SCCM Resource Library-48-Outcomes and Adverse Effects of Extreme Insulin Resistance in ICU Patients

Video Transcription

Good afternoon. My name is Ariana Bidger. I am the current PGY-2 Critical Care Pharmacy resident at Indiana University Health, and  I'll be presenting today on outcomes and adverse effects of extreme insulin resistance in ICU patients. I don't have any conflicts of interest to disclose.  So looking broadly at hyperglycemia in the ICU, we know this is a commonly observed phenomenon that can result from a variety of causes relating to baseline insulin resistance,  the stress response, and things we administer to patients in the ICU, among other things. There are several guidelines for inpatient glycemic management, both from the American  Diabetes Association, published annually, and then critical care-specific guidelines published by Critical Care Medicine in 2012.  These propose low glucose ranges as well as methods for insulin delivery. Notably though, there are no specific thresholds for diagnosis or guidance for management for  those patients with extreme insulin resistance. So with that in mind, the goal of our study was to evaluate the characteristics, hospital  course, and outcomes of critically ill patients with extreme insulin resistance, and then compare them to those with more average requirements.  Anecdotally, we had had difficulty managing these patients at our institution, and we expected that we would see more hypoglycemia as well as more difficulty achieving glucose  control. We also thought we might observe some differences in baseline characteristics that would help to identify patients at risk for this phenomenon.  This was a retrospective observational study. It took data from a four-year period between 2016 and 2019, including adult patients admitted  to ICUs at either of our downtown Indianapolis hospitals. We included patients receiving IV insulin infusions for at least 12 hours, and then  broke them up by maximum rate of insulin infusion, categorizing those as extreme, who required  greater than or equal to 35 units per hour, compared with those with lower requirements.  We justified this as kind of the 99th percentile of what we normally observed, so designating a subset of patients who were outside of the norm.  We excluded patients who were less than 18, receiving insulin infusion for treatment of suspected overdose, DKA or HHS, or those with incomplete medical records regarding insulin  infusion. Our primary endpoints were first incidents of hypoglycemia, defined as a blood glucose  less than 70, and then time to glucose control, which was defined as the first of three glucose values within the target range, separated by at least one hour.  Secondary endpoints included hypoglycemia separated by severity, so less than 50 milligrams per deciliter was our threshold for severe hypoglycemia.  Also time to first glucose within goal range, ICU and hospital length of stay, and then in-hospital mortality. The methods we used for statistical analysis are detailed below.  We also did do a post hoc matching analysis based on the following factors due to expected influence of these on attaining glucose control while in the ICU.  We started with a sample of around 9,000 patients. After screening, we were left with 32 patients in the extreme group for analysis, so quite  a small subset of the total population here, and then matched these in a three-to-one ratio with patients requiring lower amounts.  These patients were stratified by rate, designating the bottom third, middle third, and top third, contributing patients equally to the comparator group.  Demographic-wise, our mean age was 55 to 60. Notably the mean BMI was greater than 30, so indicating a large subset of the population  was obese, and then white patients were the most common race. I'm looking at past medical history, so there was significantly more type 2 diabetes in  the extreme group, 78% compared with 43% in the comparator group. We also saw poorer baseline glucose control, so median hemoglobin A1c was 8.1 in the extreme  group compared with 6.2 in the comparator. With home medications, there was significantly more insulin use, 65% in the extreme group  compared with 33% in the lower insulin requirements. Primarily, patients in the extreme group were admitted for a medical reason, so 63% compared  with 20% in the non-extreme group, which was statistically different. We've reported SOFA score at the time of ICU admission, both organ-specific and total,  with no major differences. Median SOFA was 7 to 7.5 in both groups. Looking at inpatient therapies, the most notable difference was a higher rate of steroid use  in the extreme group, so 75% compared with 39% in the non-extreme group. Looking at insulin use, the default goal range per our protocol is 100 to 150, though a custom  glucose range can be designated by the provider. Most patients in both groups used this default range, although more patients in the extreme  group had a custom goal range set, which was typically a broader range and more liberal. Duration of infusion was also significantly longer in the extreme group compared with  the non-extreme group. Looking at primary endpoints, for hypoglycemia, overall, there was more hypoglycemia in the  extreme group compared with the non-extreme group, 63% compared with 34%. This is mostly driven by mild to moderate hypoglycemia, which was statistically higher  in the extreme group compared with the non-extreme group. Severe hypoglycemia was 8 to 9% in both groups and not statistically different.  Looking at time to glucose control, so I've listed here both first glucose in range and then glucose control.  Notably, these times were longer for the extreme group in both cases than the non-extreme group.  Also point out that the non-extreme group had minimal difference in the time to first glucose in range to glucose control, showing that we achieved glucose control faster and  maintained it versus the extreme group where it took longer even once we had one glucose in the goal range.  Being at secondary endpoints, so ICU length of stay was significantly longer in the extreme group compared with the non-extreme group.  Hospital length of stay was numerically longer but not statistically different. And then in-hospital mortality was significantly higher in the extreme group compared with  the non-extreme group, which was statistically significant. As mentioned, we did do a post hoc matching analysis. The factors that we matched on are included in the footnote.  We were unable to match for reason for admission, so significantly more extreme patients were  admitted for a medical reason compared with the matched cohort. Looking overall at endpoints, so incidence of hypoglycemia was similar findings to the  main analysis with higher rates in the extreme group compared with the non-extreme matched cohort, mostly driven by mild to moderate hypoglycemia.  And then time to first glucose in range and time to glucose control were both significantly longer in the extreme group compared with the matched cohort.  Overall strengths, so this is a relatively rare phenomenon, though to our knowledge, this is the largest report of this specific subset.  We did look at clinical outcomes, both for safety and efficacy with insulin use, and we did perform a post hoc matching analysis to try and account for some baseline imbalances  between groups. Weaknesses, there is not an available definition of insulin resistance based on infusion rates,  so ours was by nature determined based on what we had commonly observed in our hospital. Insulin resistance was not directly measured, and there were some imbalances in baseline  characteristics that we were unable to match for. And then of course, this was a single center study, so it is influenced by the specific policies and practices at our institution.  So overall conclusions, patients with extreme insulin resistance were more likely to have hypoglycemia and took longer to achieve glucose control.  We may have to consider alternative approaches for patients requiring these highest rates of IV infusions.  These may include a more liberal glucose goal range, particularly for patients with poor baseline glycemic control.  And then frequent monitoring and caution given the high rates of hypoglycemia that we observed in this study. All right, so that concludes my presentation.  I appreciate your attention.  Thank you.

