The orthopedic surgeons won’t be happy about this at all. The most common orthopedic procedure is a sham….or should I say no better than a sham surgery. A study published in the New England Journal of Medicine on December 26th should change the current management of meniscal tear.
Sihvonen and colleagues randomized 70 Finnish patients without knee osteoarthritis to arthroscopic partial meniscectomy and 76 patients to a sham operation. These patients had failed at least 3 months of conventional conservative treatment. Patients with traumatic onset of symptoms or with osteoarthritis were excluded. The authors did something very interesting which both allowed them to more easily perform a sham operation and to get around the ethics of a sham operation: they did a diagnostic arthroscopy and it was during this diagnostic arthroscopy that the patient was randomized. At this point either a standard meniscectomy was performed or an elaborate sham. Everything was done the same postoperatively as far as wound care, rehab instructions, etc. The patients, those who determined the outcomes, and those who collected and analyzed the data were all blinded to study group assignment. The main outcome measure was knee pain after exercise (at 2, 6, and 12months) using a validated scale and a validated meniscus specific quality of life instrument. They also asked patients at 12 months if they would be operated on again and if they had figured out which arm of the trial they were in.
Prior to looking at the results of this study we need to make sure the study is scientifically valid. Therapeutic studies should meet the following criteria:
- Were participants randomized? YES
- Was random allocation method concealed? YES the authors used opaque envelopes
- Was intention to treat analysis used? YES
- Was the groups similar at the start of the study? YES
- Was blinding adequate? YES (only the operating room staff weren’t blinded but they didn’t participate in outcome determination)
- Were the groups treated equally apart from the intervention? YES
- Was follow-up sufficiently long and complete? YES
So I think this study is low risk for bias and I can now move to the results.
This figure shows that the 2 treatment arms had the same effect on the validated measures and the knee pain likert scale. 83% of sham surgery patients reported improvement in knee pain compared to 88.6% of meniscectomy patients. Furthermore, 96 and 93% of patients respectively, reported they would repeat the same procedure they had undergone. 5% of sham surgery patients underwent additional arthroscopy compared to 1.4% of meniscectomy patients (p=ns).
While this study is small it was adequately powered for their outcomes. One could argue the ethics of a sham operation but this methodology is the only way to truly determine if some procedures “work”. Even procedures that don’t really work can have a significant placebo effect and a sham study is a powerful way to control for that effect.
It’s important to point out that these results only apply to patients with non-traumatic degenerative meniscal tears. Interestingly, the authors did a post-hoc subgroup analysis showing that patients with sudden onset of symptoms didn’t have any different outcomes with meniscectomy over sham surgery. There are major limitations with any post-hoc analysis but suggests that now a study needs to be done in a similar fashion in traumatic injury patients.