Abstract
Introduction:
Ambulatory patients with advanced heart failure are often considered for advanced therapies, including durable mechanical circulatory support (MCS). The Interagency Registry for Mechanically Assisted Circulatory Support (INTERMACS) Profiles are a commonly used descriptor of disease severity in patients receiving MCS devices, but their role in defining the prognosis of ambulatory patients is less well-established, especially for Profiles 6 and 7.
Methods:
REVIVAL is a prospective, observational study of 400 outpatients from 21 MCS/cardiac transplant centers. Eligible patients had NYHA class II-IV symptoms despite optimal medical and electrical therapies with a recent heart failure hospitalization, heart transplant listing, or evidence of high neurohormonal activation.
Results:
The cohort included 33 (8%) INTERMACS Profile 4, 83 (21%) Profile 5, 155 (39%) Profile 6, and 129 (32%) Profile 7. Across INTERMACS Profiles, there were no differences in age, gender, ejection fraction, blood pressure, or use of guideline-directed medical therapy. A lower INTERMACS Profile was associated with more hospitalizations, greater frailty and more impaired functional capacity and quality of life. The composite endpoint of death, durable MCS, or urgent transplant at 12 months occurred in 39%, 27%, 24%, and 14% subjects with INTERMACS Profiles 4, 5, 6, and 7, respectively (p = 0.004).
Conclusions:
Among ambulatory patients with advanced heart failure, a lower INTERMACS Profile was associated with a greater burden of heart failure across multiple dimensions and a higher composite risk of durable MCS, urgent transplant, or death. These Profiles may assist in risk assessment and triaging ambulatory patients to advanced therapies.
Keywords: INTERMACS Profiles, ambulatory heart failure, death, LVAD, heart transplantation
Introduction
Patients with advanced heart failure (HF) despite optimal medical and electrical therapies suffer from a high degree of morbidity and mortality,1 and mechanical circulatory support (MCS) offers the possibility for increased quality of life and survival.2 While MCS devices were initially approved for patients with cardiogenic shock or dependent on continuous intravenous inotropic therapy, ambulatory non-inotrope dependent patients with advanced HF are increasingly considered for MCS.3–5 However, these ambulatory patients with INTERMACS (Interagency Registry for Mechanically Assisted Circulatory Support) Profiles 4 to 7 still account for less than 13% of patients implanted in 2015–16 with approved MCS devices as registered in INTERMACS.2
The greatest potential benefits from MCS in the HF epidemic are anticipated to be in ambulatory patients where elective device placement would result in lower postoperative complication rates and improved long-term survival.6,7 However, use of MCS in ambulatory HF patients is constrained in part by the lack of information and therefore equipoise regarding expected outcomes for comparable patients on contemporary medical therapy without MCS. INTERMACS Profiles have influenced candidate selection and timing for durable MCS, particularly in the hospital setting, but the utility of INTERMACS Profiles for the triage of ambulatory advanced HF patients on outpatient oral therapy has not been established.8–11
A better understanding of outcomes in these ambulatory patients with advanced HF will inform MCS patient selection, informed consent, and policy decisions. Through the Registry Evaluation of Vital Information on VADs in Ambulatory Life (REVIVAL), we sought to characterize the clinical features and 1-year outcomes of ambulatory patients with advanced HF who demonstrate high-risk features at MCS/cardiac transplant centers in the United States. We specifically focused on the impact of assigned INTERMACS Profiles at enrollment on 1-year outcomes based on the AHA Scientific Statement on decision-making in advanced heart failure12 which recommends an integrated assessment of the burden of heart failure and estimation of prognosis as part of an Annual Heart Failure Review. We hypothesized that the INTERMACS profile assignments could contribute meaningful to this estimated one-year prognosis for ambulatory advanced heart failure, thus providing clinically actionable information for these patients.
Methods
Study Population
The REVIVAL registry is a prospective, observational, multicenter patient cohort of ambulatory patients with advanced HF. Patients were enrolled from 21 centers between July 2015 and June 2016. Subjects enrolled were aged 18 to 80 years with New York Heart Association (NYHA) functional class II - IV limitation for 45 of the last 60 days, left ventricular ejection fraction ≤35%, and a HF diagnosis >12 months. All patients were on maximum tolerated doses of evidence-based HF medical and electrical therapies for ≥ 3 months or had a documented contraindication or intolerance to medication use. Enrollment required subjects to have either one high-risk feature (reduced peak oxygen uptake, reduced 6-minute walk distance, elevated natriuretic peptide level, reduced serum sodium, decreased Seattle Heart Failure Model (SHFM) one-year predicted survival, or active UNOS status 2 heart transplant); one unplanned HF hospitalization in the prior 12 months with a lesser threshold for natriuretic peptide level; or two unplanned HF hospitalizations in the prior 12 months. Patients for whom durable MCS implantation was planned were excluded from this registry. Supplementary Tables 1 and 2 detail the inclusion and exclusion criteria. An independent Observational Study Monitoring Board oversaw the conduct of the REVIVAL study. The Institutional Review Board at each center approved the study. All subjects provided written informed consent before study participation.
