RESUMO
This retrospective cohort study evaluated the effect of topical polymyxin-trimethoprim (poly) prophylaxis on the incidence of driveline infections (DLIs) in patients with continuous-flow left ventricular assist devices. All 84 cases implanted 2005-2014 with device support ≥30 days were reviewed; support ranged 1 m-5.2 yrs. Beginning 2008, poly was applied to the exit site with dressing changes. Sixty-five patients received poly (poly group) for duration of follow-up, 19 did not (no-poly); group baseline characteristics were similar. No patient developed side effects from poly. Nineteen DLIs (10 in no-poly) occurred; not using poly was a risk factor. 89% of poly group DLIs were superficial, 4 were culture-negative. DLI-related bacteremia occurred in 11% of no-poly group and 0% of poly group. Compared with no-poly, poly group demonstrated improved freedom from DLI by Kaplan-Meier analysis (P < 0.0001) and a 75% lower overall and 95% lower deep DLI incidence (P ≤ 0.001). Deep DLIs occurred in 31.6% of no-poly vs. 1.5% of poly patients (P = 0.0004), although mean support duration (1 yr) and % support >1 yr (38%) were similar. These findings, which should be confirmed with larger comparative studies, suggest that topical polymyxin-trimethoprim prophylaxis may be effective in preventing DLIs.
