Diabetic ketoacidosis: the blood tests that help identify an emergency

Diabetic ketoacidosis, or DKA, is a potentially life-threatening emergency caused by a severe lack of effective insulin. Diabetes together with vomiting, abdominal pain, deep rapid breathing, severe weakness, or drowsiness needs immediate medical assessment. For severe symptoms or altered consciousness, call emergency services: 911 in the US, or your local emergency number. Normal or only moderately raised blood sugar does not reliably rule out DKA. Do not wait for a home blood test or a fasting appointment if DKA is suspected.

What blood tests show during ketoacidosis

Insulin helps cells use glucose and restrains the production of ketones. When there is not enough effective insulin, the body breaks down more fat. The liver produces ketone bodies, particularly beta-hydroxybutyrate and acetoacetate. If they accumulate substantially, the blood can become too acidic. At the same time, high glucose often causes the kidneys to lose more water and electrolytes in the urine. DKA review

Testing therefore has to answer several questions together. Is there significant ketosis? Is the blood too acidic? How much fluid has been lost, and how have electrolytes shifted? Glucose alone cannot answer them. The 2024 international consensus report centers diagnosis on the combination of diabetes or hyperglycemia, ketosis, and metabolic acidosis. Consensus report on hyperglycemic crises

Blood beta-hydroxybutyrate reflects the acute ketone burden more directly than an ordinary urine ketone strip. Urine tests mainly detect acetoacetate and can therefore give a different picture early in DKA and during recovery. An emergency department study compared the diagnostic performance of the two approaches. Arora and colleagues, 2011

The pH measured in a blood gas test describes how acidic the blood is. Bicarbonate is an important buffer that helps absorb acids and is used up during DKA. A standard chemistry panel often reports total CO₂, most of which reflects bicarbonate. This is not the same as the carbon dioxide partial pressure measured in a blood gas test.

The anion gap is a calculated difference between measured positively and negatively charged particles. It can suggest the presence of additional acids, but it neither proves DKA by itself nor provides the best measure of recovery in every treatment setting. Creatinine and urea or blood urea nitrogen can add information about kidney function and the effects of fluid loss.

Understanding a DKA laboratory profile

Important measurements include glucose, ketones, pH or bicarbonate, sodium, and potassium. Common adult reference ranges are approximately 70–99 mg/dL for fasting glucose, 22–29 mmol/L for bicarbonate or total CO₂, 135–145 mmol/L for sodium, and 3.5–5.0 mmol/L for potassium. These are guides to laboratory values outside an acute crisis, not cutoffs you can use to rule out DKA yourself. The laboratory’s own ranges and the clinical situation matter.

In DKA, bicarbonate typically falls while ketones rise. Glucose is often very high, but it may be considerably lower with SGLT2 inhibitor use, reduced food intake, or pregnancy. This euglycemic or near-euglycemic form is an important reason not to judge symptoms from the glucose reading alone. 2024 consensus report

Sodium needs careful interpretation. High glucose pulls water out of cells into the bloodstream, which can dilute the measured sodium concentration. Meanwhile, water and salts are being lost in the urine. A low sodium result is therefore not a simple, direct measure of dehydration severity. The treatment team considers glucose and osmolality, the concentration of dissolved particles, among other factors.

Potassium can initially be normal or high in the blood even when the body’s total potassium stores are depleted. Insulin deficiency, acid-base changes, and fluid loss affect its distribution between cells and blood. Once insulin treatment starts, potassium can move rapidly back into cells and the blood concentration may fall dangerously. Close monitoring and, when needed, targeted replacement are therefore part of treatment. Do not adjust potassium on your own from a single result. Clinical review of diabetic emergencies

Typical symptoms include intense thirst, frequent urination, nausea, vomiting, abdominal pain, deep rapid breathing, and increasing tiredness. The breath sometimes smells fruity or like acetone; the absence of this smell is not reassuring. Children can deteriorate quickly, and DKA during pregnancy may occur at lower glucose concentrations.

Treatment follows more than the glucose number. A falling blood sugar does not by itself mean that ketosis and acidosis have resolved. Fluids, insulin, and electrolytes have to be adjusted together. An ordinary outpatient blood panel cannot replace the monitoring this requires.

What a laboratory profile can clarify

Glucose, total CO₂, sodium, and potassium provide important clues to a metabolic emergency. A panel containing only these four measurements cannot reliably confirm or exclude DKA. Ketone testing and assessment of acid-base status are part of the focused emergency investigation. The team also looks for a trigger, such as infection, interrupted insulin delivery, a pump problem, or previously unrecognized diabetes. Consensus report

Laboratory results are assessed alongside circulation, breathing, alertness, fluid intake, and urine output. Medicines, kidney disease, pregnancy, and sample handling can influence the findings. For example, a delayed or improperly handled sample may give a falsely low total CO₂ result. An isolated abnormal value is not automatically DKA, but compatible symptoms still make assessment urgent.

