The predictive value of platelet-to-lymphocyte ratio and lymphocyte count in recurrent small intestinal hemorrhage: a retrospective cohort study

Abstract Background The recurrence of small intestinal hemorrhage (SIH) poses a significant clinical challenge owing to the absence of reliable and accessible predictive biomarkers. This study aimed to identify a crucial biomarker for predicting SIH recurrence and establish a benchmark to assist clinical decision-making. Methods The training cohort comprised 131 newly diagnosed SIH patients who underwent endoscopy at the First Affiliated Hospital of Nanchang University. An independent validation cohort included 96 patients from the Second Affiliated Hospital of Nanchang University. Major variables (MVs) were identified using Least Absolute Shrinkage and Selection Operator (LASSO) regression. A nomogram was developed by incorporating the coefficients of MVs from logistic regression, and its discrimination, calibration, and clinical utility were validated in both cohorts. Results During one year of follow-up, the bleeding recurrence rates were 41.22% in the training cohort and 30.21% in the validation cohort. Principal causes of bleeding included inflammatory bowel disease, vascular abnormalities, small intestinal diverticula, and small intestinal tumors, among others. The nomogram, based primarily on the Platelet-to-Lymphocyte Ratio (PLR) and Lymphocyte Count (LYM), achieved area under the curve (AUC) values of 0.923 and 0.784 in the training and validation cohorts, respectively. A nomogram score of 37.4 points was preliminarily identified as a potential threshold to distinguish the risk of bleeding recurrence within one year after endoscopic treatment in patients with small intestinal hemorrhage SIH. Conclusion The nomogram developed in this study provides an effective tool for assessing the risk of recurrence after endoscopic treatment for small intestinal hemorrhage, demonstrating favorable predictive performance and offering a preliminary basis for clinical risk stratification.