Rotator cuff pain is a symptom, not a diagnosis. It is the shoulder that hurts reaching overhead, that aches through the night, that has quietly stopped doing things it used to do without being asked — usually with no single moment of injury to point at.
Before anyone attaches a label to it, one thing is worth knowing: the labels you are likely to be given overlap far more than they differ. That single fact changes how you should read whatever you get told next.
Four names for mostly one problem
Depending on who you see, the same shoulder can come home with any of these:
- Shoulder impingement or subacromial impingement syndrome — the idea that the cuff tendon is being pinched in the narrow space beneath the acromion, the bony roof of the shoulder.
- Rotator cuff bursitis or subacromial bursitis — inflammation of the bursa, the thin lubricating cushion sitting between that roof and the tendon.
- Rotator cuff tendinitis or tendinopathy — the problem located in the substance of the tendon itself.
Here is the part that rarely gets said out loud. Put an ultrasound probe on one of these shoulders and you will usually find features of more than one at the same time: a thickened bursa and a degenerative tendon, in a space that narrows when the arm comes up. The initial treatment is largely the same whichever name is used. And the label you were given frequently reflects the vocabulary of the person who examined you rather than a distinction they could actually demonstrate.
Shoulder specialists have been converging on a single umbrella term — subacromial pain syndrome — for precisely this reason. The older names survive partly out of habit and partly because separate names make for separate pages, separate visits and separate procedure codes.
One genuine exception. Everything above happens on the outside of the shoulder, under the bony roof. There is a separate problem called internal impingement that happens on the inside of the joint, in throwing athletes at the extreme of the cocking position. It is a different mechanism with different treatment, and it is covered on the SLAP tear page.
This matters to you for one practical reason: if you are told the diagnosis has changed from bursitis to impingement, very little has actually changed. The questions that decide your treatment are different ones.
What it actually feels like
The pattern is recognisable, and it is more useful than the name:
- Night pain. The symptom that brings most people in. Lying on that side becomes impossible, and the shoulder aches once the day's distractions are gone.
- A painful arc. Raising the arm out to the side hurts most through the middle of the range, roughly shoulder height, and often eases again above it.
- Pain in the outer upper arm, not on the top of the shoulder. People point to a spot halfway down the arm. Pain on the top of the shoulder points somewhere else entirely.
- Trouble reaching. Overhead, out to the side, or behind the back. Seat belts, back pockets and top shelves are the usual culprits.
- Weakness that is really pain. The arm gives way because it hurts, not because the tendon has failed. Telling those two apart is an examination finding, and it is one of the things that decides what happens next.
What it might not be
This is the section most pages skip, and it is the one that changes outcomes. A shoulder labelled impingement that is not improving is quite often a different problem wearing the wrong name:
- Frozen shoulder (adhesive capsulitis). The most consequential miss. The giveaway is loss of passive external rotation — someone else turns your arm outward and it will not go, even relaxed. Frozen shoulders get mislabelled as impingement for months, and the treatment is not the same.
- The AC joint. Pain on the top of the shoulder, tender to a fingertip on the joint itself, worst reaching across the body. Different structure, different injection, different answer.
- The neck. Cervical spine problems refer pain into the shoulder and upper arm convincingly. Pain travelling below the elbow, pins and needles, or symptoms that shift with neck position should redirect the examination.
- The biceps tendon. Pain at the front of the shoulder, often in company with cuff problems rather than instead of them.
- Calcific tendinitis. A calcium deposit inside the tendon. The presentation is unmistakable when it flares — severe, sudden, out of all proportion — and it has its own treatment path, including drainage under ultrasound.
- Glenohumeral arthritis. Stiffness and grinding rather than a painful arc.
- Underlying instability in younger and throwing athletes, where a loose shoulder produces impingement symptoms secondarily. Treating the impingement and ignoring the instability does not work.
None of this is exotic. It is the ordinary differential for a painful shoulder, and working through it is the difference between treating your shoulder and treating a label.
How the shoulder is examined here
History and examination first, because they narrow the field faster than any scan. Then diagnostic ultrasound in the same visit, performed by the physician who is going to treat you.
Ultrasound suits this problem particularly well. It shows the cuff and bursa moving, in real time, under load, in the position that actually produces the pain — which is exactly what a static scan of a shoulder lying still cannot do. It shows bursal thickening, tendon quality, calcific deposits, the AC joint and the biceps tendon in one sitting, and it allows the painful shoulder to be compared directly against the other one.
An X-ray adds the bony picture where that is relevant. An MRI earns its place when surgery is genuinely being considered, when there has been real trauma, or when the story and the examination do not agree. Ordering an MRI first on an ordinary painful shoulder tends to generate findings rather than answers, because changes seen on MRI are also common in shoulders that do not hurt.
What actually works, in order
Loading and rehabilitation
First, and not a consolation prize. This is where most of these shoulders get better, and the evidence for it is stronger than for anything that follows. It is also not rest — tendon responds to graded load, and a shoulder protected from all load usually gets worse. A programme worth doing addresses how the shoulder blade moves, restores the range that has quietly been lost, and then progressively loads the cuff. Expect weeks, not days.
Ultrasound-guided corticosteroid injection
A useful tool with honest limits. Well placed into the bursa it can produce real short-term relief, and its best use is to break a pain cycle so that rehabilitation can actually be done. The benefit tends to fade, repeated injections into the same shoulder show diminishing returns, and steroid delivered into tendon rather than around it is not neutral for that tendon. Placing it under direct ultrasound guidance rather than by surface landmarks is how you make sure it goes where it is meant to go.
