Shedding light on the effects of sodium-glucose cotransporter 2 inhibitors in the early stages of heart failure

Abstract

Heart failure (HF), which falls outside of the historical macrovascular or microvascular categorizations of diabetes complications, has been overlooked for long time in diabetic patients, despite its increasing prevalence and mortality. As originally stated in the Framingham studies, diabetes is associated with an increased risk of HF. Subsequent studies not only corroborated these findings but also identified HF as the most frequent first onset of cardiovascular involvement. The paramount role of proper management of common modifiable risk factors such as hypertension, obesity, dyslipidemia and smoking, became rapidly clear. Conversely, the impact of intensive glycemic control was more contentious. A large meta-analysis of randomized controlled trials reported a lack of effect of strict glycemic control as compared to standard care on HF-related outcomes. The considerable heterogeneity of the effect estimate and the higher risk conferred by thiazolidinediones suggested that mechanism of action of antidiabetic drugs played a key role. Furthermore, the safety concerns of pioglitazone led Food and Drug Administration to release a guidance for drug manufacturers stating that cardiovascular risk should be comprehensively evaluated during drug development. Surprisingly, in just a few years, large cardiovascular outcome trials established the beneficial cardiovascular effects of sodium-glucose cotransporter 2 inhibitors. These effects were consistent regardless diabetes and ejection fraction. Therefore, scientific community started to question the glucose-lowering and diuretic properties of sodium-glucose cotransporter 2 inhibitors as the unique mechanisms for improved outcomes. A plenty of preclinical and clinical studies identified several mechanisms besides glucose-lowering effects. However, these mechanistic studies focused on animal models and patients with established HF. If the same mechanisms account for beneficial effects in patients at risk for or with pre-HF is unknown. Grubić Rotkvić et al published an interesting work adding data in early stages HF.

Key Words: Heart failure; Sodium-glucose cotransporter 2; Sodium-glucose cotransporter 2 inhibitors; Diabetes complications; Glycemic control

Core Tip: Despite the advances in pharmacological therapy of symptomatic heart failure (HF), evidence for asymptomatic patients is poor, and sodium-glucose cotransporter 2 inhibitors are currently not recommended in pre-HF patients without diabetes. Therefore, further studies are needed to provide efficacy and mechanistic insights in patients in early stages of HF, especially considering the significant clinical and financial implications of treating HF at the beginning of its natural history.

TO THE EDITOR

Heart failure (HF), which falls outside the historical macrovascular or microvascular categorizations of diabetes complications, has long been overlooked in patients with type 2 diabetes mellitus (T2DM), despite its increasing prevalence and mortality. As originally stated in the Framingham heart study, T2DM is associated with an increased risk of HF[1]. Subsequent studies not only corroborated these findings, but also identified HF as the most frequent first onset of cardiovascular involvement[2,3]. Consequently, the paramount role of proper management of common modifiable risk factors such as hypertension, obesity, dyslipidemia, and smoking has become clear. In contrast, the impact of intensive glycemic control has been more contentious.

Previously, a large meta-analysis of randomized controlled trials (RCTs) reported that strict glycemic control had a lack of effect on HF-related outcomes when compared to standard care (odds ratio: 1.20, 95% confidence interval: 0.96-1.48)[4]. The considerable heterogeneity of the effect estimate (I² = 69%) and the higher risk for HF conferred by thiazolidinediones suggests that the mechanism of action of antidiabetic drugs plays a key role[4]. In fact, due to safety concerns regarding pioglitazone, the Food and Drug Administration has released guidelines stating that drug manufacturers should comprehensively evaluate cardiovascular risk during drug development[5].

In recent years, large cardiovascular outcome trials have established the beneficial cardiovascular effects of sodium-glucose cotransporter 2 inhibitors (SGLT2i)[6]. These effects were consistent, regardless of the diabetic status or ejection fraction[7,8]. Therefore, the scientific community has begun to consider the glucose-lowering and diuretic properties of SGLT2i as unique mechanisms for improved outcomes. Numerous preclinical and clinical studies have identified several other mechanisms affecting metabolism, bioenergetics, fibrosis, inflammation, and oxidative stress[9]. However, these mechanistic studies have focused on animal models and patients with established HF. Therefore, whether the same mechanisms account for the beneficial effects in patients “at risk of HF” or with “pre-HF” remains unknown[10].

