Ad. Monforte et al., PREDICTORS OF CYTOMEGALOVIRUS DISEASE, NATURAL-HISTORY AND AUTOPSY FINDINGS IN A COHORT OF PATIENTS WITH AIDS, AIDS, 11(4), 1997, pp. 517-524
Objective: To identify the predictors of acquiring cytomegalovirus (CM
V) disease, and to describe natural history, therapeutic management an
d autopsy findings in affected patients. Design: Observational study o
f a consecutive cohort of AIDS patients diagnosed and followed in the
same institution. Methods: All of the patients with CMV disease were i
ncluded. Statistical analyses were performed to establish the risk of
acquiring the disease at or after AIDS presentation, survival, and the
occurrence and lime of relapses in relation to maintenance therapy. T
he presence of CMV infection at autopsy was also investigated. Results
: CMV disease was diagnosed in 304 (24.8%) out of 1227 patients, its i
ncidence increasing according to the year of AIDS diagnosis. Women, ho
mosexual men, patients given zidovudine and Pneumocystis carinii pneum
onia (PCP) prophylaxis before AIDS, and severely immunodepressed patie
nts were at higher risk for the disease. CMV disease was an independen
t factor of worse survival (hazard ratio, 1.7 versus PCP; 95% confiden
ce intervals, 1.28-2.13). Patients untreated during the acute phase ha
d a 4.3 higher risk of dying than those treated. Relapses occurred les
s frequently and later in patients given continuous maintenance treatm
ent (23 out of 113; 17 months) than in untreated patients (13 out of 1
6; 3 months) or those given discontinuous therapy (22 out of 40; 7 mon
ths), whereas survival was independent from treatment. CMV infection w
as found in 97 out of 134 patients at autopsy, but was unassociated wi
th relapse. Conclusions: CMV is a severe disease whose frequency is hi
gher in severely immunodepressed patients. Continuous treatment leads
to a lower relapse rate even if it does not change survival or eradica
te the infection.