16-03-2012, 04:38 PM
Special Tests in Musculoskeletal Examination
About the authors
Paul Hattam and Alison Smeatham are physiotherapists with a
special interest in musculoskeletal medicine, having accumulated
many years of experience treating patients both in the National
Health Service and in private practice. They were among the first
extended scope physiotherapists (ESPs) to establish orthopaedic
triage in primary care during the 1990s, a model that was subsequently
evaluated and reproduced throughout the UK. They completed
a musculoskeletal master’s programme in 2002/3 and retain
a keen interest in research in their area of specialty. Both tutor regularly
on postgraduate courses for doctors and physiotherapists
in musculoskeletal medicine as well as being actively involved in
educational programmes closer to their homes. Paul now leads the
team at The Physios (www.thephysios.com) in Sheffield, UK, while
Alison works as an ESP with the specialist hip team at the Princess
Elizabeth Orthopaedic Centre in Exeter, UK.
viii
Acknowledgements
The authors would like to thank everyone who helped us see this
project through to completion. As is often the case, the size of an
undertaking only becomes evident once you have started it and
the encouragement we received from our families, friends and colleagues
was always welcome and helped to spur us on.
We benefited enormously from the collective wisdom of a
number of medical and physiotherapy colleagues from across the
UK and are particularly grateful to the following: Jill Gillespie,
Margaret Rees, Chris Blundell, and the hip and knee surgeons and
ESPs at Princess Elizabeth Orthopaedic Centre and Devon PCT for
their contributions. We appreciated the free exchange of information
and ideas with our teaching colleagues Elaine Atkins, Jill Kerr
and Emily Goodlad, even though they managed to pip us to the
post and submit the manuscript for their book first.
Nigel Hanchard contributed to the labral, tendon and impingement
tests at the shoulder. As lead author of the Cochrane prototype
diagnostic review on this topic (in development), and of the
Chartered Society of Physiotherapy’s clinical guidelines on shoulder
impingement, his familiarity with the evidence base provided a rich
resource of which we made good use. Nigel also made welcome
suggestions for the introductory chapter, all of which were very
valuable.
We would like to thank Gordon Smith who, despite agreeing to
completely unreasonable deadlines, painstakingly proof read each
section and provided correction and challenge in equal measure. We
were also very grateful for the fresh pair of eyes willingly provided
by Joy Hattam at the final proof reading stage. Our model for the
photographs, Alison Crewesmith, was good humoured throughout
a very long day and Ant Clifford had the eye to make the photos
clear and visually appealing. Space to spread out and edit the book
was generously provided by Lindsay Jackson at EEF, Sheffield.
Thanks to everyone at Elsevier, particularly Heidi Harrison
for her initial enthusiasm that helped to launch the idea and to
Veronika Watkins and Rita Demetriou-Swanwick for their endless
patience and guidance during the writing and final editing stages.
Acknowledgements ix
Paul thanks
Time away researching and writing the book inevitably meant an
increased burden on my colleagues and I would like to thank everyone
at The Physios for their encouragement and support. My family
have exhibited patience beyond measure and I am hugely grateful
to my wife, Alison, for taking everything perfectly in her stride and
helping me keep things in perspective, as well as retaining a keen
sense of humour throughout. The same can be said for Rachael,
James and Naomi who happily got on with life while Dad tapped
away on the computer. You’re amazing!
Alison thanks
I would like to thank friends and colleagues who managed to
remain interested and enthusiastic about this project over many
cups of tea and glasses of wine. My family, and in particular Mum
and Dad, continued to provide unwavering encouragement and
support even when much more important events were taking place
and my love and gratitude goes to them and to David, Karen, Tom,
Rosie, Irene and Terry.