In the recent issue of the World Journal of Cardiology, Grubić Rotkvić et al[11] published an interesting study that added data regarding the early stages of HF to the literature. Between 2019 and 2022, data was collected from two university hospitals in Zagreb, including information on biomarkers of inflammation, oxidative stress, myocardial stress, and echocardiographic parameters[11]. Despite the challenges posed by the Corona Virus Infectious Disease-2019 pandemic, the authors enrolled 64 patients with T2DM who were taking metformin[11]. Owing to suboptimal glycemic control, participants were divided into two groups to receive an additional antihyperglycemic drug, with one group receiving an SGLT2i and the other receiving a dipeptidyl peptidase-4 inhibitor (to serve as an active-controlled group)[11]. In both groups hypertension and obesity were the most common comorbidities. Nearly all patients had increased relative wall thickness and at least grade I diastolic dysfunction[10]. The only significant baseline differences included a higher blood pressure (both systolic and diastolic) and body mass index in the SGLT2i group[11]. In a previous report, Grubić Rotkvić et al[12] observed a significant increase in hematocrit and a significant decrease in body mass index in patients who were prescribed an SGLT2i. Furthermore, imaging markers of left ventricular hypertrophy (LVH) and diastolic and systolic functions were significantly improved by SGLT2i; however, no changes in any biomarkers were observed[12].

In the present study, the authors further analyzed important clinical variables that were not significantly affected by SGLT2i or dipeptidyl peptidase-4 inhibitors. Two subgroups were identified for each variable of interest according to the baseline values, and regression analyses were performed to evaluate the predictors of improved outcomes. Notably, lower myeloperoxidase (MPO) and polymerase chain reaction levels were observed in patients with higher baseline values; however, the magnitudes of these reductions were not statistically different between the treatment groups. In contrast, both systolic blood pressure (SBP) and diastolic blood pressure decreased significantly more (both P < 0.001) in the subgroups with higher baseline values (> 131 mmHg and > 80.50 mmHg, respectively). The only significant difference between treatment groups was observed in the subgroup with a baseline SBP ≤ 131 mmHg, where SBP increased slightly upon treatment with empagliflozin (P < 0.001).

Regarding biomarkers, no differences in N-terminal B-type natriuretic peptide levels were observed in this study, regardless of the baseline measurements (P = 0.519) or treatment groups (P = 0.088). Conversely, MPO and high-sensitivity C-reactive protein (hsCRP) levels decreased to a greater extent in patients with a higher burden of inflammation (≥ 91 ng/mL and ≥ 1.91 mg/L, respectively). However, although the reduction in hsCRP levels reached statistical significance (P = 0.012), the decrease in MPO levels was not statistically significant (P = 0.052). Similar results were noted for global longitudinal strain (GLS), with significant improvements in patients with a GLS > -17.84% (P < 0.001), irrespective of the treatment method. Regression analysis was performed to evaluate the predictors of improved outcomes in this study. Notably, therapy with SGLT2i was found to be the only significant independent predictor of an increase in the stroke volume index. Finally, readers should be aware of important shortcomings of this study, such as limited sample size and follow-up.

Despite the large amount of data on SGLT2i, only a few mechanistic studies have focused on patients in early stages of HF (Table 1). According to the Universal Definition of HF, endorsed by both American and European HF scientific societies, patients with diabetes, hypertension, and obesity, once recognized as “Stage A HF”, are now identified as “at risk of HF”[10]. Three studies investigated the effects of SGLT2i on biomarkers of inflammation and oxidative stress in patients with T2DM[13-15]. Reactive oxygen species negatively modulate several signaling pathways resulting in cardiac hypertrophy, fibrosis, impaired calcium homeostasis, and systemic inflammation, leading to HF progression[16]. A small prospective study conducted by Iannantuoni et al[13] evaluated 15 T2DM patients receiving empagliflozin for 24 weeks. They reported reduced levels of hsCRP and MPO in this population, which was similar to the results found by Grubić Rotkvić et al[11]. Furthermore, they reported that in this population, leukocytes expressed higher levels of interleukin-10 and glutathione and lower levels of superoxide production. Another recent retrospective study of 200 patients with T2DM reported consistent effects on CRP levels after 12 months of treatment with SGLT2i[14]. In addition, Wang et al[15] conducted an RCT assigning T2DM patients to either dapagliflozin or a placebo and found that after 1 year of follow-up, dapagliflozin significantly reduced interleukin-1 levels and limited the peripheral blood mononuclear cell maximal oxygen consumption rate, a novel marker of inflammation. On the other hand, five studies assessed the impact of SGLT2i on key echocardiographic measures enrolling patients with overt LVH or diastolic dysfunction. Both structural heart disease and diastolic dysfunction detect patients with pre-HF (prior stage B HF)[10]. An analysis of the DAPA-LVH RCT revealed that a reduction in CRP levels had a low correlation with improved GLS[17]. Conversely, in another DAPA-LVH sub-study, a reduction in the left ventricular mass index was independent of anti-inflammatory and anti-fibrotic effects and reverse remodeling was not linked to the baseline neutrophil-to-lymphocyte ratio[18]. Furthermore, in the ELUCIDATE study[19], the levels of transforming growth factor B did not differ significantly between the patients who received dapagliflozin and those who received standard care. Finally, in two RCTs evaluating the effect of dapagliflozin and empagliflozin on left ventricular function in patients with T2DM, the filling pressures at rest and after exercise improved early and significantly after treatment[20,21].

