Abstract
Aims
Pathophysiology and prognostic implications of right ventricle (RV) dysfunction in heart failure are complex and incompletely elucidated. Cardiac magnetic resonance imaging (CMR) is the reference standard for RV quantification, but its clinical implications in non-ischaemic cardiomyopathy (NICM), in the context of myocardial fibrosis and functional mitral regurgitation are not well defined. We evaluated predictors, prognostic impact, and thresholds for defining significant RV dysfunction in NICM.
Methods and results
NICM patients (n = 624) undergoing CMR assessment during 2002–2017 were retrospectively studied. CMR’s quantification of right ventricular ejection fraction (RVEF) was evaluated against the primary outcome of all-cause mortality, heart transplant, and/or left ventricular assist device implantation in threshold and multivariable analyses. Mean RVEF was 43 ± 13%, and factors associated with reduced RVEF were male sex, New York Heart Association (NYHA) class III-IV, right bundle branch block, lower left ventricular ejection fraction, higher mitral regurgitant fraction (MR-RF) and right ventricle size in NICM. RVEF per 5% increase was independently associated with the primary endpoint hazards ratio (95% confidence interval) 0.80 (0.73–0.88), P < 0.001. RVEF ≤40% was the optimal threshold associated with worse prognosis, regardless of late gadolinium enhancement (LGE) or MR-RF quantification. On the other hand, higher LGE was associated with primary endpoint in patients with RVEF ≤ 40% only, while risk associated with MR-RF was significant dampened after adjusting for RVEF.
Conclusion
RVEF provides powerful risk stratification, with RVEF ≤ 40% defining significant RV dysfunction associated with adverse outcomes in NICM. The integration of quantitative CMR measurements for RVEF, LGE, and MR-RF provides comprehensive NICM risk prognostication.
Keywords: right ventricle, cardiomyopathy, cardiac magnetic resonance imaging, mitral regurgitation
Graphical Abstract
Graphical Abstract.
Impact of right ventricular ejection fraction (RVEF) on the primary endpoint in non-ischaemic cardiomyopathy patients, including by late gadolinium enhancement (LGE) or mitral regurgitant fraction (MR-RF) categories.
Top figure—A) Cubic spine curve of the relative hazards of the primary endpoint by RVEF, adjusted for pre-specified covariates, with optimal threshold of <40%. Bottom figures—annualized primary endpoint event rate in NICM by RVEF subgroup and (B) LGE category or (C) MR-RF category. P-values are pairwise comparisons using log-rank test adjusted for multiplicity using Bonferroni-Holm method
Introduction
The right ventricle and its impact in cardiovascular diseases has had significant recent clinical and research interest, given the advances in multi-modality cardiac imaging to assess the right heart and heart failure therapies beyond targeting left ventricular systolic dysfunction.1–4 Echocardiography is the first-line imaging modality to assess the ventricle size and function using a combination of qualitative, semi-quantitative, and quantitative parameters, but is often limited because of suboptimal visualization, inaccuracies in chamber quantification and lack of standardization of measurements.5,6 Cardiac magnetic resonance imaging (CMR) has become the established reference standard for right heart chamber quantification, providing important prognostic implications for cardiovascular conditions such as heart failure, valvular heart disease, pulmonary hypertension, and congenital heart diseases.6,7 Some recent studies have demonstrated the prognostic value of RV function derived by CMR in both ischaemic and non-ischaemic cardiomyopathies.8–11 However, knowledge gaps remain regarding predictors and associations of right ventricular dysfunction, as well as the impact of LGE and function mitral regurgitation on the prognostic predictive ability for RV dysfunction to predict survival and response to treatment. Determination of predictors and associations with RV dysfunction may elucidate pathophysiologic mechanisms that can guide optimization of therapeutic interventions. This study aimed to evaluate the prevalence, associated factors, and outcomes of RV dysfunction quantified by CMR in non-ischaemic cardiomyopathy (NICM).
Methods
Study participants
The ethics approval of this study was obtained from the Institution Review Board of our institution (IRB 11–063) with a patient consent waiver. Patients having CMR with the clinical indication of NICM evaluation during 1/1/2002 and 12/31/2017 at our institution were searched. NICM was defined as heart failure with reduced ejection fraction including left ventricular ejection fraction (LVEF) < 50% and the absence of a history of myocardial infarction, prior revascularization or coronary angiography demonstrating significant obstructive coronary disease (≥70% in ≥1 epicardial coronary artery) or ischaemic scar on CMR. Other exclusion criteria included left ventricular ejection fraction ≥ 50%, prior valve surgery, primary mitral regurgitation, cardiac amyloidosis, acute myocarditis, cardiac sarcoidosis, hypertrophic cardiomyopathy, heart transplantation, and/or left ventricular assist device and congenital heart disease. We collected clinical characteristics pertaining to demographics, past medical history, symptoms, ECG, and renal function blood test results at the time of CMR from electronic medical records. If a patient had multiple CMRs, the first one during the study period was used.
