Periprocedural ketosis is common in patients with diabetes but appears unrelated to SGLT2 inhibitor use and more related to well established risk factors such as duration of fasting before surgery.
Speaking at the Australasian Diabetes Congress (ADC) 2021, Dr Jason Denny said there had been increased vigilance around periprocedural diabetic ketoacidosis and SGLT2i use since a safety alert in 2015.
This led to a 2020 consensus document between diabetes societies and the Australian and New Zealand College of Anaesthetists.
Despite this, Dr Denny said there was a paucity of evidence in the medical literature outside case reports.
Dr Denny, an anaesthetics registrar from Austin Health, presented the results of an observational study of 1,132 patients with mostly type 2 diabetes undergoing elective and emergency surgery at the hospital in 2018-19.
He said the study found a surprisingly high incidence of ketosis – 5.3% with levels >0.6 mmol/L and 2.0 % >1.0 mmol/L. More than 90% of ketosis was euglycaemic.
However the study found medication type did not associate with elevated ketones.
“Moreover, medication cessation times did not associate with elevated ketones,” he said. “SGLT inhibitors, with a median cessation duration of 62 hours, did not associate.”
Instead, the patient factors associated with ketosis were lower preoperative haemoglobin and prolonged fasting duration to food and fluid. Emergency surgery status and colonoscopy were also highly significant factors in ketosis.
The study found that each hour of fasting duration, each gram per litre decrement in albumin, and emergency and colonoscopy status, were associated with ketosis.
“For each ten hours of fasting for food, there was a three-fold increased incidence of ketosis.”
Emergency surgery status had an OR of 2.46 for ketones >1.0 mmol/L.
The study also found that in terms of complications, elevated ketones >1.0 mmol/L were associated with a significantly higher rate of breaching medical emergency team (MET) call criteria compared to ketones <1.0 mmol/L (36.5% v 6.3%).
However this was not statistically significant on multivariate adjustment, suggesting there may be other causes at play.
“What we found is what we already know – that age, higher comorbidities, low albumin, low haemoglobin associated with complications, not ketosis. These are already well known and established factors that predict complications.”
He concluded that in a low risk surgical cohort of patients with diabetes, hours fasting and nutritional status were perhaps under-recognised factors associated with ketosis and more important than diabetes medications.