INFECTIOUS MORBIDITY ASSOCIATED WITH LONG-TERM USE OF VENOUS ACCESS DEVICES IN PATIENTS WITH CANCER

Objective: To evaluate infectious morbidity associated with long-term use of venous access devices. Design: Prospective, observational study. Setting: Comprehensive cancer center at a university hospital. Participants: 1431 consecutive patients with cancer requiring 1630 venous access devices for long-term use inserted between 1 June 1987 and 31 May 1989. Measurements: Quantitative microbiologic tests to identify device-related bacteremia and fungemia, catheter tunnel infection, pocket infection in implantable port devices, and site infections; number of days the device remained in situ and time until infectious morbidity; vessel or device thrombosis and device breakage. Results: At least one device-related infection occurred with 341 of 788 (43% [95% CI, 39% to 47%]) catheters compared with 57 of 680 (8% [CI, 6% to 10%]) completely implanted ports (P less-than-or-equal-to 0.001). Device-related bacteremia or fungemia is the predominant infection occurring with catheters, whereas ports have a more equal distribution of pocket, site, and device-related bacteremia. The predominant organisms isolated in catheter-related bacteremia were gram-negative bacilli (55%) compared with gram-positive cocci (65.5%) in port-related bacteremia. The number of infections per 1000 device days was 2.77 (95% CI, 2.48 to 3.06) for catheters compared with 0.21 (CI, 0.16 to 0.27) for ports (P less-than-or-equal-to 0.001). Based on a parametric model of time to first infection, devices lasted longer in patients with solid tumors than in those with hematopoietic tumors. Ports lasted longer than catheters across all patient groups. Conclusions: The incidence of infections per device-day was 12 times greater with catheters than with ports. Patients with solid tumors were the least likely to have device-related infectious morbidity compared with those with hematologic cancers. The reasons for the difference in infectious complications is uncertain but may be attributable to type of disease, intensity of therapy, frequency with which devices are accessed, or duration of neutropenia.