RESUMO
BACKGROUND: Trimethoprim-sulfamethoxazole (TMP-SMX) is an effective treatment for Pneumocystis jirovecii pneumonia (PCP) in immunocompromised patients with and without HIV infection; however, a high incidence of adverse events has been observed. Low-dose TMP-SMX is a potentially effective treatment with fewer adverse events; however, evidence is limited. RESEARCH QUESTION: What is the efficacy and safety of low-dose TMP-SMX for non-HIV PCP compared with conventional-dose TMP-SMX after adjusting for patient background characteristics? STUDY DESIGN AND METHODS: In this multicenter retrospective cohort study, we included patients diagnosed with non-HIV PCP and treated with TMP-SMX between June 2006 and March 2021 at three institutions. The patients were classified into low-dose (TMP < 12.5 mg/kg/d) and conventional-dose (TMP 12.5-20 mg/kg/d) groups. The primary end point was 30-day mortality, and the secondary end points were 180-day mortality, adverse events grade 3 or higher per the Common Terminology Criteria for Adverse Events v5.0, and initial treatment completion rates. Background characteristics were adjusted using the overlap weighting method with propensity scores. RESULTS: Fifty-five patients in the low-dose group and 81 in the conventional-dose group were evaluated. In the overall cohort, the average age was 70.7 years, and the proportion of women was 55.1%. The average dose of TMP-SMX was 8.71 mg/kg/d in the low-dose group and 17.78 mg/kg/d in the conventional-dose group. There was no significant difference in 30-day mortality (6.7% vs 18.4%, respectively; P = .080) or 180-day mortality (14.6% vs 26.1%, respectively; P = .141) after adjusting for patient background characteristics. The incidence of adverse events, especially nausea and hyponatremia, was significantly lower in the low-dose group (29.8% vs 59.0%, respectively; P = .005). The initial treatment completion rates were 43.3% and 29.6% in the low-dose and conventional-dose groups (P = .158), respectively. INTERPRETATION: Survival was similar between the low-dose and conventional-dose TMP-SMX groups, and low-dose TMP-SMX was associated with reduced adverse events in patients with non-HIV PCP.
