Adjunctive topiramate enhances the risk of hypothermia associated with valproic acid therapy

Background: Topiramate was approved for the treatment of epilepsy in 1999 and has since been approved for the prevention of migraine headache. It is structurally different from the majority of antiepileptic medications and is pharmacodynamically unique in its ability to inhibit the enzyme carbonic anhydrase. Postmarketing reports of topiramate-associated hypothermia have occurred but this adverse event has not been well characterized. Data mining of an adverse event database was used to assist in the identification of hypothermia. Objective: We sought to explore a possible association between the concomitant use of topiramate and valproic acid and the induction of hypothermia. Methods: This was a pharmacovigilance case series survey of spontaneous hypothermia, a reported adverse event in patients treated with topiramate and valproic acid, alone and in combination. The U.S. Food and Drug Administration's Adverse Events Reporting System (AERS) database was searched for reports of hypothermia in association with the use of topiramate. A data mining algorithm was used on the AERS to identify scores for hypothermia associated with antiepileptic drugs. Results: We identified 22 unduplicated reports of hypothermia in patients exposed to topiramate. Three of the 22 were confounded by patient overdoses with multiple drugs and not considered. Use of more than one antiepileptic drug was reported in most of the remaining 19 reports. Of these 19 reports, valproic acid was mentioned in 7. Two of the 19 reports mentioned topiramate only. Eleven of the 19 patients were men. The median age of the 19 patients was 40 years (range, 31/2- 82 years). Body temperatures ranged from 29.5 degrees C (moderate hypothermia) to 35 degrees C (mild hypothermia) with a median of 34 degrees C. Eleven of 18 reports of hypothermia occurred during the cooler months (one report did not indicate the time of year in which hyopthermia occurred). Comorbid conditions included hypothyroidism in six reports, five in patients who received valproic acid concomitantly with topiramate and five reports of hyperammonemia in similarly treated patients. Data mining scores (empirical Bayes geometric mean) for antiepileptic drugs ranged from a high of 5.845 for phenobarbital to 2.956 for gabapentin. Hypothermia was reported 4.7 times more frequently when topiramate was used than was statistically expected. We have found hypothermia, defined as an unintentional drop in body core temperature to < 35 degrees C, to be associated with concomitant administration of topiramate (a carbonic anhydrase inhibitor) and valproic acid in patients who have tolerated either drug alone. Data mining analysis for topiramate showed a signal of hypothermia. Topiramate was reported 4.72 times more frequently in the database than would be statistically expected when considering all other drugs. Topiramate may act pharmacodynamically to potentiate the effects of valproic acid as a result of its ability to decrease blood HCO3- and increase blood ammonia levels.