Transcutaneously tunneled central venous lines in cancer patients: An analysis of device-related morbidity factors based on prospective data collection

Background: Long-term transcutaneous tunneled central venous catheters are frequently placed in cancer patients, accounting for significant costs and morbidity. Factors influencing outcome, though, are poorly studied. Methods: Between June 1991 and June 1993, 923 central venous tunneled cathe ters were placed in 791 patients at Memorial Sloan-Kettering Cancer Center. Placement-, device-, and patient-related parameters were charted prospecti vely (median follow-up: 120 days) and correlated to device-specific outcome events. Results: Median patient age was 28.5 years (range: 0.025 - 84.5). Disease d istribution included hematologic malignancies (64.7%), solid tumors (30.4%) , and others (4.9%). Primary indications for line access included chemother apy (72.8%), bone marrow transplantation (18.7%), total parenteral nutritio n (6.4%), and drug administration (2.1%). There were 11 insertion complicat ions (1.2%), including insertion failure (n = 6), hemorrhage (n = 4), and m alposition (n = 1). Subsequent to placement, a proven or suspected device-s pecific complication occurred in 540 lines (58.5%). Per 10,000 catheter day s, there were 17.6 infection episodes, 8.1 thrombotic complications, 6.9 in stances of catheter breakage, 3.5 accidental or inadvertent cases of displa cement, and 0.6 device leaks. Reasons for Line removal or other termination of follow-up were patient's death (32.1%), treatment end (28%), infection (19.6%), suspected infection (6.3%), displacement (6.8%), thrombosis (3.1%) , leak (1%), and others (3.1%). Median device-specific duration was 365 day s, compared with a median complication-free device-specific duration of 167 days (P < 0.0001), reflecting a highly significant device salvage rate aft er complications. Catheter tip position emerged as the dominant independent prognostic factor for reduced device-specific duration or complication-fre e device-specific duration. Conclusions: Transcutaneous tunneled central venous lines can be placed saf ely, with a considerable incidence of subsequent device-specific complicati ons, but a high salvage rate. Factors determining outcome are related to de vice placement, as well as the patient's disease status. In this study, pat ients alive 90 days after catheter placement had a 37% chance for a device complication, with a 20% chance for device loss. Future analyses of interme diate-term intravenous access should employ the measurement of device-speci fic outcome as a reference parameter to assess clinical results.