午夜福利1000集合

Which painkiller to choose? (Image: Grant Delin/Millennium Images)

It鈥檚 the first resort for anything from a child鈥檚 fever to arthritis 鈥� but paracetamol (acetaminophen) isn鈥檛 as safe or effective as we thought it was

YOU鈥橵E got a terrible headache. Niggling knee pain. An aching back. What do you reach for? Chances are that you鈥檒l open your medicine cabinet and grab some paracetamol. Half an hour or so later, you鈥檒l feel a lot better. Or will you?

Paracetamol, also known as acetaminophen, is the cure-all of our age, used to treat everything from sprained ankles to toothaches and even labour pain. It is on the first rung of the , which doctors use to treat cancer pain. We spoon it to our children to fight fever; as adults we pop it to relieve headaches or period cramps, and as we get older we鈥檙e prescribed it to soothe arthritis or backache. In the US, 27 billion doses of the drug are sold each year, and it is found in .

Given its ubiquity, you might assume that paracetamol is safe and effective 鈥� at least at the recommended dose. That鈥檚 why we lean on it more than aspirin or ibuprofen, which can irritate the stomach lining and cause bleeding. But as it turns out, this stalwart of the medicine cabinet is not quite as reliably gentle as you might think.

Paracetamol was discovered in the late 19th century, but it was rejected almost immediately because of a bizarre side effect: it seemed to turn some people blue (see timeline). That was probably because of contamination with a different drug, but as a result paracetamol was sidelined until the 1940s, when further tests showed it was good at reducing fever. Later studies concluded that it was a pretty effective painkiller too. But it really took off in the 1960s, in response to emerging concerns about the long-term side effects of aspirin and other non-steroidal anti-inflammatory drugs (NSAIDs). Today in the US, there are about 16,500 NSAID-related deaths a year in people with arthritis alone.

Paracetamol, on the other hand, we think of as relatively safe. Sure, if you take lots of tablets it could seriously damage your liver, but at the recommended dose, it鈥檚 fine, right?

This assumption is now being challenged by research suggesting that, when taken for prolonged periods, it may damage the stomach as much as NSAIDs. That might be an acceptable risk in exchange for pain relief, but in many of those who take it, paracetamol barely works better than a placebo.

Mysterious drug

How could this be? The fact is, despite its ubiquity, we still don鈥檛 really understand how paracetamol works. A leading theory is that, in part, it works like aspirin and ibuprofen, by blocking enzymes known as cyclooxygenases. These enzymes are responsible for making hormone-like compounds called prostaglandins, which trigger pain and swelling in the body as well as stimulating production of the mucous that shields our stomachs against digestive acids. NSAIDs halt the swelling process, but leave the stomach vulnerable. The suspicion was that paracetamol inhibited cyclooxygenases, but to a much lesser extent; it doesn鈥檛 reduce inflammation as these other drugs do.

Although studies in the past decade have hinted that long-term use of paracetamol might trigger internal bleeding, these findings were widely dismissed by critics who cited shortcomings of the study designs. In 2011, however, Michael Doherty of Nottingham City Hospital, UK, published a study that was harder to ignore. He followed the progress of 892 men and women with the niggling knee pain that often sets in at middle-age 鈥� usually an early symptom of osteoarthritis. Some were given paracetamol, others ibuprofen, while a third and fourth group took either a high or low-dose combination of the two.

Paracetamol is the first drug most doctors turn to for patients with such symptoms, but when Doherty looked at the blood results of those taking it, he was shocked: levels of haemoglobin, the protein that carries oxygen in the blood, were dropping fast. What鈥檚 more, their red blood cells were growing smaller and paler. The most logical explanation was that they were losing blood internally, and significant quantities of it. After three months, a fifth of them seemed to have lost the equivalent of an entire unit of blood (about 400 millilitres). That was the same amount as those taking ibuprofen 鈥� only the ibuprofen group reported feeling less pain (Annals of the Rheumatic Diseases, vol 70, p 1534).

In those combining high doses of both paracetamol and ibuprofen, the haemoglobin loss after three months was even more startling: 7 per cent of the people in that group lost the amount of haemoglobin you would find in two units of blood. The upshot: when taken for long periods, paracetamol may be just as damaging to the stomach lining as NSAID drugs are.

