Acetaminophen induced high anion gap metabolic acidosis: a potentially under-recognized consequence from a common medication

Abstract

While metabolic acidosis is one of the most common complications in patients with chronic kidney disease (CKD), there are several uncommon etiologies that are challenging to diagnose. Here, we describe a patient on peritoneal dialysis who developed high anion gap metabolic acidosis secondary to acquired 5-oxoprolinemia from acetaminophen use. While CKD is a known risk factor for developing this potentially serious complication, this case further highlights how 5-oxoproline accumulation can occur, even with therapeutic dosing of acetaminophen.

Introduction

Metabolic acidosis is commonly encountered in hospitalized patients and occurs via three mechanisms: increased acid generation, loss of bicarbonate, or lack of acid excretion by the kidneys. This condition is characterized as either an elevated anion gap (AG) or non-AG acidosis, which is calculated as the difference between the major plasma cation, sodium, and the major plasma anions, chloride and bicarbonate, calculated as follows: AG = Na⁺ – [Cl⁻ + HCO₃⁻] [1]. Non-AG causes of metabolic acidosis are due to increased bicarbonate losses in the stool (i.e., diarrheal illnesses), excessive chloride administration in intravenous (IV) fluids, or more rarely, renal tubular acidosis. These etiologies are distinguished from high AG metabolic acidosis, when unmeasured amount of anions in the blood accumulate. This state of excess acid results in consumption of bicarbonate without a relative change in sodium or chloride levels, yielding an anion gap greater than 15 mEq/L. While most pediatric cases of high AG primary metabolic acidosis are due to etiologies such as lactic acidosis, diabetic ketoacidosis, ethylene glycol ingestion, or inborn errors of metabolism, other less common etiologies should also be considered. Here, we describe a case of a high AG, primary metabolic acidosis due to acetaminophen use in a pediatric patient with chronic kidney disease (CKD) stage VD (dialysis).

Case presentation

A 12-year-old girl on peritoneal dialysis (PD) presented with a 1-week history of intermittent fevers, abdominal pain, and feeding intolerance. Prior to presentation, her home PD had been reduced due to low blood pressure, and she had received 3 days of amoxicillin–clavulanic acid for a recently diagnosed pneumonia. Her complex past history included twin–twin transfusion syndrome with subsequent hypoxicischemic injury, encephalopathy, global developmental delay, and feeding tube dependence.

In the emergency department, she was afebrile with an exam remarkable for dry mucous membranes, tachycardia, tachypnea, and a distended abdomen. Initial labs revealed an elevated white blood cell count of 18.6 thousand/mm³, hemoglobin of 9.1 g/dL, platelets of 526 thousand/mm³, sodium of 144 mEq/L, potassium 3.4 mEq/L, chloride 99 mEq/L, bicarbonate 16 mEq/L, anion gap of 29 mEq/L, BUN 75 mg/dL, and creatinine of 11.2 mg/dL. Peritoneal fluid analysis revealed a nucleated cell count of 1775/mm³ with 70% segmented nucleated cells, 1350/mm³ of red blood cells, and bacterial cultures from the fluid were obtained. Chest X-ray demonstrated a retrocardiac opacity concerning for pneumonia, and she was additionally diagnosed with bacterial peritonitis, though serial PD cultures were negative. During her hospital course, she received total parenteral nutrition with bowel rest, alkalinizing IV fluids, broad-spectrum IV, and intraperitoneal antibiotics including combinations of cefepime, vancomycin, meropenem, metronidazole, and pain control with acetaminophen and hydrocodone. PD was resumed with adequate clearance and a reduction in BUN values ranging from 40 to 50 mg/dL was achieved, consistent with her outpatient levels.

