ABSTRACT
Background
Conventional mechanical circulatory support offers limited options for patients with small or restrictive left ventricles. Left atrial inflow (LA–Ao) ventricular assist device (VAD) configuration represents an alternative to conventional apical left ventricular assist device (LVAD). However, outcomes data remain limited. To address this, we evaluated patient characteristics, procedural details, and outcomes of LA–Ao VAD configuration in this population.
Methods
A systematic search was conducted to identify case reports and series of patients with small or restrictive left ventricles who underwent LA–Ao LVAD implantation. A total of 21 reports, representing 27 patients, were included. Patient‐level data were extracted for analysis.
Results
Thirteen adult and 14 pediatric patients underwent LA–Ao VAD implantation. Median age was 57 [46–62] years in adults and 3.5 [0.8–9] years in children. Hypertrophic cardiomyopathy and restrictive cardiomyopathy were the predominant diagnoses in both adults and children, accounting for 54% (7/13) and 29% (4/14), and 46% (6/13) and 43% (6/14), respectively. Median left ventricular ejection fraction was 44% [39–53] in adults and 45% [36–61] in children, and median left ventricular end‐diastolic diameter was 42 [35–42] mm in adults and 33 [29–34] mm in children. The most used devices were HeartWare HVAD (46%, 6/13) in adults, and Berlin Heart (64%, 9/14) in children. Overall pulmonary capillary wedge pressure decreased from 23 [19–27] to 7 [0–9] mmHg (p = 0.02), and cardiac index increased from 2.0 [1.3–2.8] to 3.1 [2.7–3.7] L/min/m2 (p = 0.04). In‐hospital mortality included one fatal case (7.7%) in each group. Overall survival was 85% (11/13) in adults at median follow‐up of 6.7 [4.4–19.9] months and 86% (12/14) in children at 2.3 [1.2–8.1] months.
Conclusion
The LA–Ao configuration of VAD is a viable strategy for mechanical circulatory support in patients with small or restrictive left ventricles. It offers effective unloading of the left atrium and improved cardiac index, supporting its role as an alternative to the conventional LVAD approach.
Keywords: heart failure with preserved ejection fraction, hypertrophic cardiomyopathy, mechanical circulatory support, restrictive cardiomyopathy, ventricular assist device
Left atrial inflow ventricular assist device configuration is a viable strategy for mechanical circulatory support in patients with heart failure with preserved ejection fraction. It offers effective unloading of the left atrium and improved cardiac index, supporting its role as an alternative to the conventional apical left ventricular assist device approach.
1. Introduction
Patients with small, hypertrophied, or restrictive left ventricles (LV) usually display restrictive pathophysiology that presents unique challenges for durable mechanical circulatory support (MCS) [1]. These features are frequently encountered in patients with heart failure with preserved ejection fraction (HFpEF), a condition that currently accounts for nearly half of all heart failure cases but remains without well‐established MCS options and is underrepresented in trials evaluating MCS device performance [2, 3]. A recent INTERMACS analysis revealed that only about 2% of continuous‐flow left ventricular assist devices (LVAD) have been implanted in this population, and transplantation is performed in only a small minority [4].
A small, hypertrophied, and stiff LV results in elevated filling pressures transmitted to the left atrium (LA) and pulmonary vasculature. Chronic pressure overload contributes to progressive post‐capillary pulmonary hypertension, often with a combined pre‐ and post‐capillary component in advanced stages and associated biatrial dilation. Due to the small, hypertrophied LV, conventional LVAD implantation with LV apical inflow has a propensity for suction events, chamber collapse, and inadequate filling in these patients. To address this, passive interatrial shunt devices have been explored to lower LA pressure and indirectly reduce pulmonary vascular load, but their clinical benefit remains limited, providing modest hemodynamic improvements without significantly augmenting forward cardiac output (CO) [5].
An alternative surgical approach to apical LVAD placement involves mechanical support with inflow from the LA and outflow to the aorta (LA–Ao), thus bypassing the non‐compliant LV [5, 6]. Recognizing these potential benefits, LA–Ao LVAD support has been applied in patients with HFpEF. To date, no large‐scale studies have been conducted, and the available evidence is limited to case reports and small case series in both adult and pediatric populations. Given the absence of robust data, we aimed to systematically review and synthesize the existing clinical experience with LA–Ao LVAD implantation, focusing on patient characteristics, technical considerations, and outcomes.
2. Methods
2.1. Literature Search Strategy
A systematic search was performed in June 2025 using PubMed, Scopus, and ScienceDirect. To maximize the capture of all relevant studies published to date, the following search terms were used: (HeartMate 3 OR HM3 OR HeartMate II OR HVAD OR HeartWare OR Berlin Heart OR Berlin EXCOR OR PediMag OR LVAD OR Left Ventricular Assist Device) AND (left atrial OR biatrial OR left atrium OR transseptal OR atrial inflow OR alternative) AND (hypertrophic OR restrictive OR amyloid OR sarcoid OR HFpEF OR “heart failure with preserved ejection fraction” OR “preserved ejection fraction”). Additional studies were identified through reference screening of retrieved articles.
