Poster Presentation Australian Diabetes Society and the Australian Diabetes Educators Association Annual Scientific Meeting 2017

Historically, ketoacidosis has been thought to be a manifestation of type 1 diabetes. However, in 1987, Winter and colleagues reported a case series of ketoacidosis in African-American patients, who had phenotypic characteristics of type 2 diabetes, negative antibodies, and who subsequently ceased insulin therapy.¹ Since that initial report, ketosis-prone diabetes has become a well-known clinical entity, occurring in various ethnic groups. However, identifying who will require long-term insulin and who may be safely managed on diet or oral hypoglycaemic agents may not be obvious.   Furthermore, other factors such as starvation, critical illness, alcohol and medications must be considered as contributing factors to ketoacidosis.

We report three cases of ketoacidosis in patients with either no prior diagnosis of diabetes, or a prior diagnosis of type 2 diabetes. The first, a case of latent onset autoimmune diabetes in adulthood, the second, induced by a sodium-glucose co-transporter 2 inhibitor, and the third, ketoacidosis contributed to by starvation and critical illness. Ketone body metabolism and mechanisms of ketoacidosis are reviewed, as are classification systems of ketosis prone diabetes, highlighting the importance of autoantibody status and beta cell function in predicting long-term insulin dependence.²

Based upon this, we present a practical algorithm to a patient presenting with ketoacidosis, including initial investigations and management, and long-term management and monitoring (figure 1). This aims to prioritise patient safety, while reducing unnecessary long-term insulin.