Diabetes mellitus and hypoglycemia as a complication of intravenous insulin to treat hyperkalemia in the emergency department
Abstract
We read with interest the article by Scott NL et al. "Hypoglycemia as a complication of intravenous insulin to treat hyperkalemia in the emergency department" that have been published in the American Journal of Emergency Medicine. This interesting retrospective study reports on the rate of hypoglycemia following intravenous insulin treatment for hyperkalemia in the emergency department setting and furthermore describes which variables were independently associated with this outcome: Lower glucose level prior to insulin administration, higher doses of administrated insulin and lower doses of administered 50% dextrose were independently associated with hypoglycemia in the multivariate analysis. Age, history of diabetes, and history renal failure were not independently associated.The authors of the study have collected and analyzed a large amount of data, leading to interesting debate. Nevertheless, we would like to address them one question that may contribute to further, more detailed discussion of the issue. The authors as well as in several other studies, involved diabetes mellitus into the analysis as one disease. But diabetes mellitus is a group of metabolic diseases resulting from various pathogenic defects which may be treated by different ways and therefore a risk of hypoglycemia may differ between individual patients. Based on our clinical experience we can speculate that the highest risk of hypoglycemia after insulin administration due to treatment of hyperkalemia is associated with type 1 diabetes patients. The main reasons for our believe are
1) these patients are at higher risk of hypoglycemia due to insulin treatment itself simply said due to failure to clear circulating insulin during hypoglycaemia (this may probably explain 4% of patients experiencing hypoglycemia without i.v. insulin administration mentioned in the study) and hyperkalemia related insulin dose may interfere with previously injected insulin by a patient;
2) lower glucose threshold for release of counterregulatory hormones; and loss of normal pancreatic alpha cell responses;
3) impaired awareness of hypoglycaemia which may be present in 30% and even more patients,
4) usually good insulin sensitivity [4] (making them more sensitive to even smaller doses of insulin, moreover if they are added to their regular daily doses). Therefore we believe that it would be interesting to analyze this group of patients to evaluate the risk of hypoglycemia related to hyperkalemia treatment with insulin separately. Because above mentioned disorders 1-3 are also to some extent presented in type 2 diabetes patients, those treated with insulin or sulphonylurea derivates could also be an interesting focus on analysis. We, with great respect, suggest taking these comments into the account and also consider them, if the continuation of the study is planned.