The SHA(2)PE score: a new score for lower gastrointestinal bleeding that predicts low-risk of hospital-based intervention

Abstract

Objectives: Lower gastrointestinal bleeding (LGIB) risk scores have mainly focused on identifying high-risk patients. A risk score aimed at predicting which patients will not require hospital-based intervention may reduce unnecessary hospital admissions. The aim of the current study was to develop such a risk score. Material and Methods: A retrospective, population-based study that included patients presenting to the emergency room (ER) with LGIB from 2010 to 2013. Hospital-based intervention was defined as blood transfusion, endoscopic hemostasis, arterial embolization or surgery. The study cohort was split into train (70%) and test (30%) data. Train data were used to produce a multiple logistic regression model and a risk score that was validated on the test data. Results: Overall, 581 patients presented 625 times to the ER, mean age 61 (+/- 22), males 49%. Of train data patients, 72% did not require hospital-based intervention. Independent predictors of low-risk patients (did not require hospital-based intervention) were systolic pressure >= 100mmHg (Odds ratio [OR] 4.9), hemoglobin >12g/dL (OR 103), hemoglobin 10.5-12.0g/dL (OR 19), no antiplatelets (OR 3.7), no anticoagulants (OR 2.2), pulse <= 100 (OR 2.9), and visible bleeding in the ER (OR 3.8). When validating the score on the test data, only 2% were wrongly predicted to be low-risk, the negative predictive value was 96% and the area under curve was 0.83. Conclusions: A new risk score has been developed for LGIB that may help identify low-risk patients in the ER that can be managed in an outpatient setting, thereby lowering unnecessary hospital admissions.