Hospital volume and the occurrence of bleeding and perforation after colorectal endoscopic submucosal dissection: analysis of a national administrative database in Japan

Abstract

Background: Colorectal endoscopic submucosal dissection has gained popularity as a minimally invasive technique for the treatment of colorectal neoplasms in many countries, including Japan. However, most previous studies of endoscopic submucosal dissection had relatively small sample sizes and only included patients treated at specialized centers. Associations between hospital volume and complication rates after colorectal endoscopic submucosal dissection are still poorly understood.

Objective: Our aim was to clarify the relationships between hospital volume and the occurrence rates of bleeding and perforation after colorectal endoscopic submucosal dissection.

Design: This was a retrospective cohort study. Hospital volume was defined as the number of colorectal endoscopic submucosal dissections performed at each hospital between April 2012 and March 2013 and was categorized into the following quartiles: 1) very low-volume (18 or less patients during the year), 2) low-volume (19-35 patients), 3) high-volume (36-58 patients), and 4) very high-volume (59 or more).

Settings: This study was based on a national inpatient data from the Japanese Diagnosis Procedure Combination database.

Patients: A total of 7567 patients with colorectal endoscopic submucosal dissection were included.

Main outcome measures: Severe postoperative bleeding requiring endoscopic hemostasis or blood transfusion within 1 week after endoscopic submucosal dissection and perforation requiring open surgery were the main outcomes measured.

Results: Severe postoperative bleeding and perforation occurred in 331 (4.4%) and 13 patients (0.2%). Multivariable logistic regression analysis showed that the very high hospital volume group had a significantly lower proportion of severe postoperative bleeding than the very low hospital volume group (OR = 0.48 [95 % CI, 0.27-0.83]; p = 0.009).

Limitations: This study lacked some information on clinicopathologic features including en bloc resection, curative resection, and relapse. Individual endoscopist experience could not be analyzed.

Conclusions: The present study clearly showed a significant association between higher hospital volume and lower occurrence of severe postoperative bleeding.