Long-term changes in quality of life with LVAD support and after heart transplantation in advanced heart failure

Abstract

Background

Improving quality of life (QOL) is a major therapeutic goal for patients with advanced heart failure undergoing left ventricular assist device (LVAD) therapy or heart transplantation. In Japan, prolonged LVAD support due to donor shortage makes long-term QOL outcomes uncertain.

Objective

To evaluate longitudinal changes in QOL before and after LVAD implantation and heart transplantation, and to examine associations with physical function.

Methods

From 2013 to 2025, 95 patients underwent LVAD implantation at our institution, of whom 23 subsequently received heart transplantation. After excluding 4 patients with incomplete data, 19 were included in the final analysis. QOL was assessed using the Short Form-36 (SF-36) at baseline, during LVAD support, and after transplantation. Physical function was evaluated by grip strength, leg strength, 6-min walk distance (6MWD), and peak oxygen uptake. Physical Component Summary (PCS) and Mental Component Summary (MCS) scores were analyzed.

Results

Baseline QOL was impaired in the physical domain but relatively preserved in the mental domain (PCS 22.9 ± 10.4; MCS 49.1 ± 15.0). During LVAD support (mean 64.1 ± 14.1 months), PCS improved significantly (40.1 ± 7.1, p = 0.008), whereas MCS remained stable (52.4 ± 6.3, p > 0.99). After transplantation (mean 33.6 ± 30.4 months), PCS further improved (46.3 ± 7.9, p = 0.006), while MCS continued to remain stable (53.3 ± 5.8, p = 0.466). Post-transplant PCS showed positive but non-significant correlations with physical function measures.

Conclusion

BTT-LVAD and subsequent heart transplantation markedly improve physical QOL in patients with advanced heart failure, while mental QOL, which is relatively preserved at baseline, remains stable throughout long-term follow-up. These findings suggest that, despite stable overall mental well-being, device- and transplant-specific psychological burdens may persist and should be addressed as part of comprehensive long-term care.

Introduction

Left ventricular assist device (LVAD) implantation has become a well-established therapeutic option for patients with advanced heart failure (AHF) who are candidates for heart transplantation (HTx) [1]. As heart failure progresses, individuals often experience substantial declines in quality of life (QOL) due to dyspnea, fatigue, and reduced functional capacity. LVAD therapy has been shown to alleviate symptoms and improve QOL, either as a bridge to transplantation (BTT) or as long-term support for patients who are ineligible for transplantation [2, 3].

In Japan, the majority of LVAD implantations have historically been performed as BTT because of the limited availability of donor hearts [4]. As a consequence, Japanese patients typically remain on LVAD support for substantially longer periods than those in Western countries, creating a unique clinical environment in which long-term QOL trajectories may differ. Although destination therapy (DT) LVAD implantation became reimbursable in 2021, BTT remains the predominant indication.

While both LVAD support and HTx have demonstrated improvements in survival and symptom burden, there is a lack of detailed data describing longitudinal QOL changes across the full treatment course—from pre-LVAD baseline through prolonged LVAD support to the post-transplant period—particularly within the Japanese clinical context. Previous studies often examined shorter follow-up periods, mixed indications, or cross-sectional snapshots, limiting understanding of long-term patient-reported outcomes.

Therefore, this study aimed to comprehensively evaluate longitudinal changes in both physical and mental components of QOL among patients who underwent LVAD implantation as BTT and subsequently received HTx.

Methods

Study design and population

This retrospective observational study included patients who underwent left ventricular assist device (LVAD) implantation at our institution between October 2013 and June 2025. During this period, 95 patients received LVAD therapy, of whom 83 underwent implantation as a BTT and 12 as DT. As of June 2025, 23 BTT patients had subsequently undergone HTx. Three of these patients died shortly after HTx and one patient had missing data, resulting in a final cohort of 19 patients available for analysis.

