I was recently called to evaluate an obese patient with profound metabolic acidosis who was admitted to the hospital with change in mental status and poor oral intake for several days. Her admission labs revealed severe hyperglycemia (glucose 776 mg/dL) and normal anion gap hyperchloremic metabolic acidosis (serum pH of 7.11 and serum bicarbonate of 7 mmol/L). She was initially suspected as having diabetic ketoacidosis due to miscalculation of the anion gap by the admitting team who used the corrected serum sodium and not the actual sodium, and as expected, her acidosis did not resolve despite the correction of hyperglycemia with intravenous fluids and insulin.
Her urine pH was 6.2, urine anion gap was 12 (no ketonuria) and fractional excretion of bicarbonate was 17% suggesting the possibility of mixed renal tubular acidosis. The case appeared like a puzzle until close questioning with the family revealed that her primary care physician had started the patient on topiramate (200mg/d) few weeks back for obesity. Interestingly, there has been a recent increase in literature about the risk of hyperchloremic metabolic acidosis in adult patients on topiramate therapy.
Topiramate is primarily approved for use as an anticonvulsant in patients with partial and generalized seizures and for migraine prophylaxis but its use as an antiobesity drug is among the several “off label" indications. In clinical trials , up to 32% of patients receiving topiramate 400mg/d in divided doses had evidence of normal anion gap metabolic acidosis and up to 7% had moderate to severe metabolic acidosis (less than 17 but greater than 5 mmol/L).
So, how does topiramate cause hyperchloremic normal anion gap metabolic acidosis ? Carbonic anhydrase (CA) catalyzes the conversion of CO2 to HCO3- and H+ ion in the proximal tubule and in the type A intercalated cell of the cortical collecting duct. CAII is the most predominant isoform of CA present in the kidneys.Topimarate has been shown to inhibit the CAII activity thereby causing increased excretion of filtered bicarbonate leading to proximal renal tubular acidosis and impairment of the distal acidification leading to distal renal tubular acidosis. Patients with topimarate induced metabolic acidosis therefore tend to have an alkaline urine, positive urine anion gap, low urinary citrate levels indicating distal renal tubular impairment and increased fractional excretion of bicarbonate with elevated B2 microglobulinuria suggesting proximal tubular impairment.
Besides metabolic acidosis, topimarate use has been associated with 10 fold increased risk of nephrolithiasis and is largely due to hypocitraturia that develops in these patients with failure of renal acidification. These patients are also at risk for osteoporosis for the same reasons. Non ambulatory patients, ketogenic diets, hypovolemia and higher dosage (400mg/d) is associated with an increased risk of renal complications as outlined above. Although measurement of topiramate levels (3-25mcg/ml) is useful in diagnosing toxicity, one third of patients with therapeutic levels have evidence of hyperchloremic metabolic acidosis.
Topiramate induced metabolic acidosis usually improves with discontinuation of the drug. Baseline and periodic laboratory monitoring is needed to evaluate the patients for the development or worsening of metabolic acidosis. Prophylactic alkali therapy seems reasonable but has not been systematically studied.
Coming back to our patient, it became quite apparent that hypovolemia due to poor oral intake and osmotic diuresis probably contributed to the severity of the hyperchloremic metabolic acidosis which resolved 5 days after discontinuation of the drug and bicarbonate supplementation. Her altered sensorium also cleared. The topiramate level was not sent in a timely manner to be of any clinical utility.