Delayed versus primary sternal closure for left ventricular assist device implantation: Impact on mechanical circulatory support infections

Abstract

Background: Delayed sternal closure may be required after left ventricular assist device (LVAD) implantation due to coagulopathy or hemodynamic instability. There is conflicting data regarding infection risk.

Methods: We performed a single-center, retrospective analysis of patients who received their first LVAD between May 2012 and January 2021. Patients were divided into delayed sternal closure (DSC) and primary sternal closure (PSC) groups. We used chi-squared or Fisher Exact tests, as appropriate, to compare the incidence of postoperative LVAD-related infections (mediastinal/sternal wound) and LVAD-specific infections (driveline and pump pocket) after definitive chest closure between these two groups.

Results: A total of 327 patients met eligibility criteria, including 127 (39%) patients that underwent DSC and 200 (61%) patients that had a PSC. Demographic and clinical characteristics were similar except for an overrepresentation of men (87% vs. 75%, p = .016), Interagency Registry of Mechanically Assisted Circulatory Support class I-II patients (89% vs 66%, p < .001), patients with a previous sternotomy (43% vs 13%, p < .001), and patients with chronic kidney disease (55% vs 43%, p = .030) in the DSC group. The median DSC time was 24 (IQR: 24-48) hours. The incidence of LVAD-related mediastinal/sternal wound infection was similar between the DSC and PSC groups (4.7% vs 3.0%, p = .419). There was no difference between DSC and PSC groups in the incidence of driveline infection (6.3% vs 9%, p = .411) and pump pocket infection (1.6% vs 1.5%, p =.901), respectively.

Conclusions: DSC does not seem to increase the incidence of LVAD-related or LVAD-specific infection rates in heart failure patients undergoing device implantation surgery.

Keywords: heart; hemodynamics; mechanical support device; sternotomy; surgery.