Back to 2015 Annual Meeting

BACKGROUND:
Left ventricular assist devices (LVADs) have revolutionized the treatment of patients with end-stage cardiac disease. However, infection rates can be as high as 30-50%. While driveline infections are most common and patients can usually be managed in the ambulatory setting, infection of the LVAD pocket usually requires hospital admission and multiple surgical procedures, leading to epigastric wound breakdown, LVAD exposure, and prolonged hospital stays. In such instances, early involvement of reconstructive surgeons could be advantageous. The purpose of this study was to determine whether the use of pedicled flaps for coverage of exposed LVADs after pocket infection is associated with a decrease in the number of days spent in the hospital.
METHODS:
Data was obtained from our institution's LVAD database and individual medical records. Patients included in our study developed LVAD pocket infection between January 2009 and April 2015. Date of pocket infection (DPI) was defined as the point in time when infection involving the device pocket (purulence, wound breakdown) became obvious to the reviewer. For each patient, data on all hospital admissions after the DPI were retrieved, including reason for admission (pocket-infection-related or not), length of stay, procedures performed, and outcomes. Pedicled flaps were defined as using either omentum or rectus abdominis muscle. Pre-flap period was defined as the number of days between DPI and date of flap coverage (DFC). Post-flap period was defined as the number of days between DFC and either the patient's death or April 30th, 2015. Hospitalization ratio (HR) was defined as the percentage of time that each patient was hospitalized and was calculated for each individual's pre- and post-flap periods. Patients who did not undergo flap coverage were assigned pre-flap period only. Unpaired t-test was then used to compare the pre-flap and post-flap HR of all patients. Separate t-tests were calculated for pocket-infection-related admissions and for all admissions.
RESULTS:
During the study period, 25 patients developed device pocket infection. Twelve (48%) underwent pedicled flap coverage, of which 7 patients (58.3%) underwent rectus abdominis flap and 5 patients (41.7%) had omental flaps. There was one partial and one total flap loss. For pocket infection-related admissions, pre-flap HR dropped from 37.2%±40.9% (±standard deviation) to 18.7%±29.0% in the post-flap period, reaching statistical significance (p=0.02). In regards to all admissions, pre-flap HR dropped from 42.6%±40.3% to 22.1%±30.0% in the post-flap period, also reaching statistical significance (p=0.01).
CONCLUSIONS:
To our knowledge, we report the highest number of patients with LVAD pocket infections who underwent coverage of exposed device with pedicled flaps. Our data suggests that the use of pedicled flaps for coverage of exposed LVADs after pocket infection is associated with a decrease in the number of days spent in the hospital, including both infection-related and all admissions. Muscle and omentum flaps for coverage of exposed LVADs after pocket infection is associated with both fewer infection-related DIH and all DIH, translating to lower hospitalization cost. For patients refractory to conservative measures to manage LVAD infection, pedicled muscle or omentum flaps should be considered.