When should urine ketone testing be performed in diabetic and other at‑risk patients, and how should the results guide clinical management?

Urine Ketone Monitoring in Clinical Practice

Direct Recommendation

Blood β-hydroxybutyrate measurement should replace urine ketone testing for all clinical decision-making because urine dipsticks detect only acetoacetate, miss the predominant ketone body (β-hydroxybutyrate) in diabetic ketoacidosis, and substantially underestimate total ketone burden. 1

---

When to Test for Ketones

High-Risk Patients Requiring Immediate Testing

Perform ketone testing (preferably blood β-hydroxybutyrate) in the following situations:

• Type 1 diabetes patients when blood glucose persistently exceeds 300 mg/dL (16.7 mmol/L) 1
• During acute illness or physiological stress in any diabetic patient 1
• All pregnant patients with diabetes presenting with any metabolic concerns 1
• Symptoms suggestive of ketoacidosis: nausea, vomiting, abdominal discomfort, Kussmaul respirations, altered mental status 1
• Patients on SGLT2 inhibitors with any unexplained symptoms, as these medications increase DKA risk and can cause euglycemic DKA (glucose <250 mg/dL despite severe ketoacidosis) 1
• History of prior DKA episodes 1

---

Interpreting Blood Ketone Results

Action Thresholds (β-hydroxybutyrate)

The American College of Clinical Endocrinologists provides clear thresholds for clinical action:

• <0.5 mmol/L: No intervention required 1
• 0.5–1.5 mmol/L: Initiate sick-day management rules (increase fluids, monitor glucose every 3–4 hours, consider additional rapid-acting insulin) 1, 2
• ≥1.5 mmol/L: Seek immediate medical attention 1
• ≥3.0 mmol/L: Diagnostic of diabetic ketoacidosis when combined with appropriate clinical and laboratory findings 3, 4

Complete DKA Diagnostic Criteria

DKA requires all of the following 1:

• Plasma glucose >250 mg/dL
• Arterial pH <7.30
• Serum bicarbonate <15 mEq/L
• Positive ketones (blood or urine)
• Anion gap >10 mEq/L

---

Why Urine Ketone Testing Is Unreliable

Critical Limitations

• Urine dipsticks using nitroprusside detect only acetoacetate, not β-hydroxybutyrate, which is the predominant ketone body in DKA 1, 5
• During DKA treatment, β-hydroxybutyrate falls while acetoacetate rises, making urine tests paradoxically more positive as the patient improves 5
• Urine ketones are unreliable for diagnosing or monitoring treatment of ketoacidosis 1, 2
• Only 47.7% of hyperglycemic patients in emergency departments can provide urine samples when needed 3

False Results

False-positive urine ketones occur with:

• Sulfhydryl-containing medications (captopril, other ACE inhibitors) 1, 2
• Highly colored urine 2

False-negative urine ketones occur with:

• Test strips exposed to air for prolonged periods 1, 2
• Highly acidic urine (e.g., after large ascorbic acid intake) 1, 2

---

Benign (Physiologic) Ketonuria

Common Non-Pathological Causes

Recognize these situations where positive ketones do not indicate DKA:

• Up to 30% of first-morning urine specimens from healthy individuals show positive ketones, especially after overnight fasting 1, 2
• Up to 30% of pregnant women (with or without diabetes) have physiologic ketonuria in first-morning specimens 1, 2
• Starvation ketosis from reduced caloric intake or very low-carbohydrate diets (<50 g/day) produces β-hydroxybutyrate 0.3–4 mmol/L with normal glucose and bicarbonate ≥18 mEq/L 1
• Post-hypoglycemic ketonuria from counter-regulatory hormone surges stimulating transient lipolysis 1

---

Clinical Algorithm for Ketone Detection

Step 1: Risk Stratification

Immediate evaluation required if patient has:

• Known type 1 diabetes 1
• Current SGLT2 inhibitor use 1
• History of prior DKA 1
• Unexplained hyperglycemia with symptoms 1

Lower risk (likely benign) if:

• No diabetes history 1
• Normal blood glucose 1
• Recent decreased oral intake 1
• No abdominal pain or altered mental status 1

Step 2: Choose Testing Method

• First-line: Blood β-hydroxybutyrate measurement (point-of-care or laboratory) 1
• Avoid: Urine dipstick testing for clinical decision-making 1

Step 3: Immediate Actions Based on Results

If blood glucose >250 mg/dL with any ketones present:

• Obtain immediate blood β-hydroxybutyrate, electrolytes, arterial blood gas 6
• Check for infection (urinary tract infections precipitate ~50% of DKA cases) 1
• Assess complete DKA criteria 1

If β-hydroxybutyrate 0.5–1.5 mmol/L:

• Increase oral fluid intake 2
• Monitor blood glucose and ketones every 3–4 hours 2
• Consider additional rapid-acting insulin if levels increase 2
• If glucose is normal but ketones elevated, consume carbohydrates with insulin to suppress ketone production 2

---

Special Populations

SGLT2 Inhibitor Users

• These patients require immediate evaluation for any unexplained symptoms because SGLT2 inhibitors cause euglycemic DKA where glucose may be <250 mg/dL despite severe ketoacidosis 1
• Standard glucose thresholds for DKA do not apply 1

Pregnant Patients

• Physiologic ketonuria is common (30% of first-morning specimens) 1
• However, all pregnant diabetic patients require ketone testing during illness or metabolic stress 1
• Use blood β-hydroxybutyrate to distinguish physiologic from pathologic ketosis 1

Alcoholic Ketoacidosis

• Presents with positive urine ketones but hyperglycemia is usually absent 1
• Distinguish from DKA by checking blood glucose 1

---

Emerging Technology

Continuous ketone monitoring (CKM) via subcutaneous sensors measuring interstitial fluid β-hydroxybutyrate every 5 minutes is under development and may benefit high-risk populations including those with recurrent DKA, pregnancy with type 1 diabetes, SGLT2 inhibitor users, and socially isolated individuals 7, 5