Abstract
Background
The quality-adjusted life year (QALY) measures disease burden and treatment, combining overall survival and health-related quality of life (HRQOL). We estimated QALYs in three groups of older patients (60–80 years) with heart failure (HF) who underwent heart transplantation (HT, with pre-transplant mechanical circulatory support [HT MCS] or HT without pre-transplant MCS [HT Non-MCS]) or long-term MCS (destination therapy). We also identified factors associated with gains in QALYs through 24 months follow-up.
Methods
Of 393 eligible patients enrolled (10/1/15–12/31/18) at 13 U.S. sites, 161 underwent HT (n=68 HT MCS, n=93 HT Non-MCS) and 144 underwent long-term MCS. Survival and HRQOL data were collected through 24 months. QALY health utilities were based on patient self-report of EQ-5D-3L dimensions. Mean-restricted QALYs were compared among groups using generalized linear models.
Results
For the entire cohort, mean age in years closest to surgery was 67 (standard deviation, SD: 4.7), 78% were male, and 83% were White. By 18 months post-surgery, sustained significant differences in adjusted average+SD QALYs emerged across groups, with the HT Non-MCS group having the highest average QALYs (24-month window: HT Non-MCS=22.58+1.1, HT MCS=19.53+1.33, Long-term MCS=19.49+1.3, p=0.003). At 24 months post-operatively, a lower gain in QALYs was associated with HT MCS, long-term MCS, a lower pre-operative LVEF, NYHA class III or IV before surgery, and an ischemic or other etiology of HF.
Conclusions
Determination of QALYs may provide important information for policy makers and clinicians to consider regarding benefits of HT and long-term MCS as treatment options for older patients with HF.
Keywords: quality-adjusted life years, mechanical circulatory support, heart transplantation, quality of life, survival
INTRODUCTION
Heart failure (HF) affects approximately 10% of U.S. adults age 65 years or older.1 For patients with advanced HF, surgical treatment options include either heart transplantation (HT) or long-term mechanical circulatory support (MCS) implantation, if ineligible for HT (i.e., destination therapy [DT]). Intended to prolong overall survival and improve health-related quality of life (HRQOL),2–5 such surgical procedures may impact these endpoints in opposite directions.6, 7 Conceivably, overall survival may impact HRQOL over time following these advanced surgical therapies, while, in turn, HRQOL history may be associated with a prolonged or shortened overall survival. Reflecting on the synergistic relationship of overall survival and HRQOL, quality-adjusted life-years (QALYs) represent a simple and directly interpretable measure of the burden of disease and its treatment that combines survival and HRQOL into a single number.8
Little is known about the distribution of QALYs and risk factors associated with lower QALY gains among older patients with HF who undergo advanced cardiac surgical therapies. Research pertaining to QALYs after HT or MCS implantation is typically focused on adults of all ages, using existing datasets, and often using simulated modeling. Long et al.9 demonstrated that HT and long-term MCS improve life expectancy. Yet they also noted that lower HRQOL (i.e., health utility scores), based on post-procedure adverse events, resulted in a decrement in QALY gains, especially after long-term MCS implantation. Other literature demonstrates more improvement in QALYs with LVAD implantation (frequently long-term MCS) versus medical therapy, often due to survival gain and potentially HRQOL gain.10–14 HRQOL gain was more uncertain (and often identified as a study limitation) in these articles, because health utility scores were typically based on indirect measures (e.g., post-procedure adverse events and NYHA class) rather than direct measurement (i.e., patient self-report of HRQOL). Thus, important gaps in our knowledge of QALYs after advanced cardiac surgical therapies include a paucity of studies in older patients, wherein there may be age-based discrimination, yet who are undergoing these therapies on a more frequent basis, as compared to earlier eras;2, 15 a paucity of studies comparing QALYs among advanced cardiac surgical therapies, and lack of direct assessment of health utilities.
A valuable context in which these gaps in knowledge can be addressed is that conferred by our recently completed prospective, observational, longitudinal, multi-site NIH/NIA-funded SUSTAIN-IT study (Sustaining Quality of Life of the Aged: Heart Transplant or Mechanical Support? R01AG047416). SUSTAIN-IT participants were older (60–80 years) patients with advanced HF who underwent HT (with or without durable pre-transplant MCS) or long-term MCS.16 This report from SUSTAIN-IT aimed to: (1) estimate the group-specific QALY distribution and compare it across the three surgical therapy groups from pre- through 24 months post-operatively; and (2) identify factors associated with gains in QALYs.
MATERIALS and METHODS
Sites and Sample
Patients were recruited from 13 U.S. sites with HT and MCS programs. SUSTAIN-IT inclusion criteria required having advanced HF, being 60–80 years of age, English speaking, either awaiting HT (listed with the United Network for Organ Sharing) or being evaluated or scheduled for primary long-term MCS implantation, and being able to provide written informed consent.16 Patients underwent one of the following surgical procedures: (1) HT without pre-transplant durable MCS (HT Non-MCS), (2) HT with pre-transplant durable MCS (HT MCS), or (3) long-term MCS, if ineligible for HT. HT Patients in the HT Non-MCS group who received MCS after enrollment (n=11) remained in the HT Non-MCS group, as per initial study group assignment. MCS patients had continuous flow FDA-approved or investigational left ventricular assist devices (LVADs). Our study is compliant with the International Society for Heart and Lung Transplantation Ethics Statement. All sites received Institutional Review Board approval, and participants provided written informed consent prior to study participation.
Measures
QALYs were derived from survival data and health utility scores, combined into a single score. For health utility scores, “0” represents death and “1” represents perfect health. Importantly, health utility scores were calculated from patient self-report of HRQOL over time using the five generic EQ-5D-3L dimensions: Self-care, Usual activities, Mobility, Pain/Discomfort, and Anxiety/Depression, each of which have three response levels: (no problems, some problems, and extreme problems).17, 18 Demographic data (e.g., age, sex, race, marital status, and education) and clinical data (e.g., etiology of HF, comorbidities, and post-operative adverse events) were also collected.
