University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Daily Cotrimoxazole in severely immunosuppressed HIV-infected adults in Africa started on combination antiretroviral therapy: An observational analysis of the DART cohort
Global Health Sciences Literature Digest
Published August 30, 2010
Journal Article

Walker AS, Ford D, Gilks CF, et al. Daily Cotrimoxazole in severely immunosuppressed HIV-infected adults in Africa started on combination antiretroviral therapy: An observational analysis of the DART cohort. Lancet. 2010;375:1278-86.

In Context

Prophylactic use of cotrimoxazole is recommended for HIV-infected persons with CD4 counts <350 cells/mm3 in resource-constrained areas.(1) Cotrimoxazole is used as prophylaxis against Pneumocystis jirovecii for persons whose CD4 counts are 200 cells/mm3 or below and is discontinued when the CD4 cells increase. Antiretroviral therapy (ART) and Cotrimoxazole are routinely used together in developed countries but are less so in resource-constrained areas. Data on the effectiveness, side effects, toxicity, and adherence to ART used in combination with cotrimoxazole in such settings are limited.


To measure the effect of cotrimoxazole on survival and markers of clinical disease progression


Three clinical centers in Uganda and one in Zimbabwe

Study Design

Prospective cohort


ART-naïve (apart from use for the prevention of mother-to-child transmission), HIV-infected adults with CD4 counts <200 cells/µL who were starting ART


Survival, CD4 counts, World Health Organization (WHO) stage 3 and 4 events, malaria, body mass index, and hematological indices


The Development of Antiretroviral Therapy in Africa (DART) trial compared laboratory plus clinical monitoring (LCM) to clinically driven monitoring (CDM) of ART in an open, non-inferiority trial in which participants were randomly assigned to either arm.(2) All participants initiated treatment with co-formulated zidovudine-lamivudine plus tenofovir, abacavir, or nevirapine. Participants were seen every four weeks, and hematology, biochemistry, and CD4 cell counts were done at four weeks and every 12 weeks thereafter. In the LCM group, results were available to clinicians and in the CDM group, results (apart from CD4-cell count) could be requested if clinically indicated.

Cotrimoxazole was given once daily but was not routinely given or monitored as part of the study. Antiretroviral therapy, other medication use, and malaria episodes were recorded at each visit. Summaries for all WHO stage 4 events and deaths were reviewed by an endpoint review committee who was blinded to the participants' group assignment and CD4 count.

Marginal structural models were used to estimate the causal effects of cotrimoxazole on outcomes adjusting for time-dependent co-variates with inverse probability treatment weights that were appropriate for other time-dependent co-variates.


There were 3316 participants enrolled between January 15, 2003 and October 24, 2004 (LCM, n=1659 and CDM, n=1662). Excluded subjects included 137 who entered a pilot study of structured ART interruption. All participants had CD4 counts of <200 cells/mm3 at the time treatment began.

There were 324 (10%) participants who did not receive cotrimoxazole prophylaxis, 1959 (62%) were on cotrimoxazole when staring ART, 896 (28%) began cotrimoxazole while they were already on ART. There were 368 participants who died before the end of follow-up, and 7% of these were missing information on cotrimoxazole use and were censored at their last clinic visit. Adherence to ART was high in participants on and off cotrimoxazole.

There were 326 deaths and 14,214 person years of follow-up. Eighty-five deaths occurred within 12 weeks of starting ART. Cotrimoxazole prophylaxis was associated with a decreased risk of death in the first 12 weeks on ART regardless of whether cotrimoxazole was started prior to ART (adjusted odds ratio [AOR] 0.52, 95% confidence interval [CI]: 0.30-0.92) or at the time of ART initiation (AOR 0.46, 95% CI: 0.25-0.84). In an unweighted logistic regression model that adjusted for baseline but not time-dependent variables, the overall risk of death was reduced by 27% with the use of cotrimoxazole (OR 0.73, 95% CI: 0.56-0.96). In a marginal structural model that controlled for time-dependent variables, the reduction in overall mortality risk was 35% (OR 0.65, 95% CI: 0.50, 0.85). The benefit from cotrimoxazole decreased with increasing time on ART with no effect found after 72 weeks on ART.

Cotrimoxazole use reduced new or recurrent WHO stage 3 or 4 events by 15% (OR 0.85, 95% CI: 0.74-0.98) and the risk of malaria by 26% (OR 0.74, 95% CI: 0.63-0.88). Cotrimoxazole had no effect on WHO stage 4 events, CD4 count, or body mass index.


The results from this study support the WHO recommendations for use of cotrimoxazole prophylaxis and should be provided to all patients initiating ART and continued for 72 weeks.

Quality Rating

The study sample is likely to be representative of in-care patients with late HIV disease in eastern Africa. The exposure to cotrimoxazole varied by study site, year of entry into the study, and the type of ART. These all serve as potential sources of bias. However, the authors adjusted the analyses whenever possible to account for potential biases. Although not a clinical trial, using data from a well-conducted trial provided detailed clinical information on baseline characteristics, exposure, and outcomes as well as good follow-up of participants. Assessment of outcomes was unlikely to be biased. The multiple analysis and adjustment for potential confounders strengthens the study.

Programmatic Implications

This study provides good data in support of cotrimoxazole use among persons on ART and supports findings from other studies(3, 4) but does not provide detailed information on the diseases that Cotrimoxazole is preventing, other than malaria. As such, the mechanism of action remains uncertain. This however, should not dissuade health care workers from prescribing cotrimoxazole. Greater coverage with cotrimoxazole could result in substantial decreases in death.


  1. WHO Guidelines on Cotrimoxazole prophylaxis for HIV-related infections among children, adolescents and adults: recommendations for a public health approach. World Health Organization. 2006.
  2. DART Trial Team. Routine versus clinically driven laboratory monitoring of HIV antiretroviral therapy in Africa (DART): a randomised non-inferiority trial. Lancet 2010;375:123-31.
  3. Mermin J, Lule J, Ekwaru JP. Effect of Cotrimoxazole prophylaxis on morbidity, mortality, CD4-cell count, and viral load in HIV infection in rural Uganda. Lancet 2004;364:1428-34.
  4. Badri M, Ehrlich R, Wood R, Maartens G. Initiating Cotrimoxazole prophylaxis in HIV-infected patients in Africa: an evaluation of the provisional WHO/UNAIDS recommendations. AIDS 2001;15:1143-8.