Diabetic Ketoacidosis With a Second Patient in the Room
The DKA protocol itself is well validated. What it wasn't built around is a second patient whose distress can show up on a monitor before the mother's own labs tell the whole story.
Simone A., a 29-year-old woman at 26 weeks of her first pregnancy, runs a small alterations business out of her home and had planned to keep sewing right up until her due date. She has had type 1 diabetes for sixteen years, managed on an insulin pump, and presented to the emergency department after two days of nausea and vomiting she attributed to a stomach bug, unaware that her pump's infusion set had dislodged sometime the previous night. She arrived tachypneic, visibly dehydrated, with a fingerstick glucose of 412 mg/dL — a number that reads differently in pregnancy than out of it, since pregnancy’s accelerated ketogenesis routinely produces ketoacidosis at glucose values that would look unremarkable in a non-pregnant patient, so her 412 marks a process further along than the same figure otherwise would — and her venous blood gas confirmed a significant anion-gap metabolic acidosis with large serum ketones — diabetic ketoacidosis, now in a patient carrying a fetus whose own tolerance for prolonged maternal acidosis and dehydration is genuinely limited.
DKA management itself follows a well-validated general protocol — fluid resuscitation, insulin infusion, potassium repletion, frequent reassessment — but pregnancy changes some of the physiology that protocol was built around. Increased plasma volume and altered renal handling shift how quickly large-volume normal saline can produce a competing hyperchloremic acidosis, an effect that can muddy the anion-gap trend the team is using to judge whether the ketoacidosis itself is actually resolving. And unlike a non-pregnant patient, there's a second patient in the room whose distress can show up on a monitor well before the mother's own labs fully normalize.
The same protocol, and a second patient the labs don't show
I'd run the standard DKA protocol as validated — normal saline resuscitation, standard insulin infusion rate, hourly reassessment. Maternal stabilization by the best-evidenced method is still the primary determinant of fetal outcome, and I'm wary of improvising pregnancy-specific changes to a protocol this well studied without strong outcome data specifically supporting the change.
I'd modify the fluid choice specifically. Large-volume normal saline resuscitation can produce a hyperchloremic acidosis on top of her ketoacidosis, which would muddy our read of whether the anion gap is actually closing or just being replaced by a different acid-base problem — a balanced crystalloid, of the kind SMART tested against saline in critically ill adults, avoids that confound and lets us trust the trend we're watching.
This isn't an argument against the standard protocol broadly, just this one fluid-choice detail, given how directly it affects our ability to read her actual resolution accurately — and I'd reassess her ketones and glucose on a tighter interval than the usual adult default, given how much faster pregnancy's metabolism can shift these numbers.
Whatever the fluid and insulin plan, I want continuous fetal monitoring started right now, alongside her resuscitation, not after her labs look better. Fetal heart rate response during maternal correction is real information, not just a box to check — if her fetal tracing improves as the acidosis corrects, that tells us the resuscitation is actually working at the level that matters most for the pregnancy.
I'm not asking either of you to change your protocol on my account. I'm asking to be in the room from the start, because if the tracing shows distress that doesn't track with how her maternal numbers are trending, that's a signal worth acting on immediately, not something that waits for a formal handoff once she's "stable."
Balanced crystalloid resuscitation, standard weight-based insulin infusion, potassium repletion per protocol, and continuous fetal monitoring started at presentation with maternal- fetal medicine present from the outset. Her anion gap closed over the following eight hours with a fetal tracing that stayed reassuring throughout.
Not agreed: whether the tighter ketone/glucose reassessment interval the endocrinologist proposed should become the standing protocol for DKA in pregnancy generally, or was specific to this case's severity. The emergency medicine physician wanted more cases observed before revising the department's standard pathway; the endocrinologist felt the physiologic rationale was sufficient to adopt more broadly now. Raised as a protocol question beyond this one patient, not resolved at the bedside.