Diabetes mellitus is a chronic metabolic disease influenced by both genetic and environmental factors, with a prevalence that continues to increase annually. The current global prevalence of diabetes among individuals aged 20–79 years is estimated at 10.5% (536.6 million) and is projected to rise to 12.2% (783.2 million) by 2045, with type 2 diabetes mellitus (T2DM), characterized by insulin resistance, accounting for more than 90% of all diabetes cases (1). The coexistence of T2DM with heart failure (HF), either with preserved ejection fraction (HFpEF) or reduced ejection fraction (HFrEF), is common and is associated with a significantly increased risk of HF hospitalization, all-cause mortality, and cardiovascular mortality. The primary factors contributing to HF in T2DM include coronary artery disease, arterial hypertension, and the direct deleterious effects of T2DM on the myocardium (2). Notably, even in T2DM patients with optimal control of traditional risk factors such as glucose, blood pressure, smoking, and hyperlipidemia, the risk of HF remains significant (3). Furthermore, 50–70% of T2DM patients exhibit asymptomatic left ventricular diastolic or systolic dysfunction, which is challenging to detect at early stages due to limitations in diagnostic techniques (4). However, a well-defined framework of assessments and interventions aimed at preventing or delaying HF progression in T2DM is lacking.Recent American Heart Association/American College of Cardiology/Heart Failure Society of America(AHA/ACC/HFSA)Guidelines and the Universal Definition and Classification of Heart Failure categorize the HF process into four stages (5, 6). According to these guidelines, stage A HF is defined as the absence of structural or functional heart disease or abnormal serum biomarkers (5). All individuals with diabetes are considered at higher risk of developing HF and are classified as stage A HF. As myocardial damage progresses, many patients exhibit asymptomatic structural heart disease or elevated filling pressures, defined as stage B HF. This is followed by the onset of HF signs and/or symptoms, representing stage C HF, which ultimately progresses to severe disruption of daily life, classified as stage D HF (7). Data indicate that 5-year survival rates for HF stages are 97% for stage A, 96% for stage B, 75% for stage C, and 20% for stage D (8). Clearly, progression from stage B to stage C HF significantly worsens the prognosis. Therefore, early detection and intervention to delay or prevent the transition from preclinical HF to symptomatic HF are critical public health objectives.

In the effort to prevent the progression of T2DM-related HF, several comprehensive risk assessment tools have been developed to predict HF risk (9–11). The measurement of cardiovascular biomarkers and evaluation of traditional risk factors offer valuable opportunities for risk stratification and individualized HF risk prediction (12).Simultaneously, advancements in imaging techniques have significantly enhanced the sensitivity for detecting early, mild cardiac dysfunction (13, 14).This review explores the pathophysiology, clinical phenotypes, epidemiological features, and risk factors associated with the development of HF in T2DM. It further analyzes the components and efficacy of various risk scoring systems and emphasizes potential preventive strategies. Lastly, based on the currently available, albeit limited, evidence-based medical data, this review proposes both pharmacological and non-pharmacological preventive measures that may effectively reduce the risk of HF progression.

The drivers of myocardial dysfunction in T2DM, in addition to the common coexisting conditions of hypertension and coronary artery disease, include hyperglycemia, insulin resistance/hyperinsulinemia, and impaired glucose tolerance. These factors may exert their effects years or even decades before the clinical onset of T2DM (2, 15). The deleterious effects of these mechanisms are associated with various metabolic abnormalities, such as the deposition of advanced glycosylation end products (AGEs), lipotoxicity, and microvascular dysfunction and rarefaction (16). Glucotoxicity, resulting from hyperglycemia, induces protein glycosylation, leading to an increase in AGEs. These are produced by the non-enzymatic glycosylation of lipids, lipoproteins, and amino acids (17). AGEs alter the mechanical properties of the extracellular matrix by increasing resistance to enzymatic protein hydrolysis in connective tissue and enhancing the cross-linking of collagen and laminin. This, in turn, mediates an increase in myocardial fibrosis, decreased compliance, and left ventricular diastolic dysfunction (18). AGEs can also bind to the receptor for AGEs (RAGE), which promotes the expression of inflammatory genes and increases the production of reactive oxygen species (ROS). This contributes further to inflammation, cardiomyocyte apoptosis, fibrosis, and disturbances in the extracellular matrix, leading to adverse cardiac remodeling and dysfunction (17, 19). Hyperglycemia also exacerbates myocardial pathophysiology through the activation of the renin-angiotensin-aldosterone system (RAAS) and the sympathetic nervous system, as well as coronary microvascular dysfunction secondary to end-glycosylation (20). Insulin resistance, a hallmark of T2DM, is associated with reduced myocardial glucose uptake and compensatory increased uptake of free fatty acids (FFAs) (21). Excessive FFAs in the myocardium can lead to increased lipotoxicity and oxidative stress, resulting in damage to the myocardium (22). Furthermore, normal coronary and myocardial insulin signaling promotes the activation of coronary endothelial nitric oxide synthase (eNOS) and enhances nitric oxide(NO) bioavailability, both of which are critical for optimal coronary microvascular blood flow and myocardial function (17) ( Figure 1 ). The harmful interrelationships among these pathophysiological mechanisms may reinforce one another, forming a vicious cycle that mediates myocardial injury and cardiac dysfunction in T2DM (16).

