Retinal nerve fiber layer thickness among eyes of HIV Positive patients under evaluation for glaucoma

Background: Retinal nerve fiber layer (RNFL) thinning, as measured by Cirrus ocular coherence tomography (OCT) occurs among HIV positive patients even in the absence of infectious retinopathy, and could be a marker of glaucomatous changes, but little is known about glaucoma progression among this group. We sought to determine whether HIV parameters are associated with RNFL changes.

Methods: Retrospective study of 329 HIV positive patients under evaluation for glaucoma was performed at Weill Cornell Ophthalmology between 1/1/2008 - 3/31/2015. Selected patients had undergone at least two Cirrus OCT assessments at least six months apart with a signal strength ≥7/10, focused images, and proper centration of the optic disc within the scan circle. Included eyes had open angles, visual acuity better than 20/40, no non-glaucomatous optic neuropathies, no history of intraocular pressure lowering therapy, no macular pathology, no history of infectious retinopathy, and no history of high refractive error (≥5.00D). Demographics, HIV, and glaucoma risk factor data were collected.

Results: 55 patients (110 eyes), 21.8% female, mean age 49.1 years, met inclusion criteria. At initial ophthalmology visit, median CD4 was 494, 80% had undetectable viral load, and 97.3% were on antiretroviral therapy. Between the first and final RNFL measurements there was thinning over time in the average, superior and inferior quadrant RNFL (p=0.018, p<0.01, and p=0.012, respectively). Patients with nadir CD4 <200mm³ had a statistically significant thicker RNFL in the superior and temporal quadrants at the time of OCT1 (p=0.048 and p<0.01, respectively) and OCT2 (p=0.034 and p<0.01, respectively) compared with patients with nadir CD4 count ≥200mm³.

Conclusion: As expected, average superior, and inferior quadrant RNFL thickness was observed to decrease over time, but those with nadir CD4 count <200mm³ had thicker RNFL in the superior and temporal quadrants at the first and final measurements. RNFL thickness as measured by OCT may be falsely reassuring in the setting of HIV, perhaps due to subclinical inflammation. Larger prospective studies are needed to further evaluate the progression of glaucoma and the impact of HIV parameters.