The clinical presentation that shall not be named

When a paper is published in the shoulder pain literature, much of the social media rhetoric revolves around the diagnostic label used, rather than the main thesis of the study. When seeking ethics approval, writing up, and publishing a scientific paper, you literally HAVE to choose ONE term and thus you will automatically annoy and alienate a section of the MSK community because exactly what to label many shoulder conditions is subject to ongoing debate and disagreement. When publishing research, I often choose rotator cuff-related shoulder pain (RCRSP) for reasons I will go into below. Something that sucks the life out of me is writing at length about why I chose to use a particular term in the introduction of a paper. With the constraints of journal word count, every word is precious, and I will happily choose to dedicate more time to the aim of the paper than writing ad nauseam about the merits of various diagnostic labels. As such, in my most recent paper [13], my justification of using RCRSP was concise, which I always aim to be. However, if you did want some deeper insight into my rationale behind using certain diagnostic labels for a certain shoulder pain condition, this essay might provide some context.

Riddle me this….

A patient presents to you with non-traumatic shoulder pain. It hurts when they elevate their arm, and resisted movements into external rotation and abduction are weak and painful. Various impingement tests are positive – you know the Hawkins and Neers ones. The patient has preserved PROM and you don’t suspect an unstable shoulder, or shoulder pain arising from sinister or referred origins.

What do you label this condition?

For roughly 50 years, since the 1970s, most clinicians around the world would call this very common clinical presentation, ‘shoulder impingement syndrome’.

However, since about 2010 (yes, there were some rumblings before) there has been criticism of the veracity and usefulness of the label ‘impingement syndrome’. This scrutiny has led to the near complete refutation of shoulder impingement as a primary cause of rotator cuff pathology and whether the release or removal of the offending impinging structures, surgically or otherwise, plays any important part in improving the shoulder pain experience [1, 2].

Empirical questions and evidence aside, qualitative research draws valuable attention to how the label ‘impingement syndrome’ is construed by the patient – and surprise, surprise, it’s not that positive:

“He said there was some impingement that could have been inflammation from irritation in the shoulder ... things stopping my arm actually moving the way it should ... I could just picture something just in the joint that got in the way at a certain point and was making it difficult for the socket to move in the way it should be moving” [3]

This isn’t an essay about impingement, my thoughts are on public record here, but what do we now call this clinical presentation in the wake of the diagnostic void ushered in by the demise of ‘impingement syndrome’?

The candidates:

• Rotator cuff related shoulder pain
• Rotator cuff tendinopathy
• Subacromial shoulder pain
• Non-specific shoulder pain
• Bursitis
• Rotator cuff tear

Rotator cuff related shoulder pain (RCRSP) was formalised by Jeremy Lewis in 2016 [4] and was proposed with the express intention of moving away from definitive pathognomonic diagnostic labels. RCRSP has contemporaries in the hip/groin, such as adductor-related groin pain, which seems to be wholly accepted amongst industry experts [5]. Equally, RCRSP has gained acceptance among shoulder experts [6] and practicing clinicians [7] and is now widely used in clinical research. RCRSP can be viewed as a clinical hypothesis, made when our credence in a stiff or unstable shoulder is low, and when there are no clues that raise our suspicion of a sinister or referred into shoulder pain presentation. It's a somewhat vague term, which some love and others hate. A limitation of the term might be its reference to the rotator cuff and thus its foundation in structure, which could give rise to perceptions of structural remedies. Conversely, this may also be a strength of the term, because a pathology is NOT front and centre – rather a muscle and tendon complex, and what do muscles and tendons generally like to do – exercise! So RCRSP might be a coherent gateway into exercise-led rehabilitation rather than surgery and injections. The usefulness of a scientific theory is whether it represents an advance on pre-existing theories and RCRSP is a significant advance over shoulder impingement, rotator cuff disease, and the like. Is RCRSP the final destination? Probably not. Any good scientist should accept that their theories are ephemeral (even Einstein said this about his general theory of relativity) and that their overthrow is a triumph of scientific progress, not a failure of the theory.

Rotator cuff tendinopathy is a plausible diagnostic label for the clinical presentation outlined above. If it’s good enough to call achilles tendon pain achilles tendinopathy then why not rotator cuff tendinopathy? My concern here is the tenuous relationship between rotator cuff tendinosis observed during radiological imaging and the experience of shoulder pain [8]. Can we confidently incriminate the rotator cuff tendon as the primary cause of a person’s shoulder pain with imaging or special tests? If we can’t, then perhaps we should refrain from confidently referring to pathology in our diagnostic labels. That said, I do find myself using this term on occasion in clinical practice because most lay people have been exposed to the term ‘tendinitis’ and it’s not too hard to get to tendinopathy from tendinitis.

Subacromial shoulder pain or subacromial pain syndrome is another popular term used in research and clinical practice [7]. My aversion to the term is the invocation of subacromial – which is a meaningless word for most people with shoulder pain. Karl Popper, the famous philosopher of science, states that “we should always aim to speak as simply, as clearly and as unpretentiously as possible”, and I think referencing a strange anatomical location in the shoulder further muddies the water for a person simply wanting a diagnosis to validate their shoulder pain. If we can avoid it, and I think we can, we should.

