Metabolic ketoacidosis with normal blood glucose: A rare complication of sodium-glucose cotransporter 2 inhibitors.

Ketoacidosis is a significant and often a life-threatening complication of diabetes mellitus seen mostly in type 1 diabetes mellitus as well as occasionally in type 2 diabetes mellitus. Diabetic ketoacidosis usually manifests with high blood glucose more than 250 mg/dL, but euglycemic diabetic ketoacidosis is defined as ketoacidosis associated with blood glucose level less than 250 mg/dL. Normal blood glucose in such patients results in significant delay in diagnosis and management of diabetic ketoacidosis, thus increasing mortality and morbidity. We present a case of euglycemic diabetic ketoacidosis secondary to canagliflozin in a type 2 diabetic patient.

Introduction

Diabetic ketoacidosis (DKA) is a life-threatening complication of diabetes mellitus that mainly occurs in patients with type 1 diabetes mellitus but also in patients with type 2 diabetes mellitus. DKA is a complicated metabolic disorder characterized by ketoacidosis, marked hyperglycemia (>250 mg/dL), and anion gap metabolic acidosis which is induced as a result of a significant deficiency of insulin and rise in counter regulatory hormones. In some instances, it does occur in the presence of normal glucose levels. This latter form of DKA is known as euglycemic DKA, which is defined as ketoacidosis with blood glucose levels below 250 mg/dL.[1]

Sodium–glucose cotransporter 2 (SGLT-2) inhibitors like canagliflozin were approved by Food and Drug Administration (FDA) back in 2013 to improve glycemic control in patients with type 2 diabetes mellitus. Recent studies[1] suggest that canagliflozin is not so infrequently associated with increased incidence of euglycemic DKA in patients with type 2 diabetes mellitus. We present a rare case of euglycemic DKA in a 40-year-old female secondary to canagliflozin use. All findings were consistent with canagliflozin-induced DKA.

Case report

A 40-year-old woman with past medical history of type 2 diabetes mellitus and gastroesophageal reflux disease (GERD) presented to the emergency department because of increasing erythema, swelling, and pain in her right groin. She denied any fevers, chills, sexually transmitted disease history, or inguinal lymphadenopathy. On examination, there was redness with swelling on the right side of her mons pubis radiating down into the right labia majora with no associated lymphadenopathy. A diagnosis of cellulitis was made for which incision and drainage was performed and patient was started on IV antibiotics. Drainage culture showed methicillin-resistant Staphylococcus aureus (MRSA) and antibiotics were tailored accordingly. Her baseline investigations at the time of admission were unremarkable except for elevated white cell count of 19,000 cells/µL/cu mm. She was responding well to treatment; however, on the second day of hospitalization, her bicarbonate level dropped to 7 Meq/L. Arterial blood gas was ordered which revealed pH of 7.2, PCO2 of 21 mm Hg, and HCO3 of 8 mmol/L. Anion gap was calculated to be 18 mmol/L, but lactic acid, urine electrolytes, and kidney and liver functions were within normal limits. She was given IV fluid with bicarbonate without any improvement in her bicarbonate level. In the mean time, she started having labored breathing for which she was admitted in the intensive care unit (ICU). Beta hydroxybutyrate was ordered which was remarkably high (8.86 mmol/L). Surprisingly, her blood glucose levels were consistently below 250 mg/dL throughout the hospital stay. Her home medications included pantoprazole and canagliflozin, which were continued during the hospital stay. She had been on the same medications for years except canagliflozin that was started 1 year ago. Canagliflozin was immediately stopped and she was given supportive treatment. He bicarbonate level improved to normal within 2 days.

Discussion

In March 2013, FDA approved oral SGLT-2 inhibitors including canagliflozin, dapagliflozin, and empagliflozin to improve glycemic control in adult patients with type 2 diabetes mellitus.[2] Canagliflozin acts primarily on SGLT-2 receptors in the kidneys, inhibiting the reabsorption of filtered glucose leading to glucosuria without gastrointestinal adverse effects.[3] Recent studies demonstrated that canagliflozin 100 mg and canagliflozin 300 mg provide sustained and substantial reductions in HbA1c.[4]

SGLT-2 inhibitors have found to be associated with a number of adverse events. The major adverse effects associated with canagliflozin use are urinary tract infections, genital mycotic infections, osmotic diuresis, hypotension, and vulvovaginal pruritus. Canagliflozin was also recently found to be associated with ketoacidosis. This ketoacidosis unlike typical DKA is associated with glucose levels <250 mg/dL, which often makes diagnosis extremely difficult. The incidence of euglycemic DKA in patients with type 1 diabetes mellitus was found to be 5.1% and 9.4% treated with 100 and 300 mg of canagliflozin, respectively.[5] Unlike type 1 diabetes mellitus, the overall incidence of ketoacidosis was found to be only 0.7% in patients with type 2 diabetes mellitus.[6]

The mechanism for canagliflozin-induced ketoacidosis is unclear. SGLT-2 inhibitors act primarily on SGLT-2 receptors present in the proximal tubules of the kidneys increasing urinary glucose excretion, which in turn reduces insulin secretion from pancreatic β-cells. The decrease in circulating insulin levels results in a significant reduction in antilipolysis and consequent overproduction of free fatty acids. These free fatty acids are eventually converted to ketone bodies by β-oxidation in the liver. Evidence also suggests that the administration of SGLT-2 inhibitors stimulates the secretion of glucagon,[7] which might either be secondary to the decrease in insulin secretion or a direct stimulation of pancreatic alpha cells by SGLT-2 inhibitors.[5] Moreover, glucagon by inhibiting acetyl-CoA carboxylase and thereby increasing carnitine palmitoyltransferase I (CPT-I) activity in the liver further augments the overproduction of ketone bodies.

Patients usually present with high anion gap metabolic acidosis, increased ketone bodies, and blood glucose <250 mg/dL. Our patient with past medical history of type 2 diabetes mellitus presented with high anion gap metabolic acidosis with HCO3 in the range of 4–12 Meq/L and glucose levels ranging from 150 to 250 mg/dL and beta hydroxybutyrate 8.9 mmol/L. All these findings are consistent with euglycemic DKA.

Given that euglycemic DKA is not necessarily associated with typical manifestations of DKA induced by marked hyperglycemia, the diagnosis at times might easily be missed based only on clinical signs. Even though these patients present with normal to low blood glucose levels, severe metabolic acidosis alone has the potential to become a life-threatening condition. Increasing incidence of such events warrants thorough research work on the possible mechanism and safety profile of all SGLT-2 inhibitors including canagliflozin. Detailed clinical information on related DKA cases including the possible mechanism and pathophysiology might help to provide a more solid basis for the safe, appropriate, and wide application of this newer class of drugs.

Acidosis in a patient with type 2 diabetes mellitus on SGLT-2 inhibitors should raise the suspicion of euglycemic DKA.

Euglycemic DKA may present with severe acidosis and normal blood glucose, thus making its diagnosis cumbersome.