By Michael H. Crawford, MD, Editor
SOURCE: Chin CW, et al. Left ventricular hypertrophy with strain and aortic stenosis. Circulation 2014;130:1607-1616
Progressive, compensatory left ventricular hypertrophy (LVH) in aortic stenosis (AS) ultimately leads to myocardial injury, fibrosis, and LV dysfunction. These investigators from the United Kingdom hypothesized that the LV strain pattern on electrocardiogram (ECG) is a marker for this degenerative process and may be useful to select asymptomatic patients with severe AS for valve replacement. Three groups of patients were studied. First, patients with any severity of AS were evaluated by cardiac MRI and plasma troponin to determine the pathophysiology of the LVH strain pattern (n = 102). The findings were validated in a similar population from another UK center (n = 64). The third group was recruited from an AS lipid-lowering trial population to determine outcomes related to the ECG strain pattern (n = 140). Confounders such as bundle branch block, cardiomyopathy, AV replacement, reduced LV systolic function, and digoxin use excluded patients from these groups. In the first group, 50% of the patients had LVH on ECG, and 40% of these patients had the strain pattern. Using magnetic resonance imaging (MRI) as the gold standard, ECG had a 96% positive predictive value for LVH and an 89% negative predictive value. Those with the LVH strain pattern had the highest LV mass indices on MRI and the most severe AS (P < 0.001). Despite similar LV ejection fractions, the patients with LVH strain pattern had reduced longitudinal and midwall shortening. Also, they had higher troponin values, MRI extracellular volume, and all of them had MRI late gadolinium enhancement. MRI showed localization of fibrosis in the mid ventricular area. LVH strain was not associated with the presence of coronary artery disease (CAD). Findings in the validation group were similar. In the outcome group, 14% had LVH strain pattern, which was associated with reduced 10-year event-free adjusted survival (hazard ratio, 2.67; 98% confidence interval, 1.35-5.27; P < 0.01). The authors concluded that the ECG LVH strain pattern in patients with AS is a marker for LV midwall fibrosis and predicts a lower 10-year event-free survival.
The timing of aortic valve replacement in AS is a challenge. In the new AHA/ACC Valve Disease Guidelines (2014) class I recommendations for valve replacement in severe AS include: symptoms and a LV ejection fraction (EF) < 50% or when undergoing other cardiac surgery. The problem with symptoms is that they are subjective and hard to determine in sedentary elderly patients. LVEF < 50%, in my experience, is a late manifestation of severe AS. Clearly, earlier objective indicators are needed. Two have made it to class IIa indicators for surgery in severe AS: a heavily calcified valve and the development of hypotension during exercise testing. The former has no quantitative criteria and the latter represents some risk to the patient and is not always feasible in older sedentary individuals. Thus, the search continues.
This interesting study shed considerable new light on the subject and proposes some new criteria for valve replacement in asymptomatic severe AS patients. The simple ECG LVH with strain pattern is associated with midwall fibrosis, myocardial injury, and reduced myocardial shortening. Also, it predicts reduced event-free survival over 10 years. Although the association of ECG LVH strain and poor outcomes has been observed in prior studies, this is the longest follow-up of such patients in the literature. Previous studies have hypothesized that the LVH strain pattern may be due to myocardial ischemia, which is due to increasing LVH or CAD, or both. This study does not address coronary blood flow per se, but does show the LVH strain pattern is associated with increased fibrosis on MRI, which could be the result of ischemia or the stimulation of fibrous cell growth along with myocardial cell growth. Interestingly, LVH strain was not associated with the presence or absence of coronary artery disease in this study.
One limitation to this study is that all the subjects were Caucasians. Equatorial Africans are known to have a higher frequency of high ECG QRS voltage and other unusual ECG patterns that could alter the reliability of the LV strain pattern in such individuals. In addition, the study employed three different patient populations that could introduce other biases that may have affected the results.
So would the LVH with strain pattern be a good early indication for valve replacement? Although highly specific for myocardial fibrosis (100%), it is not very sensitive (< 70%). In the multivariate analysis done in this study, only MRI fibrosis and the severity of AS were independent predictors of ECG LVH strain. MRI may be a sophisticated and expensive tool to decide in asymptomatic severe AS patients, whom to replace their valve. On the other hand, just replacing the valve in everyone with severe AS may also make sense. Unfortunately, it is a tough sell in asymptomatic patients. As transcutaneous atrioventricular valve replacement becomes more successful, this sell may be easier.