Impact of Trainee Involvement in Management and Outcomes of Upper Gastrointestinal Bleeding

Abstract

Purpose: Upper gastrointestinal bleeding (UGIB) is common, with an annual incidence of 1 per 1,000 population. Previous studies on trainee involvement in colonoscopies have shown increased adenoma detection rates. The aim of this study was to examine whether fellow participation during management of UGIB affects the outcome when compared to management by an experienced gastroenterologist alone. Methods: This was a retrospective study of 278 cases of UGIB at a tertiary care referral center between August 2010 and July 2011. Demographics, management and outcome details were collected by chart review. Comparison was then made between cases managed by fellows with supervising staff gastroenterologists and the cases managed by a staff gastroenterologist without fellow participation. Results: Fellows were involved in management of 63% of cases, mostly in their 1st year of training. Comparison of demographics, presentation and comorbidities is shown in Table 1. Patients managed by the trainees were more likely to have had hematemesis and low BP but a higher mean hemoglobin. An upper endoscopy was performed in 87% of the cases in the trainee group and 92% of cases in the staff gastroenterologist group. There was no difference in the time to endoscopy from admission, adverse outcomes (mortality and rebleeding), length of hospital stay and cost of hospitalization between the 2 groups. Even though there was no statistically significant difference in the proportions with bleeding stigmata (p=0.3) on upper endoscopy, more patients in the trainee group had endoscopic hemostasis (p=0.02). Increased red blood cell transfusions were noted in the cases managed by the staff gastroenterologist and increased FFP transfusions were noted in trainee-managed cases. These results are summarized in Table 2.Table 1: Comparison of demographics, presentation and comorbidities in cases managed with trainees and without traineesTable 2: Comparison of management and outcomes in cases managed with trainees and without traineesConclusion: Fellow involvement was associated with increased endoscopic hemostasis. This difference is possibly due to endoscopic training involved in the management of those cases; however, no difference was noted in the overall outcome, length of hospital stay or cost of admission.