In 1972, Rubler et al. reported no evidence of coronary heart disease(CHD) in the autopsy findings of four patients with diabetic glomerulosclerosis and HF. The hearts of these patients exhibited myocardial hypertrophy and fibrosis, suggesting that metabolic factors were responsible for the observed phenomenon (23). Rubler’s observations were later supported by Regan’s study in 1977. This study involved 17 patients with T2DM, in whom CHD had been ruled out by angiography, providing clear evidence of diabetic cardiomyopathy (DCM). The study found elevated left ventricular end-diastolic pressure, reduced left ventricular compliance, and decreased left ventricular ejection fraction (LVEF) with diffuse hypokinesis (24).DCM is characterized by cardiomyocyte hypertrophy, interstitial fibrosis, and impaired coronary microvascular perfusion. In the early stages, DCM usually presents without clinical symptoms, but as the disease progresses, it leads to diastolic or systolic dysfunction. DCM is one of the primary causes of diabetes-related HF and death in patients with T2DM (25). The development of HF in T2DM is not a linear process but involves a series of evolving stages. Long-term exposure to hyperglycemia and insulin resistance (stage A HF) can eventually lead to adverse cardiac remodeling, left ventricular hypertrophy, and cardiac dysfunction (stage B HF) (26). Previous studies have shown that left ventricular diastolic dysfunction can be detected in approximately 75% of T2DM patients, even in the early stages of the disease, including those with normal blood pressure (27). The prevailing view is that left ventricular diastolic dysfunction is one of the first manifestations of DCM and is usually detected earlier than left ventricular systolic dysfunction (28). However, strain analysis and peak systolic velocity measurements have revealed subtle abnormalities in systolic function in 24% of T2DM patients without CHD or left ventricular hypertrophy (29). A recent study showed that altered systolic strain can be detected in patients with T2DM who exhibit normal diastolic function (30). This finding has led to an alternative view that diastolic dysfunction should not be regarded as the first sign of preclinical DCM (14).

Currently, two theoretical perspectives on cardiac dysfunction in DCM are debated: a single clinical phenotype and a dual clinical phenotype (31). The traditional understanding views DCM as progressing from diastolic to systolic dysfunction with structural remodeling, such as left ventricular hypertrophy. The alternative perspective suggests that diastolic and systolic dysfunction in DCM represent a single disease with two distinct phenotypes, which evolve independently into HFpEF or HFrEF (16). Abnormalities in glucose and lipid metabolism, as well as insulin resistance, coexist in obese patients with T2DM, predisposing them to the restrictive/HFpEF phenotype of DCM (diastolic insufficiency).In contrast, autoimmune-associated type 1 diabetes predisposes patients to the dilated/HFrEF phenotype (systolic insufficiency) (16).

