Glucose Abnormalities — Hyperglycemia vs Hypoglycemia

Diabetic ketoacidosis (DKA) results from absolute or relative insulin deficiency combined with excess counter-regulatory hormones (glucagon, cortisol, catecholamines, growth hormone). The absence of insulin prevents glucose uptake by cells, promotes lipolysis, and drives ketogenesis — free fatty acids are converted to ketone bodies (acetoacetate, beta-hydroxybutyrate) in the liver. The result is hyperglycemia, ketonemia, and metabolic acidosis with an elevated anion gap.

Classic DKA diagnostic criteria: glucose above 250 mg/dL, pH below 7.3 (or bicarbonate below 18 mEq/L), and positive serum or urine ketones. DKA occurs predominantly in type 1 diabetes but can occur in type 2 diabetes under conditions of severe physiologic stress (infection, MI, surgery). Common precipitants: infection (most common), insulin omission, new-onset type 1 diabetes, and acute illness.

Euglycemic DKA is a critical exception. SGLT2 inhibitors (empagliflozin, dapagliflozin, canagliflozin) promote renal glucose excretion, which keeps blood glucose near normal even as ketoacidosis develops. A patient on an SGLT2 inhibitor with nausea, vomiting, abdominal pain, and anion gap acidosis should have serum ketones checked even if glucose is below 250 mg/dL — or even below 200 mg/dL. Euglycemic DKA is frequently missed because the glucose does not trigger the clinical suspicion for DKA.

The anion gap is the key to recognizing DKA when glucose is not markedly elevated. An elevated anion gap metabolic acidosis in a patient on an SGLT2 inhibitor should prompt immediate ketone measurement. The SGLT2 inhibitor should be held during the acute illness and not restarted until the patient is eating, drinking, and metabolically stable.

Hyperosmolar hyperglycemic state (HHS) is a hyperglycemic emergency that occurs predominantly in type 2 diabetes. It is characterized by extreme hyperglycemia (glucose typically ≥600 mg/dL, often above 1000 mg/dL), severe hyperosmolality (effective osmolality >300 mOsm/kg or total osmolality >320 mOsm/kg, calculated as 2×Na + glucose/18), profound dehydration, and minimal or absent ketosis.

The pathophysiology differs from DKA: in HHS, there is enough residual insulin to suppress lipolysis and ketogenesis (preventing ketoacidosis) but not enough to prevent hyperglycemia. The extreme hyperglycemia causes an osmotic diuresis, leading to massive fluid losses — often 8–10 liters or more. The resulting dehydration and hyperosmolality cause neurological symptoms (confusion, lethargy, seizures, coma) that are the hallmark of HHS.

HHS typically develops over days to weeks, often triggered by an acute illness (infection, MI, stroke) in an elderly patient with type 2 diabetes and limited access to water. The mortality of HHS is higher than DKA — approximately 10–20% — largely because it occurs in older patients with more comorbidities and because the diagnosis is often delayed.

Treatment of HHS focuses on aggressive fluid resuscitation (isotonic saline initially, then hypotonic fluids as osmolality corrects), gradual glucose reduction (target 50–75 mg/dL per hour to avoid cerebral edema), and treatment of the precipitating cause. Insulin is used but is less central than in DKA — fluid resuscitation alone will lower glucose significantly in HHS. Potassium replacement is essential as insulin and fluid shift potassium intracellularly.

Hypoglycemia is defined by the Whipple triad: (1) symptoms consistent with hypoglycemia, (2) a low plasma glucose concentration (typically below 70 mg/dL, with clinically significant hypoglycemia below 54 mg/dL), and (3) relief of symptoms with glucose administration. All three components should be present before attributing symptoms to hypoglycemia.

Symptoms of hypoglycemia fall into two categories: adrenergic (autonomic) symptoms — tremor, palpitations, diaphoresis, anxiety, pallor — which occur at glucose levels of approximately 60–70 mg/dL and serve as warning signs; and neuroglycopenic symptoms — confusion, difficulty concentrating, slurred speech, behavioral changes, seizure, loss of consciousness — which occur at lower glucose levels (below 50 mg/dL) and indicate impaired brain function.

In patients with diabetes, the most common causes of hypoglycemia are: insulin (any type), sulfonylureas (glipizide, glyburide, glimepiride — they stimulate insulin secretion regardless of glucose level), and meglitinides (repaglinide, nateglinide). Metformin, GLP-1 agonists, DPP-4 inhibitors, SGLT2 inhibitors, and thiazolidinediones do not cause hypoglycemia when used as monotherapy — but they increase the risk when combined with insulin or sulfonylureas.

In patients without diabetes, hypoglycemia is less common and requires a systematic evaluation. Causes include: insulinoma (fasting hypoglycemia with inappropriately elevated insulin and C-peptide), factitious hypoglycemia (exogenous insulin — elevated insulin, suppressed C-peptide), reactive hypoglycemia (postprandial, often in patients after gastric bypass), and critical illness (liver failure, sepsis, adrenal insufficiency).

Hypoglycemia unawareness is the loss of the adrenergic warning symptoms that normally alert patients to a falling glucose. Patients with hypoglycemia unawareness do not experience tremor, palpitations, or diaphoresis at glucose levels of 60–70 mg/dL — they may not recognize hypoglycemia until neuroglycopenic symptoms (confusion, seizure, loss of consciousness) occur at much lower glucose levels.

Hypoglycemia unawareness develops through a process called hypoglycemia-associated autonomic failure (HAAF): recurrent hypoglycemic episodes blunt the counter-regulatory hormone response (glucagon, epinephrine) and reduce the glycemic threshold at which adrenergic symptoms occur. The result is a vicious cycle — hypoglycemia causes unawareness, which leads to more severe hypoglycemia, which further impairs the counter-regulatory response.

