CHICAGO –
After 6 months of methimazole use, the odds ratio for acute pancreatitis was 2.02, with a nonsignificant risk elevation for propylthiouracil use after a similar duration, Laszlo Hegedüs, MD, reported at the annual meeting of the American Thyroid Association.
“Ongoing methimazole, but not propylthiouracil, use is associated with an increased risk of acute pancreatitis,” he said.
Dr. Hegedüs, professor of endocrinology and metabolism at the University of Southern Denmark, Odense, said that the European Medicines Agency has noted a few postmarketing reports of acute pancreatitis in patients who received the antithyroid drug methimazole, as well as its prodrug, carbimazole. The agency has accordingly contraindicated antithyroid drug use for patients who previously experienced acute pancreatitis after receiving this drug, advising that methimazole should be “discontinued immediately” should a patient develop acute pancreatitis.
However, investigation of the antithyroid drug–pancreatitis association had been limited to aggregating those case reports, so Dr. Hegedüs and colleagues decided to use Danish medical record and registry data to investigate the association in a nationwide, controlled study that looked at both duration of therapy and total antithyroid drug use.
During the period from 1995-2018, a total of 118,649 patients who used antithyroid drugs were found in the 5.5 million individuals in the Statistics Denmark registry. Dr. Hegedüs and his colleagues also pulled in patient registry and national prescription registry data, as well as civil vital statistics data.
Of those who used antithyroid drugs, 103,825 patients used methimazole, and 14,824 used propylthiouracil. The researchers found 43,580 instances of hospitalization for first-time acute pancreatitis in the pooled antithyroid drug data. Of those, however, just 226 (0.5%) occurred in patients using methimazole, and 19 (0.04%) in those using propylthiouracil at the time of pancreatitis onset.
To ascertain the risk of acute pancreatitis in patients using antithyroid drugs for various durations, Dr. Hegedüs and his colleagues used a case-crossover study design. In the case-crossover technique, patients served as their own controls, because each patient was both exposed and not exposed to antithyroid drugs at some point during the study period. Antithyroid drugs are well suited to this study design, explained Dr. Hegedüs, because they are given for a limited time. A case-crossover design can be used with a small sample size and effectively controls for potentially confounding variables.
The odds ratio for acute pancreatitis in methimazole users after 3 months of exposure was 1.51, with a 95% confidence interval of 1.12-2.02. After 3 months of propylthiouracil exposure, the odds ratio for acute pancreatitis was 1.16 (95% CI 0.46-2.3). At 6 months, the odds ratio of 2.02 for methimazole was similarly statistically significant (95% CI, 1.50-2.78), whereas the odds ratio of 1.40 for propylthiouracil use was not significant (95% CI, 0.58-3.34).
The researchers also wanted to find out whether the cumulative drug dose affected the risk of acute pancreatitis, so they drew from the antithyroid drug population to conduct a case-control study. Here, the investigators matched data from four control patients to each case of acute pancreatitis. The researchers also controlled for sex, age, comorbidities, and prior use of drugs associated with pancreatitis.
Overall, 20% of the 692 methimazole users and their controls were men, as were 16% of the 108 propylthiouracil users, in the case-control study.
Just more than half of patients overall had a total dose exposure of 200 to 1,200 defined daily dose (DDD) – a measure developed by the World Health Organization to denote the assumed average adult dose per day of a medication – with about a quarter of patients receiving a total antithyroid drug dose more than 1,200 DDD and about 20% receiving a dose exposure of less than 200 DDD. The risk of acute pancreatitis did not increase with increased total exposure to antithyroid drugs.
“There is no evidence of a cumulative dose effect of either methimazole or propylthiouracil on the risk of acute pancreatitis,” said Dr. Hegedüs. However, “the warning of the European Medicines Agency seems justified,” he added. “The frequency of acute pancreatitis in acute methimazole users is of a similar magnitude [to that] reported for agranulocytosis,” a known, dire complication of antithyroid drug use. Patients should be advised of the potential complication and informed of signs and symptoms of acute pancreatitis, he said.
Dr. Hegedüs noted that the study had the advantage of using validated epidemiologic methods to look at drug exposure and outcomes at a nationwide scale. However, the registries from which the data were drawn also have limitations. The investigators could not determine the severity of hyperthyroidism, he said, and the relatively rare occurrence of acute pancreatitis meant that there was not sufficient statistical power to look at the subgroup of individuals who had Graves disease and to compare them with those with nodular toxic goiter.
He advised conducting a confirmatory study in an independent cohort, as well as further investigating the yet unknown mechanism of action for the link between the antithyroid drug and acute pancreatitis.
Dr. Hegedüs reported that he had no relevant conflicts of interest and reported no outside sources of funding.
SOURCE: Hegedüs, L. et al. ATA 2019, Short Call Oral Abstract 6 .