A 34-year-old woman presented to the emergency department with a chief complaint of acutely worsening abdominal pain. The patient reported the abdominal pain to be in the left lower quadrant and of several months' duration but noted that it had worsened substantially over the previous 10 days. She also reported diffuse muscle pain and nausea without emesis, no food intake over the previous 72 h, and only limited recent fluid intake consisting of ginger ale, water, Gatorade, and homemade liquor. Her last menstrual period was 2 weeks before admission. Her medical history was remarkable for posttraumatic stress disorder, bipolar disorder, gestational diabetes, a previous gastric bypass surgery, and an ambiguous history of “injury to the liver and pancreas.” She denied taking any medications or other illicit drugs, specifically denied “binge” drinking, and had recently received a tattoo from an unlicensed tattoo artist. She was homeless and resided in a tent city with her husband.
The only remarkable findings of the physical examination were tachycardia and a mildly enlarged, nontender liver. She appeared ill and smelled of a campfire but was otherwise awake, alert, and cooperative. Selected laboratory results are shown in Table 1. Low urea nitrogen and albumin concentrations suggested probable malnutrition. The observed hypocalcemia of 7.4 mg/dL could be attributed only partly to the low albumin concentration, because the albumin-corrected total Ca concentration—which is equal to the measured total [Ca] + 0.8 mg/dL × (4.0 − [Albumin]), where [Ca] is the calcium concentration in milligrams per deciliter and [Albumin] is the albumin concentration in grams per deciliter (1)—was 8.8 mg/dL, less than the institutional lower reference limit.
Opiates and tricyclic antidepressants were identified in the patient's urine. The presence of opiates was not commented on in the medical record, so whether the clinical team was aware of this finding was unknown. The records of the hospital system did not have any prescriptions for tricyclic antidepressants for the patient, so the positive result for this analyte could be due to the use of a tricyclic antidepressant obtained illicitly or with an outside prescription; alternatively, it could be due to a cross-reacting substance.
QUESTIONS TO CONSIDER
What are common causes of a combined high anion gap and increased osmolality?
What can cause ketoacidosis?
What is the appropriate approach to this patient with an increased serum isopropyl alcohol result?.
The attending physician in the emergency department requested assistance in interpreting the highly increased concentration of β-hydroxybutyrate. Furthermore, the clinical team wanted to know if detecting isopropyl alcohol in the alcohol screen implied that the homemade liquor ingested by the patient contained isopropyl alcohol.
The initial arterial blood gas results showed a pattern consistent with metabolic acidosis (pH = 7.28) that had been compensated for in part by hyperventilation, as suggested by the blood gas results. Additionally, the high β-hydroxybutyrate concentration is indicative of ketoacidosis; however, because this patient's plasma was mildly hypoglycemic and because no history of type I diabetes could be ascertained, diabetic ketoacidosis (DKA)2 was considered highly improbable (2). Rather, her medical history strongly suggested the ketoacidosis was due to alcohol consumption in conjunction with little or no nutritional intake, a condition known as alcoholic ketoacidosis (AKA) (3). Although commonly observed, AKA is often underdiagnosed in emergency departments that receive patients who are chronic alcohol abusers. It is estimated that 20% of patients presenting with ketoacidosis have AKA (4). The key factors leading to AKA include (a) starvation with glycogen depletion, (b) an increased intracellular NADH/NAD+ ratio secondary to the metabolism of alcohol, and (c) extracellular fluid volume depletion (3).
The vast majority of AKA cases present with abdominal pain, nausea, and vomiting, all of which cause patients to suddenly stop eating and reduce their liquid consumption. Common physical findings on presentation include tachycardia, tachypnea, and abdominal tenderness.
AKA is characterized in the clinical laboratory by increased serum ketones and a high anion gap. At its foundation is an underlying state of starvation that causes hepatic glycogen depletion, insulin deficiency, and increased counterregulatory hormones (glucagon, cortisol, growth hormone, and catecholamines) (5). This hormonal imbalance leads to enhanced mobilization of free fatty acids from adipose tissue and a hepatic metabolic shift from lipogenesis to lipolysis and increased gluconeogenesis. Most of the free fatty acids that enter the liver are metabolized to so-called ketone bodies: acetoacetate, β-hydroxybutyrate, and acetone. It is worth noting is that “ketone body” is partly a misnomer, because β-hydroxybutyrate, usually the most prevalent “ketone body,” is not a ketone.
AKA is particularly challenging to the patient's physiology because it has features of a positive-feedback loop. For example, patients with AKA typically present with a decreased blood volume due to prolonged emesis and/or ethanol-enhanced diuresis. This volume contraction limits the excretion of ketone bodies and organic acids, as well as increases the concentrations of lipolytic and ketogenic hormones.
Although this patient appeared to have had a high anion gap metabolic acidosis with respiratory alkalosis, other acid–base imbalances can occur in AKA. In one study, 30% of AKA patients had a coexisting metabolic alkalosis due to prolonged emesis (5). AKA can also be associated with other laboratory abnormalities, such as increased serum lactate and an osmolal gap, as well as reduced electrolyte concentrations. Both urea nitrogen and creatinine concentrations are usually increased, as are markers of liver or pancreatic injury (e.g., enzymes, bilirubin). These latter findings are most commonly due to comorbid illnesses, such as alcohol-induced hepatitis and pancreatitis. This patient exhibited all of these laboratory derangements, except for having a very low value for urea nitrogen, which was likely decreased because of the malnutrition.
