Ketoacidosis – three unusual and informative case study reports

Blood gas analysis involves determination of three parameters (pH, bicarbonate and partial pressure of carbon dioxide, pCO₂) necessary for assessment of patient acid-base status and identification of those whose acid-base is disturbed. There are four classes of acid-base disturbance: metabolic acidosis, metabolic alkalosis, respiratory acidosis and respiratory alkalosis.

Of these, metabolic acidosis, characterized by initial reduction in bicarbonate and pH along with compensatory decrease in pCO₂, is the most common. Abnormal accumulation of keto acids is one of many causes of metabolic acidosis. This condition, called ketoacidosis, is the focus of two recently published case study reports. 

The first describes two cases of diabetic ketoacidosis (DKA), the most common presentation of ketoacidosis. This is an acute and potentially life-threatening complication of diabetes (principally type 1 diabetes but also, less commonly, type 2 diabetes) that results from absolute or relative deficiency of insulin. Intercurrent illness (e.g. infection) can precipitate DKA and it can obviously also occur if prescribed exogenous insulin is not administered.

Biochemically, DKA is characterized by three findings: high anion gap metabolic acidosis (due to accumulation of keto acids that overwhelm normal blood buffering mechanisms); increased ketones (acetoacetate, β-hydroxybutyrate and acetone) in blood and urine; and increase, typically severe increase, in blood glucose concentration (hyperglycemia).

These two cases of DKA are notable because in both cases blood glucose at presentation was within normal limits; the patients were not hyperglycemic as is almost invariably the case in DKA, but euglycemic. 

The first case concerns a 21-year-old female who had been diagnosed with type 1 diabetes 5 years previously. Her regular insulin treatment was administered by an insulin pump. Two days after failure of her insulin pump this young lady presented to hospital complaining of weakness and inability to eat. She was ”moderately” dehydrated but showed no signs or symptoms suggestive of infection.

Admission blood gas analysis revealed partially compensated metabolic acidosis (reduced pH: 7.1, reduced bicarbonate: 6 mmol/L, reduced pCO₂: 14 mmHg (1.9 kPa). Electrolyte results (Na⁺, K⁺, Cl^- and HCO₃^-) indicated a raised anion gap (22 mmol/L, reference range 1-10 mmol/L). She also had increased ketones in blood and urine. All these results were strongly suggestive of DKA.

However, her blood glucose was unexpectedly well within normal limits (74 mg/dL or 4.1 mmol/L). Despite intensive investigation no evidence was found to suggest that this lady was suffering from any non-diabetic condition that can give rise to metabolic (keto) acidosis, and a diagnosis of euglycemic diabetic ketoacidosis (EDKA) was therefore eventually made. 

The second case reported in this paper concerns a 25-year-old female who had been diagnosed with type 1 diabetes at the age of 15 years. Her regular treatment involved daily insulin injections before meals and bedtime. This current admission to hospital was occasioned by recent onset of signs and symptoms of urinary tract infection (burning pain during urination, along with high-grade fever). She also reported nausea and consequent reduction in food intake. Physical examination revealed signs of dehydration. Urine testing confirmed urine tract infection. Arterial blood gas analysis revealed marked reductions in pH (7.03), bicarbonate (6.7 mmol/L) and pCO₂ (13 mmHg, 1.7 kPa).

These results along with the found high anion gap (26 mmol/L) confirmed high anion gap metabolic acidosis. This finding along with abnormally high concentration of ketones in blood and urine and history of type 1 diabetes suggested DKA. However, blood glucose was unexpectedly normal (97 mg/dL, 5.4 mmol/L). Non-diabetic causes of metabolic (keto) acidosis were excluded and a diagnosis of euglycemic diabetic ketoacidosis (EDKA) was eventually made. 

In discussion of these two case stories the authors describe how insulin deficiency can lead to accumulation of keto acids and consequent high anion gap metabolic (keto) acidosis. They explain the distinction between DKA and EDKA. They also discuss how a diagnosis of EDKA can be only made by exclusion of non-diabetic conditions that can also give rise to ketoacidosis in association with normal blood glucose. 

A separate case study report focuses on ketoacidosis occurring in the non-diabetic patient. This case concerns a 27-year-old lady who 8 weeks after giving birth to a healthy infant presented with a 4-day history of nausea and vomiting and other non-specific symptoms (general malaise, body aching and fatigue). Her poorly condition, assumed to be due to gastroenteritis, had left her unable to tolerate food and she had not eaten for 4 days. Pregnancy and birth had been uneventful and the patient had been exclusively breastfeeding her baby since birth. 

Admission venous blood gases revealed a high anion gap metabolic acidosis (pH 7.02, bicarbonate 5 mmol/L, and anion gap 37 mmol/L). The high anion gap was found on testing to be due to increased ketones in blood and urine, confirming the diagnosis of ketoacidosis. Absence of a history of diabetes and finding of normal blood glucose (3.6 mmol/L) excluded a diagnosis of diabetic ketoacidosis. The cause of ketoacidosis in this case was attributed to the combined effect of lactation, gastroenteritis and reduced food (carbohydrate) intake. 

In discussion of this case history, the authors reflect that ketoacidosis in lactating women is a rare but recognized occurrence; they cite nine previous reports. In most cases, including the one reported here, inadequate carbohydrate diet (starvation) is thought to be a significant factor in lactation-induced ketoacidosis. The report includes extended discussion of why lactating women may be at greater than normal risk of ketoacidosis, and the ketogenic effect of a low carbohydrate diet.