Patients were eligible for inclusion if they had completed at least one annual follow-up assessment of health-related quality of life (HRQoL) using the SF-36 during LVAD support and after HTx. Preoperative HRQoL assessments were not obtained in eight patients because of impaired cognitive function or dependence on mechanical circulatory support, such as intra-aortic balloon pump (IABP), extracorporeal LVAD, or mechanical ventilation. As part of the standard multidisciplinary evaluation, all candidates underwent psychiatric assessment prior to LVAD implantation and prior to HTx.

Quality of life assessment

HRQoL was assessed using the Short Form-36 (SF-36), a 36-item questionnaire originally developed for the Medical Outcomes Study that measures eight domains of health status and provides two validated composite scores, the Physical Component Summary (PCS) and the Mental Component Summary (MCS) [5]. The SF-36 has also been widely applied in LVAD and heart transplantation populations, supporting its clinical utility in this setting [6].

Assessments were performed at three predefined timepoints. The first timepoint (T0) corresponded to the baseline evaluation conducted before LVAD implantation. The second timepoint (T1) consisted of annual evaluations performed during LVAD support. The third timepoint (T2) consisted of annual evaluations performed after HTx. PCS and MCS scores derived from these assessments were used for all quantitative analyses.

Assessment of physical function

Physical function was evaluated at two distinct timepoints: 1 year after LVAD implantation and 1 year after heart transplantation. At each evaluation, grip strength and leg extension strength were measured and normalized by body weight. Six-minute walk distance (6MWD) was recorded, and cardiopulmonary exercise testing was performed to determine peak oxygen uptake (peak VO₂). These physical performance measures were used to investigate their associations with PCS and MCS scores.

Statistical analysis

Longitudinal changes in PCS and MCS across the three major timepoints (T0, T1, and T2) were analyzed using paired t-tests and repeated-measures analysis of variance (ANOVA), as appropriate. For patients who underwent multiple annual assessments during LVAD support and after HTx, additional longitudinal trends were evaluated using repeated-measures ANOVA followed by post hoc comparisons.

Associations between HRQoL indices (PCS and MCS) and physical function parameters—grip strength normalized by body weight, leg extension strength normalized by body weight, six-minute walk distance, and peak oxygen uptake—were examined using Spearman’s rank correlation coefficients (ρ) at 1 year after LVAD implantation and at 1 year after HTx. Statistical significance was defined as a two-tailed p-value < 0.05.

Exploratory analyses were also conducted to compare longitudinal QOL trajectories according to pump exchange status (exchange vs. no exchange). Because of the small sample size, these analyses were descriptive in nature and no formal hypothesis testing was performed.

All statistical analyses were conducted using IBM SPSS Statistics version 29.

Results

The mean age at LVAD implantation was 42.1 ± 10.3 years (range 23–58). A total of 19 patients (16 males, 3 females) who subsequently underwent HTx and completed the required QOL evaluations were included in the analysis. The mean duration of LVAD support before HTx was 64.1 ± 14.7 months (Table 1).

Table 1.

Baseline characteristics of the study population

Variables	Value
Demographic Variables
Age at LVAD implantation, years	42.1 ± 10.3
Gender, male (%)	16/19 (84.2)
BMI at LVAD implantation, kg/m2	20.3 ± 2.1
Clinical Variables
Diagnosis leading to implant surgery
DCM, n (%)	14 (73.7)
ICM, n (%)	3 (15.8)
Myocarditis, n (%)	2 (10.5)
LVEF, %	20.0 ± 6.8
INTERMACS profile, n (%)	Profile 1: 5 (35.7)
	Profile 3: 14 (73.7)
Initial LVAD device
HeartMate II, n (%)	19 (100)
Preoperative mechanical circulatory support, n (%)	4 (21.1)
Preoperative intubation	1 (5.3)
Indication (bridge to transplantation), n (%)	19 (100)
Support duration, months	64.1 ± 14.7

This table summarizes the demographic and clinical characteristics of 19 patients who underwent LVAD implantation as a bridge to transplantation (BTT) and were included in the final analysis. Values are presented as mean ± SD or number (percentage), unless otherwise indicated

Abbreviations: DCM dilated cardiomyopathy, ICM ischemic cardiomyopathy, LVEF left ventricular ejection fraction, BMI body mass index, INTERMACS Interagency Registry for Mechanically Assisted Circulatory Support

During the LVAD support period, 16 of 19 patients (84.2%) experienced at least one hospital readmission while awaiting transplantation. The most common causes were device-related complications, infections, and worsening heart failure symptoms.

