Abstract
Background:
Heart failure (HF) guidelines recommend assessment of left ventricular ejection fraction (LVEF) to classify patients and guide therapy implementation. However, LVEF alone may be insufficient to adequately characterize patients with HF, especially those with mildly reduced or preserved LVEF (HFmrEF/HFpEF). Recommendations on additional testing are lacking and there are limited data on use of echocardiographic features beyond LVEF in patients with HFmrEF/HFpEF.
Methods:
In patients with HFmrEF/HFpEF identified in a large US healthcare system, the association of the following metrics with mortality was evaluated: LV global longitudinal strain (LV GLS>−16), left atrial volume index (LAVi>28 ml/m2), left ventricular hypertrophy (LVH), and E/e’>13 & e’<9. A multivariable model for mortality was constructed including age, sex, and key co-morbidities followed by stepwise selection of echocardiographic features. Characteristics and outcomes of subgroups with normal vs. abnormal LV GLS and/or LVEF were evaluated.
Results:
Among 2,337 patients with complete echocardiographic data assessed between 2017–2020, the following features were associated with all-cause mortality on univariate analysis over 3 years of follow-up: E/e’ + e’, LV GLS, LAVi (all p<0.01). In the multivariable model (c-index=0.65), only abnormal LV GLS was independently associated with all-cause mortality (HR 1.35 [95% CI 1.11–1.63], p=0.002). Among patients with LVEF>55%, 498/1255 (40%) demonstrated abnormal LV GLS. Regardless of specific LVEF, patients with abnormal LV GLS demonstrated a higher burden of multiple co-morbidities and higher event rates compared with patients with normal LV GLS.
Conclusions:
In a large, real-world HFmrEF/HFpEF population, echocardiographic features, led by LV GLS, were associated with adverse outcomes irrespective of LVEF. A large proportion of patients demonstrate adverse myocardial function by LV GLS despite preserved LVEF and may represent a key cohort of interest for HF medical therapies and future clinical studies.
Keywords: heart failure with preserved ejection fraction, echocardiography, global longitudinal strain
INTRODUCTION
Heart failure (HF) guidelines recommend assessment of left ventricular ejection fraction (LVEF) to classify patients and guide therapy implementation.1,2 LVEF has been further codified in the guidelines with the development of the category of heart failure with mildly reduced ejection fraction (HFmrEF), which now determines key differences in therapy recommendations by granular LVEF categories.1–3 However, LVEF is constrained by several limitations and may be insufficient to adequately characterize patients with HF, especially those with mildly reduced or preserved LVEF (HFmrEF/HFpEF).4–7 While deeper phenotyping may aid in making decisions on medical treatments options and/or prognosis, recommendations on additional testing are lacking.
In prior research, echocardiographic features beyond LVEF have been used to discriminate patients with HFpEF from healthy populations but remain under-investigated in the patients with LVEF>40% encountered in routine practice. For instance, global longitudinal strain (GLS) is a sensitive marker of myocardial deformation and LV function.8 The prognostic utility of LV GLS for hospitalization/mortality outcomes in HFpEF has been suggested by several small secondary analyses of clinical trial datasets (~200–500 patients)9–11 and in single-center studies (~300–400 patients).12,13 Yet, there remains variability and equipoise on this question when evaluated by meta-analysis.14 Additionally, other echocardiographic features such as left ventricular hypertrophy (LVH), E/e’ ratio, e’ velocity, and pulmonary artery systolic pressure (PASP) were associated with both worse outcomes and symptoms in HFpEF in secondary trial analyses and small, single-center studies.15,16 Still, while echocardiographic features beyond LVEF (i.e. E/e’, LV GLS, PASP) are referenced in the guidelines,1,2 LVEF remains the primary recommended echocardiographic metric of phenotyping patients with HFmrEF/HFpEF in clinical practice. In this setting, we aimed to analyze the prognostic utility of multiple additional echocardiographic metrics, including LVEF and LV GLS, in a large, contemporary, real-world, HFmrEF/HFpEF population.
METHODS
Study design:
This study was a retrospective cohort study, analyzing existing electronic health record (EHR) data collected from patients receiving medical care at a large multi-hospital health system, Duke University Health System (DUHS). DUHS is comprised of Duke University Hospital, 2 large, regional community-based hospitals, and a wide array of outpatient clinics. Clinical data and outcomes were assessed using records collected from January 1, 2016 (to determine cohort eligibility) through June 30, 2021 (to provide at least 6 months of subsequent follow-up). Sequential patients meeting eligibility criteria between January 1, 2017 – December 31, 2020 were included in the cohort. The data that support the findings of this study are available from the corresponding author upon reasonable request.
Data Sources and Ascertainment
Transthoracic echocardiogram data were sourced from the C3 clinical database, which includes echocardiogram procedures conducted at Duke University Hospital, and satellite outpatient clinics. Clinical variables and outcomes were ascertained from DUHS EHR data abstracted using the Duke Enterprise Data Unified Content Explorer (DEDUCE) query tool. DEDUCE is a well-established, web-based, clinical research query tool capable of searching various facets of the current and legacy electronic medical record systems.17,18 Mortality was ascertained using the death date recorded for the patient in DEDUCE. This corresponds to the death date recorded in the DUHS EHR and is sourced from deaths recorded as part of health care delivery, deaths reported in the Social Security Administration Death Master File (provided through the National Technical Information Service) and from North Carolina death certificates.19,20 Prior internal analyses have demonstrated high sensitivity (94%) and specificity (>99%) in ascertaining mortality by this methodology compared to other gold standard techniques including the National Death Index and call center follow-up. Data from C3 and DEDUCE were linked by the patient’s DUHS electronic medical record number and dates of service. Data were extracted from source systems in December 2021, but follow-up was censored in June 30, 2021, to allow for up to 5 months latency in recording of deaths and clinical data.19,20
Study Population:
The target population was defined as all adults (age ≥18 years) with a transthoracic echocardiogram with left ventricular ejection fraction >40% between January 1, 2017 – December 31, 2020. The HFmrEF/HFpEF study population was further refined to patients with at least one distinct HF ICD 9/10 billing diagnosis (coded in any primary or secondary position) within 12 months prior to or 7 days after the qualifying echocardiogram. The following patients were excluded: prior heart transplant, presence of left ventricular assist device, severe aortic stenosis, severe mitral stenosis, and HFpEF mimickers such as infiltrative cardiomyopathies and hypertrophic cardiomyopathy. A full list of excluded HFpEF mimickers by ICD 9/10 codes is provided in Table S1. If a patient had multiple qualifying echocardiograms during the study period, they were included in the cohort at their first qualifying echocardiogram. Baseline was defined as the date of the first qualifying echocardiogram.
Study Outcomes:
The primary outcome was all-cause mortality and secondary outcomes were cardiovascular (CV) mortality and HF hospitalization. Outcomes were assessed at 90 days, 1 year, and 3 years after the qualifying echocardiogram. The primary outcome was selected given the following: 1) the importance of multiple cardiac and non-cardiac causes of morbidity/mortality in patients with HFmrEF/HFpEF, and 2) the reliability of determining the presence or absence of all-cause mortality as opposed to abstracting disease-specific mortality or disease-specific hospitalization (which are both inherently limited in retrospective EHR analyses). The secondary outcomes of CV mortality and HF hospitalization were selected in a similar manner to recent large HFpEF trials.21,22 HF hospitalization was defined as hospitalization with HF ICD 9/10 billing diagnosis, coded in primary or secondary positions. CV mortality was defined as mortality preceded within 30 days by a CV procedure or within 30 days of an outpatient visit, hospital admission, or ED visit with a CV diagnosis code.
Echocardiographic features beyond LVEF:
In addition to stratification by LVEF, the following metrics were evaluated for evidence of structural heart disease concerning for pathologic HFpEF: abnormal LV GLS (defined as > −16), left atrial enlargement (defined as left atrial volume index > 28 ml/m2), left ventricular hypertrophy (defined as any degree of LVH ≥ mild, including posterior or septal wall thickness ≥ 1.1 cm), and diastolic dysfunction (defined as the combination of both E/e’>13 + e’<9). These features were determined based on existing literature supporting their relevance in HFpEF and relevant HF guideline statements.2,23–29 All echocardiographic measures were obtained and interpreted in routine clinical practice and therefore were not blinded compared to other metrics or clinical patient characteristics. Average LV GLS from GE EchoPAC software across a 17-segment model was utilized as primary measure when available; length of line method from TomTec software was utilized as secondary metric for LV GLS when average measure was not available. Echocardiography was performed with GE (Vivid 9; GE Vingmed Ultrasound AS, Horten, Norway) or Philips (iE33; Philips Ultrasound, Bothell, WA) ultrasound systems, and LV GLS was measured with either vendor-specific software (GE EchoPAC; GE Vingmed Ultrasound AS) or vendor-independent software (VIS; TomTec 2D Cardiac Performance Analysis, Munich, Germany). LV GLS measures have demonstrated consistent reproducibility across these systems and software packages in prior analyses.30 It should be noted that the standard reporting protocol typically includes interpretation of LVEF, LVH, and left atrial size, but these features as well as GLS are not present in all echocardiogram reports depending on clinical scenarios and ability to obtain and interpret appropriate images.
