Author + information
- Received November 27, 2017
- Revision received May 6, 2018
- Accepted May 6, 2018
- Published online August 9, 2018.
- Abhinav Sharma, MDa,b,c,∗ (, )
- Lauren B. Cooper, MDd,
- Mona Fiuzat, PhDa,
- Robert J. Mentz, MDa,
- João Pedro Ferreira, MD, PhDe,f,
- Javed Butler, MD, MPH, MBAg,
- David Fitchett, MDh,
- Alan Charles Moses, MDi,
- Christopher O’Connor, MDa,d and
- Faiez Zannad, MD, PhDe
- aDuke Clinical Research Institute, Duke University, Durham, North Carolina
- bDivision of Cardiology, Mazankowski Alberta Heart Institute, University of Alberta, Edmonton, Alberta, Canada
- cDivision of Cardiology, Stanford University, Palo Alta, California
- dInova Heart and Vascular Institute, Falls Church, Virginia
- eCentre d’Investigations Cliniques Plurithématique 1433, Institut national de la santé et de la recherche médicale, Université de Lorraine, CHRU de Nancy and F-CRIN INI-CRCT, Nancy, France
- fCardiovascular Research and Development Unit, Department of Physiology and Cardiothoracic Surgery, Faculty of Medicine, University of Porto, Porto, Portugal
- gDivision of Cardiology, Stony Brook University, Stony Brook, New York
- hDivision of Cardiology, St. Michael’s Hospital, University of Toronto, Toronto, Ontario, Canada
- iNovo Nordisk A/S, Soborg, Denmark
- ↵∗Address for correspondence:
Dr. Abhinav Sharma, Duke Clinical Research Institute, 2400 Pratt Street, Durham, North Carolina 27705.
There is increasing recognition of the relationship between diabetes and heart failure (HF). Comorbid diabetes is associated with worse outcomes in patients with HF, and death from HF forms a large burden of mortality among patients with diabetes and atherosclerotic cardiovascular disease. However, there is evidence of harm relating to the risk of HF outcomes from several antihyperglycemic therapies. The absence of well-powered randomized controlled studies has resulted in significant treatment variations in the glycemic management in patients with coexisting diabetes and HF. However, there is emerging evidence from recent clinical trials suggesting that sodium-glucose–co-transporter-2 inhibitors may be used as a therapy to improve HF outcomes. In order to understand the current state of knowledge, we reviewed the evolving evidence of antihyperglycemic therapies and present strategies to optimize these therapies in patients with diabetes and HF. This analysis is based on discussions among scientists, clinical trialists, industry sponsors, and regulatory representatives who attended the 12th Global Cardiovascular Clinical Trialists Forum, Washington DC, December 1 to 3, 2016.
Diabetes is now recognized as one of the most common comorbidities among patients with heart failure (HF) (1,2). Death from HF forms a large burden of mortality among patients with diabetes and atherosclerotic cardiovascular (CV) disease (3). Diabetes poses a more than 2-fold increased risk of HF development (4). Furthermore, comorbid diabetes is associated with worse outcomes in patients with HF (5,6). Recent advances in our understanding of the underlying pathophysiology in patients with diabetes and HF have resulted in emerging paradigms of disease mechanisms (Figure 1) (1,7). There are several mechanisms of HF development in the diabetic population, including chronic hyperglycemia, insulin resistance, and the nonglucose components of the metabolic syndrome that are prevalent in patients with type 2 diabetes. Of those non–glucose-associated comorbidities, hypertension, dyslipidemia, and obesity may all contribute to the development of HF (8,9).
Despite the major public health consequences associated with diabetes and HF, there is conflicting evidence of increased HF hospitalization risk within and among various classes of antihyperglycemic therapies (10–12). Evaluation of strategies to optimize outcomes have not been adequately explored, as patients with HF are frequently excluded from antihyperglycemia drug trials (13). However, several antihyperglycemic therapies may reduce the risk of HF and CV events. In order to address these knowledge gaps, we reviewed the current evidence regarding the risks and benefits of antihyperglycemic therapies in patients with diabetes and HF, and we highlighted potential strategies to improve HF outcomes by using antihyperglycemic therapies. This analysis is based on discussions among scientists, clinical trialists, industry sponsors, and regulatory representatives who attended the 12th Global Cardiovascular Clinical Trialists Forum, Washington, DC, December 1 to 3, 2016.