Foreword
Paul Hattam and Alison Smeatham are long-time teaching colleagues
of mine. From their first musings that a new text was
needed to acknowledge and examine the growing bank of additional
tests for peripheral lesions, I have been so excited to see their
book in print knowing that they would create this clear, concise
and highly readable resource drawn from a rigorous and objective
trawl through the available evidence. They have wisely recruited
the opinion of other experts too though, acknowledging that many
tests are chiefly derived from empirical practice and should not be
discarded without due consideration for clinical experience. As we
might expect, many of the tests do not enjoy a strong, consistent
evidence base to support their application and not surprisingly, it is
the expert opinion that draws the eye to guide practice.
I have read this book because I wanted to learn from it and I
have already found it to be a valuable resource. The initial ‘About
the book’ section sets the scene for the rest of the book, providing
generic ‘ground rules’ for how to apply the tests and an explanation
of how the literature was searched and the evidence appraised
and interpreted. ‘Likelihood ratios’, combining sensitivity and specificity,
are justified as the best statistics to assess the clinical usefulness
of the diagnostic tests they have presented. The tests are
described simply and clearly and are accompanied by concise, referenced
discussion to provide context; focussing on the tests themselves
as opposed to the pathology of specific lesions. In this respect
a gap has been plugged since that tends to sit outside the remit of
other texts in musculoskeletal medicine; even those dedicated to
assessment.
The book includes tests I haven’t come across, tests I’d forgotten
and tests I’ve always been slightly confused about. The authors’
determination to unravel the complex and to keep to the facts is
both welcome and reassuring and the simple device of beginning
each test with an ‘also known as’ (aka) list defuses the confusion
where many of the tests described in the literature exist under several
different names. In this they have exposed their awareness of
just who their reader is – a busy clinician with the nous and will
to read more – but with so little time. The authors have struck an
Foreword xi
effective
balance between providing a useful pocket reference
resource and a handbook to support clinical assessment and diagnosis.
I commend their text with confidence, knowing both authors
as I do for their determination to question and challenge and the
rigour with which they will have performed their review.
A About this book
Clinical examination is the bedrock of diagnosis in musculoskeletal
(MSK) medicine. It enables the clinician to both assemble information
and to interpret the findings in order to assist in identifying
the nature and stage of the patient’s disease or injury, determine the
need for further investigations, provide a prognosis, guide treatment
and measure outcome. Its big advantage is that it can be used
at any stage in the patient’s management and reproduction of the
symptoms provides not only immediate feedback but also reassuring
evidence, to both clinician and patient, that the source of the
pain has been identified. Physical tests are quick and convenient
to perform and, in the hands of a skilled practitioner, facilitate
appropriate and cost-effective intervention.
The proliferation of additional or special tests used in MSK examination
over the years has left the clinician with a vast array of physical
tests at his or her disposal. All clinicians know that isolating the
lesion and planning effective intervention is not about performing
lots of tests for the sake of it, but instead selecting the most appropriate
examination procedures as part of the clinical reasoning process.
With so many tests in the clinical domain, combined with uncertainty
concerning their accuracy and reliability, it can be hard for clinicians
to determine whether or not the tests they have chosen are in fact
providing them with the information they need. In addition, because
tests involve both clinical and interpretative skills, results may differ
with the level of the tester’s technical expertise, thereby limiting the
generalization of results produced from individual studies.
Special tests are usually incorporated into the physical examination
once a thorough history has been taken and clinicians can use
this information to guide the selection of examination procedures.
A familiarity with the core elements of the physical examination
is therefore essential before incorporating special tests that can be
more challenging to perform and interpret. These tests are often
misreported, poorly evaluated or simply overlooked, leaving clinicians
uncertain about their application and value. Most of us have
had the experience of being confounded when a learned colleague
expresses disbelief at our omission from the patient examination of
the ‘grabem-pullem-and-twistem’ test – the one perfect test guaranteed
to diagnose the particular condition facing you! Having familiarized
yourself with this manoeuvre and used it in practice, it is
sometimes not too long before you are questioning its value and
wondering whether it is any more helpful than any other test you
might have used. So, how do we decide which tests are actually
worth using and which ones should be discarded?
