TOPLINE:

Patients with type 2 diabetes (T2D) who received SGLT2 inhibitors had a higher risk for diabetic ketoacidosis (DKA) than those who did not receive this class of drug according to results of a systematic review and meta-analysis. Elevated levels of A1c at baseline were independently associated with an increased risk for DKA among those receiving SGLT2 inhibitors.

METHODOLOGY:

• Researchers conducted a systematic review and meta-analysis to examine the relationship between baseline levels of A1c, the use of SGLT2 inhibitors, and the risk for DKA in adults with T2D.
• The analysis included 15 observational cohort studies involving 1,269,005 participants (495,854 in the SGLT2 inhibitor group and 773,151 in the non-SGLT2 inhibitor group) and seven randomized controlled trials (RCTs) involving 58,761 participants (32,656 in the SGLT2 inhibitor group and 26,105 in the placebo group).
• In observational studies, mean baseline age ranged from 56.8 to 72.0 years, baseline A1c levels ranged from 6.9% to 8.99%, and follow-up duration ranged from 0.4 to 3.1 years. In RCTs, mean baseline age ranged from 63.0 to 70.0 years, baseline A1c levels ranged from 7.1% to 8.3%, and follow-up duration ranged from 0.8 to 4.2 years.
• Studies were included only if they reported both baseline A1c levels and DKA incidence and compared SGLT2 inhibitor use with placebo or other active comparators.
• Furthermore, three of the observational studies specifically evaluated the association between A1c and risk for DKA in 153,210 patients with T2D receiving SGLT2 inhibitors.

TAKEAWAY:

• In observational studies, SGLT2 inhibitor use was associated with an increased risk for DKA compared with nonuse (risk ratio [RR], 1.33; 95% CI, 1.01-1.76); risk was higher among those with baseline A1c levels ≥ 8.3% (RR, 1.63; 95% CI, 1.46-1.81), and meta‑regression indicated significant effect modification by A1c (P = .018).
• In RCTs, SGLT2 inhibitor therapy was associated with more than a twofold increased risk for DKA (RR, 2.27; 95% CI, 1.52-3.38); when stratified by A1c levels at baseline, the pooled RRs for DKA were 2.37 (95% CI, 1.44-3.90) in the high A1c group and 2.01 (95% CI, 0.84-4.79) for the low A1c group (no significant effect modification by A1c; meta‑regression P = .73).
• In RCTs where ≥ 50% of patients were insulin users, SGLT2 inhibitor use was also associated with an increased risk for DKA (RR, 2.49; 95% CI, 1.38-4.51).
• Analysis of the three observational studies that focused on SGLT2 inhibitor recipients showed elevated baseline A1c was independently associated with higher DKA risk among users (RR, 1.50; 95% CI, 1.17-1.92).

IN PRACTICE:

“Aggregate findings from both observational studies and RCTs suggest that individuals with higher baseline [A1c] levels may be at increased risk of DKA when treated with [SGLT2 inhibitors] compared to those with better glycemic control,” the authors wrote.

“Clinicians should remain vigilant, particularly in patients with poorly controlled diabetes, and ensure that DKA risk factors are addressed,” they added.

SOURCE:

This study was led by Samuel Seidu, University of Leicester, Leicester General Hospital, Leicester, England. It was published online in Diabetes, Obesity and Metabolism.

LIMITATIONS:

The range of baseline A1c values across included studies, particularly in cardiovascular outcome trials, was relatively narrow. In observational studies, elevated baseline A1c may indicate unmeasured factors such as insulin deficiency or treatment nonadherence, contributing to residual confounding. Additionally, due to limited reporting and reliance on aggregate-level data, the analysis could not assess whether DKA risk was disproportionately higher among insulin-treated patients with elevated baseline A1c.

DISCLOSURES:

This study was financially supported by the National Institute for Health Research Applied Research Collaboration East Midlands. Two authors reported receiving speaker and advisory board honoraria, grants, and conference support or serving as consultants or speakers for various pharmaceutical companies.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.