Detection of stage B heart failure in type 2 diabetes mellitus

Heart failure (HF) is a complex clinical syndrome associated with significant mortality and morbidity, which leads to a significant burden for patients and healthcare systems. There is a well-established association between diabetes and HF that is partly but not entirely linked to coronary heart disease and hypertension. The Framingham Study firmly established the epidemiologic link between diabetes and HF. This study showed the risk of HF was increased 2.4-fold in men and 5-fold in women, and 12% diabetes already have established HF. The frequency of HF in diabetic patients is even higher among elderly adults with a 3.3% annual incidence rate. HF is usually a progressive condition. In recognition of the importance of this concept, American College of Cardiology (ACC) and the American Heart Association (AHA) have identified four stages of HF. Stage A HF (SAHF) comprises patients with any of the HF risk factors (diabetes, hypertension, coronary artery disease, metabolic syndrome and obesity) without evidence of left ventricular (LV) remodelling or low ejection fraction (EF). LV hypertrophy and reduced LVEF is associated with even greater risk of incident HF, hence, asymptomatic patients with these structural changes are categorized as having stage B HF (SBHF). Stage C and Stage D HF is clinical symptomatic HF while stage D is the end stage of HF. Patients may only move forward through the stages and not regress. The AHA and ACC recommend that an angiotensin converting enzyme inhibitor (ACEi) or angiotensin II receptor blocker (ARB) and a beta-blocker (BB) should be prescribed in the presence of Stage B HF (SBHF), to reduce the risk of developing symptomatic HF. Two considerations are important. First, the recommendation is based on the assumption that patients with non-ischaemic HF respond in the same way as the evidence involving patients with ischemic heart disease. Second, a large number of patients are classifiable as Stage B HF, but currently unrecognised. If there is a benefit in HF prevention, screening programs may need to be implemented to detect these individuals, but the optimal tools for doing so are unclear. The most commonly used measure of systolic function in clinical practice, LVEF, is not a robust index of myocardial or chamber contractility for this purpose because of its load dependence and sensitivity to chamber size. A more robust echo marker is desired. Global longitudinal strain (GLS) by speckle tracking imaging is potentially a useful marker as it measures longitudinal function being an early marker of disease and conveys more detailed information about LV systolic function than EF can provide. However, the usefulness of GLS in clinical decision-making still needs to be tested. T2DM is an important HF risk factor, and provides a readily-recognised group for screening for SBHF. This thesis investigates the clinical role and implication of early detection of HF by echocardiography in elderly asymptomatic patients with T2DM.The work in this thesis is divided into four parts: The first part (chapter 2) sought to improve the assessment of HF risk in patients with T2DM ‚Äö- a step that would be critical for effective HF screening. A systematic literature search and meta-analysis was performed to determine the effect size of each risk factor for incident HF in T2DM. Among elderly patients with T2DM, five common clinical variables are associated with significantly increased risk of incident HF and T2DM patients with these risks represent a target group for HF screening. The next part (chapter 4-8), pertains to the Tasmanian Study of Echocardiographic detection of Left ventricular dysfunction (Tas-ELF study), which sought to determine the role of early detection of HF by GLS in patients with T2DM in the Tasmanian community. Several studies were carried out to understand this. First, we assessed the association between insulin resistance (IR) and impaired exercise capacity to better understand the cause of exercise intolerance in T2DM, which is associated with LV dysfunction and adverse cardiac outcome including HF. In addition, this study supports the contribution of diabetic myocardial disease that contributes to the development of HF in diabetes. Second, longitudinal community studies were carried out to clarify the rate of progression through asymptomatic stage A to stage B HF in both diabetes and non-diabetes, in order to addresses the incidence and predictors of HF and all-cause mortality in this cohort, and define strategies for prevention of HF progression in T2DM. The result suggested the predictors of prognosis in SAHF patients due to T2DM and other causes of SAHF were different. SAHF due to T2DM had worse subclinical LV function, functional capacity and adverse outcome than other causes of SAHF. Impaired GLS was independently associated with prognosis in T2DM-SAHF, whereas not for other-SAHF patients. This study emphasised that not all types of SAHF are the same, and better targeting of interventions at the most vulnerable SAHF group ‚Äö- those with T2DM ‚Äö- seems appropriate. Third, we evaluated different echocardiographic markers including increased LV mass index, left atrial enlargement, LV diastolic dysfunction and impaired GLS as potential echocardiographic features of stage B HF in T2DM. The result suggested that GLS would be the optimal echocardiographic feature of stage B HF in T2DM for community screening. Fourth, the features associated with incident HF risk in T2DM are incompletely understood. In addition to myocardial disease, a number of other factors including mental factors are likely to influence the process. Therefore, we explored the association of depression and incident HF in elderly T2DM without any baseline cardiovascular symptoms. The result suggested that depression is prevalent in asymptomatic elderly patients with T2DM, and its role as an independent and incremental association with incident HF is an important confounder to the effect of myocardial disease. Finally, we observed the evolution of longitudinal changes of GLS among elderly T2DM patients, as little is known about the natural history of GLS over time. During our 2-year observation, the change of GLS in asymptomatic T2DM with preserved EF was only mild, but mental status had an independent association with worsening GLS. The last part (chapter 9) was a study that used a Markov model to assess the cost-effectiveness among different strategies for HF prevention from a healthcare payer perspective. The three strategies were elderly asymptomatic patients with T2DM receiving; 1) usual care; 2) primary prevention without screening; 3) screening of LVD and GLS guided prevention. The results showed that based on this Markov model, screening for asymptomatic LV dysfunction (evidenced by abnormal GLS) in elderly patients with T2DM appears cost-saving. These results could be used to inform clinical trials aimed at the early detection and treatment of LV dysfunction, with the intent of preventing the development of HF in T2DM. Conclusions. The results of the studies contained within this thesis suggest that subgroups of patients with diabetes are at the highest risk, and suitable for screening. Detection of LV dysfunction by strain imaging can predict the development of HF and all-cause of mortality in elderly asymptomatic patients with T2DM. Preventive treatment guided by GLS for HF prevention in elderly patients T2DM appears to be cost-saving.