Safety of sulfamethoxazole-trimethoprim for the treatment of bacterial infection in outpatient settings: A systematic review and meta-analysis with active comparator disproportionality analysis

Abstract

Aims: Sulfamethoxazole-trimethoprim (SMX-TMP) is a widely used antibiotic for treating bacterial infections, but its safety in adult outpatients remains understudied. This systematic review and meta-analysis evaluated the safety profile of SMX-TMP and identified critical research gaps. The pharmacovigilance study aimed to validate and extend findings from meta-analyses to better understand the real-world safety of SMX-TMP.

Methods: We searched MEDLINE and Embase up to 12 August 2024, to identify studies comparing adverse drug events (ADEs) following SMX-TMP vs. other antibiotics in adult outpatients. Meta-analyses were performed where data allowed. A pharmacovigilance study using the Food and Drug Administration Adverse Event Reporting System was conducted to supplement our findings.

Results: Our review, which included 43 studies, found SMX-TMP had a nearly 3-fold higher risk of rash compared to other antibiotics (pooled risk ratio 2.56, 95% confidence interval [1.69, 3.89], I² = 0%, n = 4458 participants, 24 randomized control trials). Pharmacovigilance data confirmed a higher frequencies of skin disorders and other ADEs compared to various comparator drugs. Compared to azithromycin, SMX-TMP was associated with a 5-fold increase in Stevens-Johnson syndrome, a 3-fold increase in toxic epidermal necrolysis, and a 10-fold increase in drug reaction with eosinophilia and systemic symptoms. Additionally, SMX-TMP showed a 10-fold increase in reports of pancytopenia, a 6-fold increase in neutropenia, a 4-fold increase in both thrombocytopenia and aplastic anaemia, a 56-fold increase in hyperkalaemia, and a 10-fold increase in hyponatraemia.

Conclusion: Our meta-analyses and pharmacovigilance study suggested SMX-TMP was associated with increased risk of ADEs compared to other antibiotics including amoxicillin/clavulanate, azithromycin and nitrofurantoin. Further robust research is essential to confirm these safety signals and guide clinical practice.

Keywords: Septra; adverse drug event; cotrimoxazole; cutaneous drug reaction.

FIGURE 1

PRISMA flow diagram of study selection from the databases MEDLINE and EMBASE. (Indexed from 1946 to 12 August 2024).

FIGURE 2

Forest plot, pooled measure of risk of rash and risk of bias assessment in 24 randomized control trials comparing SMX‐TMP to antibiotics with a similar indication. Meta‐analysis included adult outpatients with an indication of urinary tract infection (UTI) treated with SMX‐TMP (n = 2166) or a different antibiotic (n = 2292). Inverse variance was used to determine risk ratio (RR) and a random‐effects model was used to assess pooled treatment effects across studies included in each analysis. Red (−) icons indicate high risk of bias, yellow (?) icons indicate unclear risk of bias and green (+) icons indicate low risk of bias.

FIGURE 3

Forest plots and pooled risk ratios for rash according to comparator group antibiotic class. For all analyses, inverse variance was used to determine risk ratio (RR) and a random‐effects model was used to assess pooled treatment effects across studies. (A) Meta‐analysis included adult outpatients with an indication of urinary tract infection (UTI) treated with sulfamethoxazole (SMX)–trimethoprim (TMP; n = 1120) or an antibiotic from the quinolone family (i.e. ciprofloxacin, ofloxacin, norfloxacin, temafloxacin, ornalidixic acid; n = 1189). (B) Meta‐analysis includes adult outpatients with an indication of UTI treated with SMX‐TMP (n = 234) or an antibiotic from the penicillin family (i.e. ampicillin or amoxicillin; n = 225). (C) Meta‐analysis includes adult outpatients with an indication of UTI treated with SMX‐TMP (n = 136) or a different sulfa–trimethoprim combination (i.e. sulfamethopyrazine‐TMP or sulfadiazine‐TMP; n = 139). (D) Meta‐analysis includes adult outpatients with an indication of UTI treated with SMX‐TMP (n = 278) or an antibiotic from the nitrofuran family (i.e. nitrofurantoin) (n = 274).

FIGURE 4

Reporting odds ratio (ROR) and confidence interval (95%CI) for the skin disorders rash, Stevens–Johnson syndrome (SJS), toxic epidermal necrolysis (TEN), and drug reaction with eosinophilia and systemic symptoms (DRESS) using 3 different comparator drugs. Comparator drugs included (A) azithromycin, (B) amoxicillin/clavulanate and (C) nitrofurantoin. ROR > 1 and lower confidence limit >1 indicates signal detection.

FIGURE 5

Reporting odds ratio (ROR) and confidence interval (95%CI) for the blood disorders aplastic anaemia, pancytopenia, neutropenia and thrombocytopenia using 3 different comparator drugs. Comparator drugs included (A) azithromycin, (B) amoxicillin/clavulanate and (C) nitrofurantoin. ROR > 1 and lower confidence limit >1 indicates signal detection.*indicates number of cases < 3, so ROR was not calculated.

FIGURE 6

Reporting odds ratio (ROR) and confidence interval (95%CI) for the metabolic disorders hyperkalaemia and hyponatraemia using 3 different comparator drugs. Comparator drugs included (A) azithromycin, (B) amoxicillin/clavulanate and (C) nitrofurantoin. ROR > 1 and lower confidence limit >1 indicates signal detection. * Indicates fewer than 3 cases, ROR not calculated.