Clinical and treatment-related predictors of failure of pentavalent antimonial therapy for cutaneous leishmaniasis: a systematic review.
Abstract
Background
Pentavalent antimonials remain widely used for cutaneous leishmaniasis (CL), but treatment failure prolongs disease and increases patient burden. We synthesized routinely available clinical and treatment-related predictors of antimonial failure.
Methods
PubMed/MEDLINE and Europe PMC were searched through 14 August 2026 and Scopus through 30 August 2026, supplemented by citation and targeted website searches. Cohort, case-control, and prediction-model studies of antimony-treated CL were eligible. QUIPS assessed risk of bias. Heterogeneous outcomes, regimens, and overlapping cohorts precluded meta-analysis; results were synthesized following SWiM. The review was not prospectively registered.
Results
Of 1,747 records identified, 1,249 were screened and 36 reports included. Prior antimonial treatment and incomplete treatment exposure or adherence most consistently signalled poorer outcomes. Greater lesion burden often predicted failure, but thresholds varied. Lesion duration was context-dependent: prolonged disease and very early non-ulcerated lesions were associated with failure in different settings. Findings for age, sex, comorbidities, geography, and species were inconsistent or context-specific. No model had convincing external validation; a recent model yielded an internally validated C-statistic of 0.608.
Limitations
All reports had moderate or high risk of bias. Inconsistent outcome definitions and regimens, suspected cohort overlap, and lack of Embase, Web of Science, and CINAHL searches limit confidence and transportability.
Conclusions
The evidence supports documenting treatment history, measuring lesion burden, addressing adherence barriers, and arranging earlier review for higher-risk patients, rather than using an unvalidated score or automatically replacing antimony. Prospective external validation with harmonized failure definitions is needed.