A reported cuff tear feels like an answer. Most of them do not hurt.

ROTATOR CUFF, OR THE NECK?

A rotator cuff tear on your MRI does not prove it is causing your pain. Cuff tears are common in shoulders that do not hurt. Numbness in specific fingers, or weakness without pain, points to a nerve root in the neck instead. A reported rotator cuff tear feels like an answer. It is one of the least reliable answers in musculoskeletal (muscle and bone) medicine, because most of them do not hurt.

The number that should change how you read your report

Rotator cuff abnormalities on MRI are extremely common in people with no shoulder symptoms at all. How often they show up climbs steeply with age. Population studies that look at both shoulders routinely find tears in the pain-free one.

So “MRI shows a partial-thickness supraspinatus tear” describes a shoulder that belongs to somebody your age. It may be your pain source. It may be an incidental finding sitting next to the actual problem. That problem is sometimes a C5 or C6 nerve root.

What does rotator cuff pain feel like?

• Pain reaching overhead or out to the side, and reaching behind you to a seatbelt or a back pocket.
• Night pain lying on that side — one of the more telling features.
• Weakness that is painful weakness — the arm gives because it hurts, not because the muscle cannot fire.
• Pain that fades around the elbow rather than traveling into the hand.
• No numbness, no tingling, no dermatomal pattern (no line of symptoms that follows one nerve root’s skin map).

Your arm symptoms may include numbness in specific fingers, or weakness that is present without pain. If so, the cuff is not the explanation, whatever the scan says. That belongs on radiculopathy.

Why the cuff and the neck fail together

There is a mechanical reason the two so often occur together. A painful cuff changes how the shoulder blade moves. A scapula (shoulder blade) that stops rotating properly shifts load to the cervicothoracic junction (where the neck meets the upper back). That area then starts causing symptoms on its own.

That is why treating one and ignoring the other gives a partial result. It is also why a shoulder that has been rehabbed correctly sometimes leaves a neck that now needs attention. It is a sequence, not a failure.

What we do about it here

The first job is establishing which structure is producing this, because that is the question nobody has answered. Where the exam is unclear, a diagnostic injection settles it in twenty minutes.

Where the neck is the driver, that is treatable here, and the shoulder frequently settles alongside it. Where the shoulder really is dominant, you leave with that established instead of suspected. That is what makes the next appointment, wherever it is, worth having.

The scapula is the part both of them share

There is a mechanical reason these two problems keep arriving together. The shoulder blade is the base the arm works from. It only works if it rotates upward as the arm rises. When it does not, the cuff works at a mechanical disadvantage, and the load moves up to the cervicothoracic junction.

So a painful cuff produces a stiff neck. A stiff neck produces a badly moving scapula that keeps the cuff irritated. Treating either one alone gives a partial result and a recurrence. That is how people end up cycling between a shoulder specialist and a spine clinic without either being wrong.

How can a doctor tell if shoulder pain is coming from the neck?

Where the exam leaves real doubt, a diagnostic injection resolves it faster than another scan. We numb the subacromial space (the gap above the cuff tendons) and ask what happened to the arm pain while the anesthetic was working.

Pain substantially gone means the shoulder was driving it. Pain unchanged means it was not, whatever the MRI reported, and the workup turns to the cervical roots and facet joints. That is a cheap, quick and truly decisive test. It is used here instead of a second opinion on the same imaging.

How is it treated if it is the cuff, the neck or both?

If it is the cuff: conservative care matches surgery for partial-thickness cuff tears, which are about twice as common as full-thickness ones, so the cuff is treated here, without an operation. That means a guided injection to settle the pain enough to work, PRP into the tendon under ultrasound where it has stopped healing, a progressive loading program that runs for months rather than weeks, and the metabolic work that decides how well tendon repairs. We work to avoid surgery on these tears. Eubank and colleagues, synthesizing the trials in 2024, concluded: “The results of this scoping review do not support superiority of operative over nonoperative management.”

If it is the neck: which structure decides everything. A nerve root has a favorable natural history (its usual course is good) and is managed as radiculopathy. A lower cervical facet joint that refers pain over the scapula is a different problem with a different answer. It is identified by block and can be treated durably.

