Abstract
While sodium-glucose cotransporter-2 (SGLT2) inhibitors improve cardiovascular outcomes after myocardial infarction (MI), they carry the risk of euglycemic diabetic ketoacidosis (euDKA). We report the case of a 60-year-old male with type 2 diabetes mellitus who developed euDKA after a primary percutaneous coronary intervention (PCI) for ST-segment elevation myocardial infarction (STEMI). Despite exhibiting hemodynamic stability, the patient developed persistent metabolic acidosis. The diagnosis of SGLT2 inhibitor-associated euDKA was significantly delayed, as it was obscured by a combination of the patient's clinical stability, an initial normal anion gap acidosis from diarrhea, and the administration of sodium bicarbonate. This case highlights the fact that clinical stability after primary PCI might mask SGLT2 inhibitor-associated euDKA. This report also emphasizes that euDKA should be a critical diagnostic consideration in the face of persistent acidosis in STEMI patients treated with PCI and receiving SGLT2 inhibitors, even when they appear clinically stable.
Learning objectives
This case highlights that euglycemic diabetic ketoacidosis (euDKA) might be a critical and missed diagnosis in SGLT2 inhibitor-treated diabetic patients with ST-segment elevation myocardial infarction (STEMI) undergoing primary percutaneous coronary intervention (PCI). SGLT2 inhibitor-associated euDKA should be considered in diabetic patients presenting with unexplained metabolic acidosis, even if they are hemodynamically stable and their blood glucose levels are normal. Routine ketone monitoring should be considered in this setting.
Keywords: Sodium-glucose cotransporter-2 inhibitor, Euglycemic diabetic ketoacidosis, Myocardial infarction, Percutaneous coronary intervention, Metabolic acidosis
Introduction
Sodium-glucose cotransporter-2 (SGLT2) inhibitors are widely used worldwide for their beneficial effects in patients with diabetes mellitus (DM), as well as to reduce cardiovascular events in patients with heart failure and slow the progression of chronic kidney disease (CKD) [1]. Many recent studies have also reported their utility in myocardial infarction (MI) patients [1].
Despite these benefits, the potential for SGLT2 inhibitor-associated euglycemic diabetic ketoacidosis (euDKA) warrants careful attention. SGLT2 inhibitors are associated with the risk of DKA, with euDKA (plasma glucose <250 mg/dL) being reported with their use in patients with type 2 DM [2]. In these patients, near-normal or mildly elevated glucose levels delay recognition of the problem by both patients and clinicians. Furthermore, this serious condition can also be asymptomatic. Hence, serum ketones should be measured in patients receiving SGLT2 inhibitors, particularly in those presenting with nausea, vomiting, or malaise.
The practice of discontinuing SGLT2 inhibitors in acute stress conditions, such as surgery and infection, is strongly recommended across international guidelines due to their potential for causing metabolic acidosis. However, the safety of continuing SGLT2 inhibitors in MI patients is unclear. Here, we report a case of symptomatic SGLT2 inhibitor-associated euDKA after primary percutaneous coronary intervention (PCI) for ST-segment elevation myocardial infarction (STEMI).
Case report
A 60-year-old male (163 cm tall, weighing 67 kg, and with a body mass index of 25.2 kg/m2) with a medical history of type 2 DM, dyslipidemia, a history of inferior myocardial infarction treated with PCI at the age of 58 years, and multivessel coronary artery disease treated with coronary artery bypass grafting, also at the age of 58 years, presented with sudden onset of chest tightness while at the hospital for a follow-up. On physical examination, his blood pressure was 124/86 mmHg, and his pulse rate was 83 beats/min. Cardiac and respiratory sounds were normal on auscultation, and no pitting edema was noted. Since an electrocardiogram (ECG) showed ST-segment elevation in leads II, III and aVF at presentation, he was diagnosed with STEMI.
His initial laboratory findings included a serum creatinine level of 0.74 mg/dL. Emergency coronary angiography revealed total occlusion of the right coronary artery (Fig. 1). His regular medications at this time included aspirin 100 mg, dapagliflozin 10 mg, sitagliptin 50 mg, metformin 500 mg, rosuvastatin 5 mg and bisoprolol 5 mg.
Fig. 1.
Coronary angiography (A) before PCI and (B) after PCI. PCI, percutaneous coronary intervention.
