Delta-delta in DKA with vomiting-related metabolic alkalosis

Electrolyte Disordersmedium

A 28-year-old woman with type 1 diabetes has 2 days of nausea, vomiting, and abdominal pain. She has missed some insulin doses and cannot tolerate oral intake. Heart rate is 118/min and BP 92/58 mm Hg; she has deep, rapid respirations, dry mucous membranes, and capillary refill of 4 seconds. Na 132, K 4.3, Cl 88, total CO₂ 14 mEq/L; BUN 32 mg/dL; creatinine 1.43 mg/dL; glucose 574 mg/dL. Arterial pH 7.35, PaCO₂ 28 mm Hg. Which acid-base disorder best explains these findings?

  1. A.Mixed high anion gap metabolic acidosis and non-elevated gap metabolic acidosis
  2. B.Mixed high anion gap metabolic acidosis and respiratory alkalosis
  3. C.Pure metabolic acidosis
  4. D.Mixed high anion gap metabolic acidosis and metabolic alkalosisCorrect

Explanation

The findings are best explained by a mixed high anion gap metabolic acidosis and metabolic alkalosis. The anion gap is 132 − (88 + 14) = 30 mEq/L, about 18 above a normal value of 12, consistent with diabetic ketoacidosis from missed insulin. If ketoacids alone were responsible, bicarbonate would have fallen by roughly the same amount, from 24 to about 6 mEq/L. Instead it fell only 10 mEq/L, to 14. The delta-delta ratio (change in anion gap divided by change in bicarbonate) is 18/10, or 1.8, above the usual range of 1 to 2 for ketoacidosis and well above 1, meaning bicarbonate is higher than expected: a concurrent metabolic alkalosis is present. Two days of vomiting, with loss of gastric hydrochloric acid and volume contraction, explains the alkalosis, and the low chloride of 88 mEq/L supports it. Respiratory compensation is appropriate. Winter's formula predicts a PaCO₂ of 1.5 × 14 + 8 = 29 ± 2 mm Hg, and the measured value of 28 falls in that range, so there is no separate respiratory disorder. The near-normal pH of 7.35 reflects the offsetting alkalosis rather than unusually mild ketoacidosis. Recognizing the alkalosis matters because the severity of ketoacidosis is underestimated if judged by pH or bicarbonate alone, and saline replacement will correct the chloride-responsive component. A pure metabolic acidosis does not account for a bicarbonate that is 8 mEq/L higher than the anion gap rise predicts. A coexisting non-elevated gap acidosis would make the bicarbonate fall more than the anion gap rise, giving a delta-delta ratio below 1. Hyperchloremic acidosis often appears during DKA treatment when ketones are lost in urine and replaced by chloride from saline, but the opposite pattern is seen here. A concurrent respiratory alkalosis would require a PaCO₂ below the range predicted by Winter's formula. The measured value of 28 mm Hg is appropriate compensation.

Why each option

A.
A coexisting non-gap acidosis would lower bicarbonate more than the anion gap rise, giving a delta-delta below 1, the opposite of this pattern.
B.
Winter's formula predicts a PaCO2 of 29 plus or minus 2, so the measured 28 mm Hg is appropriate compensation, not a separate respiratory alkalosis.
C.
A pure metabolic acidosis cannot explain a bicarbonate 8 mEq/L higher than predicted from an anion gap rise of 18.
D.
Correct. An anion gap of 30 with only a 10 mEq/L bicarbonate fall gives a delta-delta near 1.8, revealing vomiting-related metabolic alkalosis alongside ketoacidosis.

Reference: Kitabchi AE, Diabetes Care 2009 (Hyperglycemic crises in adult patients with diabetes); Berend K, de Vries AP, Gans RO, N Engl J Med 2014 (Physiological approach to assessment of acid-base disturbances)

This is one of 1,100 questions in the MedBoardSA nephrology bank, with timed test mode, tutor mode, and progress tracking by topic.

Practise the full bank

More Electrolyte Disorders questions

Browse other topics