New onset DM 2

[bx24d]

. Urgent care, IM, FP and ED specialists are seeing an increasing number of patients with/without symptoms and hyperglycemia. So, I asked a well renowned Endocrinologist to provide a lecture to our Group on the initial office management of new onset Diabetes. I wanted to improve my understanding of when to consider starting insulin from the office as opposed to metformin/sulfonylureas. He stated that, if the patient does not have an anion gap, oral agents should be started first and reassessed over the first two weeks.

I'm a very practical guy so the problem is, I have to decide to discharge the pateint home or send to the ED before I have the results of the comprehensive metabolic panel. When an office fingerstick is 300-400 (or ">400"), and the office glyco is elevated in the setting of polyuria, polydypsia, in an otherwise stable patient, what do I do with the patient for the 24 hours that it takes to get full labs back and decide if there is a gap. Do ketones on the urine dip help me decide (starting insulin vs. oral). Is "glucose toxicity" a real clinical issue that renders oral agents relatively ineffective in these patients? Please note: I'm not talking about long term management of the sugars here; I'm talking about what to do with the patient in the first 24-48 hours of diagnosis (from a clinical and medico-legal point of view).

[Guest Paula]

I work FP on an isolated rural clinic. Here is what we do: We have POC tests for HgbA1C and I order that test. If extremely high.......10,11, 12, 13, etc., the patient gets insulin started that day and I give them their first injection. Honestly, I have not checked for the anion gap as I would have to wait for the result, too. I have found that oral agents take a really long time to be effective and I start a low dose insulin, and the insulin I pick is based on the patients ability to pay for the insulin. Sometimes it is a 70/30, sometime lantus or levimir. I might start metformin if A1C is 9. Please keep in mind my clinic is not like a "normal" clinic that has any quick access to endocrinology, ER, etc. We are way out of mainstream compared to large hospital and clinic systems, to take my words with a grain of salt.

 

I am able to daily see my patients when I start insulin, and monitor closely in those first few weeks. I believe glucose toxicity is a real issue and have not have much success with the oral agents initially if A1C higher than 10. I have been able to start insulin, monitor closely as b/s drop, start metformin, and a few patients have been able to stop the insulin and do just fine on metformin once A1C in the 6 range. However, if the random glucose is really high like 800 and A1C extremely high, like 15-18 we do refer to ER for management with the thought the patient needs hospital care for a few days, or at least we want them out of our hands and have that other set of eyes on the patient. It has only happened once that I can remember (in a type 2). Type 1 is a completely different thing, and we send them off to ER and then they get transferred to endocrinology.

 

It's worked for us. the MD here does the same.

[Guest Paula]

P.S. Our patients are not real good at following endocrinologist treatment plans. Too complicated. Most of the time if they get an endo referral they are no-shows, or start the plan, then quit going altogether and we end up managing anyway, which is how it should be. (except for type 1 of whom we have one patient).

 

I have not prescribed a sulfonylurea for several years as first line.

[True Anomaly]

Since you posted this in the EM section, I'll give a little perspective.

 

From a medico-legal EMTALA point of view, if they look good and there is no evidence of DKA or hypokalemia, there is no medical emergency.

 

Beyond that, there is no actual cut-off for when you treat hyperglycemia in the ER, or how you treat it. There have been studies that look at clinician comfortability with certain blood glucose levels, and when they'd actually start not just IV fluids but acute insulin treatment as well. But nothing in the literature says that once they're down to a certain level, you can safely discharge them home. The only thing the ADA says is that the DKA cutoff is 250 mg/dL, but of course we've all seen patients well above that who aren't in DKA.

 

I have seen other providers, and I believe I've done this once myself, start metformin on the appropriate patient (creatinine normal, no DKA, no potassium abnormalities, has good follow up, is reliable, unlikely to progress to DKA/HHNK) in the ER. I have never seen anyone start someone on insulin for the first time from the ER. If they are likely to need a home insulin regimen, currently my options are limited to putting them in an observation status and hoping I can get an endocrinology consult for appropriate therapy.

 

Just don't forget that, in light of a symptomatic patient with glucose levels going up into the 300's and 400's, you don't just need to look for evidence of ketones but also know what their potassium level is.

[cbrsmurf]

Good question. I, personally, would just start the patient on metformin (start low), send them home with precautions,go to ER if sx worsen, and f/u next day. Like true anomaly states, if they aren't in DKA (or HHS), it's not an emergency -- well that's my opinion anyways. I would like to see some literature on this topic.

 

I would also never start insulin as first-line in a type 2, symptoms or not. The single, greatest pharmaceutical reducer of cardiovascular events and mortality of pts with DM2 is metformin. If they are still having sx in 1 month (don't forget to add diet & exercise), then I would add either an incretin/DDP4 inhibitor (which usually don't do much) -> sulfonyurea and lastly insulin. All the new (and a lot of the old) studies are showing the tight glycemic control isn't as beneficial as it was once thought to be and actually more dangerous and/or detrimental to quality of life.

 

Sorry, getting off topic there, this is in the ED subforum, not FP. One question back on topic:

1 question, any role/influence for urine dipstick for ketones in the office?

[sbellin]

I agree that starting insulin from the ED or urgent care clinic is probably not the best choice; the teaching of proper injection technique and the ability to follow-up is usually not there. I’ve seen providers bothered by significant office hyperglycemia but realize that a Type 2 diabetic with a sugar of 450 and an A1c of 16 (average glucose over the last three month of 412) didn’t get there over night and it won’t be resolved by tomorrow no matter what therapy you choose; even if they’re admitted. Rapid return to euglycemia, if it could be done, in the outpatient setting is a recipe for disaster (hypoglycemia, hypokalemia). Assuming they have no acute pathology such as an infective process that’s driving the sugars higher now or symptomology which may suggest the need for admission, metformin would be an acceptable management option (if you’re convinced they’re a Type 2 and not Type 1): But even then you would be giving a starting dose which will have minimal short term effect but at least it looks like you did something. As for the insulin dosing if they’re an adult onset Type 1 10 units of Lantus may produce a profound reduction in their blood sugar whereas in a Type 2 that has severe insulin resistance it will have essentially no effect. The best thing you could do as an intervention is making a phone call to a PCP and connecting that patient directly to a provider who will manage this chronic condition in very short term follow-up. Giving them insulin or even metformin and sending them off without a connection for follow-up means that they in most cases won’t do anything. Also, remember that a majority of diabetic patients die from cardiovascular disease not hyperglycemia. So even correcting their hyperglycemia in isolation can generate in the patient and providers a false sense of security.