If it is both, which it frequently is, the order matters. Treat the dominant one and reassess at six weeks. Expect the second problem to be smaller than it looked once the first is quiet.

Why the tissue failed in the first place

Both structures are tendon, and both fail for the same underlying reasons. Diabetes is one of the strongest predictors of cuff tendinopathy, of poorer healing after repair, and of adhesive capsulitis (frozen shoulder). The same glycation (sugar bonding to collagen) that stiffens a cuff tendon stiffens the structures in the neck.

That is not an aside. It is why a shoulder rehabbed perfectly sometimes fails anyway. And it is the part of the workup that most people have never had done.

For a sore cuff, home work usually starts with the pendulum exercise and builds to band outward rotation, both taught in our physiotherapy program for neck pain.

What should I ask my doctor about shoulder and neck pain?

• An examination of both structures in the same appointment.
• A specific answer to “which structure do you think is generating this, and how do you know?”
• A diagnostic injection where the exam is truly unclear, instead of a second scan.
• And if it is the neck: which structure — a root or a facet joint — because they are treated completely differently.

Orthobiologic treatment options

Once the cuff, not the neck, is proven to be the source, a worn tendon or a partial tear that has failed exercise is a good fit for PRP placed into the tendon under ultrasound. Apart from the less common full tear that has pulled back, which gets a surgical opinion, that covers most of the cuff problems we see. A 2026 review of biologic shoulder injections found the strongest PRP evidence in rotator cuff disease.

You stay awake with local numbing, and most people drive home. It is done at our regenerative clinic, 4477 Woodson Rd in St. Louis. Other arm problems that get blamed on the neck: biceps tendonitis, shoulder arthritis, and at the elbow, tennis elbow.

Our PRP and orthobiologics overview explains the three treatments we use and where each one goes.

What people ask about a cuff tear on a report

Does a torn rotator cuff always need surgery?

No. Many tears are managed without an operation, and many are asymptomatic to begin with. Size, retraction, function and what you need the arm to do decide it — not the presence of the word “tear.” See how to tell the neck from the shoulder.

Could my shoulder pain actually be my neck?

Frequently, and it is one of the most under-recognized patterns in this area. The lower cervical facet joints and the C5-C6 roots both refer pain into the shoulder region convincingly. See shoulder pain that starts in the neck. Which fingers are involved, if any, separates these two quickly.

My injection into the shoulder did not help. What does that mean?

It is useful information. A subacromial injection that changes nothing argues against the subacromial structures being the source. It moves the question toward the neck. See the subacromial bursa. That result is diagnostic information you already paid for.

Should I have both imaged?

Only after an exam. Two scans in a person over fifty reliably produce two abnormal reports. That makes the picture less clear, not more. See telling the neck from the shoulder.

How do I know if it’s a pinched nerve or a rotator cuff tear?

Look at where the symptoms go and what they feel like. A cuff problem hurts when you reach overhead or behind you, hurts at night when you lie on that side, and fades around the elbow. The arm gives way because it hurts. A pinched nerve in the neck brings numbness in specific fingers, or weakness that is there without pain. If you have those, the cuff is not the explanation.

Does a rotator cuff tear on my MRI mean it is causing my pain?

Not on its own. Cuff tears are extremely common in people with no shoulder pain at all, and they become more common with age. Studies that scan both shoulders routinely find tears in the one that does not hurt. The tear may be your pain source, or it may sit next to the real problem, which is sometimes a C5 or C6 nerve root in the neck.

Can a rotator cuff tear cause neck pain?

It can, through the shoulder blade. A painful cuff changes how the shoulder blade moves. When the shoulder blade stops rotating properly, the load moves up to the base of the neck. That area then starts to hurt on its own. That is why treating only the shoulder or only the neck often gives a partial result. It is also why the two problems keep showing up together.

Related reading

• Is it your neck or your shoulder?
• Subacromial bursa
• AC joint pain
• Cervical radiculopathy
• Cervical facet syndrome
• Why your MRI is normal

An injection that did nothing tells you where the pain is not coming from.