Procedures
Survival and EQ-5D-3L data were either collected by sites via medical records review and patient completion of EQ-5D-3L surveys, respectively, or securely downloaded from The Society of Thoracic Surgeons Interagency Registry for Mechanically Assisted Circulatory Support (STS Intermacs). Data were collected at baseline (prior to HT or long-term MCS surgery) and post-operatively at 3, 6, 12, 18, and 24 month follow-up. Post-operative windows for HRQOL data collection were 3 months+30 days, and 6, 12, and 18 months+60 days. The 24-month window for data collection was extended to 30 months, due to data collection challenges during the COVID-19 pandemic. Long-term MCS group baseline assessments were administered after being considered and/or scheduled for long-term MCS implantation. Baseline assessments for both HT candidate groups were administered after listing for HT either while on MCS (HT MCS group) or while on medical therapy (HT Non-MCS group). In order to have pre-HT candidate baseline data as close to surgery as possible, we repeatedly collected baseline information every 6 months until HT and immediately pre-HT, if possible. Baseline data closest to HT surgery and long-term MCS implant were used for this report.
Demographic characteristics (e.g., age, sex, race, and marital status) and clinical data (e.g., etiology of HF, NYHA class, co-morbidities, and post-operative hospitalizations and adverse events) were also collected by sites or downloaded securely from STS Intermacs. Adverse events included “any event” after the index operation. MCS adverse events used Intermacs definitions (per the Intermacs Users’ Guide version date 3/28/14), which was the version available at the time of study start-up. HT adverse event definitions were per the latest HT guidelines/published articles.
Statistical Analyses
Data summaries included mean±standard deviation (SD), median (first [Q1]) and third [Q3] quartile), counts, and percentages, depending on the distribution. Comparisons by surgical therapy group were based on the 1-way ANOVA (if normally-distributed) and Wilcoxon’s rank sum test (in case of non-normality) for continuous measures, and the Chi-square or Fisher’s exact test (if entry counts were <5) for categorical measures.
EQ-5D-3L health utility scores, derived from the five EQ-5D-3L dimensions, were converted to a single summary score by applying a formula that attaches values to each of the levels in each dimension, referred to as health states, per the EQ-5D-3L User Guide.19 There are 245 possible health states. At each planned study time period while still in-study, participant missing health utility scores were singly-imputed using nearest neighbor matching based on (in decreasing order of importance): surgical therapy group, age ([60–70] or [71–80] years), NYHA class at baseline, and whenever possible, study site. If multiple matches were found, we used their average age for imputation. Box plots displaying health utility scores by surgical strategy were then created.
The Kaplan-Meier estimator was used to summarize overall survival by group, and Cox regression was used to obtain hazard ratio estimates of between group differences. Individual-level QALYs were obtained as the sum of the utility-weighted time spent in each EQ-5D-3L health state throughout follow-up.
Unadjusted and covariate-adjusted of mean-restricted QALYs were performed to compare surgical therapy groups using generalized linear models via PROC RMSTREG in SAS v 9.4 (SAS Institute, Cary NC). After adjusting for age, sex, surgical strategy, and the EQ-5D-3L baseline visual analog scale (VAS) score, other covariates were selected based on medical importance and/or statistical significance (univariable model p<0.1) and included insurance type, left ventricular ejection fraction, NYHA class, number of co-morbidities, diabetes and HF etiology. Throughout, a two-sided 5% alpha level was employed.
RESULTS
Patient Characteristics and HRQOL
Of 393 eligible patients enrolled (10/1/15–12/31/18), 161 underwent HT (n=68 HT MCS, n=93 HT Non-MCS) and 144 underwent long-term MCS; 209 patients completed the study at 24 months (47 HT MCS, 86 HT Non-MCS, and 76 long-term MCS) (Figure 1). Reasons eligible patients did not undergo advanced surgical therapies are shown in Fig S1, and reasons for post-surgical study non-completion for the three groups are detailed in Figure 1. The majority of patients in each group were male, white, married, and had more than a high school education (Table 1). Patients in the long-term MCS group were significantly older than patients in both HT groups and had a higher pre-surgical NYHA class and more co-morbidities (Table 1). At baseline, the long-term MCS group had significantly more diabetes and chronic kidney disease than the HT groups (Table 1). Very few patients (n=5) had temporary MCS in place immediately prior to surgery; almost 40% of patients in the long-term MCS group and HT Non-MCS group had intra-aortic balloon pumps in place within 48 hours of surgery without significant differences between groups, while no patients were on extracorporeal membrane oxygenation before surgery (Table 1). Patients in the long-term MCS group, HT MCS group, and HT Non-MCS group (who required MCS after enrollment) were implanted with either HeartMate II, HeartMate 3, or HeartWare LVADs (Table 2). Through 24 months follow-up, the long-term MCS group had more adverse events (i.e., major bleeding, cardiac arrhythmias, neurological dysfunction, and psychiatric episodes), and unplanned re-hospitalizations (Table 2) compared to both HT groups.
Figure 1.
Flow Diagram for Patients who underwent Long-term Mechanical Circulatory Support (MCS) or Heart Transplantation (HT) with or without pre-transplant MCS.
Table 1.
