McMaster study casts doubt on common fracture treatment
The study was completed even though the company that makes the machine pulled its funding
McMaster University has published a study on a widely-used ultrasound treatment for fractures, years after the company that sponsored the trial and builds the ultrasound machine being studied pulled funding for the research.
The study, published in the British Medical Journal, concluded that receiving low-intensity pulsed ultrasound after surgery to repair a fractured shinbone made no difference — it did not accelerate healing or improve function.
I'm certain that the industry sponsor was very hopeful it would show a benefit.- Dr. Jason Busse, McMaster researcher
The outcome of the trial was published Tuesday with a linked editorial that criticized the trial's industry sponsors Bioventus Global, a subsidiary of the UK-based Smith & Nephew, for pulling funding and trying to influence what were deemed unfavourable results.
The company stopped its funding in 2012, just before the halfway mark of the study after it did a review of its own of the results to that point.
The company Tuesday challenged the outcomes, saying the study was flawed.
Dr. Jason Busse, who led the trial, told CBC News that he found the company's statement to be "an unusual rationale," and added trials of this sort are necessary for "confidence in the clinical process."
"I'm certain that the industry sponsor was very hopeful it would show a benefit. They were hoping to use this study to provide support for sales of their device," he said, adding that the company's "backtracking" was to be expected.
The BMJ editorial, written by Dr. Xavier L. Griffin, associate professor of trauma surgery at Oxford University, said there is now sound evidence that LIPUS doesn't work, despite attempts to influence the results.
"Fortunately for patients, clinicians, and future clinical guideline groups, the results were eventually clear, despite the influence of outside agencies," wrote Dr. Griffin, who was not involved with the trial through its implementation or subsequent publication processes.
"It is time for us to make good use of [the authors'] determination and abandon this ineffective treatment."
Trial sponsored, then de-funded by manufacturers
The trial, led by Dr. Busse, associate professor in the department of anaesthesia at McMaster University, and his colleagues, was initiated and sponsored primarily by Bioventus Global, a company that produces the EXOGEN LIPUS machine, to determine the effectiveness of their product by a randomized trial completed in 43 academic trauma centres across North America.
Dr. Busse said the trial was initiated almost ten years ago, after the company reached out to him following a published review to determine the effectiveness of their product.
He and his team conducted a survey and found that almost half of orthopedic trauma surgeons in Canada used bone stimulators for at least some of the fractures that they managed, evenly split between the EXOGEN LIPUS treatment and a competing electrical stimulation product.
The treatment is one of a variety available to patients after their fractured limb has been operated on and "nailed" together. Patients are meant to use it for 20 minutes a day, by applying conducting gel to their limb, strapping on a transducer, and letting the machine apply an ultrasound signal to the affected bone for stimulation until it switches itself off.
However, he says, Bioventus Global conducted an unplanned interim analysis in late 2012 that revealed no significant difference in healing behavior between the blinded groups.
The company hastily pulled funding for the research, leaving Dr. Busse's team unable to complete its original plan for the trial.
"They saw there was no difference between the two groups, and then decided they would terminate the trial on the basis of futility," said Dr. Busse.
"We pursued it as long as we could, collecting data, but when the funding was discontinued we were unable to complete the study to what we fully planned to do, which was one year of follow up on everybody that was enrolled. So there were 72 or 73 patients that we were not able to do a complete follow-up on."
The study ultimately found no significant difference in healing outcomes for patients using the EXOGEN machine versus those who used a "sham" placebo machine.
Company rejects study results
Bioventus Global denied the findings on Tuesday, saying that the data was misleading.
The company's chief medical officer, Dr. Peter Heeckt, said that the patients enrolled in the trial showed poor compliance, only 43% on average, and many of the patients displayed uncomplicated fractures that are not necessarily the targeted users of the treatment.
"We believe that the study's results are massively confounded by two things, the most significant of them is the fact that the compliance or adherence of the patients to the treatment is very poor, and that unfortunately invalidates the study's results," said Dr. Heeckt.
"The other view is that there were not very many patients in the study who had high risk factors for impaired fracture healing," he said.
"The patients in the study were mostly male, with uncomplicated tibial fractures, and they heal very very well."
Dr. Heeckt said the results could not be seen as representative of all types of fracture injuries as a result, and said they stood by the effectiveness of their product.
Bioventus Global estimates that about 1,250 EXOGEN units are sold in Canada per year.
A spokesperson for the company said that they could not disclose how much funding was spent on research and development in individual projects, but Dr. Heeckt mentioned it would be in the millions.
Sponsor partaking in "revisionist history"
Dr. Busse said he does not believe that the patients enrolled in the trial showed a lower compliance rate than any patient would in the real world, and said the company was now engaging in "revisionist history" to minimize the fallout from the trial's findings.
"We only got the compliance data years after they had pulled out of the trial (in 2015), and we had to work very closely with them for many months to figure out what the data was saying," he said.
The original raw data provided by Bioventus Global to Dr. Busse and his team was uninterpretable, he said, and included data from machines that were not included in the trial, and some recorded data was missing altogether.
"There's no reason to believe that the compliance in our trial would be any different than patients would have outside of the trial," he added.
"If anything, compliance in our trial would be expected to be a little better, given that when we enrolled patients, we made that one of our points of discussion to get their commitment to be compliant... and we followed them on a regular basis."
He also said that a deliberate effort had been made to ensure that an equal representation of patients with varying severity of injury, and that even subgroup analyses showed no difference in effect based on gender, smoking status, presence of diabetes or even the seriousness of the injury.
Bioventus Global said it had no plans to discontinue producing the device, or change how it is marketing to orthopedic surgeons as a result of the trial.