What about non-specific shoulder pain, much like non-specific low back pain? I’m sympathetic to these non-specific terms, owing to the uncertainty or impossibility of deriving a definitive pathognomonic diagnostic label in clinical practice. However, I do have reservations about how these labels might be construed by patients. For this reason, I don’t tend to use this term in clinical practice. In a study published by Josh Zadro in 2021 [9] there was no difference in patient perceptions of needing surgery or imaging between ‘RCRSP’ and ‘episode of shoulder pain’, so there is an argument that these non-specific terms don’t really save us from unnecessary health care resource usage, which is one reason why these labels are recommended.

Speaking of the Josh Zadro study, intriguingly the best label for minimising perceptions of needing surgery or imaging was ‘bursitis’. This surprised me on first reading. Although perhaps this is because bursitis sounds innocuous relative to ‘tear’ or ‘impingement’, and inflammation has an intuitively better natural history compared to torn tissue. Another interesting feature of this study was the ‘bursitis’ label not being associated with an increased perception of needing an injection. This is not my experience in clinical practice, although maybe that is the fault of health care practitioners. So, should we call all presentations of non-traumatic, non-stiff, non-unstable, non-sinister, non-referred into shoulder pain, bursitis? Irrespective of its truth content? Maybe. Bursitis is frequently identified in people with shoulder pain, but also as frequently in those without shoulder pain [8]. My intuition is I think we can do better than just using a label to manipulate the behaviour of our patients.

What about rotator cuff tear? In the presence of a radiologically verified rotator cuff tear, and signs and symptoms correlating to this, there is an argument that a diagnosis of a rotator cuff tear is the most veridical label we have. We also have evidence that rotator cuff tears greater than 2.5cm do often correlate to shoulder pain [10]. I’m sympathetic to this. My reservation with using rotator cuff tear at scale, again, is what it means to the patient. We know that managing a rotator cuff tear, including partial and full thickness tears, with exercise or surgery is equally effective [11] but evidence suggests that patients diagnosed with a rotator cuff tear tend to erroneously believe it’s a mechanical issue that needs a mechanical fix:

“I can’t fix that [tendon tear] just by doing exercises,” ... that kind of made me make the decision a bit quicker to go ahead with the surgery. [12]

In summary, when deciding on a diagnostic label, we would do well to consider the following:

• Is the label easy to use in clinical practice?
  • Is it cogent and intelligible for patients with varying degrees of health literacy?
  • Do clinicians feel comfortable using the term?
• Does the term lend itself to unnecessary overuse of health care resources?
• What is the truth content of the term?
• Can the term be used in clinical research?
  • Remember that when studies go through ethics and peer review, it can be a barrier to have terms that encompass everything. 

 References

1. Beard, D.J., et al., Arthroscopic subacromial decompression for subacromial shoulder pain (CSAW): a multicentre, pragmatic, parallel group, placebo-controlled, three-group, randomised surgical trial. The Lancet, 2018. 391(10118): p. 329-338.
2. Park, S.W., et al., No relationship between the acromiohumeral distance and pain in adults with subacromial pain syndrome: a systematic review and meta-analysis. Sci Rep, 2020. 10(1): p. 20611.
3. Cuff, A. and C. Littlewood, Subacromial impingement syndrome - What does this mean to and for the patient? A qualitative study. Musculoskelet Sci Pract, 2018. 33: p. 24-28.
4. Lewis, J., Rotator cuff related shoulder pain: Assessment, management and uncertainties. Man Ther, 2016. 23: p. 57-68.
5. Weir, A., et al., Doha agreement meeting on terminology and definitions in groin pain in athletes. Br J Sports Med, 2015. 49(12): p. 768-74.
6. Littlewood, C., et al., Physiotherapists’ recommendations for examination and treatment of rotator cuff related shoulder pain: A consensus exercise. Physiotherapy Practice and Research, 2019. 40(2): p. 87-94.
7. Powell, J.K., et al., Physiotherapists nearly always prescribe exercise for rotator cuff-related shoulder pain; but why? A cross-sectional international survey of physiotherapists. Musculoskeletal Care, 2022.
8. Barreto, R.P.G., et al., Bilateral magnetic resonance imaging findings in individuals with unilateral shoulder pain. J Shoulder Elbow Surg, 2019. 28(9): p. 1699-1706.
9. Zadro, J.R., O’Keeffe, M.O., Ferreira, G.E., Haas, R., Harris, I.A., Buchbinder, R., Maher, C.G., Diagnostic labels for rotator cuff disease can increase people’s perceived need for shoulder surgery: an online randomised controlled experiment. JOSPT, 2021. Ahead of print.
10. Hinsley, H., et al., Prevalence of rotator cuff tendon tears and symptoms in a Chingford general population cohort, and the resultant impact on UK health services: a cross-sectional observational study. BMJ Open, 2022. 12(9).
11. Karjalainen, T.V., et al., Surgery for rotator cuff tears. Cochrane Database Syst Rev, 2019. 12: p. CD013502.
12. Malliaras, P., et al., 'Physio's not going to repair a torn tendon': patient decision-making related to surgery for rotator cuff related shoulder pain. Disabil Rehabil, 2021: p. 1-8.
13. Powell, J.K., et al., "Restoring that Faith in my Shoulder": A Qualitative Investigation of how and why Exercise Therapy Influenced the Clinical Outcomes of Individuals with Rotator Cuff-Related Shoulder Pain. Phys Ther, 2023.