It’s been almost 5 years since we last did a comprehensive column on insulin-related issues (see our November 2, 2010 Patient Safety Tip of the Week “”). More recently we’ve focused more on issues related to insulin pumps and insulin pens and related devices. But 3 recent articles again remind us of the myriad of problems that may occur with insulin.
The first, published by ISMP Canada, is a very instructive case in which insulin administered via an incorrect route for treatment of hyperkalemia resulted in profound hypoglycemia and consequent brain injury (ISMP Canada 2015). The patient was a nondiabetic patient receiving hemodialysis who was admitted to a surgical unit with a bowel obstruction. The patient was found to have significant hyperkalemia and a physician handwrote orders for salbutamol by inhalation, 1 ampule of D50W (route not specified), and regular insulin 10 units (route not specified). The D50W was given intravenously but the regular insulin was given by subcutaneous injection. When later found unresponsive, the patient’s blood glucose level was undetectable. Subsequently it was also determined that the initial potassium level was falsely elevated due to specimen hemolysis.
ISMP Canada does an excellent job describing multiple aspects of the case. But perhaps the most salient lesson is has to do with the difference in duration of action for the D50W and regular insulin administered via the subcutaneous route. For management of hyperkalemia insulin is supposed to be delivered via the IV route, where its action is immediate. But when administered via the subcutaneous route the action of the insulin is delayed and its duration is prolonged. Hence, in this case the action of insulin extended beyond the period of protection by D50W and profound hypoglycemia resulted.
The original orders were handwritten and routes were not specified by, nor subsequently clarified with, the ordering physician. The insulin order was not reviewed by a pharmacist (ward stock availability of both the D50W and insulin bypassed pharmacist review). The nurse’s familiarity with the subcutaneous route for insulin administration may have introduced confirmation bias. A second nurse reviewing the order verified the route of administration that had been on the transcription of the handwritten order (rather than seeing the original handwritten order?). And no order had been given for monitoring blood glucose levels after administration of the insulin. They also noted that different treatment approaches would likely have resulted if the hemodialysis service had been consulted and if the falsely elevated potassium level had been recognized. A significant problem was the lack of a standardized order set for management of hyperkalemia.
They provide a number of recommendations for hospitals, prescribers, healthcare IT systems, nurses and pharmacies to help avoid similar occurrences. But a key recommendation is development and use of order sets for off-label use of commonly used medications, particularly those that are high-alert medications.
This is an excellent article with multiple lessons learned. We hope you will read the article in full.
The second recent article reminds us of the continuing dangers related to U-500 insulin (Leuck 2015). The article is written from the perspective of a pharmacist and contends that any order for U-500 insulin should be considered a red flag and require review by a pharmacist. We couldn’t agree more! The article points out the two biggest reasons for errors related to U-500 insulin: (1) the syringes used and (2) the different units/mL. Unfortunately, there still are currently no syringes calibrated for the higher concentrations so healthcare workers often use this dosage form with insulin syringes that are calibrated for 100 unit per mL insulin preparations. That’s a setup for errors. The author also notes that it may be unclear when a physician writes a prescription for, say, 25 units of U-500 insulin whether they really mean 25 units or 25 mL (the latter representing 125 units of insulin). So the author recommends electronic prescribing systems and EHR’s be programmed to present an alert that would result in a pharmacist reviewing the order with the prescriber and then making sure that nursing staff understand how it is to be administered.
We discussed the syringe issue in our November 2, 2010 Patient Safety Tip of the Week “ and only be used by specially designated healthcare workers who have received specific training in use of the higher concentration product. Certainly, they should never be stored as part of “floor stock” where they might be easily mixed up with the more common 100 unit per mL preparations. We even raised the question as to whether U-500 insulin should be stocked at all in a hospital. Unfortunately, the need for that higher concentration has been increasing in recent years as more obese patients with insulin resistance have been requiring much higher insulin doses and the higher concentrations may also be needed for patients having implanted insulin pumps. We recommended that if you need to stock the higher concentration, make sure you have some mechanism (like a “hard stop”) to ensure that a pharmacist is involved in validating and preparing the dose and always have an independent double check before such preparations are administered. ”. In that column we also recommended that the 500 unit per mL preparations be stored completely separate from the other insulin preparations
The third recent article by the ECRI Institute highlights the vulnerabilities that occur at each step of the entire medication process continuum for insulin (ECRI 2015). They start at the prescribing stage where the classic problem is use of the abbreviation “U” for units. The problem arises when the “U” looks like a zero so the patient inadvertently is given 10 times too high a dose. (Note that similar problems can arise when the abbreviation “IU” for international units is used.). In our November 2, 2010 Patient Safety Tip of the Week “” we also noted that we have occasionally found these dangerous abbreviations in old order sets on computer systems or in order sets “customized” by physicians and not properly vetted. Under the prescribing stage the ECRI article also includes the risk of confusion among the many look-alike sound-alike (LASA) insulin formulations. In our previous article we gave some examples of the LASA errors and noted the importance of using tall-man lettering to help avoid some of these errors. We also highlighted the problem that occurs when computerized screens truncate choices on drop-down menus.
