Predicting intra-abdominal candidiasis in elderly septic patients using machine learning based on lymphocyte subtyping: a prospective cohort study.

Objective: Intra-abdominal candidiasis (IAC) is difficult to predict in elderly septic patients with intra-abdominal infection (IAI). This study aimed to develop and validate a nomogram based on lymphocyte subtyping and clinical factors for the early and rapid prediction of IAC in elderly septic patients. Methods: A prospective cohort study of 284 consecutive elderly patients diagnosed with sepsis and IAI was performed. We assessed the clinical characteristics and parameters of lymphocyte subtyping at the onset of IAI. A machine-learning random forest model was used to select important variables, and multivariate logistic regression was used to analyze the factors influencing IAC. A nomogram model was constructed, and the discrimination, calibration, and clinical effectiveness of the model were verified. Results: According to the results of the random forest and multivariate analyses, gastrointestinal perforation, renal replacement therapy (RRT), T-cell count, CD28+CD8+ T-cell count and CD38+CD8+ T-cell count were independent predictors of IAC. Using the above parameters to establish a nomogram, the area under the curve (AUC) values of the nomogram in the training and testing cohorts were 0.840 (95% CI 0.778-0.902) and 0.783 (95% CI 0.682-0.883), respectively. The AUC in the training cohort was greater than the Conclusion: We established a nomogram based on the T-cell count, CD28+CD8+ T-cell count, CD38+CD8+ T-cell count and clinical risk factors that can help clinical physicians quickly rule out IAC or identify elderly patients at greater risk for IAC at the onset of infection. Clinical Trial Registration: [chictr.org.cn], identifier [ChiCTR2300069020].