Prognostic value of relative wall thickness in heart failure with preserved ejection fraction: what is the best method for its calculation?

Abstract

Abstract Background The HFA-PEFF diagnostic algorithm is a recently published tool to help in the diagnosis of heart failure with preserved ejection fraction (HFpEF). One of the echocardiographic diagnostic criteria is left ventricular (LV) relative wall thickness (RWT), an index of LV concentricity. LV wall thickness can be measured by echocardiography at the posterior wall (PW) and/or the interventricular septum (IVS) in parasternal long axis view. There are three methods of RWT calculation: RWTPW= 2xPW/LV dimension at end diastole (LVDd) – the most used method, RWTIVS = 2 x IVS / LVDd and RWTPW + IVS = (PW + IVS) / LVDd. This study compares the prognostic value of these 3 methods of calculation in patients with acute HFpEF. Methods All patients admitted with acute HFpEF in a Cardiology Department during 7 years were included. RWT was considered elevated if superior to its median and was calculated with the 3 formulas. In-hospital mortality (IHM) was evaluated. The primary endpoint (EP) was a composite of all-cause mortality or hospitalization for HF during follow-up of 24 months. Statistical analysis used chi-square and Mann-Whitney U tests, binary logistic regressions, and Kaplan-Meier curves. Results 478 patients were studied (61.3% female, mean age 79.4±8.3 years). Mean RWTPW, RWTIVS and RWTPW+IVS were 0.46±0.16, 0.50±0.17 and 0.48±0.16, respectively. IHM was 3.4% and primary EP occurred in 57.8%. High RWTPW was associated with higher LV ejection fraction (LVEF) (p<0.001). Patients with high RWTIVS were older (p=0.044). High RWTPW+IVS was associated with higher left atrial area (p=0.037) and higher LVEF (p=0.002). There was no statistically significant difference between patients with high and low RWT, calculated using the 3 formulas, in other indices that are commonly used to assess diastolic function, namely in e' and E/e'. None of the 3 methods of RWT calculation was a predictor of IHM. Survival analysis showed that patients with high RWTPW had higher incidence of the primary EP (43.2% vs. 16.8%, Kaplan-Meier χ2=5.99; p=0.014), but not patients with high RWTIVS (Kaplan-Meier χ2=0.23; p=0.631) or RWTPW+IVS (Kaplan-Meier χ2=1.92; p=0.166). RWTPW was a predictor of primary EP (OR 1.81; 95% CI 1.15–2.85; p=0.011) and this result was independent from e' and E/e' (OR 2.96; 95% CI 1.08–8.10; p=0.035). Conclusion In this study comparing 3 formulas for calculation of RWT, RWTPW had better risk prediction during follow-up than RWTIVS or RWTPW+IVS. RWTPW was a predictor of all-cause mortality and hospitalization for HF, and was independent from e' and E/e', indexes that are also recommended in HFA-PEFF diagnostic algorithm. Therefore, the formula incorporating PW should be preferred in the evaluation of patients with suspected or diagnosed HFpEF. Funding Acknowledgement Type of funding sources: None.