Patient Characteristics at Baseline
| Variable | N Observed Total cohort (per group) |
Total Cohort (N=305) |
Long-Term MCS (n=144) |
HT MCS (n=68) |
HT Non-MCS (n=93) |
p-value |
|---|---|---|---|---|---|---|
| Demographic Characteristics | ||||||
| Age closest to surgery, years, mean (SD:) | 305(144,68,93) | 66.9 (SD:4.7) | 69.0 (SD: 5.2) | 65.4 (SD: 3.2) | 64.8 (SD: 2.8) | <.001 |
| Male, No. (%) | 305(144,68,93) | 238 (78%) | 113 (78%) | 55 (81%) | 70 (75%) | 0.686 |
| Race: White, No. (%) | 305(144,68,93) | 254 (83%) | 117 (81%) | 56 (82%) | 81 (87%) | 0.487 |
| Ethnicity: Hispanic/Latino, No. (%) | 299(140,68,91) | 5 (2%) | 3 (2%) | 2 (3%) | 0 (0%) | 0.301 |
| Currently Married/Domestic partner, No. (%) | 303(142,68,93) | 237 (78%) | 109 (77%) | 53 (78%) | 75 (81%) | 0.778 |
| Education > High School, No. (%) | 278(125,60,93) | 197 (71%) | 88 (70%) | 44 (73%) | 65 (70%) | 0.890 |
| Currently working (Closest to Surgery), No. (%) | 294(133,68,93) | 39 (13%) | 19 (14%) | 12 (18%) | 8 (9%) | 0.222 |
| Current primary insurance type, No. (%) | 305(144,68,93) | 0.006 | ||||
| Medicare/Medicaid | 198 (65%) | 105 (73%) | 44 (65%) | 49 (53%) | ||
| Private Insurance | 107 (35%) | 39 (27%) | 24 (35%) | 44 (47%) | ||
| Clinical Characteristics | ||||||
| Etiology of Heart Failure, No. (%) | 305(144,68,93) | 0.090 | ||||
| Ischemic Cardiomyopathy | 137 (45%) | 74 (51%) | 29 (43%) | 34 (37%) | ||
| Dilated Cardiomyopathy | 153 (50%) | 63 (44%) | 38 (56%) | 52 (56%) | ||
| Other | 15 (5%) | 7 (5%) | 1 (1%) | 7 (8%) | ||
| NYHA (Closest to Surgery), No. (%) | 305(144,68,93) | <.001 | ||||
| I and II | 47 (15%) | 1 (1%) | 37 (54%) | 9 (10%) | ||
| III and IV | 258 (85%) | 143 (99%) | 31 (46%) | 84 (90%) | ||
| Intermacs Profile (Closest to Implant), No. (%) | 219(144,65,10) | 0.009 | ||||
| Profile 1 | 25 (11%) | 11 (8%) | 12 (18%) | 2 (20%) | ||
| Profiles 2–3 | 161 (74%) | 117 (81%) | 39 (60%) | 5 (50%) | ||
| Profiles 4–7 | 33 (15%) | 16 (11%) | 14 (22%) | 3 (30%) | ||
| Not Applicable/Recorded | 86 (28%) | 0 (0%) | 3 (4%) | 83 (89%) | ||
| Left Ventricular Ejection Fraction, No. (%) | 305(144,68,93) | <.001 | ||||
| >30 (Normal to Moderate) | 20 (9%) | 6 (4%) | 3 (8%) | 11 (22%) | ||
| 20–29 (Moderate/Severe) | 75 (33%) | 45 (33%) | 12 (32%) | 18 (35%) | ||
| <20 (Severe) | 130 (58%) | 86 (63%) | 22 (59%) | 22 (43%) | ||
| Not Recorded/Documented | 80 (26%) | 7 (5%) | 31 (46%) | 42 (45%) | ||
| Inotrope Therapy (Within 48 Hours of Surgery), No. (%) | 303(143,67,93) | 190 (63%) | 116 (81%) | 12 (18%) | 62 (67%) | <.001 |
| Temporary MCS in place immediately prior to surgery, No. (%) | 305(144,68,93) | 5 (2%) | 0 (0%) | 2 (3%) | 3 (3%) | 0.102 |
| IABP (Within 48 Hours of Surgery), No. (%) | 236(143,0,93) | 88 (37%) | 53 (37%) | NA | 35 (38%) | 0.929 |
| Number of Major Comorbidities* (mean±SD) | 305(144,68,93) | 4.4+2.1 | 5.0+2.2 | 4.0+1.7 | 3.7+1.8 | <0.001 |
| Number of Major Comorbidities (binned), No. (%) | 305(144,68,93) | <.001 | ||||
| 1–2 | 59 (19%) | 21 (15%) | 12 (18%) | 26 (28%) | ||
| 3–5 | 156 (51%) | 62 (43%) | 43 (63%) | 51 (55%) | ||
| 6 or more | 90 (30%) | 61 (42%) | 13 (19%) | 16 (17%) | ||
| Arrhythmia, No. (%) | 305(144,68,93) | 185 (61%) | 93 (65%) | 42 (62%) | 50 (54%) | 0.245 |
| Hypertension, No. (%) | 305(144,68,93) | 182 (60%) | 94 (65%) | 37 (54%) | 51 (55%) | 0.168 |
| Hyperlipidemia, No. (%) | 305(144,68,93) | 180 (59%) | 90 (63%) | 40 (59%) | 50 (54%) | 0.410 |
| Diabetes, No. (%) | 305(144,68,93) | 137 (45%) | 80 (56%) | 28 (41%) | 29 (31%) | <.001 |
| Chronic Kidney Disease, No. (%) | 305(144,68,93) | 114 (37%) | 65 (45%) | 23 (34%) | 26 (28%) | 0.022 |
| Myocardial Infarction, No. (%) | 305(144,68,93) | 99 (32%) | 55 (38%) | 19 (28%) | 25 (27%) | 0.128 |
| History of Smoking, No. (%) | 293(134,66,93) | 87 (30%) | 23 (17%) | 22 (33%) | 42 (45%) | <.001 |
| Pulmonary Hypertension, No. (%) | 305(144,68,93) | 63 (21%) | 32 (22%) | 20 (29%) | 11 (12%) | 0.020 |
| Obesity (BMI >30kg/m2), No. (%) | 305(144,68,93) | 58 (19%) | 30 (21%) | 13 (19%) | 15 (16%) | 0.666 |
| History of Cancer, No. (%) | 305(144,68,93) | 44 (14%) | 26 (18%) | 6 (9%) | 12 (13%) | 0.179 |
| Cerebrovascular Accident, No. (%) | 305(144,68,93) | 44 (14%) | 23 (16%) | 11 (16%) | 10 (11%) | 0.481 |
Note: Column 2 depicts the sample size for the entire cohort by variable and within the parentheses, we have indicated the sample size for each group in the following order: long-term MCS, HT MCS, and HT Non-MCS.