Recognition of HF Hospitalization as an Important Event in AntiHyperglycemic Drug Trials
In December 2008, the U.S. Food and Drug Administration (FDA) issued guidance to pharmaceutical sponsors setting out updated expectations for on-going development of antihyperglycemic drugs (14). The primary focus of the guidance was to direct sponsors to ensure the cardiovascular (CV) safety of antihyperglycemic therapies. Prior to the guidance, approval for antihyperglycemic therapies focused on glycemic efficacy, namely the reduction of hemoglobin A1c (HbA1c). In addition, safety data were limited to outcomes derived from short-term, 6- and 12-month phase 2 and 3 randomized controlled trials. However, 2 meta-analyses identified concerns for CV safety for 2 classes of antihyperglycemic therapies: muraglitazar (the investigational dual-peroxisome proliferator-activated receptor [PPAR]-alpha and -gamma agonist, which was never approved) (15) and the FDA-approved agent rosiglitazone, a thiazolidinedione (TZD) (16). As a result of these controversial studies, the FDA and, subsequently, the European Medicines Agency mandated long-term CV safety trials as a requirement for obtaining approval of new antihyperglycemic therapies. The meta-analysis that initially suggested an increased risk of CV outcomes primarily focused on myocardial infarction [MI] and CV death. FDA guidance mandated that sponsors conduct CV outcome trials to demonstrate that antihyperglycemic therapies did not primarily increase the risk of CV major adverse cardiac event(s) (MACE), primarily focusing on a composite of CV death, MI, or stroke (14). The guidance indicates that other relevant CV events (hospitalization for acute coronary syndromes or urgent revascularization) could be considered. However, HF as a CV safety event was not suggested in the guidance. Furthermore, although the guidance mandated that patients at high risk of CV events be enrolled (including patients with relatively advanced disease, elderly patients, and patients with some degree of renal impairment), there was no requirement to enroll patients with HF. As a result of the guidance, there has been a significant increase in the number of CV outcome trials for antihyperglycemic therapies, yet none has included HF in the primary endpoint. HF is a key outcome of interest because the demonstration of an increased risk of HF hospitalization associated with TZDs (17). In more recent trials, the possibility of increased risk of HF hospitalization associated with some dipeptidyl peptidase-4 (DPP-4) inhibitors (saxagliptin and alogliptin but not sitagliptin) has reaffirmed the importance of HF outcomes among antihyperglycemia drug trials (18). The emergence of antihyperglycemic therapies, namely sodium-glucose co-transporter-2 inhibitors, that may reduce the risk of HF outcomes has resulted in significant interest in how to use these therapies as a strategy to reduce the risk of HF hospitalizations (8).
Risks and Benefits of AntiHyperglycemic Therapies in Patients With Diabetes and HF
Inclusion of patients with heart failure in antihyperglycemia drug trials.
Clinical trials of antihyperglycemic therapies often excluded patients with HF, and 33% of anti-hyperglycemic drug trials did not have a stated definition for HF events (13). Previous trials also frequently excluded all patients with any symptoms of HF or those receiving treatment for HF (Online Table 1) (19–25). Although recent CV safety trials have included more precise HF definitions, patients with more severe HF symptoms, typically those with New York Heart Association (NYHA) functional classes III to IV, were excluded (26–29). However, more recent antihyperglycemia trials typically did not have any specific HF exclusion (except for the EXAMINE [Examination of Cardiovascular Outcomes with Alogliptin versus Standard of Care] trial, which excluded patients with NYHA functional class IV) (30). Furthermore, the SUSTAIN-6 (Semaglutide in Subjects with Type 2 Diabetes) trial (31) and the LEADER (Liraglutide Effect and Action in Diabetes: Evaluation of Cardiovascular Outcome Results) trial (32) encouraged inclusion of HF patients by designating NYHA functional classes II to III and systolic/diastolic dysfunction as enrichment criteria.
