Euglycemic diabetic ketoacidosis associated with etogliflozin in post-pancreatitis diabetes: A case report

Abstract

BACKGROUND

Sodium-glucose cotransporter-2 (SGLT-2) inhibitors improve cardiovascular and renal outcomes in diabetes but may induce euglycemic diabetic ketoacidosis (euDKA) via insulin-independent mechanisms. Post-pancreatitis diabetes mellitus (PPDM) patients with impaired β-cell function face undefined risks with these agents.

CASE SUMMARY

A 29-year-old man with PPDM developed euDKA 1 week after initiating etogliflozin (5 mg/day). On admission, laboratory tests revealed blood ketones > 4.5 mmol/L, pH 7.1, and glucose 10.78 mmol/L. Discontinuation of SGLT-2 inhibitor, insulin pump therapy (basal 12 U/day, premeal bolus 4 U), aggressive hydration (6000 mL first 2 days), and nutritional support normalized ketosis and acidosis within 24 hours.

CONCLUSION

Caution is warranted with SGLT-2 inhibitors in PPDM. Insulin therapy is preferred to prevent euDKA.

Key Words: Sodium-glucose cotransporter-2; Etogliflozin; Euglycemic diabetic ketoacidosis; Acute pancreatitis; Post-pancreatitis diabetes; Case report

Core Tip: This is a first report of etogliflozin-induced euglycemic diabetic ketoacidosis (euDKA) in post-pancreatitis diabetes, which highlights risks of sodium-glucose cotransporter-2 (SGLT-2) inhibitors in β-cell compromised patients. Mechanistic analysis suggests that insulin replacement therapy should be prioritized over SGLT-2 inhibitors in post-pancreatitis diabetes mellitus to mitigate euDKA risk.

INTRODUCTION

Diabetes ketoacidosis (DKA) is a severe metabolic disorder caused by insufficient insulin and excessive counter-regulatory hormones. Clinically, blood glucose levels in ketoacidosis patients are usually 16.7–33.3 mmol/L, sometimes up to 55.5 mmol/L. The hospitalization rate of DKA patients has increased by 55% in the past decade[1]. In foreign countries, the in-hospital mortality rate of DKA in type 1 diabetes mellitus (DM) patients is 0.20%, and 1.04% in type 2 DM patients[2,3]. In China, the mortality rate of DKA patients is as high as 15.6%[4]. In recent years, sodium-glucose cotransporter-2 (SGLT-2) inhibitors have been widely used in patients with DM because of their significant cardiovascular and renal benefits. However, they may reduce blood glucose through insulin-independent glucose clearance mechanisms and hyperglucagonemia, and it cannot be ignored that this class of drugs may induce euglycemic DKA (euDKA)[5].

Acute pancreatitis (AP) is a common disease in clinical practice. Its chronic progression increases the incidence of DM. It is reported that 23% of AP patients develop DM within 3 years after discharge[6,7]. The incidence of DM at 6, 12, 18 and 24 months after AP is 3%, 7%, 9% and 11%, respectively[8]. Therefore, DM is a common complication in patients with chronic pancreatitis. DM caused by pancreatic diseases is called pancreatogenic diabetes, which the European guidelines term as type 3c diabetes[9]. Some literature also refers to it as post-pancreatitis DM (PPDM)[10]. Insulin therapy is preferred for PPDM. A recent cohort study used five categories of oral hypoglycemic drugs including SGLT2 inhibitors for PPDM and found that all effectively reduced glycosylated hemoglobin levels. However, in patients with pancreatic exocrine insufficiency, the hypoglycemic efficacy significantly decreases[11].

Etogliflozin, similar to other SGLT2 inhibitors, may induce euDKA or even cause death. However, euDKA induced by etogliflozin in PPDM patients has not been reported. This article reports the diagnosis and treatment of a patient with chronic pancreatitis who developed severe euDKA after being diagnosed with DM and treated with etogliflozin, and provides an updated literature review.

CASE PRESENTATION

Chief complaints

A 29-year-old man had dizziness and fatigue for 1 week after starting etogliflozin.

History of present illness

The patient presented to the emergency department with dizziness and fatigue for 1 week, occurring after initiating etogliflozin 5 mg/day (1 week prior). Symptoms worsened over 3 days, prompting urgent evaluation. He had no history of vomiting, abdominal pain or fever.

History of past illness

The patient was a 29-year-old man with a 4-year history of recurrent AP, who developed PPDM 1 year prior to presentation. His prior hypoglycemic regimen included irregular use of metformin, acarbose, and short-acting insulin without documented adherence issues or previous episodes of DKA. He also had a history of hyperlipidemia treated with fenofibrate. Medications before the occurrence of euDKA included metformin, acarbose, insulin, and SGLT-2 inhibitor etogliflozin (5 mg/day), which were started 1 week before admission. The patient had no history of alcohol abuse, smoking or abdominal surgery, and no known drug allergies. This medical history highlights the progression from chronic pancreatitis to PPDM and the recent introduction of etogliflozin as a potential trigger for euDKA.

