Stopping PPIs: Is Heartburn After Stopping Rebound, and How to Taper Off

PPIs block an enzyme (the proton pump) in the parietal cells of the stomach that secretes hydrochloric acid. When acid secretion is strongly suppressed, the stomach reads this as a signal to increase secretion of gastrin, a hormone that stimulates acid production. Raised gastrin acts for many weeks on enterochromaffin-like cells, which stimulate the parietal cells to work, and leads to their enlargement. As long as the drug is taken, the heightened capacity to secrete acid is held back. When the drug is stopped, the blockade disappears, and the enlarged and stimulated secretory apparatus for a time produces more acid than before treatment. This phenomenon is called rebound acid hypersecretion (RAHS).

We describe here a simplified scheme that is generally accepted in the literature, but clinical data show that the severity and duration of such an effect are limited. In one of the studies discussed below, the severity of symptoms after stopping correlated with the gastrin level at the end of treatment, which fits this mechanism well, and symptoms passed within a few weeks. It is also important that rebound is a physiological and reversible phenomenon, not a lasting dependence.

How should this be read? Both trials were done in healthy people, without reflux disease and without symptoms before treatment. That is exactly what makes them so important: they show that the drug alone, without any disease, causes transient acid-related symptoms after stopping in about four in ten people. The duration was limited: in the Swedish trial, after 4 weeks of treatment the difference disappeared in the third week after stopping, while in the Danish one, after 8 weeks of treatment, symptoms were still present in the last weeks observed, that is up to 4 weeks after stopping.

Limitations: these are small trials, symptoms were measured with questionnaires, and severity was mostly mild (in the Danish trial mean scores differed by 0.2-0.3 points). In people with reflux disease the course may look different, and the duration of the effect after longer treatment, for example years-long, is not as well studied. The evidence points to the existence of the phenomenon but does not allow an exact prediction of how long it will last in a given person.

This is the most important practical question, because the answer decides whether symptoms should simply be waited out or treated. Rebound is transient in nature: it appears in the first days after stopping, peaks in the first or second week and fades with time. A relapse of reflux disease, by contrast, is the return of symptoms that existed before treatment, and it usually does not pass and sometimes grows.

The conclusion from this trial is the opposite of that from the trials in healthy people: in people with genuine reflux disease (here: without erosions on endoscopy), heartburn returned after stopping far more often than after using the drug, and not as a short-lived rebound but as a relapse that within 6 months forced more than half of those on placebo to withdraw. At the same time the trial shows that many people managed with the drug used only when heartburn appeared, which is one of the options for coming off the drug.

The American Gastroenterological Association (AGA) has published practical guidance on stopping PPIs. The guidance is not a systematic review, and its authors themselves state that they did not formally rate the quality of evidence, but it provides an organised point of reference. The Canadian guideline, based on the GRADE method, in turn concerns adults after at least 4 weeks of treatment for heartburn, mild to moderate reflux disease or oesophagitis whose symptoms have resolved.

There is no single universal scheme. The AGA guidance allows both gradual dose reduction and abrupt discontinuation, and the Canadian guideline lists three paths: dose reduction, stopping and on-demand dosing. A systematic review of PPI discontinuation studies, based on six clinical studies, points to several important observations.

Interpretation: the proportion of successful attempts (14-64%) is wide because the studies were different and covered different groups; it means that many people can come off the drug, but some cannot. The conclusion on the advantage of gradual reduction comes from a review, and the AGA guidance treats both approaches as acceptable, so it is not yet a clear-cut finding. In practice, what is most often proposed is a dose reduction or less frequent intake for several weeks and, once symptoms have settled, the drug on demand for heartburn, but the exact scheme should be set by a doctor or pharmacist, taking the original indication into account.

One of the reasons people want to stop PPIs is fear of the adverse effects described in observational studies: infections, fractures, kidney disease or dementia. It is worth knowing the data from a randomised trial that tested these signals in a controlled way.

Heartburn after stopping a PPI can be real and needs no shame or guilt: it is a transient consequence of the drug's effect on the physiology of the stomach, which affects some people, including quite healthy ones. It is not, however, a reason to give up trying to come off the drug if there is no indication for chronic treatment. A sensible strategy has three steps: agree with a doctor whether the indication still exists, plan how to come off (gradually, at once or on demand) and wait out the period of transient symptoms with a plan to ease them. If after a few weeks symptoms do not pass, or warning signs appear, the cause has to be sought rather than automatically going back to the drug.