Deactivation of LVAD support for myocardial recovery—surgical perspectives

Abstract

Left ventricular assist devices (LVADs) are excellent therapies for advanced heart failure patients either bridged to transplant or for lifetime use. LVADs also allow for reverse remodeling of the failing heart that is often associated with functional improvement. Indeed, growing enthusiasm exists to better understand this population of patients, whereby the LVAD is used as an adjunct to mediate myocardial recovery. When patients achieve benchmarks suggesting that they no longer need LVAD support, questions related to the discontinuation of LVAD therapy become front and center. The purpose of this review is to provide a surgical perspective on the practical and technical issues surrounding LVAD deactivation.

Previous work published in the journal provides a detailed investigation of the biological and clinical perspectives regarding the mechanism and management of myocardial recovery in patients supported with left ventricular assist devices (LVADs).^1,2 The purpose of this review is to more thoroughly expand upon the practical and technical approaches to the management of the LVAD patient when it is decided that the pump is no longer needed, that is, how does one discontinue or deactivate LVAD support?

LVAD deactivation refers to any method that terminates the physiologic function of the pump; that is, the pump is turned off, regardless of whether it is removed or left in place. Deactivation can be further subclassified based on the fate of the actual pump. Explant implies its removal. This can be complete, whereby all system components are removed; or partial, whereby, at minimum, the actual pump (motor) is removed but other elements remain in place (outflow graft and/or sewing ring). Explant furthermore implies that the driveline is removed. Conversely, decommissioning entails that the pump (and therefore the driveline) remains in situ despite the termination of its physiologic activity (Table 1).

Table 1.

Proposed nomenclature for techniques associated with LVAD deactivation

Explant	Decommission
Removal of pump (including driveline)	Physiologic function disabled; pump/driveline remain in situ
Complete  - Physical removal of pump, driveline and all foreign material from left ventricle apex.	Outflow graft management  - Percutaneous/catheter-based occlusion  - Surgical ligation
Partial  - Complete removal of pump and driveline, but may have remnant sewing ring, inflow cannula at apex and/or long segment of outflow graft	Driveline management  - Cut at level of skin with incision closure (“cut and dunk”)  - Not excised (intact in situ driveline remnant)

While the intent of this review is to provide guidance as to best practices for LVAD deactivation, several a priori limitations are worth mentioning. The relative rarity of myocardial recovery with reported pump discontinuation hampers existing data that (1) are composed of small case series with diverse patient populations subject to treatment bias, (2) include an assortment of pumps, (3) have a mixture of retained LVAD system products, (4) vary in their approach to weaning protocols, and (5) are not subject to standardized postprocedural medical management. Despite these caveats, a combination of the existing literature and anecdotes forms the basis for the following surgical perspective related to LVAD deactivation.

Implant considerations

When considering the potential for myocardial recovery and concomitant surgical procedures at the time of LVAD implantation, there are several guiding principles: (1) repair organic and structural heart defects that inhibit ventricular reverse remodeling; (2) promote the overall function of the recovered heart, including the right ventricle; and (3) fix lesions at the time of implant rather than at the time of explant as one wants to decrease the trauma of the explant operation.

While not a new concept, surgeons should prepare for the next operation. Physical issues to consider include reconstruction of the pericardium (either native or prosthetic), coverage of the device, and placement of the driveline. The latter point is particularly important when considering explant. Ideally, the driveline should hug the diaphragm and not be draped across the cardiac structures. It can be quite difficult to dissect the driveline of the ventricle, especially with limited exposure. Figure 1 demonstrates 2 different patients with different post-explant outcomes. The patient in the upper panels had a partial explant whereby the driveline could not be safely removed in its entirety. He subsequently developed a smoldering infection that led to the recurrence of his heart failure and the development of a mediastinal abscess, requiring full explant and bridging extracorporeal support until he was transplanted. Conversely, the patient in the lower panel was able to have the driveline easily removed, and she remains free from heart failure recurrence after partial explant.

Figure 1.

