Refractory high anion gap metabolic acidosis due to chronic paracetamol use: a case report

Abstract

Introduction

Metabolic acidosis is a frequent finding in clinical practice, particularly among critically ill patients. While common causes of high anion gap metabolic acidosis, such as diabetic ketoacidosis, renal failure, lactic acidosis, and toxins are easy to diagnose, less typical causes require heightened clinical suspicion. We report an unusual cause of high anion gap metabolic acidosis, which required specific therapeutic measures.

Case presentation

A 45-year-old Sinhalese female with diabetes presented with severe metabolic acidosis and Kussmaul breathing. Initial treatment for sepsis and renal failure, including intravenous bicarbonate and dialysis, failed to correct the acidosis. Further evaluation revealed chronic therapeutic use of paracetamol and elevated urinary 5-oxoproline levels, which confirmed pyroglutamic acidosis. Her acidosis resolved promptly with N-acetylcysteine therapy and suspension of paracetamol.

Discussion

Chronic ingestion of therapeutic doses of paracetamol is an increasingly reported cause of high anion gap metabolic acidosis. This is a distinct entity from acidosis that results from paracetamol toxicity, which is due to liver derangement and lactic acidosis. Paracetamol depletes glutathione and this impacts the gamma glutamyl cycle, causing accumulation of 5-oxyproline/pyruvic acid. Clinicians should be aware of this rare but reversible cause in patients on long-term paracetamol therapy.

Background

Metabolic acidosis is commonly encountered in clinical practice, particularly among critically ill patients [1]. Accurate identification of its underlying cause is essential for effective management. The common causes of metabolic acidosis, particularly high anion gap metabolic acidosis (HAGMA), such as diabetic ketoacidosis, renal failure, lactic acidosis, and toxins, pose little diagnostic challenge to clinicians. We report an unusual cause of HAGMA that required specific therapeutic measures.

Case presentation

A 45-year-old Sinhalese female with a background of diabetes mellitus, hypertension, dyslipidemia, and endometriosis presented with progressive dyspnea. She reported longstanding pelvic pain and menorrhagia and was awaiting elective hysterectomy. In the days prior to admission, her exertional dyspnea worsened to breathlessness at rest. On examination, she was pale and exhibited Kussmaul breathing. The rest of the physical examination was unremarkable. Arterial blood gas (ABG) performed on admission revealed severe metabolic acidosis with normal lactate level and anion gap of 32 (pH 6.9; partial pressure of carbon dioxide (PCO₂) 8.3 mmHg; partial pressure of oxygen (PO₂) 181 mmHg; bicarbonate (HCO₃) 1.7 mmol/L; and lactate 0.6). Capillary blood sugar was 226 mg/dL, but urinary ketones were repeatedly negative. Her C-reactive protein (CRP) was 200 mg/dL, hemoglobin (Hb) 7 g/dL, white blood cell (WBC) count 22 × 10⁹/L, and serum creatinine 1.97 mg/dL. A working diagnosis of sepsis with acute kidney injury was made. Despite adequate hydration, empirical antibiotics and multiple intravenous sodium bicarbonate infusions, she remained profoundly acidotic and developed respiratory fatigue warranting elective intubation and dialysis. A subsequent contrast enhanced computed tomography (CT) scan revealed features of acute pancreatitis, and subsequently, her serum amylase was found to be 2000 U/L. She was treated for sepsis and pancreatitis and was hemodialyzed three times. Her inflammatory markers, serum creatinine, and amylase improved, and she was extubated. In the following days, she was on high doses of oral bicarbonate replacement but remained profoundly acidotic (Fig. 1), prompting further evaluation for an alternate cause for the acidosis.

Fig. 1.

Serial trends in PCO₂ (mmHg), serum creatinine (mg/dL), bicarbonate (meq/L), and lactate (mmol/L) since admission. HD hemodialysis, NAC N-acetylcysteine

Targeted history revealed daily intake of 2–3 g of paracetamol over 6 months, including during hospitalization, for chronic pelvic pain. She denied exceeding therapeutic doses, and liver function tests remained normal throughout. Suspecting pyroglutamic acidosis, paracetamol was discontinued, and a urinary organic acid assay was sent. She was also started on intravenous N-acetylcysteine (NAC) infusion. Over the next 48 hours, her acidosis dramatically improved with normalization of HCO₃ levels. The urine 5-oxoproline level, also known as pyruvic acid, was noted to be very high at 544.7% (normal < 10%), confirming the diagnosis. She was discharged home without further bicarbonate therapy and remained well at the 3-week follow-up.

Discussion and conclusion

HAGMA is due to accumulation of strong acid ions in the extracellular space. Commonly encountered causes include renal failure, diabetic ketoacidosis, and ingestion of toxic alcohols and glycols. This patient presented with persistent HAGMA despite correction of renal failure and sepsis. She had no risk factors to develop D-lactic acidosis, and the absence of neurological manifestations made this unlikely. As ketone bodies were repeatedly negative, diabetic ketoacidosis was ruled out as well. On reviewing literature, it became apparent that chronic paracetamol ingestion in nontoxic doses was being increasingly reported as a cause for HAGMA [2]. This is a separate entity from the acidosis that results from paracetamol toxicity, which is due to liver derangement and lactic acidosis [3]. In 1990, Pitt was the first to describe how chronic paracetamol ingestion leads to accumulation of pyroglutamic acid/5-oxyproline. Paracetamol causes depletion of glutathione through its metabolite N-acetyl-p-benzoquinone imine [4]. Glutathione depletion impacts the gamma glutamyl cycle in the liver by increasing the activity of γ-glutamyl cysteine synthetase and accumulation of γ-glutamyl cysteine, which is converted by γ-glutamyl cyclotransferase to pyroglutamic acid/5-oxyproline [3, 4] (Fig. 2).

