Objective:

To assess whether prolonged co-trimoxazole prophylaxis affects colonization with selected antibiotic-resistant bacteria among adults newly diagnosed with HIV in Dar es Salaam, Tanzania.

Approach:

• Design: Investigator-initiated, multicenter, double-blind, randomized, placebo-controlled CoTrimResist trial (NCT03087890).
• Participants: Eligibility includes nonpregnant adults older than 18 years with a first HIV-positive screening test and CD4 count ≥350 cells/µL; known co-trimoxazole allergy is an exclusion criterion.
• Intervention and outcomes: Participants are assigned 1:1 to daily co-trimoxazole or matching placebo for 1 year. Rectal and nasal/nasopharyngeal samples assess carriage of extended-spectrum beta-lactamase-producing E. coli and Klebsiella species, methicillin-resistant Staphylococcus aureus (MRSA), penicillin-nonsusceptible pneumococci, and vancomycin-resistant enterococci at baseline, 2 weeks, 6 months, and 1 year.

Key Findings:

• The trial includes 537 participants: 259 assigned to co-trimoxazole and 278 to placebo.
• Co-trimoxazole reduces S. aureus carriage at 2 weeks (risk difference, −7.3 percentage points; 95% CI, −12.0 to −2.6) and 6 months (−7.6; 95% CI, −13.9 to −1.3).
• MRSA carriage is reduced at 2 weeks (risk difference, −4.5 percentage points; 95% CI, −8.0 to −1.1), but the reduction does not persist.
• ESBL-producing E. coli carriage rises more rapidly in the co-trimoxazole group, although the primary analysis does not show a significant treatment effect.
• At 1 year, colonization rates do not differ significantly between groups. Mortality and serious adverse events also show no significant between-group differences.

Interpretation:

Co-trimoxazole produces temporary reductions in S. aureus and MRSA carriage without a persistent effect on resistant-bacterial colonization at 1 year. The early E. coli carriage pattern may reflect baseline imbalances and regression to the mean rather than a treatment effect.

Limitations:

• Baseline carriage and related risk factors differ between groups.
• Adherence is suboptimal and self-reported; follow-up falls to 74% at 1 year.
• Larger samples or longer follow-up may clarify unresolved effects.
• Using one swab for both nasal and nasopharyngeal sampling may reduce bacterial recovery.

Conclusion:

Prolonged co-trimoxazole prophylaxis shows transient effects on carriage but no significant persistent between-group differences in colonization at 1 year.

Sources:

• Impact of prolonged co-trimoxazole prophylaxis on colonization with antibiotic resistant bacteria in persons newly diagnosed with HIV-infection: CoTrimResist—a randomized placebo-controlled trial