Driveline position in potential left ventricular assist device explant patients. (A) Pre-explant chest radiograph with the driveline draping over the right ventricle. (B) At the time of the device explant, the driveline could not be safely removed and remained in situ. (C) Pre-explant chest radiograph with driveline tucked between the inferior wall and diaphragm. (D) Post partial explant radiograph with complete driveline removal.

With this “next operation” concept in mind, the goal of minimizing trauma and tissue manipulation during ventricular assist device (VAD) implantation may help to prepare the patient more wholly for future deactivation. The development of the physically smaller intrapericardial pumps has allowed for minimally invasive, sternal-sparing surgical implantation. These techniques not only offer reduced instrumentation and manipulation of the heart/pericardium with a subsequent reduction in the adhesive burden encountered at the time of device deactivation but have also been associated with reduced postimplant transfusion requirement, reoperation for bleeding, length of hospital stay, and early right ventricular dysfunction. Nonetheless, there are not yet data that demonstrate an association with increased overall survival or incidence of myocardial recovery.^3–7

Postimplant considerations

Postimplant considerations are numerous and have been recently reviewed.¹ Issues include LVAD speed and unloading optimization, the timing of discontinuation, rigorous uptitration of guideline-directed medical therapy, robust cardiac rehabilitation, serial echocardiographic surveillance, and weaning protocols that include both noninvasive and invasive hemodynamic studies.^8–14

While there is some consensus related to criteria for elective deactivation, little data are available to assist providers in cases involving urgent deactivation, usually associated with infection or thrombosis. In these circumstances, it is often clear what the best path forward should be if there is a minimal recovered function, then one would either move towards pump exchange or transplant; if the heart is functioning well, one would move to formal discontinuation. The gray area is found in the infected patient with an ejection fraction of 25% to 30%, a partially reverse-remodeled left ventricle, and perhaps some right ventricular dysfunction. One could argue that many heart failure patients live in such a state and simply remove VAD support; conversely, the risk of recurrent decompensation remains high. While judicious approaches to deactivating these patients can yield favorable results,¹⁵ registry data report a markedly reduced survival in nonelective explants (ISHLT, unpublished data, 2023). Hence, managing this group of patients remains an active area for investigation.

Discontinuation of LVAD support

As outlined, there are multiple ways to discontinue LVAD support, with no clear evidence to support 1 method over another (Figure 2). A 2020 systematic review sought to address this question by examining a total of 85 deactivation patients: explant (n = 61) versus decommissioning (n = 17). The authors detected no statistically significant difference in mortality, stroke, recurrence of heart failure, or need for transplant.¹⁶ Table 2 outlines the available literature with regard to LVAD deactivation. (Additional case reports are provided in Table S1.) Intuitively, it makes sense that a frail patient might be better served with decommissioning while a patient with an infection would be best served with complete explantation. With these caveats, as well as those previously mentioned, we provide an overview of the approaches to LVAD deactivation.

Figure 2.

Illustration of several techniques for LVAD discontinuation. Modified after Faerber and Doenst.³⁹ (A) Explantation with apical management using plug or direct repair. (B) HeartMate 2 partial explant with long-segment outflow graft and inflow cannula remnants in-situ. (C) Decommissioning with surgical outflow graft ligation or (D) endovascular occlusion device. (E) Decommissioned driveline transection and buried under skin.

Table 2.

Current literature related to LVAD deactivation arranged in reverse chronological order. Studies are retrospective analyses with total patients n ≥ 5 unless otherwise specified in the “Key Findings” column. (Publications with n < 5 are included in Supplementary Table 1).