Cardiac magnetic resonance imaging
Patients underwent CMR evaluation on the following CMR platforms: 1.5-Tesla (Achieva, Philips Medical Systems, Best, the Netherlands; Sonata and Avanto, Siemens Medical Solutions, Erlangen, Germany) and 3-Tesla (Ingenia, Philips Medical Systems, Best, the Netherlands) machines. CVI-42 software (Circle Cardiovascular Imaging, Calgary, Alberta, Canada) was used for quantitative CMR analyses. Left and right ventricle quantification were performed on left ventricular short axis stack series using steady state free precession, left and right ventricular end-diastolic volume indexed (LVEDVi and RVEDVi), end-systolic volume indexed (LVESVi and RVEDVi), stroke volume indexed (LVSVi and RVSVi), ejection fraction (LVEF and RVEF), left ventricular mass indexed (LVMi), and atrial volume indexed (LAVi and RAVi). In addition, mitral regurgitant volume (MR-RV) was calculated as LVSV minus aortic valve forward flow (on phase contrast sequence), and mitral regurgitant fraction (MR-RF) was calculated as MR-RV divided by LVSV. Gadolinium dimeglumine was administered intravenously at 0.2 mmol/kg and imaged 15–20 min later for late gadolinium enhancement (LGE) assessment, and quantified by the percentage of left ventricular myocardium with pixel intensity ≥5 standard deviation higher than the pixel intensity of user-defined reference normal myocardium, and distribution and pattern are also reported.12
Clinical outcomes
The primary composite endpoint of interest is all-cause death, heart transplant and/or left ventricular assist device (LVAD). Heart failure hospitalizations, implantable cardiac defibrillator implantation, cardiac resynchronization therapy, and mitral valve surgery was also recorded. All outcomes were analysed longitudinally as time-to-event variables, from the CMR date until the first event, death or date of last clinical encounter whichever is the earliest checked in January 2023 from electronic clinical records, and for deaths, the state obituary databases were also checked.
Statistical analyses
Continuous and categorical variables are presented as mean ± standard deviation or frequency (percentage). Clinical and CMR parameters associated with RVEF value were assessed in multivariable linear correlation analysis, reporting beta-coefficient, 95% confidence intervals (95%CI) and P-values. Multivariable Cox proportional hazards regression was used to assess whether RVEF and MR-RF are independently associated with the primary composite endpoints in NICM patients, after adjusting for the following covariates: age, female sex, body mass index, hypertension, diabetes, atrial fibrillation, New York Heart Association (NYHA) class 3–4, estimated glomerular filtration rate, LVEF, LGE, and time-dependent implantable cardiac defibrillators and/or cardiac resynchronization therapy device implantation. Hazards ratios (HR) were calculated with 95%CIs and P-values. Concordance index calculated for each multivariable model to assess performance. Optimal thresholds for impaired RVEF to detect the primary endpoint were determined using both the maximum Youden Index on receiver-operative characteristics curve, and restricted cubic spline function curve with 4 knots and linear tail-restricted adjusting for the above covariates in multivariable Cox regression. Kaplan-Meier survival curves were plotted by RVEF categories, and annualized primary event rates plotted based on RVEF categories and LGE or MR-RF categories from thresholds determined in a prior study,13 and P-values reported using pairwise comparisons using log-rank test adjusted for multiplicity using Bonferroni-Holm method. P-values <0.05 was deemed statistically significant and all tests were two-tailed. R version 3.6.0 was used for statistical analyses, and Prism version 8 was used for constructing figures.
Results
Cohort characteristics and outcomes
Amongst 624 NICM patients, mean age was 52.8 ± 15.9 years, and 253 (40.5%) were female (Table 1). Relevant mean CMR parameters include: RVEF 43 ± 13%, RVEDVi 85 ± 28 mL/m2, LVEF 33 ± 11%, LVEDVi 123 ± 42 mL/m2 [439 (70.4%) with dilated cardiomyopathy], MR-RF 14 ± 13% and LGE score 3 ± 7%. Cohort outcomes were reported over mean follow-up of 5.3 ± 4.5 years. The primary endpoint occurred in 113 (18.1%) patients, which included 94 (15.1%) deaths during follow-up. Mitral valve surgeries were performed in only 7 (1.1%) patients, and none transcatheter mitral valve procedures.
Table 1.
Cohort characteristics and outcomes
| Group | Total | RVEF ≥ 40% | RVEF < 40% | P-value |
|---|---|---|---|---|
| Number of patients | 624 | 301 | 323 | |
| Clinical parameters | ||||
| Age (years) | 52.8 ± 15.9 | 53.4 ± 15.0 | 52.7 ± 17.0 | 0.276 |
