ED Hyperglycemia (Without HHS or DKA): What to Lower, Who to Admit, and Who’s at Risk for Future DKA

ED Hyperglycemia (Without DKA): What to Lower, Who to Admit, and Who’s at Risk for Future DKA Bottom line up front: For adults who present to the emergency department (ED) with severe hyperglycemia but no DKA/HHS, the best-supported priorities are (1) ruling out hyperglycemic crises and precipitants, (2) symptom control and hydration, and (3) arranging reliable follow-up and appropriate outpatient therapy. Aggressively “chasing a number” in the ED has not been shown to improve short-term outcomes, and the specific discharge glucose value—within studied ranges—doesn’t predict 7-day return or hospitalization (2,3). 1) What does the evidence say about lowering glucose in the ED (when the patient is not in DKA/HHS)? Observational cohort (n=566 ED encounters). Among patients discharged with arrival glucose ≥400 mg/dL (mean 491), discharge glucose was not associated with 7-day ED revisit for hyperglycemia or hospitalization; only 2 patients developed DKA within 7 days. Authors concluded that attaining a specific glucose goal before discharge “may be less important than traditionally thought” (2). Randomized trial. Adults with hyperglycemia 400–600 mg/dL (excluding type 1 diabetes) were randomized to a discharge goal <350 mg/dL vs <600 mg/dL. ED length of stay and 7-day outcomes were similar; tighter targets marginally prolonged LOS (3). Treatment intensity vs benefit. In a paired analysis, ~10 units of subcutaneous insulin and ~1 L IV crystalloid were associated with modest average glucose reductions; IV fluids were associated with longer ED length of stay. This supports a measured approach focused on comfort, safety, and follow-up—not necessarily achieving a specific ED number (4). Bridging outpatient therapy. For stable type 2 diabetes without crisis, small prospective work supports simple discharge regimens (e.g., sulfonylurea with/without low-dose basal insulin) as a safe bridge to close follow-up (6). Clinical take. Once DKA/HHS and acute precipitants are excluded, ED glucose reduction offers symptom relief but has not shown short-term outcome advantages; avoid iatrogenic hypoglycemia and unnecessary boarding (2–4,6). 2) Is there a correlation between how high the glucose is today and future DKA? Single ED glucose as a predictor. Available data indicate no clear short-term risk signal based solely on discharge glucose among patients well enough for ED discharge (400–600 mg/dL studied); only 2/422 developed DKA within 7 days in the cohort above, and risk was not tied to the discharge number (2). “Sentinel” ED visits. Population work shows a meaningful fraction of patients admitted with DKA/HHS had an ED visit in the prior 14 days, often discharged without a glucose check or with persistent hyperglycemia—suggesting missed opportunities for intervention and follow-up, not that a specific single value predicts DKA (7). What does predict recurrent severe hyperglycemia? Younger age, prior recent hyperglycemia visit, insulin use, and very high glucose (>20 mmol/L ≈ 360 mg/dL) predict 30-day recurrent ED hyperglycemia—which is not the same as predicting DKA specifically (8). Clinical take. A single high glucose (even very high) is a poor stand-alone predictor of imminent DKA. Future DKA is driven more by pathophysiology and context (insulin deficiency/omission, intercurrent illness, access to care, SGLT2 use, etc.) than by one ED number (1,5,7). 3) Is it appropriate to discharge someone with a blood glucose of 600 mg/dL or more? HHS screen first. Modern consensus identifies HHS by severe hyperglycemia (classically >600 mg/dL), hyperosmolality, dehydration, and minimal ketosis; 2024 guidance refines diagnostic and resolution criteria and emphasizes effective or measured osmolality and mental status. Anyone near or above 600 mg/dL demands an HHS evaluation (serum osmolality, mental status, volume status, electrolytes, ketones, VBG) (1). Evidence gap for ≥600 at discharge. The RCT above allowed a “loose” ED target <600 mg/dL and found outcomes similar to <350 mg/dL, but did not study discharging at ≥600 mg/dL. There is no supportive evidence to discharge at or above 600 mg/dL, and doing so risks missing evolving HHS (3). Practical position. If ≥600 mg/dL at any time, don’t discharge until: HHS is excluded (effective osmolality below hyperosmolar range and mentation normal), volume status is corrected, glucose is clearly down-trending on a reliable regimen, a trigger is addressed, and reliable follow-up (24–72 h) plus supplies/education are in place. Absent those, admit or observe (1,3). 