Thoracotomy versus sternotomy? The effect of surgical approach on outcomes after left ventricular assist device implantation: A review of the literature and meta-analysis

Background and aim: Thoracotomy approaches to left ventricular assist device (LVAD) implantation may reduce surgical morbidity and, through preservation of the pericardial restraint over the right heart, may reduce the incidence of right ventricular failure (RVF). Methods: A meta-analysis of all original studies describing the effect of the surgical approach on postoperative outcomes after LVAD implantation was performed. Postoperative outcomes analyzed. Results: Thirteen studies were included with 692 patients undergoing a sternotomy and 373 a thoracotomy approach. Patients undergoing a thoracotomy approach had a higher comorbid status (INTERMACS 1-2: 56% vs. 44%; p = .0004), but were less likely to undergo a concomitant procedure (4% vs. 15%; p = .0002) than patients undergoing a sternotomy approach. Patients undergoing a thoracotomy approach demonstrated a reduced incidence of RVF (OR, .47; CI, .23-.97; p = .04), reexploration for bleeding (OR, .55; CI, .32-.94; p = .03), perioperative blood transfusion (SMD, -.30; CI, -.49 to -.11; p = .002), LOS (-5.57; -10.56 to -.59; p = .03), and mortality (OR, .57; CI, .33-.98; p = .04), but no difference in RVAD requirement or stroke were noted. Metaregression demonstrated that the performance of a concomitant procedure did not modify the effect of the surgical approach on the primary endpoints of RVF or RVAD requirement. Conclusions: In the current meta-analysis including over 1000 patients undergoing LVAD implantation, a thoracotomy approach was associated with a reduced incidence of RVF (but not RVAD requirement), bleeding, LOS, and mortality. No difference in stroke rates was noted. These findings not only offer additional support as to the feasibility of a thoracotomy approach for LVAD implantation but also suggest a potential superiority over a sternotomy approach.