Acute Pancreatitis Coincident with Valproate Use: A Critical Review


Address correspondence and reprint requests to Dr. J. M. Pellock at Virginia Commonwealth University Health Systems/Medical College of Virginia Hospitals, P.O. Box 980211, Richmond, VA 23298, U.S.A. E-mail:


Summary:  Purpose: Acute pancreatitis has been associated with a number of medications, including valproate (VPA). Valproate-coincident pancreatitis is uncommon and is usually associated with other risk factors; however, the United States Food and Drug Administration has issued a box warning for risk of acute pancreatitis with VPA products. We reviewed cases of pancreatitis in VPA-treated patients from the clinical database of VPA/divalproex trials and compared the incidence of elevated amylase levels between VPA- and placebo-treated patients.

Methods: Report rates of acute pancreatitis coincident with VPA use in 34 clinical trials were calculated. Incidence rates of amylase elevations above the normal range in three clinical placebo-controlled migraine trials were compared between VPA- and placebo-treated patients.

Results: Among 3,007 VPA-treated patients in 34 clinical trials, two reports of pancreatitis were considered by investigators to be probably related to VPA. Both patients recovered. Similar rates of amylase elevations were observed in VPA- (5.9%) and placebo-treated (6.1%) patients in the three migraine headache trials.

Conclusions: VPA-coincident acute pancreatitis is uncommon and idiosyncratic. Checking amylase levels in the absence of other clinical signs and symptoms provides little value for predicting pancreatitis. Physicians should be guided by clinical symptoms of pancreatitis to identify cases.