Two Toronto researchers have expressed their doubts about the efficacy of Diclegis, the only drug approved to treat morning sickness in both Canada and the United States. In a new PLOS One paper, University of Toronto physicians Rujun Zhang and Navindra Persaud say that an important study, conducted in the 1970s but never published until now, is marred by serious flaws and should no longer be used to substantiate the effectiveness of pyridoxine and doxylamine for the management of nausea and vomiting in pregnancy.
Canadian Researchers Question Diclegis Study Results
Currently the only FDA-approved medication for the treatment of morning sickness, Diclegis has a long and complicated history. The pill relies on pyridoxine, a form of vitamin B6, and doxylamine, an antihistamine, to control the nausea and vomiting experienced by a majority of pregnant women. Frequently-prescribed during pregnancy throughout the 1950s and 60s, an early combination of pyridoxine and doxylamine, marketed as Bendectin, was actually pulled from the market in the 1970s over safety concerns. The medication, now called Diclegis, was only reapproved for use during pregnancy in 2013, after researchers were able to confirm that neither active ingredient is linked to an increased risk for birth defects.
In their new study, Zhang and Persaud do not question the safety of Diclegis. Instead, the physicians return to the results of a study conducted in the 1970s, soon after Congress amended the US Food, Drug & Cosmetic Act, instructing the FDA to reassess the efficacy of all multi-ingredient medications. Clinical trials were initiated across the country, including one study that looked at the effects of Bendectin in 2,308 pregnant women. Pyridoxine and doxylamine proved effective treatments for morning sickness. A third chemical, dicyclomine, which had previously been included in Bendectin, showed little effect on nausea or vomiting and was soon dropped from the pill’s formulation.
This study, known as the “8-way” Bendectin Study, has been cited time and again in support of the continued use of pyridoxine and doxylamine. As Zhang and Persaud note in their new paper, the study’s results were even reviewed during the FDA’s recent approval of Diclegis. However, the “8-Way” Bendectin Study was never published and its data never underwent the scrutiny of peer review, NPR reports. In fact, Persaud and Zhang were forced to conduct a lengthy investigation, akin to forensics, to obtain the original study’s results. After securing the “8-way” Bendectin Study’s data records through freedom of information requests, the two physicians reconstructed the trial’s design methods and results.
“High Risk Of Bias” Undermines Early Bendectin Trial
What they found was not reassuring. Despite lasting only seven days, 31% of the clinical trial’s participants were lost to follow-up. “If data for 31 percent of the participants is missing,” Persaud told CBC, “that’s a very big problem.” In the study’s placebo group, nearly 40% of patients were lacking recorded outcome information. Beyond a sheer lack of usable data, Persaud and Zhang found insufficient explanation of how the original study’s investigators rated patient’s nausea and vomiting. Moreover, at least one of the original trial’s investigators was ultimately excluded from the study for recording data in the “absence of patient visits.”
These problems, Zhang and Persaud suggest, may explain why the study was never published. In conclusion, the physicians wrote that data from the “8-way” Bendectin Study “should not be used to support the efficacy of doxylamine, pyridoxine or dicyclomine for the treatment of nausea and vomiting during pregnancy because of the high risk of bias.”
Pressed for comment, the FDA told NPR that it is currently reviewing the findings published by Persaud and Zhang. It should be noted, however, that the federal agency’s more recent approval of Diclegis was not based substantially on the formerly-unpublished “8-way” Bendectin Study. While the FDA used the study’s results as supporting evidence, the ultimate authorization was granted based on newer research, which found that doxylamine and pyridoxine had a slight, but not negligible, effect on nausea and vomiting.