Dr Andy Franklyn-Miller

Are we fighting an uphill battle against surgery without data?

I have seen this week a variety of twitter debates lamenting the lack of research funding support for studies investigating lateral hip pain “greater trochanteric pain syndrome“ whereas osteoarthritis & total hip replacement surgery always seems to win the funding and hence more publication bias.

Simultaneously, a debate on how hip arthroscopy seems to be seeing a resurgence with anecdotally poor results, perhaps off the back of a study that very objectively assessed the operative intervention of “arthroscopy” against the less objectively measured “conservative care”.

What can we learn from the surgeons? Well their intervention is definitive – replace the hip or take a look inside the joint and cut away part of the bone or labrum. There is no question about whether the intervention should differ materially, yet in rehabilitation/ physiotherapy/sports medicine we debate endlessly and litter the research literature with interventions ranging from passive range of motion, electro modulation or stimulation, taping, massage, strength training of various types, power training, and even worse when comparing to surgery “conservative care”.

Orthopaedic surgeons give their patient a diagnosis be it hip osteoarthritis, labral tear or cam or pincer bony change supported by radiological imaging and carry out a defined procedure. There is of course a big question about whether the patient should have surgery or not, but once decided – surgery is surgery. The patient knows they are going to be “fixed” or at lest “undergo surgery” and as part of the consent process the surgeon will talk about their success rates, their failure rates and complications and likely be part of a national or at least hospital registry where this data is published.

When we look at conditions such as athletic groin pain, greater trochanteric pain syndrome, hip flexor related pain without delving into the more suspect piriformis syndrome or the recently debated deep gluteal syndrome, we struggle. We seem desperate to avoid using imaging to quantify the extent of the anatomical pathology – we argue that it doesn’t effect outcome, or tell us anything. Can we really mean this, quantifying abnormality can surely only assist both the explanation to the patient but also objective measurement. 

We claim MRI risks overdiagnosis?  It does this is if we are reliant on reports without the skill set to interpret the images ourselves, to allow patient discussion and grading the degree within our own patient cohorts. We claim MRI worries our patients unnecessarily, but if we are proposing a treatment to effect change in pain and function, doesn’t the patient deserve a diagnosis, and objective outcome measure and a timeline against data our own experience with real patients to achieve it. Do our patients not require us to increase our descriptive skills in understanding the imaging.

We are quick to blame the surgeon for over intervention, but they are producing registry data on the outcomes of their interventions – are we? How can we do this if we just don’t know how to quantify the degree of anatomical abnormality and relate this to objective outcome measures. The tendinopathy literature is a great place to start – patients are often included with inclusion criteria of palpatory pain in the tendon and reduction in a validated patient reported outcome measure.

I will argue this weekend in my talk for Sports Surgery Clinic on “conservative management for athletic groin”  that the reliability of palpatory pain has to be questioned as a means of making a diagnosis, How good are PROMS at actually telling us objectively meaningful change, we have seen that many footballers have abnormal HAGOS scores yet still play and we hang our interventions of meaningful change in how a patient perceives their pain – knowing this means something different  to each individual.Orthopaedic Surgeons may not accurately quantify the requirement or % change of re-strengthening and re-mobilisation accompanies recovery, and that perhaps these scores need to be higher than before the operation, and that strength cannot be sprayed or taped on. Many outcome studies suggest rehabilitation is essential as part of post-surgical recovery. The only head to head “surgery versus rehabilitation” paper in athletic groin pain still required all subjects to undergo post surgical rehabilitation – but in the latest systematic review of groin pain intervention we still cannot describe those interventions well enough to be repeatable, in part because of lack of objective measures and standardisation of anatomical terminology! We are shooting ourselves in the foot.

So what is that magic objective measure? Well I don’t have it, but I would suggest we could do worse than start with % of bodyweight strength gain in a related muscle group. We have all seen patients prescribed a program of ‘strength training’ who failed to gain any strength either through program design, non completion or pain, and was deemed to have failed rehabilitation – A recent example is a prescription of core stability for low back pain – how do you define the increase in core stability if you prescribe it? Length of time in a plank, Beiring sorenson? What about the anterior tilt before a plank fails – does it count or not? A recent meta-analysis in Sports medicine shows that strength gains can be made even in a patient in there 90s so can be a realistic ambition.

I would argue we could do worse than choose an intervention of strength in place of surgery, will strength gain answer everything , of course not, but it is a reasonable lace to measure an outcome- if a program is designed to increase strength and is measured against the load lifted versus patient body weight – we can see if the gain is real –and the measure is reproducible regardless of the  actual exercise prescription provided.

But we need to benchmark an anatomical description of the tendinopathy or the oedema or the fluid alongside PROMS to ensure we are comparing similar patients. Do we need to include BMI, collagen phenotype, blood type , HOMA-B and hs-CRP – well we could but that would add to a research study but to start an effective registry we need the minimum number of variables and I think that is what rehabilitation is really missing. We need to contribute, decide on a measure and form registries to judge our own outcome on a large scale and only then with an anatomical description and a standardised objective measure will we be in a position to compete.

We started an ACL-R registry at Sports Surgery Clinic in 2013 and are just about to publish 5 year re-injury data, and are planning to start a more detailed joint registry in patients at risk of total knee and hip replacement looking at strength variables, but why don’t we do it on a wider scale in lateral hip pain. It might start to give us answers, and raise the questions about what else works better. It has  given me an idea….