Inpatient insulin infusion pumps
Question from Siobhain McHale:
There is a need for policy and procedure to be written if one is to allow an inpatient to keep their insulin infusion pump while in the hospital.
Are there any hospitals that have policies and procedures in place that would allow inpatient’s to keep their insulin infusion pumps and use them during their hospital stay? Is there any precedent or research that either supports this practice or is there literature that this would be too risky?
Answer from Irl B. Hirsch, MD:
As insulin pump use continues to grow, not only in the US but around the world, this topic is receiving much more attention. Currently, there is no standard on how to best deal with CSII in the hospital setting. Importantly, there are no randomized trials (to my knowledge anyway) using CSII in patients receiving outpatient pump therapy and assessing subsequent outcomes when continued on their pump.
The situation is complicated by the degree of illness (and cognition) by the patient and therefore level of potential for self-care and the level of comfort of insulin therapy in general and CSII in particular by the treating physician. Often the treating physician in the hospital has little to no experience with CSII further complicating what the best practice may be. On top of the typical patient on a medical floor, there are futher issues to consider for the patient in surgery if the anesthesiologist is not familiar with CSII.
With this background, each hospital needs to have a written policy on how to best handle patients using CSII. There are many potential options. In my hospital for example if the patient is able to self-mange his or her diabetes we simply allow the patient to handle the insulin with the nurse recording all insulin bolus insulin for the medical record. Other hospitals require automatic referral to an endocrinologist or endocrine team (which often includes a nurse or nurse practitioner with expertise in CSII). I have also seen hospital policy demand the patient remove the pump and instead move to multiple injections (often a “sliding scale” mentality with no scheduled insulin). Obviously, this is doomed to failure. If a decision is made to stop the pump and instead use multiple injections, at the very least basal insulin (glargine or detemir) and prandial insulin (rapid acting analogue) should be the approximate doses used with the pump.
This deserves more study but due to such variability in the understanding of insulin and CSII, any study may be difficult to interpret in the “real world”.