Rates of hypoglycaemia do not differ between variable and fixed-rate intravenous infusion protocols when patients are treated for DKA, a review of Australian data has found.
However, the analysis suggested a fixed-rate protocol (FRIII) was associated with higher rates of early hypoglycaemia occurring within 12 hours of treatment, while patients receiving variable-rate protocol (VRIII) spent longer in hospital and on infusion.
The study [link here] focused on 80 patients admitted to The Canberra Hospital, which changed from using a VRIII protocol to a fixed-rate protocol (FRIII) for DKA treatment in late 2021.
The analysis compared the development of hypoglycaemia between 45 patients who received VRIII at the hospital and 35 who received FRIII, as well as tracking the time to clearance of ketones, length of hospital stay and length of time on insulin.
There was no significant difference in development of hypoglycaemia overall, with seven patients (16%) developing it in the variable group and eight (23%) in the fixed-rate group.
By comparison, a 2024 study of patients at Melbourne’s St Vincent’s Hospital [link here] found patients with DKA had a lower risk of hypoglycaemia when treated with a variable-rate infusion protocol.
The authors of the Canberra study noted in Diabetic Medicine [link here] that the rates of hypoglycaemia they identified in the VRIII treatment group (16%) were higher than what was seen in the 2024 study at St Vincent’s (between 6% and 8%) and what has been seen in previous literature.
“All participants on VRIII had ketones <0.6mmol/L at the time of hypoglycaemia, implying that DKA may have resolved,” they said when discussing the difference seen in their study.
“In those with hypoglycaemia, this occurred later in the VRIII cohort, after a median of 18 hours on the insulin infusion, with median time to ketones <0.6mmol/L in this cohort much shorter at 11.5 hours. This raises the question of whether the insulin infusion was still required at the time of hypoglycaemia for management of DKA.”
Optimal infusion regimen still not clear
The rates of hypoglycaemia in the FRIII cohort (23%) were more in line with what has been seen in previous research.
Patients on fixed-rate infusion treatments accounted for 88% of those that developed hypoglycaemia within 12 hours.
Though there was no formal rate reduction-step in the fixed protocol, treating doctors decided to reduce the rate of insulin infusion to <0.1units/kg/h after eight hours, with the change often made due to concerns about imminent hypoglycaemia.
“This insulin dose reduction likely contributed to the lower than anticipated hypoglycaemia rate we found in our study for the FRIII cohort,” the study’s authors said.
This result showed the potential of introducing a dose-reduction step as part of the FRIII protocol, they said.
Patients who received variable-rate infusion stayed in hospital for longer, though this was only statistically significant among patients with T2D.
The authors said longer hospital stays in the VRIII group did not appear related to the time taken to resolve ketosis, because this was similar between the groups.
However, they did note that patients with T2D who received variable-rate infusions were more likely to be prescribed an SGLT2i than those in the fixed-rate group.
“Prolonged ketosis and a tendency towards euglycaemia are features of SGLT2i-associated DKA,” they said.
Limitations of the study included its retrospective nature and small sample size.
Prospective or randomised clinical trials comparing fixed and variable-rate intravenous infusion protocols were needed to determine the optimal approach for the management of DKA, the authors said.