Cardiovascular magnetic resonance and positron emission tomography in the assessment of aortic stenosis

Abstract

Background Aortic stenosis is not only characterized by progressive valve narrowing but also by the hypertrophic response of the left ventricle that ensues. In this most common valvular condition novel imaging approaches (cardiovascular magnetic resonance [CMR] and positron emission tomography [PET]) have shown promise in the assessment of disease progression and risk stratification. The central aim of this thesis was to investigate the potential of CMR imaging to refine risk prediction and to improve the imaging protocol of 18F-sodium fluoride PET for aortic stenosis. Methods and Results Asymmetric wall thickening in aortic stenosis In a prospective observational cohort study, 166 patients with aortic stenosis (age 69, 69% males, mean aortic valve area 1.0±0.4cm2) and 37 age and sex-matched healthy volunteers underwent phenotypic characterisation with comprehensive clinical, imaging and biomarker evaluation. Asymmetric wall thickening on both echocardiography and cardiovascular magnetic resonance was defined as regional wall thickening ≥13 mm and >1.5-fold the thickness of the opposing myocardial segment. Asymmetric wall thickening was observed in 26% (n=43) of patients with aortic stenosis using magnetic resonance and 17% (n=29) using echocardiography. Despite similar demographics, co-morbidities, valve narrowing, myocardial hypertrophy and fibrosis, patients with asymmetric wall thickening had increased cardiac troponin I and brain natriuretic peptide concentrations (both p<0.001). Over 28 [22, 33] months of follow-up, asymmetric wall thickening was an independent predictor of aortic valve replacement or death whether detected by magnetic resonance (HR=2.15; 95 CI 1.29 to 3.59; p=0.003) or echocardiography (HR=1.79; 95 CI 1.08 to 3.69; p=0.021). Animal model of pressure overload We performed serial Cardiac Magnetic Resonance (CMR) imaging every 2-week in 31 mice subjected to pressure overload (continuous angiotensin II infusion) for 6 weeks and investigated reverse remodelling by repeating CMR 1 month following normalization of afterload (n=9). Cine CMR was used to measure left ventricular volumes, mass, and systolic function whilst myocardial fibrosis was assessed using indexed ECV (iECV) calculated from T1-relaxation times acquired with a small animal modified look-locker inversion recovery sequence. During the initial phase of increased pressure afterload indices of left ventricular hypertrophy (0.091 [0.083, 0.105] vs 0.123 [0.111, 0.138] g) and myocardial fibrosis (iECV: 0.022 [0.019, 0.024] vs 0.022 [0.019, 0.024] mL) increased in line with blood pressure measurements (65.1±12.0 vs 84.7±9.2 mmHg) whilst left ventricular ejection fraction (LVEF, 59.3 [57.6, 59.9] vs 46.9 [38.5, 49.6] %) deteriorated significantly (all p≤0.01 compared to baseline). During the reverse remodelling phase blood pressure normalized (68.8±5.4 vs 65.1±12.0 mmHg, p=0.42 compared to baseline). Whilst LV mass (0.108 [0.098, 0.116] vs 0.091 [0.083, 0.105] g) and iECV (0.034 [0.032, 0.036] vs 0.022 [0.019, 0.024] mL) improved both remained elevated compared to baseline (p<0.05). Similarly, the LVEF remained impaired 51.1 [42.9, 52.8] vs 59.3 [57.6, 59.9] %, p=0.03. There was a strong association between LVEF and iECV values during pressure overload (r=-0.88, p<0.001). Gender differences in aortic stenosis Two hundred forty-nine patients (66±13 years, 30% women) with at least mild AS were recruited from two prospective observational cohort studies and underwent comprehensive Doppler echocardiography and CMR exams. On CMR, T1 mapping was used to quantify extracellular volume (ECV) fraction as a marker of diffuse fibrosis, and late gadolinium enhancement (LGE) was used to assess focal