Background: Severe aortic stenosis (AS) is the most prevalent valve dysfunction in the ageing society. AS can induce left ventricular dysfunction. Up to one third of patients with severe AS develop an impaired left ventricular function (ejection fraction (EF) below <50%). In the AS patients with reduced EF two forms can be distinguished: 1) AS with reduced EF and high gradient (Vmax >4.0m/s; HG-AS) and 2) AS with reduced EF and low gradient (Vmax <4.0m/s; LG-AS). Differences between these subgroups has been poorly described and the underlying haemodynamic mechanism remains unclear.

Aims: In this study, we aimed to characterise patients with HG-AS and LG-AS in a multimodal way via clinical data, echocardiography, magnet resonance imaging, histology and next-generation-sequencing (NGS).

Methods:

Data for this study were gathered from the TAVR database in Goettingen. We retrospectively identified 383 patients from a clinical register and 98 patients from a TAVR study partially including left ventricular biopsies (40 with HG-AS, 58 with LG-AS) with echocardiographic LVD and severe AS, who underwent TAVR evaluation between 01/2017 - 12/2020. CMR imaging included myocardial function and tissue assessment. In addition for patients, who consented to the harvest of endomyocardial biopsies, histology and NGS was performed.

Results: Comparison of HG-AS and LG-AS in an all-comer clinical cohort showed no difference in EF (34  vs. 34%; p=0,61) and left ventricular end-diastolic diameter (51 vs. 51mm; p=0,83). More detail analysis could be performed in the well-defined patients from our TAVR study. LG-AS showed a lower stroke volume (66 vs. 55,6ml; p=0,001) and a shorter duration of flow (0.35 vs. 0,33ms; p=0.0046) and therefore a reduced cardiac output (4,7 vs. 4,0L/min; p=0,015). Here also EF (37 vs. 34; p=0,1) or left-ventricular volumes (LVEDV 119 vs. 122ml; p=0,71) were not significantly different compared to HG-AS in echocardiography and MRI. Differentiation in subgroups below/above EF of 35% did also show no differences in low parameters/gradients. Calculated PCWP from MRI showed no difference between the groups.

MRI-ECV (25.8 vs. 27.1 ml/ml²; p=0,012) was minimally increased in LG-AS, but neither MRI matrix volume (20 vs. 23.6; p=0,53) nor histologic assessment of fibrosis (23.2 vs. 25.6; p=0,73) was different. Global gene expression pattern did also not provide differences between LG-AS and HG-AS.  

LG-AS showed a higher prevalence for severe mitral regurgitation (8% vs. 24%; p=0,0001). A higher prevalence of prior PCI (17.5% vs. 44.8%; p=0.0052) or CABG (2.5% vs. 22.4%; p=0,.067), COPD (5,0% vs. 20.6%; p=0.039), right ventricular dysfunction (TAPSE 20,2 vs. 17,3mm; p=0,0019) and atrial fibrillation (30% vs. 51%; p=0.039) could be identified. Comorbidities therefore seem to be the major reason for impaired cardiac index in LG-AS patients. Follow up analyses of the patients showed a higher mortality in LG-AS vs. HG-AS.

Conclusion: Low transaortal gradient and low cardiac output in AS with reduced ejection fraction (LG-AS) is not determined by a more severely reduced systolic left ventricular function, but mainly determined by either additional cardiac comorbidities like mitral regurgitation, atrial fibrillation or right ventricular impairment. These comorbidities are probably the reason for the higher mortality even after TAVI. Treatment of these comorbidities should be considered especially in LG-AS patients.