Postprandial insulin may cut hypoglycemia rates for hospitalized adults with diabetes
Click Here to Manage Email Alerts
Key takeaways:
- Hypoglycemia rates declined after a hospital changed nutritional insulin timing from preprandial to postprandial.
- No increases in severe hyperglycemia were observed with the new policy.
Changing in-hospital nutritional insulin administration from before meals to after meals lowered hypoglycemia rates at a community hospital, according to study findings published in Diabetes Research and Clinical Practice.
In outpatient settings, most rapid-acting insulins can be used after meals, according to Andrew P. Demidowich, MD, assistant professor of medicine in the division of endocrinology, diabetes and metabolism at Johns Hopkins University School of Medicine. However, he said, standard of care in hospitals has dictated giving insulin to patients before meals, which could lead to hypoglycemia if the patients does not eat their meal. On June 28, 2019, Johns Hopkins Howard County Medical Center instituted a policy change whereby nutritional insulin would be given to patients after at least half of their meal was consumed instead of before the meal. Researchers sought to compare outcomes immediately after the policy change and those more than 3 years after the policy change with outcomes from when the old policy was in effect.
“There have been a few small- to medium-sized randomized controlled trials and retrospective studies looking at the outpatient space, but nobody has ever looked at the difference between preprandial and postprandial rapid-acting insulin administration in the inpatient space,” Demidowich told Healio. “It’s a very important gap in the medical knowledge.”
New policy reduces hypoglycemia
Demidowich and colleagues conducted a retrospective cohort study of hospitalized adults at Johns Hopkins Howard County Medical Center receiving rapid-acting nutritional insulin. The hospital changed its policy for administering nutritional insulin from before a meal was eaten to after at least 50% of a meal was consumed. Outcomes were collected before the policy change from September 2018 to June 27, 2019; immediately after the change from July 2019 to February 2020; and during a distant post-intervention period from September 2022 to August 2023. The primary outcome was the percentage of patient-days with hypoglycemia with a glucose concentration of 70 mg/dL or less. Secondary outcomes included percentage of days with moderate hypoglycemia with glucose concentration of less than 54 mg/dL, severe hypoglycemia with a glucose concentration of 40 mg/dL or less, severe hyperglycemia with a glucose concentration of 300 mg/dL or higher, and daily mean glucose concentration and length of hospital stay.
Data were collected for 1,948 patient-days before the policy change, 1,751 patient-days immediately after the policy change and 3,244 patient-days in the distant post-intervention period. In unadjusted analysis, the percentage of patient-days with hypoglycemia declined from before the policy change to the distant post-intervention period (9.4% vs. 6.7%; P < .001). No changes were observed in moderate or severe hypoglycemia or severe hyperglycemia. Mean daily glucose increased from 194.6 mg/dL with preprandial insulin administration to 199.3 mg/dL in the distant postprandial insulin period (P = .003). No change in length of hospitalization was seen.
Adults hospitalized during the distant postprandial insulin administration period were less likely to experience any hypoglycemia (adjusted OR = 0.71; 95% CI, 0.57-0.87) or severe hypoglycemia (aOR = 0.44; 95% CI, 0.24-0.82) than adults who received preprandial insulin.
Challenges and opportunities
No differences in any outcomes were observed from before the policy change to immediately after the policy change. Demidowich said hypoglycemia may not have significantly declined in the months following the policy change as hospital staff adjusted to the new policy. When insulin was administered before meals, nurses were sometimes stressed about whether a patient would eat afterward, according to Demidowich. With postprandial insulin, that stress shifted to figuring out the timing of each patient’s meal.
“Our hospital, like many hospitals, has meals on demand,” Demidowich said. “The nurse has to be vigilant with not only when the meal arrives, but when is it consumed. We haven’t come up with a good way to communicate or coordinate to nurses when the meal has been consumed.”
Despite the challenges, the researchers believe the policy change had a positive outcome and could have a major impact for the hospital down the road. In the study, the authors wrote that CMS recently added severe hypoglycemia and severe hyperglycemia as new electronic clinical quality measures for hospitals and may financially reward or penalize hospitals based on their performance on those metrics in the future.
“Every hospital in the country right now is trying to figure out how to save money,” Mihail Zilbermint, MD, MBA, associate professor of clinical medicine at the Johns Hopkins University School of Medicine and chief and director of endocrinology, diabetes and metabolism at Suburban Hospital in Bethesda, Maryland, told Healio. “[Hospitals] will probably look for any opportunity to deploy different strategies, and I think that shifting the nutritional insulin to after meals is going to help hospitals reduce their [hypoglycemia] rates.”
Findings from the study should be confirmed in prospective randomized controlled trials, according to Demidowich. He added that future studies should be conducted in both tertiary academic care settings and community hospitals, and should also assess nurse satisfaction, patient satisfaction and in-hospital complications.
For more information:
Andrew P. Demidowich, MD, can be reached at ademido1@jhmi.edu.
Mihail Zilbermint, MD, MBA, can be reached at mzilber3@jhmi.edu.
Perspective
Back to Top
This is an original study that sheds light into the challenges of insulin therapy in hospitalized patients. Since the adoption of meals on demand by hospitals many years ago, the appropriate timing of meal delivery and consumption to glucose monitoring and insulin administration has been the Achilles heel when practicing the guidelines that recommend basal-bolus insulin therapy. This study provides data that hypoglycemia can be reduced — albeit with a small increase in average glucose level, but not severe hyperglycemia — by changing the timing of the prandial insulin from premeal to post-meal.
However, I believe the more essential change was not only the timing, but the requirement of witnessing that at least 50% of the meal was consumed before insulin is administered. This is likely the most influential practice change, one that is a high risk factor for hypoglycemia in acutely ill patients and cannot be incorporated with premeal administration.
Our hospital has similar flexibility with directions for the nurse to administer the insulin either within 10 minutes premeal or immediately after the meal if intake is uncertain to allow for assessment of the intake to justify the insulin. However, this adds burden to the nursing workflow and therefore has variability in practice. In addition to insulin timing, the essential component of glucose monitoring must be done before the meal consumption for accuracy, because this also influences the correction scale dose that is often concurrently given with the prandial insulin. Given the high prevalence of hyperglycemia and that diabetes ranges up to 40% in hospitalized patients (Umpierrez GE, et al. J Clin Endocrinol Metab. 2002;doi:10.1210/jcem.87.3.8341), most of whom will need insulin therapy, this burdensome workflow can be demanding on a nursing shift.
We have many quality assurance organizations, including CMS, that require hospitals to report their outcomes on hypoglycemia and hyperglycemia to avoid harm from insulin therapy. We need more interventions that outline best practices to follow guideline recommendations and meet these quality outcomes.
Collapse
Read more about
Click Here to Manage Email Alerts