鈥淭he horrifying aspect of this is that people look at me and say 鈥榠t鈥檚 over the counter, it must be safe鈥�,鈥� says Kay Brune, a professor of pharmacology and toxicology at the University of Erlangen-Nuremberg in Germany. Brune has been campaigning to have paracetamol removed from over-the-counter sale in Germany, but has so far been unsuccessful. 鈥淏efore, physicians simply said 鈥極K, if it doesn鈥檛 work, it may not do any harm鈥�. But now we know it can do harm,鈥� he says.

Internal bleeding isn鈥檛 the only issue that鈥檚 keeping drug regulators on their toes. In January, the US Food and Drug Administration asked manufacturers to stop producing prescription drugs containing more than 325 milligrams of paracetamol per tablet because of the risk of accidental overdose. Paracetamol poisoning is responsible for nearly 80,000 visits to the emergency room in the US each year, and a third of these are people who overdosed accidentally.

Although pill packets clearly state that the maximum recommended dose is no more than 3 or 4 grams spread over 24 hours (or six to eight 500 mg tablets), because of paracetamol鈥檚 reputation for safety, some people take more than this. 鈥淭hey know they鈥檙e not supposed to take maybe six or eight tablets at a time, but they have a toothache and they just don鈥檛 want to go to the dentist,鈥� says at the US Centers for Disease Control in Atlanta, Georgia, who .

If you regularly exceed 4 g, you can quickly enter dangerous territory. During the breakdown of paracetamol, a toxin is produced that has to be mopped up by a specific enzyme in the liver, and if you take too much too fast, the supply of that enzyme quickly dwindles.

As little as 5 to 7.5 g per day can cause serious liver complications in otherwise healthy people. For people with compromised liver function due to alcoholism or liver disease, a harmful dose can be lower still. And despite the fact that the recommended maximum dose is no more than 4 g per day, roughly 6 per cent of US adults 鈥� about 14 million people 鈥� are , often in prescriptions that combine the drug with opioids to treat severe pain.

Do these risks matter? Because of the huge numbers of people who take paracetamol, and the relative ease with which it is purchased and consumed, even small risks become significant. Even so, paracetamol is valued by medical authorities 鈥� not just for treating life鈥檚 little hurts, but for persistent and potentially debilitating conditions. The UK鈥檚 National Institute for 午夜福利1000集合 and Care Excellence (NICE), the body that sets standards for medical practice, recommends paracetamol as the first-choice drug for treating the chronic pain associated with conditions like osteoarthritis and lower back pain. The American College of Rheumatology also recommends it for arthritis.

In the US, an estimated 43 million people take paracetamol each week, and nearly two-thirds of them take the drug routinely for longer than six months.

If paracetamol was effective against chronic pain, you might consider the trade-off worthwhile, but the drug has been found seriously wanting. A review of research that looked at people taking paracetamol to relieve chronic joint pain found seven studies that compared the drug with a placebo. Five of these found it to be marginally more effective, but two found no difference.

鈥淲hy are we bothering to give a drug to people that鈥檚 toxic, that has significant potential problems, when it doesn鈥檛 work?鈥� asks Andrew Moore, an anaesthetist and director of pain research at the University of Oxford. 鈥淚t鈥檚 unethical.鈥�

鈥淲hy are we bothering to give a drug to people that鈥檚 toxic, when it often doesn鈥檛 work?鈥�

Of course, placebos can themselves make people feel better: another for treating joint pain found that many people experienced moderate relief from sham treatment, particularly when it was given as an injection. For ethical reasons, doctors don鈥檛 usually prescribe placebos, so the safest active pill is often the next best thing.

鈥淚s paracetamol a safe placebo?鈥� asks John Dickson, a rheumatologist with the Redcar and Cleveland Primary Care Trust in the UK, and a consulting clinician for the 2008 NICE guidelines. 鈥淭he work Doherty did shows it is not.鈥�

In March, the Osteoarthritis Research Society International changed its paracetamol guidelines to 鈥渦ncertain鈥� to reflect growing safety concerns. And for a while at least, it looked like these concerns would be similarly heeded in the UK. When NICE issued new draft guidelines for osteoarthritis in August last year, it did away with the recommendation of paracetamol as a first resort, and flagged its potential dangers. 鈥淥n balance, the risks of paracetamol outweigh the benefits of any gain in symptom control,鈥� the report read.

