Cover Page
The handle
https://hdl.handle.net/1887/3157141
holds various files of this Leiden
University dissertation.
Author: Ng, A.
Title:
Multimodality imaging in metabolic heart disease179
Summary and conclusions
Both obese and diabetic patients can develop changes in myocardial structure and func-tion independent of coronary artery disease. Collectively known as metabolic heart dis-ease, this thesis attempted to illuminate its pathogenesis and prognostic implications. The thesis began by summarizing the various imaging modalities used to quantifying changes in myocardial structure and contractile function. Specifically, magnetic reso-nance imaging techniques such as spectroscopy and T1 mapping can be used to quantify myocardial energy substrate utilization and extent of interstitial fibrosis respectively. In addition, myocardial deformation imaging using 2-dimensional speckle tracking echocardiography provides an accurate quantification of myocardial contractility. This technique has previously been validation against the “gold standard” myocardial strain imaging using magnetic resonance imaging myocardial tagging sequence.
In Chapter 2, asymptomatic type 2 male diabetic patients without any associated diabetic complications were prospectively recruited to undergo 2-dimensional speckle tracking echocardiography. These diabetic patients were compared to male controls matched to age and body mass index. This study was first to demonstrate impaired myocardial contractility in asymptomatic diabetic patients, with an impaired longitu-dinal but preserved circumferential and radial functions. This study established global longitudinal strain as the most sensitive marker for identifying subclinical myocardial dysfunction in diabetic patients.
Chapter 3 explored the combined detrimental effects of increasing body mass index and diabetes on myocardial contractile. In this multicenter study, we demonstrated that both diabetes and increasing body mass index were independent determinants of global longitudinal strain. Secondly, the adverse impact of increasing body mass index and diabetes were additive. Finally, increasing body mass index was a stronger determi-nant of impaired global longitudinal strain compared to diabetes.
In Chapters 4 to 6, we attempted to evaluate the pathogenesis of subclinical myocar-dial dysfunction associated with diabetes and increasing body mass index. In a process analogous to fatty liver disease, diabetic patients can accumulate triglyceride within the cytoplasm of myocytes. Under normal physiological conditions, the heart uses both free fatty acids and glucose as its energy source. However, diabetes is characterized by insulin resistance and the heart preferentially uses free fatty acid for metabolism despite the prevailing hyperglycemia. The intracellular triglyceride is a source of free fatty acid and eventually, the excessive triglyceride/free fatty acid concentration exceeds the myo-cardial β-oxidative capacity. Consequently, there is accumulation of toxic intermediates such as ceramide and diacylglycerol that results in cellular dysfunction. Using proton
180
magnetic resonance spectroscopy double gated to ECG and respiratory motion, we can quantify the concentration of intramyocardial triglyceride. Furthermore, recent ad-vances in magnetic resonance imaging using post-contrast T1 mapping techniques, we can also quantify the extent of interstitial myocardial fibrosis (i.e. myocardial structural change).
In Chapter 4, intramyocardial triglyceride concentration was quantified using proton magnetic resonance spectroscopy in asymptomatic male diabetic patients, and patients were dichotomized into 2 groups based on the median intramyocardial triglyceride concentration. In this study, patients with higher intramyocardial triglyceride concen-tration was independently associated with impaired left and right ventricular global longitudinal strain.
In Chapter 5, the burden of interstitial fibrosis was quantified using post-contrast T1 maps in 50 diabetic patients without underlying coronary artery disease or macroscopic scar. Compared to healthy controls, diabetic patients had more interstitial fibrosis was reflected by a shorter post-contrast T1 time. Furthermore, post-contrast T1 time was the strongest determinant of global longitudinal strain on multivariable analysis (standard-ized β = -0.626, p < 0.001).
In Chapter 6, we evaluated the combined effects of myocardial steatosis and intersti-tial fibrosis on myocardial contractile function in patients with increasing body mass index. Importantly, patients with underlying diabetes and hypertension were excluded to avoid any potential confounding influences. In this study, epicardial fat volume (a measure of visceral adiposity) was an independent determinant of both intramyocardial triglyceride accumulation and the burden of interstitial fibrosis. On univariable analysis, epicardial fat volume, intramyocardial triglyceride concentration and interstitial fibrosis were associated with left ventricular global longitudinal strain. However, only epicardial fat volume was independently associated with global longitudinal strain on multivari-able analysis.
Finally, in Chapter 7, we aimed to determine the prognostic implications of subclinical myocardial dysfunction as identified by global longitudinal strain. In this multicenter study, “normal” global longitudinal strain cut-off value of > -17.0% was defined from 104 healthy volunteers recruited from the community. Applying this normal global longitu-dinal strain cut-off value to 397 type 2 diabetic patients with normal left ventricular ejec-tion fracejec-tion, up to 45% of patients had evidence of subclinical myocardial dysfuncejec-tion. At follow-up, the presence of subclinical myocardial dysfunction in diabetic patients was independently associated with increased all-cause mortality on follow-up (hazard ratio
181
Summary and conclusions
2.83, 95% confidence interval 1.40 – 5.71, p = 0.004). Vice versa, diabetic patients with-out subclinical myocardial dysfunction had similar survival as the general population (standardized mortality ratio 0.94, 95% confidence interval 0.52 – 1.58). Using decision curve analysis, quantifying global longitudinal strain and identifying subclinical myo-cardial dysfunction was clinically useful and provided incremental net clinical benefit for diabetic patients.
Conclusions and future directions
Echocardiographic 2-dimensional speckle tracking analysis is now a well-established tool that is widely used in routine clinical practice. It is now well accepted that global longitudinal strain is superior to “traditional” left ventricular ejection fraction in detect-ing early subtle changes and prognosticatdetect-ing adverse outcomes on follow-up. In this thesis, we determined that up to 45% of diabetic patients had evidence of subclinical myocardial dysfunction despite normal left ventricular ejection fraction or known underlying significant coronary artery disease. Using state-of-the-art cardiac magnetic resonance imaging, we have attempted to elucidate the pathophysiology of metabolic heart disease. We have identified increased epicardial adipose tissue, myocardial ste-atosis and interstitial fibrosis were associated with myocardial contractile dysfunction. Future research should aim at corroborating these findings, risk stratify patients, and identifying effective therapeutic interventions (e.g. medically or surgically induced weight loss, diabetic medications such as sodium-glucose transport protein 2 (SGLT2) inhibitors, etc.) that can potentially reverse subclinical myocardial dysfunction and improve patient outcome.
