Alberto Contreras, MD
El Paso, Texas
Alberto Contreras, MD, Alejandro Robles, MD, Antonio Mendoza-Ladd, MD
Texas Tech University Health Sciences Center, Paul L. Foster School of Medicine, El Paso, TX
Introduction: Non-variceal acute upper gastrointestinal bleeding (NVAUGIB) is one of the most common causes of hospitalization worldwide. The Glasgow-Blatchford Score (GBS) was developed to identify subjects who would benefit from hemostatic endoscopic intervention (HEI), in this and other types of GIB. According to the literature, a GBS score < 2 identifies with up to 99% accuracy subjects who do not need emergent HEI. Experience at our center suggests that the average GBS of our patients with NVAUGIB is consistently >2 and yet our rate of HEI is low. Therefore, the primary aim of this study was to accurately determine the rate of significant HEI in subjects with NVAUGIB with GBS >2. The secondary aim was to determine the cost of esophagogastroduodenoscopy (EGD) in patients in whom no significant HEI was performed.
Methods: Medical records of subjects admitted for NVAUGIB who underwent EGD from January 2015 – December 2017 were reviewed. NVAUGIB was defined as the presence of any of the following: hematemesis, hematochezia, coffee ground emesis, + FOBT, melena, drop in Hgb >2 g/dl and absence of varices on EGD. The GBS was calculated on the day of EGD on all subjects. Data obtained was compared using a Kolmogorov-Smirnov Test of Normality first to ensure normal distribution, then an independent T-test was used to evaluate for significance. EGD cost was obtained from the hospital administration.
Results: A total of 586 subjects underwent EGD for the evaluation of NVAUGIB, and their demographic characteristics are listed in Table 1. The mean GBS for subjects with ≤2 and >2 were 1 and 8.8 respectively. Of those with a GBS >2 (503/586), only 16.9% (85/503) underwent a HEI. On the other hand, only 1% (1/83) of those with GBS ≤2 (83/586), underwent HEI. Overall, subjects who received HEI (86) had higher GBS compared to those who didn’t (9.29, SD 3.51) vs. (7.42, SD 4.06) (p = < 0.05). A total of 418 EGDs without HEI were performed in subjects with GBS >2. The cost of a diagnostic EGD at our institution is approximately $2,866.00. Therefore, the cost of these procedures was approximately $1,200,000.00.
Discussion: In our population, a GBS cutoff of >2 did not identify patients requiring HEI accurately as previously reported in the literature. The cost of performing non-therapeutic endoscopic evaluation in these subjects is alarming. Further research regarding the generalizability of the GBS is needed at this time.
Citation: Alberto Contreras, MD, Alejandro Robles, MD, Antonio Mendoza-Ladd, MD. P0409 - THE GLASGOW-BLATCHFORD SCORE: PITFALLS AND COST OF GENERALIZATION. Program No. P0409. ACG 2019 Annual Scientific Meeting Abstracts. San Antonio, Texas: American College of Gastroenterology.