| Female | 253 (40.5%) | 146 (48.5%) | 107 (33.1%) | <0.001 |
| Body mass index (kg/m2) | 28.9 ± 6.6 | 28.9 ± 6.4 | 29.1 ± 6.6 | 0.633 |
| Body surface area (m2) | 2.02 ± 0.30 | 2.01 ± 0.28 | 2.04 ± 0.31 | 0.208 |
| Systolic blood pressure (mmHg) | 121 ± 19 | 123 ± 18 | 119 ± 20 | 0.015 |
| Heart rate (/minute) | 79 ± 17 | 76 ± 17 | 82 ± 18 | <0.001 |
| New York Heart Association Class | <0.001 | |||
| 1 | 219 (35.1%) | 133 (44.2%) | 86 (26.6%) | |
| 2 | 222 (35.6%) | 110 (36.5%) | 112 (34.7%) | |
| 3 | 160 (25.6%) | 55 (18.3%) | 105 (32.5%) | |
| 4 | 23 (3.7%) | 3 (1.0%) | 20 (6.2%) | |
| Hypertension | 300 (48.1%) | 145 (48.2%) | 155 (48.0%) | 1.000 |
| Diabetes | 102 (16.3%) | 46 (15.3%) | 56 (17.3%) | 0.517 |
| Atrial fibrillation | 138 (22.1%) | 56 (18.6%) | 82 (25.4%) | 0.043 |
| QRS duration (ms) | 114 ± 33 | 112 ± 30 | 115 ± 35 | 0.349 |
| Left bundle branch block | 111 (17.8%) | 47 (16.5%) | 66 (21.5%) | 0.143 |
| Right bundle branch block | 36 (5.8%) | 15 (5.0%) | 19 (5.9%) | 0.725 |
| Estimated glomerular filtration rate (mL/min/1.73m2) | 77 ± 19 | 79 ± 18 | 75 ± 20 | 0.008 |
| Beta-blocker | 428 (68.6%) | 231 (76.7%) | 197 (61.0%) | <0.001 |
| Angiotensin converting enzyme inhibitor or angiotensin receptor blocker | 370 (59.3%) | 192 (63.8%) | 178 (55.1%) | 0.028 |
| Aldosterone antagonist | 159 (25.5%) | 72 (23.9%) | 87 (26.9%) | 0.409 |
| Magnetic resonance imaging parameters | ||||
| Right ventricular end-diastolic volume indexed (mL/m2) | 85 ± 28 | 76 ± 22 | 94 ± 30 | <0.001 |
| Right ventricular end-systolic volume indexed (mL/m2) | 51 ± 25 | 37 ± 13 | 64 ± 26 | <0.001 |
| Right ventricular stroke volume indexed (mL/m2) | 35 ± 12 | 40 ± 11 | 30 ± 10 | <0.001 |
| Right ventricular ejection fraction (%) | 43 ± 13 | 53 ± 6 | 33 ± 9 | <0.001 |
| Left ventricular end-diastolic volume indexed (mL/m2) | 123 ± 42 | 115 ± 31 | 131 ± 49 | <0.001 |
| Left ventricular end-systolic volume indexed (mL/m2) | 85 ± 41 | 74 ± 28 | 95 ± 48 | <0.001 |
| Left ventricular stroke volume indexed (mL/m2) | 38 ± 12 | 41 ± 11 | 35 ± 12 | <0.001 |
| Left ventricular ejection fraction (%) | 33 ± 11 | 37 ± 8 | 29 ± 12 | <0.001 |
| Left ventricular mass indexed (g/m2) | 71 ± 25 | 66 ± 22 | 76 ± 28 | <0.001 |
| Left atrial volume indexed (mL/m2) | 46 ± 21 | 42 ± 17 | 51 ± 24 | <0.001 |
| Right atrial volume indexed (mL/m2) | 46 ± 20 | 41 ± 16 | 51 ± 22 | <0.001 |
| Mitral regurgitation volume | 12 ± 11 | 9 ± 9 | 13 ± 13 | <0.001 |
| Mitral regurgitation fraction | 14 ± 13 | 10 ± 10 | 17 ± 15 | <0.001 |
| Late gadolinium enhancement (present) | 291 (46.6%) | 124 (41.2%) | 152 (47.1%) | 0.147 |
| Late gadolinium enhancement score (%) | 3 ± 7 | 2.3 ± 5.5 | 3.3 ± 8.2 | 0.103 |
| Late gadolinium enhancement (distribution) | ||||
| Septal | 140 (22.4%) | 64 (21.3%) | 76 (23.5%) | 0.503 |
| Inferior | 87 (13.9%) | 37 (12.3%) | 50 (15.5%) | 0.298 |
| Lateral | 64 (10.3%) | 25 (8.3%) | 39 (12.1%) | 0.146 |
| Anterior | 52 (8.3%) | 22 (7.3%) | 30 (9.3%) | 0.388 |
| Right ventricular insertion point | 76 (12.2%) | 22 (7.3%) | 54 (16.7%) | <0.001 |
| Mid-myocardial | 226 (36.4%) | 96 (31.9%) | 131 (40.6%) | 0.025 |
| Epicardial | 36 (5.9%) | 16 (5.3%) | 21 (6.5%) | 0.612 |
| Subendocardial | 21 (3.4%) | 12 (4.0%) | 9 (2.8%) | 0.507 |
| Transmural | 39 (6.3%) | 13 (4.3%) | 26 (8.0%) | 0.068 |
| Outcomes | ||||
| Follow-up (years) | 4.0 ± 3.0 | 4.1 ± 2.8 | 4.0 ± 3.3 | 0.756 |
| Primary endpoint | 113 (18.1%) | 26 (8.6%) | 87 (26.9%) | <0.001 |
| Death | 94 (15.1%) | 24 (8.0%) | 70 (21.7%) | <0.001 |
| Heart transplant | 14 (2.2%) | 14 (4.7%) | 12 (3.7%) | 0.689 |
| Left ventricular assist device | 24 (3.8%) | 15 (5.0%) | 9 (2.8%) | 0.211 |
| Heart failure hospitalization | 135 (21.6%) | 54 (17.9%) | 81 (25.1%) | 0.033 |
| Implantable cardiac defibrillator | 167 (26.8%) | 67 (22.3%) | 100 (31.0%) | 0.015 |
| Cardiac resynchronization therapy | 67 (10.7%) | 39 (13.0%) | 28 (8.7%) | 0.093 |
| Mitral valve surgery | 7 (1.1%) | 1 (0.3%) | 5 (1.5%) | 0.218 |
Values are mean ± standard deviation or frequency (percentage), with student t-test and Fisher’s exact test P-values, respectively.
Factors significant associated with RVEF
Multiple linear regression was performed to determine independent covariates associated with RVEF, and results presented in Table 2. The following covariates were independently associated with higher RVEF (%) in NICM, with their beta-coefficients (95%CIs): female sex 3.20 (1.60, 4.70), P < 0.001; NYHA class 3–4 −2.57 (−4.33, −0.81), P = 0.004; right bundle branch block −4.53 (−7.77, −1.30), P = 0.006; LVEF per 10% 4.18 (3.44, 4.92), P < 0.001; MR-RF −0.90 (−1.19, −0.61), P < 0.001; and RVEDVi per 10 mL/m2 −1.36 (−1.66, −1.06), P < 0.001.
Table 2.
