A post-mortem audit of the prevalence of ketoacidosis in diabetes and alcohol dependency

Abstract

Introduction

Ketoacidosis is a form of metabolic acidosis caused by the excess production of ketone bodies. It can be commonly found in the context of both diabetes mellitus and alcohol dependency but is also associated with a myriad other aetiologies. To date the extent of risk of developing ketoacidosis is unclear in diabetic persons who are alcohol dependent and/or have abused alcohol acutely as in ‘binge drinking’.

Methodology

We performed a retrospective analysis of consecutive coroners' autopsies between 2015 and 2023 in a London Coroner's jurisdiction of individuals aged 18 years and above to assess whether significant ketoacidosis, when present, was of diabetic or alcoholic aetiology or a combination of both. Data was obtained from the medical history, circumstances of death, histology, toxicology and biochemical markers. The deceased individuals were surveyed and categorised into the following three groups: 1. Individuals with diabetes mellitus type 1or 2 and with no known history of alcohol dependency according to their medical history; 2. Alcohol dependent persons or persons abusing alcohol acutely, without diabetes mellitus; 3. Diabetic individuals with known alcohol dependency or acute abuse as in ‘binge drinking’. Individuals suffering from other causes of ketoacidosis were excluded.

Results

From an overall total of 3873 autopsies performed of persons 18 years or above over a 9-year period (2015-2023), 1021 cases were analysed and divided into three groups as stated above. Group 1 consisted of 635 diabetic individuals, with ketoacidosis present in 23 (3.6%). Group 2 consisted of 333 individuals with alcohol dependency, in which ketoacidosis was present in 17 (5.1%). Group 3 consisted of 56 individuals, with both diabetes and alcohol dependency, in which ketoacidosis was present in 7 (13%).

Conclusions

This study demonstrated a statistically significant risk of developing ketoacidosis in all individuals with both diabetes mellitus and alcohol dependency (Group 3) when compared to diabetics alone (Group 1): Risk ratio = 3.8; p = .0001). There was less but significant increased risk when females alone were compared in these two groups. Less but significant increased risk was also found in all Group 3 individuals when compared to those in the alcohol alone category (Group 2) or when all individuals or males alone were compared between Group 1 + Group 2 with those in (Group 3). There was no significant increased risk when diabetics in Group 1 were compared to persons with alcohol dependency in Group 2.