What was the primary endpoint of DEVOTE?

The primary endpoint was time to first occurrence of a three-component MACE: cardiovascular death, non-fatal myocardial infarction, or non-fatal stroke. Degludec was non-inferior to glargine U100 with a hazard ratio of 0.91 (95% CI 0.78-1.06).

Did DEVOTE show superiority for degludec on cardiovascular outcomes?

No. The trial met its non-inferiority endpoint but did not demonstrate superiority (p = 0.25 for superiority testing). The point estimate favored degludec numerically, but the confidence interval included 1.0.

How much did degludec reduce severe hypoglycemia compared to glargine?

Degludec reduced severe hypoglycemia by 40% overall (rate ratio 0.60, p <0.001) and severe nocturnal hypoglycemia by 53% (HR 0.47, p <0.001) compared to glargine U100.

Were the glycemic outcomes different between the two insulins?

HbA1c was virtually identical at trial end (both 7.5%). Fasting plasma glucose was slightly lower with degludec by about 4.4 mg/dL (p <0.001), reflecting its flatter pharmacokinetic profile.

How long did the DEVOTE trial last?

The median follow-up was 1.99 years. Maximum observation extended to approximately 2.5 years. The relatively short duration is one acknowledged limitation of the trial.

Insulin administration was open-label (patients and clinicians knew the assignment). However, MACE endpoint adjudication was performed by an independent blinded committee, reducing bias in the primary outcome assessment.

Does DEVOTE apply to patients not on insulin before?

About 15% of DEVOTE participants were insulin-naive. The subgroup analysis showed consistent non-inferiority regardless of prior insulin use (interaction p = 0.53), though the insulin-naive subgroup was smaller and the confidence interval wider.

How does degludec compare to glargine U300 (Toujeo)?

DEVOTE compared degludec only to glargine U100. The CONCLUDE trial later compared degludec to glargine U300 and found no significant difference in overall hypoglycemia rates, suggesting the advantage seen in DEVOTE may narrow against the newer glargine formulation.

Did DEVOTE find a link between hypoglycemia and cardiovascular events?

The DEVOTE 5 substudy found that severe hypoglycemia was associated with a subsequent higher risk of MACE (HR 2.05) and all-cause mortality (HR 2.23), though this was an observational association within the trial, not a causal proof.

Is degludec considered cardiovascularly protective?

No. DEVOTE established cardiovascular safety (non-inferiority), not cardiovascular protection. Unlike GLP-1 agonists or SGLT2 inhibitors that showed CV superiority in their outcome trials, degludec is considered CV-neutral.

DEVOTE Results in Detail: Numbers, Subgroups, and Time Course

Q: Were the glycemic outcomes different between the two insulins?

HbA1c was virtually identical at trial end (both 7.5%). Fasting plasma glucose was slightly lower with degludec by about 4.4 mg/dL (p <0.001), reflecting its flatter pharmacokinetic profile.

Q: How long did the DEVOTE trial last?

The median follow-up was 1.99 years. Maximum observation extended to approximately 2.5 years. The relatively short duration is one acknowledged limitation of the trial.

Q: Was DEVOTE blinded?

Insulin administration was open-label (patients and clinicians knew the assignment). However, MACE endpoint adjudication was performed by an independent blinded committee, reducing bias in the primary outcome assessment.

Q: Does DEVOTE apply to patients not on insulin before?

About 15% of DEVOTE participants were insulin-naive. The subgroup analysis showed consistent non-inferiority regardless of prior insulin use (interaction p = 0.53), though the insulin-naive subgroup was smaller and the confidence interval wider.

Q: How does degludec compare to glargine U300 (Toujeo)?

DEVOTE compared degludec only to glargine U100. The CONCLUDE trial later compared degludec to glargine U300 and found no significant difference in overall hypoglycemia rates, suggesting the advantage seen in DEVOTE may narrow against the newer glargine formulation.

Q: Did DEVOTE find a link between hypoglycemia and cardiovascular events?

The DEVOTE 5 substudy found that severe hypoglycemia was associated with a subsequent higher risk of MACE (HR 2.05) and all-cause mortality (HR 2.23), though this was an observational association within the trial, not a causal proof.

Q: Is degludec considered cardiovascularly protective?

No. DEVOTE established cardiovascular safety (non-inferiority), not cardiovascular protection. Unlike GLP-1 agonists or SGLT2 inhibitors that showed CV superiority in their outcome trials, degludec is considered CV-neutral.

By HealthRX Medical Team

Published May 25, 2026Updated May 25, 2026Last reviewed May 25, 2026

At a glance

| Parameter | Detail | |---|---| | N | 7,637 (3,818 degludec; 3,819 glargine U100) | | Population | T2D with established CVD or high CV risk | | Intervention | Insulin degludec, once daily, treat-to-target | | Comparator | Insulin glargine U100, once daily, treat-to-target | | Duration | Median 1.99 years (max ~2.5 years) | | Primary endpoint | First MACE (CV death, non-fatal MI, non-fatal stroke) | | Key result | MACE HR 0.91 (95% CI 0.78-1.06); non-inferiority confirmed (p <0.001 for non-inferiority) |

Why a CVOT for Basal Insulin?