Multivariable linear regression analysis of the associations with right ventricular ejection fraction by magnetic resonance imaging, with beta-coefficients and 95% confidence intervals (95%CI)
| Parameter | Beta | 95% CI | P-value |
|---|---|---|---|
| Age (per 1-year) | 0.03 | −0.02, 0.09 | 0.220 |
| Female | 3.20 | 1.60, 4.79 | <0.001 |
| Body mass index (per 1 kg/m2) | −0.01 | −0.13, 0.11 | 0.886 |
| Hypertension | 1.14 | −0.52, 2.79 | 0.177 |
| Diabetes | −1.56 | −3.64, 0.53 | 0.143 |
| Atrial fibrillation | −0.60 | −2.47, 1.26 | 0.525 |
| New York Heart Association class (III/IV vs. I/II) | −2.57 | −4.33, -0.81 | 0.004 |
| Estimated glomerular filtration rate (per 10 mL/min/1.73m2) | 0.08 | −0.35, 0.51 | 0.721 |
| Right bundle branch block | −4.53 | −7.77, −1.30 | 0.006 |
| Left ventricular ejection fraction (per 10%) | 4.18 | 3.44, 4.92 | <0.001 |
| Late gadolinium enhancement score (per 5%) | −0.30 | −0.84, 0.23 | 0.262 |
| Mitral regurgitant fraction (per 5%) | −0.90 | −1.19, −0.61 | <0.001 |
| Right ventricular end-diastolic volume indexed (per 10 mL/m2) | −1.36 | −1.66, −1.06 | <0.001 |
Prognostic power of RVEF in multivariable analyses
Based on both the Youden Index on receiver-operative characteristics curves and cubic spline curve (Central Illustration), RVEF ≤ 40% (area under curve of 0.69) was identified as the optimal threshold for the primary endpoint. RVEF ≤ 40% was present in 239 (38.3%) patients. Figure 1 shows Kaplan-Meier survival curves of the primary endpoint by RVEF categories below and above 40%, with significant log-rank P < 0.001.
Figure 1.
Relationships between right ventricular ejection fraction (RVEF) and the primary endpoint during follow-up non-ischemic cardiomyopathy patients. A) Cubic spine curve of the relative hazards of the primary endpoint by RVEF, adjusted for pre-specified covariates, with optimal threshold of <40%. B) Kaplan-Meier survival curves of the primary endpoint in non-ischaemic cardiomyopathy patients with RVEF ≥40% and <40%.
Multivariable analyses were then performed to determine independent predictors of event free survival (Table 3). In model 1 without RVEF, MR-RF was associated with increased adverse outcomes per 5% increase HR 1.09, 95%CI 1.02–1.16, P = 0.009, even after adjusting for LGE. In model 2 without MR-RF, RVEF per 5% increase was independently associated with the primary endpoint with HR 0.79, 95%CI 0.72–0.86, P < 0.001, adjusting for other covariates. However, when LGE, MR-RF, and RVEF were all entered together in model 3, MR-RF was no longer a significant independent predictor (P = 0.107), while RVEF per 5% increase remained independently associated with the primary endpoint HR 0.80, 95%CI 0.83–0.88, P < 0.001. Other factors independently associated with the primary endpoint in NICM were lower body mass index, diabetes and higher LGE score. Concordance indices were 0.683 for model 1, 0.724 for model 2, and 0.728 for model 3 in Table 3. Substituting RVEF for RVSV/ESV in these models, a proposed surrogate for right ventricular to pulmonary artery coupling, this ratio per 0.1 increase was also independently associated with the primary endpoint HR 0.86, 95%CI 0.81–0.91, P < 0.001 in model 2 and HR0.86, 95%CI 0.80–0.92, P < 0.001 in model 3, respectively.
Table 3.
Multivariable cox regression models of the primary endpoint reporting hazards ratio (HR) and 95% confidence intervals (95%CI). The three models include one or both of mitral regurgitant fraction and right ventricular ejection fraction on CMR
| Model 1 | Model 2 | Model 3 | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Parameters | HR | 95% CI | P-value | HR | 95% CI | P-value | HR | 95% CI | P-value |
| Age (per 1 year) | 1.01 | 0.99, 1.02 | 0.225 | 1.01 | 0.99, 1.02 | 0.230 | 1.01 | 1.00, 1.03 | 0.177 |
| Female (vs. male) | 0.82 | 0.55, 1.22 | 0.316 | 1.14 | 0.75, 1.72 | 0.537 | 1.10 | 0.73, 1.67 | 0.643 |
| Body mass index (per 1 kg/m2) | 0.96 | 0.93, 0.99 | 0.012 | 0.96 | 0.93, 0.99 | 0.021 | 0.96 | 0.93, 0.99 | 0.023 |
| Hypertension | 0.89 | 0.58, 1.38 | 0.612 | 0.90 | 0.58, 1.4 | 0.650 | 0.91 | 0.58, 1.41 | 0.662 |
| Diabetes | 2.00 | 1.23, 3.26 | 0.005 | 1.98 | 1.21, 3.23 | 0.006 | 2.01 | 1.23, 32.9 | 0.005 |
| Atrial fibrillation | 1.21 | 0.78, 1.86 | 0.400 | 1.21 | 0.79, 1.86 | 0.388 | 1.17 | 0.76, 1.81 | 0.468 |
| New York Heart Association class (per 1 class) | 1.46 | 0.96, 2.23 | 0.074 | 1.26 | 0.81, 1.95 | 0.302 | 1.23 | 0.80, 1.91 | 0.344 |
| Estimated glomerular filtration rate (per 10 mL/min/1.73m2) | 1.02 | 0.92, 1.13 | 0.740 | 1.02 | 0.92, 1.13 | 0.776 | 1.03 | 0.92, 1.14 | 0.633 |
| Implantable cardiac defibrillator and/or cardiac resynchronization therapy (time-dependent variable) | 1.16 | 0.75, 1.80 | 0.504 | 1.36 | 0.87, 2.12 | 0.181 | 1.35 | 0.86, 2.11 | 0.190 |
| Left ventricular ejection fraction (per 10%) | 0.93 | 0.77, 1.13 | 0.485 | 1.19 | 0.95, 1.48 | 0.123 | 1.20 | 0.96, 1.49 | 0.104 |
| Late gadolinium enhancement (per 5%) | 1.15 | 1.05, 1.26 | 0.003 | 1.12 | 1.03, 1.22 | 0.012 | 1.13 | 1.03, 1.23 | 0.009 |
| Mitral regurgitant fraction (per 5%) | 1.09 | 1.02, 1.16 | 0.009 | — | — | — | 1.05 | 0.99, 1.12 | 0.107 |
| Right ventricular ejection fraction (per 5%) | — | — | — | 0.79 | 0.72, 0.86 | 0.000 | 0.80 | 0.73, 0.88 | <0.001 |
Event-free survival by RVEF, LGE, and MR-RF subgroups
Figure 2 also shows annualized primary endpoint rate by RVEF categories stratified by LGE and MR-RF threshold categories. In pairwise log-rank comparisons, RVEF ≤ 40% was associated with worse prognosis compared with RVEF > 40%, both in patients with LGE < 2% and in patients with LGE > 2%. However, LGE ≥ 2% was only associated with worse prognosis when RVEF is ≤40%, but not when RVEF was >40%. On the other hand, MR-RF ≥ 20% was not significantly associated with the primary endpoint whether RVEF is >40% or ≤40%. However, RVEF ≤ 40% was also associated with worse prognosis compared with RVEF > 40%, both in patients with MR-RF < 20% and in patients with MR-RF ≥ 20%.