NYHA = New York Heart Association; Intermacs = Interagency Registry for Mechanically Assisted Circulatory Support; MCS = mechanical circulatory support; IABP = intra-aortic balloon pump; ECMO = extracorporeal membrane oxygenation; BMI=body mass index
Comorbidities include arrhythmia, hyperlipidemia, hypertension, diabetes, chronic liver disease, chronic kidney disease, myocardial infarction, pulmonary hypertension, obesity, history of cancer, CVA/Stroke, carotid disease, coronary artery diseases, obstructive sleep apnea, peripheral vascular disease, pulmonary disease (COPD/Asthma), and other medical condition.
Examples of temporary MCS include Impella and Tandem Heart.
Table 2.
Patient Device Type and Adverse Events After Surgery Through 24 Months
| Variable | N Observed Total cohort (per group) |
Total Cohort (N=305) |
Long-term MCS (n=144) |
HT MCS (n=68) |
HT Non-MCS (n=93) |
P-value |
|---|---|---|---|---|---|---|
| Device type*, No. (%) | 214(135,68,11) | 0.208 | ||||
| HeartMate II LVAD | 88 (41%) | 55 (41%) | 29 (43%) | 4 (36%) | ||
| HeartWare LVAD | 70 (33%) | 41 (30%) | 27 (40%) | 2 (18%) | ||
| HeartMate 3 LVAD | 56 (26%) | 39 (29%) | 12 (18%) | 5 (45%) | ||
| Percentage of patients with any adverse events, No. (%) | 305(144,68,93) | 248 (81%) | 127 (88%) | 50 (74%) | 71 (76%) | 0.013 |
| Number of adverse events, median (Q1, Q3) | 305(144,68,93) | 3.0 (1.0, 6.0) | 4.0 (2.0, 8.0) | 2.0 (0.0, 5.5) | 2.0 (1.0, 3.0) | <.001 |
| Number of rehospitalizations due to any adverse event, median (Q1, Q3) | 291(136,67,88) | 1.0 (0.0, 3.0) | 2.0 (1.0, 4.0) | 1.0 (0.0, 2.0) | 1.0 (0.0, 2.0) | <.001 |
| Major Infection, No. (%) | 305(144,68,93) | 131 (43%) | 68 (47%) | 31 (46%) | 32 (34%) | 0.133 |
| Major Bleeding, No. (%) | 305(144,68,93) | 88 (29%) | 72 (50%) | 8 (12%) | 8 (9%) | <.001 |
| Cardiac Arrhythmia, No. (%) | 305(144,68,93) | 70 (23%) | 48 (33%) | 12 (18%) | 10 (11%) | <.001 |
| Renal Dysfunction, No. (%) | 305(144,68,93) | 56 (18%) | 20 (14%) | 18 (26%) | 18 (19%) | 0.083 |
| Respiratory Failure, No. (%) | 305(144,68,93) | 42 (14%) | 25 (17%) | 12 (18%) | 5 (5%) | 0.019 |
| Neurological Dysfunction, No. (%) | 305(144,68,93) | 42 (14%) | 31 (22%) | 6 (9%) | 5 (5%) | <.001 |
| Worsening Heart Failure, No. (%) | 305(144,68,93) | 27 (9%) | 9 (6%) | 11 (16%) | 7 (8%) | 0.052 |
| MCS Device Malfunction, No. (%) | 305(144,68,93) | 26 (9%) | 26 (18%) | NA | NA | |
| Acute Rejection, No. (%) | 305(144,68,93) | 25 (8%) | NA | 9 (13%) | 16 (17%) | 0.492 |
| Right Heart Failure, No. (%) | 305(144,68,93) | 21 (7%) | 15 (10%) | 2 (3%) | 4 (4%) | 0.067 |
| Psychiatric Episode, No. (%) | 305(144,68,93) | 15 (5%) | 12 (8%) | 2 (3%) | 1 (1%) | 0.029 |
Note: Column 2 depicts the sample size for the entire cohort by variable and within the parentheses, we have indicated the sample size for each group in the following order: long-term MCS, HT MCS, and HT Non-MCS.
Device type for HT MCS includes patients enrolled while on MCS; device type for HT Non-MCS includes patients enrolled without a device who subsequently required MCS after enrollment and before HT.
LVAD=left ventricular assist device
EQ-5D-3L VAS scores range from worst (0) to best (100) imaginable health states. At baseline, VAS score averages+SD ranged from 45.7+22.9 to 67.8+20.7 in the three groups, differing significantly among groups with the long-term MCS group having the lowest score (Table S1). Baseline EQ-5D-3L dimension scores also differed significantly among groups, with the long-term MCS group reporting more problems within each dimension (self-care, usual activities, mobility, pain/discomfort, and anxiety/depression) than the HT groups (Table S1). At 24 months post-surgery, average+SD VAS scores were higher than at baseline for all groups, ranging from 70.2+20.3 to 84.8+13.1; the long-term MCS group still had the lowest scores. Patients reported fewer problems at 24 months; however, the long-term MCS group reported more problems for the self-care, usual activities, and mobility dimensions than the HT groups. (Table S1).
Health Utilities and Survival
Health utility scores ranging from 0 [dead] to 1 [perfect health] differed among the three groups over time. While median utility scores for all groups were higher than 0.50 at baseline and in follow-up, the long-term MCS group generally had lower health utility scores than both HT groups through 24 post-operative months (Figure 2 and Table S2).
Figure 2.
Box Plots Displaying Health Utility Scores by Surgical Strategy.
We observed lower overall survival in the long-term MCS group compared to the HT Non-MCS group through 24 months after surgery, hazard ratio (HR), 95% confidence interval (CI) 7.82 (3.11, 19.70), p<0.001 (Figure 3). The HT MCS group also had significantly lower overall survival than the HT Non-MCS group, HR and CI=2.24 (1.04, 8.95), p=0.041. Notably, there were more deaths in the long-term MCS group than in the two HT groups (Figure 3).
Figure 3.
Kaplan-Meier Survival Curves from Baseline to 24 months after Long-term Mechanical Circulatory Support (MCS) or Heart Transplantation (HT) with or without pre-transplant durable MCS.