Reference	Year	N	Device	Approach	Outflow	Apex	Key Findings
Knierim29	2023	61	HMII, HM3 HVAD, INCOR	pEx Dec.	pEx - Ligation; long Dec. - pump thrombosis	pEx- Apical plug (titanium)	Overall survival: 5 year-76.1%; 95% CI (64.2%–95.2%); Elective vs. Non-elective explant: HR 1.79 (0.58 5.53); p=0.312 Non-elective and elective deactivation = similar survival.
Bhadra37	2023	5	HM3 HVAD	Ex	Ligation; short	Double patch	Double patch technique allows preservation of sewing ring, avoids need for customized apical plug. No cases of thrombus formation during ~1.5-year observation period
Potapov42	2022	68	HMII, HM3 HVAD	pEx	Ligation; long	Apical plug (titanium)	Multicenter; Overall survival: 1 year-90.1% (84.0%–98.1%), 5 year-80.0% (68.4%–92.9%); CVA: 0%; Infection: 1.5% (n=1). 5 successful pump re-implantations through remnant sewing ring with plug removal, 1 case of infection requiring explant in case of pre-existing driveline infection.
Zhang62	2022	13	HMII HVAD	pEx	Ligation	Apical plug (unknown)
Mariani63	2022	7	HM3	pEx	Ligation	Apical plug (titanium)	Apical plug via thoracotomy offers sternotomy-sparing pEx option.
Gyoten64	2022	12	DH, EH HMII, HM3	Ex	Ligation	Suture repair	Survival free of HF recurrence (40-months): 100% Complete LVAD explant is effective and durable, inclusive of patients with active infection at the time of deactivation.
Moroni65	2022	5	HMII, HM3 HVAD	Dec.	Amplatzer x 2–3	NA	Percutaneous dec. is a viable option for VAD deactivation but may require ongoing anticoagulation.
Abdallah66	2022	6	HM3	Dec.	Amplatzer x 2	NA	Description of percutaneous dec. technique in the HM3.
Monteagudo67	2021	26	HMII HVAD	Ex pEx	Ligation	HMII - suture closure HVAD - Apical plug (titanium)	Overall survival: 1 year-88.0%; 6 year-77.0% Strict 4-phase weaning protocol combined with a less-invasive approach to LVAD explant can produce good, durable outcomes
Gyoten38	2021	12	HMII, HM3 HVAD	Ex	Ligation	Double patch	Survival free of HF recurrence or transplant: 1 year-91.7%
Gerhard33	2021	17	HMII HVAD	Dec.	1. Ligation 2. Amplatzer	NA	Dual-center; Survival free of HF requiring hospitalization, LVAD re-implant, or transplant: 1 year-94%, 2 year-87%, 3 year-78% Thromboembolism: 18% (12% CVA); Infection: 41% (all in patients with infection prior to dec.)
Birks22	2020	40	HMII	Ex pEx Dec.	Ex/pEx-Ligation Dec.-Amplatzer	Ex - patch or suture repair pEx - patch, plug, or suture	Multicenter; Survival free of VAD re-implant or transplant: 1 year-90%; 2 and 3-year-77.1% RESTAGE-HF prospective trial demonstrating feasibility of protocolized pharmacologic and mechanical unloading for the promotion of durable myocardial recovery.
Antonides68	2020	45	HMII,HM3, HVAD, Unknown	Ex pEx	-	-	EUROMACS analysis: Survival free of VAD re-implant, transplant, or recurrence of HF > NYHA III: 30-day: 100%; 2 year-88% Younger age at implant, shorter duration of HF, and non-ischemic etiology = greatest odds of recovery.
Albulushi69	2020	7	HMII, HM3 HVAD	Dec.	Amplatzer II x 3	NA	Survival free of VAD re-implant, HF recurrence, or transplant: 3 year-71% Percutaneous dec. is feasible in patients of high perioperative risk
Kapoor70	2019	7	HMII HVAD	Dec.	Ligation; long	NA	Thromboembolism: 0%; Infection: 14% (n=1, in patient with recurrent infection pre-Dec.) Reports outcomes for minimally-invasive surgical dec.
MacGowan71	2017	9	HVAD	Dec.	1. Ligation 2. Percutaneous occlusion (undisclosed)	NA	Surgical dec. of HVAD appears to have similar durability of recovery compared to Ex or pEx, but presents high infectious risk in patients with previous VAD infection.
Baldwin72	2016	27	HMII HVAD	Ex pEx Dec.	Ex-Ligation; short pEx-Ligation; long Dec.-Pump thrombosis	Ex-Suture repair pEx-Inflow ligation or Apical plug (unknown)	No significant association between approaches and post-op LOS, CVA, infection, and overall survival. Description of 4 deactivation techniques with comparison of outcomes, presents one of the earliest comparisons of different deactivation techniques.
Tchantcha-leishvili73	2016	8	HMII	Ex	Ligation	-	Survival free of VAD re-implantation or transplant at last follow-up (276 ± 240 days): 87.5%
Frazier28	2015	27	HMII HVAD	pEx Dec.	pEx-Ligation; short Dec.-Ligation Dec.-Pump thrombosis	pEx-Apical plug (felt) Dec.-NA	Survival free of VAD re-implantation transplant (1172 ± 948 days): 85.2%
Goldstein5	2012	20	HMII	-	-	-	Multicenter; Survival free of re-implant or transplant: 3.5 years-74% Factors associated with myocardial recovery: young age (<40y), non-ischemia etiology, and shorter duration (<12 months) of heart failure.
Cheung74	2012	6	HMII	pEx	Ligation; short	Apical plug (felt)	Survival free of re-implant or transplant (688 (127–1033) days): 100%
Dandel6	2005	32	Novacor, TCI BHE	pEx	Ligation	Ligation of inflow Apical plug (unknown)	Risk of HF recurrence after explant: Off-pump LVEDD > 55 mm, EF < 45%, history of HF > 5 years.
Farrar31	2002	22	Thoratec LVAD or BiVAD	-	-	-	Multicenter; Survival free of VAD re-implantation or transplant: 1 year-86%; 5 year-77%
Hetzer75	2000	23	Novacor, TCI BHE	pEx	Ligation; long	Ligation of Inflow	Analysis of pulsatile-VAD patients who underwent pEx: ~35% with recurrent HF and 29% re-implantation and/or transplant.

