SGLT2 inhibitor lowers HF hospitalization in both primary and secondary prevention T2DM patients

AHA 2018 – Chicago, IL, USA

News - Nov. 11, 2018

The Dapagliflozin Effect on Cardiovascular Events (DECLARE)-TIMI 58 Trial

Presented at the AHA congress 2018 by: Stephen Wiviott (Boston, MA, USA)

Introduction and methods

Patients with diabetes type 2 (T2DM) are at high risk for development of and complications from atherosclerotic CV events and heart failure (HF). Dapagliflozin is a selective SGLT2 inhibitor (SGLT2i) which blocks glucose and sodium resorption in the kidney. This results in a reduction of blood sugar, blood pressure and weight. Prior CV outcome trials with other SGLT2i’s have demonstrated reductions in CV and renal events, mostly in secondary prevention patients, with known atherosclerotic CVD (ASCVD). Some safety questions with SGLT2i in these other trials have been raised in relation to amputation, stroke and diabetic ketoacidosis (DKA).

The DECLARE TIMI-58 studied the effect of dapagliflozin 10 mg daily in both primary prevention (with multiple risk factors, n=10186) and secondary prevention patients (with established CVD, n=6974). Follow-up visits were performed in person every 6 months, and by telephone every 3 months. The primary safety endpoint was MACE (CVD/MI/ischemic stroke), and endpoints for both safety and efficacy were CVD or hospitalization for HF (HHF), and MACE. Median follow-up was 4.2 years (IQR: 3.9-4.4). To enroll, patients had to have a diagnosis of T2DM, with HbA1c 6.5-12% and CrCL ≥60 ml/min. Participants also had established ASCVD (ischemic heart disease, cerebrovascular disease, peripheral artery disease: secondary prevention population), or multiple risk factors for ASCVD (men ≥55 years and woman ≥60 years with at least one additional risk factor: dyslipidemia, hypertension, current tobacco use: primary prevention population). 17160 patients were randomized (1:1) at 882 centers in 33 countries. Mean eGFR was 85 ml/min/1.73m², which is higher than in most other such trials, due to a mandate that patients should have eGFR ≥60 ml/min/1.73m².

Main results

  • HbA1c showed a least square mean (LSM) difference of 0.42% (95%CI: 0.40-0.45), with lower values for dapagliflozin vs. placebo.
  • Dapagliflozin showed a weight reduction of 1.8 kg vs. placebo, which was maintained throughout the study (LSM difference: 1.8 kg, 95%CI: 1.7-2.0).
  • Systolic BP was reduced with dapagliflozin (LSM difference: 2.7 mmHg, 95%CI: 2.4-3.0), as was diastolic BP (LSM difference: 0.7 mmHg, 95%CI: 0.6-0.9).
  • The co-primary endpoints showed a reduction with dapagliflozin vs. placebo:

- CVD/HHF: 4.9% vs. 5.8% (HR: 0.83, 95%CI: 0.73-0.95, P-superiority: 0.005).

- MACE: 8.8% vs. 9.4% (HR: 0.93, 95%CI: 0.84-1.03, P-non-inferiority < 0.001, P-superiority: 0.17).

  • The first renal composite endpoint (40% reduction in eGFR, ESRD, renal or CV death) was reduced with dapagliflozin vs. placebo (4.3% vs. 5.6%, HR: 0.76, 95%CI: 0.67-0.87, P<0.001).
  • The secondary all-cause mortality endpoint was not significantly reduced with dapagliflozin (6.2% vs. 6.6%, HR: 0.93, 95%CI: 0.82-1.04, P=0.20).
  • When analyzing the different strata of enrollment, no significant interaction of existing ASCVD was seen for the effect of dapagliflozin on CV death/HHF (P-interaction: 0.99), nor on MACE (P-interaction: 0.25). Nor was there a significant interaction with a history of HF (P-interaction: 0.60).
  • Regarding safety events, significant reductions with dapagliflozin were seen for treatment-emergent serious adverse events (34.1 vs. 36.2%), major hypoglycemia (0.7 vs. 1.0%) and cancer of bladder (0.3 vs. 0.5%). Significant increases of treatment-emergent AEs leading to drug discontinuation (8.1 vs. 6.9%), DKA (0.3 vs. 0.1%) and genital infections (0.9 vs. 0.1%) were seen with dapagliflozin. No difference between treatment groups was seen in amputations and fractures.

