Step 1 topics → Endocrine → Diabetic Ketoacidosis (DKA)

Diabetic Ketoacidosis (DKA)

USMLE Step 1 trap: Chooses bicarb over K repletion when serum K is 3.1 in DKA, missing the K-before-insulin rule. Potassium must be repleted to ≥3.3 mEq/L before starting insulin, because insulin drives K⁺ into cells and can precipitate life-threatening hypokalemia.

DKA is one of the highest-yield acute complications on USMLE Step 1, and students consistently make the same dangerous error: starting insulin before ensuring potassium is at least 3.3 mEq/L, which can crash serum K to fatal levels. The core concept is insulin deficiency so profound (classic in T1DM) that glucagon runs unopposed, driving lipolysis and hepatic ketogenesis — resulting in anion gap metabolic acidosis from β-hydroxybutyrate and acetoacetate, Kussmaul respirations, and fruity breath from exhaled acetone. The exam tests whether you understand the treatment protocol's logic, not just its steps.

The exam tests DKA from multiple angles. Pure recall questions ask you to identify the lab pattern (elevated anion gap, low bicarb, elevated glucose, ketonemia). But the harder questions are clinical application: a patient in DKA has a potassium of 3.1 — what do you do first? Or: why do we not give bicarbonate routinely? These questions separate students who memorized a list from students who understand why the treatment protocol exists. USMLE Step 1 also loves the DKA-vs-HHS differential, which forces you to understand what residual insulin does and doesn't do in Type 2 diabetics.

The trickiest part of DKA is the potassium management. Total body potassium is depleted (lost in urine from osmotic diuresis), but serum K can look normal or even high on presentation because acidosis shifts K out of cells. Once you give insulin, K rushes back into cells and serum K can crash. Students who don't internalize this sequence pick the wrong answer on management questions — specifically, they reach for bicarbonate or insulin first instead of ensuring K is safe. The mechanism behind each treatment step is exactly what USMLE Step 1 rewards.

From real student decks
81%have cards covering this topic
54%have mature cards

Common misconceptions

Common mistake
Wrong: Bicarbonate should be given early in DKA to correct acidosis before or instead of potassium repletion.
Right: Potassium must be repleted to ≥3.3 mEq/L before starting insulin, because insulin drives K⁺ into cells and can precipitate life-threatening hypokalemia.
Bicarbonate does not fix the dangerous potassium problem in DKA — it actually makes it worse by driving K into cells. The critical rule is that serum potassium must be ≥3.3 mEq/L before you start insulin, because insulin activates the Na/K-ATPase pump and shifts potassium intracellularly, potentially dropping serum K to fatal levels. Think of it this way: the insulin is what actually corrects the acidosis, but you have to make the intracellular space 'safe' first by ensuring there's enough K to spare before you start pushing it in.
Common mistake
Wrong: Sodium bicarbonate should be routinely administered in DKA to correct the metabolic acidosis.
Right: Bicarbonate is not routinely given in DKA; the acidosis resolves with insulin and fluids, and bicarb can worsen hypokalemia and cause paradoxical CNS acidosis.
Bicarbonate is not a standard DKA treatment and giving it early can cause real harm. The acidosis in DKA will resolve on its own once insulin shuts down ketogenesis — you don't need to buffer it. More importantly, bicarbonate worsens hypokalemia (bicarb raises pH, which shifts K intracellularly) and can cause paradoxical CNS acidosis because CO₂ crosses the blood-brain barrier faster than bicarbonate. It's reserved for extreme, life-threatening acidosis (pH < 6.9) only.
Common mistake
Wrong: Ketone production in DKA is driven primarily by excess glucose rather than by uninhibited glucagon-stimulated lipolysis.
Right: Insulin deficiency removes inhibition of glucagon, which drives lipolysis and hepatic β-oxidation of free fatty acids into ketone bodies (β-hydroxybutyrate and acetoacetate).
Hyperglycemia itself doesn't make ketones — the liver doesn't convert glucose into ketone bodies. Ketogenesis is driven by the fatty acid supply arriving at the liver. When insulin is absent, glucagon is unopposed and activates hormone-sensitive lipase in adipose tissue, flooding the bloodstream with free fatty acids. The liver then undergoes β-oxidation of these FAs into acetyl-CoA, which gets shunted into ketone body synthesis (acetoacetate and β-hydroxybutyrate) rather than the TCA cycle. The hyperglycemia and ketoacidosis are parallel consequences of insulin deficiency, not one causing the other.
Common mistake
Wrong: HHS can present with significant ketoacidosis just like DKA because both involve severe hyperglycemia.
Right: HHS lacks significant ketoacidosis because residual insulin in T2DM is sufficient to suppress lipolysis and hepatic ketogenesis, even though it cannot prevent hyperglycemia.
The key to the DKA-vs-HHS distinction is understanding what a small amount of residual insulin can and cannot do. In HHS (Type 2 DM), patients retain just enough insulin to suppress lipolysis in adipose tissue — which prevents the flood of free fatty acids to the liver and therefore prevents ketogenesis. But that same residual insulin is insufficient to prevent hyperglycemia, so glucose climbs to extreme levels (often >600 mg/dL). In DKA (Type 1 DM), there is essentially zero insulin, so both processes go unchecked: hyperglycemia and ketoacidosis occur together. Severe hyperglycemia does not equal ketoacidosis — you need the lipolysis piece.
Free Deck audit

See if your Anki deck covers this topic.

Upload your deck →
Guided session

Stuck on this? An AI tutor that probes your understanding.

Start a session →

What the exam tests

  1. Understand the mechanistic chain from insulin deficiency → unopposed glucagon → lipolysis → hepatic β-oxidation → ketone body production → anion gap acidosis and osmotic dehydration
  2. Recognize the classic DKA clinical presentation and expected lab abnormalities: elevated anion gap, low bicarbonate, hyperglycemia, ketonemia/ketonuria, and the expected potassium shift pattern
  3. Apply the correct ordered treatment protocol: isotonic fluids first, replete potassium to ≥3.3 mEq/L before insulin, then start insulin infusion, add dextrose when glucose falls to ~200–250 mg/dL, then transition to subcutaneous insulin
  4. Distinguish DKA from HHS using lab values and clinical features — especially why HHS produces profound hyperglycemia and hyperosmolarity without significant ketoacidosis

Can you avoid these mistakes?

A 19-year-old with T1DM presents with polyuria, vomiting, and Kussmaul respirations. Labs: glucose 480, pH 7.18, bicarb 10, anion gap 26, serum K 3.0 mEq/L. You want to start insulin. What must you do first and why?
Explain mechanistically why a patient with DKA can have a normal or elevated serum potassium on admission but is actually total-body potassium depleted — and how this changes when you treat the DKA.
A 68-year-old with T2DM presents with glucose of 920 mg/dL, serum osmolality of 370 mOsm/kg, and altered mental status. His pH is 7.38 and his urine ketones are trace. Why doesn't this patient have significant ketoacidosis despite his extreme hyperglycemia?
A classmate says that giving bicarbonate early in DKA is important to quickly normalize the pH and protect the heart from acidosis. Give two mechanistic reasons why this is not standard practice and can actually worsen the clinical picture.

Related topics

Type 1 Diabetes Mellitus Hypothalamic-Pituitary-Adrenal (HPA) Axis Hypothalamic-Pituitary-Thyroid (HPT) Axis Hypothalamic-Pituitary-Gonadal (HPG) Axis Hormone Signaling (Peptide vs Steroid)

See how your Anki deck covers this topic.

Upload your deck for a free audit →