary to PDR, and 23.3% of NVG cases resulted from retinal vein occlusion. A further 15% were categorized as “other etiology.” All patients received either an Ahmed or a Baerveldt GDD device, or both. There was no significant difference between devices regarding final IOP results, but the mean number of glaucoma medications at final follow-up was 1.9 in eyes with Ahmed devices vs 1.3 in the Baerveldt group.

There were significant differences in 1-year surgical outcomes between etiologies, with 51% of PDR eyes achieving meaningful vision compared with 29% of RVO eyes. The study authors considered LogMar vision of 1.3 or 20/400 vision based on Snellen acuity the lower threshold of meaningful vision for the purpose of this analysis.

Only better VA at baseline and being on or starting a prostaglandin analogue at baseline were independently associated with vision of 20/400 or better in PDR eyes at final follow-up. For eyes with RVO, not receiving an intravitreal injection before or at the baseline visit was independently associated with meaningful final vision.

The secondary outcome measure was qualified success at final postsurgery follow-up, defined as IOP between 5 and 21 mmHg with or without IOP-reducing medication, no additional operation for elevated IOP at final follow-up, and no loss of light perception vision at 1 year. By this standard, 75% of patients overall had successful procedures, although the final IOP was lower in the other etiologies group compared with the PDR and RVO cohorts (11.9, 17.1, and 15.6 mmHg, respectively).

Overall, outcomes were better in patients with NVG caused by PDR vs RVO, and both GDD devices performed similarly in this analysis.

Limitations included the study’s retrospective nature and the large proportion of PDR eyes vs RVO and other etiologies. The highly experienced providers placing GDD may also bias the results.

“Direct prospective comparison of Baerveldt, Ahmed, and cyclophotocoagulation represents the next phase of discovery,” the authors concluded.


Medert CM, Sun CQ, Vanner E, Parrish RK 2nd, Wellik SR. The influence of etiology on surgical outcomes in neovascular glaucoma. BMC Ophthalmol. Published online December 20, 2021. doi:10.1186/s12886-021-02212-x