Abbreviations: BHE: Berlin Heart Excor, DH: DuraHeart, EH: EVAHEART, HMII: HeartMate 2, HM3: HeartMate 3, HVAD: HeartWare, TCI: Thermo CardioSystems, HeartMate, VAD: Ventricular Assist Device, BiVAD: Biventricular Assist Device, Ex: Complete Explant, pEx: Partial Explant, Dec.: Decommissioning, LVEDD: Left ventricular end-diastolic diameter, HF: Heart Failure. (-) is used when information was not provided for a given study.

Complete explant

A complete explant of the LVAD and its associated foreign material (outflow graft, apical inflow, driveline) represents the most traditional and conceptually straightforward approach to support discontinuation.^37,40 The surgical approach is usually via median sternotomy or assorted alternative incisional patterns ranging from 2 to 4 total incisions (subxiphoid, right/left anterior mini-thoracotomy, anterolateral thoracotomy, epigastric) based on surgeon preference, patient anatomy, chest hostility, and residual right ventricular function. As previously mentioned, sternotomy-sparing approaches to LVAD implants have been associated with lower rates of early postoperative right heart failure, without a difference in overall survival. Therefore, in cases of borderline right ventricle function, surgeons may consider a less-invasive approach to complete explant in lieu of sternotomy. Outflow graft ligation at the insertion site of the aorta, leaving a small remnant of graft, can be completed with heavy suture ligation and sharp transection or surgical stapler ligation. Alternatively, the entire graft can be removed with either primary or patch closure of the aorta.

A variety of techniques have been described to manage the left ventricular apex. Complete explant assumes that the sewing ring is removed. Once the pump is removed, any debris is excised, and the apex is reconstructed with patch repair or primary closure. The latter often involves a circular endoventricular suture with subsequent linear epicardial closure (Figure 3). While the utilization of cardiopulmonary bypass provides a margin of safety and exposure, removal and subsequent reconstruction can also be performed without extracorporeal support utilizing a Foley catheter for apex occlusion and subsequent direct suture repair.