In total, 21 reports met the inclusion criteria, encompassing 27 patients. All included studies were case reports or case series (Table S1). A PRISMA flow diagram summarizing the literature search and study selection process is presented in Figure 1.
FIGURE 1.
Preferred reporting items for systematic reviews and meta‐analysis (PRISMA) diagram outlining literature search strategy. [Color figure can be viewed at wileyonlinelibrary.com]
2.2. Selection Criteria
With a paucity of published cohort studies evaluating LA–Ao LVAD implantation, case reports and case series were utilized for analysis that allowed for retrieval of granular patient‐level data. Cases were included in the analysis if authors made a diagnosis of HFpEF or restrictive (RCM), hypertrophic (HCM), or histiocytoid cardiomyopathy, and patients received a durable LVAD with LA inflow placement. Patients who had an LVAD in the LA–Ao configuration for any diagnosis other than the above listed diagnoses, such as LV noncompaction cardiomyopathy, were excluded from the analysis. Non‐original articles, including reviews, editorials, and books, as well as non‐English publications, were excluded.
2.3. Data Extraction
Patient‐level data were extracted from the text, figures, and tables of selected articles and pooled for statistical analysis. Two independent reviewers (D.S. and B.F.) assessed study eligibility and extracted data on patient characteristics and relevant outcomes using a standardized data collection form. Discrepancies were resolved by a third reviewer (A.J.).
2.4. Missing Data
Not all reports included all variables intended for analysis. For missing categorical data, denominators were modified to ensure accurate proportion calculations. Unreported baseline characteristics were considered missing at random if no other reported data in those categories were present. Major adverse events were presumed absent unless explicitly stated.
2.5. Statistical Analysis
Descriptive statistics were used to summarize the data. Categorical variables were expressed as counts and percentages and analyzed with the chi‐squared test. Continuous variables were reported as medians with interquartile ranges [IQR] and compared with the Wilcoxon rank sum test. Hemodynamic parameters before and after device placement were analyzed using the Wilcoxon signed‐rank test. p values < 0.05 were considered statistically significant. All analyses were conducted using R software, version 4.5.0 (R Foundation for Statistical Computing, Vienna, Austria).
3. Results
3.1. Baseline Characteristics
A total of 27 cases were identified with LVAD implantation in the LA–Ao configuration, including 13 adults and 14 pediatric cases.
The median age of adults was 57 years [IQR, 46–62] and the median age of pediatric patients was 3.5 years [0.8–9], with a bimodal distribution of the age curve (Figure 2).
FIGURE 2.
Distribution of patient age showing bimodal shape.
In the adult cohort, 66% (6/9) were male. History of cerebrovascular accident was present in 44% (4/9) of patients. Atrial fibrillation was present in 33% (3/9) of patients, and 56% (5/9) had an implantable cardioverter defibrillator in place.
Among pediatric patients, 54% (7/13) were male. Prior sternotomy was reported in 38% (5/13) of cases. Further details of patient demographics can be found in Table 1.
TABLE 1.
Baseline patient characteristics.
| Overall (n = 27) | Adult (n = 13) | Pediatric (n = 14) | |
|---|---|---|---|
| Age (years), median [IQR] | 57 [46–62] | 3.5 [0.8–9] | |
| Male, % (n/N) | 59 (13/22) | 66 (6/9) | 54 (7/13) |
| History of cerebrovascular accident, % (n/N) | 18 (4/22) | 44 (4/9) | 0 (0/13) |
| Prior sternotomy, % (n/N) | 23 (5/22) | 0 (0/9) | 38 (5/13) |
| Atrial fibrillation, % (n/N) | 14 (3/22) | 33 (3/9) | 0 (0/13) |
| Implantable cardioverter defibrillator, % (n/N) | 27 (6/22) | 56 (5/9) | 7.7 (1/13) |
| History of malignancy, % (n/N) | 9.0 (2/22) | 22 (2/9) | 0 (0/13) |
| History of renal transplant, % (n/N) | 4.5 (1/22) | 0 (0/9) | 7.7 (1/13) |
3.2. Presentation
Among adult patients, HCM (54%, 7/13) and RCM (46%, 6/13) were the most common diagnoses. Within the RCM group, 33.3% (2/6) were attributed to amyloidosis, and sarcoidosis was reported in a single case. The median LV ejection fraction (LVEF) was 44% [39–53]. The median LV end‐diastolic diameter (LVEDD) was 42 [35–42] mm, and the median LV end‐diastolic volume (LVEDV) was 62 [46–77] mL. Valvular abnormalities were present in 38% (3/8) of patients, with two of them (66%) having at least moderate combined tricuspid and mitral regurgitation and one patient (33%) having at least moderate isolated tricuspid regurgitation. Among adults, 88% (7/8) required inotropes, and one patient (11%) was supported with an intra‐aortic balloon pump (IABP) prior to definitive device placement.
