What interventions reduce all-cause mortality in patients with diabetes?

Diabetes Interventions That Reduce All-Cause Mortality

Metformin, SGLT-2 inhibitors, and GLP-1 receptor agonists all reduce all-cause mortality in patients with type 2 diabetes, with metformin showing the strongest long-term survival benefit, followed by SGLT-2 inhibitors and GLP-1 agonists, while DPP-4 inhibitors, insulin, and sulfonylureas do not reduce mortality. 1, 2

First-Line Therapy: Metformin

Metformin is the mandatory first-line pharmacologic therapy for all patients with type 2 diabetes unless contraindicated. 2, 3

• In the UKPDS 34 trial, metformin reduced all-cause mortality by 36% (relative risk reduction, 9% to 55%, P = 0.011), translating to approximately 7 fewer deaths per 1000 patient-years compared to conventional therapy. 1

• On extended 17-year follow-up, metformin maintained a 27% reduction in all-cause mortality (7.2 deaths per 1000 patient-years, P = 0.002). 1

• A 2024 meta-analysis comparing metformin directly to SGLT-2 inhibitors in patients with cardiovascular disease found metformin offered 23.26 months of survival free from all-cause mortality versus 23.04 months with SGLT-2 inhibitors, with SGLT-2 inhibitors showing higher all-cause mortality (hazard ratio 1.308,95% CI 1.103-1.550) compared to metformin. 4

• Metformin reduces all-cause mortality compared to sulfonylureas, with observational data showing patients on metformin had longer survival than matched non-diabetic controls, while those on sulfonylureas had markedly reduced survival. 1, 5

Second-Line Therapy: SGLT-2 Inhibitors

When glycemic control remains inadequate on metformin, add an SGLT-2 inhibitor as the preferred second-line agent. 2, 3

• SGLT-2 inhibitors reduce all-cause mortality compared to usual care (high certainty of evidence). 1

• SGLT-2 inhibitors reduce all-cause mortality compared to insulin (low to moderate certainty of evidence). 1

• Prioritize SGLT-2 inhibitors specifically in patients with congestive heart failure, chronic kidney disease, or when cardiovascular mortality reduction is the primary goal. 2, 3

• SGLT-2 inhibitors also reduce major adverse cardiovascular events (moderate to high certainty), progression of chronic kidney disease (high certainty), and heart failure hospitalizations (high certainty). 1, 2

• SGLT-2 inhibitors reduce severe hypoglycemia compared to sulfonylureas and insulin (low to high certainty). 1

Alternative Second-Line Therapy: GLP-1 Receptor Agonists

GLP-1 receptor agonists are an alternative second-line option when SGLT-2 inhibitors are not preferred. 2, 3

• GLP-1 agonists reduce all-cause mortality compared to usual care (high certainty of evidence). 1

• GLP-1 agonists reduce all-cause mortality compared to DPP-4 inhibitors (moderate certainty of evidence). 1

• Prioritize GLP-1 agonists specifically in patients with increased stroke risk or when weight loss is an important treatment goal. 2, 3

• GLP-1 agonists also reduce major adverse cardiovascular events (moderate to high certainty) and stroke (high certainty). 1, 2

• GLP-1 agonists reduce severe hypoglycemia compared to sulfonylureas and insulin (low to high certainty). 1

Therapies That Do NOT Reduce All-Cause Mortality

The following interventions do not reduce all-cause mortality and should not be selected for this outcome:

• DPP-4 inhibitors do not reduce all-cause mortality compared to usual care (low to high certainty of evidence). 1

• Insulin does not reduce all-cause mortality compared to usual care (low to high certainty of evidence). 1

• Tirzepatide does not reduce all-cause mortality compared to usual care (low to high certainty of evidence). 1

• Intensive glycemic control with sulfonylureas or insulin showed no reduction in all-cause mortality in the VADT trial (HR 1.05,95% CI 0.89-1.25). 1

• The UKPDS 33 trial of intensive therapy with sulfonylureas or insulin versus conventional therapy showed only a 6% relative reduction in all-cause mortality that was not statistically significant (P = 0.44). 1

Critical Safety Consideration

When SGLT-2 inhibitors or GLP-1 agonists achieve adequate glycemic control, reduce or discontinue sulfonylureas or long-acting insulins to minimize severe hypoglycemia risk. 2, 3

• Hypoglycemic events were much more common with intensive therapy using sulfonylureas or insulin (approximately 30% vs. 1% annually in UKPDS trials). 1

• The VADT trial showed a 3-fold higher rate of hypoglycemic episodes with impaired consciousness in the intensive therapy group (9 vs. 3 episodes per 100 patient-years). 1

• Early addition of metformin to sulfonylureas resulted in an increased risk for diabetes-related death (P = 0.039) compared with continued treatment with sulfonylureas alone. 1

Practical Implementation Algorithm

1. Start all patients on metformin (unless contraindicated) plus lifestyle modifications. 2, 3

2. If glycemic control remains inadequate after 3 months, add either an SGLT-2 inhibitor or GLP-1 agonist based on:

   • Choose SGLT-2 inhibitor if: Heart failure, chronic kidney disease, or need for cardiovascular mortality reduction. 2, 3
   • Choose GLP-1 agonist if: Increased stroke risk or weight loss is a priority. 2, 3

3. Target HbA1c between 7% and 8% for most adults. 2, 3

4. When adding SGLT-2 inhibitor or GLP-1 agonist, reduce or discontinue sulfonylureas or insulin to prevent hypoglycemia. 2, 3

5. Never add DPP-4 inhibitors for mortality reduction—they provide no benefit. 1, 3

Common Pitfalls to Avoid

• Do not rely on intensive glycemic control alone (targeting very low HbA1c) to reduce mortality—the VADT and UKPDS 33 trials showed this approach does not reduce all-cause mortality and increases hypoglycemia risk. 1

• Do not combine metformin with sulfonylureas early as this increases diabetes-related death. 1

• Do not use sulfonylureas or insulin as second-line agents when mortality reduction is the goal—they are inferior to SGLT-2 inhibitors and GLP-1 agonists. 2, 3

• Monitor for vitamin B12 deficiency with long-term metformin use, especially in patients with anemia or peripheral neuropathy. 2