A meta-analysis of the three CV outcome trials with SGLT2i’s was also performed (EMPA-REG OUTCOME, CANVAS program, DECLARE-TIMI 58).

  • Based on data of the three trials, an HR for MACE of 0.86 (95%CI: 0.80-0.93) for patients with existing ASCVD (P=0.0002) was seen. Based on data of the CANVAS program and DECLARE-TIMI 58, the HR was 1.00 (95%0.87-1.16, P=0.98) for individuals without existing CVD, but with multiple risk factors.
  • For CVD/HHF, an overall HR of 0.76 (95%CI: 0.69-0.84, P<0.0001) was seen for the secondary prevention population, and of 0.84 (95%CI: 0.69-1.01, P=0.06) for the primary prevention participants.


DECLARE TIMI-58 was the largest SGLT2i trial, which included both primary and secondary prevention patients. Dapagliflozin was found to reduce CVD/HHF and did not affect the rate of MACE. The reduction in CVD/HHF was consistent regardless of baseline ASCVD or HF. Dapagliflozin was safe and generally well-tolerated, although more genital infections and DKA were seen.

In the context of the three large CV outcomes, it can be concluded that SGLT2i have moderate benefits on atherosclerotic MACE that appear confined to those with established ASCVD. SGLT2i show rubust effects on reducing the risk of HF and renal outcomes, which do not appear to depend on baseline atherosclerotic risk, prior HF or renal function. The current data with dapaglifozin extend the benefit of SGLT2i to a broader population of patients for both primary and secondary prevention.


Discussant Javed Butler first considered some indisputable facts on T2DM, among which that patients are at high risk for mortality. Traditionally, risks of macrovascular and microvascular complications are distinguished, and there is also an increased risk for HF. All of these complications are associated with worse outcomes. Prevention and treatment of these complications should therefore be central to T2DM management.

He thought of DECLARE TIMI-58 as a well-conducted trial, which included the highest proportion of patients with risk factors, but without ASCVD. It confirmed and replicated data from other studies with SGLT2i, with regard to safety, and reductions in HbA1c, blood pressure and weight.

When considering all data of the three outcome trials, he noted that those with diabetes and established CVD represent a minority of the T2DM population. No persons with only T2DM, without other risk factors, were included. The modest reduction in MACE was limited to those with established ASCVD. He interprets the data as that SGLT2i data cannot be regarded in isolation from other T2DM trials and therapies. He is of the opinion that all CV complications are important, and that the distinction between microvascular and macrovascular complications is arbitrary, as they show biological overlap. Moreover, macrovascular events does not include HF, and CKD should not be grouped in microvascular complications. To emphasize his point, he showed the conclusion of a recent study by Rawshani et al (NEJM, August 2018) that showed that if 5 risk factors were controlled, the risk of acute MI was lower, but the HR for HF remained elevated.

In conclusion, he said that for patients similar to those studied in the SGLT2i trials, these drugs should be used for HF risk reduction irrespective of their effects on MACE outcomes. For patients not adequately studied thus far, such as T2DM without other risk factors, or with manifest HF, further data are required.

- Our reporting is based on the information provided at the AHA Scientific Sessions -

The DECLARE TIMI-58 was published simultaneously in NEJMThe meta-analysis was published in The Lancet

Facebook Comments


We’re glad to see you’re enjoying PACE-CME…
but how about a more personalized experience?

Register for free