Development of a clinical risk scoring system for artesunate treatment failure prediction in malaria patients

Severe malaria remains a major cause of morbidity and mortality worldwide. Early identification of patients at high risk of poor response to treatment is essential, yet no simple clinical tool is currently available. This study aimed to develop a risk scoring system to predict failure to artesunate therapy.

This retrospective study included adult patients (aged ≥18 years) who received at least one dose of intravenous artesunate at the National Medical Center in South Korea between 2014 and 2023. Treatment failure (early or late clinical failure) was the response variable, which was defined according to WHO criteria. Candidate predictor variables included demographic, clinical, parasitological, and laboratory parameters. Odds ratios (ORs) and adjusted odds ratios (aORs) were calculated using univariate and multivariable logistic regression analyses, respectively. Final predictors were selected through backward elimination based on the Likelihood Ratio criterion, and a clinical risk scoring system was developed based on the adjusted ORs.

Among 98 patients included in the final analysis, treatment failure occurred in 12 (12.2 %). Multivariable analysis identified female sex, parasitemia >5 %, and impaired consciousness as independent risk factors. Using these variables, a risk-scoring system was constructed, and the predicted probabilities of treatment failure for patients with scores of 0, 1, 2, 3, and 4 points were 5 %, 16 %, 41 %, 72 %, and 90 %, respectively (AUROC = 0.768, 95 % CI: 0.605-0.931).

Parasitemia >5 %, impaired consciousness, and female sex were predictive of artesunate treatment failure as defined by WHO clinical criteria. The developed risk scoring system provides a practical tool for identifying high-risk patients requiring intensified monitoring and alternative treatment strategies. These findings are derived from a South Korean cohort and should be interpreted with caution when extrapolated to endemic populations.