Abstract
Left ventricular assist devices (LVADs) provide mechanical circulatory support for patients with end-stage heart failure. In some patients, mechanical support may lead to adequate recovery of myocardial function. In such cases, discontinuation of LVAD support, known as decommissioning, may be appropriate. Traditionally, LVAD decommissioning required surgical explantation, a procedure associated with significant risk and surgical complexity. More recently, a less invasive approach has emerged utilizing transcatheter outflow graft occlusion and driveline transection. Here, we present seven cases of transcatheter LVAD decommissioning highlighting the procedure and potential clinical benefits.
Learning objective
Left ventricular assist device (LVAD) decommissioning with transcatheter outflow graft occlusion is a potential management option for patients exhibiting myocardial recovery and may be preferable to surgical explantation in carefully selected patients. More studies are needed to assess long-term outcomes after transcatheter LVAD decommissioning and define optimal anticoagulation management post-procedure.
Keywords: Left ventricular assist device, Decommissioning, Transcatheter, Outflow graft, Vascular plug
Introduction
Left ventricular assist devices (LVADs) provide mechanical circulatory support for patients with end-stage heart failure (ESHF) [1]. The mechanical unloading provided by LVADs can promote myocardial recovery, making the eventual withdrawal of LVAD support possible [2]. Traditionally, LVAD decommissioning was performed either through surgical explantation, requiring redo sternotomy and cardiopulmonary bypass, or through a less invasive approach involving graft ligation via a small left thoracotomy, avoiding cardiopulmonary bypass [3]. More recently, transcatheter LVAD decommissioning, an approach that avoids chest re-entry, has been described in case reports and uses vascular plugs to occlude the outflow graft while surgically removing the driveline [4], [5] (Fig. 1). Previously, we reported transcatheter LVAD decommissioning in five patients followed for up to 1.5 years [5]. Here, we expand this experience to seven patients with over three years of follow-up.
Fig. 1.
Overview of transcatheter LVAD decommissioning including potential benefits and limitations [Created in BioRender. D, R. (2025) https://BioRender.com/j4q87nn].
LVAD, left ventricular assist device.
Decommissioning procedure
Selection for transcatheter decommissioning was determined by a multidisciplinary heart team with careful consideration of patient characteristics. Candidacy was established based on evidence of myocardial recovery demonstrated through a structured turn-down study (Table 1). Decommissioning was performed following shared decision-making, incorporating patient preferences when feasible, particularly in those electing palliative care. Under conscious sedation, arterial access was obtained, and an 8F sheath was advanced to the aortic insertion site of the outflow graft. The graft was then engaged with a Judkins Right 4 or Multipurpose diagnostic catheter and a Glidewire, and the destination sheath was advanced to the proximal anastomosis of the graft at the junction with the motor housing (Video 1a). Once the LVAD was turned off, Amplatzer Vascular Plugs (AVPs) (Abbott, Chicago, IL, USA) were deployed proximally, at the mid-graft, and distally at the aortic anastomosis (Video 1b, c). Angiography prior to release of the final plug confirmed appropriate flow reduction (Video 1d), and additional plugs were placed if necessary. Subsequently, the driveline was removed using local anesthetic. Patients continued anticoagulation due to the inflow cannula in the left ventricle.
Table 1.
Clinical characteristics of patients who underwent LVAD decommissioning, including the number of Amplatzer vascular plugs used, plug sizes and configurations, and post-procedure anticoagulation regimens.
