You have a degenerative meniscal tear on MRI, a surgeon has offered a quick clean-out, and nobody has told you what that operation actually buys you compared with the alternative. That frustration is legitimate, and it is also the exact place where the evidence gets interesting — because the best sham-controlled trials of arthroscopic partial meniscectomy (APM) for degenerative tears found that the operation performs no better than a placebo incision, which means the decision is not “surgery or nothing” but “what does a structured conservative trial look like, and what findings would justify escalating?”
What the sham-controlled evidence actually shows
The trial most people mean when they say “the sham meniscus study” is the Finnish multicenter trial led by Raine Sihvonen, published in the New England Journal of Medicine in 2013. It randomized 146 patients aged 35 to 65 with knee symptoms consistent with a degenerative medial meniscus tear and no knee osteoarthritis to either arthroscopic partial meniscectomy or sham surgery. Both groups got incisions and the same postoperative care; only one group had meniscal tissue removed. The authors reported that “the outcomes after arthroscopic partial meniscectomy were no better than those after a sham surgical procedure” (Sihvonen et al., NEJM 2013;369:2515–2524).
The numbers matter more than the headline. At 12 months, the Lysholm score improved by 21.7 points in the APM group and 23.3 points in the sham group — a between-group difference of −1.6 points (95% CI, −7.2 to 4.0). The WOMET score improved by 24.6 and 27.1 points respectively (difference −2.5; 95% CI, −9.2 to 4.1). Knee pain after exercise improved by 3.1 and 3.3 points on a 0–10 scale (difference −0.1; 95% CI, −0.9 to 0.7). There were no significant between-group differences in any primary outcome, and no significant differences in subsequent knee surgery or serious adverse events.
That trial is not an outlier. The 2022 Cochrane review by O’Connor and colleagues pooled 16 randomized trials with 2,105 participants and found high-certainty evidence that arthroscopic surgery leads to little or no difference in pain or function at three months after surgery compared with placebo surgery. In the main comparison, mean postoperative pain was 35.5 points in the arthroscopy group versus 40.1 points in the placebo group on a 0–100 scale (lower is less pain), a difference of 4.6 points (95% CI, 0.02 to 9 better). Function differed by 0.1 points (95% CI, −3.2 to 3.4). The authors concluded that arthroscopic surgery “provides little or no clinically important benefit in pain or function, probably does not provide clinically important benefits in knee-specific quality of life, and may not improve treatment success compared with a placebo procedure.”
One more line from that review is worth sitting with: “Arthroscopic knee surgery remains a common treatment for symptomatic knee osteoarthritis, including for degenerative meniscal tears, despite guidelines strongly recommending against its use.” That is the Cochrane authors’ framing, not mine, and it tells you the gap between evidence and practice is real.
What the evidence does not say
Three boundaries matter, because overreading these trials is as unhelpful as ignoring them.
It does not say no one benefits. A 4.6-point mean difference on a 0–100 pain scale is a group average. For any single person, that is indistinguishable from measurement noise. Some individuals in the sham groups improved substantially; some in the surgery groups did not. The trial tells you the average effect of removing meniscal tissue is small, not that your knee is average.
It does not cover traumatic or locked meniscal tears. The Sihvonen trial enrolled patients without knee osteoarthritis and with symptoms consistent with a degenerative medial meniscus tear. A true mechanical block — the knee that will not fully extend, or that catches and gives way with a large effusion — is a different clinical problem and was not the population studied.
It does not compare surgery against a well-dosed exercise program. The sham-controlled trials answer “does the surgery work better than a placebo incision?” They do not answer “does the surgery work better than a 12-week progressive loading program?” Those are different questions, and the second is the one a self-directed reader actually faces.
The honest gap in the conservative-care case
It would be tidy to say exercise has been proven equivalent to surgery. The evidence does not support that tidy claim. The Cochrane review found that the majority of trials without a placebo control were susceptible to bias — performance bias in 75%, detection bias in 75% — while the placebo-controlled trials were less susceptible and none were at risk of performance or detection bias. So the cleanest evidence is about surgery versus sham, not surgery versus exercise.
A 2022 systematic review of systematic reviews in Musculoskeletal Surgery concluded that “conservative treatment based on physical therapy should be the first-line management,” while also noting that most reviews revealed subgroups of patients who fail to improve after conservative treatment and find relief with surgery. That is the honest position: conservative care first is a reasonable default, not a guarantee, and the reviews themselves flag that some people do not respond.
