Abstract Form

REFERENCE NUMBER : 521
ECCATH ID : P301
8th EUROPEAN CONFERENCE ON CLINIC ASPECTS AND TREATMENT OF HIV - INFECTION
Location of research or project (country)
 
International
Thematic Areas:
8.1 Viral diseases
8.1 Viral diseases
12 EPIDEMIOLOGY/COHORT STUDIES
Title

FACTORS AFFECTING INCIDENCE OF AND SURVIVAL AFTER CMV END-ORGAN DISEASE IN PATIENTS WITH AIDS IN EUROPE.

Author: I Yust1, P Reiss, D Turner1, O Kirk2, C Katlama, A Mocroft, J van Lunzen, M Burke1, A Chiesi, JD Lundgren2 for the EuroSIDA Study Group.
1Dept of Medicine A, Tel Aviv Sourasky Medical Center, Tel Aviv, Israel, 2EuroSIDA Co-ordinating Center, Dept of Infectious Diseases, Hvidovre Hospital, Hvidovre , Denmark


Background of study: In the HAART era, there has been a decrease in the incidence of and mortality from CMV infection, including both retinal (CMVR) and extraocular (CMVEO) CMV end-organ disease (CMVD).

Objective: To study which factors affect incidence of and survival after diagnosis of CMVD.

Design: The EuroSIDA study is a multicenter prospective cohort study of 8495 HIV seropositive patients. The present study group included all AIDS subjects diagnosed as having some form of CMVD. Statistics included univariate and multivariate Cox models, the chi-squared test and the Wilcoxon or Kruskall-Wallis tests.

Results: Following a prospective diagnosis of CMVD 283 out of a total of 393 patients (72%) died. The incidence was 59 deaths per 100 patient years of follow up (PYFU) [95% Confidence limits 52.1-62.9]. The median survival (MS) was 11 months for CMVR and 7 months for CMVEO. Within the latter group, MS after gastrointestinal CMV infection was significantly higher (p < 0.0002) at 12 months than for neurological CMV infection (4.5 months) or other CMVEO (3 months). Both the incidence of CMVD and mortality from the disease have declined over the past few years. The incidence of CMVD decreased from 6.2 [5.2-7.2] in 1995 to 0.2 [0.1-0.4] in 1999; decreasing for both CMVR and CMVEO. Differences in survival according to calendar year of diagnosis were highly significant (p < 0.0001). 88% of patients with CMVD diagnosed during ‘94-‘95 died, as opposed to 50% in ‘96-’97 and 30% in ‘98-’99. By univariate analysis, starting HAART was associated with a 33% decrease in mortality (p < 0.0001). Lower body weight, CD4 cell counts, and hemoglobin (time-updated variables) were all associated with an increased mortality (p < 0.0001). Similar results were obtained by multivariate analysis and remained . Death in patients with CMVD was due to CMVEO in 16%, tumor in 13%, bacterial pneumonia in 13%, non-CMV opportunistic infection in 12%, AIDS wasting or dementia in 10%, and miscellaneous or unknown causes in 36% of cases.

Conclusion: The decrease in mortality from CMV infection is associated with several factors, including a decreased incidence of CMVD, as well as improved body weight, higher CD4 cell counts and hemoglobin levels, most of these presumably associated with HAART. Further, the site of CMV infection plays a role: patients with gastrointestinal or retinal involvement alone did better than those with neurologic or other extraocular involvement.

Authors address:

Israel Yust, Department of Medicine A, Tel Aviv Sourasky Medical Center Tel Aviv, Israel, E-mail: intmeda@tasmc.health.gov.il

011028