PURPOSE: To analyze outcome and predictive factors for patient survival and patency rates of unresectable malignant biliary obstruction treated with percutaneous transhepatic insertion of metal stents. MATERIALS AND METHODS: This is a retroslective analysis of 130 patients treated in one interventional radiological center with data collected from patient records and by telephone interviews. The procedure-related data had been prospectively documented in a computer data base. The Kaplan-Meier analysis was performed for univariate and multivariate comparison of survival and patency rates with the log-rank test used for different tumor types. Predictive factors for survival and 30-day mortality were analyzed by a stepwise logistic regression. RESULTS: Underlying causes of malignant biliary obstructions were cholangiocarcinoma in 50, pancreatic carcinoma in 29, liver metastases in 27, gallbladder carcinoma in 20, and other tumors in 4 patients. The technical success rate was 99 %, the complication rate 27 % and the 30-day mortality 11 %. Primary patency rates (406 days with a median of 207 days) did not differ significantly for different tumor types. The survival rates were significantly (p = 0.03 by log-rank test) better for patients with cholangiocarcinoma than for patients with pancreatic carcinoma and liver metastases. Multiple regression analysis revealed no predictive factor for patient survival and 30-day mortality. CONCLUSION: Percutaneous transhepatic insertion of metal biliary endoprostheses offers a good initial and long-term relief of jaundice caused by malignant biliary obstruction. Although survival rates for patients with cholangiocarcinoma are better than for other causes of malignant biliary obstruction, a clear predictive factor is lacking for patients undergoing palliative biliary stent insertion.