**Background:** Diabetic ketoacidosis (DKA) during pregnancy is an obstetric emergency associated with increased maternofetal morbidity and mortality. Pregnancy is a ketosis-prone state due to physiologic insulin resistance (driven by human placental lactogen, estrogen, growth hormone, cortisol, and TNF-alpha), accelerated starvation with lipolysis after short fasting periods, and reduced buffering capacity from baseline respiratory alkalosis. SARS-CoV-2 infection has been recognized as a potential trigger for DKA in patients with and without prior impaired glucose tolerance. The combination of pregnancy and COVID-19 may place patients at particularly high risk for DKA, yet few cases have been reported.
**Methods:** This is a single case report from a single institution. The patient was a 32-year-old woman (gravida 1, para 0) at 27 weeks gestation presenting with a 3-day history of worsening shortness of breath, congestion, loss of taste and smell, polyuria, and polydipsia. She had no significant past medical history, no prior impaired glucose tolerance, and was taking only a prenatal vitamin. Family history was notable for paternal type 1 diabetes. Two days prior to presentation, she had an abnormal glucose load test of 373 mg/dL (institutional cut-off >130 mg/dL). Biochemical evaluation on admission revealed: bicarbonate 10 mmol/L, anion gap 23 mmol/L, pH 7.2, glucose 271 mg/dL (15.05 mmol/L), beta-hydroxybutyrate 5.8 mmol/L, and hemoglobin A1c 7.4% (57 mmol/mol). SARS-CoV-2 PCR was positive. She was admitted for management of moderate DKA.
**Key Results:** Treatment included intravenous hydration, electrolyte replacement, and insulin infusion. The patient remained hemodynamically stable, did not require supplemental oxygen, and did not receive steroids or remdesivir for COVID-19. Metabolic derangements resolved by day 5 (bicarbonate 21 mmol/L, anion gap 10 mmol/L, pH 7.43, glucose 97 mg/dL, beta-hydroxybutyrate 0.1 mmol/L). She was transitioned to subcutaneous insulin at approximately 0.7 units/kg total daily dose. Glycemic control was at goal for the remainder of pregnancy on multiple daily injections. She had an uncomplicated spontaneous vaginal delivery at 38 weeks; newborn weight was 3.77 kg. Postpartum phenotypic evaluation confirmed autoimmune diabetes: positive GAD-65 antibody (0.05 nmol/L; reference <0.02), positive Zinc T8 antibody (85.6 U/mL; reference <15 U/mL), with residual beta-cell function (C-peptide 1.6 ng/mL with concurrent glucose 81 mg/dL). Six months postpartum, glycemic control remained at goal on MDI (average glucose 121 mg/dL, GMI 6.2%, coefficient of variation 23.4%, 98.6% time in range).
**Clinical Implications:** This case demonstrates that SARS-CoV-2 infection may trigger autoimmune diabetes even during pregnancy, a state normally characterized by immunotolerance. The patient's blood glucose at presentation (271 mg/dL) was just above the euglycemic DKA threshold of 250 mg/dL, and the time to resolution of metabolic derangements was longer than typically observed, suggesting a unique pathophysiology when DKA occurs in the setting of both acute SARS-CoV-2 infection and pregnancy. Clinicians should maintain a high index of suspicion for DKA in pregnant patients with COVID-19, even with modest hyperglycemia. The case also highlights the importance of postpartum biochemical evaluation to determine diabetes phenotype, as pregnancy-induced immunotolerance may blunt autoimmune disease presentation during gestation, with diagnosis only confirmed after delivery.