https://orcid.org/0000-0002-5473-0688
Abstract
This study sought to determine the independent and incremental prognostic value of semiquantitative measures of tricuspid regurgitation (TR) severity over right heart remodelling and pulmonary hypertension (PH) in heart failure with preserved ejection fraction (HFpEF).
Echocardiography was performed on 311 HFpEF patients. TR severity was defined by the semiquantitative measures [i.e. vena contracta width (VCW) and jet area] and by the guideline-based integrated qualitative approach (absent, mild, moderate, or severe). All-cause mortality or heart failure hospitalization occurred in 101 patients over a 2.1-year median follow-up. There was a continuous association between TR severity and the composite outcome with a hazard ratio (HR) of 1.17 per 1 mm increase of VCW [95% confidence interval (CI) 1.08–1.26, P < 0.0001]. Compared with patients with the lowest VCW category (≤1 mm), RV-adjusted HRs for the outcome were 1.99 (95% CI 1.05–3.77), 2.63 (95% CI 1.16–5.95), and 5.00 (95% CI 1.60–15.7) for 1–3, 3–7, and ≥7 mm VCW categories, respectively. TR severity as defined by the guideline-based approach showed a similarly graded association, but it was no longer significant in models including PH. In contrast, VCW remained independently and incrementally associated with the outcome after adjusting for established prognostic factors, as well as RV diameter and PH (fully adjusted HR 1.14 per 1 mm, 95% CI 1.02–1.27, P = 0.02; χ2 58.8 vs. 51.5, P = 0.03).
The current data highlight the potential value of the semiquantitative measures of TR severity for the risk stratification in patients with HFpEF.
Introduction
Approximately one-half of all patients with heart failure (HF) have a preserved ejection fraction (HFpEF), for which there are very few effective treatments.1 Tricuspid regurgitation (TR) is a common valvular disorder in patients with HFpEF.2,3 While increasing evidence suggests that clinically significant TR has adverse effects on the outcomes of various diseases, such as pulmonary arterial hypertension (PAH), left-sided valvular heart disease, and heart failure with reduced ejection fraction (HFrEF),4–8 its prognostic significance remains uncertain in patients with HFpEF.
Previous studies have demonstrated that increased severity of TR in HFpEF is associated with an increased risk of worse outcomes in univariable analysis, but this correlation is no longer predictable after adjusting for RV dysfunction.2,9 Another small study reported no relationship between the severity of TR and all-cause mortality in patients with HFpEF.10 One potential explanation could be related to the limitation of the qualitative assessment of TR, since grading TR does not account for the individual risk within each TR category.11 In this regard, recent studies have demonstrated the prognostic value of continuous (quantitative) TR metrics to assess severity in patients with HFrEF.6,12
Since TR in HFpEF typically develops secondary to right heart remodelling caused by elevated left ventricular (LV) filling pressure and subsequent pulmonary hypertension (PH), the prognostic impact of TR can be influenced by concomitant RV remodelling and PH. Thus, it is important to account for right heart remodelling and severity of PH to elucidate the true prognostic value of TR in patients with HFpEF.
Accordingly, the purposes of this study were (i) to determine the association between the continuous measures of TR severity, including vena contracta width (VCW) and jet area, and clinical outcomes; (ii) to assess whether they had independent and incremental prognostic value over right heart remodelling and PH; and (iii) to contrast the prognostic value of the semiquantitative TR metrics with the guideline-based qualitative approach.
Methods
Study population
This was a retrospective observational study that assessed the association between the severity of TR and clinical outcomes in patients with HFpEF. We identified 27 493 subjects who were referred to echocardiographic laboratories of the Gunma University Hospital in Maebashi, Japan (n = 17 367 [63%]), or Hokkaido University Hospital in Sapporo, Japan (n = 10 126 [37%]) between January 2014 and December 2018. Inclusion criteria for HFpEF was defined by the heart failure (HF) clinical symptoms (exertional dyspnoea, fatigue, and oedema), EF ≥50%, and evidence of abnormal LV diastolic function [directly measured pulmonary capillary wedge pressure (PCWP) >15 mmHg, B-type natriuretic peptide (BNP) levels >200 pg/mL, the ratio of early diastolic mitral inflow velocity to early diastolic mitral annular tissue velocity (E/e′) >15, left atrial (LA) volume index >34 mL/m2, or previous HF-related hospitalization].13,14 Subjects with HFrEF (EF < 50%), recovered EF (previous EF <40%), PAH, significant left-sided valvular heart disease, acute coronary syndrome, congenital heart disease, cardiomyopathies, or previous tricuspid valve surgery were excluded. From this group, patients with comprehensive echocardiographic evaluation in a compensated state of HF (at outpatient or discharge from HF hospitalization) were identified. If patients had multiple echocardiograms during the study period, the oldest exam was used as an index evaluation. This study was approved by the Clinical Research Review Board of both hospitals. The data will not be available to other researchers because the institutional ethics review boards implement the restriction for publish data sharing.
Data, including clinical demographics, past medical history, current medications, and laboratory results, were collected from a detailed chart review. Atrial fibrillation (AF) was classified as either earlier AF (patients in sinus rhythm during the echocardiographic assessment, but with a previous AF diagnosis) or current AF (patients in AF rhythm during the assessment).15
Cardiac structure and function assessment
Echocardiography was performed according to the contemporary guidelines.16 LV volume, LA volume, and EF were determined using the biplane method of disks. Stroke volume was determined from the LV outflow dimension and pulse-Doppler wave. LV diastolic function was assessed using mitral inflow velocities and septal E/e′ ratio. Right atrial (RA) pressure was estimated from the diameter of the inferior vena cava and its respiratory change.16 RV systolic pressure (eRVSP) was then calculated as 4 × peak TR velocity2 + estimated RA pressure. eRVSP was assumed to be 20 mmHg in 30 patients, where TR velocity could not be obtained. RV basal and mid-ventricular dimensions were measured at end-diastole using RV focused views. Tricuspid annular diameter was measured in an apical four-chamber view at end-diastole.11 RV systolic function was assessed by tricuspid annular plane systolic excursion (TAPSE). RA volumes were measured in the apical four-chamber view, and RA expansion index, an index of RA reservoir function, was calculated as previously described.15
Semiquantitative metrics of TR severity (VCW and colour flow jet area) were measured during mid-systole in both apical four-chamber and RV inflow views with a Nyquist limit of 50–70 cm/s, and a higher value for each was used for the primary analysis.11 TR severity was then categorized as absent (none or trivial), mild, moderate, or severe based upon the guideline-recommended multiparametric approach integrating tricuspid valve morphology, the semiquantitative indices, visual assessment, and hepatic vein flow pulse-Doppler. The severity of TR was also classified into four groups according to the VCW values (≤1, 1–3, 3–7, and ≥7 mm). Measurements of VCW and jet area represent the mean of two beats in sinus rhythm and ≥ 3 beats in AF.
Invasive haemodynamics
Right heart catheterization was performed within two weeks from echocardiographic evaluation in a subset of patients. Pressures in the RA, pulmonary artery (PA), and PCWP were measured at end-expiration. Cardiac output was determined using the thermodilution method.
Outcome assessment
All subjects were followed up from the day of their echocardiogram. The pre-specified primary end point of this study was a composite of either all-cause mortality or HF hospitalization. HF hospitalization was defined as dyspnoea and pulmonary oedema on chest X-ray requiring intravenous diuretic treatment.
