The ARNI sacubitril/valsartan has shown superior efficacy compared with enalapril in reducing CV death and HF hospitalization in patients with HFrEF, with or without diabetes. Patients were uptitrated to sacubitril/valsartan 97/103 mg twice daily within 2 to 4 weeks. The beneficial effect of sacubitril/valsartan over enalapril was consistent for patients with and without diabetes and across the spectrum of baseline HbA1c. Angiotensin-converting enzyme inhibitors (ACE-I) were the first class of medications shown to reduce mortality and morbidity and improve symptoms in patients with HFrEF. There is no difference in efficacy between patients with and without diabetes. As renin–angiotensin–aldosterone system (RAAS) inhibitors increase the risk of hyperkalemia and may acutely alter kidney function, routine surveillance of serum creatinine and potassium levels is advised. However, misinterpretation of eGFR changes as injury as opposed to understanding that it is due to favorable hemodynamic effects at the glomerular level often leads to inappropriate discontinuation of these disease-modifying agents, which should be avoided.

Two randomized placebo-controlled trials have investigated the effect of an SGLT2 inhibitor compared with placebo added to optimal medical therapy (OMT) in patients with HFrEF with and without diabetes. The DAPA-HF (Dapagliflozin and Prevention of Adverse Outcomes in HF) trial included patients if they were in New York Heart Association (NYHA) class II–IV, an LVEF ≤40% despite OMT, and had elevated NT-ProBNP (in sinus rhythm ≥600 pg/mL, in atrial fibrillation ≥900 mg/mL, or ≥400 pg/mL if they had been hospitalized for HF within the previous 12 months). Patients with T1DM or an eGFR < 30 mL/min/1.73 m ² were excluded. Therapy with dapagliflozin 10 mg once daily versus a placebo reduced the risk for the primary outcome, a composite of worsening HF (hospitalization or an urgent visit resulting in intravenous therapy for HF) or CV death, by 26% (HR, 0.74; 95% CI, 0.65–0.85). In addition, dapagliflozin reduced all-cause mortality (HR, 0.83; 95% CI, 0.71–0.97) and improved symptoms, physical function, and quality of life in patients with HFrEF. All of the clinical benefits observed were independent of baseline diabetes status and background glucose-lowering therapy and consistent across the spectrum of HbA1c.

The EMPEROR-Reduced (Empagliflozin Outcome Trial in Patients with Chronic Heart Failure with Reduced Ejection Fraction) trial evaluated empagliflozin versus a placebo and included patients with HFrEF with and without diabetes, with NYHA class II–IV, and LVEF ≤40% despite OMT; an eGFR ≥ 20 mL/min/1.73 m ² ; and an elevated NT-ProBNP (ejection fraction [EF] ≤30% or EF ≤40% and HF hospitalization within 12 months, NT-ProBNP ≥600 pg/mL; EF 31%–35%, NT-ProBNP ≥1000 pg/mL; EF 36%–40%, NT-ProBNP ≥2500 pg/mL). Empagliflozin reduced the risk for the primary outcome, a composite of CV death and HF hospitalization, by 25% versus a placebo (HR, 0.75; 95% CI, 0.65–0.86). The effect of empagliflozin on the primary outcome was consistent across patients with and without diabetes at baseline ( Fig. 16.3 ).

Effect of empagliflozin on the primary endpoint in EMPEROR-Reduced (Empagliflozin Outcome Trial in Patients With Chronic HF With Reduced Ejection Fraction). Time to first event of either cardiovascular death or HF hospitalization in (A) patients with diabetes and (B) patients without diabetes. CI , Confidence interval; HR , hazard ratio.