Will antihyperglycemic therapies increase the likelihood of new-onset or recurrent heart failure?
Clinicians have a broad menu of antihyperglycemic medications that can be used as first-, second-, or third-line therapies. However, despite the extensive number of drugs available, the optimal therapies for use in patients with diabetes and HF have not been established due to lack of high-quality randomized trial data and conflicting signals of risk and benefit among and within classes antihyperglycemic therapies (Figure 2).
Metformin and sulfonylureas
Randomized clinical trials suggest that metformin may reduce macrovascular events and that it is generally recommended as first-line treatment for patients with type 2 diabetes (19,33). No prospective randomized trial has evaluated whether metformin is the optimal first-line agent in patients with diabetes and HF. Cohort and administrative database analyses in patients with diabetes and HF suggest that metformin alone or in combination is associated with a lower mortality than sulfonylurea therapy (34,35).
Sulfonylureas improve glycemia control by increasing insulin release and, unlike other classes of antidiabetic drugs, they do not cause sodium retention. However, as with metformin, there are no randomized clinical trials assessing the CV safety of these agents specifically in patients with HF. In patients with newly diagnosed diabetes, the UKPDS (United Kingdom Prospective Diabetes Study) trial suggested that the combination of insulin and sulfonylurea versus diet-based treatment did not increase HF risk (hazard ratio [HR]: 0.91; 95% confidence interval [CI]: 0.54 to 1.52) (19). However, population-based studies have suggested the possibility of increased risk of HF hospitalizations associated with sulfonylureas compared with metformin (36).
TZDs work by improving insulin sensitivity, improving blood pressure control, optimizing lipid profiles, and potentially reducing the development of atherosclerosis. Although a 2007 meta-analysis suggested that, compared with placebo, rosiglitazone increased the odds for MI (odds ratio [OR]: 1.43; 95% CI: 1.03 to 1.98) and demonstrated a trend toward increased risk of death (OR: 1.64; 95% CI: 0.98 to 2.64), subsequent analyses have suggested no increased MI risk associated with use of rosiglitazone (37). With regard to HF outcomes, several studies have suggested that increased HF risk is associated with use of TZD. In a small randomized controlled trial of 224 patients with diabetes and HF with reduced ejection fraction (HFrEF), compared with placebo, rosiglitazone was associated with an increased risk of new or worsening peripheral edema and an increased use of HF medications associated with rosiglitazone (38). The PROACTIVE (PROspective pioglitAzone Clinical Trial In macroVascular Events) trial demonstrated a risk of HF hospitalizations associated with pioglitazone that was increased compared with placebo (pioglitazone 6%, placebo 4%, respectively; p = 0.007) (25). The RECORD (Rosiglitazone evaluated for cardiovascular outcomes in oral agent combination therapy for type 2 diabetes) trial demonstrated a doubling of fatal and nonfatal HF in patients receiving rosiglitazone (2.7% vs. 1.3%, respectively; HR: 2.1; 95% CI: 1.35 to 3.27) (22). Furthermore, in the RECORD trial, of the 61 HF cases treated with rosiglitazone, 4 patients experienced the initial HF event as a fatal event, and 30% of the surviving patients died during trial follow-up. These events were significantly increased compared with those in the control group in which 29 patients were hospitalized for HF; none of these cases were fatal initially, but 28% of the patients subsequently died. These data demonstrate that TZD-induced HF carries significant prognostic importance (17). As a result of these studies, diabetes and HF guidelines recommend not using TZD in patients with signs and symptoms of congestive HF and that initiation of these therapies is contraindicated in patients with NYHA functional classes III to IV HF (39,40). If congestive heart failure is newly diagnosed or considered likely, even in the absence of prior left ventricular dysfunction, the use of the TZD should be reconsidered (40).