Orthobiologic injection
Where the tendon itself is the problem — tendinopathy, or partial tearing within the tendon — platelet-rich plasma has a better evidence base than most of what it is advertised for. It is not a shortcut past the rehabilitation, it works alongside it, and the response builds over weeks. It is also the wrong tool entirely if the real problem turns out to be the neck, the AC joint or a frozen shoulder, which is why the diagnosis has to come first.
Surgery
Occasionally the right answer, and less often than the referral pattern suggests. See below.
The operation you may be offered, and what the evidence says
The standard operation for shoulder impingement is arthroscopic subacromial decompression — shaving bone from the underside of the acromion to open up the space the tendon runs through. It has been performed in very large numbers for decades.
It has been tested against placebo surgery, and it did not win. In randomised trials published in 2017 and 2018, patients were assigned either to a real decompression or to a placebo arthroscopy — the same anaesthetic, the same incisions, the same recovery, with nothing removed. Both trials found that decompression produced no meaningful advantage over the placebo operation.
That is about as strong a test as surgery ever gets, and it deserves to be reported plainly rather than quietly left off the page.
What it does not mean is that shoulder surgery never helps. A genuine structural tear, a calcific deposit that will not resolve, a frozen shoulder that has stopped responding, advanced arthritis — these are different problems with different answers, and some of them are operations worth having. What the evidence says is narrower and more specific: for shoulder pain labelled impingement, with no structural lesion behind it, shaving the acromion is not the answer it was once assumed to be.
When this becomes a tear conversation
Rotator cuff pain and a rotator cuff tear sit on the same spectrum, and the line between them is not always obvious at the first visit. The features that move the conversation are true weakness rather than pain-limited weakness, a definite injury rather than a gradual onset, night pain that escalates instead of settling, and a shoulder that has genuinely not shifted after a proper loading programme.
If that is closer to your situation, the detail is on the rotator cuff tear page — including eight-year outcome data from this practice on treating partial-thickness tears without surgery.
Why a surgeon is the one telling you this
Don Buford, M.D. is a board-certified orthopedic surgeon with more than 26 years in practice and over 2,000 rotator cuff repairs performed. He also performs the ultrasound and the injections personally.
That combination is the reason this page reads the way it does. Consider who is usually giving you the advice:
- A clinic that only offers injections has one answer available, and every shoulder tends to become a candidate for it.
- A surgeon who does not offer biologics has a different single answer, and a natural threshold for reaching it.
- Someone who does both has no structural reason to steer you toward either — which is what makes "this does not need an operation" a recommendation worth something.
Telling you that the commonest operation for your diagnosis does not beat placebo is information given against interest. That is rather the point.
Common questions
Is rotator cuff pain the same as shoulder impingement?
Largely, yes. Impingement, subacromial impingement syndrome, rotator cuff bursitis and rotator cuff tendinopathy describe overlapping territory rather than four separate diseases, and most shoulders labelled with one show features of the others on ultrasound. Much of the shoulder literature has moved toward a single umbrella term, subacromial pain syndrome, for exactly this reason. Which label you were given often reflects who examined you more than what is actually going on in the shoulder.
How long does rotator cuff pain take to settle?
Most shoulders improve over weeks to a few months with the right loading programme, and that timeline frustrates people who expect an injection to fix it in days. Tendon responds to graded load, and load takes time. What matters more than the calendar is the direction of travel: a shoulder that is slowly improving at six weeks usually keeps improving, and a shoulder that is completely unchanged at three months needs the diagnosis revisited rather than the same programme repeated.
Should I get an MRI for shoulder pain?
Not usually as the first step. An MRI on a painful shoulder frequently finds changes that are also present in shoulders that do not hurt, and an incidental finding can send treatment in the wrong direction. Examination plus ultrasound in the same visit answers most questions, because it shows the cuff moving under load rather than lying still. An MRI earns its place when surgery is genuinely on the table, when there has been real trauma, or when the picture does not add up.
Do cortisone shots damage the rotator cuff?
A well-placed injection into the bursa is a useful tool and not something to be afraid of. The concerns are about repetition and placement: corticosteroid delivered into tendon rather than around it is not neutral for that tendon, and repeated injections into the same shoulder have diminishing returns. Placing it under ultrasound guidance rather than by landmarks is how you make sure it goes where it is meant to go. The right use is to break a pain cycle so that rehabilitation can actually happen.
Does PRP work for rotator cuff pain?
It has a genuine role where the tendon itself is the problem — rotator cuff tendinopathy and partial tearing — which is a better-supported indication than most of what PRP is advertised for. It is not a quick fix. The response builds over weeks, it works alongside a loading programme rather than instead of one, and it is the wrong tool if the real problem turns out to be a frozen shoulder, the AC joint or the neck. Getting the diagnosis right first is what makes the difference.
Does surgery fix shoulder impingement?
The specific operation offered for impingement, arthroscopic subacromial decompression, was tested against placebo surgery in randomised trials published in 2017 and 2018, and it did not outperform the placebo. That is a strong result and it changed how this should be discussed. It does not mean shoulder surgery never helps — a genuine tear, a calcific deposit, a frozen shoulder or arthritis are different situations with different answers. It means that for pain labelled impingement with no structural lesion, decompression is not the answer it was once assumed to be.
Why does my shoulder hurt more at night?
Night pain is the symptom that brings most people in, and there are a few reasons for it. Lying down removes the daytime distractions that mask the pain, sleeping on the affected side compresses the irritated tissue directly, and lying flat lets the arm drop into a position that closes down the space the tendon runs through. Night pain that is escalating rather than settling is worth taking seriously, because it is one of the features that separates a shoulder that will improve on its own from one that will not.