Following a successful primary PCI for STEMI, the patient was admitted to the intensive care unit (ICU). His peak creatine kinase (CK) level was 1402 U/L at 12 h after symptom onset. Given his stable hemodynamic condition, oral intake was resumed on day 2, and his home medication was restarted. Despite this apparent stability, his post-procedural course was unexpectedly complicated by persistent metabolic acidosis. Vital signs and blood tests ruled out major causes, such as lactic acidosis from circulatory failure or bowel ischemia. The acidosis was initially attributed to diarrhea and was partially ameliorated by sodium bicarbonate administration. However, it persisted without a clear cause. His blood glucose level was 119 mg/dL at the time of ICU admission and was subsequently well-controlled without significant hyperglycemia. Contrast-enhanced computed tomography on day 3 showed no obvious cause for acidosis, such as bowel ischemia or abscess formation. On Day 4, the anion gap had widened to 14.9 mEq/L from 10.9 mEq/L on day 1, indicating a transition to high-anion-gap acidosis, prompting the strong suspicion of euDKA. Subsequently, dapagliflozin was discontinued on hospital day 4. An endocrinologist was consulted on hospital day 6, and subsequent investigations revealed ketosis (total serum ketones 639 μmol/L, 3-hydroxybutyrate (3-OHB) 311 μmol/L, acetoacetate 328 μmol/L) and 3+ ketonuria, although his blood glucose at that point was only mildly elevated, at 185 mg/dL (Table 1). The modest elevation in serum ketones was likely due to the timing of blood sampling, which was performed 48 h after dapagliflozin had been discontinued. Nevertheless, subsequent persistent and marked ketonuria supported the diagnosis. This discrepancy between persistent acidosis and near-normal glucose levels was the key to confirming the diagnosis of euDKA.
Table 1.
Summary of key laboratory values.
| Day 1 | Day 2 | Day 3 | Day 4 | Day 5 | Day 6 | Day 7 | Day 8 | |
|---|---|---|---|---|---|---|---|---|
| Arterial blood gas | ||||||||
| pH | 7.34 | 7.36 | 7.41 | 7.33 | 7.31 | 7.38 | 7.42 | |
| Anion gap (mEq/L) | 10.9 | 10.4 | 8.8 | 14.9 | 12.4 | 8.2 | 5.5 | |
| HCO3− (mmol/L) | 15.8 | 14.1 | 15.2 | 11.4 | 11.1 | 15.5 | 21.1 | |
| CO2 (mmHg) | 29.4 | 25.8 | 24.3 | 23.8 | 22.9 | 26.8 | 32.9 | |
| BE (mmol/L) | −4.8 | −10.4 | −8.5 | −13.4 | −14 | −8.6 | −2.6 | |
| Lactate (mmol/L) | 1.4 | 1.5 | 1.0 | 2.1 | 0.7 | 0.7 | 0.7 | |
| Glucose (mg/dL) | 127 | 130 | 89 | 111 | 79 | 106 | 127 | |
| Urine tests | ||||||||
| Ketones | N/A | N/A | 3+ | N/A | N/A | 3+ | 1+ | – |
| Specific gravity | N/A | N/A | 1.037 | N/A | N/A | 1.020 | 1.013 | 1.019 |
BE, base excess; CO2, partial pressure of carbon dioxide; HCO3−, bicarbonate.
Based on these findings, SGLT2 inhibitor-associated euDKA was diagnosed as the cause of his metabolic acidosis. Discontinuation of dapagliflozin and treatment with intravenous fluids led to a rapid resolution of both ketosis and acidosis by day 6, along with a significant improvement in his clinical status. Given the episode of SGLT2 inhibitor-associated euDKA, re-initiation of this drug class was avoided to prevent recurrence. The patient was discharged on hospital day 21 with prescription of a modified anti-diabetic regimen, including a glucagon-like peptide-1 receptor agonist-1 (semaglutide) as a safer alternative for his glycemic control. A summary of his key laboratory values during hospitalization is shown in Table 1.
Discussion
This case highlights the diagnostic difficulty of SGLT2 inhibitor-associated euDKA in patients with AMI. In this patient, the diagnosis was initially obscured and delayed by the early resumption of oral intake, diarrhea, and sodium bicarbonate administration. The diagnosis was eventually established based on persistent metabolic acidosis and ketonuria.
SGLT2 inhibitors lower blood glucose levels, while, in turn, increasing free fatty acids (FFAs) in the bloodstream. This promotes the production of ketone bodies, causing ketosis and ketoacidosis. The catecholamine surge triggered by AMI in this state further accelerates lipolysis, causing a rapid increase in both FFAs and ketone bodies. A recent study demonstrated that STEMI was a risk factor for DKA in type 2 DM patients (OR 12.8, 95 % CI 1.3–123.4, P = 0.03) [2]. In our case, in addition to the physiological stress of AMI, factors such as diarrhea and a period of reduced oral intake contributed to the development of euDKA, as both are known to promote ketogenesis.
