RESUMO
BACKGROUND: The incidence of peripherally inserted central catheter (PICC)-related adverse events has been uncertain in the setting of acute myeloid leukemia (AML) compared with the incidence of centrally inserted central catheter (CICC) adverse events. PATIENTS AND METHODS: We conducted a monocentric, randomized trial of patients with previously untreated AML. Of the 93 patients, 46 had received a PICC and 47 had received a CICC as frontline intravascular device. Thereafter, all patients underwent intensive chemotherapy for hematologic remission induction. The primary endpoint was catheter-related (CR)-bloodstream infection (BSI) and venous thrombosis (VT) rate. The secondary endpoints catheter malfunction, catheter removal, and patient overall survival. RESULTS: The CR-BSI and CR-VT rate in the PICC and CICC groups was 13% and 49%, respectively, with a difference of 36 percentage points (relative risk for CR-BSI or CR-VT, 0.266; P = .0003). The CR-BSI incidence was 1.4 and 7.8 per 1000 catheters daily in the PICC and CICC groups, respectively. Among the CR thromboses, the symptomatic VT rate was 2.1% in the PICC group and 10.6% in the CICC group. In the CICC group, 16 of the 47 patients (34%) had the catheter removed for BSI (n = 5), septic thrombophlebitis (n = 4), VT (n = 2), or malfunction (n = 5) a median of 7 days after insertion. In the PICC group, only 6 of the 46 patients (13%) required catheter removal for VT (n = 2) or malfunction (n = 4). At a median follow-up of 30 days, 6 patients in the CICC group died of CR complications versus none of the patients in the PICC group (P = .012). Using PICCs, the reduction in BSI and symptomatic VT decreased mortality from CR infection and venous thromboembolism. In contrast, the CICC approach led to early catheter removal mostly for difficult-to-treat infectious pathogens. CONCLUSION: Our data have confirmed that BSI and symptomatic VT are the major complications affecting frontline central intravascular device-related morbidity in the leukemia setting. The use of a PICC is safer than that of a CICC and maintains the effectiveness for patients with AML undergoing chemotherapy, with an approximate fourfold lower combined risk of infection or thrombosis at 30 days.