| Case | Age/Gender | Etiology of HF | Device | Days on support | Reason for decommissioning | Follow-up period (days) | Number of plugs | Proximal graft | Mid-graft | Distal graft | Aortic anastomosis | Post-procedure anticoagulation | Medical therapy |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Patient 1 | 53 y/o F | Ischemic CM | HeartMate 3 | 287 | Myocardial Recovery | 188 | 2 AVP-1 3 AVP-2 |
14 mm AVP-1 |
16 mm AVP-1 |
14 mm AVP-2(x2) |
16 mm AVP-2 |
Apixaban |
|
| Patient 2 | 43 y/o M | Non-ischemic CM | HeartMate 2 | 2626 | Myocardial Recovery | 820 | 2 AVP-1 2 AVP-2 |
14 mm AVP-1 |
16 mm AVP-1 |
16 mm AVP-2(x2) |
Apixaban | Declined | |
| Patient 3 | 45 y/o M | Alcohol-induced CM | HeartMate 3 | 657 | Myocardial Recovery | 1349 | 2 AVP-1 2 AVP-2 |
14 mm AVP-1 | 16 mm AVP-1 |
14 mm AVP-2(x2) |
Apixaban |
|
|
| Patient 4 | 51 y/o F | Peripartum CM | HeartMate 2 | 1659 | Chronic driveline infection and recurrent thrombosis | 249 | 2 AVP-1 1 AVP-2 |
- | 16 mm AVP-1 |
14 mm AVP-2 |
14 mm AVP-1 | Warfarin |
|
| Patient 5 | 63 y/o M | Non-ischemic CM | HeartMate3 | 602 | Myocardial Recovery | 1633 | 2 AVP-1 2 AVP-2 |
14 mm AVP-1 |
16 mm AVP-1 |
14 mm AVP-2(x2) | - | Apixaban |
|
| Patient 6 | 57 y/o M | Idiopathic | HeartWare | 195 | Myocardial Recovery | 918 | 1 AVP-1 1 AVP-2 |
16 mm AVP-1 |
- | 16 mm AVP-2 |
- | Warfarin |
|
| Patient 7 | 54 y/o F | Non-ischemic CM | HeartMate 2 | Unknown | Quality of life/patient-decision | Unknown | 2 AVP-1 1 AVP-2 |
16 mm (x3) | - | - | - | Unknown | None - Palliative Carea |
AVP, Amplatzer vascular plug; CM, cardiomyopathy; F, female; HF, heart failure; LVAD, left ventricular assist device; M, male.
Patient 1
A 53-year-old female with ischemic cardiomyopathy requiring HeartMate3 (HM3) (Abbott) LVAD placement presented with low flow alarms one year after implantation. A transthoracic echocardiogram (TTE) revealed recovery of LV systolic function with ejection fraction (EF) 45–50%. Right heart catheterization (RHC) showed favorable hemodynamics. She tolerated the LVAD decommissioning well and was discharged on guideline-directed medical therapy (GDMT). A repeat TTE four months later demonstrated a reduction of EF to 10–15%. She died six months post-decommissioning from unclear causes.
Patient 2
A 43-year-old male with non-ischemic cardiomyopathy (NICM) secondary to viral myocarditis leading to ESHF underwent HeartMate2 (HM2) (Abbott) implantation. Approximately seven years post-implant, he demonstrated myocardial recovery with EF 55%. He subsequently underwent LVAD decommissioning and had no recurrence of symptoms. A repeat TTE four months post-procedure demonstrated reduced EF of 30–35%. He declined medical therapy given symptomatic improvement and was lost to follow-up.
Patient 3
A 45-year-old male with NICM developed ESHF necessitating HM3 placement. Two years post-implant, his EF was improved to 40–45% and RHC showed favorable hemodynamics during turn-down study. He underwent successful LVAD decommissioning. Repeat TTE after one-year post-decommissioning demonstrated EF 40–45%.
Patient 4
A 51-year-old female with ESHF secondary to peripartum cardiomyopathy required HM2 implantation. Her course was complicated by chronic LVAD driveline infection and recurrent pump thrombosis. Due to these complications, she presented for transcatheter LVAD decommissioning. She tolerated the procedure without complications and was maintained on home inotropes with a plan for eventual HM3 implantation. She died six months later from uncertain causes.
Patient 5
A 63-year-old male with NICM and ESHF required HM3 implantation. He underwent RHC showing signs of myocardial recovery and favorable hemodynamics. He proceeded with LVAD decommissioning and has been maintained on GDMT without HF readmissions since the procedure. His most recent TTE demonstrates EF 40% four years post-procedure.
Patient 6
A 57-year-old male with NICM required placement of a HeartWare ventricular assist device (HVAD) (HeartWare Inc., Miami Lakes, FL, USA). He demonstrated myocardial recovery one year after HVAD implantation and underwent transcatheter decommissioning. While he remained symptom-free on medical therapy with an EF of 55–60%, he was later diagnosed with rectal carcinoma. He died six months post-decommissioning due to complications related to his cancer.
Patient 7
A 54-year-old female with NICM requiring HM2 implantation was admitted for a heart failure exacerbation due to medication non-adherence. The patient reported a poor quality of life with the LVAD and requested its removal for palliative purposes. She underwent transcatheter decommissioning and died one month later.
