Retrospective assessment of risk factors to predict tunneled hemodialysis catheter outcome.

ABSTRACT

PURPOSE: To identify factors that may influence the function, outcome, and complications associated with tunneled hemodialysis catheters. MATERIALS AND METHODS: Radiology reports and hemodialysis, medical, and Clinical Information System (computerized patient medical record system) records were retrospectively reviewed in 221 consecutive patients who underwent tunneled hemodialysis catheter placement by interventional radiologists between January 11, 1996 and January 13, 2000 at Dartmouth-Hitchcock Medical Center. Various patient characteristics (diabetes, smoking, hypertension, age, sex, atherosclerotic cardiovascular disease, history of cardiac catheterization, coumadin use, functional status, and obesity) were assessed for their relationship to the outcome of hemodialysis catheters. Catheter outcome was examined by calculating infection rate, thrombosis rate, fibrin formation rate, mechanical malfunction rate, and total complication rate. With these patient characteristics and outcome variables, multiple regression analysis was performed with STATA (College Station, TX) statistical analysis software. RESULTS: Of the 221 patients reviewed, 39 patients were lost to follow-up. Among the remaining 182 patients, 427 catheters were placed for a total number of 36994 catheter-days. For overall complication rate, multiple regression analysis revealed a statistically significant positive correlation only for hypertension (P =.032). Total complication rate was 0.76 events per 100 catheter-days (95% CI 0.53-1.00) for patients with a documented history of hypertension and 0.27 events per 100 catheter-days (95% CI 0.08-0.45) for patients without (P =.024, paired student t test). For patients with diabetes versus patients without, the infection rates were 0.34 episodes per 100 catheter-days (95% CI 0.15-0.53) and 0.12 episodes per 100 catheter-days (95% CI 0.06-0.18), respectively, (P =.011, paired student t test). Thrombosis rate for patients on coumadin was 0.13 events per 100 catheter-days (95% CI -0.14-0.40), while thrombosis rate for patients not taking coumadin was 0.03 events per 100 catheter-days (95% CI 0-0.05) (P =.036, paired student t test). CONCLUSION: Hypertension is a risk factor for poor outcome of tunneled hemodialysis catheters as measured by total complication rate requiring catheter removal or exchange. In this retrospective study, no other specific risk factors predicted an increased need for removal or exchange of tunneled hemodialysis catheters.