After the 2008 FDA guidance requiring cardiovascular outcome trials for new diabetes drugs, degludec needed its own safety record. An earlier regulatory review had flagged a numerical MACE imbalance in the phase 3 program, prompting the FDA to request a dedicated trial. DEVOTE was designed as a non-inferiority study with a prespecified margin of 1.3 for the upper bound of the 95% confidence interval for the hazard ratio of first MACE. The trial enrolled patients who were already at elevated cardiovascular risk, not a general T2D population, which concentrated events and shortened the required follow-up.

Primary Endpoint: Three-Component MACE

The primary publication reported 325 first MACE events in the degludec arm versus 356 in the glargine arm. The hazard ratio was 0.91 (95% CI 0.78-1.06, p <0.001 for non-inferiority). Because the upper bound of 1.06 fell well below the prespecified 1.3 margin, non-inferiority was established decisively. Superiority testing was prespecified as a secondary analysis; the p-value for superiority was 0.25, so the trial did not demonstrate that degludec was better for MACE, only that it was not worse.

MACE Components Broken Down

| Component | Degludec n (%) | Glargine n (%) | HR (95% CI) | |---|---|---|---| | CV death | 136 (3.6%) | 142 (3.7%) | 0.96 (0.76-1.21) | | Non-fatal MI | 137 (3.6%) | 152 (4.0%) | 0.90 (0.71-1.13) | | Non-fatal stroke | 65 (1.7%) | 76 (2.0%) | 0.85 (0.61-1.18) |

No single component drove the overall result. Each individual hazard ratio sat below 1.0, but none reached statistical significance on its own. The consistency across components strengthens the interpretation that degludec carries no excess cardiovascular signal.

Time-Course Pattern of MACE Events

Kaplan-Meier curves for the primary endpoint ran nearly superimposed from randomization through the full follow-up period. There was no early divergence, no late catch-up, and no crossover. The cumulative incidence at 24 months was approximately 8.5% in the degludec group and 9.3% in the glargine group per the primary trial data. The event rate was roughly linear after the first 6 months, consistent with a stable-risk population rather than one accumulating events in bursts.

This flat trajectory matters clinically. Some CVOTs show separation that widens with time (as in EMPA-REG OUTCOME for empagliflozin), suggesting a cumulative benefit. DEVOTE showed no such widening, which aligns with the interpretation that basal insulins are CV-neutral rather than CV-protective.

The HealthRX DEVOTE Signal Framework

To distinguish what DEVOTE actually proved from what it is often cited as proving, we apply a three-tier classification to each reported outcome.

| Tier | Definition | DEVOTE Outcomes in This Tier | |---|---|---| | Confirmed | Met prespecified statistical threshold | MACE non-inferiority; severe nocturnal hypo reduction | | Directional | Point estimate favors degludec, CI crosses 1.0 | Each individual MACE component; all-cause mortality (HR 0.91 to 95% CI 0.76-1.09) | | Neutral | No meaningful difference observed | HbA1c reduction; fasting plasma glucose; body weight |

This framework helps clinicians avoid over-interpreting directional signals as proven benefits. Degludec is not a cardiovascular drug. It is a basal insulin that passed its safety test and happened to show a clinically important hypoglycemia advantage.

Glycemic Control: Treat-to-Target Design

Both arms targeted a fasting plasma glucose of 71-90 mg/dL. By trial end, mean HbA1c was virtually identical: 7.5% in the degludec group and 7.5% in the glargine group (estimated treatment difference -0.01 percentage points, 95% CI -0.08 to 0.05). Fasting plasma glucose was slightly but significantly lower with degludec (difference of -4.4 mg/dL, p <0.001), likely reflecting degludec's flatter pharmacokinetic profile over 24+ hours, as described in earlier pharmacodynamic studies (Heise et al., 2012).

Mean insulin dose at end of trial was 0.82 U/kg/day for degludec and 0.84 U/kg/day for glargine, a negligible difference that confirms the treat-to-target design worked as intended.

Hypoglycemia: The Clinically Decisive Result

While MACE non-inferiority was the regulatory question, the hypoglycemia findings carry more weight for prescribing decisions.

Severe Hypoglycemia

| Measure | Degludec | Glargine | Rate Ratio or HR | |---|---|---|---| | Patients with ≥1 severe episode | 187 (4.9%) | 252 (6.6%) |, | | Severe episodes per 100 patient-years | 3.70 | 6.25 | RR 0.60 (95% CI 0.48-0.76, p <0.001) | | Time to first severe episode |, |, | HR 0.73 (0.60-0.89, p = 0.002) |

Severe Nocturnal Hypoglycemia

| Measure | Degludec | Glargine | Rate Ratio or HR | |---|---|---|---| | Patients with ≥1 severe nocturnal episode | 39 (1.0%) | 82 (2.1%) |, | | Severe nocturnal episodes per 100 patient-years | 0.62 | 1.40 | RR 0.44 (0.29-0.69, p <0.001) | | Time to first severe nocturnal episode |, |, | HR 0.47 (0.31-0.73, p <0.001) |

The 53% reduction in severe nocturnal hypoglycemia is the single most practice-changing finding from DEVOTE. Nocturnal hypoglycemia drives fear of insulin intensification, causes falls in older adults, and has been linked to cardiac arrhythmia risk. The American Diabetes Association Standards of Care now reference hypoglycemia risk as a factor in basal insulin selection, a recommendation informed in part by these data.

