Eric Robbins
Brown University Rhode Island Hospital
Providence, Rhode Island
Disclosure information not submitted.
Barry Shea, MD
Dr.
Brown University / Rhode Island Hospital, United States
Disclosure information not submitted.
Title: Diabetes Insipidus in a Critically Ill Post-Partum COVID Patient
Case Report Body:
Introduction: Diabetes insipidus (DI) is a syndrome with either inadequate secretion of, or renal unresponsiveness to, arginine vasopressin (AVP), resulting in hypotonic polyuria and polydypsia. Non-congenital forms of DI include central or nephrogenic DI, with central DI typically representing damage to the posterior pituitary. Finally, there is a transient form of DI in pregnancy, driven by placental enzymatic degradation of AVP. Pregnant or newly postpartum patients with DI may present with any of the three above-mentioned DI subtypes.
Description: A 31-year-old woman at 22 weeks gestation was admitted to the intensive care unit (ICU) with hypoxemia from SARS-COV-2. She was started on dexamethasone and remdesivir with increasing oxygen requirements, first requiring intubation and then cannulation for venovenous extracorporeal membrane oxygenation (VV-ECMO). On day six of VV-ECMO, she ejected her cervical mucus plug and had a spontaneous vaginal delivery.
Prior to and following delivery, she remained normotensive; estimated delivery-related blood loss was 500 cc. On postpartum day 5, she had 5L of urine output, and laboratory results showed a serum sodium of 155 mEq/L, urine specific gravity of 1.005 and osmolality of 228 mOsm/kg, and serum osmolality of 311 mOsm/kg. Given these results, a presumptive diagnosis of DI was made.
She was treated with intravenous desmopressin, with a post-desmopressin urine osmolality of 397 mOsm/kg, suggesting an appropriate response and thus ruling out nephrogenic DI. Transient DI of pregnancy was thought a less likely etiology, given the four days between delivery and development of DI. Approximately 24 hours prior to her diagnosis, she had severe chatter on her ECMO circuit, causing shutoff of the oxygenator and decrease in her oxygen saturation to 30%. This was quickly corrected, but it seems possible this hypoxic insult could have precipitated her DI.
Discussion: There are reports of central DI in critically ill patient with ARDS, with hypoxia-induced hypothalamic injury as the suggested etiology. Similarly, reports exist of central DI following discontinuation of vasopressin for shock or associated with HELLP or Sheehan syndromes—none of which apply to this patient. This remains the first instance we can find of postpartum central DI in COVID infection.