What follows is my practical recommendation, not a trial finding. No retrieved source tested the specific protocol below, and I am not presenting it as validated. It is a way to make your own conservative trial informative rather than vague.
Why “rest and see how it feels” produces an uninterpretable result
Most people who “try conservative care” do something undosed: they rest, they avoid stairs, they take anti-inflammatories when it flares, and they wait. Eight weeks later they have no idea whether they improved, because they never measured anything and never applied a progressive load. That is not a conservative trial. It is a delay, and it pushes people toward surgery by default rather than by evidence.
A conservative trial is only informative if it is dosed and progressed like a training program. That means sets, reps, load, frequency, and a pre-written rule for what counts as failure.
A phase-based conservative trial you can actually run
This is a template, not a prescription for your specific knee. If you have a true mechanical block, inability to bear weight, or a hot swollen joint, skip this section and get assessed. For the typical degenerative tear with intermittent pain and no locking, the structure below gives you something measurable.
Phase 1: calm the symptoms and load the quadriceps isometrically (weeks 1–2)
The goal is to keep the quadriceps active without provoking the joint. Isometric contractions are useful here because they generate force with minimal joint movement.
- Quad sets: sit with the knee straight, press the back of the knee down into a rolled towel, hold 10 seconds. 10 reps, 3 times per day.
- Straight-leg raise: lie on your back, knee straight, lift the leg 12–18 inches, hold 3 seconds, lower over 3 seconds. 3 sets of 10, once daily.
- Wall sit at a shallow angle: back against a wall, knees bent to about 30 degrees, hold 20–30 seconds. 5 holds, once daily. Stop if pain rises above 3/10 during the hold.
- Walking: as tolerated, aiming for a consistent daily step count rather than a heroic single walk. Use your phone or a pedometer to record it.
Progression rule for phase 1: move to phase 2 when you can do the full phase 1 session with pain no higher than 3/10 during the exercise and no higher than 3/10 24 hours later, on two consecutive sessions.
Phase 2: progressive band and split-squat loading (weeks 3–8)
This is where the actual training stimulus lives. You need a resistance band and a way to measure load — a bathroom scale works for split squats if you do not have weights.
- Band terminal knee extension: anchor a band behind you at knee height, loop it around the back of your knee, step forward to create tension, and straighten the knee against the band. 3 sets of 12–15, slow on the return. Progress by stepping further from the anchor or using a thicker band.
- Split squat: stand with one foot forward, lower until the front knee is bent about 45–60 degrees, then push back up. 3 sets of 8 per leg. If bodyweight is too easy, hold a dumbbell or wear a backpack with books. If it is too hard, hold a doorframe for balance and reduce the depth.
- Step-down: stand on a step about 6 inches high, lower the other heel toward the floor slowly, then push back up. 3 sets of 8 per leg. This is the single best exercise for building the eccentric quadriceps control that walking downhill and descending stairs demand.
- Single-leg balance: stand on the affected leg, hold 30 seconds, 3 times. Progress by closing your eyes or standing on a folded towel.
Frequency: three sessions per week, with at least one rest day between sessions. Load progression: add one set or a small amount of external load only when you complete all prescribed sets and reps with pain no higher than 3/10 during and 24 hours after.
Phase 3: return-to-activity criteria (weeks 9–12 and beyond)
Do not progress by calendar. Progress by criteria. You are ready to increase impact when you can do all of the following:
- 30 single-leg sit-to-stands from a standard chair in under 60 seconds, with pain no higher than 2/10.
- Descend a flight of stairs step-over-step without a handrail and without a pain flare the next day.
- Complete a 30-minute walk at a brisk pace with pain no higher than 2/10 during and 24 hours after.
- Quadriceps girth on the affected side within 1 cm of the unaffected side, measured with a tape measure 10 cm above the top of the kneecap.
If you meet those criteria, you can begin adding impact — light jogging on flat ground, then hills, then sport-specific work — using the same 24-hour pain rule.
What to measure weekly, and what the pattern means
Pick three measures and record them on the same day each week. The point is not to collect data for its own sake; it is to make your own trial interpretable.
- Pain 0–10 at 24 hours after your hardest session. This is your primary outcome. A change of 2 points or more on a 0–10 scale is generally considered clinically meaningful for an individual, though that threshold is a rule of thumb, not a trial-derived number for this exact population.