Treatment with empagliflozin also improved quality of life. A meta-analysis of the DAPA-HF and EMPEROR-Reduced trials showed a consistent reduction in HF hospitalization or CV death and all-cause mortality by SGLT2-inhibitor treatment without significant heterogeneity between trials. The combined SGLT1 and-2 inhibitor sotagliflozin was investigated in patients with T2DM who were recently hospitalized for worsening HF, irrespective of their LVEF (SOLOIST-WHF trial). Patients with an eGFR < 30 mL/min/1.73 m ² were excluded. Sotagliflozin significantly reduced the risk for the composite primary outcome (CV death, HF hospitalization, or urgent visit for HF) by 33% compared with a placebo (HR, 0.67; 95% CI, 0.52–0.85). The treatment effect was consistent across the spectrum of LVEF. Thus the SGLT inhibitors dapagliflozin, empagliflozin, and sotagliflozin are recommended, in addition to OMT (with an ARNI/ACEi/ARB, beta-blocker, and MRA), in patients with HFrEF and diabetes to reduce CV death and HF hospitalization. Three studies have investigated whether SGLT2 inhibitors can be safely started in patients hospitalized for acute HF. The EMPA-RESPONSE-AHF trial randomized 80 patients with acute HF with (approximately one-third) and without diabetes to either empagliflozin or placebo for 30 days. Treatment with empagliflozin did not affect visual analog scale dyspnea, diuretic response, NT-ProBNP levels, and length of hospital stay but was safe, increased urinary output, and reduced a combined endpoint of worsening HF, rehospitalization for HF, or death at 60 days. In the SOLOIST-WHF trial mentioned above, 1222 patients with T2DM were randomized to sotagliflozin or placebo with a median follow-up of 9 months (trial stopped prematurely due to loss of sponsor support). Sotagliflozin therapy, initiated before or shortly after discharge, resulted in significantly fewer deaths from CV causes and hospitalizations and urgent visits for HF than a placebo, with no increase in acute kidney injury. The EMPULSE trial randomized 530 hospitalized patients with and without diabetes with a primary diagnosis of acute de novo or decompensated HF, regardless of LVEF when clinically stable, to receive either empagliflozin or a placebo. More patients treated with empagliflozin had a clinical benefit (win ratio, 1.36; 95% CI, 1.09–1.68) compared with a placebo. This effect was consistent for acute de novo and decompensated chronic HF and was observed regardless of LVEF or the presence of diabetes. In these trials, very few cases of euglycemic diabetic ketoacidosis were reported; still, physicians treating patients with diabetes with SGLT2 inhibitors in this setting should be aware of this rare but potentially serious complication.

In large CVOTs, the MRAs spironolactone or eplerenone have shown to reduce death and HF hospitalization in patients with HFrEF, with consistent results in patients with or without diabetes. Eplerenone is more specific for blocking aldosterone and, therefore, causes less gynecomastia. Caution should be exercised when using MRAs in patients with impaired kidney function and in those with serum potassium concentration >5.0 mmol/L.

No dedicated trials have specifically assessed the impact of beta-blockers on HF-related outcomes in patients with HF with HFrEF who also have diabetes. However, posthoc analyses from various studies indicate that beta-blockers provide comparable benefits for HF in patients with HFrEF, with or without diabetes by reducing all-cause death and hospitalization for HF in patients with HFrEF. Treatment benefits strongly support using beta-blockers with proven efficacy in patients with HFrEF and diabetes.

Although there is insufficient evidence regarding the efficacy of thiazide or loop diuretics in reducing CV outcomes in HF patients, these diuretics are effective in preventing and treating symptoms and signs of fluid congestion with no known differentiation in efficacy or recommendation for use by diabetes status.

The place of angiotensin-II receptor blockers (ARBs) in managing HFrEF has changed over the last few years. They are now recommended for patients who cannot tolerate either ARNI or ACE-I due to serious side effects. ARBs have similar treatment effects in patients with HFrEF with or without diabetes.

Ivabradine slows the heart rate by inhibiting the If channel in the sinus node and is therefore only effective in patients with sinus rhythm. Ivabradine reduced the combined endpoint of CV death or HF hospitalization irrespective of diabetes status.

There is no evidence to support the use of this fixed-dose combination therapy in all patients with HFrEF but rather limited to self-identified Black patients as per product labeling. An RCT conducted in self-identified Black patients with HFrEF showed that adding the combination of hydralazine and isosorbide dinitrate to conventional therapy (ACE-I, beta-blocker, MRA) reduced mortality and HF hospitalization in patients with HFrEF in NYHA class III–IV. The beneficial effects were consistent in patients with or without diabetes.

Digoxin may reduce the risk of HF hospitalization in patients with HFrEF treated with ACE-Is, irrespective of diabetes status.

