An arthroscopy (key hole surgery to wash out, trim or clean up) in the knee joint is one of the most common procedures in orthopaedics, and one of the most common surgical procedures overall, with over one million performed in the USA each year. An overview of arthroscopy for knee osteoarthritis can be found in a previous blog; this post covers the sham surgery trial as part of a series on that topic.
In 2002, researchers from the USA published the results of a placebo controlled, sham surgery study of arthroscopy in patients with knee pain and osteoarthritis. They were older patients, mostly with mechanical symptoms. 180 study participants were randomised to one of three treatment arms: an arthroscopy with removal of loose bodies and torn cartilage and menisci, an arthroscopy and washout only, and a placebo only (just with the incisions). All patients had similar scars, and they were blinded to what treatment they received. Even the researchers doing the follow up didn’t know what treatment group they were in, eliminating any bias when assessing the outcome.
Over two years, there was no difference between the groups in pain or function. Patients in all groups got better; they just got better by the same amount.
The fact that patients in all groups improved is crucial. Many surgical procedures are performed on the basis that many patients get better after the treatment, and cause-and-effect is assumed (post hoc ergo propter hoc). Only comparison to a placebo can tell us whether or not the treatment has a specific effect on the patient.
Understandably, the study was criticised by surgeons who felt that the results did not fit with their experience. It has been supported by other randomised trials since, and there is some evidence that the rates of knee arthroscopy for osteoarthritis fell slightly after this study was published, but there are still many being performed, and the rates in my country are increasing. Why? Failure to accept the evidence, patient and referrer expectations, financial incentives, the desire to do something when nothing else is effective, and also because it works. It just doesn’t work any better than placebo.