[Hospital volume effects in surgical treatment of gastric cancer : Results of a prospective multicenter observational study]

Abstract

Background: The impact of hospital and surgeon volume on the treatment outcome based on data obtained from cohort and register studies has been controversially discussed in the international literature. The results of large-scale prospective observational studies within the framework of clinical healthcare research may lead to relevant recommendations in this ongoing discussion.

Material and methods: Within the framework of the prospective multicenter German Gastric Cancer Study 2 (QCGC 2), from 1 January 2007 to 31 December 2009 a total of 2897 patients with the histological diagnosis of gastric cancer from 140 surgical departments were registered and analyzed. The departments were subdivided according to the number of cases into 4 volume groups: I) <5, II) 5-10, III) 11-20 and IV) >20 patients with surgical interventions per year.

Results: Overall 1163 patients (65.6 %) underwent surgical interventions in the departments of groups III and IV. Of the patients 521 (18 %) were scheduled for neoadjuvant treatment but with no significant differences among the various volume groups. In the departments of volume groups I and II subtotal gastric resection was performed significantly more often. Transthoracic extended surgical interventions in cases of a proximal tumor site were significantly more frequent in departments from volume group IV (p <0.001). The proportion of intraoperative fresh frozen sections correlated with the case volume: group I 23.2 % vs. group IV 61.2 %. Overall hospital mortality was 6.1 % and slightly higher in volume group I with 7.8 %. The median survival time and the 5‑year survival rate showed no significant differences between the various volume groups independent of tumor stages. There was a tendency towards a longer median survival time in volume group IV only for proximal tumor sites, i.e. adenocarcinoma of the esophagogastric junction (AEG). Using Cox regression analysis hospital volume did not have an independent impact on long-term survival.

Conclusion: Hospital volume effects could only be detected for the treatment of AEG. To improve oncological long-term outcome, centralization of treatment of proximal gastric cancer appears to be recommendable.

Keywords: Gastric cancer; Hospital volume; Prospective multicenter observational study; Surgery.