Personal and family history

The patient denied any family history of malignant tumors.

Physical examination

Body temperature 36.1 °C; pulse 96 beats/minute; respiration 20 breaths/minute; blood pressure 116/68 mmHg; body mass index 18.38 kg/m². The patient showed normal development, lean body type, autonomous position, clear consciousness, poor spirit, normal speech, and appropriate answers. There was no yellowing or rash on the skin and mucous membranes. The lungs were clear on percussion, with rapid breathing, acetone odor in exhaled breath, and no dry or wet rales. Heart rate was 96 beats/minute, with regular rhythm, strong heart sounds, and no pathological murmurs in each valve area. Abdominal examination was normal, and there was no percussion pain in the kidney area. There was no edema in the extremities.

Laboratory examinations

On admission, blood ketones were elevated at > 4.5 mmol/L with metabolic acidosis (pH 7.1, carbon dioxide combining power < 5 mmol/L) and normoglycemia (glucose 10.78 mmol/L). Subsequent testing on the next day revealed profound insulinopenia (insulin < 1.6 μIU/mL, C-peptide 0.31 ng/mL), improved glucose (7.23 mmol/L), partial acidosis correction (carbon dioxide combining power 18.8 mmol/L), and persistent ketosis (blood ketones 3.67 mmol/L). Glycosylated hemoglobin was 10.2%, while autoantibodies against islet cells, insulin, and glutamic acid decarboxylase were all negative. These findings confirmed euDKA secondary to pancreatic β-cell dysfunction and SGLT-2 inhibitor use. The changes in blood ketone levels after admission are shown in Table 1.

Table 1 Changes in blood ketone levels after admission.

	1 day	2 days	3 days	4 days	5 days
Blood ketone levels (mmol/L)	> 4.5	3.67	1.08	0.43	0.16

Imaging examinations

Upper abdominal computed tomography performed prior to admission demonstrated abnormal pancreatic and peripancreatic changes consistent with chronic pancreatitis with acute exacerbation, corroborating the clinical diagnosis of pancreatitis recurrence.

FINAL DIAGNOSIS

PPDM, euDKA, chronic pancreatitis with acute exacerbation and hyperlipidemia.

TREATMENT

After admission, metformin, etogliflozin and acarbose were discontinued. Insulin pump continuous subcutaneous injection of aspart insulin was administered (total basal rate 12 U, premeal bolus 4 U each), blood glucose was closely monitored, and insulin dose was adjusted. Large amounts of fluid were infused (about 6 L/day in the first 2 days), calories were actively supplemented, fenofibrate was used for lipid-lowering treatment, and other symptomatic treatments were provided. On the next day, dizziness and fatigue significantly improved, acidosis was corrected, and blood ketone bodies were decreased.

OUTCOME AND FOLLOW-UP

The patient was discharged 1 week later, using subcutaneous injection of glargine insulin and aspart insulin, oral acarbose, and metformin. During follow-up, blood glucose was well controlled, and no recurrence of ketoacidosis occurred.

DISCUSSION

The pathophysiological mechanism of PPDM is not fully understood. Multiple factors may contribute to glucose metabolic disorders, including: Islet cell damage, AP-induced autoimmune response, hyperlipidemia, local and systemic inflammatory responses, and changes in the insulin–incretin axis[12].

Islet cell damage: Some AP patients have extensive pancreatic necrosis, leading to reduced pancreatic β-cell mass and subsequent relative insulin deficiency. Zhi et al[10] supported this theory, showing that the incidence of AP-related DM in patients with necrotizing pancreatitis was higher than in those without extensive necrosis (37% vs 11%). AP-induced autoimmune response: Although the role of immune activation in PPDM has not been widely studied, β-cell autoantibodies are positive in PPDM patients[13]. It is speculated that the intense inflammatory response in AP may cause β-cell apoptosis by exposing self-antigens such as insulin and islet nucleic acid, leading to the production of new autoantibodies[14]. Hyperlipidemia: A study involving 1257 patients showed that hyperlipidemia increased the incidence of new-onset diabetes (odds ratio 2.55)[15]. Hyperlipidemia is independently associated with the risk of AP severity[16], suggesting it is a pathogenic factor for PPDM. Local and systemic inflammatory responses: AP can cause severe local or systemic inflammatory reactions with abundant inflammatory factors. Although this inflammation is transient in most people, it becomes chronic and persistent in some patients. C-reactive protein and interleukin-6 are strongly positively correlated with DM development[17], implying that persistent inflammation after pancreatitis is a key factor in PPDM. Changes in the insulin–incretin axis: AP patients may have pancreatic exocrine dysfunction, disrupting the enteroinsular axis and impairing the secretion of incretin hormones. This leads to altered secretion of glucose-dependent insulinotropic polypeptide and glucagon-like peptide-1, resulting in hyperglycemia. Pancreatic enzyme replacement therapy can partially reverse this state[18].

