Myocardial considerations in type 2 diabetes: 2018

Heart failure (HF) continues to be one of the biggest public health problems in the 21st century, affecting 5.7 million adults in the US and accounting for one in nine deaths. Half of those developing HF die within 5 years of diagnosis. The cost associated with HF (i.e. for healthcare services, medications, and missed days of work) is estimated to be US$30 billion dollars per year.1 The leading causes of HF continue to be coronary heart disease (the most common), hypertension, and diabetes.2

In recent years, increasing numbers of patients with diabetes developing shortness of breath have been found to have left ventricular (LV) diastolic dysfunction. This is found in approximately 28% of asymptomatic and 81% of symptomatic patients with HF and is referred to as HF with preserved ejection fraction (HFpEF). Currently, there is no good treatment for HFpEF, with a number of failed trials.3 Moreover, recent landmark HF trials had only a low prevalence of diabetes patients (see Fig. 1).

Bursi et al.5 prospectively evaluated 556 patients from Olmsted County (MN, USA) using echocardiograms, finding that 44% of patients in that community had isolated LV diastolic dysfunction. At 6 months, mortality was 16% for both preserved and reduced ejection fraction (EF; age- and sex-adjusted hazard ratio 0.85; 95% confidence interval [CI] 0.61–1.19; P = 0.33 for preserved vs reduced EF).5

In summary, HF is found and well recognized in diabetes patients. It is also frequently reported as diastolic dysfunction with normal EF, commonly found in hypertensive patients with diabetes. In a recently published HF position paper,4 in current type 2 diabetes mellitus (T2DM) trials the prevalence of HF is approximately 20% (Fig. 2). Moreover, the prevalence of T2DM in HF trials is approximately 35%.4 There have been numerous HF trials completed, but only three major trials found a significant increase in adjusted all-cause and cardiovascular mortality (Table 1).

Matrix matters and insulin resistance

Patients with diabetes frequently have associated hypertension. In recent large prospective randomized cardiovascular safety trials, such as SAVOR (Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabeted Mellitus),6 EXAMINE (EXamination of cArdiovascular outcoMes with alogliptIN versus standard carE),7 TECOS (Trial EvaluatingCardivascular Outcomes with Sitagliptin),8 and EMPA-REG OUTCOME (Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes),9 hypertension was present in approximately 80% of study patients. A prospective cohort study in the US reported that, compared with normotensive patients, T2DM was almost 2.5-fold more likely to develop in subjects with hypertension.10 Similarly, people with diabetes have an increased risk of developing hypertension.11

Important research in diabetes found that one of the leading causes of hypertension in diabetes patients originates from a greater volume of exchangeable sodium due to hyperglycemia. Feldt-Rasmussen et al.12 reported that, in type 1 diabetes mellitus patients, the exchangeable sodium was approximately 2800–3000 mEq/1.73 m. This was approximately 10% higher than in healthy, normal, non-diabetic control patients, without significant differences in circulating plasma volume. The exchangeable sodium mass was positively correlated with mean arterial pressure. Elevated levels of sodium–glucose cotransporter 2 (SGLT2) may increase reabsorption of both glucose and sodium, suggesting another important benefit to the use of SGLT2 inhibitors.12-14

Insulin resistance has long been associated with hypertension, which can lead to end organ vascular (LV hypertrophy) damage associated with cardiovascular events. Moreover, many of these hypertensive diabetes patients have myocardial fibrosis (LV hypertrophy by echocardiography) with a consequent stiff, poorly compliant left ventricle, which causes diastolic dysfunction.15 Part of the risk of HF is due to myocardial fibrosis involving increased deposition of inflexible collagen, with thickening of small coronary vessels and the basement membrane.16 In the setting of elevated insulin and glucose levels, the renin–angiotensin–aldosterone system (RAAS) and sympathetic nervous system activate the transforming growth factor-β1 pathway, dysregulating extracellular matrix degradation.17

In summary, insulin-resistant patients usually have activation of the RAAS. In addition, sodium and water retention progressing to higher mean blood pressures lead to increased myocardial matrix stiffness, increasing the risk of HFpEF in most diabetes patients. Moreover, a systematic analysis of 65 studies showed evidence of association of both blood pressure and diabetes with arterial pulse wave velocity, a measure of arterial stiffness, further suggesting a role for increased myocardial matrix stiffness in hypertensive diabetes patients.18

Cardiomyocyte and power

Ischemic coronary artery disease is one of the most important causes of HF with reduced EF. However, patients with diabetes and a normal EF can present at a younger age with shortness of breath, but without evidence of coronary disease. Cellular fuel source abnormalities have been found to be another cause of diastolic dysfunction in this setting.

Increased free radicals with inflammation, endothelial cell dysfunction, and other metabolic abnormalities are common in diabetes patients. Cardiomyocyte fuel sources continue to be glucose and free fatty acids (FFAs). The balance between these two well-known fuel sources is affected in patients with diabetes. Activation of lipolysis increases FFA levels as a fuel source, leading to ketogenesis during the fasting state but, during the fed state, glucose is converted to glycogen with an increase in muscle glycogen. Recently, Ferrannini and DeFronzo suggested the potential of ketones as a “superfuel”.19 This originates, in part, from Veech,20 who advocated ketone bodies as “superfuel” for their superiority over glucose or FFAs in safeguarding efficient cardiac metabolism (Fig. 3).20 Frequently overlooked in diabetes HF are abnormalities seen in calcium metabolism. Multiple defects in transporters (e.g. troponin C, plasma Ca²⁺ pump, sarcolemmal Na⁺/Ca²⁺ exchanger, myofilament sensitivity) impair ionized calcium handling, increasing the action potential duration and prolonging diastolic relaxation time.21, 22 These abnormalities play key roles in the development of cardiac diastolic dysfunction, adding to the matrix abnormalities discussed previously.

