Rotator cuff tears: cables and crescents?

So a tear of the rotator cuff is pretty disastrous, right? It means surgery, right? Well NO it doesn't! As our understanding and knowledge of the shoulder joint improves so does our ability to identify cuff tears, but we also know that plenty do just fine without surgery.

In fact it has been well known in the medical world for quite sometime that there are a lot of people out there with tears in their rotator cuff tendons functioning normally with no pain or issues. For example Templhof et al back in 1999 published a study that investigated the shoulders of over 400 people with no pain or reported problems, and found that 23% of them had rotator cuff tears! Thats nearly 1 in 4, however, this study was done on the 'older' generation ie 50 years old and onwards, they also found the older you are the more likely you are to have a tear, no real surprise here I guess. But what is surprising is the numbers, for example over half of all the 80 year olds they looked at had cuff tears but where blissfully unaware of them!

But this phenomenon of rotator cuff tears with no symptoms or problems isn't just seen in the older generations. It is also seen in younger and more sportier populations too, Conner et al showed that up to 40% of elite over head athletes have rotator cuff tears, again with no reported problems!

So why is this? Why are there so many people out there with tears in the rotator cuff functioning normally, and what can we learn from those that have do have tears with no pain or loss of function to help those that do!

Well first we need to look at where and what is torn in the rotator cuff. The location of tears is vitally important.
Although the cuff works synergistically together and needs balance in all areas, some parts of the cuff can be classed as more 'important' than others in terms of structure and function.

For example the top or superior portion of the rotator cuff can be thought as the suspension bridge of the shoulder! Confused? Let me explain more…

Burkhart et al, first used this description when he described a thickened section found in the supraspinatus and infraspinstus tendons, in fact its nearly 3 times thicker than the rest of the tendon. Its this thickened section which he called the cable. In front of this was a thinner, poorly vascularised section of the cuff, which he called the crescent. He explains how the cable can 'bypass' forces and stresses around the crescent and transfer load between the anterior and posterior portions of the cuff, effectively connecting the front and back of the cuff together. Just like a suspension bridge cable connects one side of a bridge to another and carries the load from one pillar to the other across a span. See the images below

B= crescent C= cable S=supraspinatus I= infraspinatus BT= biceps tendon

So a tear that occurs in the 'crescent' area of the cuff, although can, for some, be initially very painful, it doesn't cause much, if any, issues to the function for the shoulder, as the cable behind it continues to take and distrubute the load and tension between the anterior and posterior rotator cuff, continuing to dynamically stabilise the humeral head, preventing superior humeral head translation on arm movement.

However, if the tear goes through the 'cable' it now effects the ability the superior cuff, it cannot function effectively with the anterior and posterior sections, it cannot contribute to balance the forces on the humeral head and so it can allow it to translate excessively superiorly under the subacromial arch causing pain and even stopping the arm from lifting completely, called pseudo paralysis, more on that here.

The crescent area of the cuff is thought to be where most cuff degeneration starts due to its poor vascularity and high demands placed on it and can cause the tendon to start to fail and tear. However, if the tear settles and stabilises and doesn't progress through the cable then the shoulder can happly function as normal. It is these cuff tears that the above studies see in people living normally with no pain or loss of function.

So how do we know if the tear is in the crescent or the cable?

Well simply put clinically we can't, and even with imaging it can be a challenge to see exactly where a cuff tear is.

So what about pain? Surely the amount of pain in the shoulder gives us a clue if its small or large cuff tear? Well actually NO it doesn't. Pain is really bad indicator of the amount of cuff damage, in fact it is a bad indicator for most things. It in fact has been seen that smaller partial cuff tears can be MORE painful than bigger tears? Gotoh et al found that when the cuff tendon is only partially torn it releases more pain mediating chemicals than when fully torn, also Carr et al found that its more likely the bursa above the cuff that generates the pain than the cuff tear itself due to its rich innervation. Anecdotally I have seen small tiny cuff tears make fully grown hard men cry like babies, and conversely seen massive huge cuff tears produce very little discomfort, so as a rule, pain is not a good indicator of tear size or a prognosis of outcome.

