Minimally invasive approaches for LVAD implantation have the potential to improve outcomes for patients requiring mechanical circulatory support, particularly those with severe comorbidities. While CS remains the standard, LVAD implantation through MI techniques, including hemisternotomy and minithoracotomy, has demonstrated safety and feasibility, enabled by miniaturized devices such as the HeartMate 3® and the now-discontinued HeartWare®. Nevertheless, limited evidence exists on the impact of the chosen approach on postoperative morbidity and mortality.

The study’s objective was to compare LVAD implantation using CS versus MI in two high-volume cardiac surgery centers. From January 2014 to December 2018, LVAD devices were implanted in 342 consecutive patients. Patient characteristics were prospectively collected, and a propensity score analysis created two comparable groups in a 1:1 match. The unmatched cohort included 241 patients who received LVAD implants via CS and 101 who underwent MI surgery. The re-sampled cohort yielded two groups, each comprising 73 patients. In the matched groups, the reoperation rate due to bleeding was 4.1% (3/73) in the MI group, compared to 17.9% (12/67) in the CS group (p = 0.018). The ICU stay was significantly shorter in the MI group than in the CS group (10.5 days vs. 4 days, p = 0.008), as was hospital stay (37 days vs. 25.5 days, p = 0.007). Cumulative mortality incidence for the CS group was 24% at 1 year and 26% at 2 years among non-transplanted patients, while it was 22.5% at 1 year and 25.2% at 2 years for the MI group.

The authors conclude that the minimally invasive surgical approach is a safe technique for LVAD implantation. MI surgery was associated with a significant reduction in postoperative bleeding complications and hospital stay duration, with no significant differences in mortality incidence.

COMMENTARY:

This observational study represents the largest series to date explicitly comparing LVAD implantation via minimally invasive approaches (left thoracotomy and hemisternotomy) to the conventional sternotomy approach. Jawad et al. associate the MI approach with reduced postoperative bleeding and shorter hospital stays. The uniformity in the anticoagulation protocol and balanced distribution of device types across both groups stand out as positive aspects of the study design.

An increasing interest exists in promoting LVAD implants that preserve the sternum as much as possible using MI approaches. The primary and most evident reason for this interest is the potential of MI approaches to enhance perioperative outcomes by minimizing bleeding and optimizing immediate postoperative mobility. Another potential benefit could be the enhanced feasibility of heart transplantation in patients who had LVAD implants as a bridge to transplant or candidacy, as the avoidance of the full midline incision reduces pericardial adhesions. A less discussed but pertinent factor is the belief among professionals that MI approaches reduce the incidence of right ventricular (RV) dysfunction. Theoretically, the wide pericardial opening required for CS can cause RV distention, which MI approaches might avoid. Indeed, this study observed a trend, though not statistically significant, toward lower incidence and severity of RV dysfunction in the MI group. This study, like many others, associates MI approaches with reduced bleeding and shorter hospital stays but, in my view, does not fully address the critical question of whether RV function (both postoperative and long-term) improves due to the preservation of the pericardium and sternum in MI approaches.

The results of this study should be interpreted with caution. It is well-known that the validity of propensity score analysis heavily depends on appropriate variable selection. A limitation of this study was the exclusion of patients requiring concomitant valve procedures, and extracorporeal circulation time data from one center was missing. There is no information on pain control, a crucial factor when evaluating outcomes involving different surgical approaches, especially when including one via thoracotomy. Additionally, the authors missed the opportunity to assess RV function more rigorously. Furthermore, before matching, the MI group had a higher INTERMACS score and a greater incidence of pump-free LVAD implants, suggesting a lower surgical risk. Lastly, most cases were performed by two highly experienced MI surgeons, limiting the generalizability of the findings. Nevertheless, this study serves as a valuable reference for future prospective studies in this field.

REFERENCE

Jawad K, Sipahi F, Koziarz A, Huhn S, Kalampokas N, Albert A, et al. Less-invasive ventricular assist device implantation: A multicenter study. J Thorac Cardiovasc Surg. 2022 Dec;164(6):1910-1918.e4. doi: 10.1016/j.jtcvs.2020.12.043.