Risk of Immune Recovery Uveitis in Patients With Cytomegalovirus Retinitis and the Acquired Immune Deficiency Syndrome

Abstract

Abstract: : Purpose: To evaluate the risk of immune recovery uveitis (IRU) in patients with cytomegalovirus (CMV) retinitis and AIDS. Methods: The Longitudinal Study of Ocular Complications of AIDS is ongoing at 19 AIDS ophthalmology centers. Enrollment data regarding the proportion of eyes with CMV retinitis which had clinically diagnosed IRU, cystoid macular edema (CME) and epiretinal membrane (ERM) were evaluated vis–a–vis potential risk factors. Results: Among 1780 patients with AIDS enrolled between 2 September 1998 and 31 August 2004, 1701 (96%) had complete immune recovery (IR) and IRU data available; CMV retinitis was present in 374 (22%) of them (539 eyes). Thirty–six patients (50 eyes) had IRU (9.6%). Among those with IR (CD4+ T cell count ≥100 cells/µL and ≥50 cells/µL above nadir), 31 had IRU (17.6%, 95% confidence interval (CI): 12.3%–24.1%). In addition to IR, ≥25% retinal involvement (OR=2.72, p=0.014) was a risk factor for IRU, whereas male gender (OR=0.26, p=0.012) and posterior pole involvement (OR=0.43, p=0.039) were associated with lower IRU risk. Active retinitis (p<0.001) and detectable CMV DNA in peripheral blood (p<0.001) were not seen in patients with IRU. Use of cidofovir (ever) (OR=1.52, 95% CI: 0.63, 3.71) and current use of anti–CMV therapy (OR=0.78, 95% CI: 0.29, 2.09) were not significantly associated with IRU. Eyes with IRU were much more likely to have CME (45.5% vs. 3.7%, p<0.001) and ERM (48.9% vs. 13.3%, p<0.001) than those without IRU, and somewhat more likely to have visual acuity (VA)≤20/50 (38.0% vs. 26.3%, p=0.077), but not VA≤20/200 (14.0% vs. 13.8%, p=0.96). Conclusions: The observed risk of IRU (17.6% of those with IR) in this multicenter study was substantial, although less than in some previous reports. Eyes with IRU had a higher burden of CME and ERM, and were more likely to have VA≤20/50 than eyes without IRU. Large retinal lesion size was confirmed as a risk factor. Large effects of cidofovir treatment (ever) and of anti–CMV therapy on risk were not observed, although moderate effects might not have been detectable with the available study power.

Keywords: AIDS/HIV • cytomegalovirus • clinical (human) or epidemiologic studies: outcomes/complications