The global prevalence of T2DM has increased by 30% over the past decade, from 333 million people in 2005 to 435 million in 2015 (32). Contemporary data suggest that the overall prevalence of HF in the general population is 11.8%, with a range of 4.7% to 13.3% (33). In numerous clinical trials involving patients with T2DM, the prevalence of HF at baseline ranged from approximately 10% to 30% (34, 35). In the Reykjavik study, the prevalence of HF among individuals with T2DM was 12%, with a higher prevalence in those over 70 years of age—16% for men and 22% for women (34). In the Kaiser Permanente population, the incidence of HF was nearly three times higher in individuals with T2DM under 75 years of age compared to those without T2DM. In the 75–84 age group, the risk of HF was twice as high in those with T2DM as in those without (35). Several factors contribute to these disparities, with gender, race, and other factors potentially influencing the observed differences. Early Framingham studies indicated that men with diabetes were twice as likely to develop HF as non-diabetic men, while women with diabetes were five times more likely to develop HF compared to non-diabetic women (36). Subsequent observational studies have demonstrated gender differences in the association between T2DM and HF risk, with an increased risk in women. Potential explanations for these gender differences include a greater burden of cardiometabolic risk factors, such as elevated body mass index (BMI) and systolic blood pressure at the time of T2DM diagnosis in women, treatment inertia that disproportionately impacts women, and differences in hormonal profiles (37, 38). Similarly, racial differences exist in the risk of HF associated with T2DM. For example, Black individuals with T2DM have a higher risk of developing HF compared to individuals of other races. Racial disparities in HF risk are largely driven by a higher burden of adverse social determinants of health, including lower incomes and limited access to healthcare among Black individuals (39).

The increased risk of HF in T2DM is partly attributable to the direct and indirect effects of hyperglycemia( Figure 1 ). The Atherosclerosis Risk in Communities (ARIC) study provides evidence of subclinical myocardial injury in individuals with prediabetes and T2DM, as assessed by high-sensitivity cardiac troponin T (hs-cTnT) assays. Subclinical myocardial injury increases progressively across the glycemic spectrum, from normoglycemia to prediabetes and diabetes. This is associated with an elevated risk of cardiovascular events, HF, and death, with the highest risk observed in patients with T2DM (40). Blood glucose levels, as measured by glycosylated hemoglobin A1c (HbA1c), serve as an independent biomarker of HF risk in patients with T2DM and even prediabetes (41–43). In T2DM patients, each 1% increase in HbA1c is associated with an 8% higher risk of HF hospitalization or HF-related death (41). In non-diabetic individuals, a 1% increase in HbA1c is associated with a 39% increased risk of HF, independent of other risk factors (43). Additionally, the duration of diabetes is an independent risk factor for HF, with each 5-year increase in the duration of diabetes corresponding to a 17% higher risk of HF (44).

Patients with T2DM often have additional comorbid risk factors for HF, which contribute to indirect myocardial injury. More than 70% of patients with T2DM have elevated blood pressure, and the coexistence of hypertension and T2DM exacerbates vascular remodeling, atherosclerosis, cardiac structural and functional abnormalities, and coronary microvascular dysfunction, all of which increase the risk of HF (45–47). Approximately 60% of individuals with T2DM also suffer from obesity, a common risk factor for both T2DM and HF (48, 49). An earlier study demonstrated that metabolic syndrome (defined by a BMI greater than 29.4 kg/m²) increased the risk of HF more than threefold over 20 years of follow-up. This increased risk persisted even after adjusting for established risk factors for HF (50). Subsequent studies have further shown that central abdominal obesity and high-fat mass are strongly associated with an elevated risk of HF in T2DM patients (51, 52). Chronic kidney disease (CKD) affects approximately 40% of T2DM patients (53).Impaired renal function and albuminuria are important independent risk factors for the development of HF in individuals with T2DM (11, 54). A recent study demonstrated that the degree of elevation in the urinary albumin-creatinine ratio (uACR) was associated with a progressively higher risk of new-onset HF in T2DM. This ranged from microalbuminuria (adjusted hazard ratio [HR], 2.21; 95% CI, 1.59–3.06) to macroalbuminuria (adjusted HR, 6.02; 95% CI, 4.11–8.80) (54). Conversely, slowing the progression of CKD in T2DM patients has been shown to reduce cardiovascular morbidity, particularly HF morbidity and mortality (55). The risk of CHD in individuals with T2DM is at least twice as high as in non-diabetic individuals. This is typically manifested as diffuse, multivessel coronary artery disease or, in some cases, as asymptomatic myocardial infarction, which is a significant cause of cardiac dysfunction, especially HFrEF (56–59). Notably, although some risk factors for HFpEF and HFrEF overlap (e.g., older age, diabetes mellitus, and a history of valvular disease), HFpEF is more commonly associated with women, obesity, and physical inactivity, whereas HFrEF is more commonly associated with men, smoking, and CHD (59–61).