- Single-leg sit-to-stand reps in 60 seconds. This is your function measure. It is objective, cheap, and sensitive to quadriceps strength changes.
- Quadriceps girth with a tape measure, 10 cm above the kneecap. This is your structural measure. It changes slowly, so do not panic over one week.
Interpretation: if your 24-hour pain is trending down and your sit-to-stand reps are trending up over three to four weeks, keep going. If nothing has changed in four weeks despite adherence, you have two options — change the load (more resistance, more range, more frequency) or reconsider the diagnosis. If pain is rising week over week despite adherence, that is a signal to get reassessed, not to push through.
When to escalate, and how to have that conversation
Escalation should be triggered by a pattern you defined in advance, not by a bad day. Before you start, write down your failure rule. A reasonable one: “If my 24-hour pain has not improved by at least 2 points and my sit-to-stand reps have not improved by at least 3, after 8 weeks of adherent phase 2 training, I will seek reassessment.”
Red flags that warrant earlier reassessment, not a wait-and-see approach:
- True locking — the knee catches and you cannot fully straighten it.
- Giving-way with a large effusion that develops over hours after a specific twist.
- Inability to bear weight.
- Night pain that wakes you, or unexplained swelling without a clear mechanical trigger.
If you do end up in a surgeon’s office after a genuine conservative trial, you are in a much stronger position than the person who arrives with an MRI and no data. You can say: “I did eight weeks of progressive loading, my 24-hour pain went from 6 to 4, my sit-to-stand went from 12 to 18, but I have plateaued and I still cannot descend stairs without a flare. What are my options?” That is a different conversation from “my knee hurts and the MRI shows a tear.”
The point is not to avoid surgery on principle
The sham-controlled evidence says that for the average patient with a degenerative meniscal tear and no osteoarthritis, removing meniscal tissue does not produce a clinically important benefit over a placebo incision. That is a strong finding, and it should change the default. But the same evidence does not tell you that you are average, and the review of reviews explicitly notes that some people fail conservative care and improve after surgery.
The goal is to make the decision with your own data in hand rather than someone else’s average. Run the conservative trial properly — dosed, measured, with a pre-written failure rule — and then decide. If you improve, you have your answer. If you do not, you have a much more specific question for a surgeon, and a much better basis for weighing the risks of an operation that the best available evidence says performs about as well as a sham.
FAQ
Does this mean I should never have meniscus surgery?
No. The trials enrolled people with degenerative tears and no knee osteoarthritis. They do not cover traumatic tears, true locking, or knees with significant osteoarthritis. And the reviews describe subgroups who fail conservative care and improve after surgery. The evidence supports changing the default, not banning the operation.
How long should I give a conservative program before deciding?
Eight to twelve weeks of dosed, progressive loading is a reasonable window, provided you are actually measuring something. If you are not measuring, you are not running a trial — you are waiting. The Cochrane review’s main comparison was at three months, which is a useful anchor for when to reassess.
What if my MRI shows a big tear?
The sham-controlled trials found no average benefit from surgery in this population regardless of tear findings, and the Cochrane review found no clinically important benefit over placebo — so the decision should rest on symptoms and response to conservative care, not on tear size alone. If you have a true mechanical block, that is a different situation and warrants prompt assessment.
Is it safe to load a knee with a degenerative meniscal tear?
The Cochrane review included eight trials comparing surgery with exercise, but its harms analysis focused on surgery; it does not establish that exercise is risk-free, so apply the 24-hour pain rule and reduce load if symptoms rise. The practical safety rule is the 24-hour pain response: if pain during exercise stays at or below 3/10 and returns to baseline within 24 hours, the load is tolerable. If it rises and stays up, reduce the load.
What does a 4.6-point difference on a 0–100 pain scale mean for me?
It means the average difference between surgery and sham is small — smaller than most people would notice. For an individual, that difference is within the range of day-to-day fluctuation. This is exactly why your own serial measurements matter more than the group mean.
Where can I read the primary sources?
The Sihvonen trial is published in the New England Journal of Medicine (2013;369:2515–2524). The 2022 Cochrane review is O’Connor D, Johnston RV, Brignardello-Petersen R, Poolman RW, Cyril S, Vandvik PO, Buchbinder R. Cochrane Database of Systematic Reviews 2022, Issue 3, Art. No.: CD014328. The review of reviews is in Musculoskeletal Surgery (2023;107:127–141).