QALYs
Unadjusted Analyses
Average QALYs since baseline were similar in the three groups at the 3- and 6-month windows after surgery (Table 3). Beginning at the 9-month window after HT or long-term MCS, sustained significant differences in average+SD QALYs emerged across groups. At the 24-month window, the average QALY was significantly higher (p<0.001) in the HT Non-MCS group (22.75±0.54 months) and the HT MCS group (20.84±0.93 months) compared with the long-term MCS group (17.95+0.74 months). Notably, at the 24-month window, there were no significant differences between the HT MCS vs HT Non-MCS QALY groups (p=0.075). QALY gains were significantly higher in the HT Non-MCS group compared to the long-term MCS group beginning at the 6-month window, while differences in QALY gains between the HT MCS group and the long-term MCS group were not significant until the 24-month window (Figure 4, Panel A). QALY gains between the HT Non-MCS group and HT MCS group were not significantly different over time (Figure 4, Panel A).
Table 3.
Average QALYs at the 3- through 24-month windows post-surgery by group (treatment strategy)
| Group | 3 Months | 6 Months | 9 Months | 12 Months | 18 Months | 24 Months |
|---|---|---|---|---|---|---|
| Unadjusted Analyses | ||||||
| HT Non-MCS | 2.88±0.06 | 5.72±0.12 | 8.56±0.19 | 11.39±0.26 | 17.07±0.40 | 22.75±0.54 |
| HT MCS | 2.82± 0.07 | 5.44±0.18 | 8.04±0.30 | 10.61±0.43 | 15.72±0.68 | 20.84±0.93 |
| Long-term MCS | 2.80±0.05 | 5.35±0.13 | 7.79±0.22 | 10.13±0.32 | 14.25±0.51 | 17.95±0.74 |
| p-value | 0.56 | 0.11 | 0.027 | 0.007 | <0.001 | <0.001 |
| Adjusted Analyses | ||||||
| HT Non-MCS | 2.83 ± 0.11 | 5.6 ± 0.25 | 8.38 ± 0.39 | 11.2 ± 0.53 | 16.86 ± 0.82 | 22.58 ± 1.1 |
| HT MCS | 2.65 ± 0.12 | 5.10 ± 0.28 | 7.53 ± 0.45 | 9.94 ± 0.63 | 14.73 ± 0.98 | 19.53 ± 1.33 |
| Long-term MCS | 2.82 ± 0.12 | 5.47 ± 0.28 | 8.04 ± 0.44 | 10.57 ± 0.61 | 15.17 ± 0.95 | 19.49 ± 1.3 |
| p-value | 0.37 | 0.18 | 0.12 | 0.07 | 0.014 | 0.003 |
QALY=quality-adjusted life years; HT=heart transplantation, MCS=mechanical circulatory support
Note: Average QALYs are calculated as mean+SD
Figure 4.
Change in QALYs Over Time: Unadjusted (Panel A) and Adjusted (Panel B).
Adjusted Analyses
Upon adjusting for age, sex, insurance type, baseline EQ-5D VAS, left ventricular ejection fraction, NYHA class, number of co-morbidities, diabetes, and HF etiology, significant differences in average QALYs emerged, with the HT Non-MCS group having the highest average+SD QALYs at both the 18- and 24-month windows (24-month window: HT Non-MCS=22.58+1.1, HT MCS=19.53+1.33, Long-term MCS=19.49+1.3, p=0.003) (Table 3). QALY gains were similar in the HT Non-MCS group compared to the long-term MCS group until the 18-month window, wherein gains were significantly lower in the long-term MCS group (Figure 4, Panel B). Gains in QALYs were less in the long-term MCS group when compared with the HT MCS group over time, yet differences between groups were not significant (Figure 4, Panel B). Lastly, QALY gains were significantly different between the HT Non-MCS group compared to the HT MCS group, favoring the HT Non-MCS group beginning at the 12-month window after surgery (Figure 4, Panel B).
Factors associated with a lower QALY gain relative to the HT Non-MCS group
Using univariable analyses (Table S3) followed by multivariable analyses, a lower gain in QALYs at the 24-month window post-operatively was associated with HT MCS, long-term MCS, a reduced LVEF prior to surgery, NYHA class III or IV before surgery, and an ischemic or other (e.g., congenital heart disease and valvular heart disease) etiology of HF (Table 4). Findings were similar for multivariable analyses at the 12-month window, except long-term MCS and ischemic etiology of HF were not significant factors (Table 4). Age, sex, baseline EQ-5D VAS, and co-morbidities were not significantly associated with QALY distributions at the 12- and 24-month windows for QALY multivariable models (Table 4).
Table 4.
Multivariable Summary of QALYs gained compared to the HT Non-MCS group after adjusting for baseline characteristics.