Data Collection
For this report, patients were followed through the 12-month visit window (12 months ± 45 days) or until earlier death, transplant or implantation of a durable mechanical support (MCS) device. Demographics, clinical characteristics, laboratory, physical assessment, echocardiography, hemodynamic, functional status, and outcome data were collected in addition to quality-of-life surveys at the time of enrollment. INTERMACS Patient Profile was assigned at enrollment and at each study visit by the treating heart failure cardiologist, according to methods previously described.13 Outcomes prospectively collected at the time of occurrence include hospitalization, stroke, heart transplant (with UNOS status at the time of transplant), MCS, and death; UNOS 1A or 1B transplant, implantation of durable MCS or death constituted the primary outcome event.
Statistical Analysis
All statistical analyses were performed at the REVIVAL Data Coordinating Center at the University of Michigan. Continuous data are presented as median (25th-75th percentiles). Univariable comparisons were performed using the Cochran-Armitage trend test for categorical variables and the Spearman correlation coefficient for continuous variables.
The primary composite endpoint of death, implantable durable MCS (or total artificial heart), or urgent heart transplantation while in United Network of Organ Sharing (UNOS) Status 1A or 1B was evaluated at the 12-month study visit (365 ± 45 days after enrollment). Survival time for patients transplanted while in UNOS Status 2 (6 patients) was censored at transplantation for all analyses. The Kaplan-Meier method was used to estimate one-year survival free of implantable durable MCS and UNOS Status 1 transplant across INTERMACS Profiles at study enrollment. Cox proportional hazard regression was done to evaluate the impact of INTERMACS Profile at study enrollment on the primary composite endpoint. Competing risk modeling was done using nonparametric Kaplan-Meier method and cause-specific hazard regression to evaluate the impact and trend of INTERMACS Profile at study enrollment on times to implantable durable MCS placement, first unplanned heart failure hospitalization, and death. Hazard ratios and the 95% confidence intervals were calculated. Competing risk, nonparametric Kaplan-Meier method was also used to estimate the cumulative incidence function (CIF) of a patient moving into one of the three mutually exclusive events of death, implantable durable MCS, or urgent transplantation.14
A two-tailed P value of <0.05 was considered statistically significant without adjustment for multiple hypotheses testing. SAS version 9.4 statistical software (Cary, NC) was used for all analyses.
Results
INTERMACS Profiles and clinical features
The distribution of INTERMACS Profiles assigned by experienced heart failure cardiologists at enrollment included 33 (8%) in Profile 4, 83 (21%) in Profile 5, 155 (39%) in Profile 6, and 129 (32%) in Profile 7. Age, sex, and race did not differ by INTERMACS Profile (Table 1). However, among the 322 patients with data on education, sicker patients (i.e., those with lower INTERMACS Profiles) were less likely to have a college education, with 11 (41%) of INTERMACS Profile 4 patients being college educated compared with 86 (78%) of INTERMACS Profile 7 patients (p=0.0003).
Table 1.
Clinical characteristics
| INTERMACS Profile | |||||
|---|---|---|---|---|---|
| Total Cohort N = 400 |
4 n = 33 |
5 n = 83 |
6 n = 155 |
7 n =129 |
|
| Age, y | 62 (54–68) | 62 (54–67) | 60 (54–68) | 62 (54–69) | 62 (57–68) |
| Female | 99 (25%) | 7 (21%) | 19 (23%) | 42 (27%) | 31 (24%) |
| Race | |||||
| White | 277 (69) | 23 (70) | 63 (76) | 96 (62) | 95 (74) |
| African American/Black | 100 (25) | 10 (30) | 13 (16) | 48 (31) | 29 (23) |
| Hispanic or Latino Education*** (n=322) | 30 (8) | 2 (6) | 11 (13) | 11 (7) | 6 (5) |
| College education | 211 (66%) | 11 (41%) | 39 (63%) | 75 (61%) | 86 (78%) |
| No college education | 111 (34%) | 16 (59%) | 23 (37%) | 48 (39%) | 24 (22%) |
| Ischemic cause | 176 (44%) | 18 (55%) | 39 (47%) | 72 (47%) | 47 (36%) |
| Diabetes mellitus, n=399 | 151 (38%) | 12 (36%) | 35 (42%) | 61 (40%) | 43 (33%) |
| Atrial fibrillation | 41 (10%) | 5 (15%) | 6 (7%) | 20 (13%) | 10 (8%) |
