From Wiki Journal Club
Jump to: navigation, search
Zinman B, et al. "Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes". The New England Journal of Medicine. 2015. 373(22):2117-28.
PubMedFull textPDF

Clinical Question

Among patients with T2DM at high risk for CV events, does daily empagliflozin 10 or 25 mg reduce CV mortality, nonfatal MI, or nonfatal strokes when compared to placebo?

Bottom Line

In this industry-sponsored trial, empagliflozin reduced the rate of CV events and slowed the progression of kidney disease among patients with T2DM and high CV risk.

Major Points

Excess mortality in T2DM is largely related to an increased incidence of CV disease.[1] In the 1998 UKPDS 34 study, metformin was associated with reduced all-cause mortality. The 2005 PROactive study not only failed to demonstrate a reduction in all-cause mortality with pioglitazone; the drug was associated with excess HF hospitalizations.[2] Subsequent studies of newer agents, including the DPP-4 inhibitor sitagliptin (TECOS, 2015[3]) and the GLP-1 inhibitor lixisenatide (ELIXA, 2015[4]), have also been unable to demonstrate a significant improvement in cardiovascular outcomes among selected patients with T2DM.

Empagliflozin is a glucose-lowering agent that inhibits the renal sodium glucose cotransporter-2 (SGLT-2 inhibitor) which confers several advantages including weight loss, reduction in blood pressure, and a low risk of hypoglycemia.[5] As the medication promotes glucosuria, it may lead to a greater incidence of GU infections and sepsis of urinary origin.

Published in 2015, the multicenter Empagliflozin, Cardiovascular Outcomes, and Mortality in Type 2 Diabetes trial (EMPA-REG OUTCOME) randomized 7,020 patients to daily empagliflozin 10 or 25 mg or placebo. At 3.1 years of follow up, empagliflozin was associated with a reduction in CV mortality, nonfatal MI, or nonfatal stroke (10.5% vs. 12.1%; P=0.04; NNT 62), as well as a reduction in all-cause mortality (5.7% vs. 8.3%; P<0.001; NNT 38) and CV mortality (3.7% vs. 5.9%; P<0.001; NNT 45). The medication was generally well tolerated. There were no differences in UTI or sepsis of urinary origin between the groups though there was a greater incidence of genital infections with the medication (6.4% vs. 1.8%; P<0.001; NNH 22). Empagliflozin was associated with reduction in weight and SBP. Follow-up data published in 2016 showed that patients on empagliflozin had lower rates of new or worsening renal dysfunction compared to placebo (12.7% vs. 18.8%; P<0.001).[6]

Interestingly, the level of HbA1c reduction with empagliflozin was modest at best (0.5%). Given that multiple previous clinical trials (ACCORD 2008, ADVANCE 2008, VADT 2009) did not demonstrate a reduction in major adverse cardiovascular events with intensive glucose-lowering therapy, it remains unclear what mechanisms drive the benefits seen with empagliflozin. The additional benefits of weight loss, blood pressure reduction, and diuresis may be contributing factors.[7] This theory is supported by the fact the reduction in cardiovascular deaths was partially driven by a reduction in deaths from heart failure, while rates of myocardial infarction remained similar.

The empagliflozin dosage groups were pooled for the analysis of empagliflozin versus placebo; however, the hazard ratios were similar for 10 mg and 25 mg for primary outcomes. It is reasonable to assume that the choice of dose in clinical practice will depend on the achievement of metabolic outcomes and tolerance of adverse effects.

Of note, the 2016 LEADER study demonstrated a CVD benefit with the GLP-1 medication liraglutide.


As of October 2015, no guidelines have been published that reflect the results of this trial.


  • Multicenter, randomized, double-blind, placebo-controlled trial
  • N=7,020
    • Empagliflozin 10mg po daily (n=2,345)
    • Empagliflozin 25mg po daily (n=2,342)
    • Placebo (n=2,333)
  • Setting: 590 sites in 42 countries
  • Enrollment: 2010-2013
  • Median follow-up: 3.1 years
  • Analysis: Intention-to-treat
  • Primary outcome: CV mortality, nonfatal MI, or nonfatal stroke


Inclusion Criteria

  • T2DM
  • Age ≥18 years
  • BMI ≤45 kg/m2
  • eGFR ≥30 mL/min/1.73 m2 by MDRD
  • CVD, defined by ≥1 of the following:
    • MI >2 months prior
    • Multivessel CAD
    • Single vessel CAD with positive stress test or UA hospitalization in prior year
    • UA >2 months prior and evidence of CAD
    • Stroke >2 months prior
    • Occlusive PAD
  • Either of:
    • No glucose-lowering agents in prior 12 weeks and A1c 7-9%
    • Stable glucose-lowering therapy in prior 12 weeks and A1c 7-10%

Exclusion Criteria

  • Uncontrolled hyperglycemia with fasting glucose >240 mg/dL (>13.3 mmol/L)
  • Liver disease with ALT, AST or ALP x3 ULN
  • Planned CV surgery or angioplasty in 3 mo
  • GFR <30 mL/min/1.73 m2 by MDRD
  • Prior surgery with chronic malabsorption (eg, bariatric) in prior 2 years
  • RBC disorders
  • Cancer
  • Anti-obesity drug treatment in prior 3 mo
  • Systemic steroids
  • Change in thyroid hormone dosage in prior 6 weeks
  • Nursing, pregnant, or child-bearing age women not using acceptable method of birth control or refusing pregnancy testing
  • Alcohol or drug abuse

Baseline Characteristics

From the pooled empagliflozin-based treatment arms.