| 12 Months | 24 Months | |||||
|---|---|---|---|---|---|---|
| Estimate | (95% CI) | p-value | Estimate | (95% CI) | p-value | |
| Surgical Strategy | ||||||
| HT Non-MCS | Reference | Reference | Reference | Reference | Reference | Reference |
| HT MCS | −1.26 | (−2.46, −0.06) | 0.040 | −3.05 | (−5.63, −0.48) | 0.020 |
| Long-term MCS | −0.63 | (−1.58, 0.32) | 0.20 | −3.09 | (−5.27, −0.92) | 0.005 |
| Age at time of surgery | 0.002 | (−0.09, 0.09) | 0.96 | 0.002 | (−0.21, 0.22) | 0.98 |
| Male | 0.23 | (−0.73, 1.2) | 0.63 | 0.51 | (−1.64, 2.66) | 0.64 |
| Insurance Type: Private Insurance vs | ||||||
| Public Insurance | 0.48 | (−0.28, 1.24) | 0.22 | 0.99 | (−0.72, 2.7) | 0.26 |
| LVEF before procedure | ||||||
| >30 (Normal to Moderate) | Reference | Reference | Reference | Reference | Reference | Reference |
| 20–29 (Moderate/Severe) | −1.35 | (−2.37, −0.33) | 0.009 | −2.99 | (−5.56, −0.43) | 0.022 |
| <20 (Severe) | −1.87 | (−2.78, −0.96) | <0.001 | −3.91 | (−6.25, −1.58) | 0.001 |
| Not recorded/not documented | −0.95 | (−1.78, −0.11) | 0.026 | −1.79 | (−3.9, 0.32) | 0.10 |
| Diabetes | −0.34 | (−1.19, 0.51) | 0.43 | −0.77 | (−2.63, 1.09) | 0.42 |
| NYHA class before procedure | ||||||
| I and II | Reference | Reference | Reference | Reference | Reference | Reference |
| III and IV | −1.44 | (−2.63, −0.24) | 0.018 | −3.31 | (−5.9, −0.73) | 0.0112 |
| Number of Comorbidities | ||||||
| 1–2 | Reference | Reference | Reference | Reference | Reference | Reference |
| 3–5 | −0.63 | (−1.58, 0.32) | 0.19 | −0.98 | (−3.09, 1.12) | 0.36 |
| 6 or more | −0.16 | (−1.29, 0.97) | 0.78 | −0.51 | (−3.06, 2.04) | 0.69 |
| HF Etiology | ||||||
| Dilated Cardiomyopathy | Reference | Reference | Reference | Reference | Reference | Reference |
| Ischemic Cardiomyopathy | −0.52 | (−1.25, 0.2) | 0.16 | −1.77 | (−3.44, −0.11) | 0.037 |
| Other | −3.51 | (−6.18, −0.84) | 0.010 | −6.70 | (−12, −1.4) | 0.013 |
| EQ-5D VAS at Baseline | −0.007 | (−0.02, 0.01) | 0.40 | −0.02 | (−0.05, 0.02) | 0.40 |
DISCUSSION
We present novel findings on QALY distributions of older patients with HF who undergo long-term MCS or HT, with or without pre-transplant MCS. All three groups experienced gains in QALYs through 24 months after surgery. However, the change in QALYs over time differed among groups, with the HT Non-MCS group having the most QALY gains by the 24-month window. Furthermore, etiology and severity of HF at baseline and surgical treatment options were associated with lower QALY gains after surgery; there was no such association by age. The analytic strength and novelty of our findings is that we used patient self-reported HRQOL data (EQ-5D-3L dimension scores) to calculate health utility scores, conducted analyses in older patients with HF, and compared and characterized QALY differences among surgical groups.
The literature supports our findings. Similar to our report, Long et al.,9 estimated QALY distributions after both HT and long-term MCS and identified QALY gains following both surgeries in adults of all ages. Other studies of adult cohorts focused on QALY gains after MCS, primarily long-term MCS, similarly reporting post-operative gains in QALYs.11–14 These studies also assessed cost-effectiveness of these advanced surgical procedures. While HT was cost-effective in the U.S.,9 cost-effectiveness of long-term MCS implantation varied by country.9, 11–14 Specifically, despite more recent analyses of cost-effectiveness with contemporary continuous flow devices, implantation of long-term MCS in publicly-funded healthcare systems in some countries, including Canada and the United Kingdom, is typically above usual thresholds set for funding.11–13 MCS is approved for coverage and reimbursement in the French healthcare system, without distinction by goal of therapy, and without a firm threshold.14 Potential factors affecting cost-effectiveness of long-term MCS included device and procedure costs, index hospital intensive care unit stay, post-implantation complications, device re-implantation, re-hospitalization, uncertain HRQOL estimates, and survival, which has improved over time with use of more contemporary devices.9–14 In a systematic review of economic evaluation of MCS from several countries, Nyet et al.10 stated that indirect assessment of health utilities was a major weakness in many studies and recommended use of directly reported HRQOL in future QALY and economic analyses. Thus, our report from SUSTAIN-IT, using patient self-report of HRQOL that considers multiple dimensions of quality of life, addresses an important limitation of previous studies using QALYs. We strongly recommend that future studies include direct measurement of health utilities.
QALYs can be used to influence policy to advocate for use of treatments, including technologies, in patient care. An issue of substantial importance in use of long-term MCS technology is that some countries have healthcare systems with a finite healthcare budget, wherein policy makers must make decisions focused on best use of resources that benefit society as a whole, although these healthcare systems are increasingly under pressure to fund implantation of long-term MCS.10–13 Neyt et al.,10 concluded that while both survival and HRQOL improve with MCS implantation, value for monies spent is more questionable, indicating that further improvement in device technology is needed.
We focused on an older age group, wherein there is a potential for discrimination regarding healthcare policies specific to use of expensive treatments, as it may be perceived that while older patients may derive similar or more HRQOL benefit from an expensive treatment than younger healthier patients, this benefit is over a shorter period of time.20, 21 Determination of QALYs over time in older patients, who are undergoing long-term MCS implantation more frequently than in past eras, may provide important information for policy makers and clinicians to consider regarding the benefit of this treatment option in these older patients. Survival is worse in older patients who undergo HT and MCS,2, 4 yet comparisons of HRQOL among older and younger patients who undergo these therapies favor older patients.5, 22 We also found in this report that age was not significantly associated with a lower gain in QALYs in older patients undergoing advanced surgical therapies through 24-months follow-up. The association of treatment strategy with lower gains in QALYs is likely driven by both the higher number of deaths and lower health utility scores in the long-term MCS group, as compared to the HT groups through 2 post-operative years. Yet, while the long-term MCS group had the lowest average QALYs through 24 months, the absolute gain in overall HRQOL (as measured by the EQ-5D-3L VAS) over time was highest in this group.