Device therapies (implantable cardioverter defibrillator [ICD], cardiac resynchronization therapy [CRT], and CRT with an implantable defibrillator [CRT-D]) have similar efficacies and risks in patients with HFrEF with or without diabetes. These therapies should be considered according to treatment guidelines in the HFrEF population. Briefly, an ICD is recommended in patients who have recovered from a ventricular arrhythmia causing hemodynamic instability and who are expected to survive for more than 1 year with good functional status in the absence of reversible causes or unless the ventricular arrhythmia occurred within 48 hours after MI, to reduce the risk of sudden death and all-cause mortality. In primary prevention, an ICD is recommend to reduce the risk of sudden cardiac death and all-cause mortality in patients with symptomatic HFrEF (NYHA class II–III) and LVEF ≤35% despite ≥3 months of optimal medical therapy, provided they are expected to survive more than 1 year with good functional status. ICD implantation is not recommended within 40 days after MI and in patients with NYHA class IV with severe symptoms refractory to pharmacological therapy unless they are candidates for CRT, left ventricular assist device, or heart transplantation. CRT is recommended for symptomatic patients with HF in sinus rhythm with a QRS duration ≥130 ms and left or right (less evidence with lower class of recommendation) bundle branch block morphology (and LVEF ≤35% despite optimal medical therapy to improve symptoms and reduce mortality). In addition, CRT rather than RV pacing is recommended for patients with HFrEF, regardless of NYHA class or QRS width, who have an indication for ventricular pacing for high-degree AV block. In contrast, a CRT is not recommended in patients with a QRS duration <130 ms who do not have an indication for pacing due to high-degree AV block. Heart transplantation could be considered in end-stage HF, but a large, prospective study of transplanted patients indicated a decreased likelihood of 10-year survival in those with diabetes.

The strongest evidence to date for the treatment of HFpEF comes from trials involving SGLT2 inhibitors. The EMPEROR-Preserved trial included patients classified as NYHA class II to IV, with an LVEF > 40% and elevated NT-ProBNP levels (> 300 pg/mL in sinus rhythm; >900 pg/mL in atrial fibrillation). Patients with an estimated glomerular filtration rate (eGFR) < 20 mL/min/1.73 m² were excluded. Compared with a placebo, empagliflozin reduced the risk of the primary outcome—a composite of cardiovascular death or hospitalization for HF—by 21%, largely due to a 29% lower risk of hospitalization for HF. This effect was independent of diabetes status, and baseline HbA1c levels did not influence outcomes.

Similarly, the DELIVER trial involved 6263 patients who were also classified as NYHA class II–IV and had an LVEF > 40%, elevated NT-ProBNP levels (> 300 pg/mL in sinus rhythm; > 600 pg/mL in atrial fibrillation), and an eGFR ≥25 mL/min/1.73 m². Dapagliflozin reduced the primary outcome—a composite of worsening HF or CV death—by 18%, primarily driven by a decrease in hospitalizations for HF. This effect was also independent of diabetes status. A meta-analysis encompassing data from both DELIVER and EMPEROR-Preserved indicated that SGLT2 inhibitors significantly lowered the composite outcome of CV death and first hospitalization for HF compared with a placebo (HR, 0.80; 95% CI, 0.73–0.87). Consistent reductions were observed for both components: CV death (HR, 0.88; 95% CI, 0.77–1.00) and first hospitalization for HF (HR, 0.74; 95% CI, 0.67–0.83). The treatment effects on this primary outcome did not vary between different LVEF subgroups and among patients with or without diabetes.

The combined SGLT1 and SGLT2 inhibitor sotagliflozin was studied in patients with T2DM who had recently been hospitalized due to worsening HF, regardless of their LVEF status; notably, 20.9% had an LVEF ≥50%. Sotagliflozin led to a reduction in the risk of the primary composite outcome—CV death, HF hospitalization, and urgent visits for HF—by 33%, demonstrating consistent effects across various baseline LVEFs.

As is standard practice in other forms of HF, diuretics should be used to manage congestion.

The GLP-1 receptor agonist semaglutide has been tested in patients with HF and preserved ejection fraction in the STEP-HFpEF DM trial. In this trial, 616 participants with obesity-related HF (BMI ≥30 kg/m ² ) and T2DM and HFpEF (LVEF ≥45%) were randomized to a placebo or the once-weekly GLP-1 receptor agonist semaglutide (2.4 mg for 52 weeks). Primary outcomes were the change from baseline in the Kansas City Cardiomyopathy Questionnaire Clinical Summary Score (KCCQ-CSS) and the change in body weight. The mean change in KCCQ-CSS was 13.7 points with semaglutide and 6.6 points with a placebo (estimated difference, 7.3 points; 95% CI, 4.1–10.4; P < 0.001), and the mean percentage change in body weight was–9.8% with semaglutide and–3.4% with a placebo ( P < 0.01) ( Fig. 16.4 ). In addition, the confirmatory secondary endpoint of difference in change in 6-minute walk distance was 14.3 m in favor of semaglutide. In addition, semaglutide led to a significantly greater reduction in hsCRP. This trial demonstrates that among patients with obesity-related HFpEF and T2DM, treatment with semaglutide significantly reduced HF-related symptoms and physical limitations in addition to a greater weight loss compared with a placebo after 1 year. Data on the improvement of the prognosis of these patients are currently lacking.

