* Large trial in Australia found no benefit from painkiller
* Findings raise questions over why it is recommended
* Low back pain the leading cause of disability worldwide (Adds fresh comment, para)
By Kate Kelland
LONDON, July 24 (Reuters) - Paracetamol, a painkiller universally recommended to treat people with acute low back pain, does not speed recovery or reduce pain from the condition, according to the results of a large trial published on Thursday.
A study published in The Lancet medical journal found that the popular pain medicine was no better than placebo, or dummy, pills for hastening recovery from acute bouts of low back pain or easing pain levels, function, sleep or quality of life.
Researchers said the findings challenge the universal endorsement of paracetamol as the first choice painkiller for lower back pain.
"We need to reconsider the universal recommendation to provide paracetamol as a first-line treatment," said Christopher Williams, who led the study at the University of Sydney in Australia.
Lower back pain is the leading cause of disability worldwide. In the United States alone, costs relating to the condition are estimated to be more than $100 billion a year.
Currently, every back pain treatment guideline in the world recommends paracetamol as the first-line analgesic and Williams said this was despite the fact that no previous studies have provided robust evidence that it works in this condition.
Tim Salomons, a pain expert at Britain's University of Reading whose own research has found that cognitive behavioural therapy could be used to treat chronic pain, said this latest study showed the challenge of treating the condition.
"It is vitally important we continuously challenge conventional wisdom about treating pain," he said in an emailed comment. "Even though paracetamol has a good safety profile, every drug has side effects. If the drug is not doing what it is being prescribed to do, pain patients might be better off without."
In Williams' trial, 1,652 people from Sydney with acute low back pain were randomly assigned to receive up to four weeks of paracetamol, either in regular doses three times a day, or as needed, or to receive placebos. All those involved received advice and reassurance and were followed up for three months.
The results showed no difference in the number of days to recovery between the treatment groups - with the average time to recovery coming out at 17 days for each of the groups given paracetamol, and at 16 days for the placebo group.
Paracetamol had no effect on short-term pain levels, disability, function, sleep quality, or quality of life, the researchers said, and the number of patients reporting negative side effects was similar in all groups.
Christine Lin, an associate professor at the George Institute for Global Health and the University of Sydney who also worked on the study, said the reasons for paracetamol failing to work for lower back pain were not well understood.
"While we have shown that paracetamol does not speed recovery from acute back pain, there is evidence that paracetamol works to relieve pain for a range of other conditions, such as headaches, some acute musculoskeletal conditions, tooth ache and for pain straight after surgery," she said in a statement about the findings.
"What this study indicates is that the mechanisms of back pain are likely to be different from other pain conditions, and this is an area that we need to study more."
Experts who were not directly involved praised the study but cautioned that guidelines should nevertheless not be changed on the basis of a single piece of research.
"More robust and consistent evidence, including verification of the results in other populations, is needed," Bart Koes and Wendy Enthoven from the Erasmus Medical Centre in the Netherlands wrote in a Lancet commentary.
They also called for more studies on whether other simple analgesics could add extra benefits on top of giving advice and reassurance to patients. (Editing by Robin Pomeroy and Catherine Evans)
Our Standards: The Thomson Reuters Trust Principles.