Background Dextrose is commonly administered with insulin during the management of hyperkalaemia to avoid hypoglycaemia. Previous research has evaluated the incidence of hypoglycaemia; however, none have reported the extent of blood glucose reduction after this regimen. The aim of this study was to better characterise the changes in blood glucose and to identify patients who may have an increased response to insulin.

Methods This was a multicentre retrospective study evaluating adult patients who received a regimen of 10 units of intravenous regular insulin plus 25 g of intravenous dextrose to manage hyperkalaemia between January 2014 and September 2016. The primary outcome was to evaluate the extent of blood glucose reduction (milligram per decilitre) up to 6 hours following the above regimen. Secondary outcomes included incidence of hypoglycaemia (blood glucose <70 mg/dL) and severe hypoglycaemia (blood glucose <40 mg/dL), and predictors of the extent of blood glucose reduction.

Results A total of 90 patients were included. The median blood glucose change over 6 hours was −24 mg/dL (IQR −53 to 6 mg/dL). Hypoglycaemia developed in 20 patients (22.2%, 95% CI 14.1% to 32.2%) and five patients (5.6%, 95% CI 1.8% to 12.5%) had severe hypoglycaemia. Patients who developed hypoglycaemia had a median baseline blood glucose of 110 mg/dL (IQR 80 to 127 mg/dL), which decreased to a median value of 52 mg/dL (IQR 40 to 60 mg/dL). Higher baseline blood glucose was significantly associated with greater blood glucose reduction (coefficient −0.36, 95% CI −0.55 to −0.18, p<0.001).

Conclusions The extent of blood glucose reduction is variable and hypoglycaemia is common. The high incidence of hypoglycaemia highlights the importance of frequent blood glucose monitoring.