Change from baseline to week 52 in the dual primary endpoints. (A) Change in KCCQ-CSS and (B) change in body weight. Shown are the observed (i.e., as-measured) mean changes from baseline in the Kansas City in the STEP-HFpEF DM trial Cardiomyopathy Questionnaire clinical summary score (KCCQ-CSS; scores range from 0 to 100, with higher scores indicating fewer symptoms and physical limitations) and percentage changes in body weight. I bars indicate the standard error, and the numbers below the graphs are the numbers of participants contributing to the mean (with variations in participants attending certain trial visits). The data at week 52* are the estimated mean changes from baseline to week 52 based on analysis of covariance ( ANCOVA ) and an imputation approach for missing data.

A similar trial has been conducted in patients with obesity-related HF without diabetes (STEP-HFpEF). A pooled analysis of both trials demonstrated that semaglutide treatment improved HF-related symptoms and physical limitations across subgroups who were versus were not treated with diuretics. Moreover, an additional analysis of these pooled trials reported that participants with a higher baseline NT-ProBNP had a similar degree of weight loss but experienced larger reduction in HF-related symptoms and physical limitations with semaglutide compared with those with NT-ProBNP.

Additional data on HF endpoints in patients with T2DM can be derived from clinical trials with the nonsteroidal mineralocorticoid receptor antagonist finerenone. Two dedicated cardiovascular outcome trials in patients with T2DM and chronic kidney disease (FIGARO DKD and FIDELIO DKD ) assessed the effect of finerenone on CV outcomes. The predefined analysis of both trials, FIDELITY, demonstrated that finerenone in combination with optimal RAAS-inhibition in patients with T2DM and CKD but without HFrEF reduced the combined CV outcomes of CV death, nonfatal MI, nonfatal stroke, or HF hospitalization significantly by 14% compared with a placebo. This effect was mainly driven by a reduction in HF hospitalization even though patients with HFrEF were excluded. The beneficial effect of finerenone on HF-related outcomes was not modified by baseline eGFR and/or UACR or the presence of HF at baseline. Overall, the prevalence of a history of HF in this patient population was 7.7%, suggesting that finerenone may be effective in the prevention of new-onset HF in T2DM and CKD.

The FINEARTS-HF (Finerenone in HF with Mildly Reduced or Preserved Ejection Fraction) trial assessed the effect of finerenone in patients with HF and mildly reduced or preserved ejection fraction. This trial enrolled patients with HF and a left ventricular ejection fraction of 40% or greater, with or without diabetes; 6001 patients were randomized to finerenone or a placebo, and the primary combined endpoint was CV death or total HF events (hospitalization/urgent visits). Though this trial was not a dedicated diabetes trial, about 40% of all patients had a medical history of T2DM. Overall, treatment with finerenone over a median follow-up of 32 months led a significant reduction of the primary endpoint with a rate ratio of 0.84 (95% CI, 0.74–0.95; P = 0.07). This effect was mainly driven by the reduction of the total number of worsening HF events with a relative risk reduction of 18% that was significant. There was a trend toward a reduction of cardiovascular death (HR, 0.93), which did not reach statistical significance. Of note, statistical significance for the primary endpoint was observed within 1 month of therapeutic initiation, a benefit that was sustained until the final follow-up. The beneficial effect of finerenone was comparable in patients with or without diabetes, suggesting that finerenone treatment is effective in reducing HF-related endpoints in patients with T2DM and HF with an ejection fraction of 40 or above. Of note, finerenone was associated with an increased risk of hyperkalemia. These findings were consistent across prespecified subgroups, including across LVEF and in those on SGLT-2 inhibitors.

Metformin is considered safe for use at all stages of HF in patients with preserved or stable, moderately reduced kidney function (i.e., eGFR > 30 mL/min/1.73 m²). Observational studies suggest that it is associated with a lower risk of death and HF hospitalization compared with insulin and sulfonylureas. However, dedicated randomized controlled CVOTs evaluating the safety and efficacy of metformin have yet to be conducted. Thus caution regarding metformin use in HF was lifted in 2006 based on data indicating no increased risk of lactic acidosis and potential clinical benefits. Current evidence supports the consideration of metformin for individuals with stable and compensated HF, particularly given the available CV outcomes data, low risk of hypoglycemia, affordability, and favorable tolerability profile.

