I recently had a patient on an insulin drip (7 units/hr) whose K level came back at 2.7. Initially, the order I was given for replacement was 40 meq Klorvess pNGT (along with 40mg of furosemide for diuresis). The pt. dumped only about 400 cc from the furosemide, but I was concerned that 40 meq was not enough replacement for a K of 2.7, and the pt. has a hx of ectopy (hx MI, CHF, EF 25%). I was given an order for an additional 20 meq pNGT, and the PA said that they weren't overly concerned about the K of 2.7 because insulin gtt can cause a false low K. This same pt. had a K of 8.5 come back the day before, which was probably inaccurate because the pt's blood clots if not tested immediately (d/t cold agglutinin disease). He had gotten Kayexalate.

I remember learning in school that one tx of hyperkalemia is insulin w/dextrose, to push K into the cells, so what they're saying does make sense, but what I'm wondering is whether there's some sort of formula to estimate the amount that the serum K level will be off d/t the insulin. Normally on my unit with cardiac pts, we like a min K of 4.0

Thanks,

Hillary

(Surgical Intermediate Care Unit)

It's true that intravenous insulin causes a potassium shift into the cells (which is why we use it to bring down a high K+).

i'm not sure why that would get labelled as a 'false' low, though - the potassium level's still low, because that intracellular K+ isn't available.

Whenever we have patients in DKA they have a hydration line, a glucose line (once the BSL drops below 15 mmol/L), an actrapid (insulin) infusion and a potassium infusion, which is titrated to their serum K+: 10mmol/hr if it's over 5.5, 20mmol/hr if it's under 5. They have four-hourly electrolyte monitoring to keep an eye on the K+ and everything else, and 6/24 gases if they came in acidotic.

Sorry, I know I've burbled on - put it down to night duty! Hope this helps :)