Insulin Regimens to Treat Hyperglycemia in Hospitalized Patients on Nutritional Support: Systematic Review and Meta-Analyses
What this study found
NPH insulin was identified as the most effective regimen for reducing mean blood glucose levels, although the clinical significance of the difference was questioned. There was considerable variability in defining hyperglycemia and hypoglycemia among studies, and few examined mortality and length of hospital stay.
- Study & population
- Seventeen studies comprising 3,260 patients were included, with a mix of randomized controlled trials and cohort studies comparing various insulin regimens for glycemic control in patients on enteral, parenteral, or combined nutritional support.
- Intervention
- This systematic review evaluated the effectiveness of different insulin regimens for managing hyperglycemia in hospitalized patients receiving nutritional support, including enteral and parenteral nutrition.
- Key limitation
- Lack of homogeneity in patient populations, methodologies, and definitions of outcomes limited direct comparisons and the ability to draw firm conclusions.
View source
Open PDFOriginal abstract
Background: The best insulin regimen to treat hyperglycemia in hospitalized patients on nutritional support (NS) is unclear. Methods: We searched electronic databases to identify cohort studies or randomized clinical trials in order to evaluate the efficacy of different insulin regimens used to treat hyperglycemia in hospitalized patients on NS on diverse outcomes: mean blood glucose (MBG), hypoglycemia, length of stay in hospital, and mortality. Results: Seventeen studies from a total of 5,030 were included. Enteral Group included 8 studies; 1,203 patients using rapid, glargine, NPH, or Premix insulin; MBG 108-225 mg/dL; hypoglycemia 0-13%. In indirect meta-analyses, NPH insulin ranked best for glucose control (MD 95% CI -2.50 mg/dL [2.65 to -2.35]). Parenteral Group included 4 studies; 228 patients using regular and glargine or NPH insulin; MBG 137-202 mg/dL; hypoglycemia 0-40%. In meta-analyses comparing regular insulin added to parenteral nutrition bag with glargine, MBG (MD 95% CI -3.78 mg/dL [-11.93 to 4.37]; I2 = 0%) or hypoglycemia frequency (RR 95% CI 1.37 [0.43-4.32]; I2 = 70.7%) did not differ. The description related to hospital length of stay and mortality was inconsistent between groups. Conclusions: The best insulin regimen to treat hyperglycemia in hospitalized patients on NS has not been established; best results using insulin regimens with NPH in enteral nutrition do not seem to be clinically relevant.