Lactate levels in our patient showed only a mild, transient increase on Day 4 (2.1 mmol/L), which was likely attributable to the physiological stress of AMI and relative hypovolemia from diarrhea. This degree of elevation was, however, insufficient to account for the magnitude of the high anion-gap metabolic acidosis observed.
The diagnostic clue was the evolution of the acidosis itself. The anion gap widened from 10.9 mEq/L on Day 1 to 14.9 mEq/L on Day 4, marking the transition from a diarrhea-related non–anion-gap acidosis to a ketone-driven high-anion-gap acidosis.
A key diagnostic challenge in this case was the discrepancy between the marked ketonuria (3+) and the only modestly elevated total serum ketone level of 639 μmol/L. The observed discrepancy could be because the blood ketone measurement was obtained more than 48 h after dapagliflozin discontinuation and after initiation of fluid therapy, suggesting the patient was already in the recovery phase of ketoacidosis. This is supported by the serum ketone composition, which showed comparable levels of 3-OHB and acetoacetate [3]. This composition is consistent with resolving ketoacidosis, unlike the marked 3-OHB predominance seen at its peak. Furthermore, the marked ketonuria can be explained by the fact that standard urine dipsticks primarily detect acetoacetate, which was still present at a significant concentration despite the patient being in the resolution phase of euDKA.
International guidelines strongly recommend the discontinuation of SGLT2 inhibitors in acute stress conditions, such as emergency surgery and severe infections [4]. However, there are no clearly established guidelines for their use following AMI or PCI. A literature search identified only five single-case reports describing SGLT2 inhibitor-associated euDKA following AMI that was treated with primary PCI (Table 2) [[5], [6], [7], [8], [9]]. These reports underscore the fact that diabetic patients receiving SGLT2 inhibitors might be at a high risk for DKA in the hyperacute phase of MI, even if hemodynamically stable.
Table 2.
Case reports of SGLT2 inhibitor-associated euDKA following ACS and PCI.
| Author | Age (years) | Sex | ACS type/PCI | SGLT2 inhibitor | Management of SGLT2 inhibitor | Onset of euDKA | Clue to diagnosis | Key diagnostic findings |
|---|---|---|---|---|---|---|---|---|
| Petersen C, et al. (2023) [5] | 28 | Male | NSTEMI/PCI | Empagliflozin | Initiated at discharge | Day 5 | Dyspnea, nausea, and vomiting | β-OHB: 9.0 mmol/dL pH 7.04 |
| Zughaib M, et al. (2023) [6] | 54 | Female | STEMI/PCI | Dapagliflozin | Continued use | Post-PCI | Nausea and vomiting | Ketonuria, ↑β-OHB (Values not reported) |
| Oriot P, et al. (2023) [7] | 77 | Female | STEMI/PCI | Empagliflozin | Continued use | Day 9 post-admission | Routine lab test (Asymptomatic) | β-OHB: 6.4 mmol/dL pH 7.22 |
| Yoshida Y, et al. (2025) [8] | 77 | Male | NSTEMI/PCI | Empagliflozin | Stopped on admission (1.5 days prior) | Post-PCI | Hemodynamic collapse | β-OHB: 4.3 mmol/dL pH 7.30 |
| Dai Z, et al. (2017) [9] | 49 | Male | STEMI/PCI | Ipragliflozin | Continued use (Stopped on diagnosis) | Acidosis present on arrival | Persistent unexplained acidosis | β-OHB: 6.8 mmol/dL pH 7.11 |
ACS, acute coronary syndrome; β-OHB, beta-hydroxybutyrate; euDKA, euglycemic diabetic ketoacidosis; STEMI, ST-segment elevation myocardial infarction; NSTEMI, non-ST-segment elevation myocardial infarction; PCI, percutaneous coronary intervention; SGLT2, sodium-glucose cotransporter 2.
While clinical trials, such as DAPA-MI and EMPACT-MI, have explored the beneficial effects of early initiation of SGLT2 inhibitors in MI patients, those who were already taking or had a plan to take SGLT2 inhibitors for MI were typically excluded [1]. A sub-analysis of the EMMY trial showed no significant difference in the incidence of adverse events, such as in glucose metabolism and DKA, when SGLT2 inhibitors were initiated within 24 h post-AMI compared to initiation at 24–48 or 48–72 h [10]. However, that study was not designed to assess outcomes in individuals with significant DKA risk factors, including a blood pH <7.32, hemodynamic instability, and ongoing SGLT2 inhibitor use.