In the pediatric patients, RCM represented 46% (6/14) of the cases, while HCM accounted for 29% (4/14). The median LVEF was 45% [36–61], and the LVEDD was 33 [29–34] mm. Valvular abnormalities were identified in 66% (8/12) of patients. Among patients with valve disease, greater‐than‐moderate isolated tricuspid regurgitation was observed in 38% (3/8) of cases, while severe aortic stenosis and at least moderate mitral stenosis were each observed in 25% (2/8) of cases. In the pediatric population, 86% (6/7) of patients required inotropic support, and 31% (4/13) received MCS prior to definitive LVAD placement. Of these, 50% (2/4) were supported with VA‐ECMO, while the remaining two patients were supported by PediMag or Berlin Heart ExCor (Berlin, Germany). The patient that was supported by the Berlin Heart prior to definitive LVAD was initially cannulated in the apical position, with support complicated by persistent LVAD filling deficiencies due to the restrictive physiology, prompting conversion to LA–Ao configuration. Further details of patient presentation and initial management are provided in Table 2.
TABLE 2.
Patient presentation on hospital admission and initial medical management and temporary mechanical circulatory support.
| Diagnosis | Overall (n = 27) | Adult (n = 13) | Pediatric (n = 14) |
|---|---|---|---|
| Hypertrophic cardiomyopathy, % (n/N) | 41 (11/27) | 54 (7/13) | 29 (4/14) |
| Restrictive cardiomyopathy, % (n/N) | 44 (12/27) | 46 (6/13) | 43 (6/14) |
| Amyloidosis | 16.6 (2/12) | 33.3 (2/6) | 0 (0/6) |
| Sarcoidosis | 8.3 (1/12) | 16.7 (1/6) | 0 (0/6) |
| Endomyocardial fibrosis | 8.3 (1/12) | 0 (0/6) | 16.7 (1/6) |
| Unspecified | 66.7 (8/12) | 50 (3/6) | 83.3 (5/6) |
| Congenital – Shones Complex, % (n/N) | 14 (3/27) | 0 (0/13) | 21 (3/14) |
| Histiocytoid cardiomyopathy, % (n/N) | 4 (1/27) | 0 (0/13) | 7.1 (1/14) |
| Echocardiographic Findings | |||
| Left Ventricle Ejection Fraction (%), median [IQR] | 44 [37–57] | 44 [39–53] | 45 [36–61] |
| Left Ventricle End Systolic Diameter (mm), median [IQR] | 24.2 [20.0–30.0] | 25 [23–30] | 22.1 [20–24.2] |
| Left Ventricle End Diastolic Diameter (mm), median [IQR] | 35.0 [32.7–42.0] | 42 [35–42] | 33 [29–34] |
| Left Ventricle End Diastolic Volume (mL), median [IQR] | 62 [46–77] | 62 [46–77] | NA |
| Interventricular Septal Thickness in diastole (mm), median [IQR] | 14 [10–15] | 14.5 [12–16] | NA |
| Valvular Abnormalities, % (n/N) | 55 (11/20) | 38 (3/8) | 66 (8/12) |
| Combined Tricuspid and mitral regurgitation ≥ Moderate | 18 (2/11) | 66 (2/3) | 0 (0/8) |
| Combined Tricuspid Regurgitation and Aortic Stenosis ≥ Moderate | 9.1 (1/11) | 0 (0/3) | 13 (1/8) |
| Isolated Tricuspid Regurgitation ≥ Moderate | 27 (5/11) | 33 (1/3) | 38 (3/8) |
| Isolated Mitral Stenosis ≥ Moderate | 18 (2/11) | 0 (0/3) | 25 (2/8) |
| Isolated Aortic Stenosis—Severe | 18 (3/11) | 0 (0/3) | 25 (2/8) |
| Pharmacologic Support Prior to definitive LVAD Implantation | |||
| Inotropic support, % (n/N) | 87 (13/15) | 88 (7/8) | 86 (6/7) |
| Milrinone | 31 (4/13) | 29 (2/7) | 33 (2/6) |
| Dobutamine | 15 (2/13) | 14 (1/7) | 17 (1/6) |
| Dopamine | 23 (3/13) | 14 (1/7) | 33 (2/6) |
| Epinephrine | 15 (2/13) | 0 (0/7) | 33 (2/6) |
| Unspecified | 46 (6/13) | 57 (4/7) | 33 (2/6) |
| Diuretics, % (n/N) | 47 (7/15) | 63 (5/8) | 29 (2/7) |
| Vasodilators, % (n/N) | 33 (5/15) | 38 (3/8) | 29 (2/7) |
| Mechanical Support prior to definitive LVAD Implantation, % (n/N) | 23 (5/22) | 11 (1/9) | 31 (4/13) |
| VA‐ECMO | 40 (2/5) | 0 (0/1) | 50 (2/4) |
| IABP | 20 (1/5) | 100 (1/1) | 0 (0/4) |
| Pedimag (BiVAD) | 20 (1/5) | 0 (0/1) | 25 (1/4) |
| Centrimag BiVAD → Berlin Heart BiVAD | 20 (1/5) | 0 (0/1) | 25 (1/4) |
3.3. Operative Characteristics
In adults, left‐sided support (LA–Ao configuration) was provided with the HeartWare HVAD (Medtronic, Minneapolis, MN, USA) (46%, 6/13), followed by HeartMate 3 LVAD (Abbott, Abbott Park, IL, USA) (23%, 3/13) and Berlin Heart (7.7%, 1/13).
