Keeping it simple during rotator cuff tendinopathy rehab

Hi all,

Back to reality and a very cold Melbourne this week. As I write this we are experiencing the coldest June day in 20 years!

Very excited about the new Mastering Lower Limb Tendinopathy Course – it’s an Oz road trip starting in Melbourne and any clinician who treats tendons is welcome, not only physios. Unique case studies based learning course.

Tendinopathy blog no 26 (subscribe here) is a rare foray into the upper limb with a look at an interested RCT by Christ Littlewood and his group from the UK. I was attracted to the thought process and design behind this study comparing a single exercise to usual physio care for management of rotator cuff tendinopathy. It brings up some interesting questions about the mechanisms of exercise as well as failure of rehab, and suggests we can perhaps take a simple rehab approach, at least some of the time. The second uses a rat model to assess the effects of ibuprofen administration on tendon cell function and mechanical properties. Somewhat informs the hot debate about whether there are any issues in recommending ibuprofen for our patients with tendinopathy – is there any penalty? Should we be doing it?

Keeping it simple with rotator cuff tendinopathy rehab

Background: As most people know this blog is about lower limb tendons, mainly because that is what I see in practice. However, I am venturing in upper limb territory this week because this study grabbed by attemtion. Chris Littlewood, a physio researcher out of Sheffield in the UK is a clever guy and his group have asked a clever question in this new study – if we strip back rotator cuff management to a single exercise, can this be as effective as a bells and whistles approach? The authors performed a randomized controlled trial comparing usual physiotherapy treatment versus only a single exercise for rotator cuff tendinopathy.

The authors argue that ‘the potential superiority of single or multiple exercises is unknown. The potential benefits of a single exercise approach include pragmatic, time saving reasons to facilitate exercise adherence. Also, the assumption that incremental benefit is gained by adding more exercises that are theoretically stressing the same tissue might not be valid and the extra burden possibly unnecessary.’

What they did: They performed a 3 centre pragmatic randomised controlled trial. They included participants that were over 18, had shoulder pain without arm referral, largely maintained range of motion (to exclude frozen shoulder), and no/minimal resting pain, shoulder pain provoked with resisted abduction or external rotation. Participants were excluded if they had shoulder surgery in the last 6 months, reasons to suspect systemic pathology or cervical repeated movement testing that affected shoulder pain and/ or range of movement. The primary outcome was the SPADI at three months post-randomisation. Secondary outcomes included the SPADI (self reported shoulder pain/function) at six and twelve months and Short Form-36 (quality of life) at three, six and twelve months.

The single exercise intervention involved 10-15 repetitions twice daily. Typically, abduction was used, starting isometric progressing to isotonic based on symptoms response. The key was no increase in symptoms upon cessation of the exercise. The load (e.g. theraband, weight) and repetitions were progressed as tolerated. Educating the patients to be able to self manage within this progressive rehab framework was the key goal, and follow up appointments were offered to facilitate this as required. Usual physio may have included advice, stretching, exercise, manual therapy, massage, strapping, acupuncture, electrotherapy, corticosteroid injection at the discretion of the treating physiotherapist.

What they found: There were 42 patients in the single exercise group, and 44 in the usual care physio group. However, at 3 months follow up there were 27 in the single exercise group (15 lost to follow up) and 33 in the usual physio group (11 lost to follow up). Those who completed follow up at 3 months were more likely to be older and report less pain/dysfunction function on the SPADI – but there were no group differences meaning that the effects of randomisation were maintained. The number of treatment sessions was very similar and not statistically different between the groups (3.1 in the single exercise and 3.4 in the usual physio groups). The table below shows the treatments provided in each group. Key differences are a greater range of rehab options in the usual physio group including ‘stabilisation’ exercises and more hands on treatment. Six (single exs) vs four (usual physio) had steroid injections by six months. Eight (single exs) vs five (usual physio) group reported medication use, mainly for pain. Adherence was only recorded in the single exercise group and was 78% but based on only 12 participants who returned their adherence diaries.

The SPADI improved significantly in each group at 3, 6, and 12 month outcomes (compared to baseline) with no significant difference between the groups at any follow up time.

A limitation of the study that the authors acknowledge is a lack of blinding that was not possible due to the NHS environment in which the study was conducted. The other concern was the large attrition even at 3 months, ie about 30%.

Clinical interpretation: The authors note that the change in SPADI at 3 months was relatively small compared to other studies, i.e. 12.4 points in the self managed group and 16.7 points in the usual physio group. The change is greater than the 10 points regarded as a minimally clinically important change, but smaller than the greater than 20-point change shown in some prior studies. This may be due to differences interventions or patient populations, but the authors also note that in the NHS they probably were delayed in starting the rehab because of NHS waiting times. There was much more change by 6 months (29.1 points in the single exercise and 23.5 points in the usual physio groups) – this indicates that patients with rotator cuff tendinopathy don’t show a really good improvement for at least 6 months – so maybe hold off on panicking and reaching for injections/surgery until then – a good take home message.

An important point is that the authors conducted a similar pilot trial in a private practice population and had a similar outcome at 3 months, suggesting the finding is not specific to an NHS population.