Statistical analysis
Data are reported as mean (SD), median (IQR), or number (%). Between-group differences were analysed using a χ2, ANOVA, or Kruskal–Wallis test. Event-free rates were assessed using the Kaplan–Meier curve analysis. The non-linear continuous relationship of VCW and jet area with the composite outcome was assessed using a Cox model. The independent prognostic power of TR severity was determined using Cox proportional hazards models, in which non-normally distributed data were log-transformed. Given the presence of 0 cm2, jet area was log-transformed after the addition of 1 cm2. The VCW was modelled as an ordinal (as described above) or a continuous variable. To account for potential confounding factors, multivariable Cox models were created on the basis of a priori knowledge: model (1) adjusted for age, sex, LA volume index, AF, mitral regurgitation (MR) severity, BNP, and cardiac implantable electrical devices (CIEDs); model (2) model 1 and RV basal diameter; and model (3) model 2 and eRVSP.6
The incremental value for the prediction of outcomes was evaluated by sequential Cox proportional hazards analysis using nested models. The change in overall −2 log likelihood ratios of the models was used to assess the increase in predictive power after adding a subsequent parameter. All tests were two-sided, with a P value <0.05 considered significant. All analyses were performed with JMP 14.0.0 (SAS Institute, Cary, NC, USA) or R version 3.6.1 (The R Foundation for Statistical Computing, Vienna, Austria).
Results
Clinical characteristics of subjects according to TR severity assessed with VCW
Of 346 patients who met the inclusion criteria, 30 and 5 patients were further excluded due to lack of follow-up data and previous tricuspid valve surgery, respectively, remaining 311 HFpEF patients for the final analysis. Table 1 displays baseline characteristics of the study subjects according to VCW categories, and Supplementary data online, Table S1 displays characteristics according to the guideline-based integrated approach. Of 311 HFpEF patients, 163 (52%), 60 (19%), and 13 patients (4%) had a mild, moderate, and severe range of VCW, respectively. TR severity assessed by VCW was associated with older age and a higher prevalence of both AF and CIEDs. There were no differences in sex, vital signs, or prevalence of other comorbidities. While the use of neurohormonal antagonists was similar between the VCW categories, there was a trend toward a higher use of diuretics with increasing TR severity. Increasing VCW was also associated with higher levels of BNP, γ-glutamyl transferase (γGT), and total bilirubin.
Patients demographics according to TR vena contracta width categories
| VCW ≤1 mm (n = 75) | VCW 1–3 mm (n = 163) | VCW 3–7 mm (n = 60) | VCW ≥7 mm (n = 13) | P value | |
|---|---|---|---|---|---|
| Age (years) | 68 ± 14 | 74 ± 12 | 77 ± 9 | 81 ± 10 | <0.0001 |
| Female, n (%) | 38 (51%) | 81 (50%) | 29 (48%) | 6 (46%) | 1.0 |
| Body mass index (kg/m2) | 24 ± 5 | 23 ± 4 | 22 ± 3 | 23 ± 4 | 0.08 |
| Vital signs | |||||
| Systolic blood pressure (mmHg) | 128 ± 23 | 128 ± 20 | 123 ± 18 | 122 ± 20 | 0.2 |
| Diastolic blood pressure (mmHg) | 70 ± 15 | 66 ± 14 | 66 ± 14 | 66 ± 10 | 0.2 |
| Heart rate (bpm) | 76 ± 18 | 72 ± 17 | 73 ± 16 | 73 ± 17 | 0.3 |
| Comorbidities | |||||
| Hypertension, n (%) | 58 (77%) | 134 (83%) | 46 (77%) | 11 (85%) | 0.6 |
| Coronary artery disease, n (%) | 21 (28%) | 42 (26%) | 8 (13%) | 1 (8%) | 0.09 |
| Earlier AF/current AF, n (%) | 8 (11%)/10 (13%) | 37 (23%)/46 (28%) | 15 (25%)/38 (63%) | 6 (46%)/7 (54%) | <0.0001 |
| Diabetes mellitus, n (%) | 30 (40%) | 55 (34%) | 18 (30%) | 2 (15%) | 0.3 |
| Cardiac implantable electrical devices, n (%) | 2 (3%) | 9 (6%) | 8 (13%) | 6 (46%) | <0.0001 |
| Medications | |||||
| ACEI or ARB, n (%) | 40 (53%) | 81 (50%) | 30 (51%) | 5 (38%) | 0.8 |
| Beta-blocker, n (%) | 28 (37%) | 68 (42%) | 31 (52%) | 3 (23%) | 0.2 |
| Diuretic, n (%) | 46 (61%) | 108 (66%) | 45 (75%) | 11 (85%) | 0.2 |
| MRA, n (%) | 25 (33%) | 59 (36%) | 24 (40%) | 5 (38%) | 0.9 |
| Laboratories | |||||
| Haemoglobin (g/dL) | 11.9 ± 2.4 | 11.5 ± 2.1 | 11.6 ± 2.5 | 10.8 ± 2.2 | 0.3 |
| Creatinine (mg/dL) | 0.9 (0.7–1.4) | 0.9 (0.7–1.2) | 1.0 (0.7–1.4) | 1.0 (0.7–1.8) | 0.8 |
| BNP (pg/mL) | 123 (47–369) | 187 (104–369) | 242 (174–391) | 282 (169–336) | 0.02 |
| Aspartate transaminase (U/L) | 23 (19–29) | 22 (18–30) | 25 (19–32) | 24 (22–34) | 0.08 |
| Alanine transaminase (U/L) | 17 (13–25) | 15 (10–22) | 16 (11–24) | 14 (11–25) | 0.1 |
| γ-glutamyl transferase (U/L) | 23 (17–48) | 26 (17–44) | 44 (27–81) | 34 (23–55) | 0.0004 |
| Alkaline phosphatase (U/mL) | 228 (186–295) | 239 (197–317) | 257 (208–332) | 301 (199–420) | 0.2 |
| Total-bilirubin (mg/dL) | 0.6 (0.5–0.8) | 0.7 (0.5–0.9) | 0.8 (0.6–1.2) | 0.8 (0.6–1.2) | 0.0001 |
| VCW ≤1 mm (n = 75) | VCW 1–3 mm (n = 163) | VCW 3–7 mm (n = 60) | VCW ≥7 mm (n = 13) | P value | |
|---|---|---|---|---|---|
| Age (years) | 68 ± 14 | 74 ± 12 | 77 ± 9 | 81 ± 10 | <0.0001 |
| Female, n (%) | 38 (51%) | 81 (50%) | 29 (48%) | 6 (46%) | 1.0 |
| Body mass index (kg/m2) | 24 ± 5 | 23 ± 4 | 22 ± 3 | 23 ± 4 | 0.08 |
| Vital signs | |||||
| Systolic blood pressure (mmHg) | 128 ± 23 | 128 ± 20 | 123 ± 18 | 122 ± 20 | 0.2 |
| Diastolic blood pressure (mmHg) | 70 ± 15 | 66 ± 14 | 66 ± 14 | 66 ± 10 | 0.2 |
| Heart rate (bpm) | 76 ± 18 | 72 ± 17 | 73 ± 16 | 73 ± 17 | 0.3 |
| Comorbidities | |||||
| Hypertension, n (%) | 58 (77%) | 134 (83%) | 46 (77%) | 11 (85%) | 0.6 |
| Coronary artery disease, n (%) | 21 (28%) | 42 (26%) | 8 (13%) | 1 (8%) | 0.09 |
| Earlier AF/current AF, n (%) | 8 (11%)/10 (13%) | 37 (23%)/46 (28%) | 15 (25%)/38 (63%) | 6 (46%)/7 (54%) | <0.0001 |
| Diabetes mellitus, n (%) | 30 (40%) | 55 (34%) | 18 (30%) | 2 (15%) | 0.3 |
| Cardiac implantable electrical devices, n (%) | 2 (3%) | 9 (6%) | 8 (13%) | 6 (46%) | <0.0001 |
| Medications | |||||
| ACEI or ARB, n (%) | 40 (53%) | 81 (50%) | 30 (51%) | 5 (38%) | 0.8 |
| Beta-blocker, n (%) | 28 (37%) | 68 (42%) | 31 (52%) | 3 (23%) | 0.2 |
| Diuretic, n (%) | 46 (61%) | 108 (66%) | 45 (75%) | 11 (85%) | 0.2 |
| MRA, n (%) | 25 (33%) | 59 (36%) | 24 (40%) | 5 (38%) | 0.9 |
| Laboratories | |||||
| Haemoglobin (g/dL) | 11.9 ± 2.4 | 11.5 ± 2.1 | 11.6 ± 2.5 | 10.8 ± 2.2 | 0.3 |
| Creatinine (mg/dL) | 0.9 (0.7–1.4) | 0.9 (0.7–1.2) | 1.0 (0.7–1.4) | 1.0 (0.7–1.8) | 0.8 |
| BNP (pg/mL) | 123 (47–369) | 187 (104–369) | 242 (174–391) | 282 (169–336) | 0.02 |
| Aspartate transaminase (U/L) | 23 (19–29) | 22 (18–30) | 25 (19–32) | 24 (22–34) | 0.08 |
| Alanine transaminase (U/L) | 17 (13–25) | 15 (10–22) | 16 (11–24) | 14 (11–25) | 0.1 |
| γ-glutamyl transferase (U/L) | 23 (17–48) | 26 (17–44) | 44 (27–81) | 34 (23–55) | 0.0004 |
| Alkaline phosphatase (U/mL) | 228 (186–295) | 239 (197–317) | 257 (208–332) | 301 (199–420) | 0.2 |
| Total-bilirubin (mg/dL) | 0.6 (0.5–0.8) | 0.7 (0.5–0.9) | 0.8 (0.6–1.2) | 0.8 (0.6–1.2) | 0.0001 |
Values are mean ± SD, median (interquartile range), or n (%).