Discussion
While earlier reports estimated the incidence of myocardial recovery in patients with LVADs at only 1–2%, more recent studies suggest that recovery rates could be as high as 40% given advances in medical therapy [2], [6]. This may lead to an increase in the number of candidates for LVAD decommissioning. Surgical LVAD explantation is associated with high perioperative (9.2%) and late mortality (15%) [7]. Less invasive options eliminate the need for redo sternotomy and have lower procedural complications [4], [5]. However, the long-term risk of leaving the device in situ is not well understood. A meta-analysis compared LVAD decommissioning with surgical explantation, analyzing data from 85 patients across 44 studies. Of these, 20% underwent minimally invasive decommissioning and 80% underwent surgical explantation. Among those who underwent minimally invasive decommissioning, only seven patients had transcatheter occlusion of the outflow graft using vascular plugs. The study found no significant differences between the two groups in thromboembolic events, infection rates, or overall survival [8]. With follow-up of just over a year, further studies with longer follow-up are needed. The study also reported a higher rate of HF recurrence in patients who underwent LVAD decommissioning compared to those treated with surgical explantation. The increased prevalence of recurrent HF in patients undergoing minimally invasive LVAD decommissioning may be attributed to selection bias, as these individuals might have been poor surgical candidates, or may result from leaving the device in situ.
In our cohort, four died during the follow-up period. Two patients underwent LVAD decommissioning for palliative indications and died one and six months after decommissioning, respectively. One patient died due to complications related to rectal cancer.
Another patient developed recurrent LV dysfunction and subsequently died from an unclear cause; in this context, potential complications related to retained hardware, including thromboembolism or infection, should be considered. Importantly, no complications were observed in any patients during the immediate post-procedural period, suggesting that LVAD decommissioning can be performed without early procedural mortality in carefully selected patients. Three patients demonstrated no evidence of recurrent LV dysfunction following decommissioning, with follow-up durations ranging from six months to four years. Two patients developed recurrent LV dysfunction; notably, one patient discontinued GDMT after symptomatic improvement, which may have contributed to the subsequent decline in ventricular function.
Unlike surgical explantation, transcatheter LVAD decommissioning leaves intracardiac and vascular hardware in place necessitating post-procedure anticoagulation. Although warfarin remains the standard anticoagulant following LVAD implantation, a recent study by Shah et al. suggested that apixaban may be a safer alternative with a lower bleeding risk [9]. In our cohort, we chose apixaban over warfarin for a majority of patients post-decommissioning and observed no cases of bleeding or cardioembolic events. Given the low number of patients in our series, further research is needed to define the optimal treatment.
It is worthwhile to highlight the technical advances of the transcatheter approach used in this case-series. Unlike previous reports that used bifemoral access and general anesthesia, we employed conscious sedation and single arterial access. We also utilized multiple AVPs to assist in minimizing retrograde flow within the outflow graft and to optimize apposition at the aortic anastomosis, minimizing the formation of a residual pouch [10]. Our initial two-plug approach was modified to routinely deploy three plugs per procedure to facilitate rapid thrombosis of the graft. In one case, a fourth plug was required when the third plug did not achieve adequate apposition at the aortic anastomosis. Rapid occlusion helps limit aortic flow regurgitation and subsequent left ventricular overload, possibly resulting in better outcomes. Complete occlusion of the outflow graft may not be immediately evident following plug deployment, particularly in the acute post-procedural period [10], and transient retrograde flow can mimic the physiology of aortic insufficiency. In this setting, intraprocedural hemodynamic monitoring, as previously described by our group, may assist in confirming adequate outflow graft occlusion. Further investigation is warranted to clarify the clinical significance of residual flow and to determine whether delayed angiographic confirmation is necessary to document subsequent thrombosis of the vascular plugs over time.
Conclusion
Transcatheter LVAD decommissioning represents a feasible and minimally invasive approach to LVAD withdrawal. Our study provides longitudinal follow-up of patients undergoing transcatheter decommissioning beyond the immediate peri-procedural period. Limitations include the small sample size, observational study design, and heterogeneity in clinical indications, which limit the generalizability of our findings and preclude definitive conclusions regarding long-term outcomes. Larger studies with longer follow-up are needed before this approach can be considered a routine alternative to surgical explantation.
Consent statement
Written informed consent was obtained from the patients.
Declaration of competing interest
The authors declare that there is no conflict of interest.
Acknowledgments
None.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.jccase.2026.07.003.
Appendix A. Supplementary data
(a) Advancement of the destination sheath into the LVAD outflow graft. (b) Positioning of the AVP-1 at the proximal outflow graft adjacent to the motor housing. (c) Positioning of two AVP-2s at the distal outflow graft near the aortic anastomosis. (d) Final angiography confirming effective occlusion of the outflow graft without residual flow.
AVP, Amplatzer vascular plug; LVAD, left ventricular assist device.