One of the authors key discussion points is that these data seem to show that we can use a single exercise and be as effective for our rotator cuff patients as a fancy multi exercise rehab program. They acknowledge their design is not really able to answer this question. What would be nice to see is this single exercise approach versus a multi exercise approach that represents perhaps what physio usually do and that is target ‘dysfunction’ in muscle capacity, movement, control, etc. I agree with the authors that a single exercise approach simply targeting load tolerance would work for some and there are huge benefits in regards to patient adherence – which tells us which patients it is probably best suited to!

I’m sure many people reading this would have seen patients who have failed simple rehab programs (e.g. just calf raises for Achilles tendinopathy – I see this all the time) and one of your thoughts may have been they failed because the rehab was not adequate enough. That may be so, but it may also be true that the simple rehab approach failed because it was not progressed based on symptom response, they failed to provide enough education/reassurance (e.g. reduced fear avoidance) or they failed to load manage well enough (i.e. take away the aggravators).

On the other hand, I am sure, and this is based on experience from the lower limb, that some people have deficits that do need to be specifically loaded, but this may be a minority. The clinical magic is in identifying the ones you can simply load within a pain monitoring framework + education + load management, as apposed to ones that need something extra.

Does ibuprofen impair tendon mechanical properties?

Background: If and how we should use ibuprofen in managing tendinopathy patients is debated. I met up with English Institute of Sport physios a couple weeks ago whilst in London for a tendon chat, and one of the hot topics we discussed was potential negative consequences of taking nonsteroidal anti-inflammatory longer term when managing a tendinopathy. Certainly it is common practice to take a short burst (2-3 weeks) to establish load tolerance during a rehab stint or intermittently over a longer period, generally pre competition or to manage pain for the inseason athlete. There is a perception that perhaps athletes can fall into the same ‘masking’ trap as post steroid injection, ie pain is better so they load too quickly and either flare pain or cause damage e.g. a partial tear or rupture.

The other question is whether the ibuprofen impairs tendon/muscle adaptation. However, studies among older adults show that ibuprofen administration (1200mg/day) may actually enhance quadriceps muscle hypertrophy and strength (Trappe et al. 2011) and did not have a negative effect on patellar tendon adaptation (Carroll et al. 2011)

What they did: This study by Rooney et al. investigated whether ibuprofen negatively impacted on tendon adaptation among rats exposed to a treadmill running protocol. The authors randomised 167 rats into an exercise and cage activity groups, and further divided exercise into acute (1hr, 10m/minute) or chronic (1 hr, 5 days/week, 10m/minute, for 2 or 8 weeks). Half the rats in each group were given ibuprofen. They were interested in the effect of the exercise interventions on both muscle and tendon adaptation. Mechanical and histological examination of tendon and muscle was performed after the exercise protocols when rats were euthanized.

What they found: The major findings of this study are summarised below

• Increased modulus at 2 and 8 wk of cage activity with ibuprofen

Modulus is simply stress divided by strain, so increased modulus means relatively greater stress or force for a smaller strain or stretch in the tendon.

• Reduced CSA at 8 weeks of cage activity with ibuprofen
• Increased modulus at 2 and 8 wk of treadmill load with ibuprofen
• Reduced CSA at 2 weeks of treadmill load with ibuprofen
• Reduced muscle fibre CSA at 2 wk in treadmill group and 8 wk in cage and treadmill groups having ibuprofen

In terms of histologic effects, ibuprofen actually increased cell density in the cage and treadmill groups at 2 weeks, and in treadmill loaded rats at 8 weeks.

Clinical interpretation: The authors conclude that taking ibuprofen does not have a detrimental effect on rat supraspinatus mechanical properties and in some instances had a positive effect, eg increased modulus. They further conclude that ‘the arachidonic acid cascade may not play a major role in the adaptations of tendon to load in a noninjurious exercise model, in contrast to results seen in acute injury models.’ But there was some evidence that ibuprofen may affect hypertrophy of sedentary and exercised supraspinatus muscles. So these findings suggest ibuprofen may have tissue specific effects.

Another interesting finding from this study is greater cell density rather than less cells or inhibition of cell proliferation shown in cell culture studies (eg Tsai et al. 2004). This highlights differences between cell culture (in vitro) and in vivo studies like the Rooney study. So are we really reducing tenocyte proliferation with ibuprofen administration in acute/painful tendinopathy? It has been suggested that maybe if we are simply having an effect on pain using paracetamol (acetaminophen) would be better. However, in the study by Carroll et al discussed above they actually found paracetamol had a detrimental effect on normal tendon adaptation (eg reduced tendon stiffness) with loading but ibuprofen did not.

So back to the original clinical point…is there any penalty to using ibuprofen intermittently to manage pain in someone with a painful tendinopathy? Perhaps the tendon does not suffer but maybe the muscle may, although the human study by Trappe seems to show muscle strength/hypertrophy benefit – so it is not clear that there is any penalty. So the only potential negative argument is pain ‘masking’ which is a considerations and highlights that we need to also apply sensible load management principles, particularly when using a burst of ibuprofen in the rehab context, but also as a longer term intermittent strategy for a competing athlete.

All the best until next time

Peter