Multivariate predictive model for predicting in-hospital mortality in HIV-associated talaromycosis: a multicenter...

Area of Science:

• Infectious Diseases
• Mycology
• Epidemiology
• Biostatistics

Background:

• Talaromycosis is a severe invasive fungal infection disproportionately affecting immunocompromised individuals, particularly those with HIV/AIDS.
• HIV-associated talaromycosis (HTM) presents a significant challenge due to high incidence and mortality rates.
• Accurate prediction of mortality risk is crucial for timely and effective clinical management of HTM patients.

Purpose of the Study:

• To develop and validate a novel nomogram model for predicting in-hospital mortality risk in patients with HIV-associated talaromycosis (HTM).
• To identify key clinical indicators and symptoms associated with mortality in HTM patients.

Main Methods:

• Retrospective analysis of 431 HTM patients across three research centers (January 2013 - December 2023).
• Patients were randomly assigned to training (70%) and validation (30%) sets.
• Logistic regression (univariate, lasso, stepwise) was used to identify predictive factors; model performance was assessed using ROC curves, calibration curves, and decision curve analysis (DCA).

Main Results:

• A total of 55/431 (12.76%) patients died during hospitalization.
• Five factors—breathlessness, elevated TB, APRI, CRP, and decreased Hb—were identified as significant predictors of HTM mortality.
• The nomogram demonstrated strong predictive performance with an AUC of 0.83 (training set) and 0.81 (validation set); calibration and DCA confirmed its clinical utility.

Conclusions:

• The developed nomogram effectively predicts in-hospital mortality in HTM patients by integrating readily available clinical data.
• This predictive tool can significantly assist clinicians in tailoring individual management strategies for HTM.
• The nomogram offers a valuable approach to improving patient outcomes and resource allocation in HTM care.