Surgical left ventricular assist device (LVAD) explantation is associated with increased risk of perioperative mortality (9.2%), myocardial infarction (10.8%), stroke (4.9%), and postexplant infections (4.3%).1 Minimally invasive surgical strategies have been explored, but still expose patients to risks associated with blood transfusion and myocardial trauma during dissection of adhesions.2 Percutaneous LVAD exclusion can potentially avoid perioperative complications associated with surgical exclusion. In this case series, we examined our institutional experience and reviewed published data for risks and benefits associated with percutaneous LVAD exclusion. A total of four patients underwent percutaneous LVAD exclusion between 2011 and 2018 at our institution. Subsequently, we performed a systematic review of all reported cases of percutaneous LVAD exclusion published in Pubmed/Medline between the years 2011 and 2018 and identified seven additional cases. The cumulative results (n = 11) are presented in Supplemental Table 1 https://links.lww.com/ASAIO/A436. The average age of all patients was 44 ± 17 years, 7 (64%) were men, and 9 (82%) had non-ischemic cardiomyopathy. The average time between LVAD implantation and percutaneous LVAD exclusion was 16 ± 9 months. Majority of the patients underwent percutaneous LVAD exclusion for myocardial recovery (n = 7, 64%); four patients had LVAD-related complications (mainly, pump thrombosis) and were deemed poor surgical candidates for exchange or explant and were not suitable for transplant listing. Either an Amplatzer Vascular Plug II (St. Jude Medical, St. Paul, Minnesota) (n = 8, 73%) or an Amplatzer Septal Occluder (St. Jude Medical, St. Paul, Minnesota) (n = 3, 27%) were used to occlude the VAD outflow or inflow grafts. The type of Amplatzer device used to occlude the grafts was mainly based on operator’s preference. The average disc diameter of Amplatzer device used to occlude the outflow graft of HeartMate II device was 19 ± 2 mm and HeartWare LVAD was 13 ± 1 mm. The majority of implantations occluded the outflow graft (n = 9, 82%), while two cases reported occluding only the inflow graft (n = 2, 18%). The occlusion device was deployed at both proximal and distal portions of the outflow graft in the majority of the cases (n = 7, 64%). All cases reported successful LVAD exclusion. Following implantation, the driveline was excised followed by closing the open end with either Gore-Tex or BioGlue Surgical Adhesive (CryoLife Inc, Kennesaw, Georgia) in most cases. None of the cases reported any procedure related direct complications or death. The average reported postprocedure follow-up was 18 ± 11 months. For postexclusion stroke prophylaxis; majority of patients received warfarin with an INR goal range between 1.5 and 3. Nonetheless, one patient was discharged on daily Aspirin 100 mg only,3 while another did not receive any anticoagulation.4 The reason for using only Aspirin was not clarified by the authors while the patient without anticoagulation had percutaneous LVAD exclusion as a palliative option. However, both patients were reported to be alive without any cerebrovascular complications on long-term follow-up. These two cases suggest that warfarin might not be needed postexclusion for stroke prophylaxis, and further studies are required to validate this observation. Two of the four patients who underwent exclusion due to LVAD-related complications died within 2 weeks of the procedure, whereas the remaining seven patients who underwent exclusion for myocardial recovery were still alive at last reported follow-up. In summary, percutaneous LVAD exclusion appears to be associated with high procedural success and minimal procedural complications, and is potentially, an attractive alternative to conventional surgical exclusion. However, it is worth mentioning a few limitations of our study. First, percutaneous LVAD exclusion can theoretically be associated with higher risk of thromboembolic complications and therefore, necessitates lifelong anticoagulation. Second, the presence of retained LVAD hardware with direct endovascular access also carries the risk of endocarditis. However, neither of the abovementioned complications were reported in our review of these patients. A larger sample size or a longer follow-up period of these cases will be necessary to establish the incidence of the abovementioned potential long-term complications. Lastly, the publication bias also needs to be kept in mind which may have resulted in overestimation of the success rate of percutaneous LVAD exclusion. Prospective studies with larger sample size and longer duration of follow-up are necessary to accurately understand the risk and benefit associated with percutaneous LVAD exclusion. Acknowledgement The authors thank Balsara K., Brinkley M., Wigger M., Menachem J., Punnoose L., and Brown Sacks S. for their valuable clinical contribution.
Read full abstract