BackgroundThe high complication rates of surgically implanted port catheter systems (SIPCS) represents a major drawback in the treatment of isolated liver neoplasms by hepatic arterial infusion (HAI) of chemotherapy. Interventionally implanted port catheter systems (IIPCS) have evolved into a promising alternative that enable initiation of HAI without laparatomy, but prospective data on this approach are still sparse. Aim of this study was to evaluate the most important technical endpoints associated with the use of IIPCS for the delivery of 5-fluorouracil-based HAI in patients with colorectal liver metastases in a phase 2-study, and to perform a non-randomised comparison with a historical group of patients in which HAI was administered via SIPCS.Methods41 patients with isolated liver metastases of colorectal cancer were enrolled into a phase II-study and provided with IIPCS between 2001 and 2004 (group A). The primary objective of the trial was defined as evaluation of device-related complications and port duration. Results were compared with those observed in a pre-defined historical collective of 40 patients treated with HAI via SIPCS at our institution between 1996 and 2000 (group B).ResultsBaseline characteristics were balanced between both groups, except for higher proportions of previous palliative pre-treatment and elevated serum alkaline phosphatase in patients of group A. Implantation of port catheters was successful in all patients of group A, whereas two primary failures were observed in group B. The frequency of device-related complications was similar between both groups, but the secondary failure rate was significantly higher with the use of surgical approach (17% vs. 50%, p < 0.01). Mean port duration was significantly longer in the interventional group (19 vs. 14 months, p = 0.01), with 77 vs. 50% of devices functioning at 12 months (p < 0.01). No unexpected complications were observed in both groups.ConclusionHAI via interventionally implanted port catheters can be safely provided to a collective of patients with colorectal liver metastases, including a relevant proportion of preatreated individuals. It appears to offer technical advantages over the surgical approach.