Twenty‐three percent (3/13) of patients received combined left‐ and right‐sided support with Berlin Heart (Figure 3). The most common access for left heart unloading was the transseptal approach from the right atrium with an intraatrial connection (92%, 12/13) (Figure 4), whereas direct LA access was utilized in one patient (7.7%) (Figure 5).
FIGURE 3.
Left‐ and right‐heart support via Berlin Heart in left atrial–to–aorta and right atrial–to–pulmonary artery configuration. Illustration by CTW. [Color figure can be viewed at wileyonlinelibrary.com]
FIGURE 4.
Left heart unloading via transseptal access from the right atrium with intraatrial connection to the left atrium. Illustration by CTW. [Color figure can be viewed at wileyonlinelibrary.com]
FIGURE 5.
Left heart unloading via direct left atrial cannulation. Illustration by CTW. Modified from [5]. [Color figure can be viewed at wileyonlinelibrary.com]
In pediatric patients, Berlin Heart was the most common definitive device (64%, 9/14), providing left‐sided support in 40% (2/5) and combined left‐ and right‐sided support in 77.8% (7/9) of cases. HeartMate 3 LVAD was employed for left‐sided support in 40% (2/5) of cases. Transseptal access with an intraatrial connection was the most common approach for left side unloading (57%, 8/14) and utilized across all devices. Further details are provided in Table 3.
TABLE 3.
Characteristics of definitive left atrial–to–aorta ventricular assist device.
| Overall (n = 27) | Adult (n = 13) | Pediatric (n = 14) | |
|---|---|---|---|
| Left‐sided only support, % (n/N) | 56 (15/27) | 77 (10/13) | 36 (5/14) |
| Heartware HVAD | 46.7 (7/15) | 60 (6/10) | 20 (1/5) |
| Heartmate 3 | 33.3 (5/15) | 30 (3/10) | 40 (2/5) |
| Berlin Heart | 20 (3/15) | 10 (1/10) | 40 (2/5) |
| Left‐ and right‐sided support, % (n/N) | 44 (12/27) | 23 (3/13) | 64 (9/14) |
| Berlin Heart | 83.3 (10/12) | 100 (3/3) | 77.8 (7/9) |
| PediMag | 8.3 (1/12) | 0 (0/3) | 11.1 (1/9) |
| CentriMag | 8.3 (1/12) | 0 (0/3) | 11.1 (1/9) |
| Approach for left heart unloading | |||
| Transseptal with intraatrial connection, % (n/N) | 74 (20/27) | 92 (12/13) | 57 (8/14) |
| Direct left atrium, % (n/N) | 11 (3/27) | 7.7 (1/13) | 14 (2/14) |
| Left atrial appendage, % (n/N) | 7.4 (2/27) | 0 (0/13) | 14 (2/14) |
| Unspecified, % (n/N) | 7.4 (2/27) | 0 (0/13) | 14 (2/14) |
| Intraatrial connection conduit | |||
| Gore‐Tex PTFE, % (n/N) | 70 (14/20) | 58 (7/12) | 88 (7/8) |
| Intergard Woven, % (n/N) | 5.0 (1/20) | 8.3 (1/12) | 0 (0/8) |
| Unspecified, % (n/N) | 25 (5/20) | 33 (4/12) | 13 (1/8) |
3.4. Postoperative Complications
In the adult subgroup, acute kidney injury occurred in 31% (4/13) and bleeding in 23% (3/13). Right heart failure (RHF) developed in two patients (15%); one patient was managed with inotropic support, while the other required temporary right ventricular assist device (RVAD) support. A single stroke of ischemic subtype was observed (7.7%, 1/13).
In the pediatric population, stroke was the most common complication (21%, 3/14), including one ischemic and two hemorrhagic events. RHF and bleeding each occurred in 14% (2/14) of cases. Among patients with RHF, one was managed with inotropic support, while the other, already on left‐sided Berlin Heart support, was converted to combined left‐ and right‐sided Berlin Heart support. Additional details of postoperative complications are provided in Table 4.
TABLE 4.
Postoperative complications.