As clinicians who treat patients on a daily basis, we decided to
dig a bit deeper and look at the evidence base for as many additional
or special tests as we could. As our primary aim was to produce a
useful guide for clinicians, we selected tests which were either in
common use, had supporting evidence or which we and our colleagues
working in a particular specialty deemed helpful.
By gathering the tests in one place, describing the technique
in detail, presenting the evidence and mixing in clinical tips and
expert opinion, we hope in this book to inform your selection of
special tests, enable you to reproduce the manoeuvres reliably and
interpret the results meaningfully.
As you will see, there is a format which is repeated for each test;
we explain the background to each section below.
B Also known a s (aka)
Where we found several names for the same test we have listed the
less well known names in the ‘also known as’ (aka) section. The use
of eponyms in this area of medicine is commonplace and we have
continued the tradition by using them in cases where the test is best
known by this name. Inevitably, there are instances of regional and
professional differences in the name given to a particular test and
we have had to make a judgement in one or two cases. There has
been debate about the relative merits of using eponyms (Matteson
& Woywodt 2007, Whitworth 2007), but since their use is ingrained
in medical practice, in our view it would be perverse to refer to the
tests in any other way.
C Technique
Ground rules
There are a number of essential principles to keep in mind when
performing these special tests:
1 You have already taken a thorough history and performed a basic physical
examination.
Also known as (aka)
Introduction
From this, you may suspect a lesion in a joint or specific tissue.
Special tests are most likely to confirm or rule out what you already
suspect and no test or investigation can replace this initial ‘sifting’
process. It serves as a vital starting point that enables the clinician
to consider differential diagnoses before selecting the most appropriate
test(s)/investigation(s) needed for the target condition.
2 Know what is ‘normal’ for your patient.
Experienced clinicians will recognize that the range of joint
movement in the normal population varies hugely with age, sex
and body type – imagine the range of ‘normal’ hip movement
in a 15-year-old gymnast, a 30-year-old prop forward and a 75-
year-old former sumo wrestler if you have any doubts! In musculoskeletal
medicine we have the advantage in most patients that
their ‘normal’ will be obvious when examining the unaffected
side. For each test, we have made the assumption that the clinician
will first have carried out the manoeuvre on the unaffected
limb in order to ascertain the patient’s normal range of movement
and response to the procedure.
3 Be selective in which tests you use.
Some tests have good evidence to support their use while
others are advocated for use in certain circumstances by expert
clinicians. In many cases, the jury is still out, and the clinician
therefore needs to exercise great care when applying the tests and
interpreting the results. Opt for quality rather than quantity –
using a huge range of extra tests is invariably counterproductive.
At best you will end up feeling confused and, at worst, your
patient will end up being investigated and treated for the wrong
condition. Using a sensitive test as part of the screening examination
for a condition (a sensitive test performed well allows you
to rule out a condition if the test is negative) and a highly specific
test to confirm a suspected diagnosis (if this is positive you
can be more confident that the condition is present) will help you
target investigations and plan treatment.
4 Use the same tests regularly.
Practice makes perfect. The more exposure you get to a particular
patient group or target condition, the more proficient you
will become at performing the test and interpreting the results.
Developing the technical skills necessary to perform the tests
well will improve the intra-tester reliability and increase your
confidence in the findings.
5 Remember that no test is diagnostic.
Of course, very few tests can be expected to conclusively rule
in or rule out any particular condition but they should add to
the index of suspicion which will inform your clinical reasoning
process.
How to do the tests
Performing the tests accurately and consistently is vital, and again
there are a few basic suggestions on how you can do this:
1 Make sure the patient is comfortable.
This will help minimize spasm, pain and voluntary muscle
activity. If the patient is apprehensive and unable to relax, performing
the test will be difficult and the result unreliable.
