“You have a rotator cuff tear.” For a lot of people, those words land like a surgical sentence — a tear is torn, torn things need fixing, so book the operating room. It is one of the most common conversations I have in clinic, and the starting point is almost always the same: the MRI report rarely decides it by itself. A torn rotator cuff is extraordinarily common, especially past middle age, and a large share of the people walking around with one have no idea it's there.
So the real question isn't “is it torn?” It's “is this tear the kind that does better with surgery, or the kind that does just as well with a good rehab program?” The evidence on that question is unusually rich — and unusually split. Two of the best randomized trials we have point in opposite directions, and the American Academy of Orthopaedic Surgeons (AAOS) published an updated clinical practice guideline in August 2025 that, read carefully, refuses to declare a single winner.1 This article walks through what those trials actually found, where they disagree, and how a working shoulder surgeon uses them to make a real recommendation for a real shoulder.
First, what kind of tear are we talking about?
Before any surgery-or-not conversation makes sense, one distinction has to be nailed down, because it changes everything that follows.
Traumatic tears
A traumatic tear happens in a specific moment — a fall onto an outstretched hand, a hard yank on the arm, a shoulder that dislocated. There was a “before” when the shoulder worked and an “after” when it didn't, and you can usually name the day. These tears tend to happen in tendon that was reasonably healthy the moment before it tore, which matters enormously for whether a repair will hold.
Degenerative tears
A degenerative tear is wear. The cuff frays and thins over years, the way a rope frays, and at some point a strand gives way. There's no single injury — the ache came on gradually, worse at night, worse reaching overhead. On MRI the tendon and the muscle behind it often already show age: thinning, and sometimes fatty infiltration, which is fat replacing muscle in a tendon that has been failing for a while. Most rotator cuff tears in adults over 60 are this kind.
The trials below studied degenerative tears almost exclusively. That's the population where the surgery-versus-rehab debate is genuinely live. For a young patient with an acute traumatic tear, the calculus is different — and I'll come back to that.
The two trials that disagree
If you only read one summary of the rotator cuff literature, you'd walk away confused, because two excellent randomized trials reached opposite headlines. Both are worth understanding, because the disagreement isn't sloppiness — it's the two trials studying different shoulders.
Moosmayer: at 10 and 15 years, repair pulled ahead
A Norwegian team led by Stefan Moosmayer randomized 103 patients with small-to-medium tears (no larger than 3 cm) to either primary tendon repair or a structured physiotherapy program, with the option to cross over to surgery later if therapy failed. They then followed these patients for a decade and a half — one of the longest randomized follow-ups in all of shoulder surgery.
At 10 years, the surgical group was doing measurably better: about 9.6 points higher on the Constant score (a 0–100 shoulder rating), roughly 16 points better on the ASES score, and about 1.8 cm less pain on a 10 cm pain scale.2 At 15 years the gap hadn't closed — it had widened slightly, to about 11.8 Constant points in favor of repair.3 Two details stand out. First, in the patients who were managed with therapy and never repaired, the average tear grew from about 16 mm to about 32 mm over the study — it roughly doubled. Second, the 14 patients who started with therapy and later crossed over to surgery ended up about 10 Constant points worse than those repaired up front. The authors' takeaway: for this specific tear, repairing it early beat waiting.
Kukkonen: for small tears in older shoulders, no difference
A Finnish team led by Juha Kukkonen ran a different randomized trial — 180 shoulders, all with small, non-traumatic supraspinatus tears (one tendon), all in patients older than 55. They compared physiotherapy alone against two surgical options. At the 1-, 2-, and now 5-plus-year marks, the groups landed in essentially the same place: no statistically significant advantage for surgery, and — notably — operating did not protect the joint from degenerating over time.4 For the older patient with a small single-tendon tear and no injury, this trial says a good rehab program is a legitimate destination, not just a stop on the way to the OR.
Why they disagree — and why that's useful
Put the two side by side and the “contradiction” mostly dissolves. Moosmayer studied a slightly younger group (down to the 50s) with tears up to 3 cm and found repair wins over 10–15 years. Kukkonen studied older patients (all over 55) with only small single-tendon tears and found a tie at 5 years. The lesson isn't “surgery works” or “surgery doesn't.” It's that the answer slides with age, tear size, and how much time you're measuring over. Younger shoulder, bigger tear, longer horizon → the scale tips toward repair. Older shoulder, small tear, pain relief as the goal → the scale tips toward a serious trial of rehab first.
What happens if I just wait?
This is the question therapy-first patients most deserve a straight answer to, because “let's try rehab” is only fair if we also watch what the tear is doing. The best natural-history data we have comes from a prospective study by Keener and colleagues that tracked degenerative cuff tears over years.5 Two findings matter for your decision. First, tears tend to enlarge over time, and the risk is higher for full-thickness tears than for partial ones. Second — and this is the part that should shape the plan — tear enlargement was linked to the muscle degenerating and to previously painless shoulders becoming painful. In other words, waiting isn't free: a tear can quietly get bigger and the muscle behind it can quietly go to fat.
That last point connects to a classic piece of shoulder science: the Goutallier grade, a 0–4 scale for how much fat has replaced the muscle behind the tendon.6 Once fatty infiltration is advanced (grade 3–4), a repair is far less likely to heal and hold, and that change is largely a one-way street — it doesn't reverse after surgery. So part of “can I wait?” is really “can I wait without letting the tear cross from repairable into not-worth-repairing?” For most small, stable tears the answer is comfortably yes. For a larger tear in a younger, active person, the clock is more real.
