This tutorial looks at the twin problems of Alcohol related and Starvation Ketoacidosis. These diagnoses are frequently missed by clinicians because 1. they attribute the metabolic acidosis to another cause e.g. lactate or acute kidney injury or 2. they do not routinely measure blood ketones. It is my view that, in any patient presenting with a plasma bicarbonate below 20mmol or mEq/L or a base deficit of -5 or greater, it is mandatory to measure blood ketones (beta-hydroxybutyrate).

I present two cases, the first is a patient who is admitted with abdominal pain and a likely upper GI bleed, with a history of an eating disorder, who has metabolic acidosis. The second patient is an alcoholicwho recently stopped both eating and consuming alcohol. She also has a metabolic acidosis. I discuss the biochemistry of alcohol metabolism and explain why alcoholics are at risk for ketoacidosis. I also explain why this is part of a paradigm of metabolic failure that, without significant attention to detail, may result in therapy that precipitates a variety of withdrawal syndromes: these include acute Wernicke’s Encephalopathy, Alcohol Withdrawal Syndrome, and acute aquaresis and Osmotic Demyelination. Alcoholic ketoacidosis almost always follows cessation of alcohol intake – and one is unlikely to make this diagnosis in a patient who, for example, presents drunk to the ED (this results in a host of other metabolic anomalies, for example hypoglycemia despite high plasma lactate).

Starvation ketoacidosis is seen in patients who are chronically malnourished or fasted for prolonged periods for surgery, in whom the pancreatic Islet cells have either atrophied or are hibernating. Careful attention must be applied to feeding and refeeding: it is imperative that the patient does not lose further lean body mass. On the other hand refeeding syndrome may result in rhabdomyolysis and death. I guarantee you’ll learn something.