Figure 3.

Partial explant of HeartMate 3 left ventricular assist device. Images are seen through a left anterior thoracotomy. Cardiopulmonary bypass is initiated through a percutaneous femoral venous cannula and direct cannulation of the outflow graft. (A) After opening the native pericardium and previously placed Gore-Tex protective barrier, the bend relief is opened, revealing the outflow graft. (B) The slide lock is removed with a hemostat allowing, (C) the pump to be removed. (D) The intraventricular surface is evaluated. (E) Pulling on sutures before cutting demonstrates large “air-knots” suggestive of gross ventricular reverse remodeling. (F) The sewing ring is excised and (G) the ventricle is inspected for debris and/or thrombus. (H) The apical defect is closed with 2 concentric endoventricular 2–0 Prolene sutures followed by (I) linear epicardial closure over felt strips.

Benefits of complete explant include the removal of all foreign material, particularly important in cases of pump infection. Indeed, these cases have shown a near-universal need for subsequent conversion from partial explant or decommissioning to complete explant and, therefore, represent really the only clear case where 1 approach has demonstrated superiority.^4,26,41,42

Limitations include technical complexity with longer operative times, significant invasiveness, and the inability to easily reimplant a new VAD in the case of HF recurrence. Because of the “size” of the operation, there might be some concerns that patients with more fragile functional status might not tolerate the heightened inflammatory milieu associated with this trauma and suffer post-cardiotomy myocardial dysfunction.

Partial explant

The partial explant is similar to the complete explant in that the pump body and driveline are entirely removed, but it differs in that some remnant material remains. Most frequently, this occurs with residual apical material (sewing ring or portions of the inflow) and outflow graft, whereby it is ligated just beyond the pump body. While sternotomy can be used, most surgical approaches are intended to be less invasive. Single (e.g., thoracotomy or subcostal) and multiple incision (e.g., anterior mini-thoracotomy + subcostal) variations have been described. Both on and off-pump variations are additionally reported.

As previously mentioned, management of the apex could be with or without retention of the sewing ring. Handmade felt and, eventually, custom-manufactured titanium and silicone plugs emerged as options for insertion into the sewing ring to complete ventriculotomy closure. Alternatively, a “double-patch” apical closure technique in which bovine pericardium is inserted through the sewing ring to seal the intraventricular surface of the left ventricle (LV) with external, topical placement of a Gore-Tex patch over the sewing ring has been described.^17,25 These closure options allow for shorter operative times, avoidance of CPB, and preserve the apex to facilitate VAD reinsertion if necessary.^43–45

Much enthusiasm exists for specifically designed titanium plugs that fit into the existing sewing rings of the HVAD and Heartmate 3 devices. The plug can be specially ordered and used on a case-by-case basis as it is neither Food and Drug Administration-approved nor CE-marked. Interestingly, a new Chinese-based pump (CorHeart 6) has a recovery plug as a part of its insertion kit. A recent multicenter European study reported results of apical management with a titanium plug in 68 patients.¹⁸ In this cohort, there were no postoperative strokes, and only 1 patient had a plug infection in the context of chronic driveline infection. At a median of 34 months, 81% (55) were alive, and 87% of those were alive without re-LVAD or transplant (n = 48; 71% of the total cohort).

Benefits of the partial explant include the opportunity for limited manipulation of the heart, amenability to minimally invasive/sternotomy-sparing approaches, potential avoidance of the need for extensive ventricular reconstruction, possible avoidance of CPB, and the potential to preserve the apex for easier VAD reimplant in the case of HF recurrence.

Limitations are primarily related to residual foreign material. Patients with a prior history of device-associated infection are at a greater risk of ongoing device-associated infection in cases of partial versus complete explant. Furthermore, the residual outflow graft and, to a lesser extent, the intraventricular material do represent foci with potential for thrombosis. In addition, retention of the fixed sewing ring potentially limits apical contractility in a recovering heart. As depicted in Figure 3, removing the ring theoretically could provide a more geometric ventricle that could continue to enforce reverse remodeling.