2 Keep yourself relaxed and comfortable.
Bony thumbs or an overly tight grip of the soft tissues is not a
pleasant experience and this will prevent the patient from relaxing
adequately. Wherever possible, keep your hands relaxed and
use them not only to carry out the manoeuvre but also to feel the
response of the tissues and patient. Ensure your position facilitates
a mechanical advantage to enable you to provide appropriate
support or resistance with the minimum effort.
3 Interpret the findings correctly.
Does the test replicate the patient’s presenting symptoms,
whether this is pain (site and nature), apprehension, clicking,
locking or paraesthesia?
4 Modify the technique depending on the condition.
In an acutely painful, severe or recent injury, make sure you
carefully apply passive or resisted stresses gradually to avoid
unnecessary provocation of pain and possible disruption of the
healing breach. In mild or chronic cases, if the initial test is painfree,
repeat the manoeuvre more strongly to replicate more vigorous
functioning of the tissue.
5 Modify the technique depending on the type of tissue.
It may be appropriate when testing inert structures such as
ligaments to start with a small amplitude movement so that an
early response to the test (i.e. pain and/or apprehension) can
be detected. Steadily increasing the amplitude, range and speed
of the test will then ensure that the capacity of the structure(s)
to restrict excessive joint play at the end-range of movement
Also known as (aka)
Introduction
has been properly evaluated. Normal ligament brings the
movement to a firm ‘stop’ and loss of this, accompanied by
increased range, is suggestive of joint instability. When testing
contractile structures, resistance to movement should be
increased slowly at first to allow assessment of the patient’s
pain and any apprehension to the test. The amount of resistance
can then gradually be increased, allowing full strength to be
assessed. Depending on the relative size and strength of the clinician
and patient, the clinician can vary the resistance applied
by using a short or long lever.
lD Clinica l c ontext
In presenting the clinical context for each test, the following search
criteria were used, resulting in a number of considerations being
made as we appraised and interpreted the evidence:
Searching the literature
Literature searches were conducted for articles in English using
the National Library for Health, Google scholar, AMED (1985 to
present), CINAHL (1982 to present), EMBASE (1974 to present)
and MEDLINE (1966 to present). Keywords and combinations of
keywords were used: diagnostic, diagnosis, examination, the joint
or structure involved (e.g. shoulder), likelihood ratio, manoeuvre,
name of test, probability, sensitivity, sign, specificity, test. A secondary
search was completed from reference lists in published
articles.
Appraising the evidence
My students are dismayed when I say to them ‘half of what you are
taught as medical students will in 10 years have been shown to be
wrong. And the trouble is none of your teachers knows which half.’
Dr Sydney Burwell, Dean of Harvard Medical School (Sackett et al 2000)
As we read through the literature, it became apparent that the
top hierarchy of evidence was lacking; with very few systematic
reviews or prospective blind comparisons that used a relevant
reference standard or a consecutive series of patients from a relevant
clinical population (Fritz & Wainner 2001). Of the available
evidence, it was naturally difficult to work out whether we were
looking at the half of the evidence which will have been proven correct
in 10 years, or the half which will not!
There are several questions that can help to identify the validity
of a study (Heneghan & Badenoch 2006, Jaeschke et al 1994, Sackett
et al 2000), and examining the answers to these questions helped us
to identify broad and repeated themes that influence the outcome
of research studies examining the reliability of MSK special tests
which deserve cognisance.
1 Is there a clearly defined question?
Most studies had a clearly defined research question as well
as specifying the target condition, which is important given the
propensity of some tests to stress more than one structure (e.g.
the active compression test at the shoulder stresses both the
acromioclavicular joint (ACJ) and the glenoid labrum but is more
specific for ACJ lesions).
2 Has the diagnostic test being examined been compared to an appropriate
reference standard?
A reference or gold standard provides confirmation that
the condition in question is present or absent. It is the measure
against which all other tests for that condition are evaluated and
so in itself should have proven accuracy. Unfortunately, this does
not exist uniformly in MSK medicine and the results of MRI or
surgery are often as close to the gold standard as we can get.