Insulin has a dose-dependent antinatriuretic effect, and mild fluid retention may be seen with use of insulin, particularly in individuals with poorly controlled glucose levels at the time of initiation (Online Ref. 1); however, unlike TZDs, it is unclear whether insulin may actually increase the risk of adverse HF events (Online Refs. 2,3). In the BARI-2D (Bypass Angioplasty Revascularization Investigation 2 Diabetes) study, insulin therapy did not result in any significant differences in HF outcomes compared with metformin and TZDs therapies (20). The ORIGIN (Outcome Reduction With Initial Glagine Intervention) trial randomized 12,537 patients with dysglycemia (defined as impaired glucose tolerance, impaired fasting glucose, or diabetes) to the basal insulin glargine or placebo therapy. Overall, insulin therapy was not associated with increased HF risk (HR: 0.9; 95% CI: 0.77 to 1.05) (23). More recent data from the ORIGIN trial suggest that insulin does not increase the risk of recurrent HF events (Online Ref. 4). Despite the lack of randomized evidence suggesting harm for HF outcome, guidelines have encouraged caution in the use of insulin in patients with HF (Online Ref. 5).
There are 3 placebo-controlled randomized controlled clinical trials that evaluated the safety of DPP-4 inhibitors in patients with type 2 diabetes (30, Online Refs. 6,7). In the SAVOR-TIMI 53 (Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus-Thrombolysis In Myocardial Infarction 53) trial (Online Ref. 6), 16 of 492 patients with type 2 diabetes at high risk of CV events were randomized to receive saxagliptin or placebo therapy. Saxagliptin was associated with an increased risk of HF hospitalization (HR: 1.27; 95% CI: 1.07 to 1.51; p = 0.007) (Online Ref. 8). The EXAMINE study randomized 5,380 patients within 15 to 90 days after MI to alogliptin or placebo therapy. Alogliptin had no impact on the composite event of CV death and hospitalization for HF (HR: 1.00; 95% CI: 0.82 to 1.21) (8,30). For patients with no baseline HF history, there was a significant increase in HF risk associated with alogliptin (2.2% vs. 1.3%, respectively; HR: 1.76; 95% CI: 1.07 to 2.90) (8). In the TECOS (Trial Evaluating Cardiovascular Outcomes with Sitagliptin) study, 14,671 subjects with type 2 diabetes and established atherosclerotic cardiovascular disease were randomized to receive sitagliptin or placebo therapy. Sitagliptin did not increase hospitalization for HF (HR: 1.00; 95% CI: 0.83 to 1.20) (9). Recently the FDA released a warning for drugs containing saxagliptin or alogliptin for HF risk.
A small mechanistic study of the DPP-4 inhibitor vildagliptin was conducted in patients with diabetes and HF. The VIVIDD (Vildagliptin in Ventricular Dysfunction Diabetes) study randomized 254 patients to receive the DPP-4 inhibitor vildagliptin versus placebo for 52 weeks (Online Ref. 9). The inclusion criteria included left ventricle ejection fraction (LVEF) of <35% and poorly controlled diabetes. The primary outcomes (change in LVEF) were similar between groups. The LV systolic and diastolic volumes were increased in the vildagliptin arm compared with those in the placebo groups. Caution is warranted in interpreting the findings of increased LV chamber volumes as the clinical relevance remains unclear. In totality, the risk of HF events was balanced between the groups.
Current society consensus statements have suggested that DPP-4 inhibitors be used with caution in patients with diabetes and HF (Online Ref. 10). The use of sitagliptin in patients with HF is likely acceptable given the absence of HF risk seen among patients with pre-existing HF (HR: 1.03; 95% CI: 0.72 to 1.36) (9).