Statistical analysis:
Patient characteristics were presented with categorical variables as counts/percentages; continuous variables as mean (SD) or median (Q1, Q3). For descriptive analyses, patients were grouped by LVEF (>40–44%, 45–49%, 50–55%, >55%) and by dividing patients based on normal/abnormal LVEF (>55/≤55%) in combination with normal/abnormal LV GLS (≤−16/>16). Differences between groups were assessed using the Chi-Square/Exact test for categorical comparisons as appropriate, and the Kruskal-Wallis test for comparison of continuous variables. The extent of available (or missing) data was summarized. Missing data was excluded from descriptive summaries.
Outcomes were evaluated as time to first event. Events were ascertained through 3 years after baseline, with early censoring at the end of data collection (June, 30 2021) or if a patient was last known to be alive but suspected lost to the health system. The patient was suspected lost to the health system 3 months after their latest diagnosis/procedure code recorded when data were accessed, if this was before the 3-year follow-up period or death date. The cumulative incidence of outcome events was summarized and compared between groups using the Kaplan-Meier method and log-rank test for all-cause mortality, and the cumulative incidence function estimate and Gray’s test for CV mortality and for HF hospitalization, accounting for the competing risk of death. To quantify the increased risk of 3-year mortality associated with each echocardiographic abnormality, the hazard ratio (95% confidence interval), p-value and concordance index were estimated using univariable Cox proportional hazards regression. The concordance index measures how often the ranking of patients by observed outcomes (time-to-death) is concordant with presence/absence of the abnormality. Estimated 3-year cumulative mortality risk was plotted across the range of each echocardiographic parameter, using estimates obtained from a Cox model with a restricted cubic spline transformation of the continuous echocardiographic parameter. In multivariable analysis, stepwise variable selection (p>0.15 to enter model, p<0.05 to stay in model) was used to identify echocardiographic abnormalities with statistically significant associations when added to a model for mortality including basic demographics and comorbidities (age, sex, body mass index, diabetes, renal disease, atrial fibrillation/flutter). The full multivariable model (without stepwise variable selection) was also constructed as a sensitivity analysis along with interaction by LV GLS software. Additional sensitivity analysis was performed with covariables of hypertension and history of coronary artery disease.
Statistical analyses were performed using SAS v9.4 (SAS Institute, Cary, NC). All hypothesis testing was two-sided with statistical significance determined at p<0.05, without any adjustment for multiple hypothesis testing. The Institutional Review Board of the Duke University Health System approved this study; no informed consent was required.
RESULTS
HFmrEF/HFpEF Cohort Development
Examining all sequential echocardiograms between January 2017 and December 2020, a total of 9,591 unique patients were identified with LVEF>40% and at least one distinct HF diagnosis within 12 months prior to or 7 days after the qualifying echocardiogram (Table S2). Of these, 694 (7%) patients were excluded due to the presence of advanced therapies (heart transplant, left ventricular assist device), 100 (1%) patients were excluded due to severe aortic stenosis or severe mitral stenosis, and 3,242 (34%) patients were excluded due to HFpEF mimickers such as infiltrative cardiomyopathies and hypertrophic cardiomyopathy (Table S1). 3,215 patients met other eligibility criteria, but were excluded due to missing echocardiographic data alone, constituting 33.5% the starting cohort of N=9,591 patients. The majority of these were missing LV GLS data (n=2,692, 84%), E/e’ or e’ were missing in n=1,574 (49%), left atrial enlargement was missing in n=754 (23%), and left ventricular hypertrophy was missing in n=86 (0.9%); these missing counts are not mutually exclusive. The final cohort included 2,337 unique patients with median follow-up of 21 months (interquartile range 9 – 33 months).
Baseline characteristics of HFmrEF/HFpEF Cohort
The analysis cohort of patients with HFmrEF/HFpEF (n = 2337) had a median age of 70 years, were 50% female, and 32% Black race (Table 1). The cohort had a high burden of co-morbidities including hypertension (92%), hyperlipidemia (76%), coronary artery disease (65%), renal disease (46%), and diabetes (45%). Median BMI was 29 kg/m2 (interquartile range [IQR] 25 – 34) and median NT-proBNP was 970 pg/mL (IQR 276 – 4174). Patients were treated with diuretics (67%), beta blockers (66%), ACEi/ARB (49%), and MRA (15%), among other cardiovascular therapies (Table 1).
Table 1.
Baseline patient characteristics stratified by LVEF
| Overall (N=2337) | LVEF >40–44% (N=232) | LVEF 45–49% (N=408) | LVEF 50–55% (N=442) | LVEF>55% (N=1255) | P-Value | |
|---|---|---|---|---|---|---|
| Patient Characteristics | ||||||
| Age (years) | 70 (61, 79) | 69 (61, 77) | 70 (59, 78) | 69 (58, 78) | 71 (62, 80) | <0.001 |
| Female | 1167 (49.9%) | 91 (39.2%) | 155 (38.0%) | 184 (41.6%) | 737 (58.7%) | <0.001 |
| Race | 0.88 | |||||
| White | 1487 (64.2%) | 143 (61.6%) | 254 (62.7%) | 292 (66.7%) | 798 (64.2%) | |
| Black | 736 (31.8%) | 80 (34.5%) | 133 (32.8%) | 129 (29.5%) | 394 (31.7%) | |
| Other | 95 (4.1%) | 9 (3.9%) | 18 (4.4%) | 17 (3.9%) | 51 (4.1%) | |
| Year of index echocardiogram | 0.89 | |||||
| 2017 | 521 (22.3%) | 55 (23.7%) | 89 (21.8%) | 86 (19.5%) | 291 (23.2%) | |
| 2018 | 656 (28.1%) | 68 (29.3%) | 115 (28.2%) | 124 (28.1%) | 349 (27.8%) | |
| 2019 | 638 (27.3%) | 61 (26.3%) | 109 (26.7%) | 132 (29.9%) | 336 (26.8%) | |
| 2020 | 522 (22.3%) | 48 (20.7%) | 95 (23.3%) | 100/ (22.6%) | 279 (22.2%) | |
| Months since prior HF diagnosis | 0 (0, 3) | 0 (0, 2) | 0 (0, 3) | 0 (0, 3) | 0 (0, 3) | 0.28 |
| Baseline Comorbidities | ||||||
| Hypertension | 2140 (91.6%) | 215 (92.7%) | 366 (89.7%) | 400 (90.5%) | 1159 (92.4%) | 0.28 |
| Diabetes | 1042 (44.6%) | 113 (48.7%) | 173 (42.4%) | 181 (41.0%) | 575 (45.8%) | 0.14 |
| Prior myocardial infarction | 595 (25.5%) | 92 (39.7%) | 149 (36.5%) | 119 (26.9%) | 235 (18.7%) | <0.001 |
| Cerebrovascular disease | 706 (30.2%) | 66 (28.4%) | 120 (29.4%) | 124 (28.1%) | 396 (31.6%) | 0.47 |
| Peripheral vascular disease | 931 (39.8%) | 90 (38.8%) | 168 (41.2%) | 183 (41.4%) | 490 (39.0%) | 0.76 |
| Chronic kidney disease | 1071 (45.8%) | 102 (44.0%) | 183 (44.9%) | 178 (40.3%) | 608 (48.4%) | 0.024 |
| Hyperlipidemia | 1774 (75.9%) | 19 (82.8%) | 301 (73.8%) | 343 (77.6%) | 938 (74.7%) | 0.035 |
| Chronic pulmonary disease | 1004 (43.0%) | 96 (41.4%) | 160 (39.2%) | 182 (41.2%) | 566 (45.1%) | 0.14 |
| Atrial fibrillation/flutter | 877 (37.5%) | 91 (39.2%) | 164 (40.2%) | 179 (40.5%) | 443 (35.3%) | 0.12 |