ACEIs/ARBs, angiotensin-converting enzyme inhibitors/angiotensin-receptor blockers; AF, atrial fibrillation; BNP, B-type natriuretic peptide; MRA, mineralocorticoid receptor antagonists; VCW, vena contracta width.
Patients demographics according to TR vena contracta width categories
| VCW ≤1 mm (n = 75) | VCW 1–3 mm (n = 163) | VCW 3–7 mm (n = 60) | VCW ≥7 mm (n = 13) | P value | |
|---|---|---|---|---|---|
| Age (years) | 68 ± 14 | 74 ± 12 | 77 ± 9 | 81 ± 10 | <0.0001 |
| Female, n (%) | 38 (51%) | 81 (50%) | 29 (48%) | 6 (46%) | 1.0 |
| Body mass index (kg/m2) | 24 ± 5 | 23 ± 4 | 22 ± 3 | 23 ± 4 | 0.08 |
| Vital signs | |||||
| Systolic blood pressure (mmHg) | 128 ± 23 | 128 ± 20 | 123 ± 18 | 122 ± 20 | 0.2 |
| Diastolic blood pressure (mmHg) | 70 ± 15 | 66 ± 14 | 66 ± 14 | 66 ± 10 | 0.2 |
| Heart rate (bpm) | 76 ± 18 | 72 ± 17 | 73 ± 16 | 73 ± 17 | 0.3 |
| Comorbidities | |||||
| Hypertension, n (%) | 58 (77%) | 134 (83%) | 46 (77%) | 11 (85%) | 0.6 |
| Coronary artery disease, n (%) | 21 (28%) | 42 (26%) | 8 (13%) | 1 (8%) | 0.09 |
| Earlier AF/current AF, n (%) | 8 (11%)/10 (13%) | 37 (23%)/46 (28%) | 15 (25%)/38 (63%) | 6 (46%)/7 (54%) | <0.0001 |
| Diabetes mellitus, n (%) | 30 (40%) | 55 (34%) | 18 (30%) | 2 (15%) | 0.3 |
| Cardiac implantable electrical devices, n (%) | 2 (3%) | 9 (6%) | 8 (13%) | 6 (46%) | <0.0001 |
| Medications | |||||
| ACEI or ARB, n (%) | 40 (53%) | 81 (50%) | 30 (51%) | 5 (38%) | 0.8 |
| Beta-blocker, n (%) | 28 (37%) | 68 (42%) | 31 (52%) | 3 (23%) | 0.2 |
| Diuretic, n (%) | 46 (61%) | 108 (66%) | 45 (75%) | 11 (85%) | 0.2 |
| MRA, n (%) | 25 (33%) | 59 (36%) | 24 (40%) | 5 (38%) | 0.9 |
| Laboratories | |||||
| Haemoglobin (g/dL) | 11.9 ± 2.4 | 11.5 ± 2.1 | 11.6 ± 2.5 | 10.8 ± 2.2 | 0.3 |
| Creatinine (mg/dL) | 0.9 (0.7–1.4) | 0.9 (0.7–1.2) | 1.0 (0.7–1.4) | 1.0 (0.7–1.8) | 0.8 |
| BNP (pg/mL) | 123 (47–369) | 187 (104–369) | 242 (174–391) | 282 (169–336) | 0.02 |
| Aspartate transaminase (U/L) | 23 (19–29) | 22 (18–30) | 25 (19–32) | 24 (22–34) | 0.08 |
| Alanine transaminase (U/L) | 17 (13–25) | 15 (10–22) | 16 (11–24) | 14 (11–25) | 0.1 |
| γ-glutamyl transferase (U/L) | 23 (17–48) | 26 (17–44) | 44 (27–81) | 34 (23–55) | 0.0004 |
| Alkaline phosphatase (U/mL) | 228 (186–295) | 239 (197–317) | 257 (208–332) | 301 (199–420) | 0.2 |
| Total-bilirubin (mg/dL) | 0.6 (0.5–0.8) | 0.7 (0.5–0.9) | 0.8 (0.6–1.2) | 0.8 (0.6–1.2) | 0.0001 |
| VCW ≤1 mm (n = 75) | VCW 1–3 mm (n = 163) | VCW 3–7 mm (n = 60) | VCW ≥7 mm (n = 13) | P value | |
|---|---|---|---|---|---|
| Age (years) | 68 ± 14 | 74 ± 12 | 77 ± 9 | 81 ± 10 | <0.0001 |
| Female, n (%) | 38 (51%) | 81 (50%) | 29 (48%) | 6 (46%) | 1.0 |
| Body mass index (kg/m2) | 24 ± 5 | 23 ± 4 | 22 ± 3 | 23 ± 4 | 0.08 |
| Vital signs | |||||
| Systolic blood pressure (mmHg) | 128 ± 23 | 128 ± 20 | 123 ± 18 | 122 ± 20 | 0.2 |
| Diastolic blood pressure (mmHg) | 70 ± 15 | 66 ± 14 | 66 ± 14 | 66 ± 10 | 0.2 |
| Heart rate (bpm) | 76 ± 18 | 72 ± 17 | 73 ± 16 | 73 ± 17 | 0.3 |
| Comorbidities | |||||
| Hypertension, n (%) | 58 (77%) | 134 (83%) | 46 (77%) | 11 (85%) | 0.6 |
| Coronary artery disease, n (%) | 21 (28%) | 42 (26%) | 8 (13%) | 1 (8%) | 0.09 |
| Earlier AF/current AF, n (%) | 8 (11%)/10 (13%) | 37 (23%)/46 (28%) | 15 (25%)/38 (63%) | 6 (46%)/7 (54%) | <0.0001 |
| Diabetes mellitus, n (%) | 30 (40%) | 55 (34%) | 18 (30%) | 2 (15%) | 0.3 |
| Cardiac implantable electrical devices, n (%) | 2 (3%) | 9 (6%) | 8 (13%) | 6 (46%) | <0.0001 |
| Medications | |||||
| ACEI or ARB, n (%) | 40 (53%) | 81 (50%) | 30 (51%) | 5 (38%) | 0.8 |
| Beta-blocker, n (%) | 28 (37%) | 68 (42%) | 31 (52%) | 3 (23%) | 0.2 |
| Diuretic, n (%) | 46 (61%) | 108 (66%) | 45 (75%) | 11 (85%) | 0.2 |
| MRA, n (%) | 25 (33%) | 59 (36%) | 24 (40%) | 5 (38%) | 0.9 |
| Laboratories | |||||
| Haemoglobin (g/dL) | 11.9 ± 2.4 | 11.5 ± 2.1 | 11.6 ± 2.5 | 10.8 ± 2.2 | 0.3 |
| Creatinine (mg/dL) | 0.9 (0.7–1.4) | 0.9 (0.7–1.2) | 1.0 (0.7–1.4) | 1.0 (0.7–1.8) | 0.8 |
| BNP (pg/mL) | 123 (47–369) | 187 (104–369) | 242 (174–391) | 282 (169–336) | 0.02 |
| Aspartate transaminase (U/L) | 23 (19–29) | 22 (18–30) | 25 (19–32) | 24 (22–34) | 0.08 |
| Alanine transaminase (U/L) | 17 (13–25) | 15 (10–22) | 16 (11–24) | 14 (11–25) | 0.1 |
| γ-glutamyl transferase (U/L) | 23 (17–48) | 26 (17–44) | 44 (27–81) | 34 (23–55) | 0.0004 |
| Alkaline phosphatase (U/mL) | 228 (186–295) | 239 (197–317) | 257 (208–332) | 301 (199–420) | 0.2 |
| Total-bilirubin (mg/dL) | 0.6 (0.5–0.8) | 0.7 (0.5–0.9) | 0.8 (0.6–1.2) | 0.8 (0.6–1.2) | 0.0001 |
Values are mean ± SD, median (interquartile range), or n (%).
