Comparison of Short-Term Outcome after Thoracotomy for Left Ventricular Assist Device Implantation Compared with Re-Do Sternotomy in Patients with Previous Cardiac Surgery

PURPOSE A thoracotomy approach for left ventricular assist device implantation is theoretically beneficial in patients who had previous sternotomy. The aim of this study is to investigate short-term outcomes of a thoracotomy approach versus a re-do sternotomy approach for left ventricular assist device implantation in patients with previous sternotomy. METHODS We retrospectively reviewed 227 patients who underwent left ventricular assist device (HeartWare, HVAD) implantation between 7/2011 and 1/2019. Of these, 52 had a previous sternotomy. Twenty-eight patients were excluded because they had an extracorporeal ventricular assist device, cardiac surgery during the same hospitalization, concomitant procedures or an alternative outflow graft anastomosis at the time of the implantation. The study cohort was divided in two groups based on incision approach: a left anterior thoracotomy approach (LAT group, N=14) and a re-do sternotomy approach (RS group, N=10). RESULTS Previous cardiac surgeries of the cohort were coronary artery bypass grafting (CABG) in 16 patients, aortic valve replacement (AVR) in 1, mitral valve (MV) replacement in 1, MV repair in 1, CABG + AVR in 1, CABG + MV repair in 1, atrial septal defect closure in 2, and heart transplant in 1. There was no difference in baseline characteristics between the groups. Median cardiopulmonary bypass time was similar, though median operative time in the LAT group was longer than that in the RS group (LAT; 267 mins vs RS; 235 mins). There was no statistical difference in postoperative short-term outcomes between the groups, except the LAT group required less postoperative red blood cell transfusion compared to the RS group (LAT; 57 % vs RS; 100 %, P = 0.024). CONCLUSION A thoracotomy approach was feasible alternative in patients with previous cardiac surgery and associated with less blood transfusion requirement compared with a re-do sternotomy approach.