In the pathophysiology of AKA, oxidative metabolism of ethanol to acetaldehyde and acetate occurs in the liver, causing a greatly increased NADH/NAD+ ratio. This abnormal NADH/NAD+ ratio has important metabolic ramifications, because to regenerate NAD+ requires that the pyruvate produced by gluconeogenesis and other pathways be converted to lactate, or else acetoacetate must be converted to β-hydroxybutyrate. Thus, the altered intracellular redox potential induced by ethanol is critical in explaining why AKA patients exhibit increased lactate and β-hydroxybutyrate concentrations.
Patients diagnosed with AKA should be treated immediately with 50 g/L glucose in normal saline to address the starvation state and lack of glucose. Insulin concentrations will consequently increase, and glucagon and other counterregulatory hormones will decrease. Eventually, this treatment stimulates the oxidation of NADH by reactivating the normal oxidative metabolism of carbohydrates, simultaneously reducing the NADH/NAD+ ratio and halting ketogenesis. Insulin therapy is not recommended unless underlying DKA is present. Thiamine is often given with glucose to ensure adequate cofactor concentrations for the enzymes involved in aerobic carbohydrate metabolism, such as pyruvate dehydrogenase. An added benefit is that thiamine can prevent Wernicke encephalopathy (4). Finally, intravenous hydration is a mainstay of therapy for AKA. Replenishing the extracellular fluid promotes normal renal function, removal of excess acids, and the return to normal bicarbonate concentrations.
The increased osmolal gap observed in this case could not be explained by ethanol alone. In such cases, it is useful to directly test for other commonly ingested, osmotically active small molecules, such as small organic alcohols. Furthermore, a suspicion of a toxic ingestion as an etiology warrants research into the identity of the ingested matter, such as consultation with the local poison center about the contents of commercial products, or even soliciting a sample of the unknown material for analysis. The “homemade liquor” in question was unavailable for analysis in this case, but a gas chromatography analysis of the patient's serum identified acetone (37 mg/dL) and isopropyl alcohol (14 mg/dL). The calculated osmolal gap of 56 mOsm/kg was above the upper limit of the reference interval by approximately 46 mOsm/kg, a difference that could be mostly explained by approximating the additive contributions of ethanol (34 mOsm/kg, using [Ethanol] in mg/dL ÷ 3.8 = 1 mOsm/kg (6)), acetone (6 mOsm/kg, using [Acetone] in mg/dL ÷ 5.8 = 1 mOsm/kg), and isopropyl alcohol (2 mOsm/kg, using [Isopropyl Alcohol] in mg/dL ÷ 6 = 1 mOsm/kg).
Although it is possible that the homemade liquor reportedly consumed by this patient contained isopropyl alcohol, there is another possible explanation for its presence. Previous studies involving acetonemic diabetic patients and acetonemic cows and rats illustrated that a highly reduced intracellular environment could cause some acetone to be biotransformed into isopropyl alcohol and concomitantly regenerate NAD+ (7, 8). None of the individuals described in these studies were exposed to isopropyl alcohol, yet all had evidence of serum isopropyl alcohol. These findings corroborated previous reports of isopropyl alcohol present in the blood of autopsy patients not previously exposed to isopropyl alcohol, as well as in vitro enzymatic studies involving alcohol dehydrogenase (9).
In summary, AKA is an often underdiagnosed condition in the US. Although its pathophysiology can be complex, the condition can be resolved rapidly with therapy, resulting in a low mortality (10). Although the clinical findings of AKA are very similar to DKA, differentiation is based on the patient's history of diabetes or alcoholism and the plasma glucose concentration on admission. Patients with AKA will usually present with normal or low glucose concentrations and a history of substantial alcohol use.
Approximately 1 month after discharge from this hospitalization, the patient was readmitted for multiple subcutaneous abscesses consistent with sequelae of injection drug use. She developed sepsis and pneumonia after surgical drainage of these abscesses, which were complicated by stroke, and died 1 week later.
POINTS TO REMEMBER
AKA occurs in alcoholics who binge drink but have little to no intake of any nonalcohol calories. This leads to glycogen depletion, ketosis, an increased NADH/NAD+ ratio, and severe dehydration.
Treatment of AKA involves simple fluid replacement with fluids containing glucose, electrolytes, and thiamine.
Patients with AKA will present with an increased anion gap in combination with an increased osmolal gap. The crucial differential diagnoses for an increase in both the anion and osmotic gaps are DKA, lactic acidosis, or ingestion of methanol or ethylene glycol.
Although AKA and DKA share a common symptomology, a patient history of substantial alcohol use along with a normal or below-normal plasma glucose concentration will indicate AKA.
↵2 Nonstandard abbreviations:
- diabetic ketoacidosis;
- alcoholic ketoacidosis.
Author Contributions: All authors confirmed they have contributed to the intellectual content of this paper and have met the following 3 requirements: (a) significant contributions to the conception and design, acquisition of data, or analysis and interpretation of data; (b) drafting or revising the article for intellectual content; and (c) final approval of the published article.
Authors' Disclosures or Potential Conflicts of Interest: No authors declared any potential conflicts of interest.
- Received for publication October 22, 2010.
- Accepted for publication January 17, 2011.
- © 2011 The American Association for Clinical Chemistry