4) Who gets DKA, and why? Core pathophysiology. DKA arises from relative or absolute insulin deficiency with counter-regulatory hormone excess driving lipolysis and ketogenesis; HHS reflects enough insulin to suppress ketones but not hyperglycemia/osmotic diuresis (1). Common precipitants (adults). Infection, missed/insufficient insulin (including pump failure), new-onset diabetes, steroid bursts, acute vascular events; socioeconomic disadvantage (low income, homelessness, under-insurance) also drives crises/readmissions (1,7). Medication-associated risk. SGLT2 inhibitors increase DKA risk—including euglycemic DKA—in type 1 (off-label) and type 2 diabetes; risk is amplified by low-carb/fasting, dehydration, alcohol, peri-operative states, and insulin-deficient phenotypes (5). Who is higher risk? Type 1 diabetes (especially younger adults), people with high A1c/poor control, prior DKA, pump issues, and those with limited access to care (1,7). 5) A practical ED approach (non-DKA/HHS hyperglycemia) Screen for crises and triggers. Vitals/mentation; CMP, β-hydroxybutyrate (or urine ketones), VBG/bicarbonate, osmolality; urinalysis/infection workup as indicated; ECG/troponin if symptoms. Use 2024 consensus criteria to exclude DKA/HHS (1). Symptom-guided treatment. IV fluids for dehydration; modest SC insulin for symptomatic relief—recognize the limited incremental glucose fall and potential LOS increase with fluids; avoid hypoglycemia (4). Disposition. Admit/observe if: HHS or DKA suspected; unreliable follow-up; significant comorbidity; social barriers; pregnancy; older/frail; SGLT2-associated euglycemic DKA concern; or inability to self-manage (1,5). Consider discharge when: Clinically stable; DKA/HHS excluded; glucose improving on a clear regimen; education provided; and rapid follow-up arranged. A specific discharge number (<350 vs <600) has not altered 7-day outcomes in studied adults—but evidence does not support discharge at ≥600 mg/dL (2,3). Bridge therapy & follow-up. For likely type 2 diabetes without crisis, consider initiating a simple oral regimen ± low-dose basal insulin with supplies and 24–72 h follow-up (6). Risk-reduction counseling. Sick-day rules, pump troubleshooting, temporary SGLT2 holds during acute illness/fasting/surgery, and social-work support for access barriers; these reduce recurrent crises (1,5,7). 6) Why the number isn’t the disease Modern guidance stresses that hyperglycemic crises are defined by physiology (ketosis, acidosis, osmolality, mental status), not by any single glucose cut-off. Treat the patient and pathophysiology; use the number to triage and trend, not as the sole determinant of disposition (1–4,7). References: Umpierrez GE, Davis GM, ElSayed NA, et al. Hyperglycemic Crises in Adults With Diabetes: A Consensus Report. Diabetes Care. 2024;47(8):1257–1275. PMID: 39052901. https://pubmed.ncbi.nlm.nih.gov/39052901/ (Open-access PMC: https://pmc.ncbi.nlm.nih.gov/articles/PMC11343900/) Driver BE, Olives TD, Bischof JE, et al. Discharge Glucose Is Not Associated With Short-Term Adverse Outcomes in ED Patients With Moderate to Severe Hyperglycemia. Ann Emerg Med. 2016;68(6):697-705.e3. PMID: 27353284. https://pubmed.ncbi.nlm.nih.gov/27353284/ Driver BE, Amin A, Greene S, et al. Comparison of Two Glycemic Discharge Goals in ED Patients With Hyperglycemia (Randomized Trial). Am J Emerg Med. 2019;37(8):1531–1536. PMID: 30316635. https://pubmed.ncbi.nlm.nih.gov/30316635/ Driver BE, Greco K, Challen E, et al. Association of ED Treatments for Hyperglycemia With Glucose Reduction and Length of Stay. J Emerg Med. 2017;53(6):791–797. PMID: 28993036. https://pubmed.ncbi.nlm.nih.gov/28993036/ Chow E, Iqbal A, Halim S, et al. Euglycemic Diabetic Ketoacidosis in the Era of SGLT-2 Inhibitors: A Review. World J Diabetes. 2023;14(9):1208–1222. PMID: 37797963. https://pubmed.ncbi.nlm.nih.gov/37797963/ (Open-access PMC: https://pmc.ncbi.nlm.nih.gov/articles/PMC10551972/) Babu A, et al. Safe and Simple ED Discharge Therapy for Type 2 Diabetes With Severe Hyperglycemia. Endocr Pract. 2009;15(6):696–704. PMID: 19625243. https://pubmed.ncbi.nlm.nih.gov/19625243/ Yan JW, Gushulak KM, Columbus MP, et al. Sentinel Visits in ED Patients With Diabetes as a Warning Sign for Hyperglycemic Emergencies. CJEM. 2018;20(2):230–237. PMID: 28738911. https://pubmed.ncbi.nlm.nih.gov/28738911/ Yan JW, et al. Risk Factors for Recurrent ED Visits for Hyperglycemia Among Patients With Diabetes. CJEM. 2017;19(5):351–359. PMCID: PMC5507935. https://pmc.ncbi.nlm.nih.gov/articles/PMC5507935/

ED Hyperglycemia (Without HHS or DKA): What to Lower, Who to Admit, and Who’s at Risk for Future DKA

More from Hanna Shanar

Understanding the Steeple Sign and Thumbprint Sign on X-Rays: Key Indicators in Diagnosing Respiratory Conditions

Protocol and recommendations for treating SCAPE (Sympathetic Crashing Acute Pulmonary Edema) in the Emergency Department

RIG, RIBS, and PIC Scores in Rib Fracture Management

Do NSAIDs Increase Bleeding Risk? Evidence Across Gastrointestinal, Intracranial, and Systemic Bleeding