Glucagon-like peptide-1 receptor agonists
Glucagon-like peptide-1 (GLP-1) is secreted by cells located in the distal intestine in response to ingestion of food. GLP-1 receptor stimulation in pancreatic beta-cells facilitates glucose-dependent insulin secretion in addition to suppression of glucagon release by alpha-cells (Online Ref. 11). The ELIXA (Evaluation of Lixisenatide in Acute Coronary Syndrome) trial evaluated the GLP-1 receptor agonist lixisenatide in patients with type 2 diabetes who had had an MI in the preceding 180 days (Online Ref. 52). Compared with placebo, lixisenatide did not significantly reduce the risk of the primary MACE (HR: 1.02; 95% CI: 0.89 to 1.17) and had no impact on HF hospitalizations (HR: 0.96; 95% CI: 0.75 to 1.23). Similar results were seen in patients with and without HF. The LEADER trial evaluated the CV safety of liraglutide in 9,340 subjects with established cardiovascular disease or CV risk factors (32). Liraglutide reduced the risk of the primary MACE outcome of CV death, nonfatal MI, and nonfatal stroke (HR: 0.87; 95% CI: 0.78 to 0.97). CV mortality was significantly reduced by 22% (HR: 0.78; 95% CI: 0.66 to 0.93). Liraglutide was associated with numerically fewer HF hospitalizations, but the difference was not statistically significant (218 [4.7%] vs. 248 [5.3%], respectively; HR: 0.87; 95% CI: 0.73 to 1.05). The SUSTAIN-6 trial, randomized 3,297 patients with diabetes and established CV disease or CV risk factors to semaglutide versus placebo therapy (31). The trial demonstrated the noninferiority of semaglutide versus placebo for the primary MACE outcome (HR: 0.74; 95% CI: 0.58 to 0.95; p for noninferiority: <0.001). Semaglutide did not statistically increase the risk of HF events (vs. placebo: 3.6% vs. 3.3%, respectively; HR: 1.11; 95% CI: 0.77 to 1.61). The EXSCEL (Exenatide Study of Cardiovascular Event Lowering) trial evaluated the CV safety of exenatide versus placebo in patients with diabetes at high risk for CV; overall, the study demonstrated noninferiority for the primary MACE outcome (HR: 0.91; 95% CI: 0.83 to 1.00). There was no increased risk of HF seen in patients randomized to exenatide (HR: 0.94; 95% CI: 0.78 to 1.13) (Online Ref. 13).
Despite the apparent safety of GLP-1 receptor agonists in patients with HF, divergent results arose when GLP-1 receptor agonists were evaluated specifically among patients with established HF. The FIGHT (Functional Impact of GLP-1 for Heart Failure Treatment) study (Online Ref. 14) randomized 300 patients with and without diabetes with reduced EF (LVEF ≤ 40%) to receive either liraglutide or placebo. Patients were also required to have had a recent HF hospitalization (within 14 days) and a pre-admission oral dose of diuretic such as furosemide, at least 40 mg or an equivalent. The primary endpoint was a global rank score across 3 hierarchical tiers: time to death, time to HF rehospitalization, and time-averaged proportional change in N-terminal pro–B-type natriuretic peptide level from baseline to 180 days. Compared with placebo, liraglutide had no significant effect on the primary endpoint (p = 0.31). However, the point estimates suggested higher risk of death or HF-related events with liraglutide in patients with diabetes (vs. placebo; 47% vs. 34%, respectively; HR: 1.54; 95% CI: 0.97 to 2.46). The LIraglutide on left LIVE (VEntricular function in chronic heart failure patients) study randomized 241 patients with and without diabetes and HFrEF (LVEF ≤45%) to receive liraglutide or matching placebo for 24 weeks (Online Ref. 15). The primary outcome measurement (change in LVEF from randomization to end of follow-up) did not differ between the liraglutide and the placebo group; however, increased adverse cardiac events (death caused by ventricular tachycardia, nonfatal ventricular tachycardia, atrial fibrillation requiring intervention, aggravation of ischemic heart disease, and worsening of HF) were seen in 12 patients (10%) treated with liraglutide compared with 3 patients (3%) in the placebo group (p = 0.04).
Reasons for divergent signals of risk in patients with HFrEF seen in the FIGHT and LIVE studies compared with those in the larger LEADER, ELIXA, SUSTAIN-6, and EXSCEL trials remain unclear. A higher risk HF patient population might have had potentially differential responses to GLP-1 receptor agonist compared with the trial populations enrolled in the CV safety studies. It is unclear whether signals of risk would emerge for patients with HFpEF. Further research will be needed to ascertain the safety of liraglutide and other GLP-1 receptor agonists in patients with established HFrEF. Despite these results, caution in interpretations across trials should be considered as these trials enrolled different patient populations and used different trial endpoints. For instance, the SUSTAIN-6 trial had a much smaller population than the LEADER population and had a shorter follow-up.