The expression of small non-coding RNA molecules (i.e. microRNAs [miRNAs]) that function in RNA silencing and post-transcriptional regulation of gene expression is increased in diabetes, resulting in abnormalities in reactive oxygen species production, calcium handing, autophagy, and matrix formation.23, 24 This may prove to be another mechanism of development of HF in diabetes.

Treatment considerations and conclusions

Heart failure in insulin-resistant (diabetes) patients is complex. Increased wall stiffness and metabolic and molecular protein abnormalities are all involved in the development of HF in diabetes. Aggressive control of known cardiovascular and diabetes risk factors is our best currently available approach, but, unfortunately, many patients do not seek medical attention until they develop these diseases. The process may begin with genetic abnormalities in some people, but continued exposure to excess nutrients with consequent increased metabolic abnormalities must be targeted early in life. The American Heart Association (AHA)/American College of Cardiology (ACC) guideline25 suggest diuretics as the initial treatment of HFpEF for symptomatic patients. Recently, the American Diabetes Association26 has suggested screening for T2DM in an effort to prevent the development of ischemic heart disease. However, it is sobering to observe that the Detection of Ischemia in Asymptomatic Diabetes Trial did not find nuclear imaging helpful in improving cardiac event rates.27 However, evidence is strong for revascularization (coronary artery bypass grafting–left internal mammary artery or percutaneous coronary intervention [PCI]) for reducing cardiovascular events and improving survival in diabetes patients with asymptomatic ischemia.

Heart failure with reduced EF in diabetes is significantly related to myocardial ischemia, whether silent or symptomatic. Nuclear testing or fractional flow reserve in the catheterization laboratory can answer the question whether these patients are having physiologic myocardial ischemia. Furthermore, asymptomatic patients with diabetes and silent myocardial ischemia are in many ways at the highest risk of an adverse outcome. In a study of 1126 asymptomatic patients, the presence of ≥10% ischemia by single photon emission computed tomography (SPECT) was independently associated with death or myocardial infarction (MI) over a median follow-up of 6.9 years (hazard ratio [HR] 2.67; 95% CI 1.31–5.44; P = 0.007).28 Hachamovitch et al.29 studied 10 627 patients with quantitative stress SPECT, with early revascularization in 671. In that study, the mean 1.9-year cardiac mortality rate in non-revascularized patients increased from 0.7% in those with no ischemia to 6.7% in those with >20% ischemia.29

In an Italian study of 2835 patients, the presence of ischemia was a strong independent predictor of mortality in patients with diabetes (HR 1.71; 95% CI 1.34–2.18; P < 0.0001).30 Revascularization by PCI is beneficial by identification of hemodynamically significant flow-limiting lesions during hyperemia with adenosine in the cardiac catheterization laboratory. Fractional flow reserve (FFR) has been correlated with death and MI in medically treated patients. In a collaborative meta-analysis, the rates of death or MI and major adverse cardiac events (MACE) were inversely related to FFR at a median follow-up of 16 months.31

In addition, the 2017 AHA/ACC guidelines suggest the use of biomarker testing with N-terminal pro B-type natriuretic peptide (NT-proBNP) useful for establishing prognosis or disease severity in chronic HF.25 Pharmacologic treatment of HF with normal EF from the current guidelines recommends global risk reduction and the use of diuretics for symptomatic relief.27 Recent exciting data from two SGLT2 inhibitor trials, namely EMPA-REG OUTCOME with empagliflozin and CANVAS32 with canagliflozin, could revolutionize the treatment of HF in diabetes, but further studies are necessary.

Disclosure

None declared.

心力衰竭(Heart failure,HF)目前仍然是21世纪最大的公共健康问题之一,  有570万名美国成年人受累,  在每9名死亡的患者中就有1名死于HF。在HF患者中有一半的人在诊断后的5年内就会死亡。据估计与HF相关的成本(即医疗保健服务、治疗药物以及失去工作的时间)每年都达到了300亿美元¹。HF的主要病因仍然是冠心病(最常见)、高血压以及糖尿病²。

近年来,  越来越多出现呼吸急促的糖尿病患者被发现合并有左心室(LV)舒张功能障碍。在大约28%无症状以及81%有症状的HF患者中可发现合并有左心室舒张功能障碍,  被称为保持射血分数的HF(HF with preserved ejection fraction,HFpEF)。目前,  针对HFpEF还没有很好的治疗方法,  只有一些失败的试验³。此外,  最近针对HF的里程碑式试验只研究了较低比例的糖尿病患者(图1)。

Bursi等⁵使用超声心动图前瞻性地对556名来自Olmsted County县(MN, USA)的患者进行了评估,  发现在该社区中有44%的患者都单独合并LV舒张功能不全。第6个月时,  在保持射血分数(EF)以及射血分数下降的患者中总死亡率为16%(保持EF与EF下降的患者相比,  年龄与性别校正之后的危险比为0.85;95%置信区间[CI]为0.61-1.19;P=0.33)⁵。

总之,  在糖尿病患者中可以发现HF,目前人们已经很好地认识到了这一点。它也经常被报告为舒张功能障碍合并正常EF,常见于合并高血压的糖尿病患者。在最近发表的HF立场声明文件中⁴,目前2型糖尿病(T2DM)试验中的HF患病率大约为20%(图2)。此外,  在HF试验中T2DM的患病率大约为35%⁴。目前已经完成了许多HF试验,  但是只有3项主要试验发现糖尿病患者校正后的全因以及心血管死亡率显著增加(表1)。