End-systolic wall stress in aortic stenosis: comparing symptomatic and asymptomatic patients

Abstract

Aims: In aortic stenosis (AS), there is poor association between symptoms and conventional markers of AS severity or left ventricular (LV) systolic function. This may reflect that symptoms arise from LV diastolic dysfunction or that aortic valve area (AVA) and transvalvular gradient do not reflect afterload. We aimed to study the impact of afterload (end-systolic wall stress [ESWS]) on the presence of symptoms in AS and to test whether symptoms are related to increased ESWS or LV remodelling.

Methods and results: In a prospective study, ESWS was estimated by measuring LV wall thickness from MRI and estimated LV end systolic pressure from echocardiographic mean gradient and systolic blood pressure in 78 patients with severe AS scheduled for aortic valve replacement and 91 patients with asymptomatic severe AS. Symptomatic patients had lower indexed AVA (0.40±0.11 vs 0.45±0.09 cm2/m2, p=0.009). They had undergone more extensive remodelling (MRI LV mass index [LVMi]: 85±24 vs 69±17 g/m2, p<0.0001), had higher tricuspid regurgitant gradient (24±8 mm Hg vs 19 ± 7 mm Hg, p=0.0001) and poorer global longitudinal strain (-15.6±3.8 vs -19.9±3.2%, p<0.0001). ESWS was higher among symptomatic patients (96±51 vs 76±25 kdynes/cm2, p=0.003). Multivariate logistic regression identified echocardiographic relative wall thickness, tricuspid gradient, mitral deceleration time, early diastolic strain rate, MRI LVMi, MRI LV end-diastolic volume index and ESWS as independently associated with being symptomatic.

Conclusion: ESWS can be estimated from multimodality imaging combining MRI and echocardiography. It is correlated with LV remodelling and neurohormonal activation and is independently associated with symptomatic status in AS.

Keywords: echocardiography; magnetic resonance imaging; valvular heart disease.

Conflict of interest statement

Competing interests: None declared.

Figures

Figure 1

Wall stress measurements. Wall stress measurements in two patients: (A) patient with symptomatic AS, systolic blood pressure 126 mm Hg, mean gradient 87 mm Hg, wall stress 239 kdynes/cm2; (B) patient with asymptomatic AS, systolic blood pressure 128 mm Hg, mean gradient 73 mm Hg, wall stress 47 kdynes/cm2. AS, aortic stenosis.

Figure 2

(A) Wall stress according to symptoms, (B) BNP according to symptoms. Some symptomatic patients had NYHA class 1, because their symptoms were either angina, syncope or abnormal exercise test. Only 90 asymptomatic patients had sufficient MRI images to assess wall stress, and BNP could only be measured in 89 asymptomatic and 77 symptomatic patients. BNP, brain natriuretic peptide.

Figure 3

Correlation between end-systolic wall stress and markers of remodelling according to relative wall thickness. BNP, brain natriuretic peptide; LV, left ventricle.

Figure 4

Receiver operating characteristic to predict symptoms. Model 1: end-systolic wall stress. Model 2: age, gender and aortic valve area index. Model 3: Model 1+additional echocardiographic variables (relative wall thickness, mitral valve deceleration time, tricuspid regurgitation and global longitudinal strain). Model 4: Model 2+MRI variables (LV end-diastolic volume index and LV mass index) and end-systolic wall stress. AUC, area under the curve; LV, left ventricle.