  • Demographics: Age 63 years, female sex 29%
    • Race: White 73%, asian 22%, black 5%
    • Ethnicity: Hispanic or Latino 18%
    • Geographic region: Europe 41%, N. America 20%, Asia 19%, Latin America 15%
  • Health measurements: BMI 31 kg/m2, BP 135/77 mmHg
  • Laboratory: HbA1c: 8.1%, Tchol 164 mg/dL, LDL 86 mg/dL, HDL 45 mg/dL, TG 171 mg/dL
    • eGFR: 74 mL/min/1.73m^2
      • eGFR >=90: 22%
      • eGFR 60 to <90: 52%
      • eGFR <60: 26%
  • Urine albumin:creatinine:
    • <30mg/g: 60%
    • 30-300 mg/g: 29%
    • >300mg/g: 11%
  • PMH/PSH: CAD 76% (single vessel 11%, multivessel 47%), MI 47%, CABG 25%, stroke 23%, PAD 21%, heart failure 10%,
    • DM: Time since diagnosis:
      • <1y: 3%
      • 1-5y: 15%
      • 5-10y: 25%
      • >10y: 57%
  • Glucose-related medications: Metformin 74%, insulin 48% (median daily dose 54 units), SU 43%, DPP-4 inhibitor 11%, TZD 4%, GLP-1 agonist 3%
    • Monotherapy: 29%
    • Dual therapy: 48%
  • Other medications: ACE-inhibitor/ARB 81%, B-blocker 65%, diuretic 44%, CCB 33%, anti-aldosterone 7%, statin 77%, fibrate 9%, ezetimibe 4%, niacin 2%, ASA 83%, clopidogrel 11%, VKA 6%


  • Randomization in a 1:1:1 ratio to a group with stratification by A1c, BMI, renal function, and geographic region:
    • Empagliflozin 10 mg
    • Empagliflozin 25 mg
    • Placebo
  • All doses were taken daily after a 2 week open-label placebo run-in period
  • Modifications to non-study glucose-modifying medications were permitted after week 12 to reach local therapeutic guidelines
  • Other cardiovascular risk factors, including dyslipidemia and hypertension, were treated at the discretion of investigators to local guidelines


Comparisons are empagliflozin (pooled) vs. placebo. Noninferiority had a margin of 1.3 for the hazard ratio. P-values are for superiority except where specified.

Primary Outcomes

CV mortality, nonfatal MI, or nonfatal stroke
10.5% vs. 12.1% (HR 0.86; 95% CI 0.74-0.99; superiority P=0.04, noninferiority P<0.001; NNT 62)

Secondary Outcomes

CV mortality, nonfatal MI, nonfatal stroke, or UA hospitalization
12.8% vs. 14.3% (HR 0.89; 95% CI 0.78-1.01; superiority P=0.08, noninferiority P<0.001; NNT 67)
All-cause mortality
5.7% vs. 8.3% (HR 0.68; 95% CI 0.57-0.82; P<0.001; NNT 38)
CV mortality
3.7% vs. 5.9% (HR 0.62; 95% CI 0.49-0.77; P<0.001; NNT 45)
HF hospitalization
2.7% vs. 4.1% (HR 0.65; 95% CI 0.50-0.85; P=0.002; NNT 71)
HF hospitalization or CV mortality excluding fatal stroke
5.7% vs. 8.5% (HR 0.66; 95% CI 0.55-0.79; P<0.001; NNT 36)

A full list of secondary outcomes are presented in Table 1 of the main text. Outcomes that did not reach statistical significance for differences between groups include 1. MI excluding silent MI, 2. nonfatal MI excluding silent MI, 3. silent MI, 4. UA hospitalization, 5. coronary revascularization procedures, 6. strokes, 7. nonfatal strokes, 8. TIAs.