At baseline (pre-LVAD), the mean PCS and MCS scores were 22.9 ± 10.4 and 49.1 ± 15.0, respectively. During LVAD support, regular SF-36 assessments demonstrated substantial improvement in PCS with relative stability in MCS. In the 0–365-day interval, the mean PCS and MCS were 40.8 ± 10.0 and 52.2 ± 10.3, respectively. In the 366–730-day interval, the corresponding values were 37.9 ± 9.3 and 53.0 ± 6.5. Annual repeated-measures analysis showed that PCS significantly improved during LVAD support (mean 40.1 ± 7.1, p = 0.008), whereas MCS remained stable (mean 52.4 ± 6.3, p > 0.99) (Fig. 1). No additional significant changes were observed in subsequent follow-up evaluations (PCS, p = 0.852; MCS, p = 0.674) (Fig. 2).

Fig. 1.

Changes in SF-36 PCS and MCS scores across major timepoints. SF-36 PCS and MCS scores measured at baseline (T0: pre-LVAD), during LVAD support (T1), and after heart transplantation (T2). Bars represent mean values with SD. Brackets indicate statistically significant differences between timepoints (*p < 0.05). Abbreviations: PCS, Physical Component Summary; MCS, Mental Component Summary

Fig. 2.

Annual changes in SF-36 scores from 1 to 6 years after LVAD implantation. Mean PCS and MCS scores with SD from 1 to 6 years after LVAD implantation. p-values (PCS: p = 0.852; MCS: p = 0.674) reflect the results of repeated-measures analyses assessing changes over time. Higher scores indicate better HRQoL

Following HTx, SF-36 assessments again demonstrated significant gains in PCS and continued stability in MCS. In the 0–365-day post-HTx interval, the mean PCS and MCS were 46.2 ± 10.5 and 54.9 ± 6.0, respectively. Annual analysis showed that PCS significantly improved after HTx compared with the LVAD support phase (46.3 ± 7.9, p = 0.006), while MCS remained stable at 53.3 ± 5.8 (p = 0.466). Across the extended post-transplant follow-up (1–6 years), neither PCS (p = 0.905) nor MCS (p = 0.502) showed significant additional change (Fig. 3). Repeated-measures ANOVA confirmed a significant time effect for PCS but not for MCS.

Fig. 3.

Annual changes in SF-36 scores from 1 to 6 years after heart transplantation. Mean PCS and MCS scores with SD from 1 to 6 years after HTx. p-values (PCS: p = 0.905; MCS: p = 0.502) indicate the absence of significant changes during post-transplant follow-up. SF-36 scores range from 0 to 100, with higher scores reflecting better HRQoL

In a sensitivity analysis using linear mixed-effects models, longitudinal trajectories of PCS and MCS were similar to those observed in the ANOVA, although the fixed effect of time did not reach statistical significance, likely due to the limited sample size and unbalanced number of observations.

Exploratory subgroup analyses examined the potential influence of pump exchange on QOL trajectories. Seven patients (36.8%) underwent pump exchange due to complications such as pump thrombosis or mechanical failure. Both PCS and MCS trends appeared broadly similar between patients with and without pump exchange, and no interpretable differences were observed because of the small sample size (Fig. 4).

Fig. 4.