Table 1 Main studies evaluating the effects of sodium-glucose cotransporter 2 inhibitors on biomarkers and echocardiographic parameters in patients at risk for or with pre-heart failure.

Ref.	Year	Study design	Cohort	Mean HbA1c	Prevalent SGLT2i	Main outcomes
Iannantuoni et al[13]	2019	Single-center, prospective, observational	15 T2DM patients	After 6 months, 8% to 7%	Empagliflozin	After 6 months, hsCRP decrease1, MPO decrease1, IL-10 increase1, GSR increase1, CAT mRNA increase1, GSH increase1
Brown et al[17]	2021	Single-center, prospective, randomized, double-blind, placebo-controlled	47 T2DM patients with LVH	NA	Dapagliflozin	After 12 months, GLS increase1
Rau et al[21]	2021	Single-center, prospective, double-blind, placebo-controlled, parallel groups, randomized	42 T2DM patients	After 3 months, 7.5% to 7.1%	Empagliflozin	After 3 months, E decrease1, E/A decrease1, E/e’ decrease1
Shim et al[20]	2021	Single-center, prospective, randomized, double-blind, placebo-controlled	58 T2DM patients with at least grade I diastolic dysfunction	NA	Dapagliflozin	After 6 months, DR25W increase1, DR50W increase1, e’25W increase1, E/e'50W decrease1
Lin et al[19]	2024	Single-center, prospective, open-label, randomized, active-controlled	76 T2DM patients	After 6 months, 7.8% to 7.1%	Dapagliflozin	After 6 months, LVESV decrease1, LVMI decrease1, LVDd decrease1
Gherbon et al[14]	2024	Single-center, retrospective	200 T2DM patients	After 12 months, 8.4% to 7.3%	Dapagliflozin, Empagliflozin	After 12 months, CRP decrease1, SBP decrease1, DBP decrease1, NT-proBNP decrease1
Wang et al[15]	2024	Single-center, prospective, randomized, double-blind, placebo-controlled	62 T2DM patients	NA	Dapagliflozin	After 12 months, IL-1B decrease1, PBMC decline1
Dihoum et al[18]	2024	Single-center, prospective, randomized, double-blind, placebo-controlled	60 T2DM patients with LVH	After 12 months, 7.8% to 7.2%	Dapagliflozin	After 12 months, CRP decrease1, IL-1B decrease, LVM decrease1, GLS increase1
Grubić Rotkvić et al[11,12]	2024	Two-center, prospective, non-randomized	64 T2DM patients	After 6 months, 8.5% to 6.7%	Dapagliflozin, Empagliflozin	After 6 months, decrease IVST decrease1, E/e’ decrease1, decT decrease1, SVi increase1, GLS increase, hsCRP and MPO decrease

¹P < 0.05.

HbA1c: Hemoglobin A1c; SGLT2i: Sodium-glucose cotransporter 2 inhibitors; T2DM: Type 2 diabetes mellitus; hsCRP: High-sensitivity C-reactive protein; MPO: Myeloperoxidase; IL: Interleukin; GSR: Glutathione S-reductase; CAT: Catalase; GSH: Glutathione; LVMI: Left ventricular mass index; GLS: Global longitudinal strain; DR: Diastolic reserve; LVESV: Left ventricular end-systolic volume; LVDd: Left ventricular diastolic diameter; SBP: Systolic blood pressure; DBP: Diastolic blood pressure; NT-proBNP: N-terminal B-type natriuretic peptide; PBMC: Peripheral blood mononuclear cell; LVH: Left ventricular hypertrophy; LVM: Left ventricular mass; IVST: Interventricular septum thickness; decT: Deceleration time; SVi: Stroke volume index; NA: Not applicable.

Despite advances in pharmacological therapy for symptomatic HF, evidence for asymptomatic patients is poor and strictly limited to soft outcomes such as biomarkers and echocardiographic measures. In addition, SGLT2i are currently not recommended for pre-HF patients without T2DM. Therefore, further studies are required to provide mechanistic insights into the early stages of HF in patients with T2DM and evaluate the optimum treatment methods for these patients, especially considering the significant clinical and financial implications of HF treatment in its early stages.