Background: Recent meta-analyses of published controlled studies concluded that adult patients with cancer randomly assigned to receive parenteral nutrition had higher rates of infectious complications than control subjects.
Methods: The infection risk associated with parenteral nutrition was assessed in 310 pediatric patients with cancer. These patients had central venous access devices (CVAD), Hickman/Broviac (H/B) catheters, or implantable subcutaneous ports in place for the delivery of chemotherapy and supportive care.
Results: The median duration of CVAD placement was 363 days; a total of 450 patient years (i.e., the sum of the total years of catheters experienced from all patients studied) were examined. Overall, the infection rate was 0.06 infections/100 days. During the period of parenteral nutrition administration, the rate increased to 0.5 infections/100 days. Among patients who received parenteral nutrition, there were no significant differences in any clinical parameter between the patients who developed an infection and those who did not. When evaluating the entire study population, infection was more likely to occur in patients who had acute nonlymphocytic leukemia (P < 0.01) or H/B catheters (P < 0.01), or who received parenteral nutrition (P < 0.02); there was no relationship between infection and catheter duration, days hospitalized, or days neutropenic (absolute neutrophil count < 0.5 x 10(9)/l). Only CVAD type and parenteral nutrition retained significance in a multivariate Cox proportional hazards model. After adjustment for diagnosis and CVAD type, the risk of infection was 2.4-fold greater in patients given parenteral nutrition (95% confidence interval 1.5 to 3.9; P < 0.001).
Conclusion: These data confirm that administration of parenteral nutrition is associated with an increased risk of infection in children who have CVAD in place for cancer therapy.