Analysis and validation of a nomogram for predicting the risk of acute respiratory failure for Pneumocystis Pneumonia patients without HIV

<bold>Objective:</bold>We explored the risk factors of acute respiratory failure (ARF) and proposed a model to predict the risk of ARF for non-HIV related Pneumocystis Pneumonia (PCP) patients. <bold>Methods:</bold> In this multi-center, retrospective study in 6 secondary or tertiary academic hospitals in China, 120 PCP patients were screened from the Dryad database for the development of a predictive model. A total of 49 patients from Peking University People's Hospital were collected for external validation. Crucial clinical features of these patients are selected applying univariate and multivariate logistic regression analysis. We established an intuitive nomogram. Calibration curve, decision curve analysis (DCA) and clinical impact curve (CIC) were used to evaluate the model’s performance. <bold>Results:</bold> A cohort of 120 patients formed the training set for the development of this nomogram, with 49 patients constituting its validation set. Univariate and multivariate logistic regression analysis identified three independent risk factors associated with ARF, including fever, dyspnea, and use of antibiotics. We finally established a nomogram with four factors: Age, fever, dyspnea, and use of Antibiotics. The development group had an AUC of 0.829, while the validation group had an AUC of 0.857, indicating good diagnostic accuracy. In the plotting and analysis of the calibration curve, DCA as well as CIC, our model demonstrated good clinical performance. <bold>Conclusion:</bold> We present a trust-worthy nomogram for predicting ARF for non-HIV related PCP patients.