| HF diagnosis >5 years | 228 (57%) | 17 (52%) | 47 (57%) | 79 (51%) | 85 (66%) |
| Prior heart txp evaluation | 111 (28%) | 10 (30%) | 25 (30%) | 45 (29%) | 31 (24%) |
| Ejection fraction, % | 20 (16–25) | 19 (15–20) | 20 (17–26) | 20 (16–25) | 20 (17–25) |
| 6MWT, m, n=368*** | 341 (281–403) | 303 (219–354) | 327 (263–372) | 342 (281–392) | 371 (317–447) |
| Gait speed, sec, n=366*** | 4.1 (3.5–5.0) | 4.3 (4.1−5.5) | 4.3 (3.8−5.4) | 4.1 (3.4−4.8) | 4.0 (3.3−4.6) |
| Hand grip strength, kg, n=381 | 66 (49–88) | 70 (45–91) | 68 (53–88) | 60 (43–83) | 77 (51–96) |
| NYHA Class*** | |||||
| I | 6 (2%) | 0 | 0 | 4 (3%) | 2 (2%) |
| II | 113 (28%) | 1 (3%) | 8 (10%) | 42 (27%) | 62 (48%) |
| III | 240 (60%) | 24 (73%) | 61 (74%) | 92 (59%) | 63 (49%) |
| IIIb | 31 (8%) | 6 (18%) | 9 (11%) | 14 (9%) | 2 (2%) |
| IV | 10 (3%) | 2 (6%) | 5 (6%) | 3 (2%) | 0 |
| # HF admissions in prior yr** | 1.1 ± 1.4 | 1.6 ± 1.6 | 1.1 ± 1.5 | 1.2 ± 1.4 | 0.8 ± 1.2 |
| UNOS Status 2 listing | 15% (60) | 12% (4) | 13% (11) | 19% (29) | 12% (16) |
| BMI, kg/m2, n=398 | 29 (25–34) | 27 (24–31) | 31 (26–36) | 29 (26–35) | 28 (25–33) |
| SBP, mmHg*, n=398 | 107 (98–117) | 104 (98–113) | 105 (98–114) | 106 (98–118) | 108 (99–120) |
| HR, bpm, n=398 | 72 (64–82) | 76 (70–83) | 73 (64–81) | 71 (65–84) | 71 (64–82) |
| JVP, cm H2O*, n=319 | 7 (6–9) | 8 (5−10) | 8 (6–10) | 7 (6–9) | 7 (6–8) |
| Serum sodium, mmol/L | 138 (136–140) | 138 (137−142) | 138 (136–140) | 139 (136–141) | 138 (136–140) |
| Creatinine, mg/dL | 1.3 (1.1–1.6) | 1.3 (1.1−1.7) | 1.3 (1.1–1.7) | 1.3 (1.1–1.6) | 1.3 (1.0–1.6) |
| Estimated GFR, ml/min | 55 (42–70) | 55 (44−73) | 55 (42−71) | 54 (42–70) | 56 (43–69) |
| Uric acid, mg/dL*, n=369 | 8.1 (6.7–9.6) | 9.0 (7.7−10.7) | 7.8 (6.7−9.6) | 8.2 (6.6–9.8) | 7.6 (6.5–9.0) |
| Seattle HF Model Score*** | 0.530 (−0.138–1.203) | 1.226 (0.670–2.037) | 0.839 (0.100–1.462) | 0.530 (−0.014–1.152) | 0.132 (−0.356–0.809) |
| Beta-blocker | 382 (96%) | 30 (91%) | 81 (98%) | 148 (96%) | 123 (95%) |
| ACE inhibitor | 177 (44%) | 17 (52%) | 34 (41%) | 68 (44%) | 58 (45%) |
| ARB | 101 (25%) | 6 (18%) | 23 (28%) | 46 (30%) | 26 (20%) |
| ARNI | 58 (15%) | 3 (9%) | 11 (13%) | 21 (14%) | 23 (18%) |
| ACEI, ARB, or ARNI | 336 (84%) | 26 (79%) | 68 (82%) | 135 (87%) | 107 (83%) |
| Hydralazine plus nitrate | 61 (15%) | 6 (18%) | 11 (13%) | 22 (14%) | 22 (17%) |
| MRA | 294 (74%) | 26 (79%) | 57 (69%) | 112 (72%) | 99 (77%) |
| Loop diuretic | 368 (92%) | 30 (91%) | 76 (92%) | 143 (92%) | 119 (92%) |
| CRT+ICD | 196 (49%) | 18 (55%) | 43 (52%) | 65 (42%) | 70 (54%) |
| ICD alone | 201 (50%) | 14 (42%) | 40 (48%) | 88 (57%) | 59 (46%) |
| History of appropriate ICD shock | 80 (20%) | 8 (24%) | 21 (25%) | 28 (18%) | 23 (18%) |
Values reported as median (25th – 75th percentile) or number (percent).
P < 0.05
P <0.01
P <0.001
Cochran-Armitage test for trend for categorical variables and Spearman test of association for continuous variables.
HF, Heart failure; Tx, transplant; 6MWT, 6 minute walk test; NYHA, New York Heart Association; SBP, systolic blood pressure; HR, heart rate; JVP, jugular venous pressure; SHFM, Seattle Heart Failure Model; yr, year; UNOS, United Network for Organ Sharing; BMI, body mass index; GFR, glomerular filtration rate.
Patients in different INTERMACS Profiles at enrollment had no difference in ejection fraction but did differ in NYHA functional class. While 63 (49%) of the INTERMACS Profile 7 patients were NYHA Class III and none were NYHA Class IV, 24 (73%) of the INTERMACS Profile 4 patients were NYHA Class III and 2 (6%) were NYHA Class IV (p < 0.0001; Table 1). Patients with lower INTERMACS Profiles at enrollment also had more frequent hospitalizations for heart failure in the year prior to study enrollment, with a mean of 1.6 prior HF hospitalizations in INTERMACS Profile 4 patients versus a mean of 0.8 hospitalizations in INTERMACS Profile 7 patients (p = 0.001).
Patients with lower INTERMACS Profiles at enrollment had other markers of decreased functional capacity and greater frailty (Table 1). The median 6-minute walk distance progressively decreased as INTERMACS Profiles worsened: 371, 342, 327, and 303 meters in INTERMACS Profiles 7, 6, 5, and 4, respectively (p<0.0001). Patients with lower INTERMACS Profiles also had greater impairment in gait speed (p=0.0002) (Table 1).