Importantly, QALYs should not be used on an individual level, as these aggregate data do not consider what may be best for the patient, based on physical examination, tests, etc., nor do they consider individual patient preferences, values, and treatment goals.20 Treatments that improve patient HRQOL without extending life (or extend life less compared to other treatments) are essential for countless patients with chronic diseases.21 Thus, while older patients who undergo long-term MCS, when ineligible for HT, do not have QALY gains over time as high as older patients who undergo HT, they none-the-less have gains in QALYs. Shared decision making, when considering these surgical treatment options, must entail a discussion reflecting treatment benefits, risks, and alternatives, and patient preferences and goals of treatment.23
Limitations of our study in these older patients include a fairly homogeneous cohort by sex, race, and education. Furthermore, the follow-up period was ~24 months post-operatively; longer follow-up is needed. Lastly, only 26% of all MCS devices were HeartMate 3 LVADs, which have different adverse event profiles and rates of survival compared to the other two devices.24–26 Thus, QALYs for patients implanted with HeartMate 3 devices require independent study. An important strength of our study was direct measurement of health utilities.
CONCLUSIONS
Our findings depict QALY gains in older patients with advanced HF who undergo HT or long-term MCS, using contemporary continuous flow durable MCS. Determination of QALYs may provide important information for policy makers and clinicians to consider regarding the benefits of HT and long-term MCS as treatment options for older patients with advanced HF.
Supplementary Material
SOURCES OF FUNDING
This work was sponsored by the National Institutes of Health, National Institute on Aging (NIA), Sustaining quality of life of the aged: Heart transplant or mechanical circulatory support? (SUSTAIN-IT) (R01AG047416, Grady KL [PI]); ClinicalTrials.gov ID: NCT02568930
FINANCIAL CONFLICT OF INTEREST STATEMENT
Kathleen L. Grady, NIH/NIA and NIH/NHLBI grants; payment of registration fees for scientific meetings (AHA, HFSA, ACC, ISHLT); leadership role (member ISHLT Board of Directors, Foundation Board, Research Oversight Committee, Governance Committee, Leadership Advisory Forum, and chair Grants and Awards Committee).
Mary Amanda Dew, NIH grants.
Francis D. Pagani, none.
John A. Spertus, NIH grants; research contracts from Janssen and Abbott Vascular and BMS; royalties or licenses for SAQ, KCCQ-12, and PAQ; consulting fees from Alnylam, BMS, Sanofi, Edwards, Bayer, Terumo, Cytokinetics, Imbria, Janssen, Merck, Pfizer, Lilly, United Healthcare; participation on the NIH funded RECOVER DSMB; Board member of Blue Cross Blue Shield of Kansas City.
Eileen Hsich, NIH/NHLBI grant.
Melana Yuzefpolskaya, Abbott Educational grant; Abbott Speakers Bureau.
Brent Lampert, DO, none.
James K. Kirklin, MD, Intellectual properties for IT software development in registry database design developed at and licensed from the University of Alabama at Birmingham; Chair of DSMB for Xeltis cardiac conduit clinical trial, Chair of DSMB for Carmat TAH clinical trial, and Chair- XVIVO Clinical Safety Monitoring Board - Tx preservation technology; President; World Society for Pediatric and Congenital Heart Surgery; Common stock in Kirklin Solutions Co. Database development and analytics; receive partial salary support in my role as Director of the Data Center for STS Intemacs/Pedimacs Registries.
Michael Petty, payment for lectures for Abbott Labs.
Andrew Kao, MD, honoraria from Kansas Association of Sleep Professionals and the American Board of Internal Medicine; support for attending meetings for CareDX, American Board of Internal Medicine, and United Network for Organ Sharing; participation on the CareDX DSMB or Advisory Board. Clyde Yancy, participation on a DSMB or Advisory Board for the NHLBI Cardiothoracic Surgery Clinical Trials Network.
Justin Hartupee, honoraria for lectures from Bristol Myers Squibb; payment for expert testimony; support for attending meetings by Abbott.
Salpy V. Pamboukian, MD, participation on the Clinical Events Committee, Chair Carmat Total Artificial Heart.
Maryl Johnson, MD, none.
Margaret Murray, none.
Tingqing Wu, none.
Adin-Cristian Andrei, NH/NIA grant.
Non-standard abbreviations:
- DT
destination therapy
- EQ-5D-3L VAS
Euroqol-5 dimension-3L visual analog scale
- HF
heart failure
- HRQOL
health-related quality of life
- HT
heart transplantation
- LVAD
left ventricular assist device
- LVEF
left ventricular ejection fraction
- MCS
mechanical circulatory support
- QALY
Quality-adjusted life year
- STS Intermacs
Society of Thoracic Surgeons Interagency Registry for Mechanically Assisted Circulatory Support
- SUSTAIN-IT
Sustaining Quality of Life of the Aged: Heart Transplant or Mechanical Support?
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Contributor Information
Kathleen L. Grady, Northwestern University, Chicago, IL;.
Mary Amanda Dew, University of Pittsburgh, Pittsburgh, PA;.
Francis D. Pagani, University of Michigan, Ann Arbor, MI;.
John A. Spertus, University of Missouri-Kansas City, Kansas City, MO;.
Eileen Hsich, Cleveland Clinic, Cleveland, OH;.
Melana Yuzefpolskaya, Columbia University, New York, NY;.
Brent Lampert, Ohio State University, Columbus, OH;.
James K. Kirklin, Kirklin Solutions, Hoover, AL;.
Michael Petty, University of Minnesota Medical Center, Fairview, MN;.
Andrew Kao, St. Luke’s Health System, Kansas City, MO;.
Clyde Yancy, Northwestern University, Chicago, IL;.
Justin Hartupee, Washington University, St. Louis, MO;.
Salpy V. Pamboukian, University of Washington, Seattle, WA;.
Maryl Johnson, University of Wisconsin, Madison, WI;.
Margaret Murray, University of Wisconsin, Madison, WI;.
Tingqing Wu, Northwestern University, Chicago, IL;.
Adin-Cristian Andrei, Northwestern University, Chicago, IL;.