Figure 2.
Annualized primary endpoint event rate in non-ischaemic cardiomyopathy (NICM) by right ventricular ejection fraction (RVEF) category and (A) late gadolinium enhancement (LGE) category or (B) mitral regurgitant fraction (MR-RF) category. P-values are pairwise comparisons using log-rank test adjusted for multiplicity using Bonferroni-Holm method.
Discussion
Previously referred to as the forgotten ventricle, several studies over the last decade have demonstrated the increasing prevalence and adverse prognostication of RV dysfunction when concomitant with a number of other cardiovascular diseases.8–11,14 Despite these findings, the interactions between RVEF with LGE and MR-RF, and their combined impact on prognosis remain incompletely understood. This study focused on quantitative CMR-evaluation of the RV in 624 NICM patients, and elucidated several important findings. Important clinical and CMR parameters that are associated with reduced RVEF include male sex, NYHA class, reduced LVEF, and increased RVEDVi and MR-RF. Reduced RVEF was independently associated with adverse prognosis in terms of higher rates of the primary endpoint of death, heart transplant and/or LVAD implantation. The optimal RVEF prognostic threshold was ≤40% which occurred in a significant minority 38.3% of patients. Furthermore, those with RVEF ≤ 40% had worse prognosis regardless of their quantified myocardial scar or functional mitral regurgitation category. However, knowledge gaps remain regarding predictors and associations of right ventricular dysfunction, as well as the impact of LGE and functional mitral regurgitation on the prognostic predictive ability for RV dysfunction to predict survival and response to treatment.
Right ventricular dysfunction pathophysiology and predictors
The pathophysiology of RV dysfunction remains complex, in part because of the heterogeneous range of aetiologies, broadly classified as conditions that directly impair RV contractility such as intrinsic RV cardiomyopathy, that causes RV volume overload or that lead to RV pressure overload.3,15 RV dysfunction is common in NICM, with heterogeneous aetiologies, such as intrinsic RV myopathy, secondary dysfunction as a result of chronic pressure/volume overload resulting for LV dysfunction/functional mitral regurgitation/pulmonary venous congestion/arrhythmia. While numerous studies have demonstrated how LGE of the LV provides strong prognostic and diagnostic insights in patients with myocardial disease, the relationship of LV LGE with RV dysfunction has not been entirely elucidated. Our study demonstrates that preserved RV function, mitigates risk associated with LV LGE, and that risk associated with RV dysfunction is further accentuated when LV LGE is present. The prevalence of RV dysfunction is our study was similar to a prior meta-analysis which reported this occurred in 47.2% (2234/4732) across 11 studies.16 Male sex was found to be associated with reduced RVEF (beta-coefficient −3.4%) in this study. This was also observed in other cardiomyopathy studies, and slightly lower mean RVEF values by 3–4% occur even in healthy men compared to women using CMR.10,17,18 We also found several markers of heart failure severity such as higher NYHA class, reducing LVEF and greater RVEDVi and MR-RF to be associated with lower RVEF, of which LVEF and functional MR severity have also been identified elsewhere.8,10,11 We like others also identified RBBB to be associated with reduced RVEF, while other studies have reported duration of NICM, RV LGE, haemodynamic measures of higher mean pulmonary arterial and right atrial pressures, pulmonary vascular resistance and lower cardiac index to also predict lower RVEF.8,10
Prognostic significance of RVEF in NICM, in the context of LGE and functional mitral regurgitation
Importantly, reduced RVEF was independently associated with the primary composite endpoint, even upon adjustment of other clinical and CMR covariates in the multivariable models. The findings are similar to the prior CMR study from our institution of ischaemic cardiomyopathy patients, where the adjusted hazards ratio for mortality was 1.17 per 10% drop in RVEF.8 In terms of NICM, one CMR study of 250 patients found RVEF < 45% to be associated with their primary endpoint of all-cause death or heart transplantation, with unadjusted and adjusted hazards ratios of 5.90 and 4.24 respectively.9 Another CMR also used RVEF < 45% to define RV dysfunction to be independently associated with composite primary endpoint (HR 3.19) and ventricular arrhythmias alone (HR 6.48).19 A third CMR study found RVEF to be independently associated with cardiac mortality with adjusted hazards ratio of 0.96.10 These results are similar to our findings. However, we uniquely adjusted for MR-RF and LGE score as quantitative measures of functional mitral regurgitation and myocardial scar respectively in our multivariable model, both of which are established CMR biomarkers of adverse prognosis in cardiomyopathies.13,20–22 Another study also found that reduced RVEF in implantable cardioverter defibrillator candidates was independently associated with defibrillator shock and death.23 These findings highlight the importance of RV function assessment in all NICM patients, at least as much if not greater than LVEF assessment, including in risk stratification for device implantation.
