Hypoglycaemia due to insulin therapy for the management of hyperkalaemia in hospitalised adults: A scoping review.

Change log
Chothia, Mogamat-Yazied  ORCID logo  https://orcid.org/0000-0002-9801-1300
Humphrey, Toby 
Schoonees, Anel 
Chikte, Usuf Mohamed Ebrahim 
Davids, Mogamat Razeen  ORCID logo  https://orcid.org/0000-0003-4900-0231

INTRODUCTION: Hyperkalaemia is a very common electrolyte disorder encountered in hospitalised patients. Although hypoglycaemia is a frequent complication of insulin therapy, it is often under-appreciated. We conducted a scoping review of this important complication, and of other adverse effects, of the treatment of hyperkalaemia in hospitalised adults to map existing research on this topic and to identify any knowledge gaps. MATERIALS AND METHODS: We followed the PRISMA-ScR guidelines. Studies were eligible for inclusion if they reported on any adverse effects in hospitalised patients ≥18-years-old, with hyperkalaemia receiving treatment that included insulin. All eligible research from 1980 to 12 October 2021 were included. We searched Medline (PubMed), Embase (Ovid), the Cochrane Library, CINHAL, Africa-Wide Information, Web of Science Core Collection, LILACS and Epistemonikos. The protocol was prospectively registered with the Open Science Framework (https://osf.io/x8cs9). RESULTS: Sixty-two articles were included. The prevalence of hypoglycaemia by any definition was 17.2% (95% CI 16.6-17.8%). The median timing of hypoglycaemia was 124 minutes after insulin administration (IQR 102-168 minutes). There were no differences in the prevalence of hypoglycaemia when comparing insulin dose (<10 units vs. ≥10 units), rate of insulin administration (continuous vs. bolus), type of insulin (regular vs. short-acting) or timing of insulin administration relative to dextrose. However, lower insulin doses were associated with a reduced prevalence of severe hypoglycaemia (3.5% vs. 5.9%, P = 0.02). There was no difference regarding prevalence of hypoglycaemia by dextrose dose (≤25 g vs. >25 g); however, prevalence was lower when dextrose was administered as a continuous infusion compared with bolus administration (3.3% vs. 19.5%, P = 0.02). The most common predictor of hypoglycaemia was the pre-treatment serum glucose concentration (n = 13 studies), which ranged from < 5.6-7.8 mmol/L. CONCLUSION: This is the first comprehensive review of the adverse effects following insulin therapy for hyperkalaemia. Hypoglycaemia remains a common adverse effect in hospitalised adults. Future randomised trials should focus on identifying the optimal regimen of insulin therapy to mitigate the risk of hypoglycaemia.

Research Article, Medicine and health sciences, Biology and life sciences, Physical sciences, Research and analysis methods
Journal Title
PLoS One
Conference Name
Journal ISSN
Volume Title
Public Library of Science