ACEIs/ARBs, angiotensin-converting enzyme inhibitors/angiotensin-receptor blockers; AF, atrial fibrillation; BNP, B-type natriuretic peptide; MRA, mineralocorticoid receptor antagonists; VCW, vena contracta width.
Left and right heart remodelling and dysfunction according to VCW
LV volume, mass, and cardiac output were similar among the groups whereas increasing TR severity was associated with decreased mitrals’ tissue velocity (Table 2). With increasing VCW, there was a progressive increase in the prevalence of significant MR. LA volume index and mitral E-wave both increased with increasing VCW severity. E/e′ ratio was similar among the groups.
Cardiac structure and function according to TR vena contracta width categories
| VCW ≤1 mm (n = 75) | VCW 1–3 mm (n = 163) | VCW 3–7 mm (n = 60) | VCW ≥7 mm (n = 13) | P value | |
|---|---|---|---|---|---|
| Left heart | |||||
| LV end-diastolic volume (mL) | 87 ± 36 | 93 ± 35 | 85 ± 35 | 85 ± 40 | 0.4 |
| LV mass index (g/m2) | 100 ± 30 | 109 ± 32 | 105 ± 34 | 106 ± 26 | 0.2 |
| LV ejection fraction (%) | 60 ± 7 | 61 ± 7 | 62 ± 6 | 61 ± 4 | 0.5 |
| Stroke volume (mL) | 56 ± 19 | 59 ± 19 | 55 ± 21 | 50 ± 15 | 0.2 |
| Cardiac output (L/min) | 4.1 ± 1.4 | 4.2 ± 1.6 | 3.9 ± 1.4 | 3.6 ± 1.1 | 0.4 |
| Mitral E (cm/s) | 75 ± 26 | 84 ± 28 | 91 ± 27 | 102 ± 21 | 0.0007 |
| Mitral e′ (cm/s) | 5 ± 2 | 6 ± 2 | 6 ± 3 | 7 ± 2 | 0.0004 |
| Mitral s′ (cm/s) | 6 ± 1 | 6 ± 2 | 5 ± 1 | 5 ± 1 | 0.0002 |
| E/e′ ratio | 15 ± 6 | 16 ± 7 | 16 ± 6 | 15 ± 5 | 0.6 |
| LA volume index (mL/m2) | 39 ± 17 | 55 ± 33 | 76 ± 48 | 75 ± 24 | <0.0001 |
Mitral regurgitation (%) Absent/ mild/ moderate | 72%/24%/4% | 42%/53%/5% | 23%/65%/12% | 15%/54%/31% | <0.0001 |
| Right heart | |||||
| eRVSP (mmHg) | 23 ± 6 | 31 ± 10 | 39 ± 11 | 43 ± 9 | <0.0001 |
| TAPSEa (mm) | 18 ± 5 | 18 ± 5 | 16 ± 5 | 14 ± 5 | 0.01 |
| RV basal diameter (mm) | 32 ± 7 | 35 ± 7 | 40 ± 8 | 46 ± 8 | <0.0001 |
| RV mid-diameter (mm) | 26 ± 6 | 28 ± 6 | 32 ± 7 | 36 ± 7 | <0.0001 |
| TV annular diameter (mm) | 22 ± 5 | 25 ± 5 | 31 ± 6 | 36 ± 5 | <0.0001 |
| TR vena contracta width (mm) | 0.4 ± 0.4 | 1.8 ± 0.5 | 4.3 ± 0.9 | 9.4 ± 2.1 | <0.0001 |
| TR jet area (cm2) | 0 ± 0 | 1 ± 1 | 5 ± 2 | 14 ± 7 | <0.0001 |
| Estimated right atrial pressure (mmHg) | 4 ± 2 | 5 ± 3 | 5 ± 3 | 11 ± 5 | <0.0001 |
| Right atrial max area (cm2) | 13 ± 4 | 16 ± 6 | 22 ± 8 | 38 ± 13 | <0.0001 |
| Right atrial max volume (mL) | 30 ± 15 | 42 ± 25 | 70 ± 40 | 163 ± 82 | <0.0001 |
| Right atrial expansion index (%) | 61 (38–87) | 50 (25–87) | 31 (15–48) | 16 (9–24) | <0.0001 |
| VCW ≤1 mm (n = 75) | VCW 1–3 mm (n = 163) | VCW 3–7 mm (n = 60) | VCW ≥7 mm (n = 13) | P value | |
|---|---|---|---|---|---|
| Left heart | |||||
| LV end-diastolic volume (mL) | 87 ± 36 | 93 ± 35 | 85 ± 35 | 85 ± 40 | 0.4 |
| LV mass index (g/m2) | 100 ± 30 | 109 ± 32 | 105 ± 34 | 106 ± 26 | 0.2 |
| LV ejection fraction (%) | 60 ± 7 | 61 ± 7 | 62 ± 6 | 61 ± 4 | 0.5 |
| Stroke volume (mL) | 56 ± 19 | 59 ± 19 | 55 ± 21 | 50 ± 15 | 0.2 |
| Cardiac output (L/min) | 4.1 ± 1.4 | 4.2 ± 1.6 | 3.9 ± 1.4 | 3.6 ± 1.1 | 0.4 |
| Mitral E (cm/s) | 75 ± 26 | 84 ± 28 | 91 ± 27 | 102 ± 21 | 0.0007 |
| Mitral e′ (cm/s) | 5 ± 2 | 6 ± 2 | 6 ± 3 | 7 ± 2 | 0.0004 |
| Mitral s′ (cm/s) | 6 ± 1 | 6 ± 2 | 5 ± 1 | 5 ± 1 | 0.0002 |
| E/e′ ratio | 15 ± 6 | 16 ± 7 | 16 ± 6 | 15 ± 5 | 0.6 |
| LA volume index (mL/m2) | 39 ± 17 | 55 ± 33 | 76 ± 48 | 75 ± 24 | <0.0001 |
Mitral regurgitation (%) Absent/ mild/ moderate | 72%/24%/4% | 42%/53%/5% | 23%/65%/12% | 15%/54%/31% | <0.0001 |
| Right heart | |||||
| eRVSP (mmHg) | 23 ± 6 | 31 ± 10 | 39 ± 11 | 43 ± 9 | <0.0001 |
| TAPSEa (mm) | 18 ± 5 | 18 ± 5 | 16 ± 5 | 14 ± 5 | 0.01 |
| RV basal diameter (mm) | 32 ± 7 | 35 ± 7 | 40 ± 8 | 46 ± 8 | <0.0001 |
| RV mid-diameter (mm) | 26 ± 6 | 28 ± 6 | 32 ± 7 | 36 ± 7 | <0.0001 |
| TV annular diameter (mm) | 22 ± 5 | 25 ± 5 | 31 ± 6 | 36 ± 5 | <0.0001 |
| TR vena contracta width (mm) | 0.4 ± 0.4 | 1.8 ± 0.5 | 4.3 ± 0.9 | 9.4 ± 2.1 | <0.0001 |
| TR jet area (cm2) | 0 ± 0 | 1 ± 1 | 5 ± 2 | 14 ± 7 | <0.0001 |
| Estimated right atrial pressure (mmHg) | 4 ± 2 | 5 ± 3 | 5 ± 3 | 11 ± 5 | <0.0001 |
| Right atrial max area (cm2) | 13 ± 4 | 16 ± 6 | 22 ± 8 | 38 ± 13 | <0.0001 |
| Right atrial max volume (mL) | 30 ± 15 | 42 ± 25 | 70 ± 40 | 163 ± 82 | <0.0001 |
| Right atrial expansion index (%) | 61 (38–87) | 50 (25–87) | 31 (15–48) | 16 (9–24) | <0.0001 |
Values are mean ± SD, median (interquartile range), or n (%).
