Roy Daniel Pollom, MD Treating hospitalized patients with diabetes for the past 30 years, I have learned to appreciate there are preventable causes of hyper and hypoglycemia. A paper written in 2005 enumerated the preventable causes of hyper and hypoglycemia in great detail (1). In this blog I will highlight what are commonly seen as the most common causes of severe hypoglycemia in the acute care setting. One of the most common causes of hypoglycemia is the continuation of home medications without a careful assessment of their appropriate use in the acute care setting. Often nurses report that a patient is admitted on industrial doses of insulin with no details about timing of the last dose. Medication reconciliation is part of the culture of safety caregivers provide to patients when they enter and leave our acute care facilities. On a good day the R.N. can confirm both the type of insulin used and the amount and timing of the last dose given. This effort prevents dose stacking which causes hypoglycemia due to the layering effect of insulin given too closely to the last dose. Often what is not confirmed is the caloric content of the patient’s home diet. Some patients have ramped up their basal insulin to the point where they are feeding their dose of insulin throughout the day and night to prevent hypoglycemia. Such a strategy inadvertently delays the timely use of more physiologic insulin regimens which include basal and prandial insulin for meal coverage plus prn correction doses. In the outpatient setting basal insulin’s are commonly titrated to extraordinary levels in an attempt to improve glycemic control. Once the patient is hospitalized in a calorie controlled environment, continuation of such large basal doses precipitates hypoglycemia. A patient history of frequent nocturnal hypoglycemia at home is a clue that the basal insulin dose needs to be reduced to prevent hypoglycemia in the hospital. I will commonly use 0.3-0.4 x the patient’s kilogram weight to derive a starting basal insulin dose when the home dose is frightfully high. Once hospitalized, the patient’s total insulin needs will often be more affected by their calorie controlled diet than the stress of their acute illness and bed rest. I refer to this as “The Frig Factor” and its influence on the selection of an initial basal dose can be profound. The ADA has recommended that basal insulin analogs be the basal insulin’s of choice and that NPH insulin has little place in the acute care setting. Prior insulin clamp studies by Tim Heise (2) have shown the shot to shot intrasubject variability ranges from 68% for NPH down to 27% with insulin detemir and 38% for insulin glargine. Comparator trials between NPH and the basal insulin analogs have shown up to a 50% reduction in nocturnal hypoglycemia with the basal analogs (3). Combinations of regular insulin and NPH are commonly associated with nocturnal and late morning hypoglycemia. Peaking insulin’s like NPH as well as the pre- mix insulin’s can cause major hypoglycemia in the late morning when lunch is delayed or skipped due to a change in patient status or testing. Every year patients are admitted to the hospital and develop severe hypoglycemia during the day of admission caused by a sulfonylurea (SU) that was taken the night prior to admission. For this same group, the stage is set for further episodes of hypoglycemia when patients on SUs are made NPO, are too sick to eat, or have altered meal schedules due to testing or changes in their medical status. In a US National Health and Wellness Survey, 54% of internet responders reported moderate to severe hypoglycemia on OAD treatment for diabetes. SU regimens in addition to basal insulin are also associated with much higher incidence of hypoglycemia. One possible solution to this preventable cause of hypoglycemia is to prohibit the use of SUs in the hospital until the caregiver can confirm the patient is eating without hesitation and has no planned meal time interruptions. The same qualifiers might suffice for a trial on premix insulin prior to discharge. Certain adverse drug events are as predictable as they are preventable.
|