Sodium-glucose co-transporter-2 inhibitor
Sodium-glucose co-transporter (SGLT)-2 facilitates glucose and sodium movement across cell membranes in the proximal renal tubule. Inhibition of SGLT-2 results in insulin-independent improvements in glycemic control due to glycosuria of approximately 70 to 80 g/day (Online Refs. 16,17.). The EMPA-REG OUTCOME (Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes) trial was a CV safety trial of the SGLT-2 inhibitor empagliflozin (Online Ref. 18). The trial randomized 7,020 patients with type 2 diabetes and established CV disease to receive 10 or 25 mg of empagliflozin or placebo. Empagliflozin reduced the primary MACE endpoint compared with that with placebo therapy (10.5% vs. 12.1%, respectively; HR: 0.86; 95% CI: 0.74 to 0.99). Furthermore, empagliflozin reduced the risk of HF admissions compared with placebo (4.1% vs. 2.7%, respectively; HR: 0.65; 95% CI: 0.50 to 0.85). Among the patients with a baseline history of HF, empagliflozin therapy was associated with a numerically lower rate of HF hospitalization (10.4% vs. 12.3%, respectively; HR: 0.75; 95% CI: 0.48 to 1.19) and CV mortality (8.2% vs. 11.1%, respectively; HR: 0.71; 95% CI: 0.43 to 1.16) (Online Ref. 19). Adverse events consistent with congestive HF such as edema were reported in a higher proportion of patients treated with placebo (216 of 2,333 patients [9.3%]) than with empagliflozin (9.3% vs. 4.5%, respectively).
The CANVAS (Canagliflozin Cardiovascular Assessment Study) program integrated 2 clinical trials with a total of 10,142 patients with type 2 diabetes and high CV risk. Patients were randomized to receive canagliflozin or placebo, and the trial demonstrated a significant reduction in the risk of CV death, nonfatal MI, or nonfatal stroke (26.9 vs. 31.5, respectively, per 1,000 patient-years; HR: 0.86; 95% CI: 0.75 to 0.97). No interaction was seen between patients with and without a baseline history of HF (interaction p = 0.51). An unexpected finding of an increased risk of toe or metatarsal amputation was identified (6.3 vs. 3.4, respectively, per 1,000 patient-years; HR: 1.97; 95% CI: 1.41 to 2.75). Randomization to canagliflozin was associated with a reduced risk of HF hospitalization (5.5 vs. 8.7, respectively, per 1,000 patient-years; HR: 0.67; 95% CI: 0.52 to 0.87). Furthermore, patients with a history of HF appear to have derived a greater magnitude of benefit from canagliflozin with regard to reduction in the risk of CV death and HF hospitalization than patients without a history of HF (Online Ref. 20). Similar results have been seen in population-level studies (Online Ref. 21). The ability of SGLT-2 inhibitors to optimize volume status through glycosuria and also inhibit the sodium-hydrogen exchanger in the kidneys and the heart may result in a cascade of responses including increased natriuresis, decreased myocardial fibrosis, and increased cardiac contractility (Figure 3) (Online Refs. 22,23).