Long-term PCS and MCS trajectories with and without pump exchange during LVAD support. PCS and MCS trajectories from 1 to 6 years after LVAD implantation, comparing patients with pump exchange (n = 7; solid line, black squares) and without pump exchange (n = 12; dashed line, gray circles). Points represent mean scores with SD. Analyses are descriptive due to the small sample size. Higher scores indicate better HRQoL

At 1 year after HTx, PCS showed moderate, although non-significant, positive correlations with 6MWD (ρ = 0.491, p = 0.125) and Grip/BW (ρ = 0.445, p = 0.170). These associations, while non-significant, may indicate a clinically meaningful relationship between physical function and perceived physical health. No meaningful correlations were observed between MCS and any physical function measure (Table 2).

Table 2.

Correlations between PCS/MCS and physical function

Timepoint	QOL index	Physical measure	Spearman ρ	n	p-value
VAD 1Y	PCS	Grip/BW	0.476	8	0.233
		Leg/BW	0.262	8	0.531
		6MWD	0.226	9	0.559
		PeakVO2	0.359	5	0.553
	MCS	Grip/BW	0.048	8	0.911
		Leg/BW	−0.286	8	0.493
		6MWD	0.100	9	0.797
		PeakVO2	−0.359	5	0.553
HTx 1Y	PCS	Grip/BW	0.445	11	0.170
		Leg/BW	0.336	11	0.312
		6MWD	0.491	11	0.125
		PeakVO2	−0.143	7	0.760
	MCS	Grip/BW	−0.191	11	0.574
		Leg/BW	−0.409	11	0.212
		6MWD	−0.373	11	0.259
		PeakVO2	−0.143	7	0.760

Spearman correlation coefficients (ρ), sample sizes (n), and p-values are presented for associations between HRQoL scores (PCS and MCS) and physical function metrics, including grip strength normalized by body weight (Grip/BW), lower limb strength (Leg/BW), 6-min walk distance (6MWD), and peak oxygen consumption (Peak VO₂). Analyses were performed at 1 year after LVAD implantation and at 1 year after heart transplantation.

Abbreviations: PCS Physical Component Summary, MCS Mental Component Summary

Discussion

Among patients who underwent LVAD implantation as a BTT and subsequently received HTx, longitudinal HRQoL was evaluated in 19 individuals using the SF-36. Before LVAD implantation, the PCS score was markedly impaired (22.9 ± 10.4), whereas the MCS score was relatively preserved (49.1 ± 15.0). During LVAD support, PCS improved significantly to 40.1 ± 7.1, while MCS remained stable at 52.4 ± 6.3. After HTx, PCS further improved to 46.3 ± 7.9, whereas MCS again remained stable (53.3 ± 5.8). These findings demonstrate that prolonged BTT-LVAD support and subsequent HTx are associated with sustained improvements in physical HRQoL, while mental HRQoL—already relatively preserved at baseline—remains stable over long-term follow-up.

Improvements during LVAD support and after HTx

The present data show that LVAD support yields substantial improvement in physical HRQoL. The increase in PCS reflects relief from the profound functional limitations of end-stage heart failure and aligns with prior reports showing improved physical status after LVAD implantation and HTx [7, 8]. In contrast, MCS did not parallel PCS improvement. Because baseline MCS values were close to Japanese population norms, this stability is best interpreted as a ceiling effect rather than a lack of psychological benefit.

Stable MCS after HTx should not be interpreted as a failure of psychological recovery. Instead, it may reflect preserved coping capacity, cultural patterns in self-reporting, and psychological screening performed before LVAD implantation and transplantation. Additionally, generic HRQoL instruments such as the SF-36 may not adequately capture device-specific or transplant-specific psychosocial stressors, including concerns related to driveline dependence, device malfunction, and long-term uncertainty. Thus, while overall mental well-being remained stable, subtle psychosocial burdens may persist and warrant targeted intervention.