Hypoglycemia unawareness is a safety-critical condition. Patients with unawareness are at high risk for severe hypoglycemia, loss of consciousness, and hypoglycemia-related accidents (falls, motor vehicle accidents). Management requires: relaxation of glycemic targets (accepting higher A1c to reduce hypoglycemia frequency), structured hypoglycemia avoidance (identifying and eliminating triggers), and CGM with low-glucose alarms.

Hypoglycemia unawareness can be partially reversed by 2–3 weeks of strict hypoglycemia avoidance — the counter-regulatory response recovers when hypoglycemic episodes are eliminated. This requires accepting temporarily higher glucose levels and A1c. CGM is essential for detecting nocturnal hypoglycemia and providing early warning before symptoms develop.

Stress hyperglycemia is elevated blood glucose during acute illness in patients without known diabetes. It results from counter-regulatory hormone release (cortisol, glucagon, catecholamines, growth hormone) that promotes gluconeogenesis and glycogenolysis while impairing insulin-mediated glucose uptake. It is common in hospitalized patients — occurring in approximately 30–40% of non-diabetic patients admitted to the ICU.

Stress hyperglycemia is associated with worse outcomes in hospitalized patients: increased infection risk, impaired wound healing, longer ICU stays, and higher mortality. Whether treating stress hyperglycemia improves outcomes is less clear — the NICE-SUGAR trial demonstrated that intensive insulin therapy (target glucose 81–108 mg/dL) in ICU patients increased mortality compared to conventional therapy (target below 180 mg/dL), largely due to hypoglycemia.

A1c is the key test for distinguishing stress hyperglycemia from previously undiagnosed diabetes in a hyperglycemic hospitalized patient. A normal A1c (below 5.7%) in a hyperglycemic patient confirms stress hyperglycemia — the patient did not have elevated glucose before the acute illness. An elevated A1c (6.5% or above) in a hyperglycemic patient suggests previously undiagnosed diabetes that was unmasked by the acute illness.

Patients with stress hyperglycemia and a normal A1c should have fasting glucose checked after discharge to confirm resolution. Approximately 20–30% of patients with stress hyperglycemia will be found to have prediabetes or diabetes on follow-up testing — the acute illness may have unmasked underlying glucose dysregulation that was not yet symptomatic.

DKA management has three simultaneous priorities: volume resuscitation, insulin to suppress ketogenesis, and electrolyte repletion — particularly potassium. These must be managed in parallel, not sequentially.

Volume resuscitation: isotonic saline (0.9% NaCl) is the initial fluid at 1–1.5 L/hour for the first 1–2 hours, then transitioned to 0.45% NaCl or lactated Ringer's based on corrected sodium. Fluid deficit in DKA is typically 3–6 liters. Aggressive early resuscitation is essential — but the rate is adjusted based on clinical response and comorbidities.

Insulin: do not start insulin until potassium is above 3.5 mEq/L. Hypokalemia worsens with insulin administration because insulin drives potassium into cells. The standard protocol is a continuous IV insulin infusion at 0.1 units/kg/hour. When glucose falls below 200–250 mg/dL, add dextrose to the IV fluids to allow continued insulin infusion until the anion gap closes and ketones clear.

Potassium: virtually all DKA patients are total-body potassium depleted despite a normal or elevated serum potassium on presentation (acidosis shifts potassium extracellularly). As insulin is given and acidosis corrects, potassium will fall. Aggressive potassium repletion is required throughout the infusion. Target potassium 4–5 mEq/L.

Resolution criteria: anion gap closure (not just glucose normalization), bicarbonate above 18, and pH above 7.3. Glucose normalizes before the anion gap closes — this is the most common reason for premature insulin discontinuation. Transition to subcutaneous insulin only after oral intake is established and the anion gap has closed.

Euglycemic DKA (SGLT2 inhibitor-associated): glucose may be near-normal despite active ketoacidosis. The diagnosis requires checking a beta-hydroxybutyrate and anion gap in any patient on an SGLT2 inhibitor presenting with nausea, vomiting, or malaise — even with a glucose below 250 mg/dL.

HHS management: the primary intervention is aggressive fluid resuscitation — deficits are typically 8–10 liters. Isotonic saline is used initially, then transitioned to hypotonic fluids once hemodynamics are stable. Insulin infusion is used but at lower rates than DKA, and glucose should be lowered slowly (no faster than 50–75 mg/dL/hour) to avoid cerebral edema from rapid osmolality shifts. Electrolyte monitoring and repletion are required throughout.

Hypoglycemia treatment: the 15-15 rule applies to conscious patients who can swallow — 15 grams of fast-acting carbohydrate (glucose tablets, 4 oz juice, regular soda), recheck in 15 minutes, repeat if still below 70 mg/dL. For patients who cannot swallow or are unconscious: IV dextrose (D50W, 25 mL IV) is first-line in the hospital setting. Glucagon (1 mg IM or SC) is the out-of-hospital option when IV access is unavailable.

After treating hypoglycemia, identify and address the cause. Insulin and sulfonylureas are the agents most commonly responsible. Sulfonylurea-induced hypoglycemia can be prolonged — patients may require observation and a dextrose infusion for 12–24 hours because the drug continues to stimulate insulin secretion after the initial glucose correction.

Escalation: DKA with pH below 7.0, altered mental status, hemodynamic instability, or inability to tolerate oral intake requires ICU-level care. HHS with severe hyperosmolality (above 380 mOsm/kg), altered mental status, or hemodynamic instability similarly requires intensive monitoring. Recurrent DKA or hypoglycemia warrants endocrinology consultation and a structured diabetes management review.