The rotator cuff is four muscles and their tendons wrapping the top of the upper arm bone, holding the ball centred in the socket while the larger muscles move the arm. A tear means one of those tendons has pulled partly or completely away from its attachment. What follows is the part most pages leave out: which tears actually need repairing, which do not, and how that decision gets made.
Two things are worth knowing before anything else, because they change how you should read your own scan.
First, a tear on imaging is a finding, not a diagnosis. Rotator cuff tears are found regularly in shoulders that do not hurt, and they turn up more often the older the shoulder is. Plenty of people are walking around with a tear they will never know about. If your MRI report says "rotator cuff tear," that sentence on its own does not tell you whether you need an operation.
Second, a full-thickness tear does not knit itself back together. Tendon does not bridge a gap the way a cut in skin closes. But that is not the same as saying you need surgery — a great many people become comfortable and functional without repair, because the shoulder compensates and the pain settles. Those two facts sit together uncomfortably, and the space between them is where the real decision lives.
What it actually feels like
Most people do not describe a tear. They describe a shoulder that has stopped cooperating:
- Night pain — the symptom that most often brings people in. Lying on that side becomes impossible, and the shoulder aches when the day's distractions are gone.
- Trouble reaching — overhead, out to the side, or behind the back. Seat belts, back pockets and top shelves are the everyday casualties.
- Weakness that is distinct from pain. Not "it hurts to lift this" but "the arm will not do it."
- A sudden event, in a minority. A fall, a catch, a hard pull, and immediate loss of strength — this pattern behaves differently from the slow degenerative kind and is treated more urgently.
The distinction between pain and weakness matters more than it sounds. A shoulder that is weak because it hurts often recovers with time and rehabilitation. A shoulder that is weak because the tendon is not attached is a different problem.
How the shoulder is examined here
The examination comes first, and it is not a formality. Which movements hurt, which are weak, whether the weakness is real or pain-limited, how the shoulder blade moves, whether the neck is contributing — these decide what the imaging means.
Ultrasound is used during the visit itself. That is not a substitute for an MRI and it is not in competition with one; the two answer different questions.
An MRI is a detailed static picture of a shoulder lying still inside a magnet. Ultrasound shows the cuff moving — in real time, under load, in the position that actually produces the pain. It allows the injured side to be compared directly against the other shoulder in the same visit. And when an injection is indicated, it turns the injection from a landmark-guided procedure into a directly visualised one, with the needle on screen the whole way in.
For a full-thickness tear where surgery is being considered, an MRI is still the study that shows tear size, retraction and the condition of the muscle. Both get used, for what each is good at.
Partial-thickness tears are a different problem
The single most useful thing to know about your scan is whether the tear goes all the way through the tendon.
A partial-thickness tear means the tendon is damaged and partly torn but still attached to the bone. There is no gap. The tissue is frayed, degenerative, or split through part of its depth — and the fibres that remain are still doing their job. These are far more common than most patients realise, they account for a large share of the shoulders that hurt for months without a clear injury, and they are the tears where biologic treatment has a real evidence base rather than a marketing claim.
A full-thickness tear means the tendon has pulled off completely and there is a gap between tendon and bone. That gap does not close on its own, and no injection closes it either.
The distinction matters because it changes the entire conversation. A partial tear is a tendon that can potentially be helped to heal. A full-thickness tear is a mechanical problem that either gets repaired or gets worked around. Providers who blur the two — advertising "heal your rotator cuff without surgery" without saying which kind of tear they mean — are the reason this field has a credibility problem.
Everything in the next section, including the eight-year outcome data from this practice, is about partial-thickness tears.
What the evidence supports, and what it does not
This field has a marketing problem, so it is worth separating what is advertised from what is actually published — including work from this practice.
The longest follow-up in the literature is Dr. Buford's
In July 2026 Dr. Buford published a pilot study of bone marrow aspirate concentrate injection for patients with symptomatic partial-thickness rotator cuff tears. At eight years it is the longest follow-up in the published literature with documented healing of the cuff.