Available in 212 patients.
eRVSP, estimated right ventricular systolic pressure; LA, left atrial; LV, left ventricular; RV, right ventricular; TAPSE, tricuspid annular plane systolic excursion; TR, tricuspid regurgitation; VCW, vena contracta width.
Cardiac structure and function according to TR vena contracta width categories
| VCW ≤1 mm (n = 75) | VCW 1–3 mm (n = 163) | VCW 3–7 mm (n = 60) | VCW ≥7 mm (n = 13) | P value | |
|---|---|---|---|---|---|
| Left heart | |||||
| LV end-diastolic volume (mL) | 87 ± 36 | 93 ± 35 | 85 ± 35 | 85 ± 40 | 0.4 |
| LV mass index (g/m2) | 100 ± 30 | 109 ± 32 | 105 ± 34 | 106 ± 26 | 0.2 |
| LV ejection fraction (%) | 60 ± 7 | 61 ± 7 | 62 ± 6 | 61 ± 4 | 0.5 |
| Stroke volume (mL) | 56 ± 19 | 59 ± 19 | 55 ± 21 | 50 ± 15 | 0.2 |
| Cardiac output (L/min) | 4.1 ± 1.4 | 4.2 ± 1.6 | 3.9 ± 1.4 | 3.6 ± 1.1 | 0.4 |
| Mitral E (cm/s) | 75 ± 26 | 84 ± 28 | 91 ± 27 | 102 ± 21 | 0.0007 |
| Mitral e′ (cm/s) | 5 ± 2 | 6 ± 2 | 6 ± 3 | 7 ± 2 | 0.0004 |
| Mitral s′ (cm/s) | 6 ± 1 | 6 ± 2 | 5 ± 1 | 5 ± 1 | 0.0002 |
| E/e′ ratio | 15 ± 6 | 16 ± 7 | 16 ± 6 | 15 ± 5 | 0.6 |
| LA volume index (mL/m2) | 39 ± 17 | 55 ± 33 | 76 ± 48 | 75 ± 24 | <0.0001 |
Mitral regurgitation (%) Absent/ mild/ moderate | 72%/24%/4% | 42%/53%/5% | 23%/65%/12% | 15%/54%/31% | <0.0001 |
| Right heart | |||||
| eRVSP (mmHg) | 23 ± 6 | 31 ± 10 | 39 ± 11 | 43 ± 9 | <0.0001 |
| TAPSEa (mm) | 18 ± 5 | 18 ± 5 | 16 ± 5 | 14 ± 5 | 0.01 |
| RV basal diameter (mm) | 32 ± 7 | 35 ± 7 | 40 ± 8 | 46 ± 8 | <0.0001 |
| RV mid-diameter (mm) | 26 ± 6 | 28 ± 6 | 32 ± 7 | 36 ± 7 | <0.0001 |
| TV annular diameter (mm) | 22 ± 5 | 25 ± 5 | 31 ± 6 | 36 ± 5 | <0.0001 |
| TR vena contracta width (mm) | 0.4 ± 0.4 | 1.8 ± 0.5 | 4.3 ± 0.9 | 9.4 ± 2.1 | <0.0001 |
| TR jet area (cm2) | 0 ± 0 | 1 ± 1 | 5 ± 2 | 14 ± 7 | <0.0001 |
| Estimated right atrial pressure (mmHg) | 4 ± 2 | 5 ± 3 | 5 ± 3 | 11 ± 5 | <0.0001 |
| Right atrial max area (cm2) | 13 ± 4 | 16 ± 6 | 22 ± 8 | 38 ± 13 | <0.0001 |
| Right atrial max volume (mL) | 30 ± 15 | 42 ± 25 | 70 ± 40 | 163 ± 82 | <0.0001 |
| Right atrial expansion index (%) | 61 (38–87) | 50 (25–87) | 31 (15–48) | 16 (9–24) | <0.0001 |
| VCW ≤1 mm (n = 75) | VCW 1–3 mm (n = 163) | VCW 3–7 mm (n = 60) | VCW ≥7 mm (n = 13) | P value | |
|---|---|---|---|---|---|
| Left heart | |||||
| LV end-diastolic volume (mL) | 87 ± 36 | 93 ± 35 | 85 ± 35 | 85 ± 40 | 0.4 |
| LV mass index (g/m2) | 100 ± 30 | 109 ± 32 | 105 ± 34 | 106 ± 26 | 0.2 |
| LV ejection fraction (%) | 60 ± 7 | 61 ± 7 | 62 ± 6 | 61 ± 4 | 0.5 |
| Stroke volume (mL) | 56 ± 19 | 59 ± 19 | 55 ± 21 | 50 ± 15 | 0.2 |
| Cardiac output (L/min) | 4.1 ± 1.4 | 4.2 ± 1.6 | 3.9 ± 1.4 | 3.6 ± 1.1 | 0.4 |
| Mitral E (cm/s) | 75 ± 26 | 84 ± 28 | 91 ± 27 | 102 ± 21 | 0.0007 |
| Mitral e′ (cm/s) | 5 ± 2 | 6 ± 2 | 6 ± 3 | 7 ± 2 | 0.0004 |
| Mitral s′ (cm/s) | 6 ± 1 | 6 ± 2 | 5 ± 1 | 5 ± 1 | 0.0002 |
| E/e′ ratio | 15 ± 6 | 16 ± 7 | 16 ± 6 | 15 ± 5 | 0.6 |
| LA volume index (mL/m2) | 39 ± 17 | 55 ± 33 | 76 ± 48 | 75 ± 24 | <0.0001 |
Mitral regurgitation (%) Absent/ mild/ moderate | 72%/24%/4% | 42%/53%/5% | 23%/65%/12% | 15%/54%/31% | <0.0001 |
| Right heart | |||||
| eRVSP (mmHg) | 23 ± 6 | 31 ± 10 | 39 ± 11 | 43 ± 9 | <0.0001 |
| TAPSEa (mm) | 18 ± 5 | 18 ± 5 | 16 ± 5 | 14 ± 5 | 0.01 |
| RV basal diameter (mm) | 32 ± 7 | 35 ± 7 | 40 ± 8 | 46 ± 8 | <0.0001 |
| RV mid-diameter (mm) | 26 ± 6 | 28 ± 6 | 32 ± 7 | 36 ± 7 | <0.0001 |
| TV annular diameter (mm) | 22 ± 5 | 25 ± 5 | 31 ± 6 | 36 ± 5 | <0.0001 |
| TR vena contracta width (mm) | 0.4 ± 0.4 | 1.8 ± 0.5 | 4.3 ± 0.9 | 9.4 ± 2.1 | <0.0001 |
| TR jet area (cm2) | 0 ± 0 | 1 ± 1 | 5 ± 2 | 14 ± 7 | <0.0001 |
| Estimated right atrial pressure (mmHg) | 4 ± 2 | 5 ± 3 | 5 ± 3 | 11 ± 5 | <0.0001 |
| Right atrial max area (cm2) | 13 ± 4 | 16 ± 6 | 22 ± 8 | 38 ± 13 | <0.0001 |
| Right atrial max volume (mL) | 30 ± 15 | 42 ± 25 | 70 ± 40 | 163 ± 82 | <0.0001 |
| Right atrial expansion index (%) | 61 (38–87) | 50 (25–87) | 31 (15–48) | 16 (9–24) | <0.0001 |
Values are mean ± SD, median (interquartile range), or n (%).