| Coronary artery disease | 1523 (65.2%) | 178 (76.7%) | 285 (69.9%) | 294 (66.5%) | 766 (61.0%) | <0.001 |
| Baseline Vital Signs | ||||||
| Body mass index (kg/m2) | 29 (25, 34) | 28 (25, 32) | 28 (24, 32) | 29 (25, 33) | 29 (25, 34) | <0.001 |
| Baseline heart rate (bpm) | 70 (63, 80) | 74 (64, 82) | 70 (63, 82) | 71 (64, 80) | 70 (63, 80) | 0.068 |
| Baseline SBP (mmHg) | 131 (118, 147)* | 128 (116, 144) | 133 (118, 148) | 130 (119, 149) | 132 (119, 146) | 0.28 |
| Baseline DBP (mmHg) | 72 (64, 81)* | 73 (64, 80) | 73 (64, 82) | 73 (65, 82) | 71 (63, 80) | 0.061 |
| Baseline Cardiovascular Medications | ||||||
| Aspirin | 1154 (49.4%) | 109 (47.0%) | 205 (50.2%) | 232 (52.5%) | 608 (48.4%) | 0.42 |
| Statin | 1251 (53.5%) | 127 (54.7%) | 226 (55.4%) | 254 (57.5%) | 644 (51.3%) | 0.12 |
| Beta blocker | 1532 (65.6%) | 164 (70.7%) | 281 (68.9%) | 291 (65.8%) | 796 (63.4%) | 0.065 |
| ACEi/ARB | 1138 (48.7%) | 121 (52.2%) | 222 (54.4%) | 228 (51.6%) | 567 (45.2%) | 0.003 |
| ARNI | 35 (1.5%) | 11 (4.7%) | 10 (2.5%) | 7 (1.6%) | 7 (0.6%) | <0.001 |
| Calcium channel blocker | 1070 (45.8%) | 90 (38.8%) | 164 (40.2%) | 208 (47.1%) | 608 (48.4%) | 0.004 |
| Diuretics | 1566 (67.0%) | 143 (61.6%) | 241 (59.1%) | 295 (66.7%) | 887 (70.7%) | <0.001 |
| ‐Loop diuretics | 1229 (52.6%) | 111 (47.8%) | 191 (46.8%) | 234 (52.9%) | 693 (55.2%) | 0.011 |
| ‐Thiazide diuretics | 358 (15.3%) | 33 (14.2%) | 48 (11.8%) | 67 (15.2%) | 210 (16.7%) | 0.11 |
| ‐Other diuretics | 804 (34.4%) | 70 (30.2%) | 114 (27.9%) | 169 (38.2%) | 451 (35.9%) | 0.004 |
| P2Y12 inhibitor | 541 (23.1%) | 58 (25.0%) | 106 (26.0%) | 111 (25.1%) | 266 (21.2%) | 0.12 |
| Oral anti-coagulant | 491 (21.0%) | 48 (20.7%) | 91 (22.3%) | 95 (21.5%) | 257 (20.5%) | 0.87 |
| MRA | 341 (14.6%) | 40 (17.2%) | 70 (17.2%) | 69 (15.6%) | 162 (12.9%) | 0.083 |
| SGLT2 inhibitors | 24 (1.0%) | 3 (1.3%) | 5 (1.2%) | 6 (1.4%) | 10 (0.8%) | 0.57 |
| Hydralazine | 548 (23.4%) | 52 (22.4%) | 79 (19.4%) | 117 (26.5%) | 300 (23.9%) | 0.097 |
| Nitrates | 238 (10.2%) | 27 (11.6%) | 40 (9.8%) | 53 (12.0%) | 118 (9.4%) | 0.39 |
| Baseline Laboratory Measures | ||||||
| Creatinine | 1.1 (0.9, 1.7) | 1.2 (0.9, 1.8) | 1.1 (0.9, 1.7) | 1.2 (0.9, 1.7) | 1.1 (0.9, 1.7) | 0.26 |
| eGFR | 59.0 (36.8, 81.1) | 59.8 (34.7, 79.6) | 62.4 (40.7, 81.0) | 59.0 (37.7, 83.7) | 57.6 (35.7, 81.0) | 0.47 |
| BUN (mg/dL) | 19 (13, 28) | 19 (14, 28) | 18 (14, 27) | 18 (13, 29) | 19 (13, 28) | 0.76 |
| Sodium (mmol/L) | 138 (136, 140) | 138 (136, 140) | 138 (136, 140) | 138 (136, 140) | 138 (136, 140) | 0.74 |
| Hemoglobin (g/dL) | 11.5 (9.9, 13.1) | 11.6 (10.1, 13.2) | 11.5 (9.7, 13.3) | 11.4 (9.7, 13.1) | 11.6 (10.0, 13.0) | 0.55 |
| BNP (pg/mL) | 213 (77, 563)** | 320 (89, 877) | 310 (133, 834) | 246 (86, 596) | 175.4 (71, 436) | 0.004 |
| NT-proBNP (pg/mL) | 970 (276, 4174)** | 1790 (384, 8496) | 1459 (417, 5750) | 1168 (265, 4876) | 758.5 (236, 2938) | <0.001 |
Available in 1865/2337 patients
BNP available in 617/2237 patients, NT-proBNP available in 1711/2337 patients
Median (Q1,Q3) for continuous variables; n (%) for discrete variables
ACEi, angiotensin-converting enzyme inhibitor; ARB, angiotensin II receptor blocker; ARNI, angiotensin receptor–neprilysin inhibitor; BNP, B-type natriuretic peptide; BUN, blood urea nitrogen; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; HF, heart failure; MRA, mineralocorticoid receptor antagonist; NT-proBNP, N-terminal pro-B-type natriuretic peptide; SBP, systolic blood pressure; SGLT2, sodium-glucose cotransporter-2
Compared to included patients, those excluded due to missing echocardiographic data tended to be older (age 72 vs 70 years), less commonly Black (28% vs 32%), closer to their most recent billed HF diagnosis code (2.6 vs 3 months), more commonly demonstrated atrial fibrillation/flutter (51% vs 38%), had a higher body mass index (median of 31 vs 29 kg/m2) and higher heart rate (median of 75 vs 70 bpm), and higher NT-proBNP (median of 1388 vs 970 pg/mL) (Table S3). Patients excluded also tended to more commonly have their index echocardiogram performed inpatient (58% vs 38%) and also tended have lower survival over 3-year follow-up compared to those with complete echocardiographic data (Figure S1).
Echocardiographic features associated with outcomes
Among 2,337 patients with complete echocardiographic data, 1082 (46%) had LVEF ≤55%, and the median (interquartile range) was −15.0 (−17.6 to −12.5) for LV GLS, 39 ml/m2 (30 to 50 ml/m2) for LAVi, and 12 (9.0 – 7.0) for E/e’, respectively. Abnormal LV GLS (>−16) was present in 1393 (60%), abnormal left atrial enlargement (LAVi>28) in 1835 (79%), mild/moderate/severe LVH in 1910 (82%), and abnormal E/e’ (E/e’>13 and e’<9) in 918 (39%). In univariable analysis, the following dichotomized features were associated with increased all-cause mortality risk over 3 years of follow-up: abnormal E/e’ + e’ (HR 1.59, 95% CI [1.22–1.90]), abnormal LV GLS (HR 1.42 [1.18–1.71]), and elevated LAVi (HR 1.39 [1.09–1.77]) (all p<0.01, Table 2). The presence of LVH and LVEF reduced below 55% were not significantly associated with all-cause mortality (Table 2). When assessing in a continuous fashion, E/e’, e’, LV GLS, LAVi, and LVEF all demonstrated increasing mortality risk at increasing abnormal values (Figure 1, Figure S2). Through stepwise selection built on a clinical multivariable model (age, sex, key co-morbidities – Table 3) for all-cause mortality, LV GLS emerged as the only echocardiographic feature that was independently associated with all-cause mortality (aHR 1.35 [1.11–1.63], p=0.002). There was no interaction by LV GLS software (GE EchoPAC 79.5% of cases, TomTec 21.5%; p-value for interaction = 0.91). Abnormal E/e’ + e’, LVH, LAVi, and LVEF each did not individually demonstrate significant association with all-cause mortality in this multivariable model to warrant inclusion by stepwise selection. The concordance index of this multivariable model was 0.65. The complete multivariable model prior to stepwise selection is shown in Table S4 as well as sensitivity analysis with hypertension and CAD in Table S5, both with similar results.
Table 2:
Univariable association between echocardiographic features of pathologic HFmrEF/HFpEF and all-cause mortality
| Pathologic Feature of HFmrEF/HFpEF | Hazard Ratio (95% CI) | P-Value | Concordance Index |
|---|---|---|---|
| Abnormal global longitudinal strain (LV GLS)* | 1.42 (1.18, 1.71) | <0.001 | 0.54 |
| Left atrial enlargement by LAVi** | 1.39 (1.09, 1.77) | 0.008 | 0.52 |
| Left ventricular hypertrophy (LVH) † | 1.20 (0.93, 1.53) | 0.16 | 0.51 |
| E/e’>13 + e’<9 | 1.59 (1.33, 1.90) | <0.001 | 0.55 |
| Left ventricular ejection fraction (LVEF) ≤ 55% | 1.13 (0.94, 1.35) | 0.18 | 0.51 |
Abnormal LV GLS defined as > −16
LAVi (left atrial volume index) defined as > 28 ml/m2
LVH ≥ mild, defined as posterior or septal wall thickness ≥ 1.1 cm
Figure 1: Cumulative incidence of mortality at 3 years stratified by LVEF and LV GLS.
