Assessment and risk prediction in patients with aortic stenosis: insights from cardiovascular magnetic resonance
Chin, Calvin Woon Loong
MetadataShow full item record
BACKGROUND Aortic stenosis affects not only the valve but also the myocardium. In response to the increased afterload, left ventricular hypertrophy initially occurs as a compensatory response to maintain wall stress and cardiac output but ultimately, decompensation and heart failure ensues. The transition from adaptation to decompensation is driven by myocyte death and myocardial fibrosis. The aims of the thesis are to investigate cardiovascular magnetic resonance assessment of disease severity and myocardial fibrosis, and explore its relationship with other biomarkers of disease activity and clinical outcome in patients with aortic stenosis. METHODS AND RESULTS The conventional assessment of aortic stenosis relies heavily on two-dimensional and Doppler echocardiography but there are inherent limitations in echocardiography that can affect the severity classification. I demonstrated that cardiovascular magnetic resonance offered a more accurate estimation of left ventricular volumes and mass, and excellent myocardial characterization. Indeed, inaccurate stroke volume estimation by Doppler echocardiography and inconsistent thresholds in current guidelines accounted for more than 40% of patients with discordant small-area, lowgradient aortic stenosis. These data may explain the variable prognosis reported in this unique group of patients, and argue for more accurate assessment of borderline cases with cardiovascular magnetic resonance. Late gadolinium enhancement imaging detects focal areas of established myocardial fibrosis. In many conditions, including aortic stenosis, a more diffuse form of fibrosis predominates, which is potentially reversible and not readily identified by late gadolinium enhancement. Recently several myocardial T1 mapping approaches have been developed to quantify diffuse fibrosis. Using a standardized and systematic approach, I compared several commonly used T1 mapping techniques and identified that extracellular volume had the best profile (reproducibility and discriminatory potential) for the identification of diffuse fibrosis in patients with aortic stenosis. Cardiac troponin is a structural protein present in the cardiac myocytes. Recent advances in assay technology have substantially improved sensitivity, allowing quantification of troponin concentrations with a high degree of precision in everyone. In more than 250 patients with aortic stenosis, I demonstrated that cardiac troponin I concentrations were independently associated with markers of left ventricular decompensation (hypertrophy and fibrosis) and predicted clinical outcome in patients with aortic stenosis. This suggests that myocardial fibrosis detected by cardiovascular magnetic resonance is consequent on myocardial injury secondary to left ventricular decompensation. Left ventricular hypertrophy with strain pattern on a 12-lead electrocardiogram is associated with poor outcome in patients with aortic stenosis, but the mechanism of this electrocardiographic pattern has not been described. In more than 300 patients with aortic stenosis, I demonstrated that these characteristic repolarization abnormalities were a highly specific marker of focal mid-wall myocardial fibrosis (specificity of 99% and sensitivity of 54%). Moreover, the prognostic value of this electrocardiographic pattern was again confirmed with markedly worse long-term outcomes in these patients. CONCLUSION I have demonstrated that cardiovascular magnetic resonance can assist in the assessment of disease severity in patients with aortic stenosis and discordant echocardiographic findings. Moreover, I have validated the assessment of diffuse myocardial fibrosis, as well as, demonstrated the close association between myocardial fibrosis and biomarkers of myocardial injury and electrocardiographic strain pattern that predicted an adverse outcome in patients with aortic stenosis.