Emergency room visits for problems from diabetes drugs   (December 5, 2007)

There's a recent article, Medication Use Leading to Emergency Department Visits for Adverse Drug Events in Older Adults, that surprised me a bit: among people who are 65 years of age or older and who are seeking emergency department care for adverse drug events (side effects), there are three commonly prescribed drugs that accounted for more than one third of such visits. And one of them is insulin (the others were warfarin and digoxin). Additionally, other medications on the list include several diabetes pills, such as metformin, glyburide, and glipizide. Although it's not stated clearly, the authors hint that the reason that insulin and the diabetes pills landed on the list is their ability to cause hypoglycemia -- rather than other possible reasons for drug side effects such as allergic reactions. And they don't describe whether the hypoglycemic events were severe, life-threatening, or fatal: but hypoglycemia sometimes can be, and that's a major concern for every person with diabetes (PWD).

Why would oldsters on insulin or other diabetes therapies be likely to land in ERs? I assume it'd be for the same reason as it would be for anyone with diabetes: hypoglycemia due to a mismatch in medication doses vs. food intake and/or exercise level. Perhaps it's a bit more likely for oldsters if the patient has other issues such as dementia, or if he/she has fallen and fractured a hip due to dizziness from hypoglycemia, or needs a smaller dose because of limited kidney or liver function, but overall, I think that hypoglycemia in patients of any age is simply a mismatch of meds/food/exercise.

What strategies could we use to decrease the frequency of hypoglycemia? I'd like to list a few.

1. Remain a little sweet. In the bad old days, we often encouraged patients to "remain a little sweet" (have hyperglycemia or trace urine glucose levels), because of the risks associated with hypoglycemia and the inability to detect it rapidly before home sugar testing became available. But now-a-days, we doctors encourage our patients to aim for the tightest possible control to avoid complications such as blindness and stroke. This brings up a strategy for use in a small subset of patients: if the person with diabetes has overwhelming dementia, plus blindness, plus amputations, and/or other disasters such as terminal cancer, the risk of hypoglycemia is clearly real, and the risk of future diabetic complications is diminished considerably by comparison. For such patients, keeping the patient out of DKA and hyperglycemic non-ketotic coma on the one hand and hypoglycemia on the other means it's okay to be a little sweet.
2. Better education of physicians and other health professionals. Diabetes docs and Certified Diabetes Educators know the necessity of teaching patients about hypos, but the average non-specialist simply doesn't have the time to teach everything to every patient. Hence my conclusion: the average doc should be taught to send every diabetes patient to a diabetes specialist or CDE. This has to be impressed upon physicians by their peers and professional organizations, by insurance companies, and by patients asking for referrals.
3. Better education of patients and family members, preferably by CDEs. Numerous examples come to mind: First, is there glucagon at home? There should be, for every patient who's taking insulin, unless they're living alone somewhere on a desert island. Family members or next-door neighbors will have to be taught how to inject it, and where it's stored. Glucagon could be life-saving in some scenarios: e.g., if a blizzard prevents ambulances to get to a residence where the patient collapsed after shoveling snow. Is the patient and family aware of night-time hypos and how to prevent them? That nocturnal hypoglycemia may occur following daytime exercise? That taking rapid-acting insulin at bedtime may cause a hypo at night? Is the bed-partner aware that thrashing and sweating may indicate the PWD is low? Does the patient have ready access to glucose-containing products to treat hypoglycemia promptly so a mild episode doesn't become severe? Have all family members been taught about recognition and treatment of hypos?
4. More attention to drug interactions, by pharmacists, physicians, and educated patients. Example: if the patient is only on metformin alone, it isn't likely to have hypoglycemia. However, treatment with metformin plus other diabetes drugs that do cause hypoglycemia may make hypos more likely and more severe. And occasionally, metformin plus non-drug reasons can cause hypos: such as when caloric intake is deficient, or when strenuous exercise is not compensated by caloric supplementation, or when there's heavy alcohol use. Finally, there are numerous other drug interactions with non-diabetes-drugs that might impact the frequency or severity of hypoglycemia: for instance, the class of drugs called "beta blockers" may make hypos more difficult for patients to recognize.

So, can hypos in the elderly (and other PWD) be diminished? My guess is yes, if sufficient attention is paid to this risk in the treatment plans that should be set up for each patient. It's up to all of us - physicians, CDEs, pharmacists, families, and patients themselves - to help assure that every person with diabetes has a clear, well-defined, workable, individualized treatment plan, and to make sure that prevention, recognition, and early treatment of hypoglycemia is part of the plan.