LVEF = left ventricular ejection fraction; LV GLS = left ventricular global longitudinal strain
Table 3.
Multivariable mortality model by stepwise selection
| Covariate | HR (95% CI) | P-Value |
|---|---|---|
| Age (years), per year | 1.02 (1.01,1.03) | <0.001 |
| Female | 1.14 (0.95,1.36) | 0.17 |
| Body mass index, per kg/m2 | 0.97 (0.96,0.99) | <0.001 |
| Diabetes mellitus | 1.22 (1.00,1.48) | 0.045 |
| Chronic kidney disease | 1.84 (1.51,2.25) | <0.001 |
| Atrial fibrillation/flutter | 1.20 (1.00,1.45) | 0.054 |
| Abnormal LV GLS (>−16) | 1.35 (1.11,1.63) | 0.002 |
LV GLS, left ventricular global longitudinal strain
In the setting of normal LVEF (>55%) (n=1255), 498 patients (40%) demonstrated decreased LV GLS (Table 4). Patients with preserved LVEF and impaired LV GLS were more likely to be Black race (41% vs 26–33% in the three other LVEF + LV GLS cohorts, Table 4) and demonstrate higher body mass index (median BMI 30 vs 28–29 kg/m2). This patient cohort had a higher burden of chronic kidney disease (54% vs 28–46%), diabetes (49% vs 37–45%), hyperlipidemia (80% vs 71–78%), and hypertension (95% vs 88–91%). These patients demonstrated an intermediate elevation in NT-proBNP (1125 vs 593–1580 pg/mL). This cohort demonstrated higher diuretic utilization (74% vs 59–68%) but lower rates of MRA (11% vs 14–17%) and ARNI (1% vs 0.3–2.7%) therapy. This cohort demonstrated higher event rates of all-cause mortality, cardiovascular mortality, and HF hospitalization, compared to patients with normal LV GLS irrespective of LVEF (Figure 2–3).
Table 4:
Baseline patient characteristics, stratified by LVEF (preserved/reduced) and LV GLS (abnormal/normal)
| Overall (N=2337) | LVEF≤55% and LV GLS>−16 (N=895) | LVEF>55% and LV GLS>−16 (N=498) | LVEF≤55% and LV GLS≤−16 (N=187) | LVEF>55% and LV GLS≤−16 (N=757) | P-Value | |
|---|---|---|---|---|---|---|
| Patient Characteristics | ||||||
| Age (years) | 70 (61, 79) | 70 (59, 78) | 71 (62, 81) | 69 (56, 77) | 71 (62, 80) | 0.0006 |
| Female | 1167 (49.9%) | 353 (39.4%) | 283 (56.8%) | 77 (41.2%) | 454 (60.0%) | <0.001 |
| Race | <0.001 | |||||
| White | 1487 (64.2%) | 561 (63.1%) | 276 (55.9%) | 128 (68.8%) | 522 (69.7%) | |
| Black | 736 (31.8%) | 294 (33.1%) | 201 (40.7%) | 48 (25.8%) | 193 (25.8%) | |
| Other | 95 (4.1%) | 34 (3.8%) | 17 (3.4%) | 10 (5.4%) | 34 (4.5%) | |
| Year of Index Echocardiogram | 0.39 | |||||
| 2017 | 521 (22.3%) | 180 (20.1%) | 116 (23.3%) | 50 (26.7%) | 175 (23.1%) | |
| 2018 | 656 (28.1%) | 258 (28.8%) | 131 (26.3%) | 49 (26.2%) | 218 (28.8%) | |
| 2019 | 638 (27.3%) | 252 (28.2%) | 148 (29.7%) | 50 (26.7%) | 188 (24.8%) | |
| 2020 | 522 (22.3%) | 205 (22.9%) | 103 (20.7%) | 38 (20.3%) | 176 (23.2%) | |
| Months since prior HF diagnosis | 0 (0, 3) | 0 (0, 3) | 0 (0, 2) | 1 (0, 4) | 0 (0, 4) | <0.001 |
| Baseline Comorbidities | ||||||
| Hypertension | 2140 (91.6%) | 817 (91.3%) | 475 (95.4%) | 164 (87.7%) | 684 (90.4%) | 0.002 |
| Diabetes | 1042 (44.6%) | 398 (44.5%) | 246 (49.4%) | 69 (36.9%) | 329 (43.5%) | 0.023 |
| Prior MI | 595 (25.5%) | 315 (35.2%) | 106 (21.3%) | 45 (24.1%) | 129 (17.0%) | <0.001 |
| Cerebrovascular Disease | 706 (30.2%) | 262 (29.3%) | 166 (33.3%) | 48 (25.7%) | 230 (30.4%) | 0.21 |
| Peripheral Vascular Disease | 931 (39.8%) | 373 (41.7%) | 203 (40.8%) | 68 (36.4%) | 287 (37.9%) | 0.31 |
| Chronic kidney disease | 1071 (45.8%) | 411 (45.9%) | 270 (54.2%) | 52 (27.8%) | 338 (44.6%) | <0.001 |
| Hyperlipidemia | 1774 (75.9%) | 698 (78.0%) | 400 (80.3%) | 138 (73.8%) | 538 (71.1%) | <0.001 |
| Chronic Pulmonary Disease | 1004 (43.0%) | 367 (41.0%) | 225 (45.2%) | 71 (38.0%) | 341 (45.0%) | 0.13 |
| Atrial Fibrillation/Flutter | 877 (37.5%) | 362 (40.4%) | 192 (38.6%) | 72 (38.5%) | 251 (33.2%) | 0.021 |
| Coronary Artery Disease | 1523 (65.2%) | 644 (72.0%) | 326 (65.5%) | 113 (60.4%) | 440 (58.1%) | <0.001 |
| Baseline Vital Signs | ||||||
| Body Mass Index (kg/m2) | 29 (25, 34) | 28 (25, 32) | 30 (26, 34) | 29 (25, 33) | 29 (25, 34) | 0.0002 |
| Baseline heart rate (bpm) | 70 (63, 80) | 72 (64, 82) | 72 (64, 81) | 67 (61, 77) | 69 (62, 78) | <0.001 |
| Baseline SBP (mmHg) | 131 (118, 147)* | 131 (119, 149) | 133 (119, 149) | 127 (116, 144) | 131 (118, 145) | 0.11 |
| Baseline DBP (mmHg) | 72 (64, 81)* | 73 (64, 82) | 72 (64, 80) | 74 (65, 80) | 71 (62, 80) | 0.057 |
| Baseline Cardiovascular Medications | ||||||
| Aspirin | 1154 (49.4%) | 469 (52.4%) | 274 (55.0%) | 77 (41.2%) | 334 (44.1%) | <0.001 |
| Statin | 125 (53.5%) | 509 (56.9%) | 283 (56.8%) | 98 (52.4%) | 361 (47.7%) | <0.001 |
| Beta Blocker | 1532 (65.6%) | 616 (68.8%) | 333 (66.9%) | 120 (64.2%) | 463 (61.2%) | 0.010 |
| ACEi/ARB | 1138 (48.7%) | 483 (54.0%) | 245 (49.2%) | 88 (47.1%) | 322 (42.5%) | <0.001 |
| ARNI | 35 (1.5%) | 24 (2.7%) | 5 (1.0%) | 4 (2.1%) | 2 (0.3%) | <0.001 |
| Calcium Channel Blocker | 1070 (45.8%) | 389 (43.5%) | 281 (56.4%) | 73 (39.0%) | 327 (43.2%) | <0.001 |
| Diuretics | 1566 (67.0%) | 568 (63.5%) | 370 (74.3%) | 111 (59.4%) | 517 (68.3%) | <0.001 |
| ‐Loop Diuretics | 1229 (52.6%) | 454 (50.7%) | 295 (59.2%) | 82 (43.9%) | 398 (52.6%) | 0.001 |
| ‐Thiazide Diuretics | 358 (15.3%) | 118 (13.2%) | 88 (17.7%) | 30 (16.0%) | 122 (16.1%) | 0.13 |
| ‐Other Diuretics | 804 (34.4%) | 298 (33.3%) | 200 (40.2%) | 55 (29.4%) | 251 (33.2%) | 0.016 |
| P2Y12 inhibitor | 541 (23.1%) | 239 (26.7%) | 122 (24.5%) | 36 (19.3%) | 144 (19.0%) | 0.001 |
| Oral Anti-Coagulant | 491 (21.0%) | 196 (21.9%) | 119 (23.9%) | 38 (20.3%) | 138 (18.2%) | 0.090 |
| MRA | 341 (14.6%) | 152 (17.0%) | 56 (11.2%) | 27 (14.4%) | 106 (14.0%) | 0.032 |
| SGLT2 Inhibitors | 24 (1.0%) | 12 (1.3%) | 5 (1.0%) | 2 (1.1%) | 5 (0.7%) | 0.60 |
| Ivabradine | 3 (0.1%) | 0 (0.0%) | 0 (0.0%) | 1 (0.5%) | 2 (0.3%) | 0.064 |
| Hydralazine | 548 (23.4%) | 218 (24.4%) | 131 (26.3%) | 30 (16.0%) | 169 (22.3%) | 0.030 |
| Nitrates | 238 (10.2%) | 104 (11.6%) | 61 (12.2%) | 16 (8.6%) | 57 (7.5%) | 0.013 |
| Baseline Laboratory Measures | ||||||
| BUN (mg/dL) | 19 (13, 28) | 19 (14, 29) | 20 (13, 31) | 15 (12, 22) | 18 (13, 26) | <0.001 |
| Sodium (mmol/L) | 138 (136, 140) | 138 (136, 140) | 138 (136, 140) | 138 (136, 140) | 138 (136, 140) | 0.85 |
| Hemoglobin (g/dL) | 11.5 (9.9, 13.1) | 11.5 (9.8, 13.2) | 11.4 (9.8, 13.0) | 12.0 (10.2, 13.2) | 11.6 (10.1, 12.9) | 0.31 |
| BNP (pg/mL) | 213 (77, 563)** | 309 (106, 776) | 214 (102, 477) | 158 (40, 445) | 140 (63, 395) | <0.001 |
| NT-proBNP (pg/mL) | 970 (276, 4174)** | 1580 (394, 7100) | 1125 (343, 3789) | 703 (155, 2618) | 593 (206, 2452) | <0.001 |
| Creatinine | 1.1 (0.9, 1.7) | 1.2 (0.9, 1.8) | 1.2 (0.9, 1.9) | 1.0 (0.8, 1.4) | 1.1 (0.9, 1.5) | <0.001 |
