Assessment of aortic stenosis using modern non-invasive imaging techniques
Dweck, Marc Richard Leslie
Introduction. Aortic stenosis is characterised both by progressive narrowing of the valve and the hypertrophic response of the left ventricle. The purpose of this thesis was to study the contribution of inflammation and calcification to valve narrowing using Positron Emission and Computed Tomography (PET/CT) and to investigate the hypertrophic response using cardiovascular magnetic resonance (CMR). Methods. PET/CT studies. Patients with aortic sclerosis and mild, moderate and severe stenosis were prospectively compared to matched control subjects. Aortic valve severity was determined by echocardiography. Calcification and inflammation in the aortic valve and coronary arteries were assessed by sodium 18-‐fluoride (18F-‐NaF) and 18-‐fluorodeoxyglucose (18F-‐FDG) uptake using PET. CMR studies. Consecutive patients with moderate or severe aortic stenosis undergoing CMR were enrolled into a registry. Patients who received gadolinium contrast were categorised into absent, mid-‐ wall or infarct patterns of late gadolinium enhancement (LGE) by blinded independent observers. Patients follow-‐up was completed using patient questionnaires, source record data and the National Strategic Tracing Scheme. After excluding those patients with concomitant triggers to LV remodeling, the extent and patterns of hypertrophy were investigated based upon measurements of indexed LV mass, indexed LV volume and the relative wall mass. Results. PET/CT studies. 121 subjects (20 controls; 20 aortic sclerosis; 25 mild, 33 moderate and 23 severe aortic stenosis) were studied. Quantification of tracer uptake within the valve demonstrated excellent inter-‐observer reproducibility with no biases and limits of agreement of ±0.21 (18F-‐NaF) and ±0.13 (18F-‐FDG) for maximum tissue-‐to-‐background ratios (TBR). Activity of both tracers was higher in patients with aortic stenosis than control subjects (18F-‐NaF: 2.87±0.82 vs 1.55±0.17; 18F-‐ FDG: 1.58±0.21 vs 1.30±0.13; both P<0.001). 18F-‐NaF uptake displayed a progressive rise with valve severity (r2=0.540, P<0.001) with a more modest increase observed for 18F-‐FDG (r2=0.218; P<0.001). Amongst patients with aortic stenosis, 91% had increased 18F-‐NaF (>1.97) and 35% increased 18F-‐ FDG (>1.63) uptake. Increased 18F-‐NaF uptake was also observed in the coronary arteries in a subset of patients with atherosclerosis. These patients (n=40) had higher rates of prior cardiovascular events (p=0.016) and angina (p=0.023), and higher Framingham risk scores (p=0.011). CMR studies. 143 patients (aged 68±14 years; 97 male) were followed up for 2.0±1.4 years and 27 died. Compared to those with no LGE (n=49), univariate analysis revealed that patients with mid-‐wall fibrosis (n=54) had an eight-‐fold increase in all-‐cause mortality despite similar aortic stenosis severity and coronary artery disease burden. Patients with an infarct pattern (n=40) had a six-‐fold increase. Mid-‐wall fibrosis (HR 5.35 [95% CI 1.16-‐24.56]; P=0.03) emerged as an independent predictor of all cause mortality by multivariate analysis. The pattern of LV remodelling was studied in 91 patients (61±21 years; 57 male) and displayed wide variation comprising normal ventricular geometry (n=11), concentric remodelling (n=11), asymmetric remodelling (n=11), concentric hypertrophy (n=34), asymmetric hypertrophy (n=14) and LV decompensation (n=10). The magnitude of the hypertrophic response was unrelated to the severity of aortic valve narrowing. Conclusions. Modern imaging techniques have provided important insights in to the pathology underlying aortic stenosis and suggest that valvular calcification and myocardial fibrosis have a key role. Both represent important potential targets for future therapeutic interventions.