Additional Outcomes

Mean difference in A1c at compared to placebo
Baseline A1c was 8.5%.
At week 12
Empagliflozin 10 mg: -0.54% (95% CI -0.58% to -0.49%)
Empagliflozin 25 mg: -0.60% (95% CI -0.64% to -0.55%)
At week 94
Empagliflozin 10 mg: -0.42% (95% CI -0.48% to -0.36%)
Empagliflozin 25 mg: -0.47% (95% CI -0.54% to -0.41%)
At week 206
Empagliflozin 10 mg: -0.24% (95% CI -0.40% to -0.08%)
Empagliflozin 25 mg: -0.36% (95% CI -0.51% to -0.20%)
Mean difference in weight
Both doses of the medication were associated with about 2 kg of weight loss during follow up. Details are on page 57 of the supplementary appendix.[8]
Mean difference in BP
Both doses of the medication were associated with about 4 mmHg reduction in SBP during follow-up. It was associated with an initial 1-2 mmHg reduction in DBP though that was lost at about week 136. Details are on page 58-59 of the supplementary appendix.[8]
New medications added after baseline
In addition to baseline therapies, which are listed in baseline characteristics above.
Any: 19.5% vs. 31.5%
Insulin: 5.8% vs. 11.5%
DPP-4 inhibitor: : 5.6% vs. 8.3%
SU: 3.8% vs. 7.0%
Metformin: 3.7% vs. 4.8%
TZD: :1.2% vs. 2.9%
GLP-1 agonist: 1.4% vs. 2.4%

Subgroup Analysis

For the primary outcome.

<65: No difference
≥65: Favors empagliflozin
P-value for interaction 0.01
Hgb A1c
<8.5%: Favors empagliflozin
≥8.5%: No difference
P-value for interaction 0.01

There were no significant interactions for sex, race, BMI, BP control, eGFR, urine albumin:creatinine, CV risk, insulin use, statin use, antihypertensive therapy use, ACE-inhibitor or ARB use, beta-blocker use, or diuretic use.

Adverse Events

90.2% vs. 91.7% (P<0.001)
Severe: 23.5% vs. 25.4% (P<0.05)
Serious: 38.2% vs. 42.3% (P<0.001)
Death: 3.8% vs. 5.1% (P<0.01)
Leading to discontinuation: 17.3% vs. 19.4% (P<0.01)
Hypoglycemia: 27.6% vs. 27.9% (NS)
UTI: 18.0% vs. 18.1% (NS)
Complicated UTI: 1.7% vs. 1.8% (NS)
Above defined as pyelonephritis, sepsis of urinary origin or UTI-related serious event.
Urosepsis: 0.4% vs. 0.1% (NS)
Genital infection: 6.4% vs. 1.8% (P<0.001)
Above per 88 MedDRA definitions, the authors don't explicitly state which were included.
Males: 5.0% vs. 1.5% (P<0.001)
Females: 10.0% vs. 2.6% (P<0.001)
Volume depletion: 5.1% vs. 4.9% (NS)
ARF: 5.2% vs. 6.6% (P<0.01)
AKI: 1.0% vs. 1.6% (NS) (P<0.05)
DKA: 0.1% vs. <0.1% (NS)
VTE: 0.6% vs. 0.9% (NS)
Fracture: 3.8% vs. 3.9% (NS)


  • Individual empagliflozin arms did not reach statistical significance in outcomes compared to placebo
  • No study has found similar results or side effects (UTI) with any other medication in this class (SGLT2 inhibitors)
  • Statistically significant reductions in the primary outcome were found only in certain subgroups, e.g., Age ≥65, A1C <8.5%, Asian race, BMI <30.


  • Boehringer Ingelheim (BI) and Eli Lilly, manufacturers of empagliflozin (Jardiance) and empagliflozin/metformin (Synjardy)were the primary sources of funding for the study
  • BI paid for medical writing and analyzed the study data, and the study's steering committee included employees of BI
  • All researchers were paid by either BI or Eli Lilly for either consulting or data monitoring

Further Reading

  1. Tancredi M, et al. "Excess mortality among persons with type 2 diabetes." The New England Journal of Medicine. 2015;373:1720-1732.
  2. Dormandy JA, et al. "Secondary prevention of macrovascular events in patients with type 2 diabetes in the PROactive Study (PROspective pioglitAzone Clinical Trial In macroVascular Events): a randomised controlled trial." The Lancet. 2005;366(9493):1279-1289.
  3. Green JB, et al. "Effect of sitagliptin on cardiovascular outcomes in type 2 diabetes." The New England Journal of Medicine. 2014;373(3):232-242.
  4. Pfeffer MA, et al. "Lixisenatide in Patients with Type 2 Diabetes and Acute Coronary Syndrome." N Engl J Med. 2015 Dec 3;373(23):2247-57.
  5. Valentine, V. "The Role of the Kidney and Sodium-Glucose Cotransporter-2 Inhibition in Diabetes Management."Clinical Diabetes.2012; 30(4):151-155.
  6. http://www.ncbi.nlm.nih.gov/pubmed/27299675 Wanner C, et al. "Empagliflozin and Progression of Kidney Disease in Type 2 Diabetes." NEJM. 2016 Jul 28;375(4):323-334.]
  7. Abdul-Ghani M et al. SGLT2 Inhibitors and Cardiovascular Risk: Lessons Learned From the EMPA-REG OUTCOME Study. Diabetes Care 2016. 39:717-25.
  8. 8.0 8.1 Supplementary appendix