| eGFR | 59.0 (36.8, 81.1) | 58.5 (36.2, 79.6) | 54.6 (31.4, 78.1) | 70.6 (50.1, 92.1) | 60.1 (38.2, 82.3) | <0.001 |
Available in 1865/2337 patients
BNP available in 617/2237 patients, NT-proBNP available in 1711/2337 patients
Median (Q1,Q3) for continuous variables; n (%) for discrete variables
Figure 2. Kaplan Meier curves for all-cause mortality by LVEF and LV GLS status.
LVEF = left ventricular ejection fraction; LV GLS = left ventricular global longitudinal strain
Figure 3: Cumulative Incidence Rates of Clinical Outcomes by LVEF and LV GLS status.
LVEF = left ventricular ejection fraction; LV GLS = left ventricular global longitudinal strain
DISCUSSION
In a large population of patients with HFmrEF/HFpEF, echocardiographic features of pathologic HF, led by reduced LV GLS, abnormal E/e’ + e’, and elevated LAVi, were associated with adverse outcomes irrespective of quantitative LVEF. After incorporating demographic and clinical comorbidity factors, reduced LV GLS emerged as the only echocardiographic feature that was independently associated with all-cause mortality. Among patients with preserved LVEF, a large subset (~40%) demonstrated adverse myocardial function by LV GLS. This cohort of patients demonstrated higher event rates of all-cause mortality, CV mortality, and HF hospitalization, compared to patients with normal LV GLS regardless of specific LVEF.
To our knowledge, this represents the largest, contemporary, real-world analysis of echocardiographic features of patients with HFmrEF/HFpEF. Prior research has established several echocardiographic features beyond LVEF as useful in diagnosing and prognosticating patients with HFpEF.23 These include average E/e’,15,31 septal/lateral e’,16 PASP/TR velocity,15,32 LAVi,32,33 LV mass index,33 LVH,15 relative wall thickness,15 and LV GLS.10–12,14,32,34 These features have a wide range of sensitivity/specificity for diagnosing HFpEF,23 and have been incorporated into combination scores with clinical variables to predict likelihood of HFpEF (HFA-PEFF and HF2PEF).23,35 LV GLS, in particular, has emerged as a particularly useful measure of myocardial function in HFpEF through secondary analyses of clinical trial datasets (~200–500 patients)9–11, small single-center studies (~300–400 patients),12,13 and observational registries,31 Further, while a recent meta-analysis demonstrated variability in association between GLS and outcomes (only 4 of 9 studies indicating a clear relationship).14 one of the largest of these studies evaluated 447 patients enrolled in the TOPCAT trial and showed that LV longitudinal strain had the strongest association with clinical outcomes among echocardiographic variables and remained prognostic after adjustment for multiple clinical/echocardiographic variables; a trend towards improvement in LV longitudinal strain was noted in the spironolactone arm in the Americas cohort.10 Several other echocardiographic variables were also prognostic led by e’ and E/e’.
This study is consistent with data from several of these prior studies and establishes several useful findings in a large, modern, real-world cohort. First, the prevalence of impaired LV GLS was 60% when LV GLS was measured among a cohort of patients with LVEF>40%. This finding correlates closely with the range of 52–67% from analyses of clinical trials including TOPCAT, RELAX, and PARAMOUNT.9–11 This present study also underscores the limited utility of quantitative LVEF in differentiating outcomes among patients with LVEF>40%. Further, it demonstrates that, while several features (E/e’, LAVi, LV GLS) have prognostic utility, abnormal LV GLS is the only feature that remains associated with outcomes after adjusting for clinical variables and considering multiple echocardiographic metrics. LV GLS is known to have strong reproducibility compared to other metrics,36 as well as a strong geometric and mathematical rationale for superior reflection of systolic function in patients with HFpEF.37 In this setting, the findings of this study further add to the body of literature underscoring the association between LV GLS and clinical outcomes in HFmrEF/HFpEF. These findings also establish a definitive foundation to support investigation of differential drug/device response across LV GLS spectrum, which could prove more useful than LVEF in this regard as well
Second, the generalizability of these findings is restricted to the cohort of patients with available LV GLS data and, notably, this excluded over half (58%) of this tertiary care cohort of carefully identified patients with HFmrEF/HFpEF. The reasons these metrics were not measured in a subset of our cohort are unknown; of note, there was no strong temporal trends across the study period (i.e. LV GLS missing in 58% in 2017 and in 55% in 2020). These data highlight an underutilization of LV GLS in this population, which appeared to be older with higher co-morbidity burden and worse outcomes, often assessed by index echocardiography in the inpatient setting. While patients with acute HFpEF may present technical challenges for imaging modalities, there is strong evidence that LV GLS can be utilized in this setting with acute, decompensated HF.12,34 Risk stratification of this population may become increasingly important with the improvement in evidence of pharmacologic therapy for HFmrEF/HFpEF and the growing evidence of the utility of in-hospital initiation of therapies.38,39 There is also growing evidence for the utility of LV GLS on the other end of the HFpEF ‘spectrum’ as it has been recently shown that impaired strain may have prognostic value in predicting HF admissions in stage A-B HF.40 It appears that the future growth of LV GLS could be focused at these different ends of the HFpEF spectrum, where additional risk stratification and monitoring could add significant value.
Third, these data demonstrate that there is a large cohort of patients with preserved LVEF and impaired LV GLS (i.e. 40% of those with preserved LVEF), who have a unique clinical profile and adverse outcomes. In this study, this cohort had higher rates of systemic comorbidities including obesity and higher rates of non-ischemic cardiomyopathy. Event rates including all-cause and CV mortality as well as HF hospitalization were among the highest in this cohort in both the short-term (90 days) and long-term follow-up (1–3 years). More research is warranted to better define the potential unique pathophysiology leading to these cardiac functional abnormalities and adverse outcomes. This cohort of patients may represent a key cohort of interest for medical therapy for HF and future clinical studies.
