New Kids on the Block: SGLT2 Inhibitors and Euglycemic DKA

SGLT2 inhibitors are oral hypoglycemic drugs approved for use in patients with T2D. Sodium-glucose cotransporters are a family of glucose transporters found in the intestinal mucosa (SGLT1) and in the proximal tubules of nephrons (SGLT2). Figure 1 illustrates how SGLT’s function as an efficient filter for glucose reabsorption. By piggybacking off the energy gradient created by Na+/K+-ATPase pumps on the basolateral membrane, glucose molecules are transported across the apical membrane via secondary active transport. Then, GLUT uniporters pump the glucose into the peritubular capillaries for reabsorption.

By inhibiting SGLT2 in the setting of hyperglycemia, there is a saturation of the transporters, an overall reduction in glucose reabsorption from the tubular lumen, and increased urinary excretion of glucose – thus reducing plasma glucose levels.^2-4

Fig. 1: Normal SGLT function in proximal renal tubules

(Wright EM, Loo D, Hirayama B. Physiological Reviews 2011; 91(2): 733-794.)

Although the mechanism is unclear, another benefit of SGLT2 inhibitors is lowering of blood pressure and weight loss. Because of its non-insulin-dependent mechanism of action, SGLT2 inhibitors are an exciting therapeutic option for those with inadequate glycemic control without causing hypoglycemia.⁴ Despite the promising data currently presented regarding the SGLT2 inhibitors, there is still a lack of data regarding long-term safety and therapeutic outcomes.

SGLT2 inhibitor-induced euDKA was first reported by Peters, et al. in a case report of 13 episodes in nine patients. Seven patients had T1D (used off-label) and two patients had T2D. Due to the absence of significant hyperglycemia, there was a delay in recognition of euDKA, and the patients were worked up as food poisoning, migraines, and other non-specific disease states. The authors concluded that euDKA was triggered by initially decreased or unchanged insulin doses, low caloric and fluid intake, and concurrent illnesses. In combination with these triggers, other clinical signs and symptoms associated with euDKA were any malaise, nausea, vomiting, shortness of breath, and ketonuria or ketonemia.

So how exactly does euDKA happen?

The pathogenesis of DKA has already been well-established. By definition, DKA can occur when glucose levels > 250 mg/dL. Briefly, traditional DKA is characterized by a hyperglycemic state with absolute or relative insulin deficiency plus increased counter-regulatory hormones. This imbalance leads to decreased glucose utilization – therefore catalyzing hepatic gluconeogenesis, glycogenolysis, and lipolysis. Free fatty acids are released and delivered to the liver for β-oxidation, ketone bodies are produced while continuing to increase plasma glucose levels – thus, propagating a vicious cycle.⁷ (Fig 2) Eventually, this ketotic and hyperosmolar state leads to anion gap acidosis and severe dehydration.

Fig 2 & 3: DKA and euDKA pathophysiology

EGP = endogenous glucose production; TGD = tissue glucose disposal; UGCr = urinary glucose clearance rate

(Rosenstock J, Ferrannini E. Diabetes Care 2015; 38(9): 1638-1642.)

In contrast to traditional DKA, SGLT2 inhibitor-induced ketoacidosis has a different pathophysiology and mechanism. Euglycemic DKA was first defined in 1973 by Munro, et al. as DKA but with plasma glucose < 300 mg/dL in T1D patients only. The primary cause was thought to be reduced carbohydrate availability in conjunction with inadequate insulin therapy.⁹ SGLT2 inhibitors cause a rapid and prolonged urinary excretion of glucose – approximately 20-40% of consumed carbohydrates are lost. This excessive excretion causes a carbohydrate and volume deficiency, thereby decreasing the stimulus for insulin release and worsening dehydration. Through unknown mechanisms, SGLT2 inhibitors also increase glucagon, cortisol, and epinephrine, furthering insulin resistance, lipolysis, and ketogenesis. Although the half-life of these drugs is approximately 12 hours, the pharmacodynamics effects and terminal half-life may actually be closer to 24 hours. For this reason, gluconeogenesis does not contribute to the hyperglycemia generally associated with traditional DKA, since plasma glucose continues to be filtered out and excreted. In general, insulin resistance, endogenous glucose production, and underutilization of glucose in tissues are milder in euDKA than in DKA.^{6, 7, 9}

Management of euDKA is essentially the same as traditional DKA. Insulin therapy with aggressive fluid resuscitation should be initiated. SGLT2 inhibitor therapy should be held and not restarted without a proper endocrine evaluation. Two patients in the Peters, et al. study had recurrences of euDKA with rechallenge of their SGLT2 inhibitor therapy and required readmission for treatment.⁸

Invokana (canagliflozin)

Invokamet (canagliflozin + metformin)
Farxiga (dapagliflozin)
Xigduo XR (dapagliflozin + metformin)
Jardiance (empagliflozin)
Glyxambi (empagliflozin + linagliptin)
Synjardy (empagliflozin + metformin)

https://www.accessdata.fda.gov/scripts/medwatch/index.cfm?action=reporting.home

• SGLT2 inhibitors are a non-insulin dependent novel class antidiabetic agents approved for use in Type 2 diabetes patients.
• SGLT2 inhibitors decrease plasma glucose levels by increasing urinary glucose excretion.
• DKA is driven by insulin resistance and underutilization of glucose by tissues. SGLT2 inhibitor-induced euglycemic DKA is driven by excessive glucose excretion and resulting decreased insulin production, which decreases tissue uptake of glucose. Ketogenesis occurs the same in both disease states via lipolysis and beta-oxidation.
• Clinical clues suggesting euDKA: using SGLT2 inhibitor, recent decreased food and fluid intake, recently reduced insulin regimen, general malaise, nausea, vomiting, ketonuria and/or ketonemia, elevated anion gap.
• Clinicians should be aware of these patterns and signs to appropriately identify euDKA and provide rapid treatment.
• Report SGLT2 inhibitor-related events to MedWatch.