The tear type matters, so it is worth stating plainly: these were partial-thickness tears — tendon that is damaged and partly torn but still attached to the bone. That is a different problem from a full-thickness tear, where the tendon has pulled off completely and there is a gap to close. Healing a partial tear is biologically plausible. Closing a full-thickness gap with an injection is not, and nothing below should be read as claiming it.
Over those eight years:
- 32% of patients fully healed their partial-thickness tears
- 48% had tears that were stable or smaller
- 20% ultimately came to surgery, at an average of three years after treatment
That deserves to be read honestly in both directions. Four out of five patients avoided an operation over eight years and a third healed the partial tear outright — a materially better picture than the blanket claim that biologics do nothing for a cuff. It is also a pilot study rather than a large randomised trial, and one patient in five still ended up in the operating room. Evidence worth taking seriously, not a guarantee.
What has not changed is the limit of the biology. A large, retracted, full-thickness tear sitting on an atrophied muscle is not going to be drawn back to the bone by an injection, and a provider suggesting otherwise is describing something no literature supports. The cases worth treating biologically sit well short of that.
Where biologics have the strongest support in this shoulder:
- Rotator cuff tendinopathy and partial-thickness tears — where the tendon is degenerative or partly torn but still attached. This is the honest indication, it is what the eight-year data above is drawn from, and it accounts for a large share of painful shoulders.
- Augmenting a surgical repair — adding biologic material at the repair site to support healing. An active area of research rather than a settled one, and it should be described that way.
- Buying time and function in a shoulder where surgery is not the right answer — because of the tear pattern, the tissue quality, or the patient's own circumstances.
If you have been diagnosed with rotator cuff tendinopathy or a partial-thickness tear, this is the specific situation that eight-year study was built around — and the conversation worth having before anyone books you an operation.
The options, all of them
A consultation here is not a pre-decided route to an injection or to an operation. There are four broad paths, and the honest answer is that all four are correct for somebody.
Rehabilitation and load management
The first-line answer for most degenerative tears, and a genuinely effective one. Work directed at the shoulder blade and the muscles that remain intact can restore a great deal of function around a tear that is never going to be repaired. The commonest reason it fails is that it was never really done.
Corticosteroid injection
Useful for breaking a pain cycle so that rehabilitation can happen. Worth understanding as a short-term measure rather than a treatment for the tear, and worth being cautious about repeating — steroid is not neutral for tendon tissue, particularly if surgery may follow.
Orthobiologic injection
Platelet-rich plasma or bone marrow aspirate concentrate, prepared in the office from your own tissue and placed under ultrasound guidance. Best supported for tendinopathy and partial-thickness tears. Not a substitute for repair when repair is what the shoulder needs.
Arthroscopic rotator cuff repair
Reattaching the tendon to bone through small incisions with a camera. The right answer for an acute traumatic tear in a functioning shoulder, for a tear causing weakness that matters to the patient's life, and for younger patients where leaving a tear to enlarge carries a real cost.
Not every tear is repairable. Longstanding tears with retracted tendon and atrophied muscle sometimes cannot be brought back to the bone, and pretending otherwise helps nobody. Those shoulders have their own set of options, and that conversation is a different one.
How the choice actually gets made
This is the part worth reading twice, because it is where practices genuinely differ.
The decision is not "tear, therefore repair." It is a weighing of the tear pattern and size, whether the onset was traumatic or degenerative, the quality of the tendon and the muscle, whether weakness is real or pain-driven, the age and demands of the patient, and — not least — what that particular person needs the shoulder to do.
Why a surgeon is the one saying this
Dr. Buford has been a shoulder surgeon for more than 26 years and has performed over 2,000 rotator cuff repairs. He also published the longest documented follow-up on biologic treatment of rotator cuff tears in the literature.
That combination is rare, and it is the whole argument of this page. Very few physicians can describe both routes from direct experience — most can only describe the one they perform.
It cuts in both directions, which is the point:
- A clinic that only injects has one answer to every shoulder. It cannot tell you that your tear needs repairing, because it cannot repair it.
- A surgeon who does not offer biologics has the opposite single answer, and a strong incentive to reach for it.
- Someone who does both has no structural reason to prefer either — which is what makes "an injection is unlikely to help your shoulder" or "you do not need the operation you were quoted" a recommendation you can weigh rather than a sales position.
