Mortality and heart failure in patients with severe tricuspid regurgitation. Impact of RV volumes and function by CMR

Background and objectives

Right ventricle (RV) dilatation and dysfunction are established criteria for intervention in patients with significant tricuspid regurgitation (TR); however defined thresholds to support intervention are lacking. As a result the optimal timing for surgery in TR remains controversial and surgery is commonly undertaken at a late stage.

Purpose

To describe predictive cut-off values of RV size and function of poor prognosis in asymptomatic patients with significant TR.

Methods

Consecutive patients in stable clinical condition evaluated in the Heart Valve Clinic with significant TR (severe, massive or torrential TR) undergoing a Cardiac Magnetic Resonance (CMR) study were included. Conventional parameters of biventricular volume and function were assessed in all patients. A combined endpoint of hospital admission due to right heart failure and cardiovascular mortality was defined.

Results

75 patients were included in this study (age 75±8 years, 75% female, 91% functional TR). During a median follow-up of 3 years (IQR: 1.4–3.9 years), 39% of the patients (n=29) experienced the combined endpoint. After adjusting for age and LVEF in a multivariate Cox proportional model, RV-EDV and RVEF were independently associated with cardiovascular mortality and heart failure. Thresholds of RV-EDV ≥100 ml/m2, RV-ESV ≥40 ml/m2 and RVEF ≤58% held the best accuracy to predict outcomes (figure 1). Regression spline model for RVEF and outcomes are presented in the figure 2. They show that RV function negatively impacted event-free survival, with an increase in the HR spline function near the crossing value (red line, RVEF ≤58%). In multivariable analysis, following adjustment for age and LVEF, a value of RVEF ≤58% and RV-EDV ≥100 ml/m2, was associated with 2.29, and 3.91-fold increased risk of heart failure or cardiovascular death respectively (RVEF Hazard Ratio (HR): 2.29 [1.06–4.9], p=0.03, and RV-EDV HR: 3.91 [1.56–9.82], p=0.004).

Conclusion

RV size and function are crucial for determining optimal timing for TR intervention. For the first time, cut-off values of RV volume and function are defined in a cohort of consecutive patients based on outcome data. Proposed values provide a basis for prospective studies to establish definitive optimal surgical timing for severe TR.

Funding Acknowledgement

Type of funding sources: None. Figure 1. ROC and Cox regressions analysisRegression spline curve

Prognostic impact of different parameters of right ventricular systolic function in patients with significant tricuspid regurgitation. A cardiac magnetic resonance study

Background

Right ventricle (RV) dysfunction represent an established criteria for intervention in patients with significant tricuspid regurgitation (TR). RV ejection fraction (RVEF) by Cardiac Magnetic Resonance (CMR) is considered the gold standard of RV function; however it is influenced by changes of preload conditions and may remain unaffected until late stages in severe TR. Novel measures of RV function such as RV longitudinal shortening (RV-LS) and effective RV ejection fraction (eRVEF) may be earlier markers of RV dysfunction.

Purpose

To compare the prognostic impact of conventional and novel parameters of RV systolic function.

Methods

Consecutive patients in stable clinical condition evaluated in the Heart Valve Clinic with significant TR (severe, massive or torrential TR) undergoing a CMR study were included. In addition to conventional parameters of biventricular volume and function, RV-LS and eRVEF were assessed as novel parameters of RV function. RV-LS was assessed in the 4-chamber view by measuring the displacement of the tricuspid annulus during the cardiac cycle. The length between the epicardial border of the LV apex and the middle of a line connecting the origins of the tricuspid valve leaflets was measured in both end-systole and end-diastole. Effective RVEF (eRVEF) is a measure of RV global systolic function but corrected by TR volume. Both formulas are represented in figure 1. A combined endpoint of hospital admission due to right heart failure and cardiovascular mortality was defined

Results

75 patients were included in this study (age 75±8 years, 75% female, 91% functional TR) During a median follow-up of 3 years (IQR: 1.4–3.9 years), 39% of the patients (n=29) experienced the combined endpoint. RV-LS and eRVEF identified higher rates of RV dysfunction than RVEF. RV-LS of ≥−14% and eRVEF of ≤34% were associated with impaired prognosis (figure 2). After adjustment of age and LVEF, both eRVEF (adjusted HR per abnormal value: 5.29 95% CI, [2.25–12.4]) and RV-LS (adjusted HR per abnormal value: 3.46, 95% CI, [1.13–9.17]) were significantly associated with outcomes. Among all parameters of RV function, eRVEF was the strongest predictor of outcomes, incremental to RVEF (Δ C-statistic 0.139 [0.040–0.237], p=0.005).

Conclusion

RV function is crucial for determining optimal timing for TR intervention. RV-LS and eRVEF identify higher rates of RV dysfunction beyond RVEF. Among all measures of RV function, eRVEF held the strongest association with outcome, incremental to RVEF.