The administration of guideline-directed medical and electrical therapy was similar across assigned INTERMACS Profiles, as was the incidence of prior appropriate ICD shocks (Table 1). Patients with lower INTERMACS Profiles had decreased systolic blood pressure (Spearman rs= 0.11, p=0.03) and higher jugular venous pressure (Spearman rs= −0.12, p=0.03) (Table 1). Laboratory findings were generally comparable across INTERMACS profiles 4–7, with no differences in serum sodium, serum creatinine, and estimated glomerular filtration rate. Patients in INTERMACS Profile 4 did have higher uric acid levels (p=0.02) (Table 1).
Quality of life
Generic health-related quality of life as assessed by the EuroQol-5D instrument was impaired in this population of ambulatory advanced HF patients. REVIVAL subjects had “some” or “extreme” limitation in mobility (53%), usual activities (62%), self-care (10%), pain/discomfort (50%), and anxiety/depression (40%; Table 2). Patients with lower baseline INTERMACS Profiles more frequently noted “some” or “extreme” limitation in mobility, usual activity, and self-care. The degrees of pain/discomfort and anxiety/depression were comparable across INTERMACS Profiles. Patient assessment of their overall quality of life using the EuroQol visual analog scale score had a median of 65, with lower median scores (indicating worse quality of life) in patients in lower INTERMACS Profiles (p=0.0002) (Table 2). Similar results were noted with the HF-specific Kansas City Cardiomyopathy Questionnaire, where patients with lower INTERMACS Profiles had worse overall scores, clinical scores, total symptom scores, and physical limitations (p<0.0001) (Table 2).
Table 2.
Quality of life questionnaires
| INTERMACS Profile | |||||
|---|---|---|---|---|---|
| Total Cohort n= 400 |
4 n = 33 |
5 n = 83 |
6 n = 155 |
7 n = 129 |
|
| EuroQol-5D (Some problems or more) | |||||
| Mobility**, n=372 | 198 (53%) | 20 (63%) | 54 (69%) | 77 (54%) | 47 (40%) |
| Self-Care*, n=372 | 37 (10%) | 2 (6%) | 13 (17%) | 17 (12%) | 5 (4%) |
| Usual Activities***, n=372 | 232 (62%) | 25 (78%) | 58 (74%) | 96 (67%) | 53 (45%) |
| Pain/ Discomfort, n=372 | 186 (50%) | 15 (47%) | 43 (55%) | 71 (50%) | 57 (48%) |
| Anxiety/ Depression, n=372 | 148 (40%) | 13 (41%) | 39 (50%) | 54 (38%) | 42 (35%) |
| EQ-5D index score, n=372*** | 0.82 (0.71–0.86) | 0.80 (0.71–0.85) | 0.79 (0.71–.83) | 0.83 (0.71–0.86) | 0.83 (0.78–1.00) |
| Health State VAS***, n=367 | 65 (50–75) | 60 (46–73) | 60 (50–70) | 65 (50–75) | 70 (60–80) |
| Kansas City Cardiomyopathy Questionnaire, n=373 | |||||
| Overall score*** | 64 (48–78) | 60 (43–70) | 53 (38−66) | 63 (48−78) | 73 (60–86) |
| Clinical score*** | 69 (55–83) | 63 (48–72) | 62 (45−74) | 71 (52−82) | 77 (64–91) |
| Total symptom score*** | 72 (55–86) | 67 (43–78) | 66 (43−77) | 73 (54−87) | 79 (65–94) |
| Physical limitation*** | 67 (50–83) | 63 (42–77) | 54 (42−75) | 67 (46−83) | 79 (60–92) |
| Personal Health Questionnaire score***, n=367 | 6 (2–10) | 6 (4–11) | 8 (4–12) | 6 (2–9) | 5 (2−8) |
| State-trait-anxiety inventory score, n=363 | 35 (26−46) | 33 (23−45) | 39 (29−49) | 35 (27−45) | 34 (26−45) |
Values reported as median (25th – 75th percentile) or number (percent).
P < 0.05
P <0.01
P <0.001.
Cochran-Armitage test for trend for categorical variables and Spearman test of association for continuous variables
VAS, visual analog scale
INTERMACS Profiles over time
Among 246 survivors on medical therapy who presented for follow-up, the distribution of INTERMACS Profiles at baseline, 2, 6, and 12 months is shown in Figure 1. The distribution of INTERMACS Profiles of survivors improved over time with almost 80% of patients demonstrating INTERMACS Profiles 6 and 7 after 2, 6, and 12 months of enrollment. This in part reflects survivorship bias-patients with lower INTERMACS Profiles at enrollment more frequently required LVAD implantation or urgent transplant – and the relatively stable INTERMACS Profiles of those who did not require advanced therapies during the follow-up period.
Figure 1.
Distribution of INTERMACS Profiles at enrollment, 2, 6, and 12 months among survivors on medical therapy who presented for clinical follow up visits.
Outcomes by INTERMACS Profile
Of the 400 subjects enrolled in the study, two (2) provided no follow-up information beyond enrollment, leaving 398 evaluable subjects, among whom 90 combined primary outcome events occurred: 30 (8%) deaths, 14 (4%) urgent transplants, and 46 (1%) durable MCS implants (44 LVADs and 2 BiVADs). Among these MCS recipients, INTERMACS Patient Profiles at the time of implant were 11% Profile 1, 17% Profile 2, 43% Profile 3, 22% Profile 4 and 7% Profile 5 and none in Profile 6–7. The competing events curve depicting the primary outcomes of death, durable MCS, and urgent transplantation is shown in Figure 2. This includes censored data on 25 evaluable subjects without a primary outcome event whose follow-up did not extend through the 1-year visit window, including 4 subjects who were lost to follow-up (median follow-up 234 (186–303) days), 15 subjects who withdrew from the study (median follow-up 165 (77–238) days), and 6 subjects who received UNOS status 2 transplants (median follow-up 111 (32–152) days).