Optimizing Therapies in Patients With Diabetes and HF
The STENO-2 (Intensified Multifactorial Intervention in Patients With Type 2 Diabetes and Microalbuminuria) trial demonstrated that aggressive risk factor modification involving glycemic control, blood pressure, and lipid profile improved clinical outcomes among patients with type 2 diabetes mellitus (Online Ref. 24). Such multifactorial treatment strategies may improve outcomes in patients with diabetes and HF, yet those strategies have never been tested in a well-powered randomized controlled trial. As a result, there is little evidence to guide clinicians with regard to glycemic targets in patients with diabetes and HF. Current HF and diabetes guidelines recommend lowering HbA1c concentration to <7% in patients with HF (Class IIa, Level of Evidence: A to reduce microvascular complications, B to reduce macrovascular complications) (39, Online Refs. 5,25). Guidelines also suggest that less stringent glycemic targets goals (such as HbA1c <8% [64 mmol/mol]) may be appropriate for patients with a history of severe hypoglycemia, limited life expectancy, advanced microvascular or macrovascular complications, and extensive comorbid conditions (Online Ref. 25). HF may be considered a significant comorbid condition, given the poor prognosis associated with patients who have both diabetes and HF. Further research will be needed to determine optimal strategies to individualize glycemic control among patients with diabetes and HF. Recent guidelines have not specifically suggested treatment strategies beyond avoidance of TZDs in patients with diabetes and HF (39; Online Refs. 5, 25). The absence of such recommendations arises from a lack of robust randomized controlled trial evidence in this patient population. Metformin is generally recommended as the first-line agent due to the relative safety of these agents in patients with diabetes and HF. With SGLT-2 inhibitors, given the possible reduction in the risk of MACE and HF outcomes in patients with established HF, we recommend these as the second-line agents (Central Illustration). The increased risk of toe and metatarsal amputation with canagliflozin should be considered when using this therapy; however, the increased risk has not been seen with empagliflozin. Liraglutide reduces the risk of CV outcomes in patients with established atherosclerotic CV disease, but caution should be used with GLP-1 receptor agonists in patients with recent HF hospitalization, given the signals of increased risk of adverse events seen in the FIGHT trial. Other classes of antihyperglycemic therapies do not have proven mortality benefit and should be considered only when SGLT-2 inhibitors and liraglutide cannot be used. Finally, a team-based approach including a diabetologist and the patient’s primary care provider is essential to ensuring optimal continuity of care.
Given the results of the EMPA-REG OUTCOMES and the CANVAS trials, there is significant interest in exploring the role of SGLT-2 inhibitors in patients with HF, regardless of the presence of diabetes (Online Table 2). The EMPEROR-Reduced (EMPagliflozin outcomE tRial in Patients With chrOnic heaRt Failure With Reduced Ejection Fraction; NCT03057977) and EMPEROR-Preserved (EMPagliflozin outcomE tRial in Patients With chrOnic heaRt Failure With Preserved Ejection Fraction; NCT03057951) trials are 2 examples of CV outcome trials that are evaluating the impact of SGLT-2 inhibitors in HF patients. For the first time, many of these trials are evaluating HF hospitalization as part of the composite of primary outcomes. Furthermore, several mechanistic studies are evaluating the impact of SGLT-2 inhibitor on invasive hemodynamics, body sodium content, cardiac biomarkers, and even arrhythmia burden among patients with HF (Online Table 2). If these trials successfully demonstrate improved HF outcomes with use of SGLT-2 inhibitor, this would represent a significant shift in our therapeutic strategies to manage and treat HF patients.
Although these trial are currently underway, there are still several additional questions that warrant further evaluation. Although metformin is recommended as the first-line agent, given the absence of a clear mortality reduction signal, future research should evaluate SGLT-2 inhibitors as potential first-line therapy. Pragmatic strategy trials of mono- and dual-therapy antihyperglycemic drug therapies, strategy trials of new antihyperglycemic therapies compared with therapies that have mortality benefit (Online Ref. 26), and identifying optimal time of therapy initiation (e.g., during hospitalization vs. in outpatients) are also needed. Patients with stage D HF often cannot tolerate HF therapies; therefore, evaluation of the safety and efficacy of SGLT-2 inhibitors in this population is warranted.
Given the significant public health consequences of diabetes and HF, strategies to optimize and treat these patients are desperately needed. As the number of antihyperglycemic and HF therapies increases, pragmatic trials to help guide clinicians as to which therapies to use and when to initiate them will be needed. The encouraging results of the SGLT-2 inhibitor trials suggest that these therapies may become the cornerstone in the treatment of patients with diabetes and HF. The extensive knowledge gaps that continue to exist can be addressed through enhanced collaboration among academics, industries, and government in order to help improve the lives of patients with diabetes and HF.