Interestingly, we report for the first time that although RVEF < 40% was associated with worse prognosis regardless of the LGE or MR-RF category, higher LGE score was only associated with worse prognosis in patients with RVEF ≤ 40%, but not if RVEF > 40%. These findings imply the LGE’s prognostic value in NICM patients is predominantly in those with RV dysfunction rather than those without. Furthermore, the highest risk NICM group was patients with RV dysfunction and significant LGE who thus warrant the most immediate clinical attention. Interestingly, although MR-RF was associated with the primary endpoint, but this association is attenuated when RVEF is entered into the model. This was somewhat surprising as MR-RF had been an established adverse prognosticator in other CMR cardiomyopathy studies, although most of those did not adjust for RVEF.20,21,24 These findings suggest that right ventricular dysfunction may be an important mediator in the adverse prognosis associated with functional mitral regurgitation, such that risk associated with MR-RF’s is modulated by RVEF. Our findings demonstrate that risk associated with MR RF ≥ 20% is significantly attenuated when RVEF > 40% vs. RVEF ≤ 40%. The finding is supported by the significant correlation between MR-RF and RVEF in our linear regression analysis. Our findings emphasize the importance of routine quantification of RVEF, LGE, and MR-RF in NICM patients having CMR, and that prognostic interpretation of these parameters should be made based on their combination rather than individually.
Clinical implications
There are a number of clinical implications from this study. Our study demonstrates the importance of assessing the presence and degree of RV dysfunction in all cardiomyopathy patients. RV dysfunction was prevalent and associated with specific clinical and imaging parameters and highlight the importance of RV functional assessment in patients with non-ischaemic cardiomyopathy. Our study demonstrates the value of quantitative RV function assessment by CMR, in addition to comprehensive left heart and mitral valve quantitative measures. Patients undergoing CMR with known or suspected cardiomyopathy should have quantitative RV assessment in addition to the left ventricle, given its powerful prognostic value. While RVEF is currently not part of any of the major heart failure risk scores or guidelines for risk stratification in patients with heart failure,1,2,25 further studies are needed to determine if incorporation of RVEF into future risk models might result in improved predictive accuracy. Furthermore, our study elucidates novel relationships between LGE, MR-RF, and RVEF and their relative and combined impact on risk, which have important clinical implications in regards to response to device and valve interventions. Management strategies for RV dysfunction specifically remain limited given the lack of proven therapies to improve survival, but should follow current recommendations including maximizing medical therapy for left ventricular dysfunction, optimizing volume status with diuretics, and in selected cases consider pulmonary vasodilators, inotropes and mechanical support.1–3 Treatments of associated factors that can worsen RV function such as arrhythmias, valvular heart disease, inflammation, and chronic lung disease may also be beneficial.26 Finally, prior studies have indicated RV dysfunction and RV-pulmonary arterial coupling to be associated with adverse prognosis in patients with functional MR undergoing transcatheter mitral interventions,27 and have demonstrated greater RV dysfunction in men similar to our study,28 Further studies are necessary to determine how sex differences in RV dysfunction and right sided ventriculo-vascular coupling impact appropriate treatment strategy for functional MR.
Limitations
This is a retrospective cohort study, but represents real-world data from a large single-centre CMR study. NICM patients referred for CMR may not be representative of all NICM patients. NICM aetiology was not available in many patients especially early on to analyse. A minority of patients had atrial fibrillation (8.2%) at time of CMR which may limit accuracy of CMR quantification. More advanced CMR techniques such as parametric mapping, extracellular volume, and right and left ventricular strain were not evaluated. We focused on CMR rather than other imaging modalities, and did not analyse echocardiography and right heart catheterization parameters including right heart and pulmonary pressures and RV to pulmonary arterial coupling. Biomarkers such as brain natriuretic peptide were unavailable especially for patients prior to 2010 to allow for analyses. Serial CMR was rarely undertaken to assess chamber remodelling over time. There was inconsistent information regarding cause of death to be analysed. A minority of patients from other states and countries referred to our institution have limited follow-up. Treatments for NICM patients have evolved over time and are not accounted for in this analysis, and thus the current study does not reflect impact of angiotensin receptor/neprilysin inhibitor and sodium–glucose cotransporter-inhibitor therapy, which were not widely utilized during the study period.
Conclusion
In conclusion, RV dysfunction is prevalent in NICM patients. Male sex, higher NYHA class, RBBB, reduced LVEF, increased RVEDVi, and MR-RF were significantly and independently associated with RVEF. RVEF decrease was independently associated with higher rates of the primary endpoint adjusting for clinical and CMR covariates. RVEF ≤ 40% was the optimal prognostic threshold, and was associated with adverse prognosis regardless of the LGE score or MR-RF category, while the LGE score was only associated with adverse prognosis in patients with RVEF ≤ 40% but not in patients with RVEF > 40%. Our findings add to our understanding of the prognostic value and associated factors of CMR-derived RV assessment in NICM patients, and stresses the importance of interpreting of RVEF, LGE, and MR-RF in combination rather than separately.
