Log In
Forgot username? Forgot password? New User? Sign Up Free
SCCM is performing maintenance on its websites. For the best browsing experience, please use Microsoft Edge or Safari. Those using Chrome or Firefox may experience access issues at this time.
Patients taking SGLT2 inhibitors are at increased risk for diabetic ketoacidosis (DKA). Clinicians often miss the diagnosis of DKA when blood glucose levels are near normal. Patients with euglycemic DKA are more likely to develop hypoglycemia after initiation of insulin infusion therapy.
Critically ill patients are at risk of complications when glycemic control is poorly managed. The risk of mortality rises with either hyperglycemia, generally defined as a blood glucose level greater than 140 mg/dL, or hypoglycemia, defined as a blood glucose level less than 70 mg/dL.1 To avoid complications from hyper- and hypoglycemia, the optimal glycemic target is generally between 140 and 180 mg/dL.2 Diabetic ketoacidosis (DKA) is a life-threatening condition consisting of blood glucose level greater than 250 mg/dL, ketosis, anion gap metabolic acidosis, and clinical evidence of dehydration. Clinicians may also be misled by the presence of pseudonormoglycemia. Euglycemic DKA (EDKA) is a variant of DKA in which blood glucose levels are near normal despite metabolic acidosis and ketosis. This typically occurs in the setting of insulin deficiency, reduced gluconeogenesis, physiologic stress, or infection. Sodium-glucose cotransporter 2 (SGLT2) inhibitors have been increasingly reported as a cause of DKA and EDKA.3 Five SGLT2 inhibitors (canagliflozin, dapagliflozin, ertugliflozin, empagliflozin, and bexagliflozin) are currently approved by the U.S. Food and Drug Administration (FDA) to achieve target hemoglobin A1C levels while improving other cardiovascular and renal risk factors.4 A study by Sell et al compared EDKA to hyperglycemic DKA.5 The researchers performed a retrospective review of adult patients admitted to an academic medical center in Michigan from August 2015 through October 2022 for DKA based on pH less than 7.3, serum bicarbonate less than18 mmol/L, and anion gap greater than 10 mmol/L. Patients were then divided into EDKA (< 250 mg/dL) versus hyperglycemic DKA (> 250 mg/dL) cohorts based on the presenting glucose level. Patient demographics, resource utilization, and clinical outcomes (length of stay, disposition) were collected. Researchers also collected the time from emergency department presentation to first pH greater than 7.3 and first bicarbonate greater than 18 mmol/L. Among the 629 patients with DKA, 44 had EDKA and 585 were hyperglycemic. All the patients were managed in an emergency department-based intensive care unit (ICU). Patients with starvation ketosis were differentiated from patients with EDKA based on the resolution of ketosis with glucose supplementation and the lack of a diabetes diagnosis before or during their admission. Mean age and gender demographics were relatively similar. Patients with EDKA generally had higher pH (7.17 vs. 7.14), higher bicarbonate (11.9 mmol/L vs. 10.4 mmol/L), lower anion gap (21.6 mmol/L vs. 26.3 mmol/L, P < 0.05), and lower serum potassium levels (4.3 mmol/L vs. 5.3 mmol/L, P < 0.001). The etiologies of EDKA typically included insulin use shortly before hospitalization (57%), poor oral intake (29%), and SGLT2 inhibitors (14%). The average time to obtain a bicarbonate level greater than 18 mmol/L after insulin administration was similar in both groups (12.3 hrs vs. 12.1 hrs). After initiation of insulin infusion, patients with DKA had a significantly higher rate of hypoglycemia with blood glucose levels less than 70 mg/dL (18.2% vs. 4.8%, P = 0.02) and hypokalemia (27.3% vs. 19.1%, P = 0.23). The duration of insulin infusion was 5.9 hours shorter in the EDKA group. Hospital length of stay was shorter for patients with EDKA (2.2 days vs. 3.9 days, P = 0.03) but similar to ICU length of stay (16.4 hrs vs. 16.6 hrs). Patients with EDKA had a higher rate of discharge from the emergency department after resolution of DKA (57% vs. 33%), and the remaining patients were admitted to medical wards or the ICU. None of the patients with EDKA required subsequent ICU admission, while 5.1% of the hyperglycemic patients did. None of the euglycemic patients with DKA died with therapy, while five (0.9%) of the hyperglycemic patients died. EDKA is responsible for about 3.2% of DKA admissions, which poses a challenge to clinicians because the diagnosis of DKA may be overlooked in a patient with a metabolic acidosis and ketonemia.6 Like patients with DKA, the significant volume depletion in patients with EDKA occurs due to the osmotic diuresis from glucosuria and is further exacerbated by reduced oral intake and vomiting.7 The concept of relative dysglycemia, a diabetic complication that occurs when the body’s ability to detect and respond to hypoglycemia increases, has recently gained recognition in critically ill patients. Okazaki et al reported significantly increased mortality when the maximum and minimum glycemic ratio increases during the first 24 hours after ICU admission.8 Since the American Heart Association guidelines gave SGLT2 inhibitors a Class 1 recommendation for its use in patients with heart failure, SGLT2 inhibitor use has increased.9 Clinicians need to be aware of the association of SGLT2 inhibitors with euglycemic and hyperglycemic DKA. While the FDA’s adverse reporting system revealed a sevenfold increased risk of DKA in the setting of SGLT2 inhibitor use, studies on the association between SGLT2 inhibitors and EDKA remain scarce.10 Clinicians should be attentive to EDKA in patients with diabetes, metabolic acidosis, and ketosis, especially those taking SGLT2 inhibitors. Clinicians should also be attentive to the higher risk of hypoglycemia soon after insulin infusion therapy is initiated in patients with EDKA. References:
Posted: 9/17/2024 | 0 comments
Log in to Comment