What the 2025 AAOS guideline actually says
The updated AAOS guideline is refreshingly honest about the split. For symptomatic small-to-medium full-thickness tears, it gives a strong recommendation that either surgery or physical therapy meaningfully improves patient-reported outcomes — both are legitimate first moves.1 It notes, at moderate strength, that when a repair actually heals, those patients tend to do better than with therapy alone. And it flags the caveat that sits underneath the whole debate: with therapy alone, tear size, muscle atrophy, and fatty infiltration can progress over five to ten years. It does not tell you that everyone needs surgery, and it does not tell you that surgery is pointless. It tells you the decision is yours to make with a surgeon who knows your specific shoulder.
What I weigh in clinic
When a patient hands me an MRI report that says “rotator cuff tear,” I'm running through a sequence out loud with them. Here's the actual reasoning.
1. Did this happen in a moment, or over years?
A genuine traumatic tear in a shoulder that worked fine the week before moves me toward earlier repair — the tendon is usually healthy enough to heal, and repairing it promptly protects the muscle before it can degenerate. A gradual, degenerative ache buys us room to try rehab first.
2. How old is the shoulder, and what do you need it to do?
A 52-year-old electrician who works overhead all day and a 74-year-old who wants to sleep and lift a coffee cup are not the same decision, even with an identical MRI. The Kukkonen trial speaks directly to the older, lower-demand shoulder; the Moosmayer trial speaks more to the younger, higher-demand one. I try to match your shoulder to the trial that actually studied it.
3. How big is the tear, and is it changing?
Small, single-tendon, stable tears are the ones where rehab holds up well in the data. For a bigger tear, or one we can see enlarging on repeat imaging, I'm more inclined to repair before it moves out of the repairable zone.
4. What does the muscle look like?
If the muscle behind the tendon is still healthy, a repair has a real chance of healing. If it's already largely fat (advanced Goutallier grade), a standard repair is much less likely to hold, and we talk plainly about whether the goal should be pain relief and function through other means rather than chasing a repair that the biology won't support.
5. Have we given rehab a real, supervised trial?
“I tried some exercises” and “I did twelve weeks of a structured, progressive, supervised program” are different things, and only the second one counts as a fair test. For the degenerative tears where rehab is a reasonable first move, that's usually where we start — and if you're not turning the corner, the conversation about surgery is then an informed one rather than a reflex.
None of this is a formula. Two of those five answers can point one way and three the other, and the recommendation lives in the weighing. What I try never to do is let a single line on a radiology report — “full-thickness tear” — make the decision that a full picture of your shoulder should make.
The bottom line for the layperson
- A rotator cuff tear on MRI is common and does not automatically mean you need surgery — plenty of people have one and never need an operation.
- The kind of tear matters most: a sudden, traumatic tear in a younger, active shoulder leans toward earlier repair; a gradual, degenerative tear in an older shoulder is often reasonable to rehab first.
- The best long-term trial (15 years) found repair pulled ahead for tears up to 3 cm — while a separate trial found no difference for small tears in patients over 55. Both are right, for different shoulders.
- Waiting isn't free: tears can enlarge and the muscle can turn to fat, and past a point a repair may no longer hold — so if you wait, the tear should be watched, not ignored.
- The 2025 AAOS guideline says surgery and physical therapy are both legitimate starting points for most small-to-medium tears. The right choice depends on your age, your tear, your muscle quality, and what you need your shoulder to do — decide it with a surgeon, not from the MRI report alone.
If you've been told you have a rotator cuff tear and you're not sure whether it needs fixing, that uncertainty is reasonable — the evidence itself is split, and the answer really does depend on the specifics of your shoulder. You can read more about how rotator cuff repair works, see what the recovery looks like in our piece on sleep after rotator cuff surgery, or request a visit to review your imaging together.
Sources
- American Academy of Orthopaedic Surgeons. “Management of Rotator Cuff Injuries: Evidence-Based Clinical Practice Guideline.” Adopted August 2025. AAOS CPG (PDF)
- Moosmayer S, Lund G, Seljom US, et al. “At a 10-Year Follow-up, Tendon Repair Is Superior to Physiotherapy in the Treatment of Small and Medium-Sized Rotator Cuff Tears.” J Bone Joint Surg Am. 2019;101(12):1050–1060. DOI: 10.2106/JBJS.18.01373
- Moosmayer S, Lund G, Seljom US, et al. “Fifteen-Year Results of a Comparative Analysis of Tendon Repair Versus Physiotherapy for Small-to-Medium-Sized Rotator Cuff Tears.” J Bone Joint Surg Am. 2024;106(19):1785–1796. DOI: 10.2106/JBJS.24.00065
- Kukkonen J, Ryösä A, Joukainen A, et al. “Operative versus conservative treatment of small, nontraumatic supraspinatus tears in patients older than 55 years: over 5-year follow-up of a randomized controlled trial.” J Shoulder Elbow Surg. 2021;30(11):2455–2464. DOI: 10.1016/j.jse.2021.03.133
- Keener JD, Galatz LM, Teefey SA, et al. “A Prospective Evaluation of Survivorship of Asymptomatic Degenerative Rotator Cuff Tears.” J Bone Joint Surg Am. 2015;97(2):89–98. DOI: 10.2106/JBJS.N.00099
- Goutallier D, Postel JM, Bernageau J, et al. “Fatty muscle degeneration in cuff ruptures. Pre- and postoperative evaluation by CT scan.” Clin Orthop Relat Res. 1994;(304):78–83. PMID: 8020238
This article reflects Dr. Dewan's reading of the cited evidence at the time of publication. It is educational content, not medical advice. Your specific case may differ — schedule a consultation to discuss your imaging and history.