Also, of note, the lowest rate of adverse outcomes was identified in patients with LVEF 41–55% and preserved LV GLS, even lower than in patients with LVEF >55% and preserved LV GLS. This is an intriguing and perhaps surprising finding with several possible explanations. First, given the known inter- and intra-observer variability of LVEF estimate and the established fluctuation in LVEF depending on load conditions, it could be theorized that the LVEF estimate in these cases was less ‘reliable’ in these cases, and that the ‘true’ prognosis was predicted by LV GLS. Second, it is notable that pharmacologic guideline-directed HF therapy was higher in this population despite lower diuretic requirements (compared to patients with preserved LVEF and preserved LV GLS) so it is possible that these therapy differences contributed to lower outcomes rates. Lastly, this was the smallest cohort in the study and the outcome findings may be artifactual and may not persist in larger cohorts. Still, these findings add fodder to the ongoing debate on how heavily the HF community should continue to rely on granular LVEF for stratification/classification in patients with LVEF>40% (HFmrEF/HFpEF). While LVEF has been described as an imperfect but critical phenotyping feature of patients with HFmrEF/HFpEF,41–44 others have argued for a stronger movement beyond LVEF to phenotype these patients.6,45–47 The study supports an intermediate reality in that, for many patients with HFmrEF/HFpEF, LVEF must be relied upon given the lack of regular utilization of features such as LV GLS in modern practice, but, in this setting, there is growing evidence that LV GLS should be further studied, prioritized, and incorporated with particular opportunities for growth in acute, decompensated HFmrEF/HFpEF.
Limitations
This study is strengthened by leveraging a large, diverse, both community and tertiary care cohort of real-world patients with HFmrEF/HFpEF, and by including depth of granularity regarding the clinical and echocardiographic features of this cohort. Additionally, the primary outcome of all-cause mortality was determined from multiple institutional, state, and national sources to ensure accuracy across all care settings. This study also has several limitations. First, the degree of missingness for factors such as LV GLS precluded the ability to perform imputation strategies and required exclusion of patients with missing data. This introduces potential selection bias and limits generalizability to the included cohort described above. Second, the study design was unable to distinguish between secondary regurgitant lesions in the setting of HFpEF and primary regurgitant lesions alone responsible for directly leading to HFpEF pathophysiology; therefore, these patients were included in the cohort (2.35% with severe mitral regurgitation and 0.43% with severe aortic regurgitation).We were also unable to account for dynamic changes in echocardiographic features such as LVEF or LV GLS over time; while this is typical of prognostic echocardiographic analyses, future research will be improved by analysis of trends, including modification by therapy, and association with outcomes.
CONCLUSIONS
In a large, real-world HFmrEF/HFpEF population, echocardiographic features, led by LV GLS, are associated with adverse outcomes irrespective of LVEF. Impaired LV GLS is common (60%) in patients with HFmrEF/HFpEF, but remains unmeasured on index echocardiographic evaluation in over half of these patients. A large subset of patients demonstrated abnormal myocardial function by LV GLS despite preserved LVEF >55%. These patients have a unique clinical profile with higher burden of systemic comorbidities and demonstrate prominent adverse outcomes; this phenotype of patients with HFmrEF/HFpEF may represent a key cohort of interest for therapy implementation and future clinical studies.
Supplementary Material
CLINICAL PERSPECTIVES.
What is New?
In this large cohort of patients with HFmrEF/HFpEF, LV GLS emerged as the most prognostic echocardiographic feature with regards to mortality and hospitalization outcomes. Nearly half of patients with preserved LVEF demonstrated abnormal LV GLS. This cohort of patients demonstrated adverse outcomes similar to patients with impaired LVEF and abnormal LV GLS and significantly worse than patients with normal LV GLS regardless of specific LVEF.
What are the Clinical Implications?
These findings suggest that LV GLS may serve as a superior discriminator of outcomes in patients with HFmrEF/HFpEF as compared to other echocardiographic features including LVEF. Patients with HFmrEF/HFpEF and abnormal LV GLS have a unique clinical profile and may represent a key cohort of interest for future studies.
DISCLOSURES
Dr. Peters is supported by the National Heart Lung and Blood Institute (T32HL069749) and has received honoraria from Cytokinetics. Dr Felker has received research grants from the National Heart, Lung, and Blood Institute, American Heart Association, Amgen, Bayer, Bristol Myers Squibb, Merck, Cytokinetics, and CSL Behring; has acted as a consultant to Novartis, Amgen, Bristol Myers Squibb, Cytokinetics, Medtronic, Cardionomic, Boehringer Ingelheim, American Regent, Abbott, AstraZeneca, Reprieve, Myovant, Sequana, Windtree Therapeutics, and White-Swell; and has served on clinical endpoint committees/data safety monitoring boards for Amgen, Merck, Medtronic, EBR Systems, V-Wave, LivaNova, Siemens, and Rocket Pharma. Dr. Mentz has received research support and honoraria from Abbott, American Regent, Amgen, AstraZeneca, Bayer, Boehringer Ingelheim/Eli Lilly, Boston Scientific, Cytokinetics, Fast BioMedical, Gilead, Innolife, Medtronic, Merck, Novartis, Relypsa, Respicardia, Roche, Sanofi, Vifor, Windtree Therapeutics, and Zoll. Dr. DeVore reports research funding through his institution from the American Heart Association, Amgen, Biofourmis, Bodyport, Cytokinetics, American Regent, Inc, the NHLBI, and Novartis. He also provides consulting services for and/or receives honoraria from Abiomed, Amgen, AstraZeneca, Cardionomic, InnaMed, LivaNova, Natera, Novartis, Procyrion, Story Health, Vifor, and Zoll.
Footnotes
REFERENCES:
- 1.Heidenreich PA, Bozkurt B, Aguilar D, Allen LA, Byun JJ, Colvin MM, Deswal A, Drazner MH, Dunlay SM, Evers LR, Fang JC, Fedson SE, Fonarow GC, Hayek SS, Hernandez AF, Khazanie P, Kittleson MM, Lee CS, Link MS, Milano CA, Nnacheta LC, Sandhu AT, Stevenson LW, Vardeny O, Vest AR, Yancy CW, Beckman JA, O’Gara PT, Al-Khatib SM, Armbruster AL, Birtcher KK, Cigarroa JE, de las Fuentes L, Dixon DL, Fleisher LA, Gentile F, Goldberger ZD, Gorenek B, Haynes N, Hlatky MA, Joglar JA, Jones WS, Marine JE, Mark DB, Mukherjee D, Palaniappan LP, Piano MR, Rab T, Spatz ES, Tamis-Holland JE, Wijeysundera DN, Woo YJ. 2022 AHA/ACC/HFSA Guideline for the Management of Heart Failure. J Card Fail. 2022;28:e1–e167. [DOI] [PubMed] [Google Scholar]
- 2.McDonagh TA, Metra M, Adamo M, Gardner RS, Baumbach A, Böhm M, Burri H, Butler J, Celutkiene J, Chioncel O, Cleland JGF, Coats AJS, Crespo-Leiro MG, Farmakis D, Gilard M, Heymans S. 2021 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J. 2021;42:3599–3726. [DOI] [PubMed] [Google Scholar]
- 3.Bozkurt B, Coats AJ, Tsutsui H, Abdelhamid M, Adamopoulos S, Albert N, Anker SD, Atherton J, Böhm M, Butler J, Drazner MH, Felker GM, Filippatos G, Fonarow GC, Fiuzat M, Gomez-Mesa JE, Heidenreich P, Imamura T, Januzzi J, Jankowska EA, Khazanie P, Kinugawa K, Lam CSP, Matsue Y, Metra M, Ohtani T, Francesco Piepoli M, Ponikowski P, Rosano GMC, Sakata Y, SeferoviĆ P, Starling RC, Teerlink JR, Vardeny O, Yamamoto K, Yancy C, Zhang J, Zieroth S. Universal Definition and Classification of Heart Failure: A Report of the Heart Failure Society of America, Heart Failure Association of the European Society of Cardiology, Japanese Heart Failure Society and Writing Committee of the Universal Definition. J Card Fail. 2021;27:P387–413. [DOI] [PubMed] [Google Scholar]
