External validation of the Malaria Scoring System in a non-endemic emergency department.

The incidence of malaria is increasing even in non-endemic regions, posing a diagnostic challenge. Limited awareness of the disease may result in delayed diagnosis, potentially leading to severe complications. This study aims to externally validate the Malaria Scoring System (MSS), a diagnostic tool originally developed and validated in Sweden, to identify patients at higher risk of plasmodial infection. The MSS incorporates key clinical and laboratory parameters, including recent travel to Sub-Saharan Africa, hematological abnormalities (thrombocytopenia, anemia, lymphopenia, neutropenia), and high fever. A score of ≥ 3 is associated with an increased probability of malaria. A retrospective, single-center study was conducted at the Emergency Department of Santa Croce e Carle Hospital (Cuneo, Italy), including all patients tested for malaria between January 2013 and July 2024. The diagnostic performance of the MSS was assessed in terms of discrimination using receiver operating characteristic (ROC) curve analysis and calibration via the Hosmer-Lemeshow test and calibration belt analysis. A total of 365 patients were included, of whom 114 (31.2%) were diagnosed with malaria. Fever was more frequent among malaria patients (96.5% vs. 84.1%, p < 0.001). Laboratory findings revealed significant differences, with a higher prevalence of thrombocytopenia (78.1% vs. 8.4%, p < 0.001), lymphopenia (71.9% vs. 32.7%, p < 0.001), and anemia (53.5% vs. 35.1%, p < 0.001) in malaria patients. The MSS was significantly higher in malaria cases (median score: 4 [3; 5]) compared to non-malaria cases (median score: 1 [1; 2], p < 0.001). The MSS demonstrated excellent discriminatory ability, with an area under the ROC curve of 0.902 (95% CI 0.869-0.935). The calibration of the MSS was evaluated using the Hosmer-Lemeshow test (X = 1.68, p = 0.195). This external validation study confirms that the MSS exhibits strong discriminatory performance and good calibration for predicting malaria infection, supporting its potential utility in clinical practice.