Low mortality and encouraging survival rates led to increasing numbers of liver resections, especially for colorectal cancer metastases, during the last two decades. It has been shown that mortality after liver surgery correlates with hospital volume. We present our recent experience with liver resection. Subgroup analyses were performed comparing the periods before and after the establishment of a new liver surgery programm with a different surgical staff. Methods: From 1998 until 2005 358 liver resections were performed. Perioperative and survival data are documented prospectively since 2002. In 2001 a new liver surgery programm was established with refinements in operative techniques (e.g. parenchyma dissection with CUSA) and perioperative care. The 358 resections were undertaken for colorectal cancer (CRC) metastasis (n=178), hepatocellular carcinoma (n=60), other metastasis (n=39) and various others (n=81). Analyses of the perioperative outcome were compared between the periods before (P1; n=100) and after 9/2001 (P2; n=258). Further analyses were performed in patients with metastases of CRC including data on long-term survival. Results: The mean annual number of liver resection increased from 24 (P1) to 59 (P2). The relative frequency of major resections (classical hemihepatectomy or extended hemihepatectomy) increased from 40% (P1) to 54% (P2). Mortality decreased from 9.1% (P1) to 3.9% (P2; p=0.06), overall complication rates were 56% and 50%, respectively (n.s.). The frequency of patients requiring perioperative blood transfusions decreased from 64% to 26% (p<0.001). In the subgroup of patients with primary liver resection for CRC-metastases (n=156) the frequency of major resections increased from 38% in P1 to 56% in P2. Mortality decreased from 10% in P1 to zero in P2 (p<0.01) whereas morbidity showed only a slight reduction (56% vs. 43%; p=0.1). Since 2001 117 consecutive primary liver resections for CRC-metastases were performed without mortality, and none of these patients required relaparotomy for bleeding. Survival data after primary resection for CRC-metastases were available in 138 patients. Three and five year survival were 65% and 52%, respectively (patients at risk: 38 after three and 11 after five years). Conclusions: Modern surgical and perioperative techniques together with a high case load reduce mortality and major complications in liver surgery, despite more extended resections. In the view of a multidisciplinary approach the very low operative mortality and relative good oncological outcome in CRC liver metastases should encourage medical oncologists to present all eligible patients to the hepatic surgeon.