Researchers have tested fake acupuncture, which often seems to work as well “real” acupuncture. But fake arthroscopic knee surgery?
Seems there would be enormous practical and ethical barriers to pulling off such a trial. Would the control arm still get anesthetized? Would they still go home with incision marks?
According to a paper in the New England Journal of Medicine, researchers in Finland experimented on a group of 146 people with knee pain. All had been diagnosed with damage to the piece of cartilage known as the meniscus. About half of the patients got real surgery and the other half, indeed, got anesthetized and cut. The doctors still inserted the scopes into the control patients to affirm that they indeed had meniscal damage. The trial showed that real surgery only had slight benefits over fake, and no benefits after a year.
This paper may have gotten more coverage if not for the holiday. The study's authors reported that 700,000 people get knee surgery for meniscal tears every year.
The New York Times covered the study in detail, making the key point that the result does not pertain to people who get surgery to treat acute injuries:
The Finnish study does not indicate that surgery never helps; there is consensus that it should be performed in some circumstances, especially for younger patients and for tears from acute sports injuries. But about 80 percent of tears develop from wear and aging, and some researchers believe surgery in those cases should be significantly limited.
According to the Times, most patients had a spinal anesthesia but some had a general. That’s a lot to ask of these volunteers. And yet, according to the study, they were happy with their treatments even when they found out they were in the control arm.
The Times included a detailed interview with an orthopedic surgeon who had a plausible explanation for the result: Perhaps the fake surgery patients did as well as those who got meniscal repair because the meniscal tears were not causing their pain. And whatever was causing the pain got better:
“Take 100 people with knee pain; a very high percentage have a meniscal tear,” said Dr. Kenneth Fine, an orthopedic surgeon who also teaches at George Washington University. “People love concreteness: ‘There’s a tear, you know. You have to take care of the tear.’ I tell them, ‘No. 1, I’m not so sure the meniscal tear is causing your pain, and No. 2, even if it is, I’m not sure the surgery’s going to take care of it.”
That doctor was quoted saying that he had a meniscal tear but, “I’m not going to let anybody operate.” His quote suggests that at least some doctors were already concerned about overuse of the procedure, or did not always deem it the best option.
The Wall Street Journal also covered the story, starting with this lead paragraph:
A fake surgical procedure is just as good as real surgery at reducing pain and other symptoms in some patients suffering from torn knee cartilage, according to a new study that is likely to fuel debate over one of the most common orthopedic operations.
If the doctor quoted in the Times is correct in suggesting that the torn knee cartilage wasn’t the primary cause of the pain, however, then the patients in the study would be better described as having been diagnosed with torn knee cartilage, rather than suffering from it. Sounds like we really don’t know what was causing their suffering.
The Journal version of the story did have a nice quote from one of the study authors:
"Doctors have a bad tendency to confuse what they believe with what they know."
And it had some interesting background on the use of placebos to test surgery:
Simulated surgery, which also is used in some medical-device studies, is less common than the use of drug placebos, in part because making incisions into the skin carries risks, such as infection or bleeding. Roughly 150 clinical trials begun this year included a sham procedure, compared with about 2,500 studies that used a drug placebo, according to a search of a clinical trials database run by the National Institutes of Health.
Wouldn’t the use of anesthesia be an issue too?
WebMd also covered the story and put more emphasis on the placebo effect:
THURSDAY, Dec. 26, 2013 (HealthDay News) — Improvements in knee pain following a common orthopedic procedure appear to be largely due to the placebo effect, a new Finnish study suggests.
Attributing their recovery to the placebo effect is an assumption that doesn’t appear to be supported by the evidence. What the study didn’t tell us is what would have happened had these patients not received either real or sham surgery. We can’t assume that patients who didn’t get either "treatment" would fail to improve over the course of a year.
More generally, just because subjects get a placebo in a trial and get better doesn’t necessarily prove the placebo caused the recovery, or that the placebo effect caused the recovery of the treatment group. The placebo effect is a powerful thing, but so is the body’s ability to heal. And sometimes there are other factors do consider.
There were two key questions I couldn’t find an answer for in any of the stories. First, what did existing research say on how often the type of knee pain in question does go away on its own? Some stories included information on a previous study showing that physical therapy worked as well as surgery. The second unanswered question: What was the post-operative procedure for patients in the study? Did the patients rest more or use ice or heat or physical therapy that they wouldn’t have used had they not undergone the fake/real procedures? I imagine if I thought I’d just had knee surgery I’d behave somewhat differently, and I’d expect some sort of post-op instructions from the docs. Seems whatever the post-op procedure was deserves a closer look.
[…] not just fake pills—placebo surgeries have also shown a surprising amount of […]