Based on current understanding of early screening for HF in T2DM, the following process is summarized ( Figure 2 ): First, regular screening is recommended.The ADA guidelines suggest annual screening (84). Second, a clinical risk score can be useful as an initial step in screening. For cardiac biomarker screening, although it is reasonable to conduct it annually as recommended by the ADA, the clinical validity and cost-effectiveness of this approach for the entire T2DM population require further evaluation. Echocardiography may also be considered directly, without prior cardiac biomarker assessment, in patients with high clinical risk scores. This is because the presence of cardiac structural and functional abnormalities in these individuals can identify stage B HF, even in the absence of biomarker testing, necessitating aggressive intervention (7, 26). Third, elevated clinical risk or abnormal levels of cardiac biomarkers (e.g., BNP ≥50 pg/mL, NT-proBNP ≥125 pg/mL, or hs-cTn ≥99th percentile), in combination with echocardiogram evaluation, further subdivides stage B HF (B1-B4). This should prompt aggressive risk factor modification, lifestyle interventions, increased physical activity, weight loss, and early initiation of SGLT2 inhibitors, which should be progressively intensified as the stage B classification advances (7). Fourth, if comorbid CKD is present (uACR ≥30 mg/g or reduced eGFR), a combination of SGLT2 inhibitors and finerenone should be initiated immediately. Finally, patients at high risk should seek evaluation from a cardiologist at or after starting the recommended preventive measures, especially those with stage B3 or B4 HF.

Over the past decade, advancements in the field of HF in T2DM have primarily focused on the treatment of symptomatic HF, with limited evidence supporting the assessment and prevention of stage B HF progression to stage C HF. Specifically, tools for early risk stratification and the application of biomarkers have lacked large-scale validation, limiting the formation of clinical consensus. Existing predictive models primarily rely on clinical variables (e.g., age, gender, medical history), without fully incorporating subclinical cardiac structural changes or functional abnormalities (e.g., myocardial strain abnormalities).Additionally, the development of these models is based on regional and ethnic heterogeneity, which constrains their accuracy and generalizability. Large-scale cohort studies stratified by age, sex, and ethnicity are warranted in T2DM populations. By integrating clinical variables, novel biomarkers, and high-sensitivity indicators of cardiac dysfunction, these studies will enable the development of efficient predictive models. The discovery, clinical validation, and application of effective biomarkers are essential. Notably, as a validated surrogate marker of insulin resistance, the triglyceride-glucose (TyG) index significantly correlates with early subclinical HFpEF risk in T2DM and predicts HFpEF prognosis (173–175). This biomarker offers a novel strategy for identifying stage B HF and stratifying risk in T2DM. Emerging proteomic, metabolomic, and epigenetic technologies reveal promising biomarkers including sST2, galectin-3, and non-coding RNAs, which correlate with early myocardial fibrosis or metabolic dysregulation (176–178). However, their prognostic utility for HF progression in T2DM requires validation.Novel ultrasound techniques such as speckle-tracking echocardiography enhance diagnostic accuracy for stage B HF by detecting subtle myocardial strain abnormalities. Continued innovation in cardiac ultrasound will undoubtedly refine HF risk assessment. Moreover, recent research has achieved substantial progress in elucidating mechanisms underlying diabetes-induced myocardial damage, with key insights into mitochondrial dysfunction, apoptosis and ferroptosis regulation (179, 180). Developing precisely targeted therapeutic agents against these mechanisms holds promise for early and precise prevention of T2DM-mediated HF progression.

As a major cardiovascular complication in T2DM, HF poses significant clinical challenges.Advances in clinical risk stratification, biomarker profiling, and echocardiographic techniques for predicting HF onset, coupled with the development of novel therapeutics such as SGLT2 inhibitors, MRAs, and GLP-1 receptor agonists, hold promise for reducing the contemporary burden of HF in T2DM. Nevertheless, effective tools for screening high-risk individuals and targeted interventions to mitigate T2DM-driven pathophysiological mechanisms of HF have yet to be established.Consequently, bridging these translational gaps demands sustained advancement across developing accurate HF risk-stratification tools, identifying robust biomarkers and next-generation echocardiographic innovations, creating molecularly targeted therapeutics and ensuring their effective clinical translation.