In situations where the reference standard has demonstrated
validity, and when it either forms part of the accepted diagnostic
process or treatment, its use in research is fairly straightforward.
A good example is the use of arthroscopy in meniscal lesions at
the knee where it not only serves a purpose in delivering appropriate
surgical treatment but coincidentally provides the most
accurate way of diagnosing a meniscal lesion. It therefore provides
an excellent reference standard against which preoperative
physical tests (such as McMurray’s) can be measured. However,
other conditions and injuries provide a stiffer challenge. What
would be a suitable reference standard in the case of grade I
medial collateral ligament sprain at the knee? The valgus stress
test is widely accepted as a reliable diagnostic test although using
Clinical context
Introduction
arthroscopy in this instance would clearly be inappropriate and
unlikely to make it past the ethics committee! As a result, the valgus
stress test at the knee is a widely used and accepted test but
has no evidence to support its use.
A further problem can be encountered if the reference standard
selected has not itself been evaluated sufficiently before
being used as a measure. For example, in early studies on
femoro-acetabular impingement at the hip, physical MSK tests
were measured against findings on MRI and MR arthrography,
the results of which are now known to correlate only moderately
with those of arthroscopy, the current, generally recognized gold
standard for this condition (Malanga & Nadler 2006).
The lack of an appropriate and/or validated reference standard
does appear to be a factor in the lack of evidence and this
may be a reason why so many of the ‘bread and butter’ tests
used in everyday practice have no evidence to support their
use, for example tennis and golfer’s elbow and simple ligament
sprains at the knee and ankle.
3 Has the diagnostic test been evaluated on a spectrum of patients?
Ideally the patients studied should cover the full spectrum of
the condition, replicating the range in the population on whom
the test would be used in practice. However, the vast majority
of studies on MSK tests are done in a hospital setting and are
therefore likely to be carried out on subjects with more advanced
pathology than that encountered in primary care. This may contribute
towards a reduction in the number of false positives and
true negatives in the sample, thereby over-estimating sensitivity
and under-estimating specificity which, in turn, generates a spectrum
bias.
Patients with co-morbidities that mimic the target condition
should also be included, again replicating the circumstances
in which the test is likely to be used, rather than evaluating the
results against asymptomatic patients. The accuracy of the test varies
significantly when tested in patients with other co-morbidities
(e.g. the use of McMurray’s test in patients with a history of a
meniscal tear against those with underlying osteoarthritis).
4 Has the reference standard been applied to all patients?
To achieve stringent data on the accuracy of a test, the reference
standard should be applied to all subjects, including those
who have a negative test (and are therefore considered unlikely
to have the condition being targeted). Application of the reference
standard, however, could carry unacceptable risk and cost.
Thus, in many studies of diagnostic accuracy, patients with negative
test results are subjected to alternative, less invasive but
sub-optimal ‘reference standards’ or may not be subjected to
a reference test at all, preventing true negatives and, therefore,
specificity from being calculated.
5 Have the intended use, the physical performance and the definition of a
positive and negative test been explained?
It is interesting to see how tests evolve over time and change
from the original description. Adaptations of tests are often used
in studies without making this explicit or describing the test adequately
so that it can be replicated by other workers. The use of
an adapted test means that comparison with other studies that
used the original or other adaptations is not possible and these
issues are rife in studies of MSK diagnostic tests.
6 Is the test validated in independent groups of patients?
It is worth checking whether the original description of a
test’s performance is validated by independent workers as the
originator(s) of a test often report high levels of diagnostic accuracy
that are not reproduced by independent researchers.
7 How can you apply the evidence to your own patients?
Applying the results of the test to your patient population
needs to be done with caution as there are many variables to be
taken into consideration: patient population, location of study
(i.e. primary or secondary care), skill level of clinicians performing
the test, etc. Will the test(s) assist in the management of your
patient population and lead to improved diagnostic accuracy
and treatment?