Stroke-associated pneumonia risk prediction: Development and internal validation of a preliminary nomogram

Stroke-associated pneumonia (SAP) is a common complication of acute stroke, highlighting the need for early risk identification. This study aimed to develop and internally validate a preliminary SAP nomogram using routinely available clinical and hematological parameters. This single-center retrospective study included 650 patients randomly divided into training (n = 454) and internal validation (n = 196) cohorts. Baseline blood samples were obtained within 24 hours of admission and before SAP onset; the outcome was SAP within 7 days of stroke onset. Least absolute shrinkage and selection operator (LASSO) regression followed by multivariable logistic regression was used to select predictors and construct the nomogram. SAP occurred in 190 patients (29.2%). Older age, atrial fibrillation, higher white blood cell count, lower lymphocyte percentage, and lower hemoglobin were independently associated with SAP. The area under the receiver operating characteristic curve (AUC) was 0.772 in the training cohort and 0.738 in the internal validation cohort, with respective 95% confidence intervals of 0.723-0.820 and 0.662-0.815. The optimism-corrected training AUC was 0.762. The validation Brier score was 0.181, and decision curve analysis suggested potential clinical net benefit. This preliminary nomogram may support early SAP risk stratification. However, the single-center retrospective design and omission of key neurological predictors, including stroke severity and dysphagia, necessitate prospective multicenter validation before clinical implementation.