Clinician vs. Reading Center Assessment of Cytomegalovirus Retinitis Lesion Area

Abstract

Abstract: : Purpose: To compare clinician and fundus photograph reading center (FPRC) assessments of cytomegalovirus (CMV) retinitis area at baseline and change in lesion area during follow–up, and to correlate these assessments with visual field scores. Methods: From patients enrolled in 3 of The Studies of the Ocular Complications of AIDS (SOCA) research studies, we identified a subset of 95 eyes with CMV retinitis restricted to zones 1 and 2 by clinician evaluation. Standard study data included CMV retinitis lesion area from (1) clinician estimate (nearest 5%) during the ophthalmic exam and (2) FPRC estimate (continuous) from an array of 60–degree photos. Visual field was measured with a Goldmann perimeter using the DRS protocol. Results: Mean area of CMV retinitis lesions at baseline by clinician estimate was 12.8% of total retinal area and by FPRC estimate was 6.2% of total retinal area (P<0.001 for difference). Clinician and FPRC area estimates were correlated (Pearson ρ= 0.78, P<0.001), although systematically offset (Concordance ρ=0.54, P<0.001). Both clinician and Reading Center area estimates were inversely correlated with visual field scores (Spearman ρ= –0.38 and –0.52, respectively, both P<0.001). Mean change in area over 3–month intervals was +1.2% by clinicians and +1.1% by FPRC (P=0.73 for difference, Concordance ρ=0.42, P<0.001). Change in visual field did not correlate significantly with change in retinitis area by either clinician or FPRC. Conclusions: Clinician estimates of baseline CMV retinitis area were greater than FPRC estimates, but both correlated significantly with visual field score. There was agreement between clinician and FPRC determinations of change in area over time, an important clinical indicator of success of therapy for CMV retinitis. Awareness of similarities and differences between clinician and FPRC estimates of lesion area should improve the ability of clinicians to apply research data in practice.

Keywords: AIDS/HIV • cytomegalovirus • clinical (human) or epidemiologic studies: treatment/prevention assessment/controlled clinical trials