Heart failure with preserved left ventricular (LV) systolic function accounts for approximately half of all new cases of heart failure (HF) in the elderly. When coronary and valvular heart disease is excluded, HF with preserved systolic function is largely a consequence of hypertensive heart disease. The pathophysiology likely involves diastolic dysfunction, in which case the diagnosis of diastolic HF can be invoked, however comprehensive echo-Doppler assessment is absent from most studies. Furthermore, using conventional echocardiographic techniques, hypertensive heart disease is only detected when gross structural changes or (less reliably) when poorly defined changes in LV filling have occurred, and patients are already at increased cardiovascular risk. Early stages of diastolic HF are not well characterized, largely due to group heterogeneity related to poor selection criteria, as well as a lack of widely available sensitive, load-independent non-invasive techniques for assessment of diastolic dysfunction. The overall hypothesis of the studies undertaken in this thesis is that sensitive new echocardiographic techniques can make a major contribution to the understanding of myocardial dysfunction at an early stage in the development of hypertensive heart disease. Further, this thesis hypothesizes that these new technologies will facilitate the evaluation of novel diagnostic strategies, and allow quantification of the response to novel treatments in patients with hypertensive heart disease and early diastolic HF.
The thesis initially reviews the clinical aspects and pathophysiology of hypertensive heart disease and diastolic HF, emphasizing the role of myocardial fibrosis, which may be a target for intervention. This is followed by a discussion of the current state- of-the-art in echocardiographic assessment of diastolic function, and potential new approaches which might allow more sensitive tissue characterization. The third chapter describes the methodologies used in this thesis, with particular attention to the quantitative assessment of long-axis LV segmental function and measurement of arterial compliance.
Chapter 4, a cross-sectional study, evaluates the clinical and echo-Doppler characteristics of patients with suspected diastolic HF using new parameters of diastolic function. Of 938 consecutive patients referred for assessment of LV function, 6% had diastolic HF defined by clinical HF, normal systolic function and no coronary disease. Using comprehensive Doppler echocardiography, diastolic dysfunction was confirmed in the majority of these patients with impaired relaxation as the dominant filling pattern, but with a significant proportion demonstrating advanced diastolic dysfunction (pseudonormal filling). These results indicate that when defined with strict patient selection and rigorous echo-Doppler criteria, isolated diastolic HF is less common than previously suggested. In addition, the identification of diastolic dysfunction is significantly related to the sophistication of the echocardiographic techniques used.
A hypertensive response to exercise (HRE) may reflect an early stage of chronic hypertension, and provides a model with which to validate new echocardiographic tools for the assessment of early myocardial dysfunction. In chapter 5, new quantitative measures of LV long-axis systolic function (strain imaging and ultrasound backscatter) and conventional echocardiography are assessed in 41 patients with normal ejection fraction and a HRE, and compared with 17 matched controls. Patients without resting hypertension (n=19) had reduced LV strain and backscatter parameters indicating systolic dysfunction, but normal exercise performance, diastolic function and LV mass compared with controls. Patients with resting hypertension had abnormalities in all parameters. These results indicate that a HRE is associated with isolated systolic dysfunction, which may represent the earliest abnormality in hypertensive heart disease. In addition, quantitative new echocardiographic techniques appear to be useful for evaluating subtle abnormalities of myocardial function in hypertensive heart disease.
Chapter 6 uses new techniques to assess the relationship of diastolic dysfunction and arterial compliance in hypertensive patients with suspected diastolic HF. Seventy medically treated hypertensive patients with exertional dyspnoea and 15 normotensive controls were studied. Arterial compliance was highest in controls and became progressively lower in hypertensives with progressively worse diastolic function. After adjusting for age, gender, body size, blood pressure (BP) and ventricular hypertrophy, arterial compliance was independently related to diastolic dysfunction. It is therefore concluded that reduced arterial compliance is an independent predictor of diastolic dysfunction in hypertensive heart disease and should be considered a potential target for intervention in diastolic HF.
As the new ultrasound technologies are highly specialised, this thesis sought to determine whether similar information could be obtained from B-type natriuretic peptide (BNP), which is a marker of LV filling pressures. Chapter 7 therefore examines the diagnostic value of BNP for predicting LV diastolic dysfunction in hypertensive patients with suspected diastolic HF. Quantitative echocardiographic techniques similar to those used in chapter 5 were applied to 72 ambulatory hypertensive subjects with exertional dyspnoea and ejection fraction >50%. BNP values were increased in patients with abnormal versus normal diastolic function and related independently to BP, LV systolic strain-rate (SR), left atrial function and age. However, the majority of patients with diastolic dysfunction had BNP levels within the normal range thus limiting the diagnostic value of resting BNP values in stable diastolic HF.
It was hypothesized that the inability of BNP to predict abnormal diastolic function in chapter 7 may have reflected relatively normal patient filling pressures at rest. As diastolic HF is characterized by dyspnoea due to increased LV filling pressures during stress, chapter 8 further examines the relationship of exercise-induced increases in BNP to LV filling pressures and parameters of cardiovascular performance in suspected diastolic HF. The increment of BNP with exercise was associated with better exercise capacity, LV systolic and diastolic function, and left atrial function. BNP was higher in patients with evidence of elevated filling pressures at peak exercise indicating that peak exercise BNP may be useful to identify exercise-induced elevation of filling pressures (and hence diastolic HF) in hypertensive patients undergoing exercise testing for investigation of exertional dyspnoea.
Finally, in chapter 9, the quantitative echo techniques which were validated as sensitive markers of myocardial dysfunction in chapter 5 are applied in an interventional study of hypertensive patients with early diastolic HF. Following randomization to spironolactone (an aldosterone antagonist) or placebo for 6 months, actively treated patients demonstrated increases in long-axis myocardial systolic function. Increases in long-axis strain were independent of BP changes, and were accompanied by a decrease in posterior wall thickness. The results demonstrate that aldosterone antagonism can improve myocardial function in hypertensive heart disease and may provide a specific treatment for diastolic HF.