20 November 2017

Rethinking prolonged PPI use

Clinical Gastro

The sharp, burning feeling of reflux is so unpleasant it drives millions of people around the world to heavy-duty drugs that tell the stomach to stop performing one of its key roles – producing acid.

By switching off the proton pumps in the gastric cells, proton-pump inhibitors (PPIs) disrupt the production of acid and protect us from the damage too much of it wreaks on the delicate lining of the oesophagus, for instance.

This has been tremendously useful in saving people from the pain of reflux and ulcers caused by NSAID therapy, and preventing progression of more serious diseases, such as Barrett’s oesophagus, cementing the place of PPIs as one of the most popular drugs here and abroad in the three decades since they first became available.

In Australia, PPI use increased by 1318% in the 10years after 1995. They now regularly make the top-10 leaderboard for PBS-subsidised prescribed drugs, accounting for around 20 million prescriptions a year.

As well-tolerated and safe as the medications appear to be in the short-term, more and more papers have been published in recent years raising concerns about potentially dangerous side effects associated with prolonged PPI use.

Research also suggests that one, and in some populations maybe two, in every three people taking PPIs may be doing so unnecessarily.

In the last few years, the RACGP and the Gastroenterological Society of Australia have made special mention of prolonged PPI use as part of their Choosing Wisely recommendations, asking doctors to taper patients down to the lowest effective dose and to deprescribe entirely if possible.

“Even though GORD is often a chronic condition, over time the disease may not require acid suppression and it is important that patients do not take drugs that are no longer necessary,” the society says.

They point to research now linking long-term use to fractures, pneumonia, enteric infections, vitamin and mineral deficiencies, and acute interstitial nephritis – particularly among older patients in whom PPI use is high.

Even in the last fortnight, a study published in the journal, Gut, found that PPI use more than doubled the risks of developing stomach cancer compared with non-users in patients previously treated for Helicobacter pylor infection.

PPIs are often used as part of the treatment of H. pylori, which itself is a cause of corpus atrophy and hence gastric cancer. But this link between PPI use and gastric cancer post infection was something new.

To better understand whether PPIs themselves had anything to do with stomach cancer, UK and Hong Kong researchers compared the cancer rates among users and non-users who had already been successfully treated for the H. pylori infection with a triple therapy of antibiotics and PPIs.

The large population-based study of 63,000 adults found that not only were gastric cancer rates twice as high for PPI users than non-users, but the likelihood of developing cancer escalated with higher doses and longer duration of therapy.

Daily use of PPIs was linked to a five-fold increase in gastric cancer compared with non-use, and the risk was eight-fold higher for those using the medication for three years or more.

Microbiologist Associate Professor Richard Ferrero has been studying H. pylori for 30 years and says the findings raise concerns about PPI use and overuse in the community.

“In our society, more and more people want a quick fix,” Professor Ferrero says. He laughs, saying he sees this firsthand with his wife, who has been treated for a H. pylori infection in the past: “My wife will pop a PPI when she’s got reflux, and I’ll be telling her when I go home she should be sticking more to the Mylantas.”

But much of the research linking prolonged PPI use to various ailments is still emerging, and often contradictory.

Even though this study builds on the idea that there is some characteristic of PPIs responsible for gastric cancer, it is still correlational, the head of the Gastrointestinal Infection and Inflammation research group at the Hudson Institute says.

However, it does appear to be backed up by trials on animals, which indicate there is something about the combination of PPIs and H. pylori behind the development of cancer. Professor Ferrero points to research on gerbils given a H. pylori infection, which found significantly more precancerous lesions in animals given the medication than those which weren’t.

H. pylori infection itself increases the risk of developing stomach cancer, but of those who have been infected only around 1% to 2% go on to develop it, Professor Ferrero explains. Nevertheless, of everybody who is infected and develops gastritis, and up to one in five go on to develop peptic ulcer disease, including  duodenal ulcers.

So what makes PPIs so particularly risky when compared with other acid-suppressing medications such as H2 receptor antagonists?