- 4.Gevaert AB, Tibebu S, Mamas MA, Ravindra NG, Lee SF, Ahmad T, Ko DT, Januzzi JL, Van Spall HGC. Clinical phenogroups are more effective than left ventricular ejection fraction categories in stratifying heart failure outcomes. ESC Hear Fail. 2021;8:2741–2754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Mele D, Nardozza M, Ferrari R. Left ventricular ejection fraction and heart failure: an indissoluble marriage? Eur J Heart Fail. 2018;20:427–430. [DOI] [PubMed] [Google Scholar]
- 6.Ferreira JP, Packer M, Butler J, Zannad F. Reconsidering the ejection fraction centric view of pharmacologic treatment for heart failure. Eur J Heart Fail. 2022;48–53. [DOI] [PubMed] [Google Scholar]
- 7.Cikes M, Solomon SD. Beyond ejection fraction: An integrative approach for assessment of cardiac structure and function in heart failure. Eur Heart J. 2016;37:1642–1650. [DOI] [PubMed] [Google Scholar]
- 8.Gorcsan J, Tanaka H. Echocardiographic assessment of myocardial strain. J Am Coll Cardiol. 2011;58:1401–1413. [DOI] [PubMed] [Google Scholar]
- 9.Kraigher-Krainer E, Shah AM, Gupta DK, Santos A, Claggett B, Pieske B, Zile MR, Voors AA, Lefkowitz MP, Packer M, McMurray JJV, Solomon SD. Impaired systolic function by strain imaging in heart failure with preserved ejection fraction. J Am Coll Cardiol. 2014;63:447–456. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Shah AM, Claggett B, Sweitzer NK, Shah SJ, Anand IS, Liu L, Pitt B, Pfeffer MA, Solomon SD. Prognostic importance of impaired systolic function in heart failure with preserved ejection fraction and the impact of spironolactone. Circulation. 2015;132:402–414. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.DeVore AD, McNulty S, Alenezi F, Ersboll M, Vader JM, Oh JK, Lin G, Redfield MM, Lewis G, Semigran MJ, Anstrom KJ, Hernandez AF, Velazquez EJ. Impaired left ventricular global longitudinal strain in patients with heart failure with preserved ejection fraction: insights from the RELAX trial. Eur J Heart Fail. 2017;19:893–900. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Buggey J, Alenezi F, Yoon HJ, Phelan M, DeVore AD, Khouri MG, Schulte PJ, Velazquez EJ. Left ventricular global longitudinal strain in patients with heart failure with preserved ejection fraction: outcomes following an acute heart failure hospitalization. ESC Hear Fail. 2017;4:432–439. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Pellicori P, Kallvikbacka-Bennett A, Khaleva O, Carubelli V, Costanzo P, Castiello T, Wong K, Zhang J, Cleland JGF, Clark AL. Global longitudinal strain in patients with suspected heart failure and a normal ejection fraction: Does it improve diagnosis and risk stratification? Int J Cardiovasc Imaging. 2014;30:69–79. [DOI] [PubMed] [Google Scholar]
- 14.Morris DA, Ma XX, Belyavskiy E, Aravind Kumar R, Kropf M, Kraft R, Frydas A, Osmanoglou E, Marquez E, Donal E, Edelmann F, Tschöpe C, Pieske B, Pieske-Kraigher E. Left ventricular longitudinal systolic function analysed by 2D speckle-tracking echocardiography in heart failure with preserved ejection fraction: A meta-analysis. Open Hear. 2017;4:1–12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Shah AM, Claggett B, Sweitzer NK, Shah SJ, Anand IS, O’Meara E, Desai AS, Heitner JF, Li G, Fang J, Rouleau J, Zile MR, Markov V, Ryabov V, Reis G, Assmann SF, McKinlay SM, Pitt B, Pfeffer MA, Solomon SD. Cardiac structure and function and prognosis in heart failure with preserved ejection fraction. Circ Hear Fail. 2014;7:740–751. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Ibrahim IM, Hafez H, Al-Shair MHA, El Zayat A. Echocardiographic parameters differentiating heart failure with preserved ejection fraction from asymptomatic left ventricular diastolic dysfunction. Echocardiography. 2020;37:247–252. [DOI] [PubMed] [Google Scholar]
- 17.Horvath MM, Winfield S, Evans S, Slopek S, Shang H, Ferranti J. The DEDUCE Guided Query tool: Providing simplified access to clinical data for research and quality improvement. J Biomed Inform. 2011;44:266–276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Horvath MM, Rusincovitch SA, Brinson S, Shang HC, Evans S, Ferranti JM. Modular design, application architecture, and usage of a self-service model for enterprise data delivery: The Duke Enterprise Data Unified Content Explorer (DEDUCE). J Biomed Inform. 2014;52:231–242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Carnicelli AP, Clare R, Hofmann P, Chiswell K, Devore AD, Vemulapalli S, Felker GM, Sarocco P, Mentz RJ. Characteristics and outcomes of patients with heart failure with reduced ejection fraction after a recent worsening heart failure event. J Am Heart Assoc. 2021;10:e021276. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Carnicelli AP, Clare RM, Hofmann P, Chiswell K, DeVore AD, Vemulapalli S, Felker GM, Kelsey AM, DeWald TA, Sarocco P, Mentz RJ. Clinical trajectory of patients with a worsening heart failure event and reduced ventricular ejection fraction. Am Heart J. 2022;245:110–116. [DOI] [PubMed] [Google Scholar]
- 21.Solomon SD, McMurray JJV, Anand IS, Ge J, Lam CSP, Maggioni AP, Martinez F, Packer M, Pfeffer MA, Pieske B, Redfield MM, Rouleau JL, van Veldhuisen DJ, Zannad F, Zile MR, Desai AS, Claggett B, Jhund PS, Boytsov SA, Comin-Colet J, Cleland J, Düngen H-D, Goncalvesova E, Katova T, Kerr Saraiva JF, Lelonek M, Merkely B, Senni M, Shah SJ, Zhou J, Rizkala AR, Gong J, Shi VC, Lefkowitz MP. Angiotensin–Neprilysin Inhibition in Heart Failure with Preserved Ejection Fraction. N Engl J Med. 2019;381:1609–1620. [DOI] [PubMed] [Google Scholar]
- 22.Anker SD, Butler J, Filippatos G, Ferreira JP, Bocchi E, Böhm M, Brunner–La Rocca H-P, Choi D-J, Chopra V, Chuquiure-Valenzuela E, Giannetti N, Gomez-Mesa JE, Janssens S, Januzzi JL, Gonzalez-Juanatey JR, Merkely B, Nicholls SJ, Perrone SV., Piña IL, Ponikowski P, Senni M, Sim D, Spinar J, Squire I, Taddei S, Tsutsui H, Verma S, Vinereanu D, Zhang J, Carson P, Lam CSP, Marx N, Zeller C, Sattar N, Jamal W, Schnaidt S, Schnee JM, Brueckmann M, Pocock SJ, Zannad F, Packer M. Empagliflozin in Heart Failure with a Preserved Ejection Fraction. N Engl J Med. 2021;385:1451–1461. [DOI] [PubMed] [Google Scholar]
- 23.Pieske B, Tschöpe C, de Boer RA, Fraser AG, Anker SD, Donal E, Edelmann F, Fu M, Guazzi M, Lam CSP, Lancellotti P, Melenovsky V, Morris DA, Nagel E, Pieske-Kraigher E, Ponikowski P, Solomon SD, Vasan RS, Rutten FH, Voors AA, Ruschitzka F, Paulus WJ, Seferovic P, Filippatos G. How to diagnose heart failure with preserved ejection fraction: the HFA–PEFF diagnostic algorithm: a consensus recommendation from the Heart Failure Association (HFA) of the European Society of Cardiology (ESC). Eur J Heart Fail. 2020;22:391–412. [DOI] [PubMed] [Google Scholar]
- 24.Caballero L, Kou S, Dulgheru R, Gonjilashvili N, Athanassopoulos GD, Barone D, Baroni M, Cardim N, De Diego JJG, Oliva MJ, Hagendorff A, Hristova K, Lopez T, Magne J, Martinez C, De La Morena G, Popescu BA, Penicka M, Ozyigit T, Carbonero JDR, Salustri A, Van De Veire N, Von Bardeleben RS, Vinereanu D, Voigt JU, Zamorano JL, Bernard A, Donal E, Lang RM, Badano LP, Lancellotti P. Echocardiographic reference ranges for normal cardiac Doppler data: Results from the NORRE Study. Eur Heart J Cardiovasc Imaging. 2015;16:1031–1041. [DOI] [PubMed] [Google Scholar]