Available in 212 patients.
eRVSP, estimated right ventricular systolic pressure; LA, left atrial; LV, left ventricular; RV, right ventricular; TAPSE, tricuspid annular plane systolic excursion; TR, tricuspid regurgitation; VCW, vena contracta width.
In contrast to the modest impacts on the LV, TR severity was associated with remarkable abnormalities in RV structure and function, including higher eRVSP, larger RV and TV annular diameters, and lower TAPSE (Table 2). Particularly, increasing VCW was correlated with larger RV basal and mid-diameters (r = 0.44 and r = 0.38, both P < 0.0001) and higher eRVSP (r = 0.65, P < 0.0001). The TR severity was also associated with RA remodelling and dysfunction. RA volumes were progressively increased and RA reservoir function, as assessed by RA expansion index, was decreased with increasing VCW. Estimated RA pressure was increased as the VCW category progressed. Sensitivity analysis categorized patients according to the guideline-based integrated approach, which showed essentially similar results (Supplementary data online, Table S2).
Among the subgroup of patients undergoing invasive right heart catheterization within 2 weeks from index echocardiographic evaluation (n = 27), mean PCWP (13 ± 7 mmHg), PA pressures (PA systolic pressure 33 ± 11 mmHg and PA mean pressure 22 ± 8 mmHg), and cardiac index (2.7 ± 1.1 L/min/m2) were within normal range while RAP was mildly elevated (7 ± 5 mmHg). Sensitivity analysis performed separately among patients with and without invasive assessment showed similar baseline characteristics, suggesting that the presence of right heart catheterization did not significantly influence the results (Supplementary data online, Table S3).
TR severity and clinical outcomes
Over a median follow-up time of 2.1 (interquartile range 0.9–3.4) years, there were 101 (32%) primary end points (37 all-cause deaths and 64 HF hospitalizations). Kaplan–Meier curves show that event-free survival varied based on the semiquantitative VCW categories in which even the mild category (VCW 1–3 mm) was associated with higher event rates when compared with the lowest VCW category (Figure 1A). In an unadjusted Cox proportional hazards model (Table 3), event risk was more than two-fold in the moderate category (VCW 3–7 mm) and more than four-fold in the severe category (VCW ≥7 mm) when compared with the lowest VCW category. It should be noted that the risk was nearly doubled, even in the mild category (VCW 1–3 mm). Measures of RV remodelling and dysfunction as well as PH were all associated with the composite outcome of all-cause mortality or HF hospitalization [RV basal diameter per 1 mm, hazard ratio (HR) 1.03, 95% confidence interval (CI) 1.01–1.06, P = 0.005; TAPSE per 1 mm, HR 0.95, 95% CI 0.91–0.99, P = 0.02; ln eRVSP per 1 unit, HR 4.48, 95% CI 2.52–8.00, P < 0.0001].
(A) Compared with heart failure with preserved ejection fraction (HFpEF) patients in the lowest vena contracta width (VCW) category (≤1 mm), patients in the mild TR category (VCW 1–3 mm) had an increased risk of the composite outcome of all-cause mortality or heart failure hospitalization (P = 0.039 vs. ≤1 mm). (B) TR severity defined by the guideline-based qualitative approach showed similar graded associations with the outcome.
Univariable and multivariable cox proportional hazard for the association with the composite outcome
| Unadjusted | Model 1 | Model 2 (Model 1 + RV diameter) | Model 3 (Model 2 + eRVSP) | |||||
|---|---|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | |
| TR grading by VCW categories | ||||||||
| VCW ≤1 mm | Ref | – | Ref | – | Ref | – | Ref | – |
| VCW 1–3 mm | 1.81 (1.02–3.21) | 0.04 | 2.05 (1.09–3.88) | 0.03 | 1.99 (1.05–3.77) | 0.04 | 1.45 (0.74–2.85) | 0.3 |
| VCW 37 mm | 2.34 (1.23–4.44) | 0.009 | 2.81 (1.26–6.25) | 0.01 | 2.63 (1.16–5.95) | 0.02 | 1.73 (0.74–4.05) | 0.2 |
| VCW ≥7 mm | 4.22 (1.79–9.96) | 0.001 | 5.73 (1.97–16.6) | 0.001 | 5.00 (1.60–15.7) | 0.006 | 2.84 (0.87–9.33) | 0.08 |
| Continuous measures of TR severity | ||||||||
| VCW, per 1 mm | 1.17 (1.08–1.26) | <0.0001 | 1.21 (1.09-1.33) | 0.0001 | 1.20 (1.07-1.32) | 0.0006 | 1.14 (1.02-1.27) | 0.02 |
| Ln TR jet area | 1.64 (1.29–2.07) | <0.0001 | 2.03 (1.45-2.86) | <0.0001 | 2.00 (1.38-2.89) | 0.0002 | 1.65 (1.12-2.43) | 0.01 |
| TR grading by an integrated approach | ||||||||
| TR none/trivial | Ref | – | Ref | – | Ref | – | Ref | – |
| TR mild | 1.59 (0.99–2.57) | 0.05 | 1.77 (1.01–3.12) | 0.046 | 1.73 (0.96–3.10) | 0.07 | 1.16 (0.61–2.21) | 0.7 |
| TR moderate | 1.99 (1.13–3.52) | 0.02 | 2.49 (1.20–5.18) | 0.01 | 2.41 (1.12–5.15) | 0.02 | 1.58 (0.71–3.51) | 0.3 |
| TR severe | 3.92 (1.95–7.86) | 0.0001 | 4.55 (1.83–11.3) | 0.001 | 4.24 (1.53–11.7) | 0.005 | 2.17 (0.74–6.41) | 0.2 |
| Unadjusted | Model 1 | Model 2 (Model 1 + RV diameter) | Model 3 (Model 2 + eRVSP) | |||||
|---|---|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | |
| TR grading by VCW categories | ||||||||
| VCW ≤1 mm | Ref | – | Ref | – | Ref | – | Ref | – |
| VCW 1–3 mm | 1.81 (1.02–3.21) | 0.04 | 2.05 (1.09–3.88) | 0.03 | 1.99 (1.05–3.77) | 0.04 | 1.45 (0.74–2.85) | 0.3 |
| VCW 37 mm | 2.34 (1.23–4.44) | 0.009 | 2.81 (1.26–6.25) | 0.01 | 2.63 (1.16–5.95) | 0.02 | 1.73 (0.74–4.05) | 0.2 |
| VCW ≥7 mm | 4.22 (1.79–9.96) | 0.001 | 5.73 (1.97–16.6) | 0.001 | 5.00 (1.60–15.7) | 0.006 | 2.84 (0.87–9.33) | 0.08 |
| Continuous measures of TR severity | ||||||||
| VCW, per 1 mm | 1.17 (1.08–1.26) | <0.0001 | 1.21 (1.09-1.33) | 0.0001 | 1.20 (1.07-1.32) | 0.0006 | 1.14 (1.02-1.27) | 0.02 |
| Ln TR jet area | 1.64 (1.29–2.07) | <0.0001 | 2.03 (1.45-2.86) | <0.0001 | 2.00 (1.38-2.89) | 0.0002 | 1.65 (1.12-2.43) | 0.01 |
| TR grading by an integrated approach | ||||||||
| TR none/trivial | Ref | – | Ref | – | Ref | – | Ref | – |
| TR mild | 1.59 (0.99–2.57) | 0.05 | 1.77 (1.01–3.12) | 0.046 | 1.73 (0.96–3.10) | 0.07 | 1.16 (0.61–2.21) | 0.7 |
| TR moderate | 1.99 (1.13–3.52) | 0.02 | 2.49 (1.20–5.18) | 0.01 | 2.41 (1.12–5.15) | 0.02 | 1.58 (0.71–3.51) | 0.3 |
| TR severe | 3.92 (1.95–7.86) | 0.0001 | 4.55 (1.83–11.3) | 0.001 | 4.24 (1.53–11.7) | 0.005 | 2.17 (0.74–6.41) | 0.2 |
Model 1 = age, sex, Ln LA volume index, AF, MR severity, Ln BNP, and intra-cardiac device; Model 2 = model 1 + RV basal diameter; Model 3 = model 2 + Ln eRVSP.
