A retrospective cohort study has identified a statistically significant association between the use of nucleoside and nucleotide reverse transcriptase inhibitors (NRTIs)—which are commonly used in the treatment of HIV and hepatitis B—and an increased risk of primary open-angle glaucoma (POAG).
Researchers from the University of Pennsylvania analyzed data from 305,441 participants who were aged 40 years or older and had linked electronic health records from the NIH All of Us database. A cohort of 718 NRTI users was identified. Patients were excluded if they had a diagnosis of POAG prior to NRTI use. These users were matched with 7,180 nonusers in a 1:10 ratio using propensity score matching that was then adjusted for key covariates including age, race, sex at birth, HIV and hepatitis B diagnosis, and family history of POAG. Additional sensitivity analyses were conducted to control for self-reported eye care utilization and geographical differences in ophthalmologist availability. Their findings were recently published in Ophthalmology.
Key Findings
POAG was diagnosed in 4.32% of NRTI users compared to 2% of matched nonusers (unadjusted odds ratio [OR] = 2.21, 95% confidence interval [CI]=1.48–3.28, P<.001). Use of any NRTIs was still significantly associated with increased risk of developing POAG after adjusting for HIV, hepatitis B, and family history of POAG (OD=1.84, 95% CI=1.22–2.77, P=.004). The risk was 2.30 (95% CI=1.07–4.96, P=.033) after adjusting for self-reported eye doctor visits.
Self-reported visits to eye doctors were similar between groups—29.8% for NRTI users and 29.9% for nonusers. The distribution of NRTI users was not correlated with ophthalmologist density by ZIP code.
The researchers found no significant association between NRTI use and risk of cataracts or angle-closure glaucoma.
Among HIV-negative NRTI users (n=275), no statistically significant association with POAG was observed. The role of HIV infection may, therefore, warrant further investigation, the authors suggested, writing, “Due to the low n of HIV- NRTI users in our study, we are not equipped to rule out the possibility that HIV infection, including its duration and severity, contributes to the association between NRTI use and POAG risk we describe here.”
An association between NRTI use and mitochondrial toxicity is well documented, and this study builds on prior evidence from animal models and human genetic studies that has linked mitochondrial abnormalities with POAG, the researchers noted. Their findings also complement prior work from an Australian study that found higher rates of intraocular pressure–lowering medication use in younger males who were on antiretroviral therapy, which included NRTIs among other classes.
Limitations and Future Directions
Although the study employed rigorous statistical controls, its retrospective nature limits causal interpretation. The small number of HIV-negative NRTI users also restricts definitive conclusions about the independent effect of NRTIs apart from HIV-related factors.
The authors conclude that prospective studies can investigate causality and explore underlying mechanisms, including the role of NRTI-induced mitochondrial toxicity in glaucomatous neurodegeneration. GP