- 25.Ponikowski P, Voors AA, Anker SD, Bueno H, Cleland JGF, Coats AJS, Falk V, González-Juanatey JR, Harjola VP, Jankowska EA, Jessup M, Linde C, Nihoyannopoulos P, Parissis JT, Pieske B, Riley JP, Rosano GMC, Ruilope LM, Ruschitzka F, Rutten FH, Van Der Meer P, Sisakian HS, Isayev E, Kurlianskaya A, Mullens W, Tokmakova M, Agathangelou P, Melenovsky V, Wiggers H, Hassanein M, Uuetoa T, Lommi J, Kostovska ES, Juilliere Y, Aladashvili A, Luchner A, Chrysohoou C, Nyolczas N, Thorgeirsson G, Weinstein JM, Lenarda A Di, Aidargaliyeva N, Bajraktari G, Beishenkulov M, Kamzola G, Abdel-Massih T, Celutkiene J, Noppe S, Cassar A, Vataman E, AbirKhalil S, van Pol P, Mo R, Straburzynska-Migaj E, Fonseca C, Chioncel O, Shlyakhto E, Zavatta M, Otasevic P, Goncalvesova E, Lainscak M, Molina BD, Schaufelberger M, Suter T, Yılmaz MB, Voronkov L, Davies C. 2016 ESC Guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J. 2016;37:2129–2200m. [DOI] [PubMed] [Google Scholar]
- 26.Abraham WT, Ponikowski P, Brueckmann M, Zeller C, Macesic H, Peil B, Brun M, Ustyugova A, Jamal W, Salsali A, Lindenfeld JA, Anker SD, Abraham W, Lindenfeld JA, Brueckmann M, Salsali A, Welty F, Clayton T, Greenberg B, Konstam M, Lees K, Palmer M, Parhofer K, Pedersen T, Carson P, Freston J, Kaplowitz N, Lewis J, Mann J, Petrie J, Agostoni P, Butler J, Desai A, Filippatos G, Howlett J, Wranicz J, Mas JR, Cardoso JS, Störk S. Rationale and design of the EMPERIAL-Preserved and EMPERIAL-Reduced trials of empagliflozin in patients with chronic heart failure. Eur J Heart Fail. 2019;21:932–942. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Anker SD, Butler J, Filippatos GS, Jamal W, Salsali A, Schnee J, Kimura K, Zeller C, George J, Brueckmann M, Zannad F, Packer M, Packer M, Butler J, George J, Brueckmann M, Perrone S, Nicholls S, Janssens S, Bocchi E, Giannetti N, Verma S, Jian Z, Gomez Mesa JE, Spinar J, Böhm M, Merkely B, Chopra V, Senni M, Taddi S, Tsutsui H, Chuquiure E, La Rocca HPB, Ponikowski P, Vinereanu D, Sim D, Choi DJ, Juanatey JRG, Squire I, Butler J, Januzzi J, Pina I, Pocock SJ, Carson P, Doehner W, Miller A, Haas M, Pehrson S, Komajda M, Anand I, Teerlink J, Rabinstein A, Steiner T, Kamel H, Tsivgoulis G, Lewis J, Freston J, Kaplowitz N, Mann J, Petrie M, Bernstein R, Cheung A, Green J, Januzzi J, Kaul S, Ping CLS, Lip G, Marx N, McCullough P, Mehta C, Rosenstock J, Sattar N, Scirica B, Tsutsui H, Wanner C, Welty FK, Parhofer KG, Clayton T, Pedersen TR, Lees KR, Konstam MA, Greenberg B, Palmer M. Evaluation of the effects of sodium–glucose co-transporter 2 inhibition with empagliflozin on morbidity and mortality in patients with chronic heart failure and a preserved ejection fraction: rationale for and design of the EMPEROR-Preserved Trial. Eur J Heart Fail. 2019;21:1279–1287. [DOI] [PubMed] [Google Scholar]
- 28.Nagueh SF. Left Ventricular Diastolic Function: Understanding Pathophysiology, Diagnosis, and Prognosis With Echocardiography. JACC Cardiovasc Imaging. 2020;13:228–244. [DOI] [PubMed] [Google Scholar]
- 29.Obokata M, Reddy YNV, Borlaug BA. Diastolic Dysfunction and Heart Failure With Preserved Ejection Fraction: Understanding Mechanisms by Using Noninvasive Methods. JACC Cardiovasc Imaging. 2020;13:245–257. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Risum N, Ali S, Olsen NT, Jons C, Khouri MG, Lauridsen TK, Samad Z, Velazquez EJ, Sogaard P, Kisslo J. Variability of global left ventricular deformation analysis using vendor dependent and independent two-dimensional speckle-tracking software in adults. J Am Soc Echocardiogr [Internet]. 2012;25:1195–1203. Available from: 10.1016/j.echo.2012.08.007 [DOI] [PubMed] [Google Scholar]
- 31.Donal E, Lund LH, Oger E, Hage C, Persson H, Reynaud A, Ennezat PV, Bauer F, Drouet E, Linde C, Daubert C. New echocardiographic predictors of clinical outcome in patients presenting with heart failure and a preserved left ventricular ejection fraction: A subanalysis of the Ka (Karolinska) Ren (Rennes) Study. Eur J Heart Fail. 2015;17:680–688. [DOI] [PubMed] [Google Scholar]
- 32.Donal E, Lund LH, Oger E, Bosseau C, Reynaud A, Hage C, Drouet E, Daubert JC, Linde C. Importance of combined left atrial size and estimated pulmonary pressure for clinical outcome in patients presenting with heart failure with preserved ejection fraction. Eur Heart J Cardiovasc Imaging. 2017;18:629–635. [DOI] [PubMed] [Google Scholar]
- 33.Zile MR, Gottdiener JS, Hetzel SJ, McMurray JJ, Komajda M, McKelvie R, Baicu CF, Massie BM, Carson PE. Prevalence and significance of alterations in cardiac structure and function in patients with heart failure and a preserved ejection fraction. Circulation. 2011;124:2491–2501. [DOI] [PubMed] [Google Scholar]
- 34.Park JJ, Park JB, Park JH, Cho GY. Global Longitudinal Strain to Predict Mortality in Patients With Acute Heart Failure. J Am Coll Cardiol. 2018;71:1947–1957. [DOI] [PubMed] [Google Scholar]
- 35.Reddy YNV, Carter RE, Obokata M, Redfield MM, Borlaug BA. A simple, evidence-based approach to help guide diagnosis of heart failure with preserved ejection fraction. Circulation. 2018;138:861–870. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Farsalinos KE, Daraban AM, Ünlü S, Thomas JD, Badano LP, Voigt JU. Head-to-Head Comparison of Global Longitudinal Strain Measurements among Nine Different Vendors: The EACVI/ASE Inter-Vendor Comparison Study. J Am Soc Echocardiogr. 2015;28:1171–1181.e2. [DOI] [PubMed] [Google Scholar]
- 37.Stokke TM, Hasselberg NE, Smedsrud MK, Sarvari SI, Haugaa KH, Smiseth OA, Edvardsen T, Remme EW. Geometry as a Confounder When Assessing Ventricular Systolic Function: Comparison Between Ejection Fraction and Strain. J Am Coll Cardiol. 2017;70:942–954. [DOI] [PubMed] [Google Scholar]
- 38.Rao VN, Murray E, Butler J, Cooper LB, Cox ZL, Fiuzat M, Green JB, Lindenfeld JA, McGuire DK, Nassif ME, O’Brien C, Pagidipati N, Sharma K, Vaduganathan M, Vardeny O, Fonarow GC, Mentz RJ, Greene SJ. In-Hospital Initiation of Sodium-Glucose Cotransporter-2 Inhibitors for Heart Failure With Reduced Ejection Fraction. J Am Coll Cardiol. 2021;78:2004–2012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Greene SJ, Butler J, Fonarow GC. In-hospital initiation of quadruple medical therapy for heart failure: making the post-discharge vulnerable phase far less vulnerable. Eur J Heart Fail. 2022;24:227–229. [DOI] [PubMed] [Google Scholar]
- 40.Haji K, Huynh Q, Wong C, Stewart S, Carrington M, Marwick TH. Improving the Characterization of Stage A and B Heart Failure by Adding Global Longitudinal Strain. JACC Cardiovasc Imaging. 2022;15:1380–1387. [DOI] [PubMed] [Google Scholar]
- 41.Lund LH, Pitt B, Metra M. Left ventricular ejection fraction as the primary heart failure phenotyping parameter. Eur J Heart Fail. 2022;24:1158–1161. [DOI] [PubMed] [Google Scholar]
- 42.Lund LH, Vedin O, Savarese G. Is ejection fraction in heart failure a limitation or an opportunity? Eur J Heart Fail. 2018;20:431–432. [DOI] [PubMed] [Google Scholar]
- 43.Khan MS, Shahid I, Fonarow GC, Greene SJ. Classifying heart failure based on ejection fraction: imperfect but enduring. Eur J Heart Fail. 2022;24:1154–1157. [DOI] [PubMed] [Google Scholar]
- 44.Reddy YNV. Using the ejection fraction to guide medical therapy: a necessary but imperfect first step. Eur J Heart Fail. 2022;24:779–781. [DOI] [PubMed] [Google Scholar]
- 45.Greenberg B, O’Connor CM, Felker GM. Classifying Heart Failure in the 21st Century: Matching Taxonomy With Science. JACC Hear Fail. 2021;9:771–773. [DOI] [PubMed] [Google Scholar]
- 46.Maurer MS, Packer M. How Should Physicians Assess Myocardial Contraction?: Redefining Heart Failure With a Preserved Ejection Fraction. JACC Cardiovasc Imaging. 2020;13:873–878. [DOI] [PubMed] [Google Scholar]
- 47.Patel RB, Vaduganathan M, Greene SJ, Butler J. Nomenclature in heart failure: a call for objective, reproducible, and biologically-driven terminology. Eur J Heart Fail. 2018;20:1379–1381. [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.