Two thousand repairs is also what makes the honest negative useful. Knowing what a repaired cuff looks like at two years, and which tears do badly, is the same knowledge that identifies the shoulder that should never have been operated on in the first place.
Re-tear: what it means and what it does not
A repaired cuff can fail to heal, or heal and then tear again. It is worth understanding honestly rather than discovering later.
Re-tear is more likely with larger tears, with poorer tendon tissue, with longstanding tears where the muscle has already atrophied, and in people who smoke or have poorly controlled diabetes. Reported rates vary widely between published series, largely because studies define re-tear differently and image at different intervals — a number quoted without that context is not telling you much.
The part that surprises people: an imaging re-tear does not automatically mean a bad outcome. A proportion of patients whose repair shows a defect on a later scan are nonetheless substantially better than they were before surgery, with less pain and better function. The scan and the shoulder are not the same thing — which is where this page started.
The modifiable risks are worth naming because they are actionable. Smoking and blood sugar control are not incidental to tendon healing.
What recovery involves
Recovery after repair is longer than most people expect, and the reason is biological rather than surgical. The tendon has to heal to bone, and that healing cannot be hurried by working harder.
Broadly: a sling for several weeks, then a staged return of motion, and only after that any strengthening. Strength typically continues improving for months beyond the point at which the shoulder already feels much better. The specific timeline depends on tear size and tissue quality and is set at the pre-operative visit rather than improvised afterwards.
Recovery after an orthobiologic injection is a different scale entirely — soreness for a few days and a week away from significant loading. Full detail is on the PRP page. One rule applies to both: do not take ibuprofen, naproxen or other anti-inflammatories afterwards unless you have been told to. They work directly against the healing response.
Common questions
Does every rotator cuff tear need surgery?
No, and this is the single most misunderstood point about rotator cuff tears. Many tears are found on imaging in shoulders that do not hurt, and their frequency rises steadily with age. A tear on a scan is a finding, not a diagnosis. What decides treatment is the combination of the tear, the examination, how the shoulder actually functions, and what the patient needs it to do.
Can a rotator cuff tear heal on its own?
A full-thickness tear does not knit back together on its own. That is not the same as saying it cannot become comfortable and functional. Many people do well without repair, because the surrounding muscles compensate and the pain settles. Partial-thickness tears and tendinopathy are a different situation, and are where biologic treatment has the most support.
Will PRP heal my rotator cuff tear?
No injection available today has been shown to close a full-thickness rotator cuff tear. Any provider telling you otherwise is describing a result the published literature does not support. Where PRP has a genuine role is in tendinopathy and partial-thickness tears, and as an augment at the time of surgical repair. Those are real indications; regrowing a torn tendon across a gap is not.
Why use ultrasound if I have already had an MRI?
An MRI is a static scan of a shoulder lying still. Ultrasound shows the cuff moving, under load, in the position that actually hurts, and it allows the two shoulders to be compared in the same visit. It also makes any injection a directly visualised procedure rather than a landmark-guided one. The two studies answer different questions and are not in competition.
What is a rotator cuff re-tear?
It is a repair that fails to heal, or heals and then tears again. Re-tear is more common with larger tears, poorer tissue quality, longstanding tears with muscle atrophy, smoking and diabetes. Reported rates vary widely between studies because the definitions and imaging follow-up differ. A re-tear on imaging does not automatically mean a poor result: some patients with an imaging re-tear are still substantially better than before surgery.
How long is recovery after rotator cuff repair?
Longer than most people expect. A sling for several weeks, then a staged return of motion before any strengthening, because the repair has to biologically heal to bone before it can be loaded. Meaningful strength typically continues improving for months afterwards. The timeline depends on tear size and tissue quality and is set at the pre-operative visit, not guessed at afterwards.
Who performs the injection and the surgery?
Don Buford, M.D., a board-certified orthopedic surgeon, performs both personally. Injections are placed under direct ultrasound guidance. Because the same physician offers the injection and the operation, the recommendation is not constrained by what happens to be on the menu.
Request an appointment or call (877) 777-8883.