Fig. 2.

Effect of chronic paracetamol consumption on the gamma glutamyl cycle, leading to accumulation of pyruvic acid

Compared with similar published cases, our patient had an exceptionally low pH (6.9) and bicarbonate level (1.7 mmol/L) and is among the most severe [1, 5, 6] Table 1 presents a summary of some of the cases published to date.

Table 1.

Clinical and biochemical characteristics of reported cases of 5-oxoproline (pyroglutamic) acidosis associated with chronic paracetamol use

Reference	Acetaminophen dose (g/day)	pH (7.31–7.41)	pCO2 (mmHg, 41–51)	HCO3 (meq/L, 22–29)	Anion gap (meq/L, 6–12)	Serum creatinine (mg/dL, 0.74–1.35)	Glucose (mg/dL, 70–99)	Lactate (mmol/L, 0.5–2.2)	5-oxoproline (mmol/mol Cr, < 70)	Urine ketones
Trevor-Jones E. et al	4	7.11	19	9.1	22	3.5	146	1.2	700	N/A
Case 1, Hunter et al	4	N/A	N/A	9	20	2.1	85	0.7	+  +  +	+  +
Case 2, Hunter et al	1	7.51	16	15	27	1.66	91	1.5	+  +  +	N/A
Case 3, Hunter et al	1	N/A	N/A	14	16	0.86	N/A	0.9	+  +  +	Negative
van den Bersselaar L.R. et al	3	7.16	53	18	18	1.79	N/A	1.1	1721	N/A
Hundemer G.L. et al	1.3	7.29	27	12	28	0.9	N/A	0.8	17,455	Negative
Case 1, Kortmann W. et al	4	7.12	11	3.5	30	N/A	N/A	0.7	16,623	Negative
Case 2, Kortmann W. et al	3	7.29	23	10	29	2.07	N/A	1.6	1050	N/A
Howie S. et al	3	7.26	9.5	4.2	25	0.89	49	1.6	N/A	Negative
Fenves A.Z. et al	2.7	7.44	14	8	35	2.2	159	4.3	2470	Negative
Our patient	2–3	6.9	8.3	1.7	32	1.97	226	0.6	500	Negative

References in order: [1, 5–10]

Although pyroglutamic acidosis due to chronic paracetamol ingestion is an underrecognized entity, emerging case reports suggest that it may be more common than previously thought. Studies estimate that 5-oxoproline acidosis may account for approximately 5% of unexplained high anion gap metabolic acidosis cases in tertiary care hospitals [9]. Despite this, true prevalence remains unknown, as routine urine 5-oxoproline testing is not widely performed. Sex differences in glutathione transferase activity, and lower glutathione stores in women make women more susceptible to pyroglutamic acidemia [1]. On review of 22 cases, 82% occurred in women [2–4]. Other recognized causes of glutathione depletion include malnutrition and sepsis [11], which make patients more susceptible to developing acidosis. Our patient was not clinically malnourished, with a body mass index (BMI) of 23.4 kg/m², and was not hypoalbuminemic. Management involves discontinuing paracetamol, and some reports have shown benefit in using NAC [7]. NAC directly replenishes intracellular glutathione, restoring feedback inhibition on the gamma-glutamyl cycle and reducing 5-oxoproline accumulation [6]. Unlike dialysis or sodium bicarbonate therapy, which merely address acidemia, NAC resolves the underlying metabolic disturbance. Case series have shown that NAC therapy leads to rapid correction of acidosis, even in the absence of detectable serum paracetamol levels, reinforcing its role beyond hepatotoxicity [9].

This case underscores the importance of a systematic approach to refractory HAGMA. Clinicians should consider pyroglutamic acidosis in patients on long-term paracetamol therapy presenting with unexplained high anion gap metabolic acidosis, as early recognition and targeted therapy can result in rapid clinical improvement.

List of abbreviations

ABG

Arterial blood gas

PCO₂

Partial pressure of carbon dioxide

PO₂

Partial pressure of oxygen

HCO₃

Bicarbonate

Hb

Hemoglobin

WBC

White blood cell

CRP

C-reactive protein

CT scan

Computed tomography scan

NAC

N-acetylcysteine

HAGMA

High anion gap metabolic acidosis

AST

Aspartate aminotransferase

ALT

Alanine aminotransferase

ALP

Alkaline phosphatase

IV

Intravenous

NaHCO₃

Sodium bicarbonate

mg/dL

Milligrams per deciliter

mmHg

Millimeters of mercury

U/L

Units per liter

5-OP

5-Oxoproline (also known as pyroglutamic acid)

g/L

Grams per liter

Author contributions

DL, DP, JM, DG, and PJ were involved in the clinical management of the patient. DL, DG, AB, and PJ were involved in the drafting of the manuscript. Figure 2 was developed by the authors. All authors approved the final submission.

Funding

This study is self-funded.

Data availability

All data generated or analyzed during this study are included in this published article (and its supplementary information files).

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent is available for review by the Editor-in-Chief of this journal.

Competing interests

The authors declare that they have no competing interests.