fibrosis. There was no difference in age between women and men (66±15 vs 66±12 years, p=0.78). However, women presented a better cardiovascular risk profile than men with less hypertension, dyslipidemia, diabetes, and coronary artery disease (all p≤0.10). As expected, LV mass index measured by CMR was smaller in women than in men (p<0.0001). Despite fewer comorbidities, women presented larger ECV fraction [29.0 (27.4-30.6) vs. 26.8 (25.1-28.7) %, p<0.0001] and similar LGE [4.5 (2.3- 7.0) vs. 2.8 (0.6-6.8) %, p=0.20] than men. In multivariable analysis, female sex remained an independent determinant of higher ECV fraction and LGE (both p≤0.05). Prior CT angiography for PET Forty-five patients (age 67.1±6.9 years, 76% males) underwent CTA (CTA1) and combined 18F-NaF PET/CTA (CTA2) imaging within 14 [10,21] days. We fused CTA1 from visit one with 18F-NaF PET from the second visit (PET) and compared visual pattern of activity, maximal standard uptake values (SUVmax) and target to background (TBR) measurements on (PET/CTA1) fused versus hybrid (PET/CTA2) data. On PET/CTA2, 226 coronary plaques were identified. Fifty-eight coronary segments from 28 (62%) patients had high 18F-NaF uptake (TBR>1.25), whilst 168 segments had lesions with 18F-NaF TBR ≤1.25. Uptake in all lesions was categorized identically on co-registered PET/CTA1. There was no significant difference in 18F-NaF uptake values between PET/CTA1 and PET/CTA2 (SUVmax: 1.16±0.40 vs. 1.15±0.39, p=0.53; TBR:1.10±0.45 vs. 1.09±0.46, p=0.55). The intraclass correlation coefficient for SUVmax and TBR was 0.987 (95%CI 0.983 to 0.991) and 0.986 (95%CI 0.981 to 0.992). There was no fixed or proportional bias between PET/CTA1 and PET/CTA2 for SUVmax and TBR. Cardiac motion correction of PET scans improved reproducibility with tighter 95% limits of agreement (±0.14 for SUVmax and ±0.15 for TBR vs. ±0.20 and ±0.20 on diastolic imaging; p<0.001). Delayed PET imaging Twenty patients (67±7years old, 55% male) with stable coronary artery disease underwent coronary CT angiography and PET/CT both 1 h and 3 h after the injection of 266.2±13.3 MBq of 18F-NaF. We compared the visual pattern of coronary uptake, maximal background (blood pool) activity, noise, standard uptake values (SUVmax), corrected SUV (cSUVmax) and target to background (TBR) measurements in lesions defined by CTA on 1h vs 3h post injection 18F-NaF PET. On 1h PET 26 CTA lesions with 18F-NaF PET uptake were identified in 12 (60%) patients. On 3h PET we detected 18F-NaF PET uptake in 7 lesions which were not identified on the 1h PET. The median cSUVmax and TBR values of these lesions were 0.48 [interquartile range (IQR) 0.44-0.51] and 1.45 [IQR, 1.39-1.52] compared to -0.01 [IQR, -0.03-0.001] and 0.95 [IQR, 0.90-0.98] on 1h PET, both p<0.001. Across the entire cohort 3h PET SUVmax values were similar to 1h PET measurements 1.63 [IQR, 1.37-1.98] vs. 1.55 [IQR, 1.43-1.89], p=0.30 and the background activity was lower 0.71 [IQR, 0.65-0.81] vs. 1.24 [IQR, 1.05-1.31], p<0.001. On 3h PET, the TBR values, cSUVmax and the noise were significantly higher (2.30 [IQR, 1.70-2.68] vs 1.28 [IQR, 0.98-1.56], p<0.001; 0.38 [IQR, 0.27-0.70] vs 0.90 [IQR, 0.64-1.17], p<0.001 and 0.10 [IQR, 0.09-0.12] vs. 0.07 [IQR, 0.06-0.09], p=0.02). The median cSUVmax and TBR values increased by 92% (range: 33-225%) and 80% (range: 20-177%). Conclusions In aortic stenosis, asymmetric wall thickening is associated with adverse prognosis, in this condition there are significant differences in the fibrosis burden between male and female patients and the adverse remodeling of the ventricle can be reproduced in a simple animal model of pressure overload. For 18F-NaF PET utilizing a CT angiography acquired before the PET acquisition enables adequate uptake quantification and delayed emission scanning facilitates image analysis.