U.K. Data Tie Acetaminophen to Increased Ulcer Risk

Widespread belief holds that the popular over-the-counter painkiller acetaminophen does not cause stomach ulcers, but a new study from Great Britain puts that in doubt.

General practice records from 1998 to 2018 in the U.K show that, among some 180,000 people age 65 and older who received prescriptions for acetaminophen (also known as paracetamol in Britain), risks for peptic ulcers, bleeding from ulcers, and any type of lower gastrointestinal (GI) bleeding were increased from 20% to 36% compared with more than 400,000 people not prescribed the drug, according to Jaspreet Kaur, PhD, MPH, of the University of Nottingham in England, and colleagues.

Acetaminophen use was also associated with increased rates of more general health problems including heart failure, chronic kidney disease, and hypertension, the group reported in Arthritis Care & Research.

“Despite its perceived safety, acetaminophen is associated with several serious complications,” Kaur and colleagues wrote. “Given its minimal analgesic effectiveness, the use of acetaminophen as the first-line oral analgesic for long-term conditions in older people requires careful reconsideration.”

Although acetaminophen’s safety has largely been taken for granted — WebMD, for example, declares flatly that it “won’t lead to peptic ulcers” — the authors noted that evidence to the contrary has been accumulating. A decade ago, Britain’s National Institute for Health and Care Excellence cited studies suggesting that the agent does, in fact, modulate the same cyclooxygenase-related pathways as do nonsteroidal anti-inflammatory drugs (NSAIDs) such as ibuprofen, with the potential for increasing risk for adverse GI effects. And in 2022, the institute formally dropped acetaminophen from its recommended treatments for osteoarthritis, both for safety and because it didn’t appear very effective.

Kaur and colleagues conceived their study to examine the risks in the general senior population. They examined the U.K.’s Clinical Practice Research Datalink, looking for people age 65 and older issued at least two acetaminophen prescriptions during a 6-month period without a concomitant order for another analgesic. They also included only those patients who had not received any acetaminophen scripts in the 12 months preceding the first qualifying prescription, “to avoid prevalent users.” Follow-up for adverse events, as recorded in patients’ charts, began 12 months after the first prescription.

Each acetaminophen user was then matched in varying ratios to patients not qualifying as a user (either “prevalent” or by the preceding inclusion criteria), by age, sex, and general practice location. In all, the analysis comprised 180,483 users and 402,478 non-users. Kaur’s group also matched users 1:1 with non-users by propensity scoring, with 79,024 patients in each group.

Mean patient age was about 75 and just over 60% were women. Without propensity matching, users differed substantially from non-users in many characteristics. Users also took opioids, NSAIDs, aspirin, proton pump inhibitors, and other medication types at higher rates; they were also more likely to be overweight or obese, to be current or former smokers, and to have comorbidities. Propensity matching, however, greatly diminished these differences.

After propensity matching and adjustment for remaining covariates, Kaur and colleagues calculated the following hazard ratios for users versus non-users:

• Uncomplicated peptic ulcers: HR 1.20, 95% CI 1.10-1.31
• Peptic ulcer bleeding: HR 1.24, 95% CI 1.16-1.34
• Lower GI bleeding: HR 1.36, 95% CI 1.29-1.46
• Heart failure: HR 1.09, 95% CI 1.06-1.13
• Chronic kidney disease: HR 1.19, 95% CI 1.13-1.24
• Hypertension: HR 1.07, 95% CI 1.04-1.11

These risks increased with the number of prescriptions, the authors emphasized. For example, hazard ratios for patients with nine or more acetaminophen prescriptions reached 1.39 (95% CI 1.22-1.58) for uncomplicated peptic ulcers, and 1.44 (95% CI 1.29-1.60) for peptic ulcer bleeding, relative to acetaminophen non-users.

Although the study data obviously could not identify a biological mechanism to explain these risks, Kaur and colleagues offered some speculation. One, they suggested, “prolonged acetaminophen ingestion might inhibit prostacyclin synthesis in humans, resulting in gastrointestinal lesions and bleeding,” perhaps because the drug can block excretion of a prostacyclin metabolite, signalling that renewed synthesis is not needed.

With regard to the apparent adverse renal effect, the researchers cited previous studies suggesting that the drug can cause acute tubular necrosis with long-term use.

The study did come with an important limitation: “there is no provision for recording over-the-counter prescriptions” in the database, the researchers acknowledged, and it is possible that users (and non-users, too) also bought acetaminophen on their own. But in Britain, patients 65 and older get acetaminophen for free when it’s formally prescribed. As well, acetaminophen use following that first 6-month period was not tracked in the study.

Please enable JavaScript to view the comments powered by Disqus.