Acute Pancreatitis with Liraglutide – Methods

An analysis of MedWatch adverse event reports submitted to the Food and Drug Administration in the first two years following approval

June 5, 2014

PART II: METHODS

Data source

The study utilized data from the FDA’s AERS (now known as FAERS), a post-marketing surveillance database that contains spontaneous adverse event and medication error reports submitted to the FDA for drug and therapeutic biologic products.[14]

We searched the AERS database for all reports that listed either liraglutide or Victoza as the “primary suspect” drug in causing “pancreatitis acute” (the coded term for acute pancreatitis in the Medical Dictionary for Regulatory Activities, or MedDRA). We inserted an asterisk before and after each drug name to ensure that all reports using variations of the drug name were captured in the search. Our search included all reports received by the FDA between February 1, 2010, and December 31, 2011, the most recent data available at the time of data collection. The full MedWatch reports from all cases that matched these criteria were subsequently obtained from the FDA through a Freedom of Information request.

Data extraction

From the MedWatch reports, we extracted the following demographic data: age, weight, gender, and country of origin. We also collected the following data pertinent to the adverse drug reaction: dose of liraglutide at the time of the pancreatitis event, the duration on liraglutide before the onset of the event, whether the adverse event subsided after discontinuation of liraglutide (positive dechallenge), whether the adverse event reappeared after reintroduction of liraglutide (positive rechallenge), the presence of objective evidence to confirm the diagnosis of acute pancreatitis, and the outcome of the event (e.g., hospitalization or death).

Naranjo scale as a measure of causal likelihood

To assess the likelihood that liraglutide caused the acute pancreatitis events described in the MedWatch reports, we used the Naranjo adverse drug reaction probability scale (Table 1). The 10-item Naranjo questionnaire, published in 1981, is a systematic method for determining the likelihood of a causal relationship between a medication and an adverse event in a given patient.[15]

The causal criteria in the Naranjo questionnaire include the presence of a plausible temporal relationship (adverse event occurring after drug administration), a positive dechallenge, a positive rechallenge, plausible alternative causes, and objective evidence to confirm the diagnosis of the adverse event. Points were assessed based on the presence or absence of each of these and other factors, and a final score (range -4 to 13) was generated that indicated the overall likelihood of a causal connection: a score of ≤0 is considered “doubtful,” 1-4 “possible”, 5-8 “probable,” and ≥9 “definite” evidence of causality.[16]

Objective evidence of acute pancreatitis was defined by either an elevation of amylase or lipase to at least three times the upper limit of the normal range (≥3X ULN) or imaging studies confirming the diagnosis. A patient was considered to have an alternative cause for acute pancreatitis if any of the following were reported: active gallstone disease, alcohol abuse, triglyceride levels >500 mg/dL,[17] concomitant treatment with a medication known to be associated with acute pancreatitis,[18] or a prior history of either chronic pancreatitis or idiopathic acute pancreatitis.

Information provided within each MedWatch report was independently analyzed and rated by two authors (EB and SA) based on the questions and scoring methodology in the Naranjo scale; both reviewers were blinded to each other’s initial rating. Disagreements between the two researchers (in 18% of total cases) were resolved through consensus among four authors (EB, SA, MC, and SW).

Next Page » Part III: Results


[14] FDA Adverse Event Reporting System (FAERS; formerly AERS). FDA/CDER, Silver Spring. http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Surveillance/AdverseDrugEffects/default.htm. Accessed April 25, 2014.

[15] Naranjo CA, Busto U, Sellers EM, et al. A method for estimating the probability of adverse drug reactions. Clin Pharmacol Ther 1981, 30: 239-45.

[16] Ibid.

[17] Murphy MJ, Sheng X, MacDonald TM, Wei L. Hypertriglyceridemia and acute pancreatitis. JAMA Intern Med 2013, 173: 162-4.

[18] We considered two categories of drugs, which were taken concomitantly with liraglutide at the time of the reported event, as alternative drug causes for pancreatitis: 1) any incretin mimetic other than liraglutide, due to the extensively documented association between these drugs and pancreatitis, and 2) any drug with at least one published case of positive rechallenge of acute pancreatitis. A list of such drugs was obtained from the following two sources: a) Drugs classified as a “definite” cause of pancreatitis in Table 2 from: Nitsche C, Maertin S, Scheiber J, et al. Drug-induced pancreatitis. Curr Gastroenterol Rep 2012, 14: 131-8; and b) Class IA and IB drugs in Table 2 from: Badalov N, Baradarian R, Iswara K, et al. Drug-induced acute pancreatitis: an evidence-based review. Clin Gastroenterol Hepatol 2007, 5: 648-61; quiz 644.