Incidence of Cytomegalovirus (CMV) Retinitis in Second Eyes of Patients With the Acquired Immune Deficiency Syndrome and Unilateral CMV Retinitis
Accepted 4 January 2005.
Purpose
To evaluate the risk and risk factors for developing cytomegalovirus (CMV) retinitis in previously uninvolved second eyes among patients with unilateral CMV retinitis.
Design
Cohort study.
Methods
settings: Single-center academic AIDS ophthalmology practice. patient population: Three hundred seventy-six consecutive patients with AIDS and unilateral CMV retinitis were followed from the time of CMV retinitis diagnosis for the development of second-eye retinitis. experimental procedures: Demographic and clinical characteristics were noted at baseline. Use of highly active antiretroviral therapy (HAART) and immune recovery in response to HAART were noted prospectively. main outcome measure: Development of CMV retinitis in a previously uninvolved eye.
Results
Ninety-one percent of subjects received systemic anti-CMV treatment. Second-eye retinitis occurred in 26.1%/person-year (19.6% within the first 6 months), less than half the rate previously reported in untreated groups. Initial CD4+ T cell count >12 cells/μl, use of HAART, and initial posterior pole involvement were associated with 64%, 46%, and 41% reductions in incidence vis-à-vis comparison groups. Benefit from HAART was limited to that subset who developed immune recovery of a degree expected to restore innate control of CMV (a rise in the CD4+ T cell count by >50 cells/μl to a level >100 cells/μl).
Conclusions
The risk of second-eye retiniti is substantial in patients with unilateral CMV retinitis but appears to be reduced by anti-CMV therapy and by HAART-induced immune recovery. Patients are at highest risk when CD4+ T cell counts are very low and in the months immediately after CMV retinitis diagnosis.
aDepartment of Ophthalmology, The Johns Hopkins University School of Medicine, Baltimore, Maryland
bDepartment of Medicine, The Johns Hopkins University School of Medicine, Baltimore, Maryland
cDepartment of Epidemiology, The Johns Hopkins University Bloomberg School of Public Health, Baltimore, Maryland
dDepartment of Biostatistics, The Johns Hopkins University Bloomberg School of Public Health, Baltimore, Maryland
Inquiries to John H. Kempen, MD, PhD, 550 North Broadway, Suite 700, Baltimore, MD 21205; Fax: (410) 955-0629
Supported in part by grants EY00386 and EY07127 (J.H.K.) and EY10268 and EY00405 (D.A.J.).
ABSTRACT