Despite improvements in her overall condition, she had persistent high AG acidosis and tachypnea. On day 8 of her hospitalization, her pH was 7.34, bicarbonate was 12 mEq/L, and anion gap ranged from 20–29 mEq/L despite adequate PD, TPN with maximum acetate, and additional alkalinizing fluids. Additional workup included a normal calculated osmolar gap at 2 mOsms/L and normal/undetected serum levels of beta-hydroxybutyrate, salicylate, ethylene glycol, and methanol. Serial lactate levels were checked and were normal (i.e., less than 2 mmol/L). She had not received IV iron, isoniazid, or propylene glycol-containing medications. Finally, the patient’s medications were reviewed, revealing scheduled enteral acetaminophen administration every 4–6 h with an average dosing of 48.5 mg/kg/day over the 8 days of her hospitalization. This raised suspicion for pyroglutamic acidosis, and acetaminophen was discontinued with 150 mg/kg of IV N-acetylcysteine given once, followed by 10 mg/kg of enteral N-acetylcysteine for three doses 12 h apart. Additionally, the patient’s urine organic acid profile showed “a large excretion of pyroglutamic acid” (as our lab reports only qualitative results), confirming the diagnosis of pyroglutamic acidemia. Within 48 h of stopping acetaminophen and administering N-acetylcysteine, the patient’s metabolic acidosis showed rapid improvement with a bicarbonate value of 22 mEq/L, the anion gap reduced to 15 mEq/L, and the patient’s lactate remained normal at 0.9 mmol/L. The urine organic acid profile was not repeated to check for normalization of pyroglutamic acid given the patient’s clinical response.

Discussion

Pyroglutamic acidosis is a potentially life-threatening metabolic disorder that occurs in states of glutathione depletion due to rare inherited autosomal recessive enzyme deficiencies as well as acquired states [2]. Given that few studies have systematically evaluated the incidence of acquired pyroglutamic acidosis and that acetaminophen is such a widely utilized medication, this case report demonstrates that pyroglutamic acidosis is a potentially under-recognized complication.

As shown in Fig. 1, pyroglutamic acid, also known as 5-oxoproline, is an intermediate metabolite in the γ-glutamyl cycle. When glutathione is depleted, γ-glutamylcysteine synthetase is upregulated and results in increased levels of intermediate products shunted toward production of 5-oxoproline which is normally excreted in the urine. While 5-oxoprolinase degrades 5-oxoproline back into glutamic acid, this requires ATP and becomes a redundant, high-energy process. Thus, patients with reduced glomerular filtration, as well as low ATP states due to poor nutrition, are at risk for clinically significant 5-oxoproline accumulation [2]. Not only large doses of acetaminophen are known to deplete glutathione; prolonged acetaminophen use even at therapeutic levels has also been shown to consume glutathione and cysteine stores generating an environment favoring 5-oxoprolinemia [2]. While this complication has been reported to predominantly occur in females due to hypothesized sex-specific differences in acetaminophen metabolism, this remains poorly understood [3]. Additionally, certain antibiotics, such as flucloxacillin, are proposed to inhibit the enzyme 5-oxoprolinase and increase the risk of pyroglutamic acidemia [4], with other low-glutathione states that may also increase risk including alcoholism, sepsis, and hepatic impairment [2].

Fig. 1.

The γ-glutamyl cycle. 5-oxoproline (i.e., pyroglutamic acid), is an intermediate metabolite of the γ-glutamyl cycle synthesized by γ-glutamyl cyclotransferase. While glutathione normally inhibits γ-glutamylcysteine synthetase, depletion of glutathione upregulates this key enzyme, increasing γ-glutamyl phosphate and γ-glutamyl cysteine levels, which are then shunted to 5-oxoproline production. Thus, clearance of accumulated metabolites requires ATP for degradation of 5-oxoproline back into glutamic acid by 5-oxoprolinase and renal excretion

The patient described in this case report had several risk factors precipitating her acquired pyroglutamic acidosis, including baseline malnutrition due to several days of feeding intolerance, female sex, infection, CKD stage VD, and frequent acetaminophen use. As described in this case, correction of the metabolic derangements primarily involves the removal of the offending agent [2, 4, 5]. Though N-acetyl cysteine has been used, and, mechanistically, should alleviate the metabolic burden by repleting cysteine levels and ultimately glutathione levels, data for its benefit in acetaminophen induced pyroglutamic acidosis is lacking [2]. As described in the case above, cessation of acetaminophen and N-acetylcysteine administration resulted in the resolution of the patient’s otherwise refractory acidosis within 48 h. While recurrent pyroglutamic acidosis due to acetaminophen has been previously suggested [5], given the multi-factorial nature of this acquired complication, reinitiating the use of acetaminophen in a patient similar to as described in this case would require careful risk/benefit consideration and close monitoring.

Summary.

What is new?

• Pyroglutamic acidosis can occur with therapeutic acetaminophen administration in at-risk pediatric patients, resulting in high anion gap metabolic acidosis.

• Patients with kidney dysfunction are at increased risk for pyroglutamic acidosis, requiring a high degree of suspicion for appropriate diagnosis and management.