| Overall (n = 27) | Adult (n = 13) | Pediatric (n = 14) | p value | |
|---|---|---|---|---|
| Bleeding, % (n/N) | 19 (5/27) | 23 (3/13) | 14 (2/14) | 0.56 |
| Acute kidney injury, % (n/N) | 19 (5/27) | 31 (4/13) | 7.1 (1/14) | 0.11 |
| Right heart failure, % (n/N) | 15 (4/27) | 15 (2/13) | 14 (2/14) | 0.94 |
| Inotropes | 50 (2/4) | 50 (1/2) | 50 (1/2) | 1.00 |
| Temporary RVAD | 25 (1/4) | 50 (1/2) | 0 (0/2) | 0.25 |
| Durable RVAD | 25 (1/4) | 0 (0/2) | 50 (1/2) | 0.25 |
| Stroke, % (n/N) | 15 (4/27) | 7.7 (1/13) | 21 (3/14) | 0.32 |
| Ischemic | 50 (2/4) | 100 (1/1) | 33 (1/3) | |
| Hemorrhagic | 50 (2/4) | 0 (0/1) | 67 (2/3) | |
| Infection, unspecified, % (n/N) | 11 (3/27) | 7.7 (1/13) | 14 (2/14) | 0.59 |
| Pneumonia, % (n/N) | 3.7 (1/27) | 7.7 (1/13) | 0 (0/14) | 0.29 |
| Sepsis, % (n/N) | 3.7 (1/27) | 0 (0/13) | 7.1 (1/14) | 0.33 |
| Seizure, % (n/N) | 3.7 (1/27) | 0 (0/13) | 7.1 (1/14) | 0.34 |
3.5. Hemodynamic Changes After LA‐Ao VAD Placement
For all patients, pulmonary capillary wedge pressure (PCWP) significantly decreased from 23 [19–27] to 7 [0–9] mmHg (p = 0.02), and cardiac index increased from 2.0 [1.3–2.8] to 3.1 [2.7–3.7] L/min/m2 (p = 0.04). Pulmonary artery systolic pressure significantly decreased from 43 [40–48] to 27 [25–34] mmHg (p = 0.02), and diastolic pressure decreased from 22 [20–32] to 11 [8–15] mmHg (p = 0.02). Pulmonary vascular resistance also declined from 5.5 [4.4–9.5] to 2.8 [2.7–3.3] Wood units, although this reduction did not reach statistical significance (p = 0.15). Additional hemodynamic changes associated with LA–Ao VAD placement are listed in Table 5.
TABLE 5.
Hemodynamic changes after left atrial–to–aorta ventricular assist device.
| n = 27 | p value | ||
|---|---|---|---|
| Before | After | ||
| Pulmonary capillary wedge pressure (mmHg), median [IQR] | 23 [19–27] | 7 [0–9] | 0.02 |
| Pulmonary vascular resistance (Wood), median [IQR] | 5.5 [4.4–9.5] | 2.8 [2.7–3.3] | 0.15 |
| Systemic systolic blood pressure (mmHg), median [IQR] | 92 [86–94] | 118 [106–130] | 1.00 |
| Systemic diastolic blood pressure (mmHg), median [IQR] | 64 [52–66] | 80 [74–85] | 0.50 |
| Cardiac index (L/min), median [IQR] | 2.0 [1.3–2.8] | 3.1 [2.7–3.7] | 0.04 |
| Central venous pressure (mmHg), median [IQR] | 14 [10–18] | 10 [8–16] | 0.73 |
| Pulmonary artery systolic pressure (mmHg) median [IQR] | 43 [40–48] | 27 [25–34] | 0.02 |
| Pulmonary artery diastolic pressure (mmHg), median [IQR] | 22 [20–32] | 11 [8–15] | 0.02 |
3.6. Outcomes
In‐hospital/30‐day mortality was similar between groups, with a single fatal case in each. Heart transplant listing rates were comparable between adults (85%, 11/13) and pediatric patients (86%, 12/14; p = 0.94). Transplantation rate was higher among adults without reaching statistical significance (adults: 64%, 7/11 vs. pediatrics: 50%, 6/12; p = 0.51), as was time to transplantation (adults: 4.0 [3.3–4.0] months; pediatrics: 2.5 [1.2–6.0] months; p = 0.35). Overall survival was 85% (11/13) in adults at a median follow‐up of 4.3 [3.3–12.9] months and 86% (12/14) in pediatric patients at a median follow‐up of 1.4 [1.0–3.6] months. Additional details regarding post LA–Ao VAD placement may be found in Table 6.
TABLE 6.
Patient outcomes following left atrial–to–aorta ventricular assist device.
| Overall (n = 27) | Adult (n = 13) | Pediatric (n = 14) | p value | |
|---|---|---|---|---|
| Listed for heart transplant, % (n/N) | 85 (23/27) | 85 (11/13) | 86 (12/14) | 0.94 |
| Transplanted, % (n/N) | 57 (13/23) | 64 (7/11) | 50 (6/12) | 0.51 |
| Time on LVAD support (months), median [IQR] | 3.6 [2.0–4.0] | 4.0 [3.3–4.0] | 2.5 [1.2–6.0] | 0.35 |
| Duration of LVAD support (months), median [IQR] | 3.0 [1.3–5.0] | 4.0 [2.8–10.8] | 1.4 [1.0–3.6] | 0.03 |
| In hospital/30‐day mortality, % (n/N) | 7.4 (2/27) | 7.7 (1/13) | 7.1 (1/14) | 0.96 |
| Follow‐up time (months), median [IQR] | 3.6 [1.3–8.1] | 4.3 [3.3–12.9] | 1.4 [1.0–3.6] | 0.03 |
| Pump exchange due to pump thrombosis, % (n/N) | 7.4 (2/27) | 15 (2/13) | 0 (0/14) | 0.13 |
| Driveline infection, % (n/N) | 3.7 (1/27) | 7.7 (1/13) | 0 (0/14) | 0.29 |
| Overall survival, % (n/N) | 85 (23/27) | 85 (11/13) | 86 (12/14) | 0.94 |
4. Discussion
This systematic review evaluates the available published experience with durable LVAD support in the LA‐Ao configuration among patients with small or restrictive left ventricles. This approach provided a technically feasible option for device placement in this population, resulted in significant improvements in hemodynamic parameters, and was associated with an acceptable complication profile. These outcomes enabled most patients to meet criteria for heart transplant listing, with acceptable transplantation rates and optimal short‐term overall survival.