Figure 2.
Competing events. Mutually exclusive endpoints of death, durable MCS, or urgent transplant were assessed through the end of the 12-month visit window. The cumulative incidence of each event at 410 days is shown.
Patients with lower INTERMACS Profiles at enrollment had a significantly higher incidence of the primary composite endpoint of death, durable MCS, or urgent transplant (Figure 3; p = 0.002). At the end of the 12-month study visit window (410 days), the primary composite endpoint rate was 40%, 29%, 25%, and 14% for subjects with INTERMACS Profiles 4, 5, 6, and 7, respectively (Figure 3).
Figure 3.
Composite primary endpoint of death, durable MCS, or urgent transplant according to INTERMACS Profile at enrollment. Estimated survival was assessed through the end of the 12-month visit window using the Cox proportional hazard method. Incidence of the primary endpoint at 410 days is shown. The composite primary endpoint was significantly different by INTERMACS Profile (P = 0.0002).
The risk of durable MCS placement by the 12-month study visit was also significantly higher for patients in lower INTERMACS Profiles (Figure 4; p = 0.011). The risk of death alone was not significantly associated with INTERMACS Profile. By the end of the 12-month visit window, the CIF estimates for death were 12%, 11%, 8%, and 5% for subjects with INTERMACS Profiles 4, 5, 6, and 7, respectively (p=0.29). The competing risk of urgent transplant by INTERMACS Profile could not be assessed as there were only 14 urgent transplants in this cohort within the first year.
Figure 4.
Durable MCS placement according to INTERMACS Profile at enrollment. Incidence of durable MCS placement was measured through the end of the 12-month visit window using the competing risks model. Incidence of durable MCS placement at 410 days is shown. Incidence of durable MCS placement was significantly different by INTERMACS Profile ( P = 0.011).
Heart failure hospitalization was less likely in patients with higher baseline INTERMACS Profiles (Figure 5; p = 0.0005). By the 12-month visit window, first HF hospitalization had CIF estimates of 37%, 47%, 28%, and 24% in INTERMACS Profiles 4, 5, 6, and 7, respectively (Figure 5).
Figure 5.
Time to first heart failure hospitalization by INTERMACS Profile. This was assessed through the end of the 12-month visit window using the competing risks model. Incidence of HF hospitalization at 410 days is shown. Incidence of first heart failure hospitalization was significantly different by INTERMACS Profile (P = 0.0005).
Figure 6 presents the risk of the primary composite outcome, MCS and first heart failure hospitalization by enrollment INTERMACS Profile by the 12-month visit window. Patients in INTERMACS Profile 4, 5, and 6 had a 3.7-fold, 2.2-fold, and 1.8-fold higher risk of death, durable MCS, or urgent transplant compared to patients in INTERMACS Profile 7. Patients in INTERMACS Profile 4 and 5 had 4.3-fold and 3.1-fold increased risk of durable MCS placement compared to patients in INTERMACS Profile 7. Patients in INTERMACS Profile 4 and 5 had 2.0-fold and 2.5-fold increased risk of first HF hospitalization compared to patients in INTERMACS Profile 7.
Figure 6.
Risk of the primary composite outcome, MCS, and first heart failure hospitalization by INTERMACS Profile. This was assessed during the 12-month visit window. Cox proportional hazard regression was used to estimate the risk of the primary composite outcome. Competing risk modeling, using cause-specific hazard regression, was used to estimate the risk of MCS and first heart failure hospitalization.
Discussion
Among ambulatory patients with advanced HF, we observed a significant gradient of disease severity across INTERMACS Profiles assigned by experienced heart failure cardiologists at study enrollment. Significant differences between the Profiles in key parameters, such as NYHA functional class and 6-minute walk distance, help to validate the Profiles. A lower baseline INTERMACS Profile was associated with greater functional limitation, greater risk at one year of the composite endpoint (death, durable MCS, or urgent transplant), and greater risk of HF hospitalization.
The present study represents the largest prospectively observed cohort of high-risk ambulatory advanced HF patients in INTERMACS Profiles 4 to 7 being aggressively medically optimized at VAD/transplant centers. The INTERMACS Profile reflects an integrated clinical assessment and provides a graded measure of risk. The impact of assigned INTERMACS Profiles at enrollment on 1-year outcomes provides clinically actionable information for ambulatory advanced heart failure patients, as recommended in the AHA Scientific Statement on decision-making in advanced heart failure.12 An Annual Heart Failure Review involves assessment of patient symptom burden and quality of life to inform prognosis. The INTERMACS profile assignments reported here can thus offer insight into this estimated prognosis for ambulatory advanced heart failure within a 1-year time horizon.
