Development of a predictive tool for hemophagocytic lymphohistiocytosis/macrophage activation syndrome risk in adult-onset Still's disease.

BACKGROUND: Hemophagocytic lymphohistiocytosis/macrophage activation syndrome (HLH/MAS) is one of the most common complications of adult-onset Still's disease (AOSD), and a leading cause of organ failure and mortality in these patients. Early diagnosis and management are crucial but can be challenging. Our study aims to develop a tool for predicting the risk of HLH/MAS in AOSD patients. METHODS: The study recruited AOSD patients admitted to Ruijin Hospital from May 2018 to May 2024. Clinical features and laboratory test results were collected within the first 3 days of hospitalization. Predictor variables were selected using LASSO regression, and a logistic regression model was subsequently developed to predict HLH/MAS. Model performance was assessed through calibration curves, receiver operating characteristic (ROC) curve, decision curve analysis (DCA), and bootstrap internal validation. RESULTS: This retrospective cohort study included 190 patients with AOSD. Those who developed HLH/MAS had a higher incidence of fever and splenomegaly at admission. They also exhibited more significant reductions in blood cell counts, elevated inflammatory markers, liver dysfunction, and coagulopathy compared to patients without HLH/MAS. We developed the HLH/MAS Risk Score, which included five key predictors-body temperature, platelet count, D-dimer levels, systemic score, and initial glucocorticoid response-to assess the HLH/MAS risk in hospitalized AOSD patients. This model was visualized and demonstrated robust predictive performance, achieving an AUC of 0.86. Comparative analysis with other MAS-related models confirmed that the HLH/MAS Risk Score had competitive discriminative power. CONCLUSIONS: In this study, we developed the HLH/MAS Risk Score, a predictive tool designed to assess the risk of HLH/MAS in AOSD patients during hospitalization. The model demonstrated strong predictive performance and provides a reliable, easy-to-use tool for the early identification of patients at high risk for HLH/MAS. Further external validation is needed to confirm its clinical applicability across diverse populations.