We read with interest the study published in October 2022 by Murugesan et al. on euglycemic diabetic ketoacidosis (EDKA) during cardiac surgery linked to SGLT2i therapy.[1] We would like to highlight a few points from their work.
First, the authors have diagnosed EDKA based on hyperketonemia [presumably a beta hydroxy-butyrate (BHB) estimate] with a cut-off of 0.6 mmol/L. While hyperketonemia in the setting of SGLT2i and major surgery should alert the clinicians about impending DKA, inferring DKA without explicit biochemical evidence of DKA is a major limitation of the study. It is unclear if a BHB level of 0.6 mmol/L without acidosis, which may partially be related to lengthy fasting, is a cause for immediate concern. Indeed, governing bodies recommend a BHB cut-off of 1.0 mmol/L to consider the likelihood of periprocedural DKA associated with SGLT2i.[2] BHB values more than 1.0 mmol/L should be interpreted in conjunction with other metabolic parameters such as base excess to delineate the progression from ketosis to ketoacidosis and to facilitate further management.[2] Ideally, other standard DKA parameters such as pH and bicarbonate should also have been contemplated for the diagnosis. Without such parameters, the results may reflect hyperketonemia without ketoacidosis. The caveats of relying on ketonuria in the diagnosis of DKA in this scenario require special mention. Ketonuria is less reliable than hyperketonemia as SGLT2i therapy results in increased renal reabsorption of ketone bodies, thus hyperketonemia can manifest in the absence of ketonuria.[3]
Second, the incidence rate is derived from a retrospective study on a small heterogenous sample, wherein the key aspect of preprocedural drug cessation was inconsistent. Ideally, the calculation of incidence requires a large prospective study where the drug will be ceased uniformly in the study cohort with similar baseline characteristics and exposed to surgeries of similar magnitude. Furthermore, in those studies, measures should be in place to identify early changes reflective of an evolving DKA based on clear definitions throughout the perioperative period, which includes preprocedural screening. We acknowledge that such a robust study design to estimate this perioperative adverse event will be challenging.
Third, it is shown that the timing of cessation of SGLT2i had a minimal impact on the occurrence of EDKA based on the statistics applied. The small sample size, inherent limitations of a retrospective study design, and measurement of only serum ketones limit this inference. It should be appreciated that DKA can still occur despite following the guidelines of preprocedure withholding.[4] DKA risk is not just attributable to the medication alone but also significant is a multitude of patient-related and perioperative factors.[5] Adhering to the pharmacologically applied principles is one of many elements mitigating DKA risk.
Fourth, in our earlier systematic review on this topic across 47 cases,[5] we identified 42 cases of EDKA and the remaining five were of DKA with elevated blood sugars (Hyperglycemic DKA). The authors have misinterpreted this as an incident rate of EDKA as 89%. The absence of a clear denominator is an integral limitation in exploring frequency estimates in such reviews and we acknowledged this in the review.
The authors conclude that patients who require insulin preoperatively are more at risk of EDKA from SGLT2i. We believe this to be an epiphenomenon and it probably reflects differences in age, body mass index, insulin resistance, and long-term glycaemic control between the two groups. Finally, there is a possibility that the patients who received glucose insulin infusion were unevenly matched. This would have been a potential confounding factor in the mechanism of ketogenesis.
Vigilance in identifying the early markers of DKA and management pathways based on international guidelines tailored to local resources will mitigate this perioperative adverse event.
Financial support and sponsorship
EJM is the recipient of the 2023 Royal Adelaide Hospital Early Career Research Fellowship.
Conflicts of interest
There are no conflicts of interest.
REFERENCES
- 1.Murugesan KB, Balakrishnan S, Arul A, Ramalingam S, Srinivasan M. A retrospective analysis of the incidence, outcome and factors associated with the occurrence of euglycemic ketoacidosis in diabetic patients on sodium glucose co-transporter-2 inhibitors undergoing cardiac surgery. Ann Card Anaesth. 2022;25:460–5. doi: 10.4103/aca.aca_47_21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Australian Diabetes Society: Periprocedural diabetic ketoacidoss (DKA) with SGLT2 inhibitor use in people with diabetes. [[Last accessed on 2023 Jun 18]]. Available at https://www.diabetessociety.com.au/wp-content/uploads/2023/05/ADS-ADEA-ANZCA-NZSSD_DKA_SGLT2i_Alert_Ver-May-2023.pdf .
- 3.Milder DA, Milder TY, Kam PCA. Sodium-glucose co-transporter type-2 inhibitors: Pharmacology and peri-operative considerations. Anaesthesia. 2018;73:1008–18. doi: 10.1111/anae.14251. [DOI] [PubMed] [Google Scholar]
- 4.Meyer EJ, Mignone E, Hade A, Thiruvenkatarajan V, Bryant RV, Jesudason D. Periprocedural euglycemic diabetic ketoacidosis associated with sodium-glucose cotransporter 2 inhibitor therapy during colonoscopy. Diabetes Care. 2020;43:e181–4. doi: 10.2337/dc20-1244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Thiruvenkatarajan V, Meyer EJ, Nanjappa N, Van Wijk RM, Jesudason D. Perioperative diabetic ketoacidosis associated with sodium-glucose co-transporter-2 inhibitors: A systematic review. Br J Anaesth. 2019;123:27–36. doi: 10.1016/j.bja.2019.03.028. [DOI] [PubMed] [Google Scholar]