In adults ≥65 with type 2 diabetes, GLP-1 receptor agonists reduce major cardiovascular events, and SGLT2 inhibitors reduce heart-failure hospitalizations and kidney outcomes, with benefits that grow with age; sulfonylureas markedly increase hypoglycaemia.

Background

Type 2 diabetes is increasingly common in older adults, who face heterogeneous risks, multimorbidity, functional/cognitive impairments, and higher susceptibility to adverse events. Selecting glucose-lowering medications that maximize patient-centered cardiovascular, renal, cognitive, and safety outcomes is essential. This systematic review and network meta-analysis focused on outcomes directly relevant to older adults.

Patients

• Population: Adults aged ≥65 years with type 2 diabetes mellitus.
• Sample size: 22 randomized controlled trials; 41,654 participants.
• Risk profile: Many trials enrolled patients with established or high cardiovascular/renal risk; subgroup analyses by age (including ≥75 and ≥80 years when available).

Intervention

• Glucose-lowering drug classes: SGLT2 inhibitors, GLP-1 receptor agonists (GLP-1RAs), DPP-4 inhibitors, metformin, sulfonylureas (SU), and acarbose.

Control

• Placebo or active comparators among the above drug classes.

Outcome

• Primary (patient-centered):
  • 3-point major adverse cardiovascular events (3P-MACE: nonfatal MI, nonfatal stroke, cardiovascular death)
  • Hospitalization for heart failure (HHF)
  • Renal composite outcome (≥40% eGFR decline, end-stage renal disease, or renal death)
• Secondary:
  • Myocardial infarction, stroke, cardiovascular death, unstable angina hospitalization, coronary revascularization
  • Albuminuria progression, acute kidney injury (AKI), eGFR change, urinary tract infection
  • Cognition (MMSE, MoCA)
  • Hypoglycaemia (any, severe), diabetic ketoacidosis
  • Glycaemic control (HbA1c, FPG, 2h-PPG), body weight, BMI, systolic/diastolic blood pressure
  • Adverse events (any AE, drug-related AE, discontinuations, serious AEs), all-cause mortality

Study Design

Systematic review and frequentist network meta-analysis of randomized controlled trials (PRISMA 2020 compliant; protocol registered in PROSPERO CRD42023437068). Risk of bias assessed with Cochrane RoB 2.0; confidence in network estimates with CINeMA.

Level of Evidence

Level I (systematic review and network meta-analysis of RCTs); overall CINeMA confidence generally low-to-moderate.

Follow up period

Approximately 24 weeks to 5.4 years across trials; large cardiovascular/renal outcome trials typically 2–4+ years.

Results

Primary outcomes

• 3P-MACE
  • GLP-1RAs vs placebo: RR 0.83 (95% CI, 0.71–0.97)
  • NNT (illustrative): ≈40–60 over ~3–5 years, assuming baseline 3P-MACE risk 10–15% in older high-risk populations (ARR ≈1.7–2.6%).
• Hospitalization for heart failure (HHF)
  • SGLT2 inhibitors vs placebo: RR 0.66 (95% CI, 0.57–0.77)
  • NNT (illustrative): ≈37–60 over ~2–4+ years, assuming baseline HHF risk 5–8% (ARR ≈1.7–2.7%).
• Renal composite outcome
  • SGLT2 inhibitors vs placebo: RR 0.69 (95% CI, 0.53–0.89)
  • NNT (illustrative): ≈32–65 over ~2–4+ years, assuming baseline renal composite risk 5–10% (ARR ≈1.6–3.1%).

Secondary outcomes

• Cardiovascular components
  • GLP-1RAs: MI RR 0.69 (95% CI, 0.48–0.99); coronary revascularization RR 0.66 (95% CI, 0.49–0.89).
  • SGLT2 inhibitors: Highest ranking for HHF prevention; mixed effects for 3P-MACE across individual agents.
• Renal components
  • SGLT2 inhibitors: Albuminuria progression RR 0.71 (95% CI, 0.59–0.85); AKI RR 0.70 (95% CI, 0.53–0.92).
  • DPP-4 inhibitors: Albuminuria progression RR 0.89 (95% CI, 0.81–0.98); eGFR reduction MD −3.46 mL/min/1.73 m² (95% CI, −5.37 to −1.55).
• Cognition
  • DPP-4 inhibitors vs SU (single RCT): Better MMSE (MD 4.08; 95% CI, 4.02–4.14) and MoCA (MD 1.20; 95% CI, 1.12–1.28).
• Glycaemic control and hemodynamics
  • HbA1c/FPG/2h-PPG: GLP-1RAs, SGLT2 inhibitors, metformin, SU, and DPP-4 inhibitors all reduced HbA1c and postprandial glucose; GLP-1RAs primarily lowered 2h-PPG (less effect on FPG).
  • Blood pressure (SGLT2 inhibitors): SBP MD −4.45 mmHg (95% CI, −6.64 to −2.25); DBP MD −1.17 mmHg (95% CI, −2.07 to −0.26).
• Weight
  • GLP-1RAs: MD −3.87 kg (95% CI, −5.54 to −2.21).
  • SGLT2 inhibitors: MD −1.85 kg (95% CI, −2.42 to −1.27).
  • Sulfonylureas: MD +1.09 kg (95% CI, 0.13 to 2.06).
• Hypoglycaemia
  • Sulfonylureas: Any hypoglycaemia RR 4.19 (95% CI, 3.52–4.99); severe hypoglycaemia RR 7.06 (95% CI, 3.03–16.43).
• Adverse events and mortality
  • All-cause mortality generally comparable to placebo across classes.
  • Several agents increased any AEs and discontinuations vs placebo; serious AEs and serious drug-related AEs did not consistently increase.
• Age-specific effects
  • Cardiovascular and renal benefits increase with age; the number needed to treat decreases in older subgroups.

Limitations

• Many trials enrolled patients with established/high CV or renal risk; applicability to lower-risk older adults is uncertain.
• Insufficient RCT data on neuropathy and retinopathy in older adults.
• Few trials in newly diagnosed patients; most were add-on therapies.
• Comparative power may be limited for drugs with similar mechanisms; heterogeneity across agents within classes (e.g., SGLT2 inhibitors).
• One small trial informed cognition; further dedicated studies are needed.
• Older adults with severe frailty/comorbidity often excluded from RCTs, limiting generalizability.

Funding

Ministry of Health and Welfare (MOHW104-TDU-B-211-113-003; MOHW106-TDU-B-211-113-001); Ministry of Science and Technology (MOST 106-2314-B-075-051-MY3; MOST 109-2314-B-010-061-; MOST 110-2634-F-A49-005); National Yang Ming Chiao Tung University (E107F-M01-0501); Yin Yen-Liang Foundation Development and Construction Plan (School of Medicine, National Yang Ming Chiao Tung University). Funders had no role in study design, data collection/analysis, interpretation, or manuscript writing.

Citation

Pan S-Y, Su E-L, Huang C-J, Chuang S-Y, Chiang C-E, Chen C-H, Cheng H-M. Evaluation of glucose-lowering medications in older people: a comprehensive systematic review and network meta-analysis of randomized controlled trials. Age and Ageing. 2024;53:afae175. doi:10.1093/ageing/afae175.