Abstract
Introduction. Talaromyces marneffei (TM) is a common opportunistic infection among patients with AIDS, characterized by rapid systemic dissemination and a high mortality rate. Early identification of patients at risk of death is critical to improving clinical outcomes.Hypothesis/Gap Statement. Despite the severity of disseminated TM infection (DPSM), few predictive tools exist to assess mortality risk in affected AIDS patients. A clinical prediction model incorporating novel inflammatory markers may help guide timely intervention.Aim. This study aimed to identify independent risk factors for mortality in AIDS patients with DPSM and to develop and validate a nomogram for individualized risk prediction.Methodology. A retrospective study was conducted on 174 AIDS patients with DPSM and complete clinical data admitted to Hangzhou Xixi Hospital between January 2013 and June 2024. A training cohort of 104 patients was used to identify mortality-related risk factors via logistic regression and to construct a predictive nomogram. The remaining 70 patients constituted a validation cohort to evaluate the model using area under the curve (AUC), decision curve analysis (DCA) and calibration curves.Results. The overall mortality rate was 18.97% (33/174). Effusion, bone marrow suppression, systemic inflammation and malnutrition were significantly associated with fatal outcomes (P<0.05). Multivariate logistic regression identified white blood cell count, C-reactive protein-to-prealbumin ratio and procalcitonin-to-albumin ratio as independent risk factors for mortality. The nomogram based on these predictors showed strong discriminative power in both training and validation cohorts (AUC=0.89 and 0.78, respectively). DCA demonstrated the clinical utility and net benefit of the model.Conclusion. This study identified key predictors of mortality in AIDS patients with DPSM and developed a validated nomogram incorporating novel inflammatory markers. The tool offers potential value for individualized risk assessment and clinical decision-making.