Notably, while there was a monotonic relationship between INTERMACS profiles and the primary combined endpoint of death, durable MCS, or urgent transplant, the relationship between INTERMACS profiles and quality of life and heart failure hospitalizations demonstrated some variation. We could not identify a statistically significant relationship between INTERMACS profile and death alone. However, because the study excluded patients with LVAD contraindications, we suspect that death was prevented by LVAD and urgent transplant in many of these patients. INTERMACS Profile 5 patients had numerically worse scores on several quality of life measures and they also had a higher incidence of heart failure hospitalizations than patients in other INTERMACS Profiles. These findings may be due to the play of chance, especially given the smaller number of patients in Profiles 4 and 5. However, perceived worse quality of life could result in more heart failure hospitalizations, as patients who feel worse may be more likely to seek medical attention resulting in heart failure admissions. However, while the INTERMACS Profile 5 patients felt worse and were more often hospitalized, they had a lower incidence of death, durable MCS, or urgent transplantation than INTERMACS Profile 4 patients, suggesting that the treating physicians’ perception of their illness severity (INTERMACS Profile 4 vs 5) might have been a better indicator of overall prognosis than the patients’ perception of their symptoms.
Comparable findings on the relationship between INTERMACS Profiles and outcomes were noted in the smaller Medical Arm of Mechanically Assisted Circulatory Support (MedaMACS) study15 of 166 patients, as well as in a larger single-center study.16 The latter included 969 ambulatory heart failure patients retrospectively identified with INTERMACS Profiles adjudicated by chart review and noted that lower INTERMACS profiles confer an increased risk of subsequent 3-year mortality and heart failure hospitalization.16 Our findings are complementary and offer additional insights. First, in our study, the treating clinician assigned INTERMACS Profiles prospectively, confirming the validity of prospective assignments as well as the accurate phenotyping of our cohort. Second, in our study, 71% of subjects were INTERMACS Profiles 6 and 7 compared with 46% of subjects in MedaMACS15, allowing us to extend the predictive power of INTERMACS Profiles to less advanced ambulatory chronic HF patients. Finally, our study spanned 21 hospitals, confirming the generalizability of the findings to multiple VAD/transplant centers.
The ROADMAP study also included a study population of ambulatory non-inotrope dependent advanced HF patients, but with a critical difference: in ROADMAP, all patients considered the choice between LVAD and ongoing optimal medical management before follow-up began3,17. As expected, patients choosing to receive an LVAD were more likely to be in INTERMACS Profile 4 and had lower baseline quality of life. Adverse events were higher with LVAD in the first year, though not statistically different by year 2, driven primarily by bleeding events. LVAD recipients were more likely to be hospitalized, though there was no difference in 1- and 2-year survival in the intention-to-treat analysis. LVAD recipients in ROADMAP were more likely than medically treated patients to experience clinically meaningful improvements in health-related quality of life, functional status and emotional well-being. While the REVIVAL registry does not offer a comparison (LVAD) group, it does offer complementary information to ROADMAP, including nearly 4 times as many medically managed patients across a broader phenotypic range, offering better understanding into the natural history of ambulatory advanced HF.
While a substantial proportion of patients with lower INTERMACS Profiles required VAD or urgent transplant in the year following enrollment, over 90% of survivors were assigned Profiles 5–7 at the 1-year visit. In 2015–2016, of the 5400 registered VAD implants in INTERMACS, only 1.6% were in patients with Profiles 5–7 at the time of LVAD implantation.2 Much of this reluctance stems from the unacceptably high adverse event rates with MCS devices. Additional expansion of durable MCS into ambulatory patients in INTERMACS Profiles 5–7 HF will require not just improved technology with fewer adverse events, but also better prognostic tools to identify those at greatest risk of dying without MCS, and those who are most likely to have favorable outcomes with VAD support. Recent attention to adverse events on LVAD therapy—particularly to pump thrombosis and stroke—prevented equipoise for randomizing ambulatory advanced heart failure patients in the REVIVE-IT trial to medical versus currently approved device therapy.19 As a prospective randomized trial is likely not feasible in this population, we continue to rely on judgment and experience in the triage of INTERMACS Profile 5–7 patients to advanced HF therapies.
Whether durable MCS therapy should be considered for HF patients in INTERMACS Profile 5–7 remains unclear. An analysis of the previously described MedaMACS registry matched with Profile 4 to 7 patients with LVADs from the INTERMACS registry indicated that survival was similar for medical and LVAD therapy in the overall cohort, which included the lower severity Profiles 6 and 7, but survival was better with LVAD therapy among patients in Profiles 4 and 5.18 The current findings are an important extension with a larger samples comprising more Profile 6–7 patients. This allows us to demonstrate that Profiles 6 and 7 have distinct and clinically relevant trajectories and better map the transition from Stage C to D heart failure.
The knowledge that those patients assigned to INTERMACS Profiles 4–5 in particular are at increased risk of death and LVAD implantation by one year suggests that these patients should be flagged by treating clinicians for more close follow-up by advanced heart failure cardiologists for early signs of deterioration and preparedness discussions about their goals, values, and preferences for advanced heart failure therapies, especially LVAD support.
