Background: Three central venous catheters types are commonly used in pediatric cancer: totally implantable catheters (ICs; eg, mediports, portacaths), tunneled externalized catheters (eg Broviac® or Hickman® catheters [Bard Medical]) and peripherally inserted central catheters. While previous studies have examined risks for catheter-related infections, this is the first large multicenter study to compare catheter-associated infectious morbidity through patient outcomes and hospital utilization.
Methods: A historical cohort analysis was conducted using data from the Pediatric Health Information System reporting 1167 hospitalizations of neutropenic patients with childhood cancer and a central venous catheters. Multivariate analyses controlled for age, ethnicity, gender, malignancy category and transplant status. Outcomes included incidence of serious bacterial infection, mortality, duration of hospitalization, time and use of intensive care unit and antibiotic usage.
Results: Neutropenic cancer patients with totally ICs (n = 429) have a statistically significant approximately 50% decrease in length of hospitalization (P < 0.001), risk for requiring an intensive care unit stay (P = 0.002), documented serious bacterial infection (P = 0.001) and days on antibiotics (P < 0.001) when compared with patients with tunneled externalized catheters (n = 463). Similar differences were found comparing hospitalizations of patients with ICs to those with peripherally inserted central catheters (n = 275). No difference in mortality was observed among catheters groups.
Conclusions: In neutropenic pediatric oncology patients with a central venous catheters, ICs are associated with the least hospital utilization including shortest duration of hospitalization, intensive care unit time and antibiotic therapy when compared with tunneled external catheters and peripherally inserted central catheters. The impact of differences in catheter-associated infectious morbidity on patient outcomes and hospital utilization should be included in clinical decision making.