The real-world safety profile of sodium-glucose co-transporter-2 inhibitors among older adults (≥ 75 years): a retrospective, pharmacovigilance study

Abstract

Background: As indications for sodium-glucose co-transporter-2 inhibitors (SGLT2i) are expanding, a growing number of older adults have become candidates for treatment. We studied the safety profile of SGLT2i among older adults.

Methods: A retrospective, pharmacovigilance study of the FDA's global database of safety reports. To assess reporting of pre-specified adverse events following SGLT2i among adults (< 75 years) and older adults (≥ 75), we performed a disproportionality analysis using the sex-adjusted reporting odds ratio (adj.ROR).

Results: We identified safety reports of 129,795 patients who received non-insulin anti-diabetic drugs (NIAD), including 24,253 who were treated with SGLT2i (median age 60 [IQR: 51-68] years, 2,339 [9.6%] aged ≥ 75 years). Compared to other NIAD, SGLT2i were significantly associated with amputations (adj.ROR = 355.1 [95%CI: 258.8 - 487.3] vs adj.ROR = 250.2 [79.3 - 789.5]), Fournier gangrene (adj.ROR = 45.0 [34.5 - 58.8] vs adj.ROR = 88.0 [27.0 - 286.6]), diabetic ketoacidosis (adj.ROR = 32.3 [30.0 - 34.8] vs adj.ROR = 23.3 [19.2 - 28.3]), genitourinary infections (adj.ROR = 10.3 [9.4 - 11.2] vs adj.ROR = 8.6 [7.2 - 10.3]), nocturia (adj.ROR = 5.5 [3.7 - 8.2] vs adj.ROR = 6.7 [2.8 - 15.7]), dehydration (adj.ROR = 2.5 [2.3 - 2.8] vs adj.ROR = 2.6 [2.1 - 3.3]), and fractures (adj.ROR = 1.7 [1.4 - 2.1] vs adj.ROR = 1.5 [1.02 - 2.1]) in both adults and older adults, respectively. None of these safety signals was significantly greater in older adults (P_interaction threshold of 0.05). Acute kidney injury was associated with SGLT2i in adults (adj.ROR = 1.97 [1.85 - 2.09]) but not in older adults (adj.ROR = 0.71 [0.59 - 0.84]). Falls, hypotension, and syncope were not associated with SGLT2i among either adults or older adults.

Conclusion: In this global post-marketing study, none of the adverse events was reported more frequently among older adults. Our findings provide reassurance regarding SGLT2i treatment in older adults, although careful monitoring is warranted.

Keywords: Diabetes; Diabetic ketoacidosis; Fournier gangrene; Heart failure; Older adults; Sodium-glucose co-transporter-2 inhibitors.

Fig. 1

Study flowchart. FAERS- FDA adverse event reporting system

Fig. 2

SGLT2-inhibitors related adverse events among adults and older adults. Disproportionality analysis of SGLT2-inhibitors-related adverse events as compared to other non-insulin antidiabetic drugs from the full database. A lower limit of the adj.ROR 95% CI above 1 is the conventional threshold for significant signal detection. Adults are patients aged 18–75, older adults are 75 years or older. adj.ROR- Sex-adjusted reporting odds ratio; AKI- Acute kidney injury; CI- Confidence interval; DKA- Diabetic ketoacidosis; NIAD- non-insulin antidiabetic drugs; SGLT2- Sodium-glucose co-transporter 2

Fig. 3

Selected safety signals stratified by SGLT2-inhibitor agent. Selected statistically significant safety signals were stratified by SGLT2-inhibitor type (canagliflozin: 10,115 adults and 757 older adults; dapagliflozin: 4,560 adults and 539 older adults; empagliflozin: 7,078 adults and 1,034 older adults). RORs of every drug-AE pair were compared to non-insulin antidiabetic drugs from the full database (88,502 adults and 17,040 older adults). A lower limit of the ROR 95% CI above 1 is the conventional threshold for significant signal detection. Adults are patients aged 18–75, older adults are 75 years or older. ROR- Reporting odds ratio; AKI- Acute kidney injury; CI- Confidence interval; DKA- Diabetic ketoacidosis; GU- genitourinary; NIAD- non-insulin antidiabetic drugs; SGLT2- Sodium-glucose co-transporter 2