Decommissioning

Decommissioning of the LVAD implies retention of the pump body in situ but requires management of the outflow graft to avoid closed-loop regurgitant flow from the aorta into the LV, hence precipitating ventricular dilation and failure. Management of the driveline in decommissioning consists of 2 distinct approaches: (1) transection with closure of the overlying skin and subcutaneous tissue and (2) no intervention, driveline remains in situ. The former has been more frequently reported in the literature; however, in cases of well-incorporated drivelines without a prior history of infection, some surgeons have elected to leave the driveline in situ in lieu of intervening on a structure that has otherwise been problem-free.

Surgical decommissioning

Multiple techniques and incisions have been described, but the general concept is the localization of the outflow graft with an incision that allows for easy access and some method of surgical ligation (stapler or suture). Following ligation, the LVAD is expected to thrombose over the next 48 to 72 hours.³³

Benefits of this approach include the avoidance of cardiopulmonary bypass and cardiac manipulation, minimal invasiveness, and the preservation of the apex in the event of VAD reimplantation.

Limitations are related to the burden of remnant foreign material. Just as there is a theoretical risk of thromboembolic complications in cases of partial explant, that risk remains in the decommissioned patient. Additionally, there is the risk of infection of the residual material. There has been some debate as to the impact on contraction dynamics of the LV related to the remnant inflow within the apex, but this has not been substantiated in the follow-up of patients who have undergone both elective and nonelective decommissioning (e.g., cases of pump thrombosis with incidentally discovered recovery).^15,26

Percutaneous decommissioning

Percutaneous decommissioning of the LVAD represents the least invasive option of support discontinuation. While general anesthesia is commonly utilized, there are reports of this procedure being performed with conscious sedation.⁴⁶ The procedure involves percutaneous access and the deployment of an endovascular occlusive device (most commonly Amplatzer or Amplatzer II plug) within the outflow graft. The number of plugs required is not standardized and ranges from 1 to 3. Other devices such as the Watchman or Lambre left atrial appendage occluder and Amplatzer Septal Occluder have been used with success; however, these are far less common.^47–49

Benefits of the percutaneous approach to decommissioning include its minimally invasive nature, complete avoidance of surgical manipulation of the heart, preservation of the LV apex in the event of VAD reimplantation, and even the ability to perform emergent outflow occlusion in cases of impending pump failure to avoid regurgitant flow and resultant LV overload while alternative replacement durable mechanical circulatory support becomes available.⁵⁰

Limitations of this approach are again related to the high burden of residual foreign material and its associated increased risk for subsequent infection and thromboembolic complications. A 2-center study with broad LVAD deactivation experience highlights these risks.¹⁶ For patients with a previous history of infection during LVAD support (n = 8, 47% of the cohort), 88% experienced a recurrent infection postdecommissioning (n = 7, 41% of the cohort).²⁶ Thus, either a history of or concern for, infection represents a relative contraindication to device decommissioning. With respect to thromboembolism, 3 patients (18%) experienced thromboembolic complications despite therapeutic anticoagulation (n = 2) or aspirin monotherapy (n = 1), which does represent a signal of increased risk in the decommissioning population. Further confounding this issue is a lack of evidence-based approach to anticoagulation/antiplatelet therapy in this population.⁵¹

Pediatric LVAD explant

While most pediatric patients supported with LVADs are bridged to transplant, approximately 9% (7.5%−11.3%) will demonstrate myocardial recovery.^52,53 Unfortunately, information in this population is limited, and the literature primarily consists of registry-based retrospective studies, case series, case reports, and cohort studies.⁵⁴ A recent systematic review analyzed 37 studies to identify the incidence of pediatric VAD-assisted myocardial recovery and the durability of recovery after VAD deactivation.⁵² Recovery rates with explant were reported between 0% and 38.8%, with an overall rate of 8.7% (81/928 patients). The duration of follow-up varied widely with a maximum follow-up of 19.1 years. Overall survival was 85.5% (71/83 patients) and during this follow-up period, 3 patients required VAD reimplants and 2 underwent heart transplants. A number of themes are associated with pediatric recovery: Naturally transient diseases (myocarditis, post-cardiotomy cardiogenic shock), pulsatile VAD, and shorter duration of VAD support before recovery parameters are met. Moreover, previous work has identified age (< 2 years) and body surface area (< 0.53 m²) as indicative of a greater likelihood of recovery.^55,56