So if we can't use pain what about function, well yes this does help inform us if there is a crescent or cable tear. Pseudo paralysis and drop or lag tests where you can't hold your arm in certain positions does tell us with some reliabilty that the cuff could be badly torn with cable involvement. BUT don't jump the gun, these tests are not perfect and I have seen many shoulders with pseudo paralysis and even lag signs in the first few days of pain starting that can resolve spontaneously within a few weeks just due to the pain levels.

Imaging can be helpful in determining the size and location of cuff tears. MRI is usually the best way of visualising the exact location of cuff tears, as well as the quality of the muscle behind the tear, important in deciding what type of surgery will or won't help. Ultrasound scanning is also useful for looking at the rotator cuff, it can be used dynamically and its quicker and cheaper to use, but it can also be tricker to interpret, but you can see a lot on ultrasound scanning including the cable. I recently found this paper by Morag et al that describes how to visualise them. I have been using ultrasound scanning in my clinics for over three years now in an attempt to aid my diagnosis of shoulder pains and I'm still learning and still often send for an MRI or second opinion to confirm what I think I may have, or have not seen.

So in summary, if you perosnally have, or see a patient with shoulder pain and think it maybe a rotator cuff tear, don't jump the gun and don't think it automatically needs surgery, it needs investigation to determine if its in the crescent or cable combined with a host of other factors before a prognosis can be made. Also remember a lot of rotator cuff tears can settle and manage very well with very little intervention. And finally remember that pain levels and cuff tears are completely unrelated.

As always thanks for reading

Happy exercising


This is article is intended for information purposes only, if in doubt please consult your doctor or physiotherapist for further advice.





  1. Great post here again Adam. Do you you know of any research looking into whether the location of a RTC tear effects a patient’s prognosis with non-operative management?
    A group of orthopedic surgeons in the US have reported satisfactory results with non-operative care including physical therapy for the majority of patients with atraumatic full thickness RTC tears. Location of the tear within the tendon was not mentioned in these studies.

    • Hi Bill
      Thanks for your kind comments, and thanks for the links to the studies. I did see one of those papers.
      I am unaware on any papers that specifically describe the locations of tears as a prognosis of outcome eg crescent v cable, most describe them as either full thickness or partial and small, medium, large or massive in size as well as either articular surface or bursal sided tears (with articular surfaces usually being seen in the younger patients), but again I cannot find any studies that say just due to location of the tear has a high predictive value of poor or good outcomes with either operative or conservative treatment.
      However from experience the decision whether to operate is not just based on location and size of tear, but also factors such as age, quality of the muscle behind the tear and the desired use of the shoulder joint in the future, for example an 76 year old with a full thickness tear including cable and pseudo paralysis of the arm, but with poor health and who only wants less pain and able to do ADL activities will probably not have surgery, whereas the 48 year builder with a similar tear and who needs to work for another 20 years would probably have the surgery
      Each case is individual and all the factors should be considered when deciding conservative or operative treatment, not just location and pain etc.

      • Adam,
        I agree this is an individualized decision for surgery with multiple factors at play. It is also very interesting that patients who have RTC repair surgery are often found to have a subsequent tear on MRI yet still have satisfactory outcomes. What do you think is the mechanism for their improvement? Is surgery the ultimate placebo? Is it the progressive loading with post operative rehabilitation? Other factors?

      • Hi Bill
        That’s a very good question and one I’m not entirely sure about! I do have few theories thou, firstly most cuff surgery repairs involve removal of the sub acromial bursa, we know this is a highly sensitive structure & generates a lot if pain, when its gone and pain isn’t present then we have a happy patient.
        Secondly an acromial decompression can also be done for cuff tear repairs, this again can help reduce pain and improve function even in the presence of a cuff tear
        Finally as you mention I think the enforced rest after surgery and subsequent rehab loading the cuff also helps reduce pain and increase function of the shoulder regardless if there is a small tear present
        What do you think?

  2. Adam,
    I am really not sure either. I think the bursal theory has some merit. I listened to Jeremy Lewis’s podcast on BJSM and he cited the research linked below stating that bursectomy alone had similar results to bursectomy plus SAD for patients with SAIS. It would be interesting to do a similar study with RTC repairs. I also think a placebo controlled surgery study for RTC repairs might also be really interesting.

Related news