In patients with HFpEF, alternative inflow strategies have garnered attention in response to the technical limitations posed by conventional apical cannulation. LV cavity dimensions have long been recognized as a major predictor of outcomes in conventional LVAD therapy, with an LVEDD of 50 mm often considered a threshold for poor filling, inflow obstruction, and high procedural risk [4, 7]. Consistent with this, an INTERMACS analysis of patients with HCM showed that LVEDD below 50 mm was associated with significantly worse one‐year survival compared with patients above this threshold (35% vs. 88%) [8]. Although apical cannulation may remain appropriate in the “burn‐out” form of HCM, where progressive ventricular dilation allows safer apical access, all adult patients in the present analysis had LVEDD values below 5.0 cm (median 42 mm). Accordingly, the LA inflow site was favored over the LV apex in this population. Utilizing the LA for inflow cannulation in this population effectively bypasses the restrictive LV cavity and enables stable circulatory support without the need for apical reconstruction [9]. Additionally, two cases included in our analysis describe conversion from apical cannulation to LA inflow due to persistent filling difficulties or suction events [10, 11]. These experiences underscore the utility of the LA–Ao configuration, not only as a primary strategy but also as a salvage approach in anatomically prohibitive cases.
Hemodynamic improvements were substantial and clinically meaningful. Patients demonstrated significant reductions in PCWP and pulmonary artery pressures, as well as increased cardiac index following implantation. These changes represent more than physiological endpoints, as they enabled patients to meet criteria for heart transplant listing [12]. In many cases, elevated pulmonary pressures and high transpulmonary gradients normalized within days to weeks after implantation, restoring candidacy in patients who were previously ineligible for transplant.
The predominance of continuous‐flow LVADs in the adult population likely contributed to procedural consistency and reduced device‐related morbidity, while pediatric patients were supported with paracorporeal devices, most commonly Berlin Heart, which remains the standard of care in this population due to its versatility in smaller patients and ease of anatomical accommodation [13, 14].
The transseptal approach represented the predominant method for LA unloading compared with direct LA or LA appendage cannulation in both adults and pediatric populations. Importantly, in patients requiring combined left‐ and right‐sided support, the transseptal approach did not interfere with the insertion or function of RA inflow cannulas [11, 15]. In these cases, two Berlin Heart pumps were employed, maintaining adequate decompression of both atria and contributing to balanced biventricular support.
Among adults, acute kidney injury was the most frequently observed complication, possibly associated with systemic manifestation of heart failure. RHF, while present, occurred in 15% of patients. This rate appears lower than the 24% incidence of RHF reported among conventional apical LVAD recipients in a recent INTERMACS analysis [16]. These findings may suggest a potential hemodynamic advantage of LA inflow, possibly related to reduced RV afterload without direct alteration of ventricular geometry or interdependence. In contrast to adults (23%), a significantly higher proportion of pediatric patients (64%) required combined left‐ and right‐sided support. The etiology of this difference remains unclear; however, it may be attributable to the higher propensity for genetic causes of cardiomyopathy in pediatric patients [17]. This higher prevalence of RHF in children underscores the need for early, tailored intervention. Management strategies for RHF included inotropic support, temporary RVAD implantation, and durable right‐sided support with the Berlin Heart, in which an existing left‐sided Berlin Heart configuration was converted to biventricular support through additional right‐sided cannulation. While no inhaled pulmonary vasodilators were reported among the analyzed cases, inhaled nitric oxide and epoprostenol have been described in the literature as adjunctive therapies for acute RHF because of their selective pulmonary vasodilatory effects [18, 19].
LA–Ao VAD placement represents the currently available MCS strategy for patients with HFpEF whose LV dimensions preclude apical cannulation. Although the reported outcomes appear favorable, this approach is not ideal. The risk of intracavitary ventricular thrombosis associated with complete LV bypass and subsequent blood stasis remains a major concern [5, 7]. Fried et al. demonstrated that, among patients supported with LVAD in conventional configuration, LV thrombosis was associated with a significantly higher risk of stroke and short‐term mortality compared with patients without LV thrombosis [20]. In the pooled data, LV thrombosis was not observed, possibly reflecting strict monitoring of anticoagulation regimen and pump speed adjustment to allow periodic aortic valve opening [12, 21]. Nevertheless, stroke was observed in the analyzed cases, with hemorrhagic events predominating among pediatric patients. Although postoperative elevation in systemic blood pressure is a recognized concern for hemorrhagic complications in patients receiving LVAD support [22], blood pressure data were not available for the patients with hemorrhagic stroke. As such, suboptimal postoperative blood pressure management could be a contributing factor, and close monitoring and optimization of blood pressure should be emphasized.
