# Renal Protection at a Metabolic Cost: A Systematic Review and Meta‐Analysis of Perioperative Use of Sodium–Glucose Cotransporter 2 Inhibitors

## TL;DR

Using SGLT2 inhibitors before surgery may protect the kidneys and reduce deaths, but could also increase the risk of ketoacidosis, which is masked by varying study results.

## Contribution

This study provides a meta-analysis showing a paradoxical benefit-risk profile of SGLT2 inhibitors in surgery, highlighting their renoprotective effects and hidden ketoacidosis risks.

## Key findings

- Perioperative SGLT2 inhibitors reduced acute kidney injury and mortality odds.
- Sensitivity analyses revealed increased risk of euglycemic ketoacidosis.
- Lower pH, base excess, and blood glucose levels were observed in SGLT2 inhibitor users.

## Abstract

Concerns about diabetic ketoacidosis (DKA) and euglycemic ketoacidosis (eKA) are balanced against possible organ‐protective benefits in the debated perioperative management of sodium‐glucose cotransporter‐2 (SGLT2) inhibitors. This meta‐analysis compared the perioperative clinical and laboratory outcomes associated with perioperative exposure to SGLT2i.

Through July 31, 2025, we searched PubMed, Web of Science, Scopus, and CENTRAL for observational studies and randomised controlled trials comparing the outcomes of preoperative use of SGLT2 inhibitors with non‐use in patients undergoing cardiac or non‐cardiac surgery. We pooled data using a random‐effects model and investigated heterogeneity using leave‐one‐out sensitivity analyses. PROSPERO‐ID: CRD420251155809.

There were 10 studies comprising 246,242 patients. Due to considerable heterogeneity, the primary pooled analysis revealed no significant association between SGLT2 inhibitor use and either eKA (OR 4.86; p = 0.11) or DKA (OR 2.21; p = 0.11). However, a significant increase in the risk of eKA (OR 1.11; p < 0.001) and DKA (OR 5.33; p < 0.001) was observed using leave‐one‐out sensitivity analysis to identify outliers. On the other hand, the usage of SGLT2 inhibitors was associated with a statistically significant decrease in both mortality (OR 0.73; p = 0.006) and acute renal injury (OR 0.68; p < 0.0001). The SGLT2 inhibitor group had significantly lower perioperative pH, base excess, and blood glucose levels.

The use of perioperative SGLT2 inhibitors poses a clinical paradox between significant renoprotection and survival advantages and a latent risk of ketoacidosis concealed by considerable heterogeneity. While metabolic monitoring is essential, current surgeries requiring more prolonged withholding may need to weigh metabolic risk against the drug's significant benefit in reducing acute kidney injury and mortality.

Across 246,542 surgical patients, perioperative SGLT2 inhibitor use was associated with a lower odds of acute kidney injury and mortality. In contrast, ketoacidosis outcomes (DKA and euglycemic ketoacidosis) showed no statistically significant difference overall, but appeared more frequent in selected sensitivity and subgroup analyses, consistent with an euglycemic masking effect. These findings underscore a perioperative paradox requiring individualised risk stratification and vigilant metabolic monitoring.