Although reports of LVAD deactivation in the pediatric literature provide scant technical details, the dominant approach for VAD discontinuation is a complete explant. In general, less residual material in the chest is considered better and safer, given the expected decades to come after an explant. Management of the ventricular apex includes primary repair with or without the support of cardiopulmonary bypass^57–59 or the use of titanium plugs if available. Regarding outflow management, 1 report describes oversewing the outflow cannula graft, while another describes primary interrupted suture repair of the aorta following complete outflow graft removal.^58–60 Pediatric surgical device decommissioning in an HVAD-supported patient has been described with subsequent complete explant 6 years later.⁶¹ This unique case highlights the point that it is unlikely that a decommissioned device will stay silent for the entire duration of human lives.

Postsupport care and metrics of successful recovery

Despite functional recovery that allows for pump deactivation, these patients remain advanced heart failure patients. Ongoing use of guideline-directed medical therapy with titration to maximally tolerated doses is essential. Furthermore, routine postdiscontinuation assessment of myocardial function is necessary to monitor for HF recurrence. There is no standard approach, but serial echocardiography with progressive lengthening of the intervals between exams is common.⁶² Other strategies include the use of clinical symptom screening, physical examination, biochemical testing, and cardiopulmonary exercise testing.

Current metrics of successful postdeactivation recovery focus on survival free of HF recurrence as evidenced by the need for cardiac transplant or LVAD reimplantation. The RESTAGE-HF (Remission from Stage D Heart Failure) experience represents the most recent prospective data for post-VAD patients and notes 1 and 3-year rates of 90% and 77%, respectively.⁸ Systematic reviews have compiled similar results.^16,63 Moving forward, other elements to be incorporated into composite end-points should include heart failure hospitalizations, thromboembolic events, infection of retained products, and quality of life assessments.

Ongoing issues/future research questions

Currently, no data exist to clearly prescribe the best method for pump deactivation. Lacking a randomized, clinical trial comparing explant with decommissioning, we are left with anecdotes and small case series. Despite the perception that surgical explant appears medieval, most patients are hospitalized for less than 1 week and are up on their feet quickly; they are usually in much better condition than when they underwent implantation. Removing the pump and the driveline is attractive as it avoids the need for anticoagulation and reduces the risk of infection. While decommissioning is certainly easier and less traumatic, the signal for increased thromboembolism and infection is noteworthy.

A number of issues still need to be better understood and include, but are not limited to (1) what is the best form and length of duration of postdeactivation anticoagulation? (2) what is the optimal postdeactivation medical and echocardiographic screening? (3) should patients receive a defibrillator with a postdeactivation ejection fraction of 45%? (4) how do we screen for and proactively identify cases of subclinical LVAD infection, both before and after deactivation? (5) what is the role of the right ventricle and its function on the durability of the deactivated LVAD patient?

Fortunately, the field has coalesced to create the VAD Wean Registry, an international consortium of > 55 LVAD centers that are retroactively and prospectively enrolling recovery/deactivation patients. Indeed, early analysis suggests no difference in survival between the respective methods of deactivation (Snehal Patel, ISHLT 2023). The registry is in the process of enhancing their data acquisition and expanding its numbers. It is hoped that an ongoing analysis will provide more granular data regarding a wide array of outcomes categorized by technique of discontinuation. Quite likely, there will not be a singular dominant strategy but rather a spectrum of approaches tailored to the needs of individual patients.

Supplementary Material

Supplementary data

Supplementary data associated with this article can be found in the online version at doi:10.1016/j.healun.2024.05.005.