Acknowledgements
Not applicable
Contributor Information
Tom Kai Ming Wang, Section of Cardiovascular Imaging, Heart, Vascular and Thoracic Institute, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195, USA; Section of Cardiovascular Imaging, Imaging Institute, Cleveland Clinic, Cleveland, OH, USA.
Duygu Kocyigit, Section of Cardiovascular Imaging, Heart, Vascular and Thoracic Institute, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195, USA.
Nicholas Chan, Section of Cardiovascular Imaging, Heart, Vascular and Thoracic Institute, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195, USA.
Donna Salam, Section of Cardiovascular Imaging, Heart, Vascular and Thoracic Institute, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195, USA.
Mustafa Turkmani, Section of Cardiovascular Imaging, Heart, Vascular and Thoracic Institute, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195, USA.
Jennifer Bullen, Department of Quantitative Health Sciences, Cleveland Clinic, Cleveland, OH, USA.
Zoran B Popović, Section of Cardiovascular Imaging, Heart, Vascular and Thoracic Institute, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195, USA; Section of Cardiovascular Imaging, Imaging Institute, Cleveland Clinic, Cleveland, OH, USA.
Christopher Nguyen, Section of Cardiovascular Imaging, Imaging Institute, Cleveland Clinic, Cleveland, OH, USA.
Brian P Griffin, Section of Cardiovascular Imaging, Heart, Vascular and Thoracic Institute, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195, USA.
W H Wilson Tang, Section of Heart Failure and Transplantation Medicine, Heart, Vascular and Thoracic Institute, Cleveland Clinic, Cleveland, OH, USA.
Deborah H Kwon, Section of Cardiovascular Imaging, Heart, Vascular and Thoracic Institute, Cleveland Clinic, 9500 Euclid Avenue, Cleveland, OH 44195, USA; Section of Cardiovascular Imaging, Imaging Institute, Cleveland Clinic, Cleveland, OH, USA.
Funding
None declared.
Data availability
The data underlying this article will be shared on reasonable request to the corresponding author.
References
- 1. Heidenreich PA, Bozkurt B, Aguilar D, Allen LA, Byun JJ, Colvin MM et al. 2022 AHA/ACC/HFSA guideline for the management of heart failure: a report of the American College of Cardiology/American Heart Association joint committee on clinical practice guidelines. J Am Coll Cardiol 2022;79:e263–421. [DOI] [PubMed] [Google Scholar]
- 2. McDonagh TA, Metra M, Adamo M, Gardner RS, Baumbach A, Böhm M et al. 2021 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure. Eur Heart J 2021;42:3599–726. [DOI] [PubMed] [Google Scholar]
- 3. Konstam MA, Kiernan MS, Bernstein D, Bozkurt B, Jacob M, Kapur NK et al. Evaluation and management of right-sided heart failure: a scientific statement from the American Heart Association. Circulation 2018;137:e578–622. [DOI] [PubMed] [Google Scholar]
- 4. Amsallem M, Mercier O, Kobayashi Y, Moneghetti K, Haddad F. Forgotten No more: a focused update on the right ventricle in cardiovascular disease. JACC Heart Fail 2018;6:891–903. [DOI] [PubMed] [Google Scholar]
- 5. Rudski LG, Lai WW, Afilalo J, Hua L, Handschumacher MD, Chandrasekaran K et al. Guidelines for the echocardiographic assessment of the right heart in adults: a report from the American society of echocardiography endorsed by the European association of echocardiography, a registered branch of the European Society of Cardiology, and the Canadian society of echocardiography. J Am Soc Echocardiogr 2010;23:685–713. quiz 86–8. [DOI] [PubMed] [Google Scholar]
- 6. Lang RM, Badano LP, Mor-Avi V, Afilalo J, Armstrong A, Ernande L et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American society of echocardiography and the European association of cardiovascular imaging. J Am Soc Echocardiogr 2015;28:1–39.e14. [DOI] [PubMed] [Google Scholar]
- 7. Kramer CM, Barkhausen J, Bucciarelli-Ducci C, Flamm SD, Kim RJ, Nagel E. Standardized cardiovascular magnetic resonance imaging (CMR) protocols: 2020 update. J Cardiovasc Magn Reson 2020;22:17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Sabe MA, Sabe SA, Kusunose K, Flamm SD, Griffin BP, Kwon DH. Predictors and prognostic significance of right ventricular ejection fraction in patients with ischemic cardiomyopathy. Circulation 2016;134:656–65. [DOI] [PubMed] [Google Scholar]
- 9. Gulati A, Ismail TF, Jabbour A, Alpendurada F, Guha K, Ismail NA et al. The prevalence and prognostic significance of right ventricular systolic dysfunction in nonischemic dilated cardiomyopathy. Circulation 2013;128:1623–33. [DOI] [PubMed] [Google Scholar]
- 10. Pueschner A, Chattranukulchai P, Heitner JF, Shah DJ, Hayes B, Rehwald W et al. The prevalence, correlates, and impact on cardiac mortality of right ventricular dysfunction in nonischemic cardiomyopathy. JACC Cardiovasc Imaging 2017;10(10 Pt B):1225–36. [DOI] [PubMed] [Google Scholar]
- 11. Fahmy AS, Csecs I, Arafati A, Assana S, Yankama TT, Al-Otaibi T et al. An explainable machine learning approach reveals prognostic significance of right ventricular dysfunction in nonischemic cardiomyopathy. JACC Cardiovasc Imaging 2022;15:766–79. [DOI] [PubMed] [Google Scholar]