Given the limited existing data and the experience from this case, it is essential to establish clear guidelines with specific criteria for resuming SGLT2 inhibitors in the post-AMI and post-PCI settings, including monitoring of blood or urinary ketones and blood gas analysis.
A limitation of this case report is that the serum ketone levels measured on day 6 were not significantly elevated, despite marked ketonuria on days 4 and 6, and urine specific gravity being at the upper end of the normal range.
In conclusion, this case highlights that the risk of SGLT2 inhibitor-associated euDKA should not be underestimated, particularly in patients with major physiological stressors, such as AMI.
This case contributes to the critically small body of existing evidence and highlights the urgent need for establishing clear guidelines for the safe use of SGLT2 inhibitors following AMI and PCI in diabetic patients. Further research into SGLT2 inhibitor-associated euDKA during AMI is warranted to prevent this serious complication.
Consent statement
Informed consent was obtained from the patient for publication of this case report, including the accompanying images.
Declaration of competing interest
The authors declare that there is no conflict of interest.
References
- 1.Maremmani M., Ebrahimi R., Centola M., Achilli F., Capone V., Bossone E., et al. Association of sodium-glucose cotransporter-2 inhibitors with mortality across the spectrum of myocardial infarction: a systematic review and meta-analysis. Cardiovasc. Diabetol. 2025;24(1):29. doi: 10.1186/s12933-025-02592-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Zhao Z., Zhao F., Zhang Y., Hu X., Li J., Tian C., et al. Risk factors of dapagliflozin-associated diabetic ketosis/ketoacidosis in patients with type 2 diabetes mellitus: a matched case-control study. Diabetes Res. Clin. Pract. 2023;196 doi: 10.1016/j.diabres.2023.110236. [DOI] [PubMed] [Google Scholar]
- 3.Laffel L. Ketone bodies: a review of physiology, pathophysiology and application of monitoring to diabetes. Diabetes Metab. Res. Rev. 1999;15(6):412–426. doi: 10.1002/(sici)1520-7560(199911/12)15:6<412::aid-dmrr72>3.0.co;2-8. [DOI] [PubMed] [Google Scholar]
- 4.Dhatariya K.K. The management of diabetic ketoacidosis in adults- an updated guideline from the Joint British Diabetes Society for Inpatient Care. Diabet. Med. 2022;39(6) doi: 10.1111/dme.14788. [DOI] [PubMed] [Google Scholar]
- 5.Petersen C., Gyabaah F., Sotelo J., Yohanna S., Deoker A. A case of prolonged recovery for post-percutaneous coronary intervention (PCI) sodium-glucose cotransporter-2 (SGLT2) inhibitor-induced euglycemic diabetic ketoacidosis in a 28-year-old. Cureus. 2023;15(9) doi: 10.7759/cureus.45180. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zughaib M., Basharat B., Small D. A case of STEMI-induced euglycemic diabetic ketoacidosis in a patient receiving a sodium glucose cotransporter-2 inhibitor. JACC: Case Reports. 2023;11 doi: 10.1016/j.jaccas.2023.101792. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Oriot P., Hermans M.P., Beauloye C., Rogghe P.A., Noel S., Paternotte E. Unsuspected diabetic ketoacidosis after myocardial infarction in a patient treated with SGLT2 inhibitor increased length of stay in the hospital: how can it be prevented? A case report. Eur. Heart J. Case Rep. 2023;7(8) doi: 10.1093/ehjcr/ytad336. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Yoshida Y., Fujimoto Y., Konishi H., Sato E., Iwata H., Ouchi S., et al. Euglycemic diabetic ketoacidosis in a patient with type 2 diabetes and preserved insulin secretion on empagliflozin. Intern. Med. 2025;64(14):2183–2188. doi: 10.2169/internalmedicine.4752-24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Dai Z., Nishihata Y., Kawamatsu N., Komatsu I., Mizuno A., Shimizu M., et al. Cardiac arrest from acute myocardial infarction complicated with sodium-glucose cotransporter 2 inhibitor-associated ketoacidosis. J. Cardiol. Cases. 2017;15(2):56–60. doi: 10.1016/j.jccase.2016.10.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.von Lewinski D., Kolesnik E., Aziz F., Benedikt M., Tripolt N.J., Wallner M., et al. Timing of SGLT2i initiation after acute myocardial infarction. Cardiovasc. Diabetol. 2023;22(1):269. doi: 10.1186/s12933-023-02000-5. [DOI] [PMC free article] [PubMed] [Google Scholar]