Postoperative anticoagulation regimens were inconsistently reported across the pooled case reports, limiting quantitative analysis. When reported, the target INR was generally 2.5–3.5, with strategies varying by device type and institutional protocol. HeartMate 3 LVAD patients were commonly managed with warfarin, whereas HeartWare HVAD and Berlin Heart patients more often included combined antiplatelet therapy with either warfarin or direct oral anticoagulants.
An emerging alternative for patients with HFpEF is the left atrial assist device (LAAD), which is positioned at the level of the mitral valve and provides circulatory support in series rather than in parallel [5]. By maintaining physiologic arterial pulsatility, this configuration may reduce the risk of blood stagnation and subsequent LV thrombus formation. Although this technology remains in the experimental stage, early in vivo results have demonstrated promising hemodynamic results [23].
5. Limitations
This study is limited by its retrospective design and reliance on published case reports and series, introducing reporting and publication bias. Small sample size and incomplete echocardiographic and hemodynamic data restricted comprehensive analysis. Further prospective, multicenter studies with standardized data collection are needed to better define patient selection, device strategy, and outcomes in this understudied population.
6. Conclusion
The LA–Ao configuration of VAD is a viable strategy for MCS in patients with small or restrictive left ventricles. It offers targeted unloading of the left atrium and favorable hemodynamics, supporting its potential role as an alternative to the conventional LVAD approach.
Author Contributions
Aryana J. Jones: Data analysis/interpretation, Writing – original draft, Writing – review and editing. Dylan Smoot: Data extraction, Statistics. Fady Bassem Fayek: Data extraction. Daler Rahimov: Writing – review and editing. Chelsey T. Wood: Figure Artist. Taufiek Konrad Rajab: Supervision, Writing – review and editing. Howard J. Eisen, Rene Alvarez, Eduardo Rame, Keshava Rajagopal, Charles W. Hoopes: Supervision. Vakhtang Tchantchaleishvili: Conceptualization, Supervision, Writing – review and editing.
Disclosure
Vakhtang Tchantchaleishvili serves on the advisory board of EvaHeart.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Table S1: Articles included in systematic review.
References
- 1. Buono M. G. D., Buckley L., and Abbate A., “Primary and Secondary Diastolic Dysfunction in Heart Failure With Preserved Ejection Fraction,” American Journal of Cardiology 122, no. 9 (2018): 1578–1587, 10.1016/j.amjcard.2018.07.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Mehra M. R., Uriel N., Naka Y., et al., “A Fully Magnetically Levitated Left Ventricular Assist Device — Final Report,” New England Journal of Medicine 380, no. 17 (2019): 1618–1627, 10.1056/NEJMoa1900486. [DOI] [PubMed] [Google Scholar]
- 3. Rogers J. G., Pagani F. D., Tatooles A. J., et al., “Intrapericardial Left Ventricular Assist Device for Advanced Heart Failure,” New England Journal of Medicine 376, no. 5 (2017): 451–460, 10.1056/NEJMoa1602954. [DOI] [PubMed] [Google Scholar]
- 4. Patel S. R., Saeed O., Naftel D., et al., “Outcomes of Restrictive and Hypertrophic Cardiomyopathies After LVAD: An INTERMACS Analysis,” Journal of Cardiac Failure 23, no. 12 (2017): 859–867, 10.1016/j.cardfail.2017.09.011. [DOI] [PubMed] [Google Scholar]
- 5. Gordon J. S., Blazoski C. M., Wood C. T., et al., “Mechanical and Interventional Support for Heart Failure With Preserved Ejection Fraction: A Review,” Artificial Organs 46, no. 11 (2022): 2109–2117, 10.1111/aor.14275. [DOI] [PubMed] [Google Scholar]
- 6. Brahmbhatt D. H., Noly P. E., Fung N. L., Billia F., Rao V., and Badiwala M. V., “Trans‐Atrial Inflow Cannula Configuration for Durable HeartWare Left Ventricular Assist Device in Hypertrophic Cardiomyopathy: A Case Series and a Word of Caution,” JHLT Open 9 (2025): 100265, 10.1016/j.jhlto.2025.100265. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Grupper A., Park S. J., Pereira N. L., et al., “Role of Ventricular Assist Therapy for Patients With Heart Failure and Restrictive Physiology: Improving Outcomes for a Lethal Disease,” Journal of Heart and Lung Transplantation 34, no. 8 (2015): 1042–1049, 10.1016/j.healun.2015.03.012. [DOI] [PubMed] [Google Scholar]
