Coupling is about work: matched as chamber, not really as fibre

This editorial refers to ‘Distinctive left ventricular–arterial and right ventricular–pulmonary arterial coupling observed in patients with heart failure and a higher left ventricular ejection fraction range’, by N. Ohte et al., https://doi.org/10.1093/ehjci/jeae033. Coupling, like that between individuals and companies, determines work performance. Ventricular–arterial coupling (VAC) determines cardiovascular performance and efficiency, and is conventionally described as the balance between the elastances of the ventricle (Ees) and the arterial tree (Ea) on the pressure–volume (PV) plane. When Ees matches Ea (Ees = Ea), stroke work (chamber work) is maximized. This is considered optimal, hence normal coupling.1 Ohte et al.2 investigated left- and right-sided VACs in heart failure (HF) patients with left ventricular (LV) ejection fraction (EF) > 40%, speculating that the VACs are associated with the worse prognosis in the higher LVEF category. The patients were analysed in three LVEF categories: HFEF-40–50%, 50–60%, and 60–%. The last two categories are known as HF with preserved EF (HFpEF). In HFpEF, concentric LV remodelling/hypertrophy increases Ees to match high Ea (Ees↑=Ea↑, pseudo-normalized coupling), called ‘Coupling disease’.3 This maladapted geometry causes blood pressure lability and load hypersensitivity. The European Society and Association stated in the consensus document4 that VAC is clinically useful in HF with LVEF < 50%, but ‘less informative’ in HFpEF due to the pseudo-normalization. However, Ohte et al. argue that in the HFEF-60–%, left-sided VAC is not only pseudo-normalized, but Ees is also overcompensated to Ea, resulting in excessive coupling (Ees>>Ea). Ohte et al. non-invasively evaluated left-sided VAC as Ees/Ea, where Ees was calculated as end-systolic pressure (Pes)/end-systolic volume (Ves) (see Supplementary data online, Appendix S1). Ees is originally measured invasively with catheterization, defined as the slope of linearized Pes–Ves relationship (ESPVR) during load manipulation. The volume-axis intercept of the ESPVR is termed V0 (unstressed volume), i.e. Ees = Pes/(Ves − V0).5,6 That is, Ohte et al. assumed V0 = 0 for all patients. Using the invasive method, Rosch et al.7 investigated left-sided VAC in the two HFpEF categories studied by Ohte et al. (see Supplementary data online, Table S1) Both articles had almost similar LVEF and geometry, but Rosch et al. reported the VAC of 1.01 for the HFEF-60–% category, whereas Ohte et al. found 2.12 (excessive coupling, likely due to much higher Ees of 5.19 vs. 1.85 in Rosch et al.). Invasive linearized ESPVR analysis in the physiologic data range often yields a negative (i.e. non-physiologic) V0, which may vary with ventricular mass and geometry.5,8 Therefore, Rosch et al.’s V0 values were applied to Ohte et al.’s subjects, and VAC was recalculated as Ees = Pes/(Ves − V0). Both articles showed near-identical VACs for the HFpEF categories (recalculated Ees/Ea = 0.96, pseudo-normalized coupling in HFEF-60–%). Ohte et al. may have overestimated Ees by fixing V0 to zero, misinterpreting the left-sided VAC as excessive. The method they used (Ees = Pes/Ves, called ‘pseudo Emax’6) consistently overestimates Ees obtained by the invasive method in human research by an average of 2.2 mmHg/mL.8 ‘V0 definitely cannot be disregarded’.6 ‘Ees and V0 must be presented in interpreting contractility’.5 Therefore, the aforementioned consensus document4 recommends using the single-beat method8 for non-invasive Ees estimation (online calculator9 available). Ohte et al. evaluated right-sided VAC using the ratio of tricuspid annular plane systolic excursion (TAPSE) to pulmonary arterial systolic pressure (PASP) and concluded that it was ineffective (i.e. Ees < Ea). However, ‘TAPSE/PASP was not always tightly correlated to the VAC’10 and might be suggested to be the inverse of Ea (see Supplementary data online, Appendix S2). Examining their other data would provide a more objective approach. Ohte et al. reported that SV was similar across all categories, and PASP was higher in the LVEF-higher categories. This suggests that pulmonary Ea was also higher (Pes↑/SV→=Ea↑). Since TAPSE was also similar across them, if TAPSE itself reflects VAC,10 then right ventricular (RV) Ees was also higher (Ees↑/Ea↑=coupling→, i.e. pseudo-normalized coupling). Besides, RV fractional area change (FAC) was higher in the LVEF-higher categories, while SV remained constant. This suggests concentric remodelling of the RV (SV→/RV chamber size↓=RVFAC↑). It is intriguing that these findings suggest that the right-sided VAC may also exhibit characteristics of Coupling disease. Focus shifts from VAC to workload. The PV loop for chamber work is derived from the force–length relationship, assuming that mural force governs pressure.6 On the other hand, the pressure–strain (PS) loop for myocardial work index (MWI) reflects the ‘loading condition’.11 In Ohte et al.’s data, does arterial pressure accurately reflect the loading across different ventricular geometries? As shown in the Supplementary data online, Figure S1, the Pes is similar for all categories (111, 113, 113 mmHg), but end-systolic myofibre stress4 decreases with higher LVEF and concentricity (30, 26, and 22 kPa, respectively). This suggests arterial pressure alone cannot fully capture the loading across diverse ventricular geometries. Hence, the myocardium cannot be ‘neglected’2 in this context. PV- and PS-relationships remain essential concepts. The pressure-axis elegantly reveals ‘acute’ effects by handgrip,7 resynchronization, and ischaemia11 in a ‘single’ patient. The volume-axis shows diastolic reserve.6 The strain-axis is normalized by cardiac size. The stress-axis facilitates the comparison of ventricular loading across different geometries. Figure 1 illustrates LV loops (red, Coupling disease; blue, normal) at equal SV and arterial pressure on PV, PS, and myofibre stress–strain (SS) planes. Each loop area represents chamber work, MWI, and myofibre work index, respectively. In Coupling disease, the PV loop shifts left without area change (red triangle), while the loop area on the PS-plane shrinks due to impaired strain (black arrow),12 and further shrinks on the SS-plane due to the stress reduction (white arrow). Assuming strain ∝ contractile force (CF)/stress, the loop area [(CF/stress) × stress] on SS-plane may proxy CF. Thus, in Coupling disease, ventricular geometric changes may be used as a lever to maintain chamber work at low CF. Loop area differences between coupling disease (red) and normal (blue) ventricle on the three planes. P, chamber pressure; V, chamber volume; ε, myocardial strain; σ, myofibre stress. See text for details. Importantly, myocardial/myofibre stiffness can be estimated from the SS-relationship.5 As coupling occurs throughout systole, not just at end-systole, comprehensive analysis combining profiles of time-varying the stiffness and arterial load4 may elucidate ineffective VACs in Coupling disease. Ohte et al.'s study on the link between both-side VACs and prognosis in HF patients with LVEF > 40% remains inconclusive due to methodological issues, but highlights the importance of the VACs in determining work performance and suggests shifting the perspective from pressure to stress for understanding and treating cardiovascular pathologies in HFpEF. Supplementary data are available at European Heart Journal - Cardiovascular Imaging online. None declared. The data that support the findings of this commentary are available from the corresponding author.