Simple imaging of the right heart phenotype in heart failure

This editorial refers to ‘Pressure-volume loop validation of TAPSE/PASP for right ventricular arterial coupling in heart failure with pulmonary hypertension’ by A. Schmeisser et al., pp. 168–176. Schmeisser et al.1 show that in patients with heart failure with reduced ejection fraction (HFrEF) the tricuspid annular systolic plane excursion (TAPSE) is correlated to the ratio of end-systolic to arterial elastances (Ees/Ea) measured on an invasively measured right ventricular (RV) pressure–volume (PV) loop, without added value of its correction for systolic pulmonary artery pressure (PASP). This is good news. It has indeed been better realized in recent years that the ‘right heart phenotype’ in HF is of poor prognosis,2,3 but there is still uncertainty on how best to capture it with simple non-invasive measurements. Ventricular function is defined by plotting P as a function of V. On so obtained PV loops, inotropy (or systolic function) is defined by the maximum or end-systolic value of P/V and lusitropy (or diastolic function) by the end-diastolic value of P/V. A PV loop also allows for the definition of a lumped parameter afterload as the ratio between end-systolic pressure and stroke volume, and thus makes it possible to define the coupling of RV function to the pulmonary circulation by a ratio of Ees/Ea.4 The measurement of the Ees/Ea ratio is gold standard for the quantification of RV-PA coupling. Its normal value is around 1.5. Uncoupling is associated with increased RV dimensions when the Ees/Ea decreases by >50% to critical values of around 0.7–0.8, which herald a shift from homoeometric (systolic function) adaptation to heterometric (dimension or Starling) adaptation.5 The system thus has reserve before the occurrence of RV dilation along with increased filling pressures and clinical symptomatology of systemic congestion, or right heart phenotype. As elaborated on by Schmeisser et al., invasive measurements of Ees and Ea are too demanding to assess RV-PA coupling in clinical practice. A few years ago, Guazzi et al.6,7 introduced the echocardiography of TAPSE/PASP ratio instead, reasoning that TAPSE could be a surrogate of Ees and PASP an indirect estimate of Ea. The TAPSE/PASP was shown to be a potent predictor of outcome in HF6,7 and superior to other composite echocardiographic indices in the assessment of RV-PA coupling.7,8 Rigorously defined prognostic cut-off values for TAPSE and TAPSE/PASP in HF were derived, at 17 mmHg and 0.38 mm/mmHg, respectively.6,7,9 It is interesting that exactly the same cut-off values were confirmed by Schmeisser et al.1 Normal values range from 18 to 30 mm, mean 24 mm for TAPSE, and 0.4–2.0 mm/mmHg, mean 1.2 mm/mmHg for TAPSE/PASP.10 End-systolic elastance is commonly assumed to be a load-independent measurement of systolic function. But this is true only for beat-to-beat changes in loading as were initially described by Starling. Within minutes, a ventricle adapts by adjusting its contractility to preserve flow-output without dimension changes.4 The Ees/Ea expresses the adequacy of this adaptation. Thus, Ees may be in the normal range but abnormal by failing to increase in an adapted fashion. The same is obviously true for any surrogate measure of contractility, including TAPSE or ejection fraction (EF) [assessed at echocardiography by a fractional area change (FAC)], the difference is in the sensitivity to afterload. Not surprisingly, Schmeisser et al. show that both TAPSE and FAC are correlated to Ees/Ea but less tightly so than to Ees alone. The inevitable relationships between relative dimension changes during the cardiac cycle, such as TAPSE, EF and FAC, and Ees/Ea, are demonstrated by re-writing the ratio of elastances as EF/(1 − EF).11 Decreased RV-PA coupling is associated with parallel decreases of Ees/Ea and EF.5,11 Critical values of RV EF associated with a clinical diagnosis of right heart failure are in the range of <35–40%.5,11 Is the TAPSE/PASP ratio redundant with respect to TAPSE alone? The results reported by Schmeisser et al. suggest that this is the case.1 It has actually been previously shown that both the TAPSE and the TAPSE/PASP ratio may independently predict outcome in HFrEF.5,9 The study by Schmeiser et al. only reported on HFrEF patients. Even though 67% of them had a mean pulmonary artery pressure (PAP) higher than 25 mmHg allowing for a diagnosis of pulmonary hypertension, pulmonary vascular resistance hardly exceeded the limits of normal and the diastolic pressure gradient remained low. In spite of this minimal increase in afterload, the Ees/Ea was decreased to the severely decompensated range with values ranging from 0.4 to 0.9 or >50% decrease from normal. Similar observations of RV-PA uncoupling by predominantly altered contractile function in the presence of trivial increase in afterload have been reported in experimental over-pacing induced dilated cardiomyopathy.12 It is understandable that adjusting markedly depressed systolic function for minimally increased afterload is of limited added value for prognostication. But it is possible that the prognostic capability of TAPSE/PASP becomes better than that of TAPSE alone in patients with pulmonary vascular remodelling and increased PAP by more than expected from increased wedged PAP. Another important aspect to consider is a better preservation of RV contractility by positive systolic ventricular interaction in HF with preserved (p) EF.13 A decreased TAPSE/PASP in these patients would then indicate a worse prognosis on more advanced alteration in both terms of the ratio. Schmeisser et al. are to be congratulated for their contribution to a better understanding of the prognostic relevance of the RV in HF, and how to measure it noninvasively in clinical practice. Referring echocardiographic measurements to gold standard invasively measured PV loops in a total of 110 HF patients is a remarkable catheterization laboratory ‘tour de force’. Their emphasis on TAPSE or, as they show it, equally relevant FAC is clinically useful and strengthened by their results. Whether TAPSE (or FAC) alone is better or equally valid than corrected for PASP is likely in most patients with HFrEF, but needs to be still assessed in HFrEF with more severe pulmonary hypertension and/or HFpEF. Conflict of interest: none declared. The opinions expressed in this article are not necessarily those of the Editors of EHJCI, the European Heart Rhythm Association or the European Society of Cardiology.