Abstract
OBJECTIVE: To comprehensively analyze the relevant factors influencing the treatment efficacy of Nocardia farcinica pneumonia, construct and validate a prediction model to provide a scientific basis for clinical treatment, and realize visual prediction using a nomogram. METHODS: The clinical data of 150 patients with Nocardia farcinica pneumonia collected from January 2020 to December 2024 were selected and divided into a training set (n = 105) and a validation set (n = 45) at a ratio of 7:3. The data covered patients' basic information, laboratory examination indicators, imaging features, and treatment regimens. Risk factors were screened by univariate and multivariate logistic regression in the training set to construct a nomogram model. The receiver operating characteristic curve (ROC) and calibration curve were plotted to evaluate the model's efficacy and were validated in the validation set. Decision curve analysis (DCA) was used to evaluate the clinical value. RESULTS: In the training set, 26 cases (24.32%) exhibited poor treatment response, while 11 cases (25.25%) were identified in the validation set. Multivariate analysis identified serum albumin levels, empyema, cavitary lesions, and antibiotic regimens (sulfonamides/cephalosporins/carbapenems) as independent factors influencing the therapeutic efficacy of Nocardia farcinica pneumonia. In the training and validation sets, the model achieved C-index values of 0.849 and 0.831, with areas under the ROC curve (AUC) of 0.849 (95% CI: 0.764-0.935) and 0.831 (95% CI: 0.580-1.000), respectively. The sensitivity and specificity were 0.772 and 0.895 in the training set, and 0.773 and 0.857 in the validation set, indicating predictive capability for treatment outcomes. The nomogram model exhibited excellent predictive accuracy upon calibration curve analysis. Decision curve analysis (DCA) further confirmed its high clinical utility. CONCLUSION: Beyond confirming the role of host immunity and inflammation, this study develops and validates the first nomogram that integrates baseline albumin, empyema, cavitation, and antibiotic choice to quantitatively predict individual treatment failure risk in Nocardia farcinica pneumonia. This tool provides an immediately applicable visual guide for early risk stratification and personalized therapy selection, addressing a significant gap in the management of this complex infection.