- 12. Setser RM, Bexell DG, O’Donnell TP, Stillman AE, Lieber ML, Schoenhagen P et al. Quantitative assessment of myocardial scar in delayed enhancement magnetic resonance imaging. J Magn Reson Imaging 2003;18:434–41. [DOI] [PubMed] [Google Scholar]
- 13. Wang TKM, Kocyigit D, Choi H, Anthony CM, Chan N, Bullen J et al. Prognostic power of quantitative assessment of functional mitral regurgitation and myocardial scar quantification by cardiac magnetic resonance. Circ Cardiovasc Imaging 2023;16:e015134. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Cremer PC, Zhang Y, Alu M, Rodriguez LL, Lindman BR, Zajarias A et al. The incidence and prognostic implications of worsening right ventricular function after surgical or transcatheter aortic valve replacement: insights from PARTNER IIA. Eur Heart J 2018;39:2659–2667. [DOI] [PubMed] [Google Scholar]
- 15. Thandavarayan RA, Chitturi KR, Guha A. Pathophysiology of acute and chronic right heart failure. Cardiol Clin 2020;38:149–60. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Iglesias-Garriz I, Olalla-Gómez C, Garrote C, López-Benito M, Martín J, Alonso D et al. Contribution of right ventricular dysfunction to heart failure mortality: a meta-analysis. Rev Cardiovasc Med 2012;13:e62–9. [DOI] [PubMed] [Google Scholar]
- 17. Kawel-Boehm N, Hetzel SJ, Ambale-Venkatesh B, Captur G, Francois CJ, Jerosch-Herold M et al. Reference ranges (“normal values”) for cardiovascular magnetic resonance (CMR) in adults and children: 2020 update. J Cardiovasc Magn Reson 2020;22:87. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Kawut SM, Lima JA, Barr RG, Chahal H, Jain A, Tandri H et al. Sex and race differences in right ventricular structure and function: the multi-ethnic study of atherosclerosis-right ventricle study. Circulation 2011;123:2542–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Becker MAJ, Cornel JH, van de Ven PM, van Rossum AC, Allaart CP, Germans T. The prognostic value of late gadolinium-enhanced cardiac magnetic resonance imaging in nonischemic dilated cardiomyopathy: a review and meta-analysis. JACC Cardiovasc Imaging 2018;11:1274–84. [DOI] [PubMed] [Google Scholar]
- 20. Cavalcante JL, Kusunose K, Obuchowski NA, Jellis C, Griffin BP, Flamm SD et al. Prognostic impact of ischemic mitral regurgitation severity and myocardial infarct quantification by cardiovascular magnetic resonance. JACC Cardiovasc Imaging 2019;1313:1489–1501. [DOI] [PubMed] [Google Scholar]
- 21. Tayal B, Debs D, Nabi F, Malahfji M, Little SH, Reardon M et al. Impact of myocardial scar on prognostic implication of secondary mitral regurgitation in heart failure. JACC Cardiovasc Imaging 2021;14:812–22. [DOI] [PubMed] [Google Scholar]
- 22. Klem I, Klein M, Khan M, Yang EY, Nabi F, Ivanov A et al. Relationship of LVEF and myocardial scar to long-term mortality risk and mode of death in patients with nonischemic cardiomyopathy. Circulation 2021;143:1343–58. [DOI] [PubMed] [Google Scholar]
- 23. Jimenez-Juan LA-O, Ben-Dov N, Goncalves Frazao CA-O, Tan NA-O, Singh SA-O, Dorian P et al. Right ventricular function at cardiac MRI predicts cardiovascular events in patients with an implantable cardioverter-defibrillator. Radiology 2021;301:322–329. [DOI] [PubMed] [Google Scholar]
- 24. Kwon DH, Kusunose K, Obuchowski NA, Cavalcante JL, Popovic ZB, Thomas JD et al. Predictors and prognostic impact of progressive ischemic mitral regurgitation in patients with advanced ischemic cardiomyopathy: a multimodality study. Circ Cardiovasc Imaging 2016;9:e004577. [DOI] [PubMed] [Google Scholar]
- 25. Levy WC, Mozaffarian D, Linker DT, Sutradhar SC, Anker SD, Cropp AB et al. The Seattle heart failure model: prediction of survival in heart failure. Circulation 2006;113:1424–33. [DOI] [PubMed] [Google Scholar]
- 26. January CT, Wann LS, Calkins H, Chen LY, Cigarroa JE, Cleveland JC Jr et al. 2019 AHA/ACC/HRS focused update of the 2014 AHA/ACC/HRS guideline for the management of patients with atrial fibrillation: a report of the American College of Cardiology/American Heart Association task force on clinical practice guidelines and the heart rhythm society in collaboration with the society of thoracic surgeons. Circulation 2019;140:e125–51. [DOI] [PubMed] [Google Scholar]
- 27. Brener MI, Grayburn P, Lindenfeld J, Burkhoff D, Liu M, Zhou Z et al. Right ventricular-pulmonary arterial coupling in patients with HF secondary MR: analysis from the COAPT trial. JACC Cardiovasc Interv 2021;14:2231–42. [DOI] [PubMed] [Google Scholar]
- 28. Tello K, Richter MJ, Yogeswaran A, Ghofrani HA, Naeije R, Vanderpool R et al. Sex differences in right ventricular-pulmonary arterial coupling in pulmonary arterial hypertension. Am J Respir Crit Care Med 2020;202:1042–6. [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.
Data Availability Statement
The data underlying this article will be shared on reasonable request to the corresponding author.