Four DPP-4 inhibitors—sitagliptin, saxagliptin, alogliptin, and linagliptin—have been evaluated in dedicated placebo-controlled CV safety trials involving patients with T2DM who are either at high risk for ASCVD or already diagnosed. Saxagliptin was found to significantly increase the risk of HF hospitalization and is therefore not recommended for patients with diabetes who have or are at increased risk for HF. Alogliptin showed a nonsignificant trend toward an increase in HF hospitalizations. In contrast, sitagliptin and linagliptin demonstrated a neutral effect on HF outcomes. Vildagliptin has not been tested in a CVOT; however, it did not significantly affect LVEF but was associated with an increase in LV volumes in a small study. So if DPP-4 inhibitors are to be used, to avoid HF risk, priority should be given to sitagliptin or linagliptin.

SGLT2 inhibitors have been investigated in different populations with diabetes, ranging from patients with ASCVD or multiple ASCVD risk factors to patients recently hospitalized for worsening HF, with increasing absolute risk reductions for HF-related outcomes as patient risk increases.

As outlined above, in dedicated HF trials, dapagliflozin and empagliflozin reduced CV death and HF hospitalization in patients with HFrEF with or without diabetes, and sotagliflozin reduced CV death and HF hospitalization in patients with T2DM and recent hospitalization for HF of any etiology. Moreover, empagliflozin and dapagliflozin reduced the risk of CV death or HF hospitalization in patients with HFmrEF or HFpEF. While the EMPA-REG OUTCOME (empagliflozin) and VERTIS CV (ertugliflozin) trials investigated patients with T2DM and established ASCVD risk, the CANVAS Trials Program (canagliflozin) and DECLARE-TIMI 58 trial (dapagliflozin) included patients with established ASCVD or multiple ASCVD risk factors. In all of these placebo-controlled CVOTs of SGLT2 inhibitors, only a small proportion of patients had a baseline history of HF. Empagliflozin reduced the risk of HF hospitalization by 35% in patients with and without previous HF. Canagliflozin also significantly reduced the risk of HF hospitalization by 33%. Dapagliflozin significantly reduced the combined endpoint of CV death and HF hospitalization, a result driven mainly by lower rates of HF hospitalization. This effect was independent of preexisting HF. Ertugliflozin did not reduce the combined endpoint of CV death and HF hospitalization, although there was a significant reduction in HF hospitalization and repeated hospitalizations. In addition, four trials investigated the effect of SGLT2 inhibitors in patients with T2DM with moderate or worsening CKD (CREDENCE [canagliflozin] and SCORED [sotagliflozin]) with and without diabetes (DAPA-CKD [dapagliflozin] and EMPA-KIDNEY [empagliflozin]). In these patients at high risk for HF, a consistent risk reduction of CV death or HF hospitalization was observed, ranging from 23% to 31%. A meta-analysis of six outcome trials of four SGLT2 inhibitors in patients with T2DM (EMPA-REG OUTCOME, CANVAS Program [two trials], DECLARE-TIMI, CREDENCE, VERTIS CV) demonstrated a 32% reduction in HF hospitalization, with no heterogeneity between trials; the effect on HF hospitalization was independent of ASCVD.

Eight CVOTs have been conducted with GLP-1RAs in patients with T2DM, revealing a prevalence of established HF ranging from 9% to 24%. Most GLP-1RAs demonstrated a neutral effect on the risk of HF hospitalization in placebo-controlled RCTs assessing the CV safety of glucose-lowering medications in patients with T2DM, even though these agents increased their heart rate by 3 to 5 beats per minute. Additionally, a meta-analysis encompassing eight trials with a total of 60,080 patients found that GLP-1RAs reduced HF hospitalization compared with a placebo. The AMPLITUDE-O trial, which compared efpeglenatide to a placebo, showed a nominally significant reduction in hospitalizations for HF. This trial featured stratified randomization based on baseline or anticipated use of SGLT2 inhibitors and reported the highest prevalence (15.2%) of SGLT2 inhibitor use among GLP-1RA studies. An exploratory analysis from AMPLITUDE-O suggested that the efficacy and safety of efpeglenatide were independent of concurrent SGLT2 inhibitor use concerning HF hospitalization.