References
- 1.Rose E.A., Moskowitz A.J., Packer M., Sollano J.A., Williams D.L., Tierney A.R., Heitjan D.F., Meier P., Ascheim D.D., Levitan R.G., Weinberg A.D., Stevenson L.W., Shapiro P.A., Lazar R.M., Watson J.T., et al. The REMATCH trial: rationale, design, and end points. Randomized evaluation of mechanical assistance for the treatment of congestive heart failure. Ann Thorac Surg. 1999;67:723–730. doi: 10.1016/s0003-4975(99)00042-9. [DOI] [PubMed] [Google Scholar]
- 2.Birks E.J., Drakos S.G., Patel S.R., Lowes B.D., Selzman C.H., Starling R.C., Trivedi J., Slaughter M.S., Alturi P., Goldstein D., Maybaum S., Um J.Y., Margulies K.B., Stehlik J., Cunningham C., et al. Prospective multicenter study of myocardial recovery using left ventricular assist devices (RESTAGE-HF [Remission from Stage D Heart Failure]): medium-term and primary end point results. Circulation. 2020;142:2016–2028. doi: 10.1161/CIRCULATIONAHA.120.046415. [DOI] [PubMed] [Google Scholar]
- 3.Baldwin A.C., Sandoval E., Letsou G.V., Mallidi H.R., Cohn W.E., Frazier O.H. Surgical approach to continuous-flow left ventricular assist device explantation: a comparison of outcomes. J Thorac Cardiovasc Surg. 2016;151:192–198. doi: 10.1016/j.jtcvs.2015.08.024. [DOI] [PubMed] [Google Scholar]
- 4.Albulushi A., Goldsweig A.M., Stoller D., Delaney J.W., Um J., Lowes B., Zolty R. Percutaneous deactivation of left ventricular assist devices. Semin Thorac Cardiovasc Surg. 2020;32:467–472. doi: 10.1053/j.semtcvs.2020.01.012. [DOI] [PubMed] [Google Scholar]
- 5.Moroni F., Shah K.B., Quader M.A., Klein K., Smallfield M.C., Parris K.E., Gertz Z.M. Percutaneous approach to left ventricular assist device decommissioning. Catheter Cardiovasc Interv. 2022;100:169–174. doi: 10.1002/ccd.30230. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kirklin J.K., Pagani F.D., Kormos R.L., Stevenson L.W., Blume E.D., Myers S.L., Miller M.A., Baldwin J.T., Young J.B., Naftel D.C. Eighth annual INTERMACS report: special focus on framing the impact of adverse events. J Heart Lung Transplant. 2017;36:1080–1086. doi: 10.1016/j.healun.2017.07.005. [DOI] [PubMed] [Google Scholar]
- 7.Phan K., Huo Y.R., Zhao D.F., Yan T.D., Tchantchaleishvili V. Ventricular recovery and pump explantation in patients supported by left ventricular assist devices: a systematic review. ASAIO J. 2016;62:219–231. doi: 10.1097/MAT.0000000000000328. [DOI] [PubMed] [Google Scholar]
- 8.Choi J.H., Weber M.P., Horan D.P., Luc J.G.Y., Phan K., Patel S., Rizvi S.A., Maynes E.J., Reeves G.R., Entwistle J.W., Morris R.J., Massey H.T., Tchantchaleishvili V. Left ventricular assist device decommissioning compared with explantation for ventricular recovery: a systematic review. ASAIO J. 2020;66:17–22. doi: 10.1097/MAT.0000000000000926. [DOI] [PubMed] [Google Scholar]
- 9.Shah P., Looby M., Dimond M., Bagchi P., Shah B., Isseh I., Rollins A.T., Abdul-Aziz A.A., Kennedy J., Tang D.G., Klein K.M., Casselman S., Vermeulen C., Sheaffer W., Snipes M., et al. Evaluation of the hemocompatibility of the direct oral anticoagulant apixaban in left ventricular assist devices: the DOAC LVAD study. JACC Heart Fail. 2024;12:1540–1549. doi: 10.1016/j.jchf.2024.04.013. [DOI] [PubMed] [Google Scholar]
- 10.Wang W., Li H., Tam M.D., Zhou D., Wang D.X., Spain J. The amplatzer vascular plug: a review of the device and its clinical applications. Cardiovasc Intervent Radiol. 2012;35:725–740. doi: 10.1007/s00270-012-0387-z. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
(a) Advancement of the destination sheath into the LVAD outflow graft. (b) Positioning of the AVP-1 at the proximal outflow graft adjacent to the motor housing. (c) Positioning of two AVP-2s at the distal outflow graft near the aortic anastomosis. (d) Final angiography confirming effective occlusion of the outflow graft without residual flow.
AVP, Amplatzer vascular plug; LVAD, left ventricular assist device.