AF, atrial fibrillation; BNP, B-type natriuretic peptide; CI, confidence interval; eRVSP, estimated right ventricular systolic pressure; HR, hazard ratio; RV, right ventricular; TR, tricuspid regurgitation; VCW, vena contracta width.
Univariable and multivariable cox proportional hazard for the association with the composite outcome
| Unadjusted | Model 1 | Model 2 (Model 1 + RV diameter) | Model 3 (Model 2 + eRVSP) | |||||
|---|---|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | |
| TR grading by VCW categories | ||||||||
| VCW ≤1 mm | Ref | – | Ref | – | Ref | – | Ref | – |
| VCW 1–3 mm | 1.81 (1.02–3.21) | 0.04 | 2.05 (1.09–3.88) | 0.03 | 1.99 (1.05–3.77) | 0.04 | 1.45 (0.74–2.85) | 0.3 |
| VCW 37 mm | 2.34 (1.23–4.44) | 0.009 | 2.81 (1.26–6.25) | 0.01 | 2.63 (1.16–5.95) | 0.02 | 1.73 (0.74–4.05) | 0.2 |
| VCW ≥7 mm | 4.22 (1.79–9.96) | 0.001 | 5.73 (1.97–16.6) | 0.001 | 5.00 (1.60–15.7) | 0.006 | 2.84 (0.87–9.33) | 0.08 |
| Continuous measures of TR severity | ||||||||
| VCW, per 1 mm | 1.17 (1.08–1.26) | <0.0001 | 1.21 (1.09-1.33) | 0.0001 | 1.20 (1.07-1.32) | 0.0006 | 1.14 (1.02-1.27) | 0.02 |
| Ln TR jet area | 1.64 (1.29–2.07) | <0.0001 | 2.03 (1.45-2.86) | <0.0001 | 2.00 (1.38-2.89) | 0.0002 | 1.65 (1.12-2.43) | 0.01 |
| TR grading by an integrated approach | ||||||||
| TR none/trivial | Ref | – | Ref | – | Ref | – | Ref | – |
| TR mild | 1.59 (0.99–2.57) | 0.05 | 1.77 (1.01–3.12) | 0.046 | 1.73 (0.96–3.10) | 0.07 | 1.16 (0.61–2.21) | 0.7 |
| TR moderate | 1.99 (1.13–3.52) | 0.02 | 2.49 (1.20–5.18) | 0.01 | 2.41 (1.12–5.15) | 0.02 | 1.58 (0.71–3.51) | 0.3 |
| TR severe | 3.92 (1.95–7.86) | 0.0001 | 4.55 (1.83–11.3) | 0.001 | 4.24 (1.53–11.7) | 0.005 | 2.17 (0.74–6.41) | 0.2 |
| Unadjusted | Model 1 | Model 2 (Model 1 + RV diameter) | Model 3 (Model 2 + eRVSP) | |||||
|---|---|---|---|---|---|---|---|---|
| HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | HR (95% CI) | P value | |
| TR grading by VCW categories | ||||||||
| VCW ≤1 mm | Ref | – | Ref | – | Ref | – | Ref | – |
| VCW 1–3 mm | 1.81 (1.02–3.21) | 0.04 | 2.05 (1.09–3.88) | 0.03 | 1.99 (1.05–3.77) | 0.04 | 1.45 (0.74–2.85) | 0.3 |
| VCW 37 mm | 2.34 (1.23–4.44) | 0.009 | 2.81 (1.26–6.25) | 0.01 | 2.63 (1.16–5.95) | 0.02 | 1.73 (0.74–4.05) | 0.2 |
| VCW ≥7 mm | 4.22 (1.79–9.96) | 0.001 | 5.73 (1.97–16.6) | 0.001 | 5.00 (1.60–15.7) | 0.006 | 2.84 (0.87–9.33) | 0.08 |
| Continuous measures of TR severity | ||||||||
| VCW, per 1 mm | 1.17 (1.08–1.26) | <0.0001 | 1.21 (1.09-1.33) | 0.0001 | 1.20 (1.07-1.32) | 0.0006 | 1.14 (1.02-1.27) | 0.02 |
| Ln TR jet area | 1.64 (1.29–2.07) | <0.0001 | 2.03 (1.45-2.86) | <0.0001 | 2.00 (1.38-2.89) | 0.0002 | 1.65 (1.12-2.43) | 0.01 |
| TR grading by an integrated approach | ||||||||
| TR none/trivial | Ref | – | Ref | – | Ref | – | Ref | – |
| TR mild | 1.59 (0.99–2.57) | 0.05 | 1.77 (1.01–3.12) | 0.046 | 1.73 (0.96–3.10) | 0.07 | 1.16 (0.61–2.21) | 0.7 |
| TR moderate | 1.99 (1.13–3.52) | 0.02 | 2.49 (1.20–5.18) | 0.01 | 2.41 (1.12–5.15) | 0.02 | 1.58 (0.71–3.51) | 0.3 |
| TR severe | 3.92 (1.95–7.86) | 0.0001 | 4.55 (1.83–11.3) | 0.001 | 4.24 (1.53–11.7) | 0.005 | 2.17 (0.74–6.41) | 0.2 |
Model 1 = age, sex, Ln LA volume index, AF, MR severity, Ln BNP, and intra-cardiac device; Model 2 = model 1 + RV basal diameter; Model 3 = model 2 + Ln eRVSP.
AF, atrial fibrillation; BNP, B-type natriuretic peptide; CI, confidence interval; eRVSP, estimated right ventricular systolic pressure; HR, hazard ratio; RV, right ventricular; TR, tricuspid regurgitation; VCW, vena contracta width.
After adjusting for age, gender, LA volume index, AF, MR severity, BNP, and the presence of CIEDs (Model 1), an increasing VCW category was associated with a heightened risk for a composite outcome. The graded association of the VCW category with the outcome remained significant after adjusting for a measure of RV enlargement (Model 2), without attenuation of point estimates. However, the association failed to remain significant after further adjustment for pulmonary pressures (Model 3).
Similar graded associations between TR severity and the composite outcome were observed in both univariable and multivariable models using the guideline-based integrated multiparametric approach (Table 3), with the exception of a marginal association between mild TR and the outcome in Model 2. As with the VCW category, the severity of TR was no longer significantly associated with the outcome in the Model 3 which included eRVSP.
We found that the association between VCW and the outcome was continuous, with a steeper increase in the outcome with a larger VCW (Figure 2A). A continuous measure of VCW was associated with the composite of all-cause mortality or HF hospitalization with a crude HR of 1.17 (95% CI 1.08–1.26, P < 0.0001) per 1 mm increment. Greater VCW remained significantly associated with the outcome in multivariable Models 1 and 2. Of note, a continuous measure of VCW was predictive of the outcome even after further adjusting for eRVSP (Model 3).
There were continuous associations of semiquantitative metrics of TR severity with the outcome, with a steeper increase in hazards ratios with a larger VCW and jet area. Abbreviations as in Figure 1.
Similar findings were observed in analyses using TR jet area, with a continuous association between jet area and the outcome (Figure 2B). In multivariable Models 1–2, a continuous measure of jet area was also associated with a heightened risk for the composite outcome (Table 3) and remained significantly associated after additional adjustment for eRVSP.
In a sensitivity analysis excluding patients with CIED (n = 25), all key associations between the TR severity and outcomes remained significant, with the exception of a marginal association of continuous VCW with the outcomes in Model 3 (P = 0.05) (Supplementary data online, Figure S1;Supplementary data online, Table S4).