Following HTx, further physical recovery was evident, as reflected by an additional increase in PCS. In contrast, MCS again showed no material change, supporting the interpretation that mental HRQoL was preserved rather than improved. This dissociation between physical and mental components is consistent with previous findings in Japanese patients undergoing prolonged BTT-LVAD support [7, 8].

Pump exchange and HRQoL

LVAD therapy is often complicated by adverse events, including frequent rehospitalization and, in some patients, pump exchange [4]. In this study, pump exchange occurred in 36.8% of patients—reflecting the historical reliance on second-generation axial-flow pumps such as HeartMate II during prolonged BTT in Japan. As third-generation centrifugal pumps (e.g., HeartMate 3) now predominate, pump exchange in this cohort should be interpreted as an era-specific phenomenon.

Given the small sample size and variable timing, the pump exchange subgroup analysis was exploratory and descriptive. No definitive conclusions about its causal impact on HRQoL can be drawn, and the findings should not be generalized to current device practice.

Physical function and perceived QOL

At 1 year after HTx, PCS demonstrated positive—but non-significant—associations with objective physical function measures such as grip strength and 6MWD. A dissociation between objective performance and perceived physical health has been reported previously [8, 9] and may reflect limited sample size, residual frailty, or the multifactorial nature of perceived health status. In addition, within the Japanese clinical context, long-term LVAD support often leads to relatively stable physical function during the chronic phase [10], which may also contribute to the observed pattern.

Limitations and future directions

This study has several limitations. First, the final sample size was small (n = 19) and represents a highly selected subgroup of LVAD recipients—those who survived long-term LVAD support, remained eligible for HTx, and were able to complete repeated assessments. Patients who died while on support, remained on destination therapy, or were too ill to participate were not represented. Consequently, the present findings likely reflect a best-case scenario and should not be extrapolated to all LVAD patients.

Second, the retrospective observational design precludes establishing causal relationships between LVAD support, HTx, and HRQoL. Because of the limited sample size, multivariable adjustment for potential confounders such as comorbidities, frailty, organ dysfunction, or rehabilitation intensity was not feasible.

Third, although the SF-36 is a validated and widely used HRQoL instrument, heart failure–specific questionnaires were not employed. Disease-specific symptoms and psychosocial concerns unique to LVAD or transplant recipients may therefore not have been fully captured.

Fourth, data on psychiatric comorbidities, psychological support, and antidepressant use were not systematically available and could not be incorporated into the analysis, potentially affecting the interpretation of MCS stability.

Fifth, all patients were supported with HeartMate II devices. Given the predominance of HeartMate 3 in current practice, findings—particularly those related to pump exchange—should be viewed as reflective of a historical device era.

Finally, the pump exchange analysis was descriptive due to small sample size and heterogeneous timing; no firm conclusions can be drawn regarding its impact on HRQoL.

Future studies should include larger, multicenter cohorts to enhance generalizability and permit multivariable analyses. Incorporating heart failure–specific HRQoL instruments and standardized psychiatric evaluations will allow for more comprehensive assessment of physical and psychological recovery during prolonged LVAD support and after HTx in the Japanese population.

Conclusion

This study demonstrated that LVAD support and subsequent HTx significantly improve the physical component of HRQoL, while the mental component remains relatively stable. These findings underscore the importance of integrating psychosocial support alongside physical rehabilitation in the long-term care of patients with advanced heart failure, particularly in countries such as Japan, where waiting times for transplantation are prolonged.

Authors’ contributions

K.K. designed the study, collected data, performed the statistical analysis, prepared the tables and figures, and drafted the manuscript. T.Y., Y.T., Y.H., Y.N., and M.M. contributed to study conception, clinical supervision, and critical revision of the manuscript. D.T. and K.H. contributed to data collection and patient follow-up. All authors reviewed, edited, and approved the final manuscript.

Funding

This research received no financial support.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was approved by the Nagoya University Clinical Research Review Committee (approval number 2019–0179, approved on August 23, 2019). Informed consent was waived due to the retrospective design, as approved by the ethics committee.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.