Home Search Abstracts View Session E-mail Abstract Author


Session 130 Poster Abstracts
HIV Care in Different Settings
Session Day and Time: Wednesday, 1-4 pm
Room: Hall B


806    
Delayed Diagnosis of HIV Infection and Late Initiation of ART in the Swiss HIV Cohort Study
M Wolbers1, H Bucher1,2, H Furrer3, M Rickenbach4, M Cavassini4, R Weber5, P Schmid6, E Bernasconi7, B Hirschel8, Manuel Battegay*2, and the Swiss HIV Cohort Study
1Basel Inst for Clin Epidemiology, Switzerland; 2Univ Hosp Basel, Switzerland; 3Univ Hosp Berne, Switzerland; 4Univ Hosp Lausanne, Switzerland; 5Univ Hosp Zurich, Switzerland; 6Cantonal Hosp, St Gall, Switzerland; 7Regional Hosp Lugano, Switzerland; and 8Univ Hosp Geneva, Switzerland

Background:  Delays in diagnosis and late initiation of ART lead to substantial HIV-related morbidity and mortality.
Methods:  We defined 2 subpopulations of the Swiss HIV cohort:  To investigate delayed HIV diagnosis, we included 1915 patients with HIV diagnosis within 3 months prior to cohort registration (group A); to study late ART initiation despite cohort follow-up, we included 1730 treatment-naive patients with CD4 cell counts ≥200 cells/μL before their second cohort visit (group B). In group A, prognostic factors for a low initial CD4 cell count were examined with a median regression model. In group B, we compared patients initiating ART with CD4 cell counts ≥200 cells/μL to those dropping to CD <200 cells/μL with a logistic regression model. Time to ART uptake was studied in both groups with Kaplan-Meier curves and Cox models.
Results:  Median initial CD4 cell count for patients in group A was 331 cells/μL; 31% and 10% were <200 and <50 cells/μL, respectively. Independent risk factors for lower initial CD4 cell counts were older age and non-Caucasian race, whereas patients with men who have sex with men, intravenous drug users, and patients living alone had higher CD4 cell counts (all p ≤0.003, except for living alone: p = 0.04). In group B, 30% initiated ART with CD4 cell counts ≥200 cells/μl and 18% dropped to CD4<200 cells/μL before starting ART; only 26 patients dropped to CD4<50 cells/µL without prior ART. A sub-Saharan country of origin was independently associated with a higher probability of initiating ART with CD4 ≥200 cells/μL (OR = 2.91, 95%CI 1.54 to 5.49) during cohort follow-up; other covariates including intravenous drug usage were non-significant. Median CD4 counts at ART initiation were 207 cells/µL and 253 cells/µL in groups A and B, respectively. Median times to ART initiation from HIV diagnosis in group A patients with initial CD4<200 cells/µL and from first CD4<200 cells/µL in group B patients were 35 and 63 days, respectively. In both groups, the value of the CD4 cell count <200 cells/μL independently predicted the time to ART initiation (HR = 0.63 [95% CI 0.54 to 0.72] and HR = 0.73 [95%CI 0.55 to 0.98] per +100 cells/μL, respectively).
Conclusions:  Low CD4 cell counts at ART initiation and, particularly, very low CD4 cell counts (<50 cells/µL) before starting ART are predominantly due to late presentation. As guidelines favor starting ART before CD4 drop <200 cells/µL an earlier diagnosis and a stringent follow-up are paramount.