Selection for surgery and survival of synchronous colorectal liver metastases; a nationwide study

Resection of colorectal liver metastases (CRLM) is associated with improved survival but we currently have limited population-based data on selection for surgery.

Patients in the Swedish Colorectal Cancer Register reported with liver metastases at diagnosis in 2007-2011 were identified. Clinical characteristics including American Society of Anesthesiologists classification, type of hospital and health care region were retrieved. Linkage to the National Patient Register and Statistics Sweden provided information on liver resection and socioeconomic variables.

Synchronous CRLM was found in 4243/27,990 (15.2%) patients, of whom 1094 (25.8%) also had concurrent lung metastases. Of 3149 patients with liver-only metastases, 556 (17.8%) were subjected to liver resection. The resection rate varied by subsite; right-sided 11.7%, left-sided 19.7% and rectal cancer 22.7% (p = 0.001). It varied by type of hospital 14.1-23.6%, by region 11.5-22.7%, and was 19.8% in men and 14.9% in women (all p < 0.001). The adjusted odds were 0.74 (0.59-0.93) for females, 0.58 (0.46-0.74) for general district and 0.50 (0.37-0.68) for district hospital patients, and there were large regional differences. Patients >75 years were very unlikely to receive liver surgery 0.22 (0.15-0.32). In patients subjected to liver surgery, median survival was 57 months, 5-year survival rate was 45.4%, and those with left-sided colon cancer had the best outcome (48.8%; p = 0.02). Five-year hazard ratio for patients not subjected to liver surgery was 4.3 (3.7-5.0).

Nationwide outcome after resection of synchronous CRLM was impressing but ambiguous selection mechanisms and inaccessibility need to be resolved. The implications of subsite deserve further attention.