Rotator Cuff Tear: Do You Always Need Surgery?

When the scan report lands in a patient’s lap and the word “tear” is sitting there in black and white, the assumption is almost automatic. Something is torn, so something must be stitched. I see the relief, and then the worry, cross a patient’s face in the same minute: relief that there is finally a name for the pain, worry that an operation is now unavoidable. So let me say the most important thing first, plainly. A rotator cuff tear does not automatically mean you need surgery. For a great many people it does not. But “not always” is not the same as “never,” and the whole skill lies in telling those two apart.

The rotator cuff is a set of four flat tendons that wrap over the top of the shoulder and hold the ball of the joint snugly in its socket while your arm lifts, reaches and rotates. When one of those tendons tears, the shoulder can become painful, weak, or both. What surprises most patients is how common these tears are, and how many of them cause no trouble at all.

Most cuff tears are quieter than you would expect

Rotator cuff tears are largely a feature of the ageing tendon, much as grey hair is a feature of the ageing scalp. In a community screening study in Japan where everyone who turned up had an ultrasound of both shoulders, full-thickness tears were found in just over a fifth of people overall, and the rate climbed steadily with age, from essentially none below the age of fifty to more than a third of people in their eighties. Strikingly, about two-thirds of all the tears found caused no symptoms whatsoever. The owners did not know they were there.

I dwell on that figure because it reframes the whole question. If you could line up a hundred people in their seventies and scan every shoulder, you would find a great many tears in people living perfectly active lives, gardening, lifting grandchildren, never once thinking about their shoulders. A tear seen on a scan is therefore not, by itself, a problem that must be repaired. It is a finding that has to be matched against how the shoulder actually behaves and how much it troubles you.

The split that changes everything: degenerative versus traumatic

The single most useful question I ask is not “how big is the tear?” but “how did it happen?” The answer sorts most patients into one of two groups, and the two are treated quite differently.

A degenerative tear creeps up on you. There is no single dramatic moment. The tendon, thinned by years of use and a less generous blood supply with age, gives way gradually. The shoulder grumbles, aches at night, and slowly loses some strength. These are the tears that dominate the numbers above, and they are the ones where surgery is most often not the first answer.

A traumatic tear is different in character. The shoulder was working well, and then a defined event, a fall onto an outstretched hand, a heavy lift, a sudden jerk, tore a tendon that may have been largely healthy until that instant. The arm is often suddenly and genuinely weak. These tears, particularly when they are full-thickness and the patient is younger and active, sit firmly in the group where earlier surgical repair deserves serious consideration.

There is also the matter of depth. A partial-thickness tear involves only part of the tendon’s thickness; a full-thickness tear goes all the way through. Most partial tears, and many small full-thickness degenerative ones, are managed without surgery in the first instance. The treatment, in other words, is decided not by the word “tear” alone but by the story behind it, the depth, the size, and the person attached to the shoulder.

What the trials actually show for degenerative tears

This is where good research has genuinely sharpened the answer, so let me lay it out honestly rather than from habit.

Several well-conducted randomised trials have compared surgical repair with a structured physiotherapy programme for symptomatic degenerative tears. A Finnish trial randomised patients with non-traumatic supraspinatus tears to physiotherapy alone, to a smoothing of the overlying bone with physiotherapy, or to a full repair with physiotherapy. At two years there was no meaningful difference between the three groups in pain or function, and the authors concluded that conservative treatment is a reasonable first option for these tears in older patients. When the same patients were followed for more than five years, the conclusion held: surgery had not proven superior for these small, single-tendon tears.

A Norwegian trial compared primary tendon repair with physiotherapy (with the option of later surgery if physiotherapy failed) in tears up to three centimetres. The honest nuance here is important. Early on the two groups were very close. But as the researchers followed these patients out to five, ten, and fifteen years, a modest advantage for repair emerged and slowly widened, and in the physiotherapy group the tears tended to enlarge over time. Even so, roughly a quarter to a third of the physiotherapy patients ended up choosing surgery across fifteen years, which means most did not.

Pull these together with a meta-analysis of the randomised evidence and a clear, usable message appears. For degenerative cuff tears, surgery produces, on average, a small additional improvement over good non-surgical care, and only around one in eight conservatively treated patients later crosses over to an operation. The success rate of non-surgical treatment is high enough that a surgeon should be selective, choosing surgery for the patients most likely to gain from it rather than offering it to everyone with a tear.

So how do I actually decide? The two paths

When a patient sits in front of me, I am, in effect, sorting their shoulder onto one of two paths. Neither path is a guarantee, and patients move between them as the shoulder declares itself over time. But the framework is honest, and it is the same one I would use for my own family.

A few threads run through both paths. New, real weakness after a clear injury pulls a shoulder firmly towards Path B; it suggests a tendon that has genuinely let go rather than one that is merely sore. Tear size and progression matter, because a tear that is enlarging, or muscle that is starting to waste and fill with fat, is a tear whose repair window may be closing. And the person matters: a sixty-eight-year-old who wants a comfortable shoulder for daily life and a thirty-five-year-old labourer who must carry loads overhead are not the same problem, even with an identical scan.

What “non-surgical treatment” actually involves

When I recommend the non-surgical path, I am not recommending that you do nothing, and I am certainly not recommending rest alone. A frozen, unused shoulder gets worse, not better. A good programme is active and specific: a physiotherapist guides you through exercises that strengthen the muscles around the cuff and restore smooth movement, so that the remaining healthy tendons learn to do more of the work. This takes a fair trial of several months, done properly, before anyone can say it has failed.

Where pain is the main obstacle to doing that rehab, a single corticosteroid injection into the shoulder can give useful short-term relief of pain and a window in which to work, and the guidelines support its considered use. I am cautious about repeating injections many times over, because there is reason to think that multiple steroid injections may weaken the tendon and make any later repair less secure. One well-timed injection to unlock a rehab programme is a tool; a string of them as a substitute for a plan is not.

And when surgery is the right call

None of this is an argument against surgery. Rotator cuff repair is a good operation that I perform regularly, and for the right shoulder it is clearly the better choice. The strongest case is the acute, traumatic, full-thickness tear in an otherwise healthy shoulder, especially in a younger or active person, where a prompt repair aims to restore strength and protect the joint for the decades ahead. Surgery also earns its place when a fair, well-structured trial of non-surgical care has genuinely been given and the shoulder remains painful and weak, or when follow-up shows the tear is enlarging.

The guidelines reflect exactly this balance. They recognise that both physiotherapy and surgery meaningfully improve symptoms for small and medium tears, that a repair which heals tends to give somewhat better function than physiotherapy or a repair that fails to heal, and that for partial tears a trial of physiotherapy comes first, with surgery reserved for those who do not settle. In other words, the formal recommendations say what the trials say: tailor the treatment to the tear and the person, do not reach for the same answer every time.

What I tell my patients

If you take one thing from this, let it be this. The question is not “is my cuff torn,” because at a certain age a fair number of perfectly happy shoulders are. The question is “is this tear, in this shoulder, in my life, behaving in a way that surgery would improve.” For many degenerative tears the honest answer is that a good rehabilitation programme, with watchful follow-up, is the sensible first move. For a smaller group, those with a fresh significant injury, real weakness, a large or enlarging tear, or a high-demand shoulder, the honest answer points towards repair, and sooner rather than later.

That is not fence-sitting. It is the actual shape of the evidence. Anyone who tells you that every cuff tear must be operated on, or that no cuff tear ever needs surgery, is choosing a slogan over your shoulder.