Where's the infection?

I work in the ER and had a diabetic pt today. Per the nursing home, pt hasn't been able to maintain her blood sugar. She's had continual bouts of hypoglycemia and has been given 3 vials of glucagon within a 24h period. They called the ambulance when she became unresponsive. Medics checked her blood sugar upon arrival and it was in the 90s but she was still unresponsive. Pt awoke en route to the hospital, rechecked sugar; it was 83. Upon arrival pt was aox3. Labs were done, urine was collected, pt had a CXR, and abdominal/pelvic CT. The CXR and CT were unremarkable. Pts WBC was 15.7, electrolytes normal and all other labs were WNLs, urine was negative.

What's the source of infection?
This happened to a patient of mine. Turned out to be related to an older cardiac med but I can't remember the name of it...while it may be something else consider looking into it!

What was her skin like? Did she recently have an invasive procedure? Pain anywhere? What was her PMH?

I believe the med was sotalol (betapace)

Sometimes the body's stress response can elevate the wbc

^ That. I remember when I worked cardiac, it was completely expected that post MI patients would have elevated WBC counts.Some other less obvious sources of infection can be osteomyelitis (not uncommon in diabetics) and endocarditis (which can exist subclinically for a really long time before the person becomes symptomatic).

Look for WBC formula: if there is "left move" ( over 85% neutrophiles, maybe some bands), then it is probably infection. Possibilities are multiple: osteomyelitis, dental stuff, chronic synus infection, skin/perineal issues among others. One negative UA does not exclude things like chronic pyelonephritis.If WBC elevated but mostly because lymphocytes, it is stress related leucocytosis, combined effect of glucagon and endogenic corticosteroids.

Wouldn't an infection tend to elevate the blood sugar?

Infection process can cause elevation of blood glucose (by the same cortisol action, mostly) in the absence of other influences, such as wrong admixture of insulins or not changing Lantus dose if patient is not eating as much as usual.I would wonder who gives this patient his insulin, if the device is working properly, who and how checks blood sugar. I saw way too many "hypoglycemias" caused by aides not letting ethanol dry out before pocking and then using the very first drop of blood.

Quote from KatieMIInfection process can cause elevation of blood glucose (by the same cortisol action, mostly) in the absence of other influences, such as wrong admixture of insulins or not changing Lantus dose if patient is not eating as much as usual.I would wonder who gives this patient his insulin, if the device is working properly, who and how checks blood sugar. I saw way too many "hypoglycemias" caused by aides not letting ethanol dry out before pocking and then using the very first drop of blood.

A1C gives "overall" blood glucose level for the last 120 days. It is not sensitive to short term changes, either to the low or high side, as hemoglobin glycosylation is non-enzyme dependent and happens with the set rate. So if patient had CBG> 250 most of the time over the last 3 months but had several REAL episodes of CBG < say, 70 that lasted several hours each, his A1C level will be still high. If there was a FALSE hypoglycemia due to wrong technique, his real blood sugar will be still higher, and A1C high as well. There are tests that allow to detect level of each of insulins patient is receiving to diagnose intentional or non-intentional overdose but they are rarely done and only in handful of places.In my place, we see cases as OP described all the time and wast majoity, if not all of them, based on wrong use of sliding scale and underuse of basal insulin like Lantus. Patient is not feeling well/eating well, Lantus dose decreased, CBG goes up, short- or ultrashort insulin administered as sliding scale causing sharp fall of glucose in 10 min., this causes increase of secretion of cortisone and glucagon as our body has no idea why the life-sustaining glucose is disappearing so quickly, in 3-4 hours the short insulin action is over and CBG checked again, found to be higher still, the higher dose of short-acting insulin administered as sliding scale. The show can go on till patient misses a meal and/or sliding scale dose happens to be high enough so clinical hypoglycemia happens. Subsequent treatment with glucagone either stops the chain if someone takes a look on the whole picture and gets it right, or rocks the boat even harder, the glucagon-induced hyperglycemia is "treated" again by sliding scale, and eventually patient goes to hospital.

I would check out the skin--what did her feet look like/in between toes, that type of thing. Also dental stuff--even if she has dentures, what does her mouth underneath look like?I think the glucagon (3 of them) can cause an increase in WBC's as well.

If it were my will, I would expressively prohibit sliding scale short-term insulins. Either do it as it supposed to be done (CBG checked when patient's plate is there and he desided what he will and won't eat, glucose-elevating effect calculated and appropriate dose of ultra-short term insulin given while already eating, with periodic measurements right before meals and at night and correction of basal insulin dose done accordingly) or just put everyone on pumps, they are not so expensive nowadays. Just don't treat anyone's diabetes as if clinical pharmacology does't exist.