In the sequential Cox proportional hazards analysis (Figure 3), a model based on clinical and left heart factors (age, sex, LA volume index, AF, MR severity, and BNP: left heart model) was associated with adverse outcome in HFpEF (χ2 31.1, P < 0.001). Addition of RV basal diameter and eRVSP to this model significantly improved the predictive power (χ2 51.5 vs. 31.1, P < 0.001). The prognostic value was further improved by adding VCW as a continuous variable (HR 1.14 per 1 mm, 95% CI 1.02–1.27, P = 0.02; χ2 58.8 vs. 51.5, P = 0.03), whereas it was not improved by adding qualitative TR severity (P = 0.44). Similar incremental prognostic value was obtained by adding TR jet area (HR 1.68 per 1 unit, 95% CI 1.14–2.48, P = 0.009; χ2 58.5 vs. 51.5, P = 0.009).
Addition of right ventricular (RV) basal diameter and estimated RV systolic pressure (eRVSP) significantly increased the model based on age, sex, LA volume index, AF, MR severity, and BNP. The prognostic value was further improved by adding VCW to the previous model, whereas it was not improved by the addition of the qualitative TR severity.
Discussion
This is the first study to investigate the association of TR severity, as assessed by semiquantitative measures of TR severity (including VCW and jet area), with clinical outcomes in patients with HFpEF. The major findings were (i) there was a continuous association between the semiquantitative measures of TR and the relative risk for the outcome, with a steeper increase in the outcome with a larger VCW and jet area; (ii) the semiquantitative measures of TR were associated with the composite outcomes of all-cause mortality or HF hospitalization independent of age, sex, LA volume index, AF, MR severity, BNP, and the presence of CIEDs. Of note, the association remained significant after further adjusting for RV enlargement and pulmonary pressures; (iii) VCW provided an incremental prognostic value over the model based on clinical and left heart factors, right heart remodelling, and PH; and (iv) TR severity as defined by the guideline-based qualitative approach did not remain significant after adjusting for pulmonary pressures. The current data highlight the potential value of the semiquantitative measures of TR severity for the risk stratification in patients with HFpEF.
Previous studies
Increasing evidence suggests that clinically significant TR has a negative effect on clinical outcomes in various heart diseases, including aortic stenosis, PAH, and HFrEF.6–8 However, the prognostic significance of TR in patients with HFpEF remains uncertain.2,9,10,17 Previous studies have demonstrated that TR severity is associated with clinical outcomes in univariable analysis, but this association is no longer predictable after adjusting for RV dysfunction.2,9 In a recent analysis of HFpEF, there appeared to be a relationship between TR severity and impact on mortality.4 However, data regarding pulmonary pressure and RV size and function were not available and were not accounted for in the multivariable model. These data suggest that observed relationship between TR and outcomes in HFpEF might be attributed to concomitant RV remodelling and PH ,2,3,18 and emphasize the need for accounting for the two factors to elucidate the true impact of TR on clinical outcomes.
Current study
In the current study, we demonstrated for the first time that increased TR severity reflected by increased VCW was associated with adverse outcomes both independently and incrementally over the established prognostic markers, as well as RV remodelling and PH. One potential reason for the discrepant results between ours and other studies may be related to the assessment of TR severity.2,9,10 As opposed to the current study which assessed TR using semiquantitative measures, previous studies only assessed the severity of TR qualitatively. While the guideline-based qualitative approach comprehensively evaluates the severity and pathophysiology of TR, categorizing TR does not account for the individual risk within each TR category. In this regard, we demonstrated a continuous association of VCW and jet area with the composite outcome. This indicates that the qualitative approach is unlikely to allow for individual risk stratification of such a continuous relationship. In fact, we observed that TR grading assessed by either the guideline-based integrated grading approach or even VCW category was not associated with outcomes after further adjusting for eRVSP. In contrast, continuous metrics of VCW and TR jet area remained significant predictors after adjusting for right heart remodelling and severity of PH. These results highlight the prognostic value of the semiquantitative (continuous) measures of TR in patients with HFpEF and suggest the potential role for risk stratification in clinical practice as compared with qualitative assessments of TR severity. The semiquantitative approach may also impact on therapeutic decision making.
We also demonstrated a dose-dependent association between TR and outcomes (Figure 1A). Notably, even HFpEF patients with mild TR (VCW 1–3 mm) had a nearly two-fold relative risk compared with patients without TR. This finding is concordant with previous studies in HFrEF,17,19 and may suggest a potential for earlier intervention. In patients with mild TR, volume-reducing therapies might be effective, either through aggressive diuretics, mineral corticoid receptor antagonists, or as is currently being investigated, sodium-glucose co-transporter 2 inhibitors.20,21 The semiquantitative metrics of TR severity may be also used to define an indication or timing for TR treatment. Further study is required to identify optimal VCW cut-offs to initiate treatment.
Beyond the aforementioned prognostic value of semiquantitative measures of TR, the current data have potentially important clinical implications. Despite the high prevalence and its pathophysiologic significance in HFpEF, TR is usually left untreated in most patients.2,3,19,22 This may be related to a limited availability of treatment options and in part to the perception of TR as a bystander or even mere reflection of the disease progression in this syndrome. Our results suggest that TR itself drives poor outcomes in patients with HFpEF rather than being a mere surrogate for right heart remodelling and PH. The potential underlying mechanisms could be related to worsening right heart remodelling by inducing a volume overload (forming a vicious cycle) and systemic venous congestion contributing to renal and liver impairment.23 Indeed, we found that TR severity was directly correlated with hepatobiliary enzyme elevation, which reflects hepatic congestion and associates with outcomes in patients HFpEF.24 The association between TR and the outcomes could also be mediated by reducing cardiac output.22
Therapeutic implication
The current study suggests that TR may be a potential therapeutic target in patients with HFpEF, in which there is no proven treatment. Treating TR may break the vicious cycle between TR and right heart remodelling, mitigate systemic venous congestion, increase LV filling and cardiac output, and thus improve outcomes in patients with HFpEF.3 There are few therapeutic options for TR. Guideline-directed treatment is limited to diuretics, and surgery for secondary TR is rarely performed because of a unacceptably high perioperative mortality.19 The limitation of treatment options and resulting substantial under treatment have generated the recent development of novel transcatheter tricuspid valve intervention (TTVI). Early non-randomized trials have demonstrated that TTVIs may be safe and effective at reducing TR, and may even be associated with better survival and reduced HF hospitalization compared with current medical treatment options.25,26 The current data might support implementation of clinical trials to test safety and efficacy of TTVIs in HFpEF patients with moderate or severe TR.
Limitations
This is a retrospective study from tertiary referral centres, and as such, has inherent flaws related to selection and referral bias. The sample size was small, especially in the higher VCW categories, and this could bias the overall results, increase the risk for failing to detect a significant group difference that might be apparent in a larger sample (Type II error), and limit subgroup analyses. TR severity was assessed using the semiquantitative measures of VCW and jet area based upon image availability. Quantitative measures of TR severity, such as effective regurgitant orifice area and regurgitant volume using the PISA, or 3-dimensional Doppler vena contracta area would have provided greater prognostic value; however, they are not commonly clinically used and were therefore not available in the current study.
Conclusions
TR severity, as assessed by semiquantitative measures, is both independently and incrementally associated with the composite outcomes of all-cause mortality or HF hospitalization over multiple prognostic factors, as well as right heart remodelling and PH in patients with HFpEF. In contrast, TR severity as defined by the qualitative approach did not remain significant after adjusting for PH. The current data highlight the potential prognostic value of the semiquantitative measures of TR severity, which may help therapeutic approaches for TR in patients with HFpEF.
Supplementary data
Supplementary data are available at European Heart Journal - Cardiovascular Imaging online.
Funding
M.O. received research grants from the Fukuda Foundation for Medical Technology, the Mochida Memorial Foundation for Medical and Pharmaceutical Research, Nippon Shinyaku, and the Japanese Circulation Society. K.N. was supported by a Fellowship (N0025231) from the Heart Foundation of Australia.
Conflict of interest: none declared.