At the transcribing stage the ECRI article highlights the problems related to illegible writing and use of acronyms. They also highlight the problem of using outdated medication lists during medication reconciliation when the patient arrives at the hospital. Obviously, insulin doses can be changed frequently and careful reconciliation on admission (and discharge or transfer) is very important. Equally important when medication reconciliation is done on admission is determining accurately when the patient took their last dose of insulin (and their last meal). That is one of the most common errors we find with medication reconciliation. One other problem we have encountered is that many notes, even in the electronic medical record, are still dictated and transcribed. If a physician does not carefully review those transcriptions, errors in insulin dose or preparation may be propagated to other settings.
At the dispensing stage the ECRI article emphasizes failure to double check. That double check should include the insulin product, the concentration, and the dosage. Care needs to be taken here to avoid look-alike sound-alike (LASA) errors. Our November 2, 2010 Patient Safety Tip of the Week “” discussed the many reported errors in which heparin and insulin vials are confused.
At the administration stage the ECRI article discusses incorrect doses, formulations, and concentrations plus LASA errors. It also includes here issues with misuse of insulin pens (see our many columns on insulin pen issues listed at the end of today’s column) and problems with erroneous recording of patient’s glucose. Problems with patient identification appear here. They also note here the issue of timing of meals (or lack thereof) in relation to insulin administration.
Here we again would add failure to double check as a hazard. In our October 16, 2012 Patient Safety Tip of the Week “What is the Evidence on Double Checks?” we did recommend that double checks be done before administering high-risk medications such as insulin. It is important here also not to over-rely on technology. One problem encountered several times is when the physician prescribes the wrong insulin dose. The nurse may recognize it might be too high a dose but the bar coding system tells the nurse that this is the correct patient and this is the correct insulin preparation and dose that is in the computer. So use of a barcoding system does not obviate the need for double checks.
At the monitoring stage the ECRI article notes problems such as failure to monitor a patient’s response to insulin so that appropriate changes in insulin dose are not made.
We refer you to our November 2, 2010 Patient Safety Tip of the Week “” for numerous other important details and links to some excellent resources. It also deals with issues such as problems that can arise during hospital transports and handoffs, problems with co-management of patients, use of insulin drips, and avoiding “sliding scale” insulin coverage. But we think most of the recommendations listed in that column are still worth repeating here:
ISMP reminds us that not all healthcare providers have a good understanding of insulin dosing. A 2011 report by ISMP noted instances where residents or other physicians were administering insulin and made significant errors (ISMP 2011). It is a reminder that many physicians have never received formal training on insulin administration. We should never assume providers know how to administer insulin.
In 2013 the American Society of Health-System Pharmacists (ASHP) Research and Education Foundation convened a multidisciplinary panel of experts to address safe insulin practices. The panel made 10 recommendations (Cobaugh 2013) that address the areas of prescribing, dispensing and storage, administering, monitoring, evaluating and planning:
And while most of our discussion has focused on insulin errors in hospitals, we should be reminded that they also occur in the ambulatory setting. A review of data from poison control centers showed a substantial trend of increasing reports of unintentional therapeutic errors involving insulin from 2000 to 2009 (Spiller 2011). These were primarily in adults age 40 and older but females were disproportionately involved and events occurred more often in late evening. It is not clear whether this reflects the types of error or simply who uses the poison control centers. Unfortunately, the study did not include data on the nature of the therapeutic errors.
When hospitals ask us what are good topics for FMEA’s (failure mode and effects analyses) we often suggest insulin management as an excellent one. Problems can occur at so many stages of the whole insulin management system that doing a FMEA on this can be very rewarding.
Some of our prior columns highlighting the safety issues of insulin, insulin pumps, insulin pens and similar devices:
November 2, 2010 “Insulin: Truly a High-Risk Medication”
September 18, 2012 “Insulin Pump Safety”
February 26, 2013 “Insulin Pen Re-Use Incidents: How Do You Monitor Alerts?”
April 2013 “More Tips on Insulin Pen Safety”
April 2014 “Insulin Pens - Again”
March 10, 2015 “FDA Warning Label on Insulin Pens: Is It Enough?”
ISMP Canada. Wrong-Route Incident Involving Insulin and Dextrose Prescribed for Hyperkalemia. ISMP Canada Safety Bulletin 2015; 15(2): 1-4 March 5, 2015
Leuck S. Dosing Dangers of U-500 Insulin. Pharmacy Times 2015; April 7, 2015
ECRI Institute. Insulin Errors Occur Across the Medication Use Process. ECRI Institute’s PSO Monthly Brief. February 2015
FDA. FDA Drug Safety Communication: FDA requires label warnings to prohibit sharing of multi-dose diabetes pen devices among patients. FDA Safety Announcement 2015; February 25, 2015
ASHP (American Society of Health-System Pharmacists) and Hospital and Health-System Association of Pennsylvania. Professional practice recommendations for safe use of insulin in hospitals. Rockville (MD): American Society of Health-System Pharmacists
Cobaugh DJ, Maynard G, Cooper L, et al. Enhancing insulin-use safety in hospitals: Practical recommendations from an ASHP Foundation expert consensus panel. Am J Health-Syst Pharm 2013; 70(16): 1404-1413
ISMP (Institute for Safe Medication Practices). Misadministration of IV insulin associated with dose measurement and hyperkalemia treatment. ISMP Medication Safety Alert! Acute Care Edition. August 11, 2011
Spiller HA, Borys DJ, Ryan ML, et al. Unintentional Therapeutic Errors Involving Insulin in the Ambulatory Setting Reported to Poison Centers. The Annals of Pharmacotherapy 2011; 45(1): 1-6
Print “PDF version”