They have different modes of action, Professor Ferrero explains.

“PPIs are more effective at shutting down acid production in the stomach, and there is some evidence that they can have an effect on gastrin production in the stomach.”

The hormone gastrin causes changes in the stomach, causing cell proliferation and that proliferation can lead to cancer.

Nevertheless, Professor Ferrero wants to be careful not to create a Catalyst-style panic around the well-tolerated and well-used drugs, emphasising their importance for the right indication. But using the powerful acid suppressant for mild reflux, or long-term without indication, is taking a “bazooka-like” approach to the problem, he says.

PPIs can turn around the life of someone with severe reflux disease, and have been a wonderful tool in preventing common problems such as oesophageal strictures, and even oesophageal cancer, Associate Professor Simone Strasser, spokesperson for the Gastroenterological Society of Australia, says.

But as part of her work at the Royal Prince Alfred Hospital’s liver unit with patients with advanced liver disease, Professor Strasser says she is forced to take people off unnecessary PPIs and lower their doses “all the time”.

“The issue with PPIs is that people take them for less severe indications – for a bit of reflux here and there – and stay on them long term,” she says.

Most of these patients aren’t preventing any long-term complications by using them, undermining the risk-benefit calculation needed to justify prolonged use.

Professor Strasser says there has been growing concern in the last few years about the impact of prolonged PPIs on the movement of bacteria across the bowel wall, known as bacterial translocation.

This unexpected movement of bacteria increases the risk of infection in people with cirrhosis, as well as increasing the risk of complications such as encephalopathy, she adds.

It is particularly concerning because patients experiencing these complications state associated with cirrhosis are more likely to die than their counterparts.

“There have been studies now linking the use of PPIs in advanced liver disease with recurrent encephalopathy, recurrent infections and reduced survival,” she says.

A study published in Hepatology just last year of advanced liver disease patients found that PPI use was associated with an almost 90% increased risk of getting encephalopathy and a 70% increased risk of getting spontaneous bacterial peritonitis.

“So the focus in hepatology and liver disease management is to constantly review the indications for a PPI and to take people off them and to reduce those sorts of adverse outcomes.”

This kind of deprescribing may also help prevent other complications being hinted at in the research.

Stomach acid also acts as a defence against pathogens and diminishing its availability  can allow bugs to be ingested, potentially causing infections such as pneumonia or gastroenteritis.

Last year, a meta-analysis of nine trials including 121,000 cases of pneumonia found that the risks of infection were higher among people who had used high-dose PPIs, even among those using the drugs for less than a month.

The research is not clear cut, however. Another case-controlled review of a similar size failed to find such a link after confounders had been accounted for, and a retrospective cohort study found hospitalisations for community acquired pneumonia were no higher among new NSAID users who were also taking the acid suppressant, compared with ones who weren’t.

So just how sure can we be that PPIs are the cause of the problem, and they are not just a marker of a person with multiple health concerns?

Professor Strasser says the research is not clear cut. For many conditions, studies have emerged saying PPIs are a problem, and others saying they aren’t. “So it all comes down to the quality of the study and the populations involved.”

In August, an independent, non-profit organisation, The Medical Letter, published its critical appraisal of the research into safety issues around PPIs, including links to fracture, hypomagnesemia and QT prolongation, kidney disease, vitamin B12 deficiency, iron deficiency, community-acquired pneumonia, dementia and mortality.

Fracture risk is one of the most well-known associations with the medication, with the US health regulator insisting on warning labels to caution against long-term use because of this risk.

Last year, a meta-analysis of 118 observational studies, including almost 250,000 fracture cases, found that patients taking PPIs for both the short and long term were at an increased risk of fracture.

PPI use led to a 26% greater risk of hip fractures, 58% greater risk of spine fractures and an overall risk of fracture 33% higher than non-use.

One explanation for this is that lowering gastric acidity may impair the body’s ability to absorb calcium and other minerals and nutrients.

Similarly, the absorption of folate and vitamin B12 may be impaired in people who take PPIs, which could increase fracture risk by changing homocysteine levels.