Yet by the time the final version was published in February, the old advice had been reinstated. This was partly down to objections raised by doctors about having few alternative options, though NICE says it is also awaiting the results of a more comprehensive review of over-the-counter painkillers by the UK鈥檚 Medicines and 午夜福利1000集合care Products Regulatory Agency. Dickson, like others, was disappointed. 鈥淚f paracetamol isn鈥檛 safe, we shouldn鈥檛 be prescribing it,鈥� he says.

Of course, most of us don鈥檛 take paracetamol every day; it鈥檚 a drug we reach for when we develop a headache or sprain an ankle. And for acute pain of that nature, paracetamol performs reasonably well, if not as spectacularly as its popularity might suggest. Pharmacists measure the effectiveness of painkillers by looking at whether they can reduce your reported sensation of pain by at least 50 per cent, and by counting how many people would need to take it for one person to experience this level of relief compared with placebo. This is known as the number needed to treat (NNT).

Effective relief

For example, in the case of the moderate pain of a sprained ankle, 3.8 people would need to take a standard 1 g dose of paracetamol (2 tablets) for one of them to get effective relief. For a standard 400-milligram dose of ibuprofen, the NNT is 2.5 (see table).FIG-mg29710502.jpg

Most people suffering from acute pain are unlikely to take these drugs for more than a few days, so the risk of internal bleeding is less of a concern than in those taking it for prolonged periods. But, given that paracetamol isn鈥檛 as effective as some alternatives for short-term pain, it could make more sense to take one of them, or a combination of drugs that work through different pathways, such as paracetamol plus ibuprofen.

Should we do away with paracetamol entirely? Most experts believe it鈥檚 still a useful tool in the arsenal against fevers, headaches and sore muscles because, in the people for whom it does work, it tends to work fairly well. It鈥檚 just that, as with many analgesics, the chances are hit-and-miss that it will work for you 鈥� possibly because everyone鈥檚 body is slightly different.

However, when it comes to chronic pain, it could be time for a rethink. Moore suggests measuring your pain, tracking whether a drug makes a difference, and if it doesn鈥檛, quickly moving on. 鈥淔rankly, with paracetamol, if it鈥檚 not going to work within a week, it鈥檚 never going to work with you,鈥� he says.

Indeed, a spokeswoman for McNeil Consumer 午夜福利1000集合care, which makes Tylenol in the US, points out that the drug鈥檚 label clearly states that consumers should stop use and ask a doctor if they have pain that gets worse or lasts more than 10 days.

Of course, the ideal would be to develop a paracetamol variant that worked better and had fewer drawbacks. Stuart Bevan and David Andersson at King鈥檚 College London recently found that when paracetamol is given, one of its break-down products activates a protein on the surface of nerves in the spinal cord and reduces their ability to transmit pain signals. If confirmed, targeting this protein could be a promising starting point.

Pharmaceutical companies are also researching and developing new analgesics. But given the huge regulatory hurdles for over-the-counter drugs, few are focusing on that market. 鈥淚t is more likely that medicines currently available on prescription would become available over the counter, as they will already have a good amount of safety data,鈥� says Roger Knaggs at the University of Nottingham, UK.

Still, it鈥檚 possible that a promising alternative already exists. Just as paracetamol was consigned to a dusty back room for half a century, other analgesics may have been overlooked or condemned for the wrong reasons. Safety hurdles today are much higher than when drugs like paracetamol were first approved. If it were a new drug, says Moore, it probably wouldn鈥檛 get approval.

It could also be that some drugs which fail to win approval are doing so because of poor study design, rather than serious flaws with the drugs themselves. Robert Dworkin at the University of Rochester in New York, is the director of an initiative with the FDA that is taking a second look at analgesics that didn鈥檛 pass muster in earlier clinical trials. It is currently focused on prescription-strength drugs, but Dworkin says a similar approach could work for over-the-counter remedies too.

In the meantime, what should you do with the paracetamol in your own cupboard? For short-lived aches and pains, the advice hasn鈥檛 changed much. 鈥淚f you follow the instructions and if you don鈥檛 take it in too-large doses, paracetamol is very safe,鈥� says Bevan.

But for ongoing pain, it may be time to start looking for alternatives. With any drug, there鈥檚 a risk that side effects will outweigh benefits. For paracetamol, we need to decide which risks are still worth taking.

Leader: 鈥�Don鈥檛 kill the painkiller, educate the people鈥�

This article appeared in print under the headline 鈥淐ure-all no more鈥