Funding Acknowledgement

Type of funding sources: None. Figure 1. RV-LS and eRVEF calculationFigure 2. Kaplan Meier Curves

Mortality and heart failure in patients with severe tricuspid regurgitation. Risk stratification based on TR severity by CMR

Background

Tricuspid regurgitation (TR) is related to poor prognosis independently of the etiology. A new scale classification has been proposed to better characterize the grading of more than severe TR. Massive and torrential TR seem to have worse prognosis based on echocardiographic studies.

Purpose

To stratify patients' risk based on TR severity by CMR

Methods

Consecutive patients in stable clinical condition evaluated in the Heart Valve Clinic with significant TR (severe, massive or torrential TR) with a contemporaneous echo and CMR were included. TR severity was evaluated by vena contracta (VC) and ERO method, using EPIQ system (Phillip Medical system Andover, Massachusets) and by TR regurgitant fraction (TRF) using a 1.5 Tesla CMR Philips scanner. A combined endpoint of hospital admission due to right heart failure and cardiovascular mortality was defined.

Results

75 patients were included in this study (age 75±8 years, 75% female, 91% functional TR). According to echocardiography 75% were severe, 15% were massive and 10% were torrential TR. Patients with massive and torrential TR showed higher RV end-diastolic volume and lower RVEF. A TRF >50% held the best accuracy to define massive / torrential TR. During a median follow-up of 3 years (IQR: 1.4 – 3.9 years), 39% of the patients (n=29) experienced the combined endpoint. After adjusting for age and LVEF in a multivariate Cox proportional model, TRF was independently associated with cardiovascular mortality and heart failure (hazard ratio per 1%=1.05, [1.02–1.08], p<0.001). Patients were stratified in 3 groups according to TR severity. Patients with TRF ≤40% showed the lowest incidence of events and those with TRF ≥50% experienced the worse prognosis (log rank=0.001, figure)

Conclusion

Patients with massive/torrential TR are populations at higher risk of heart failure and mortality. Risk stratification strategies may identify the patients who benefit the most of intensive therapeutic treatments and intervention on the tricuspid valve. New classification scheme should be included in CMR grading scales.

Funding Acknowledgement

Type of funding sources: None. Kaplan Meier curve. Risk stratification

P5566Prognostic value of quantitative assessment of tricuspid regurgitation. Correlation between echocardiography and cardiac magnetic resonance

Background

Significant tricuspid regurgitation (TR) is related to poor prognosis independently of the etiology. TR severity and right ventricular (RV) size and function are determinant in the evaluation of patients with RT and are independently related to outcomes. While TR severity is commonly evaluated with echocardiography (echo), cardiac magnetic resonance (CMR) is the gold standard to study the RV. The association between CMR and echocardiographic measures of quantitative TR is unknown.

Purpose

Our aim was to evaluate the association between the most commonly used methods in both techniques: biplane vena contracta (VC) and effective regurgitant orifice (ERO) parameters evaluated by echo and TR volume (TRV) and TR regurgitant fraction (TRF) by CMR; secondly we aimed to evaluate the prognostic value of each parameter.

Methods

Consecutive patients in stable clinical status with significant TR evaluated in the Heart Valve Clinic between 2015–2018 with a contemporaneous echo and CMR were included. TR severity was evaluated by VC and ERO method, using EPIQ system and by VRF and TRF using a 1.5 Tesla CMR Philips scanner. End-point included cardiovascular mortality, tricuspid valve surgery or heart failure.

Results

A total of 36 patients were included (mean age was 72±7 years, 72% females, 94% functional TR). Both VC and ERO showed moderate to strong and significant correlations with VRF and TRF (table). During a median follow up of 20 months [IQR: 10–29], 38% of the patients reached the combined end point (n=7 developed right heart failure, n=11 underwent tricuspid valve surgery, and n=2 died). Patients with events showed a larger ERO and higher VRF and TRF (p<0.01 for all) and a tendency to larger VC (p=0.06). PISA, VRF and TRF were prognostic factors of the combined endpoint (PISA per 0.1 cm2, HR: 282 [3.9–20362], p=0.01; VC per 1 mm, HR 1.27 [0.98–1.64] p=0.06; VRF per 1ml: HR: 1.02 [1.005–1.025], p=0.003; FRT per 1%, HR: 219.5 [4.8–9897], p=0.06). A value of PISA of 0.42, of VRF of 46 ml and FRV of 43% reached the best accuracy to predicted poor outcomes (p<0.01 for all).

Table 1. Bivariate correlations ERO VC Regurgitant volume by CMR R=0.57, p=0.004 R=0.55, p=0.003 Regurgitant fraction by CMR R=0.61, p<0.001 R=0.56, p=0.01

Conclusion

Validated echocardiographic parameters of TR are significantly correlated with quantitative measures by CMR. PISA by echo, and VRF and FRV by CMR are predictive of impaired prognosis. Further studies confirming our CMR cut-off values of poor outcomes are needed for clinical implementation.