Effect of fasting ketosis and type 1 diabetes on alcohol breath analysis technology used in Australia

Abstract

Breath-alcohol analysis devices (breathalysers) provide an approximated blood alcohol concentration (aBAC) from a breath sample. Both acetone consumption and acetone vapour have been demonstrated to produce false-positive aBAC measurements in the absence of ethanol. There is concern that individuals predisposed to elevated breath-acetone (BrAce) concentrations, such as those with type 1 diabetes, may produce falsely elevated aBAC results. No previous investigations have evaluated the impact of ketone production on aBAC readings secondary to fasting in those with type 1 diabetes. This formed the study’s aim. The breathalysers tested were the Alcolizer LE5 (AL5) and Lion Intoxilyzer 8000 (LI8), which are used throughout Australia. The study was approved by St Vincent’s Hospital Human Research Ethics Committee (2019/ETH08728). Twenty participants were recruited into two equal groups: type 1 diabetes and controls. Written informed consent was obtained from each participant before inclusion. Participants were required to fast overnight and abstain from alcohol for 24 h prior to their visit. Participants with type 1 diabetes were advised to take their usual basal insulin dose but withhold their short-acting insulin on the morning of their visit. At the visit, blood ethanol, 3-βhydroxybutyrate (BHB) and glucose concentrations were measured. Each participant then performed three breathtests, a passive and standard AL5 test and evidential LI8 test (devices supplied by NSW Police). Groups were well matched for age, sex and body mass index (BMI; Table 1). The mean HbA1c of the type 1 diabetes group was 7.3% (56 mmol/mol). All blood ethanol measurements were below the lowest detectable threshold (<0.03%). The type 1 diabetes group had a nonstatistically significant fourfold higher mean blood BHB concentration than the controls (using Mann–Whitney tests). No breath-test produced a positive reading. This is the first study to investigate the impact of fasting ketone body production on aBAC readings in those with type 1 diabetes, providing a deeper understanding of this growing body of literature. Despite the overnight fast and secondary ketone production, no breath-test produced a false-positive reading. This may be due to the low levels of ketosis producing low BrAce concentrations. It is unlikely that our participants experienced similar BrAce levels to previous studies where high concentrations of acetone vapours were utilised or where subjects directly ingested acetone. A limitation was the mean HbA1c of 7.3% (56 mmol/mol), as poorer glycaemic control is associated with a higher risk of ketosis and ketoacidosis. As such, future studies should investigate the occurrence of false-positive aBAC readings in states of higher ketone concentrations (e.g. carbohydraterestriction diets). Another direction could be to evaluate the impact of sodium glucose co-transporter 2 inhibitors use, given their increased risk of ketoacidosis and increasing popularity. Our results suggest that false-positive aBAC readings secondary to fasting ketosis in those with and without type 1 diabetes are unlikely. This should help alleviate concerns in those with type 1 diabetes about producing false-positive aBAC readings secondary to fasting ketosis in Australia.