Dr. Sharma has received support from Bayer-Canadian Cardiovascular Society, Alberta Innovates Health Solution, Bristol-Myers Squibb-Pfizer, Roche Diagnostics, and Takeda. Dr. Mentz has received support from U.S. National Institutes of Health (NIH) grants U10HL110312 and R01AG045551-01A1 and from Amgen, AstraZeneca, Bayer, Bristol-Myers Squibb, GlaxoSmithKline, Gilead, Luitpold, Medtronic, Merck, Novartis, Otsuka, and ResMed; has received honoraria from HeartWare, Janssen, Luitpold Pharmaceuticals, Merck, Novartis, ResMed, and Thoratec/St. Jude; and has served as an advisory board member for Amgen, Luitpold, Merck, and Boehringer Ingelheim. Dr. Butler has received support from NIH, European Union, and Patient Centered Outcomes Research Initiative (PCORI); and has consulted for Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Meyers Squibb, CVRx, Janssen, Luitpold, Medtronic, Novartis, Relypsa, Vifor, and ZS Pharma. Dr. Moses is an employee of Novo Nordisk; has received Continuing Medical Education (CME) from Boehringer; has consulted for Boehringer, Lilly, Amgen, Merck, Sanofi, Novartis, AstraZeneca; is a steering committee member for the EMPA REG Outcome trial; and has been data monitoring and safety board (DMSB) for the SUSTAIN 6 and PIONEER 6 studies. Dr. Zannad has received honoraria from Takeda Development Center. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
John R. Teerlink, MD, served as Guest Editor for this paper.
- Abbreviations and Acronyms
- confidence interval
- U.S. Food and Drug Administration
- heart failure
- hazard ratio
- major adverse cardiac event(s)
- New York Heart Association
- odds ratio
- peroxisome proliferator-activated receptor
- Received November 27, 2017.
- Revision received May 6, 2018.
- Accepted May 6, 2018.
- Dei Cas A.,
- Khan S.S.,
- Butler J.,
- et al.
- Sharma A.,
- Zhao X.,
- Hammill B.G.,
- et al.
- Sharma A.,
- Green J.B.,
- Dunning A.,
- et al.
- Cavender M.A.,
- Steg P.G.,
- Smith S.C.,
- et al.
- Lawson C.A.,
- Jones P.W.,
- Teece L.,
- et al.
- Sharma A.,
- Demissei B.G.,
- Tromp J.,
- et al.
- Jorsal A.,
- Wiggers H.,
- McMurray J.J.V.
- McGuire D.K.,
- Van de Werf F.,
- Armstrong P.W.,
- et al.
- Udell J.A.,
- Cavender M.A.,
- Bhatt D.L.,
- Chatterjee S.,
- Farkouh M.E.,
- Scirica B.M.
- Sharma A.,
- Bhatt D.L.,
- Calvo G.,
- Brown N.J.,
- Zannad F.,
- Mentz R.J.
- ↵U.S. Department of Health and Human Services, Food and Drug Administration, Center for Drug Evaluation and Research (CDER). Guidance for Industry Diabetes Mellitus — Evaluating Cardiovascular Risk in New Antidiabetic Therapies to Treat Type 2 Diabetes; December 2008. Available at: https://www.fda.gov/downloads/Drugs/Guidances/ucm071627.pdf. Accessed March 12, 2018].
- McMurray J.J.V.,
- Gerstein H.C.,
- Holman R.R.,
- Pfeffer M.A.
- Dormandy J.A.,
- Charbonnel B.,
- Eckland D.J.A.,
- et al.
- Raz I.,
- Wilson P.W.F.,
- Strojek K.,
- et al.
- Eurich D.T.,
- Weir D.L.,
- Majumdar S.R.,
- et al.
- Inzucchi S.E.,
- Masoudi F.A.,
- McGuire D.K.
- Tzoulaki I.,
- Molokhia M.,
- Curcin V.,
- et al.
- Stone J.C.,
- Furuya-Kanamori L.,
- Barendregt J.J.,
- Doi S.A.R.
- Dargie H.J.,
- Hildebrandt P.R.,
- Riegger G.A.J.,
- et al.
- Bozkurt B.,
- Aguilar D.,
- Deswal A.,
- et al.
- Nesto R.W.,
- Bell D.,
- Bonow R.O.,
- et al.,
- for the American Heart Association/American Diabetes Association
- Central Illustration
- Recognition of HF Hospitalization as an Important Event in AntiHyperglycemic Drug Trials
- Risks and Benefits of AntiHyperglycemic Therapies in Patients With Diabetes and HF
- Optimizing Therapies in Patients With Diabetes and HF
- Future Directions