REFERENCES
- 1.Tsao CW, Aday AW, Almarzooq ZI, et al. Heart Disease and Stroke Statistics – 2022 Update: A Report from the American Heart Association. Circulation 2022;145(8):e153–e639. [DOI] [PubMed] [Google Scholar]
- 2.Khush K, Cherikh W, Chambers D, et al. The International Thoracic Organ Transplant Registry of the International Society for Heart and Lung Transplantation: Thirty-sixth Adult Heart Transplantation Report—2019; Focus Theme: Donor and recipient size match. J Heart Lung Transplant 2019;38(10):1056–66. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Goldstein DJ, Bello R, Shin JJ, et al. Outcomes of cardiac transplantation in septuagenarians. J Heart Lung Transplant 2012;31(7):679–85. [DOI] [PubMed] [Google Scholar]
- 4.Kormos RL, Cowger J, Pagani FD, et al. The Society of Thoracic Surgeons INTERMACS database annual report: evolving indications, outcomes, and scientific partnerships. J Heart Lung Transplant 2019;38:114–26. [DOI] [PubMed] [Google Scholar]
- 5.Grady KL, Naftel D, Myers S, et al. Change in health-related quality of life from before to after destination therapy mechanical circulatory support is similar for older and younger patients: Analyses from INTERMACS. J Heart Lung Transplant 2015;34:213–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Andrei AC, Murray S. Regression models for the mean of the quality-of-life-adjusted restricted survival time using pseudo-observations. Biometrics 2007;63(2):398–404. [DOI] [PubMed] [Google Scholar]
- 7.Andrei AC, Grady KL. Visualization and dynamics of multidimensional health-related quality-of-life-adjusted overall survival: a new analytic approach. Qual Life Res 2014;23:1411–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Torrance G, Feeny D. Utilities and quality-adjusted life years. Intl J Tech Assess 1989;5, 559–75. [DOI] [PubMed] [Google Scholar]
- 9.Long E, Swain G, Mangi A. Comparative survival and cost-effectiveness of advanced therapies for end-stage heart failure. Circ Heart Fail 2014May;7(3):470–8. [DOI] [PubMed] [Google Scholar]
- 10.Neyt M, Van den Bruel A, Smit Y, De Jong N, Vlayen J. The cost-utility of left ventricular assist devices for end-stage heart failure patients ineligible for cardiac transplantation: A systematic review and critical appraisal of economic evaluations. Ann Cardiothorac Surg 2014. Sep;3(5):439–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Chew D, Manns B, Miller R, Sharma N, Exner D. Economic Evaluation of Left Ventricular Assist Devices for Patients with End Stage Heart Failure Who Are Ineligible for Cardiac Transplantation. Can J Cardiol. 2017Oct;33(10):1283–91. [DOI] [PubMed] [Google Scholar]
- 12.Schueler S, Silvestry S, Cotts W, et al. Cost-effectiveness of left ventricular assist devices as destination therapy in the United Kingdom. ESC Heart Fail 2021Aug;8(4):3049–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lim H, Shaw S, Carter A, et al. A clinical and cost-effectiveness analysis of the HeartMate 3 left ventricular assist device for transplant-ineligible patients: A United Kingdom perspective. J Heart Lung Transplant 2022. Feb;41(2):174–86. [DOI] [PubMed] [Google Scholar]
- 14.Tadmouri A, Bloomkvist J, Landais C, Seymour J, Azmoun A. Cost-effectiveness of left ventricular assist devices for patients with end-stage heart failure: Analysis of the French hospital discharge database. ESC Heart Fail. 2018Feb;5(1):75–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Caraballo C, DeFilippis E, Nakagawa S, et al. , Clinical Outcomes After Left Ventricular Assist Device Implantation in Older Adults: An INTERMACS Analysis. J Am Coll Cardiol HF 2019;7:1069–1078. [DOI] [PubMed] [Google Scholar]
- 16.Grady KL, Andrei AC, Elenbaas C, Warzecha A, Baldridge A, Kao A, Spertus JA, Pham DT, Dew MA, Hsich E, et al. Health-related quality of life in older patients with advanced heart failure: findings from the SUSTAIN-IT study. J Am Heart Assoc 2022;11(4):e024385. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.EuroQol Group. EuroQol--a new facility for the measurement of health-related quality of life. Health Policy 1990;16(3):199–208. [DOI] [PubMed] [Google Scholar]
- 18.Brazier J, Jones N, Kind P. Testing the validity of the Euroqol and comparing it with the SF-36 Health Survey questionnaire. Qual Life Res 1993;2(3):169–180. [DOI] [PubMed] [Google Scholar]
- 19.EQ-5D-3L User Guide 2018. https://euroqol.org/publications/user-guides. Accessed 11/27/23.
- 20.La Puma J, Lawlor EF. Quality-adjusted life-years. Ethical implications for physicians and policymakers. J Am Med Assoc 1990Jun6;263(21):2917–21. [DOI] [PubMed] [Google Scholar]
- 21.Sullivan S, Lakdawalla D, Devine B, et al. Alternatives to the QALY for Comparative Effectiveness Research. Health Affairs APRIL 21, 2023. 10.1377/forefront.20230419.896238. [DOI] [Google Scholar]
- 22.Shamaskin A, Rybarczyk B, Wang E, et al. Older patients have better quality of life, adjustment and adherence than younger patients 5 years post heart transplantation. J Heart Lung Transplant 2012May;31(5):478–84. [DOI] [PubMed] [Google Scholar]
- 23.Allen L, Stevenson L, Grady K, et al. Decision-making in advanced heart failure: A scientific statement from the American Heart Association. Circ. 2012April17;125(15):1928–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Mehra M, Goldstein DJ, Cleveland JC, et al. Five-Year Outcomes in Patients With Fully Magnetically Levitated vs Axial-Flow Left Ventricular Assist Devices in the MOMENTUM 3 Randomized Trial. J Am Heart Assoc 2022Sep27;328(12):1233–42. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Cho SM, Moazami N, Frontera JA. Stroke and Intracranial Hemorrhage in HeartMate II and HeartWare Left Ventricular Assist Devices: A Systematic Review. Neurocrit Care 2017Aug;27(1):17–25. [DOI] [PubMed] [Google Scholar]
- 26.Mihalj M, Heinisch PP, Schober P, et al. Third-generation continuous-flow left ventricular assist devices: a comparative outcome analysis by device type. ESC Heart Fail 2022Oct;9(5):3469–82. [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.