In this case, the patient's C-peptide and insulin levels were significantly below the reference range, likely due to repeated local and systemic inflammatory reactions causing pancreatic β-cell damage. Although the patient's anti-islet cell, anti-insulin and anti-glutamic acid decarboxylase antibodies were all negative, we cannot exclude the possibility that repeated β-cell destruction exposed abundant endogenous proteins, leading to the production of unknown antibodies mediating β-cell damage.

DKA is caused by absolute or relative insulin deficiency and elevated glucagon levels. These hormonal changes accelerate β-oxidation of free fatty acids (FFAs), leading to excessive ketone body production, while reducing ketone utilization in other tissues[19]. SGLT-2 inhibitors lower blood glucose by increasing urinary glucose excretion, reducing pancreatic β-cell insulin secretion, thereby decreasing the antilipolytic effect of insulin and stimulating FFA production, exacerbating the above process and further increasing blood ketone levels[20]. SGLT-2 inhibitors can also trigger glucagon secretion by activating potassium channels, promoting ketone production[20]. These drugs may simulate starvation states and induce increased ketogenesis, shifting energy metabolism from glucose to lipid[21]. Additionally, the human SGLT family (SLC5) includes 12 genes, 10 of which are Na⁺/substrate cotransporters involved in glucose, inositol, ketone, and anion transport. SGLT-2 inhibitors may interact with these cotransporters, affecting renal ketone excretion[22]. In this case, the patient used metformin, acarbose and insulin, without developing significant ketoacidosis. However, after adding etogliflozin, euDKA rapidly developed. There are several possible reasons: (1) Pancreatic β-cell dysfunction in PPDM: SGLT-2 inhibitors further reduced insulin secretion; (2) Impaired renal ketone clearance by SGLT-2 inhibitors; and (3) Drug-induced starvation-like state promoting lipid metabolism and ketone accumulation. Reduced renal gluconeogenesis due to increased urinary glucose excretion, resulting in normal blood glucose levels despite ketosis[20]. Additionally, α-glucosidase inhibitors delayed carbohydrate absorption, and pancreatic exocrine insufficiency aggravated carbohydrate intake impairment, potentially contributing to euDKA.

Cases of PPDM and SGLT-2-inhibitor-induced euDKA have been reported in domestic and foreign journals, but there are no reports of PPDM patients using SGLT-2 inhibitors. Etogliflozin, the fourth SGLT2 inhibitor globally launched in December 2017, has proven hypoglycemic efficacy with low hypoglycemia risk and dual cardiovascular and renal protective effects, leading to increasing use in type 2 DM patients. To avoid inappropriate use endocrinologists should thoroughly review patient histories. For patients with a history of pancreatitis, islet function should be promptly evaluated, paying attention to body habitus, energy intake, and stress status. We should avoid empirically selecting SGLT2 inhibitors based on type 2 DM treatment strategies. During follow-up, islet function changes should be monitored and insulin therapy adjusted for significant β-cell decline. For gastroenterologists and emergency physicians: Monitor blood glucose in patients with recurrent pancreatitis and assess type 3c diabetes. Use insulin instead of oral hypoglycemic agents (especially SGLT2 inhibitors) for hyperglycemic patients, and promptly refer for islet function assessment and endocrine consultation to avoid severe complications like euDKA/DKA. Notably, while SGLT-2 inhibitors are unapproved for type 1 diabetes and special subtypes like PPDM in China, international guidelines support their use in type 1 diabetes. Both conditions involve islet dysfunction, but unlike type 1 diabetes with irreversible autoimmune β-cell damage, PPDM lacks associated autoantibodies, allowing potential islet recovery. This suggests that SGLT-2 inhibitors should be carefully used in PPDM under strict monitoring. Based on type 1 diabetes treatment experience, these drugs may be tried in PPDM while monitoring glucose, ketones, islet function, and pancreatitis markers. This could control blood sugar and maximize cardiovascular/renal benefits. However, further research is needed to confirm their safety and efficacy in PPDM.

CONCLUSION

Caution is warranted with SGLT-2 inhibitors in PPDM. Insulin therapy is preferred to prevent euDKA.