- 8. Liang L. W., Lumish H. S., Sewanan L. R., et al., “Advanced Heart Failure Therapies for Hypertrophic Cardiomyopathy,” JACC Heart Failure 11, no. 11 (2023): 1473–1480, 10.1016/j.jchf.2023.07.004. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Lanmueller P., Eulert‐Grehn J. J., Schoenrath F., et al., “Durable Mechanical Circulatory Support in Patients With Heart Failure With Preserved Ejection Fraction,” Interactive Cardiovascular and Thoracic Surgery 33, no. 4 (2021): 628–630, 10.1093/icvts/ivab144. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Magnetta D. A., Reichhold A., Thrush P. T., Monge M., Webster G., and Joong A., “Biventricular Assist Device Support for Intractable Arrhythmias From Histiocytoid Cardiomyopathy,” ASAIO Journal 68, no. 11 (2022): e207–e210, 10.1097/MAT.0000000000001715. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Marey G. M., Said S. M., Ameduri R., et al., “Berlin Excor Cannulation of Left Atrial Appendage in Left Ventricular Restrictive Physiology: A Novel Bailout Strategy,” ASAIO Journal 67, no. 9 (2021): e157–e159, 10.1097/MAT.0000000000001330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Kiamanesh O., Rankin K., Billia F., and Badiwala M. V., “Left Ventricular Assist Device With a Left Atrial Inflow Cannula for Hypertrophic Cardiomyopathy,” JACC Case Reports 2, no. 13 (2020): 2090–2094, 10.1016/j.jaccas.2020.10.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Jawaid O., Salerno C., and Ravichandran A., “Left Ventricular Assist Device and the Current State of the Art: HeartMate 3 at 5 Years,” Heart Failure Clinics 20, no. 1 (2024): 83–89, 10.1016/j.hfc.2023.05.005. [DOI] [PubMed] [Google Scholar]
- 14. Almond C. S., Morales D. L., Blackstone E. H., et al., “Berlin Heart EXCOR Pediatric Ventricular Assist Device for Bridge to Heart Transplantation in US Children,” Circulation 127, no. 16 (2013): 1702–1711, 10.1161/CIRCULATIONAHA.112.000685. [DOI] [PubMed] [Google Scholar]
- 15. Kari F. A., Hörer J., and Michel S., “Biventricular Assist Device Implant Using Biatrial Cannulation for Restrictive Cardiomyopathy,” Multimedia Manual of Cardiothoracic Surgery (2024): 2024, 10.1510/mmcts.2024.039. [DOI] [PubMed] [Google Scholar]
- 16. Kapelios C. J., Lund L. H., Wever‐Pinzon O., et al., “Right Heart Failure Following Left Ventricular Device Implantation: Natural History, Risk Factors, and Outcomes: An Analysis of the STS INTERMACS Database,” Circulation. Heart Failure 15, no. 6 (2022): e008706, 10.1161/CIRCHEARTFAILURE.121.008706. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Das B., Deshpande S., Akam‐Venkata J., Shakti D., Moskowitz W., and Lipshultz S. E., “Heart Failure With Preserved Ejection Fraction in Children,” Pediatric Cardiology 44, no. 3 (2023): 513–529, 10.1007/s00246-022-02960-7. [DOI] [PubMed] [Google Scholar]
- 18. Motoji Y., Kitamura T., Mishima T., et al., “Inhaled Nitric Oxide Therapy Is Effective in Improving Right Ventricular Function in Patients Receiving ECPELLA Support,” General Thoracic and Cardiovascular Surgery 74, no. 1 (2026): 28–37, 10.1007/s11748-025-02181-8. [DOI] [PubMed] [Google Scholar]
- 19. Ghadimi K., Cappiello J. L., Wright M. C., et al., “Inhaled Epoprostenol Compared With Nitric Oxide for Right Ventricular Support After Major Cardiac Surgery,” Circulation 148, no. 17 (2023): 1316–1329, 10.1161/CIRCULATIONAHA.122.062464. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Fried J. A., Lumish H., Zuver A. M., et al., “Presence of Left Atrial or Left Ventricular Thrombus at the Time of CF‐LVAD Implantation Is Associated With Increased Post‐Operative Risk of Stroke or Death,” Journal of Heart and Lung Transplantation 38, no. 4 (2019): S68–S69, 10.1016/j.healun.2019.01.156. [DOI] [Google Scholar]
- 21. Maeda K., Nasirov T., Rosenthal D. N., and Dykes J. C., “An Alternative Approach by HeartWare Ventricular Assist Device in Hypertrophic Cardiomyopathy,” Annals of Thoracic Surgery 106, no. 5 (2018): e231–e232, 10.1016/j.athoracsur.2018.04.065. [DOI] [PubMed] [Google Scholar]
- 22. Willey J. Z., Boehme A. K., Castagna F., et al., “Hypertension and Stroke in Patients With Left Ventricular Assist Devices (LVADs),” Current Hypertension Reports 18, no. 2 (2016): 12, 10.1007/s11906-015-0618-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Miyagi C., Kuroda T., Kuban B. D., et al., “A Less‐Invasive Left Atrial Assist Device Concept for Diastolic Heart Failure: First in Vitro and in Vivo Assessment,” JTCVS Open 21 (2024): 180–190, 10.1016/j.xjon.2024.08.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1: Articles included in systematic review.