Current issues of ACP Journal Club are published in Annals of Internal Medicine


Oral ganciclovir prevented cytomegalovirus disease in AIDS

ACP J Club. 1996 Nov-Dec;125:57. doi:10.7326/ACPJC-1996-125-3-057

Source Citation

Spector SA, McKinley GF, Lalezari JP, et al., for the Roche Cooperative Oral Ganciclovir Study Group. Oral ganciclovir for the prevention of cytomegalovirus disease in persons with AIDS. N Engl J Med. 1996 Jun 6;334:1491-7.



To determine the effectiveness of oral ganciclovir for the prevention of cytomegalovirus (CMV) disease in patients with AIDS.


Randomized, double-blind, placebo-controlled trial with median 367-day follow-up.


19 centers in the United States.


725 adults (median age 39 y, 82% white, 99% men) who were infected with HIV and had either 2 CD4+ cell counts of ≤ 50 cells/mm3 or, in those with a history of an AIDS-defining opportunistic infection, ≤ 100 cells/mm3; all patients also had CMV infection confirmed by positive antibody test or urine culture. Exclusion criteria were CMV disease, active gastrointestinal disease, absolute neutrophil count < 750 cells/mm3, platelet count < 50 000/mm3, creatinine clearance rate < 70 mL/min, or a Karnofsky score < 60.


Patients were stratified by CD4+ cell count and allocated in a 2:1 ratio to ganciclovir, 1000 mg 3 times/d (n = 486), or placebo (n = 239). All patients could receive antiretroviral therapy and oral acyclovir. Patients were seen every 2 months for follow-up that included dilated-eye examinations by an experienced ophthalmologist.

Main outcome measures

The primary outcome was 12-month incidence of confirmed CMV disease. Secondary outcomes were CMV-related retinitis, colitis, esophagitis, gastroenteritis, hepatitis, pneumonia, polyradiculopathy, and other CMV disease.

Main results

Treatment with ganciclovir led to fewer patients with CMV disease than did placebo (14% vs 26%, P < 0.001). {This absolute risk reduction (ARR) of 12% means that 8 patients would need to be treated (NNT) with ganciclovir for 12 months (compared with placebo) to prevent 1 patient from developing CMV disease, 95% CI 5 to 17; the relative risk reduction (RRR) was 46%, CI 27% to 60%.}* Treatment with ganciclovir also led to a lower incidence of CMV retinitis at 12 months than did placebo (12% vs 24%, P < 0.001) {ARR 12%; NNT 8, CI 5 to 17; RRR 50%, CI 30% to 64%}*. The groups did not differ for incidence of colitis, esophagitis, gastroenteritis, hepatitis, pneumonia, polyradiculopathy, or other CMV disease. Neutropenia (≤ 750 cells/mm3) occurred in more patients who received ganciclovir than in those who received placebo (26% vs 13%, P < 0.001). The groups did not differ for survival (21% for ganciclovir vs 26% for placebo, P = 0.14).


Oral ganciclovir was effective in the prevention of cytomegalovirus disease in patients with AIDS.

Source of funding: Roche Pharmaceuticals.

For article reprint: Dr. S.A. Spector, University of California, San Diego, Clinical Sciences Building, 9500 Gilman Drive, La Jolla, CA 92093-0672, USA. FAX 619-534-7411.

*Numbers calculated from data in article.


Most patients with AIDS are latently infected with CMV, but only about 20% to 30% develop CMV disease. The decision to prescribe a prophylactic regimen for CMV depends not only on the risk for disease and the efficacy of the regimen but also on the side effects, cost-effectiveness, and effect on quality of life.

This multicenter study supports the theory that oral ganciclovir is effective in preventing CMV disease in HIV-infected patients with low CD4+ cell counts. The primary benefit was prevention of CMV retinitis. This study involved a diligent search for CMV retinitis, which may or may not have produced visual symptoms. The second and yet-to-be-published trial from Community Programs for Clinical Research on AIDS (CPCRA) (1) reached a different conclusion: Prophylaxis with oral ganciclovir was ineffective in preventing symptomatic CMV disease. These discrepant results may partly be explained by differences in CD4+ counts and in how patients were followed. The CPCRA trial only evaluated patients for CMV disease if they developed symptoms.

In analyzing these studies from the perspective of quality of life, it is reasonable to ask whether preventing asymptomatic or minimally symptomatic CMV retinitis with a very expensive and potentially toxic agent is a reasonable goal. Because neither study showed an effect on survival, the decision to use this prophylactic regimen should be based on quality of life and cost-effectiveness. Until studies that focus on these questions are done, widespread use of oral ganciclovir prophylaxis cannot be recommended.

Chris E. Forsmark, MD
University of FloridaGainesville, Florida, USA


1. Brosgart CL, Craig C, Hillman D, et al. In: Program and abstracts of the 35th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Francisco, September 17-20, 1995.