A new model (Alfalfa-Warfarin-GIB) for predicting the risk of major gastrointestinal bleeding in warfarin patients.

Abstract

This study aimed to analyze the factors influencing warfarin-related major gastrointestinal bleeding (GIB) and to develop a score that would provide a reference for assessing the risk of major GIB associated with warfarin treatment. This was a retrospective analysis of clinical and follow-up data from warfarin-treated patients. Scores were analyzed using logistic regression. The area under the subject working characteristic curve (AUC), sensitivity, specificity, and Hosmer-Lemeshow test were used to evaluate the scoring performance. A total of 1591 patients who met the requirements for warfarin use were included in this study, and 46 developed major GIB. After univariate analysis as well as multivariate logistic regression analysis, nine factors were found to be associated with increased risk of major GIB, namely age ≥ 65years, history of peptic ulcer, history of major bleeding, abnormal liver function, abnormal renal function, cancer, anemia, labile international normalized ratio, and combination of antiplatelet agents/non-steroidal anti-inflammatory drugs. The Alfalfa-Warfarin-GIB score was constructed using these nine factors. The AUC and Bootstrap method-corrected AUC of the Alfalfa-Warfarin-GIB score were 0.916 (95% CI: 0.862-0.970, P < 0.001) and 0.919 (95% CI: 0.860-0.967, P < 0.001), respectively, which were higher than those of the HAS-BLED score (AUC = 0.868, 95% CI: 0.812-0.924, P < 0.001). Based on nine risk factors, the Alfalfa-Warfarin-GIB score was constructed to predict the risk of warfarin-related major GIB. The newly developed Alfalfa-Warfarin-GIB score has a better predictive value than the HAS-BLED score and may be an effective tool to help reduce the occurrence of major GIB in patients on warfarin.