Two small RCTs have specifically evaluated GLP-1RAs in patients with HF with HFrEF. The LIVE trial randomly assigned 241 patients with chronic, stable HFrEF—both with and without diabetes—to receive either liraglutide or placebo. After 24 weeks, there were no changes observed in LVEF, quality of life, or functional class; however, serious adverse cardiac events were more prevalent in the liraglutide group (12 events [10%] vs. 3 events [3%] for placebo; P = 0.04). The FIGHT trial (Functional Impact of GLP-1 for HF Treatment) involved randomly assigning 300 patients—with and without diabetes—who had HFrEF and a recent hospitalization for HF to either liraglutide or a placebo. After 180 days of treatment, there was no significant difference between groups regarding the primary outcome measures: time to death, time to rehospitalization for HF, and average proportional change in NT-ProBNP levels from baseline to day 180. Furthermore, there was a nonsignificant difference in rehospitalizations for HF between groups (63 [41%] in the liraglutide group vs. 50 [34%] in the placebo group; HR, 1.30; 95% CI, 0.89–1.88; P = 0.17).

Additional data on HF outcomes in patients with T2DM are available from the FLOW trial. The FLOW trial was the first dedicated kidney outcome study, enrolling 3534 adults with CKD. Semaglutide resulted in a 24% reduction in the incidence of the primary kidney composite endpoint, which included kidney failure, a sustained 50% decline in eGFR, or CV or kidney death (HR, 0.76; 95% CI, 0.66–0.88). Additionally, an 18% relative risk reduction in 3P-MACE—comprising CV death, MI, or stroke—was observed (HR, 0.82; 95% CI, 0.68–0.98). These CV risk reductions were consistent with findings from the SELECT trial involving a non-CKD cohort at high CV risk, where a 20% reduction in 3P-MACE was reported.

Secondary analyses of the FLOW trial also indicated that semaglutide reduced the risk of HF events or CV death by 17%, regardless of baseline HF history, and showed no treatment heterogeneity across subgroups classified by KDIGO risk. Furthermore, the use of SGLT2 inhibitors alongside semaglutide was relatively low in the FLOW trial (15.6%), and secondary subset analyses demonstrated that concomitant SGLT2 inhibitor use did not significantly affect treatment outcomes on kidney and CV composite endpoints.

Secondary analyses from the SELECT trial, assessing the effect of semaglutide versus a placebo in 17,604 patients with a BMI ≥27 kg/m ² and ASCVD but no diabetes, suggest that treatment with semaglutide reduced MACE and composite HF endpoints in those with ( n = 4286) and without clinical HF ( n = 13,314), regardless of HF subtype.

Overall, based on the totality of current evidence, GLP-1RAs are safe and potentially beneficial in patients with T2DM and HF.

In patients with T2DM and advanced HF, the use of insulin is independently linked to a significantly worse prognosis. Two basal insulins have been formally evaluated in dedicated CVOTs.

In the ORIGIN trial, 12,537 participants (mean age 63.5 years) at high CV risk, including those with impaired fasting glucose, impaired glucose tolerance, or T2DM, were randomized to receive insulin glargine titrated to achieve a fasting blood glucose level of ≤95 mg/dL (≤5.3 mmol/L) or standard care. After a median follow-up of 6.2 years, insulin glargine showed a neutral effect on HF hospitalizations. The DEVOTE trial was a double-blind study comparing ultra-long-acting, once-daily insulin degludec with insulin glargine U100 and included 7637 patients with T2DM who had established ASCVD or were at high ASCVD risk. Treatment with insulin degludec did not result in any significant difference in HF hospitalization rates compared with insulin glargine; however, prior HF was independently associated with a higher risk of future HF hospitalization.

Thiazolidinediones increased the risk of HF hospitalization in several trials and are not recommended in patients with diabetes and symptomatic HF.

Data regarding the effects of sulfonylureas on HF are inconsistent. Two retrospective cohort studies involving 111,971 patients with diabetes suggest an adverse safety profile, indicating a 20% to 60% higher mortality rate and a 20% to 30% higher risk of HF compared with metformin. However, trials such as UKPDS, NAVIGATOR (Nateglinide And Valsartan in Impaired Glucose Tolerance Outcomes Research), and ADOPT (A diabetes outcome progression trial) did not show any increased signals for HF. Additionally, data from the CAROLINA trial, which compared linagliptin—shown not to increase the risk of HF hospitalization versus a placebo in the CARMELINA trial—with glimepiride indicated no elevated risk of HF hospitalization associated with this sulfonylurea.