RESUMO
GOALS OF WORK: Prospective clinical study to evaluate patients suffering from solid tumor using a totally implanted venous access device (TIVAD) to determine: (1) if there is a relationship between cutaneous contamination at port insertion site and catheter-related bloodstream infection (CRBI); (2) development modalities of CRBI; (3) if there is a relationship between chemotherapy administration modalities by push/ bolus versus continuous infusion and CRBI. PATIENTS AND METHODS: We studied 41 consecutive patients who needed a TIVAD positioned for chemotherapy administration by bolus/ push or continuous infusion. In every patient, we performed blood cultures from blood samples from port catheters and cutaneous cultures from cutaneous tampons of the skin surrounding the implant area on the first (T0) and eight day (T1) postoperatively, after 1 month (T2), and after 3 months (T3) from insertion. MAIN RESULTS: The study was completed on 40 patients; in one case, the port was removed at T2 for septic complications. We obtained four positive blood cultures (two, 5%), two in the same patient, all caused by staphylococcus. Positive cutaneous tampons were 21 (13%) in 11 patients (27%); the four CRBI occurred in this group of patients with none in the remaining 30 patients (73%) for a total number of 120 tampons (p<0.01). In two cases, the same germ was isolated from both the skin and blood. None of the patients presented a local infection of the subcutaneous pocket. Positive cutaneous cultures decrease over time: T0-T2; 24-5%; T1-T3, 20-5% (p<0.04). There were no differences in CRBI incidence and positive cutaneous tampons between the two chemotherapy administration modalities. CONCLUSIONS: Cutaneous microbial flora has a primary role in CRBI development within TIVADs; there is a relationship between cutaneous colonization and CRBI; colonization reaches its maximum during the first days after catheterization in which the use of the system is at high risk; colonization occurs both via extraluminal and endoluminal routes; there is no difference in CRBI incidence between bolus and continuous infusion administration.
