Hospital volume and mortality in liver resections for colorectal metastasis using population-based administrative data.

Abstract

Approximately one-quarter of patients with colorectal carcinoma develop colorectal liver metastases (CRLM). Surgical treatment with curative intent by hepatic resection is the standard medical care. While some studies with small sample sizes have investigated the relationship between hospital procedure volume and in-hospital mortality for this diagnosis, no population-based study has been conducted. The present study was aimed at closing this gap. Based on administrative population-based hospital discharge data (Diagnosis Related Group Statistic), patients diagnosed with CRLM and treated with hepatic resection from 2011 to 2015 were identified. The hospital operation-volume effect on risk-adjusted in-hospital mortality was examined by logistic regression models. During the study period, 5900 patients with CRLM were treated with hepatic resection, of whom 189 (3.2%) died before hospital discharge. Hospitals of different operation-volume quartiles did not differ in terms of mortality rates. Sensitivity analysis investigating the volume-mortality relationship separately for every resection procedure showed no clear result. Procedure frequencies vary among hospitals of different volume quartiles, with low-volume hospitals performing systematically more low-risk procedures (in terms of reduced mortality rate), than high-volume hospitals. Based on almost complete German hospital discharge data, the results did not confirm unconditional volume-outcome relationship for CRLM patients.