Video Summary

The study evaluated the characteristics, hospital course, and outcomes of critically ill patients with extreme insulin resistance in comparison to those with lower insulin requirements. The researchers found that patients with extreme insulin resistance were more likely to experience hypoglycemia and it took longer for them to achieve glucose control. They suggested considering alternative approaches for patients with high insulin infusion rates, such as a more liberal glucose goal range and frequent monitoring. The study revealed that patients with extreme insulin resistance had longer ICU stays and higher in-hospital mortality rates. However, the study had some limitations, including being a single-center study and imbalances in baseline characteristics between the groups.

Asset Subtitle

Pharmacology, Endocrine, 2021

Asset Caption

The Society of Critical Care Medicine's Critical Care Congress features internationally renowned faculty and content sessions highlighting the most up-to-date, evidence-based developments in critical care medicine. This is a presentation from the 2021 Critical Care Congress held virtually from January 31-February 12, 2021.

Arianna Vidger

Introduction/Hypothesis: Extreme insulin resistance is a phenomenon reported in some ICU patients. Clinical data to guide management of these patients are lacking. The goal of this study was to describe the characteristics, hospital course, and outcomes of critically ill adult patients with extreme insulin resistance.

Methods: Adult ICU patients on IV insulin infusions between January 2016 and December 2019 were eligible for inclusion in this retrospective observational cohort study. Patients on insulin infusion <12 hours, receiving insulin infusion for suspected overdose or acute hyperglycemic crisis, and those with incomplete medical records were excluded. Extreme insulin resistance was defined as being on an insulin infusion with at least one charted rate â‰¥35 units/hour. Patients with extreme insulin resistance were included in a 1:3 ratio to patients without extreme insulin resistance. The primary endpoints were rate of hypoglycemia and time to glucose control.

Results: One hundred twenty-eight patients were included, 32 had extreme insulin resistance. Baseline characteristics varied between groups. Patients in the extreme group were more likely to have type 2 diabetes (78% v. 43%,p< 0.05), be on insulin prior to admission (65% v. 33%,p< 0.01), have a higher hemoglobin A1c (median 8.1 v. 6.9,p=0.019), be admitted for a medical reason (63% v. 20%,p<0.001), and receive steroids during admission (75% v. 39%,p<0.001). Median admission SOFA scores were similar between groups (7.5 v. 7,p=0.433). Any hypoglycemia (63% v. 34%,p=0.005) and moderate hypoglycemia (50-70 mg/dL) (53% v. 26%,p<0.01) were more common in the extreme group. Severe hypoglycemia (<50 mg/dL) was similar between groups (9% v. 8%,p =0.86). The extreme group had a longer time to glucose control (median 19.8 v. 5.7hr,p<0.001), received higher maximum insulin infusion rates (median 39.8 v. 7.6 units/hr,p<0.001), had a longer ICU length of stay (median 9.6 d v. 5.1d,p=0.027), and had a higher mortality rate (34% v. 15%,p=0.014).

Conclusions: ICU patients with extreme insulin resistance had higher rates of hypoglycemia and took longer to achieve glucose targets compared to those without extreme insulin resistance. An individualized approach may be required to avoid hypoglycemia in this patient population.

Meta Tag

Content Type Presentation

Knowledge Area Pharmacology

Knowledge Area Endocrine

Knowledge Level Advanced

Membership Level Select

Tag Diabetes

Tag Insulin

Year 2021

Keywords

critically ill patients

extreme insulin resistance

hypoglycemia

glucose control

ICU stays

	Society of Critical Care Medicine 500 Midway Drive Mount Prospect, IL 60056 USA Phone: +1 847 827-6888 Fax: +1 847 439-7226 Email: support@sccm.org	Contact Us About SCCM Newsroom Advertising & Sponsorship DONATE	MySCCM LearnICU Patients & Families Surviving Sepsis Campaign Critical Care Societies Collaborative	GET OUR NEWSLETTER
© Society of Critical Care Medicine. All rights reserved. \| Privacy Statement \| Terms & Conditions The Society of Critical Care Medicine, SCCM, and Critical Care Congress are registered trademarks of the Society of Critical Care Medicine.