This study has important limitations. First, we were unable to assess inter-observer variability for the assignment of INTERMACS profile. The Profiles were assigned by experienced advanced heart failure physicians practicing at high volume centers and the observed outcomes have the expected dose-response relationship. Moreover, If present, interobserver variability would be expected to bias the observed findings to the null rather than improve the predictability of the profiles. Second, this observational analysis focused on the risk of death, urgent transplant or LVAD implantation by INTERMACS Profile. As outcomes were not collected after the endpoint was achieved, we cannot speculate on the benefit of LVAD by INTERMACS Profile, nor can we speculate on the impact of INTERMACS Profiles on long-term quality of life. Third, these high-risk ambulatory advanced HF patients were enrolled at quaternary care VAD/cardiac transplant centers, so results may not be applicable outside this setting. Even so, INTERMACS profiles at the time of durable MCS were similar to those recorded in INTERMACS20 suggesting implant thresholds similar to the larger MCS community. In addition, absolute event rates in the different INTERMACS Profiles should be interpreted contingent upon study entry criteria and within the context of overlapping clinical features between profiles. Finally, because no multiple comparison adjustments were made to p-values, the results should be interpreted cautiously.
Through the REVIVAL registry, we sought to characterize the clinical features and outcomes of ambulatory patients with advanced HF who demonstrate high-risk features while on a strategy of oral medical therapy at MCS/transplant centers. Among ambulatory patients with advanced HF, the INTERMACS Profile as assigned by experienced heart failure cardiologists, a simple, clinically-derived classification system, provides prognostic information. A lower INTERMACS Profile was associated with greater functional limitation and an increased risk of death, durable MCS or urgent cardiac transplant and heart failure hospitalization at one year. As INTERMACS profiling provides a convenient shorthand that encapsulates relevant prognostic information across multiple clinical domains, these profiles may assist in triaging ambulatory patients to advanced HF therapies.
Supplementary Material
Acknowledgements
Supported by funding from the National Institutes of Health, National Heart, Lung, and Blood Institute (NHLBI Contract Number: HHSN268201100026C) for REVIVAL and the National Center for Advancing Translational Sciences (NCATS Grant Number: UL1TR002240) for the Michigan Institute for Clinical and Health Research (MICHR). The views expressed in this manuscript are those of the authors and do not necessarily represent the views of the National Heart, Lung, and Blood Institute; the National Institutes of Health; or the U.S. Department of Health and Human Services.
The authors wish to acknowledge the singular contributions of Laurie Rigan to the REVIVAL study.
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 citable 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.
Disclosures
Palak Shah reports Grant Support: American Heart Association/Enduring Hearts Scientist Development Grant, Merck, Haemonetrics, and Medtronic. Consulting: NuPulse CV and Ortho Clinical Diagnostics.
Keyur Shah reports being a consultant for Medtronic and Syncardia.
Jeff Teuteberg reports speaking and ad board with Medtronic.
Douglas Horstmanshof reports being a speaker/research grant for Abbott Medical.
Keith Aaronson reports the following related to the submitted work: NIH/NHLBI (contract) and outside the submitted work: Medtronic (research support and consultant [Independent Physician Quality Panel member), Abbott (research support), Procyrion (consultant), NuPulseCV (consultant), Bioventrix (research support).
No conflicts: Michelle Kittleson, Anuradha Lala, Rhondalyn McLean, Salpy Pamboukian, Jennifer Thibodeau, Nisha Gilotra, Wendy Taddei-Peters, Thomas Cascino, Neal Jeffries, Lynne Stevenson, Douglas Mann, Garrick Stewart, Blair Richards, Shokoufeh Khalatbari.
Contributor Information
Michelle M Kittleson, Smidt Heart Institute, Cedars-Sinai Medical Center, Los Angeles, CA.
Palak Shah, Department of Medicine, Inova Heart and Vascular Institute, Falls Church, VA.
Anuradha Lala, Department of Internal Medicine, Mount Sinai Hospital, New York, NY.
Rhondalyn C McLean, Department of Medicine, University of Pennsylvania, Philadelphia, PA.
Salpy Pamboukian, Department of Medicine, University of Alabama, Birmingham, AL.
Douglas A Horstmanshof, IACC Advanced Cardiac Care Dept, INTEGRIS Baptist Medical Center, Oklahoma City, OK.
Jennifer Thibodeau, Dept of Internal Medicine, University of Texas SW Medical Center, Dallas, TX.
Keyur Shah, Department of Internal Medicine, Virginia Commonwealth University, Richmond, VA.
Jeffrey Teuteberg, Department of Medicine, Stanford University, Palo Alto, CA.
Nisha A Gilotra, Department of Cardiology, Johns Hopkins Hospital, Baltimore, MD.
Wendy C Taddei-Peters, Div of Cardiovascular Sciences, National Heart, Lung, and Blood Institute, Bethesda, MD.
Thomas M Cascino, Division of Cardiovascular Medicine, University of Michigan, Ann Arbor, MI.
Blair Richards, Michigan Institute for Clinical and Health Research, University of Michigan, Ann Arbor, MI.
Shokoufeh Khalatbari, Michigan Institute for Clinical and Health Research, University of Michigan, Ann Arbor, MI.
Neal Jeffries, Division of Cardiovascular Sciences, National Heart, Lung, and Blood Institute, Bethesda, MD.
Lynne W Stevenson, Department of Medicine, Vanderbilt University, Nashville, TN.
Douglas Mann, Department of Internal Medicine, Washington University, St. Louis, MO.
Keith D Aaronson, Division of Cardiovascular Medicine, University of Michigan, Ann Arbor, MI.
Garrick C Stewart, Cardiovascular Medicine, Brigham and Women’s Hospital, Boston, MA.
