Many patients with hepatic tumors cannot benefit from resection owing to the difficult anatomic sites of their lesions. Some of these patients might be eligible for ex vivo liver resection and autotransplantation. This procedure consists of complete hepatectomy, extracorporeal liver resection, and autotransplantation of the remnant liver.

Four databases were searched for studies reporting cases of ex vivo liver resection and autotransplantation. Outcomes of this procedure were evaluated by meta-analysis of proportions with random effects model and individual participant data analysis.

Fifty-three studies were assessed. Meta-analysis revealed an R0 resection rate of 93.4% (95% confidence interval: 81.0–97.9%, I² = 0%), a frequency of major surgical complications of 24.5% (95% confidence interval, 16.9–34.3%, I² = 26%), a 30-day mortality of 9.5% (95% confidence interval: 5.9–14.9%, I² = 0%), and a 1-year survival of 78.4% (95% confidence interval: 62.2–88.8%, I² = 64%). We were able to obtain the individual participant data in 244 patients; R0 resection was achieved in 98.6%, with no obvious difference between analyzed subgroups. The 30-day mortality and 1-year survivals were 7.9% and 82.1%, respectively. For groups with malignant and nonmalignant tumors, the 30-day mortalities were 11.3% vs. 6.3% (P = .181), and 1-year survivals were 65.0% vs. 89.7% (P < .001). When comparing those with malignant versus those with nonmalignant lesions, major surgical complications occurred in 50.0% vs. 21.0%; P < .001). Regression analysis revealed that outcomes of patients with benign tumors were better compared with those with malignant tumors (1-year survival, odds ratio: 4.629; 95% confidence interval: 2.181–10.097, P < .001).

Ex vivo liver resection and autotransplantation facilitates radical treatment in selected patients with conventionally unresectable hepatic tumors and normal liver function. The outcomes of treatment of malignant lesions appear to be less satisfactory.