Or possibly it has something to do with the direct action PPIs have on skeletal cells such as osteoclasts, which have their own proton pumps. Less bone turnover could lead to worse bone resorption, but scientists are, as yet, unclear on exactly how it would affect bone strength and fracture.

Another prospective study published in 2012 backed up the causal link between PPIs and fracture, after showing that even after controlling for risk factors such as femoral neck bone density, people using PPIs were 40% more likely to have a nontraumatic fracture compared with people who did not use PPIs. This study, which followed up 9400 people for 10 years, found people on PPIs were quicker to have a fracture than those not taking the drug.

That said, there have also been other analyses, such as the Nurses’ Health Study of 80,000 postmenopausal women, which have found no link to fracture risk.

Then there is hypomagnesemia, a condition that occurred rarely, but demonstrably, in a 2015 meta-analysis of 110,000 patients across nine different observational studies.

The condition, which leads to weakness, cardiac arrhythmias, convulsions and hypotension, has also been flagged by health regulators as something to watch for among patients taking the drugs.

According to The Medical Letter, the condition normally goes hand in hand with conditions such as hypokalemia and hypocalcemia, and potentially even QT interval prolongation and torsades de pointes.

Kidney disease has also come under the microscope after major studies found higher rates of chronic kidney disease among PPI users.

A prospective cohort study of 10,000 patients with an estimated glomerular filtration rate (eGFR) of ?60 mL/min/1.73 m2 found that there was a 45% higher risk of chronic kidney disease among PPI users.

Another retrospective study of 115,000 patients initiating PPIs or H2 blockers found serum creation doubled in PPI users, eGFR levels dropped, and end stage renal disease and acute kidney injury were more likely.

It’s not clear what the mechanism is though, according to the authors of The Medical Letter.

Then just last month, University of California researchers found links between PPIs and liver disease, which they attributed to the changes in the gut bacteria.

While alcohol is to blame for around half of all cirrhosis-related deaths, the authors say the medication may exacerbate the problem by changing the gut microbiota and allowing more harmful bugs to take hold.

Their research suggests that without the gastric acid, enterococcus bacteria proliferates in the intestine, where it is then able to move to the liver and cause inflammation and liver disease. They investigated the possible link from two angles: a population study in humans and a controlled trial in mice.

The researchers first created mouse models of non-alcoholic fatty liver disease and nonalcoholic steatohepatitis and then used either PPIs or genetic engineering to blood acid production in the stomach.

Testing the gut microbiome of the mice revealed those with gastric acid suppression had more enterococcus bacteria in their intestines and translocation to the liver. Similarly, when they colonised other mice with an excess of the same bacteria, they found that just like the mice on gastric acid suppressants, the excess bacteria increased liver inflammation and injury.

More enterococcus bacteria in the mice meant more progression in alcohol-induced liver disease, NAFLD and NASH.

Then, when the researchers looked at the stool samples of 5000 patients with chronic alcohol abuse, they found enterococcus was higher among those taking PPIs.

Among the human population, PPIs also appeared to be an independent risk factor for  liver disease among alcoholics. Active PPI users had a 21% chance of being diagnosed with alcoholic liver disease, compared with 16% in previous users and 12% in never users.

This indicates that people who chronically abuse alcohol have an 8% higher absolute risk of developing liver disease if they are currently taking the drugs.

Reducing or eliminating PPI use where possible is not only  consistent with the principle of quality use of medicine, but also prevents harms in groups like older patients, who we know have high rates of PPI use and potentially dangerous polypharmacy, Professor Strasser says.

“We know as people get older, medications end up getting added but very rarely get subtracted, and we know now there are interactions between PPIs and other drugs and some drugs actually rely on stomach acid for absorption,” she says, pointing to the expensive hepatitis C drugs that need a certain level of stomach acidity to work.

“You need stomach acid, it’s why we’re built this way,” she says. “And to suppress it long term, has not surprisingly, got long-term impacts.”