REFERENCES
- 1.McMurray JJ. Clinical practice. Systolic heart failure. The New England journal of medicine 2010;362:228–38. [DOI] [PubMed] [Google Scholar]
- 2.Kirklin JK, Pagani FD, Kormos RL, et al. Eighth annual INTERMACS report: Special focus on framing the impact of adverse events. J Heart Lung Transplant 2017;36:1080–6. [DOI] [PubMed] [Google Scholar]
- 3.Estep JD, Starling RC, Horstmanshof DA, et al. Risk Assessment and Comparative Effectiveness of Left Ventricular Assist Device and Medical Management in Ambulatory Heart Failure Patients: Results From the ROADMAP Study. Journal of the American College of Cardiology 2015;66:1747–61. [DOI] [PubMed] [Google Scholar]
- 4.Cowger J, Shah P, Stulak J, et al. INTERMACS profiles and modifiers: Heterogeneity of patient classification and the impact of modifiers on predicting patient outcome. J Heart Lung Transplant 2015. [DOI] [PMC free article] [PubMed]
- 5.Baldwin JT, Mann DL. NHLBI’s program for VAD therapy for moderately advanced heart failure: the REVIVE-IT pilot trial. Journal of cardiac failure 2010;16:855–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Boyle AJ, Ascheim DD, Russo MJ, et al. Clinical outcomes for continuous-flow left ventricular assist device patients stratified by pre-operative INTERMACS classification. J Heart Lung Transplant 2011;30:402–7. [DOI] [PubMed] [Google Scholar]
- 7.Jorde UP, Kushwaha SS, Tatooles AJ, et al. Results of the destination therapy post-food and drug administration approval study with a continuous flow left ventricular assist device: a prospective study using the INTERMACS registry (Interagency Registry for Mechanically Assisted Circulatory Support). Journal of the American College of Cardiology 2014;63:1751–7. [DOI] [PubMed] [Google Scholar]
- 8.Gorodeski EZ, Chu EC, Chow CH, Levy WC, Hsich E, Starling RC. Application of the Seattle Heart Failure Model in ambulatory patients presented to an advanced heart failure therapeutics committee. Circulation Heart failure 2010;3:706–14. [DOI] [PubMed] [Google Scholar]
- 9.Sartipy U, Goda A, Yuzefpolskaya M, Mancini DM, Lund LH. Utility of the Seattle Heart Failure Model in patients with cardiac resynchronization therapy and implantable cardioverter defibrillator referred for heart transplantation. American heart journal 2014;168:325–31. [DOI] [PubMed] [Google Scholar]
- 10.Adlbrecht C, Hulsmann M, Neuhold S, Strunk G, Pacher R. Prognostic utility of the Seattle Heart Failure Score and amino terminal pro B-type natriuretic peptide in varying stages of systolic heart failure. J Heart Lung Transplant 2013;32:533–8. [DOI] [PubMed] [Google Scholar]
- 11.Zafrir B, Goren Y, Paz H, et al. Risk score model for predicting mortality in advanced heart failure patients followed in a heart failure clinic. Congestive heart failure 2012;18:254–61. [DOI] [PubMed] [Google Scholar]
- 12.Allen LA, Stevenson LW, Grady KL, et al. Decision making in advanced heart failure: a scientific statement from the American Heart Association. Circulation 2012;125:1928–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Stevenson LW, Pagani FD, Young JB, et al. INTERMACS Profiles of Advanced Heart Failure: The Current Picture. The Journal of Heart and Lung Transplantation 2009;28:535–41. [DOI] [PubMed] [Google Scholar]
- 14.McGiffin DC, Naftel DC, Kirklin JK, et al. Predicting outcome after listing for heart transplantation in children: comparison of Kaplan-Meier and parametric competing risk analysis. Pediatric Heart Transplant Study Group. J Heart Lung Transplant 1997;16:713–22. [PubMed] [Google Scholar]
- 15.Stewart GC, Kittleson MM, Patel PC, et al. INTERMACS (Interagency Registry for Mechanically Assisted Circulatory Support) Profiling Identifies Ambulatory Patients at High Risk on Medical Therapy After Hospitalizations for Heart Failure. Circulation Heart failure 2016;9. [DOI] [PubMed] [Google Scholar]
- 16.Samman-Tahhan A, Hedley JS, McCue AA, et al. INTERMACS Profiles and Outcomes Among Non-Inotrope-Dependent Outpatients With Heart Failure and Reduced Ejection Fraction. JACC Heart failure 2018;6:743–53. [DOI] [PubMed] [Google Scholar]
- 17.Starling RC, Estep JD, Horstmanshof DA, et al. Risk Assessment and Comparative Effectiveness of Left Ventricular Assist Device and Medical Management in Ambulatory Heart Failure Patients: The ROADMAP Study 2-Year Results. JACC Heart failure 2017;5:518–27. [DOI] [PubMed] [Google Scholar]
- 18.Ambardekar AV, Kittleson MM, Palardy M, et al. Outcomes with ambulatory advanced heart failure from the Medical Arm of Mechanically Assisted Circulatory Support (MedaMACS) Registry. The Journal of Heart and Lung Transplantation 2019;38:408–17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Pagani FD, Aaronson KD, Kormos R, et al. The NHLBI REVIVE-IT study: Understanding its discontinuation in the context of current left ventricular assist device therapy. The Journal of Heart and Lung Transplantation 2016;35:1277–83. [DOI] [PubMed] [Google Scholar]
- 20.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]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.