ABSTRACT
OBJECTIVE: To evaluate clinical and laboratory results in HIV-infected patients with complete viral suppression under HAART (highly active antiretroviral treatment) for whom treatment was interrupted and to identify risk factors associated with prolonged (i.e., successful) treatment interruption. METHODS: Retrospective study of patients who interrupted therapy while on HAART with a plasma HIV RNA <400 copies/mL. Multivariate regression analysis was performed to identify factors associated with a prolonged interruption (more than 6 months). RESULTS: 36 treatment interruptions in 30 patients were analyzed. Patients' mean age was 42 years, 83% were men, and 60% were infected through homosexual contact. Median CD4 cell count at initiation of HAART was 292/mm3 and median viral load 43,000 copies/mL. Reported reasons for HAART discontinuation included patient or clinician choice (n=21) or drug toxicity (n=15). Median CD4 cell count when treatment interruption began was 606/mm3, and its median duration was 14 months. During treatment interruption, adverse clinical events or laboratory findings occurred in 9 patients (30%), all of whom had a CD4 cell count nadir < 300/mm3. When HAART resumed, median CD4 cell count was 302/mm3, and median viral load 59,800 copies/mL. Plasma HIV RNA dropped to <400 copies/mL in all patients within 4 months of resuming treatment. In multivariate analysis, the factors associated with resuming HAART within 6 month of treatment interruption were: HAART including non-nucleoside analog (adjusted Hazard Ratio [aHR]: 3.6, 95% CI: 1.2-10.6, p = 0.02), a CD4 cell count nadir < 300 (adjusted Hazard Ratio [aHR]: 5.5, 95% CI: 2.0-26, p = 0.0057), undetectable plasma HIV RNA for longer than 21 months at the interruption (aHR: 7.2, 95% CI: 2-26, p = 0.002). This probability was 45.5% in patients with a CD4 cell count nadir < 300 and 14.3% in the others (p 0.10). CONCLUSION: Antiretroviral treatment should be interrupted only with caution in patients with a CD4 cell count nadir <300/mm3 because of the risk of adverse clinical events or laboratory findings.