DKA is not the same as every form of ketosis. Fasting or a very low-carbohydrate diet can raise ketones without diabetic ketoacidosis. Conversely, presumed nutritional ketosis is not an adequate explanation for vomiting, changed breathing, or drowsiness. Alcohol-related ketoacidosis and other causes of acidosis also require medical care.

Frequently asked questions

What is a blood test for diabetic ketoacidosis?

It assesses glucose, ketones, acid-base status, and electrolytes, with other measurements added as needed. The aim is to identify a dangerous metabolic disturbance and guide treatment. A routine profile containing glucose, CO₂, sodium, and potassium alone is incomplete for this purpose.

When is this testing needed?

When DKA is suspected, for example in someone with diabetes and persistent vomiting, abdominal pain, deep rapid breathing, or a change in consciousness. DKA can also be the first presentation of previously undiagnosed diabetes. Testing then belongs in urgent care, not a scheduled general wellness check.

Can I test for it at home?

People with diabetes can check glucose and, where appropriate, blood ketones according to their agreed sick-day or emergency plan. This can help identify risk early. If DKA is suspected, home readings do not replace assessment and treatment in a medical facility. An arranged home blood draw or mail-in kit would be too slow and would not provide the necessary monitoring.

How often should testing be done?

There is no fixed “DKA screening schedule” for people without symptoms. During illness, self-monitoring follows the person’s diabetes plan. In hospital, relevant measurements are repeated closely during DKA because insulin and fluid treatment can change electrolyte concentrations quickly.

What can affect the results?

Insulin deficiency, infections, surgery, vomiting, and fluid loss are important influences. SGLT2 inhibitors can contribute to DKA without such a marked rise in glucose. Impaired kidney function changes potassium and bicarbonate. Deep, rapid breathing releases more carbon dioxide, lowering its partial pressure as a response to acidosis. That measurement must be distinguished from total CO₂ on a chemistry panel.

Do I need to fast before the blood draw?

Do not wait to complete a fast if DKA is suspected. Assessment needs to happen immediately. For a planned, non-emergency test, fasting may be useful for certain glucose questions, but is usually unnecessary for sodium, potassium, or bicarbonate alone. The treatment team should know about medicines and the timing of the last insulin dose.

Can lifestyle changes affect these values?

Food, fluids, exercise, and alcohol affect metabolism. But DKA cannot be treated at home by drinking more water, exercising, or changing diet. Exercise may be inappropriate when insulin is insufficient and ketones are present. Prevention includes an agreed sick-day plan and reliable access to insulin. Do not independently omit insulin simply because nausea has reduced food intake.

How should I interpret the result?

The key is the combination of ketones and acidosis in the context of diabetes or raised glucose. Low bicarbonate and sodium or potassium shifts help describe the situation but are not diagnostic by themselves. DKA can occur with normal blood potassium or only moderately elevated glucose. A compatible pattern requires immediate medical assessment.

Sources

• Kitabchi AE, Umpierrez GE, Miles JM, Fisher JN (2009). Hyperglycemic crises in adult patients with diabetes. Diabetes Care, 32(7), 1335-1343. https://doi.org/10.2337/dc09-9032 PubMed
• Umpierrez G, Davis GM, ElSayed NA, Fadini GP, Galindo RJ, Hirsch IB, Klonoff DC, McCoy RG, Misra S, Gabbay RA, Bannuru RR, Dhatariya KK (2024). Hyperglycemic crises in adults with diabetes: A consensus report. Diabetes Care, 47(8), 1257-1275. https://doi.org/10.2337/dci24-0032 PubMed
• Dhatariya KK, Glaser NS, Codner E, Umpierrez GE (2020). Diabetic ketoacidosis. Nature Reviews Disease Primers, 6(1), 40. https://doi.org/10.1038/s41572-020-0165-1 PubMed
• Umpierrez G, Korytkowski M (2016). Diabetic emergencies—Ketoacidosis, hyperglycaemic hyperosmolar state and hypoglycaemia. Nature Reviews Endocrinology, 12(4), 222-232. https://doi.org/10.1038/nrendo.2016.15 PubMed
• Arora S, Henderson SO, Long T, Menchine M (2011). Diagnostic accuracy of point-of-care testing for diabetic ketoacidosis at emergency-department triage: β-hydroxybutyrate versus the urine dipstick. Diabetes Care, 34(4), 852-854. https://doi.org/10.2337/dc10-1844 PubMed