Ending an 8-week course of a proton pump inhibitor caused rebound heartburn, acid regurgitation, and dyspepsia in healthy volunteers, Dr. Christina Reimer and her colleagues reported.
Rebound symptoms were clinically significant in the study subjects, causing mild to moderate discomfort for at least 2 weeks after withdrawal of daily PPI therapy—a “remarkable” finding given that 40% of the study subjects had never experienced such symptoms before, wrote Dr. Reimer of the department of medical gastroenterology at Copenhagen University and her associates.
Although ongoing PPI therapyis indicated only in patients who have persistent gastro-esophageal reflux or who are taking NSAIDs, physicians have long reported that it is difficult to reduce or withdraw the drug after a 4-week or 8-week course of empiric treatment for dyspepsia. This has led some to suspect that perhaps physiologic changes induced by the drug itself cause hypersecretion after the PPI is withdrawn.
To investigate the issue, Dr. Reimer and her colleagues performed a double-blind trial in 120 healthy volunteers who had no acid-related symptoms. Sixty subjects were randomly assigned to receive 40 mg of esomeprazole daily for 8 weeks followed by 4 weeks of identical placebo tablets (withdrawal period), and the other 60 subjects were assigned to 12 weeks of placebo.
All were given antacid tablets to be used as rescue medication if they developed acid-related symptoms during the study.
The study subjects completed weekly electronic questionnaires that included the Gastrointestinal Symptom Rating Scale (GSRS). They also were assessed using the 36-item Short Form Health Survey (SF-36) and gave blood samples so that plasma levels of gastrin and chromogranin A could be tracked as indirect measures of gastric acid suppression and enterochromaffin cell mass.
After active treatment was withdrawn but subjects still thought they were taking the drug, “a significant and increasing difference between the PPI and the placebo group” was seen in dyspepsia and reflux symptom scores. A total of 44% of the PPI group reported heartburn, acid regurgitation, or dyspepsia, compared with only 15% of the placebo group.
The proportion of PPI-treated subjects who reported acid-related symptoms was 22% during the second week after withdrawal, 22% during the third week after withdrawal, and 21% during the fourth week after withdrawal. In contrast, the proportions of placebo subjects who reported such symptoms were 7%, 5%, and 2%, respectively.
The differences between the two groups also were significant for each of the individual components of the acid-related GSRS scores. In contrast, GSRS scores for diarrhea and constipation were not significantly different.
In addition, 52% of the active-treatment group reported using rescue antacids after PPI withdrawal. In contrast, only 11% of the placebo group used rescue antacids.
“Our study results reveal for the first time that profound acid inhibition with a PPI for 8 weeks induces acid-related symptoms in a significant proportion of subjects after withdrawal of therapy,” the investigators said.
Moreover, the increased use of rescue antacids after PPI withdrawal “attests to the clinical relevance of the observed symptoms,” they added.
Other researchers have investigated whether tapering the use of PPIs might help long-term users discontinue the drugs. In one clinical trial, “tapering over 3 weeks did not have a significant effect on the proportion that successfully withdrew treatment, compared to instant discontinuation,” they said.
In their study, subjects continued to report acid-related symptoms 4 weeks after withdrawal, and two other studies of PPI withdrawal showed increased acid secretion at least 8 weeks after PPIs were discontinued. Therefore, the duration of rebound acid hypersecretion remains unknown at present and tapering the medication does not appear to help.
“Our results justify the speculation that PPI dependency could be one of the explanations for the rapidly and continuously increasing use of PPIs,” they noted.
AstraZeneca supplied the medication and placebo used in this study.