Subgroup Analyses for MACE

The trial prespecified 17 subgroup analyses for the primary MACE endpoint. Key findings:

| Subgroup | HR (95% CI) | Interaction p-value | |---|---|---| | Age <65 years | 0.87 (0.71-1.08) | 0.43 | | Age ≥65 years | 0.95 (0.77-1.17) |, | | Baseline HbA1c <8.4% | 0.90 (0.73-1.10) | 0.80 | | Baseline HbA1c ≥8.4% | 0.93 (0.74-1.16) |, | | Prior insulin use: yes | 0.93 (0.79-1.09) | 0.53 | | Prior insulin use: no | 0.80 (0.54-1.18) |, | | eGFR <60 mL/min | 0.88 (0.70-1.12) | 0.67 | | eGFR ≥60 mL/min | 0.93 (0.76-1.13) |, |

No interaction p-value reached significance. The non-inferiority finding was consistent regardless of age, renal function, baseline HbA1c, geographic region, or prior insulin use.

Limitations the Authors Acknowledged

The DEVOTE investigators were explicit about several constraints:

Short median follow-up (1.99 years). Longer-duration effects, beneficial or harmful, could not be captured.
Treat-to-target design equalized glycemia. Any CV benefit from better glucose control would be masked by design.
Open-label insulin titration. While MACE adjudication was blinded, patients and clinicians knew the assigned insulin, which could influence hypoglycemia reporting.
Comparator was glargine U100, not U300. Glargine U300 (Toujeo) has its own flatter profile and lower nocturnal hypoglycemia rate versus U100. Whether degludec's hypoglycemia advantage persists against U300 remains an open question addressed only by the smaller CONCLUDE trial, which showed no significant difference in overall hypoglycemia between degludec and glargine U300.
Population was high-CV-risk. Generalizing the MACE result to lower-risk T2D patients is reasonable but not formally tested.

Post-Trial Analyses: DEVOTE Sub-Studies

Several prespecified substudies extended the primary findings:

DEVOTE 3 examined day-to-day fasting glucose variability. Degludec produced 36% lower visit-to-visit variability in fasting plasma glucose, and higher variability was independently associated with higher risk of severe hypoglycemia and MACE (Zinman et al., 2018).
DEVOTE 5 assessed the relationship between severe hypoglycemia and subsequent CV events. Severe hypoglycemia was associated with a higher risk of MACE (HR 2.05) and all-cause mortality (HR 2.23) in the following observation period.

These substudies did not change the primary conclusions but reinforced the clinical relevance of the hypoglycemia advantage: less glycemic variability, fewer severe lows, and a lower burden of hypoglycemia-associated cardiovascular risk.

What This Means for Prescribing

DEVOTE answers one question with certainty: degludec is cardiovascularly safe. It answers a second question with strong evidence: degludec causes meaningfully less severe hypoglycemia, especially at night, compared to glargine U100. It does not answer whether degludec is superior to glargine U300 for hypoglycemia, nor whether it offers CV protection beyond neutrality.

For clinicians choosing between basal insulins in patients with established CVD or high CV risk, the hypoglycemia data from DEVOTE tilt the balance toward degludec when nocturnal hypoglycemia history, fall risk, or hypoglycemia unawareness are part of the clinical picture. Cost and formulary access remain the primary counterweights, as degludec is typically more expensive than glargine biosimilars now available in most markets.

Frequently asked questions

›

References

Marso SP, McGuire DK, Zinman B, et al. Efficacy and Safety of Degludec versus Glargine in Type 2 Diabetes. N Engl J Med. 2017;377(8):723-732. PubMed
Zinman B, Marso SP, Poulter NR, et al. Day-to-day fasting glycaemic variability in DEVOTE: associations with severe hypoglycaemia and cardiovascular outcomes (DEVOTE 2 and DEVOTE 3). Diabetologia. 2018;61(1):48-57. PubMed
Philis-Tsimikas A, Klonoff DC, Engmann Kurtz C, et al. CONCLUDE: A Randomized Clinical Trial of Insulin Degludec versus Glargine U300 in Insulin-Treated Type 2 Diabetes. Diabetes Care. 2020;43(6):1234-1241. PubMed
Heise T, Hermanski L, Nosek L, et al. Insulin degludec: four times lower pharmacodynamic variability than insulin glargine under steady-state conditions in T1DM. Diabetes Obes Metab. 2012;14(9):859-864. PubMed
American Diabetes Association Professional Practice Committee. Standards of Care in Diabetes, 2023. Diabetes Care. 2023;46(Suppl 1). PubMed
FDA Label: Insulin Degludec (Tresiba). AccessData