What practical skills do rheumatologists of the future need?

BaillieÁre's Clinical Rheumatology Vol. 14, No. 4, pp. 635±648, 2000

doi:10.1053/berh.2000.0104, available online at http://www.idealibrary.com on

3 What practical skills do rheumatologists of the future need? Alexander Fraser

MB, BCh, MRCPI

Specialist Registrar

Douglas J. Veale*

MD, FRCPI, FRCP(Lond)

Senior Lecturer, Honorary Consultant Rheumatology Rehabilitation Research Unit, Department of Rheumatology, Old Home, Leeds General In®rmary, Great George Street, Leeds, LS1 3EX, UK

In this chapter, we consider future practical skills required for rheumatologists. While dicult to predict against a background of rapid technological advance and successive changes to health-care provision world wide, a number of questions are examined. The ®rst question is what core skills are essential in the curriculum? This has been addressed in at least one joint European e€ort by UEMS. Great diversity in both clinical practice and training was found across Europe; clearly, the di€erence across continents may prove even more signi®cant. Second, the role of arthroscopy is considered, the evidence for its therapeutic bene®t in clinical rheumatology practice being inconclusive. Issues concerning diagnostic methods including electrophysiology and ultrasound (US) are also discussed in this chapter. There is evidence to support the use of electrophysiology in routine diagnosis for speci®c diseases. US has become popular as technology improves. It is cheap but highly operator dependent, and the feasibility of rheumatologists using US in the clinic remains to be proved. In conclusion, health care is changing rapidly, and training must adapt, and is adapting, to meet its challenges. A number of opportunities will present to the rheumatologist of the future, but the feasibility of these in routine clinical practice remains to be seen. Key words: practical skills; training/curriculum; arthroscopy; imaging.

Rheumatology has grown rapidly as a specialty in recent years, with an expansion in consultant and specialist training numbers, now estimated at 1400 in the UK. During the same short period, there have been a number of signi®cant changes in the health service not just in the UK, but also in other European countries. Greater integration within Europe, at social and cultural levels in addition to political, has also prompted strategies to examine di€erences in the education and training of rheumatologists across the continent. This process has occurred at the level of undergraduate education and postgraduate training, the latter resulting in the production of the Union EuropeÂenne des Medecins Specialistes (UEMS) core curriculum.1 In North America *Corresponding author. 1521±6942/00/040635+14 $35.00/00

c 2000 Harcourt Publishers Ltd. *

636 A. Fraser and D. J. Veale

there has also been a process of examining perceived need in relation to the adequacy of training.2 Technological advances have also expanded explosively in both the clinical and the laboratory setting, leading to a rapid increase in the potential, and wider availability, of many investigative techniques. Some of these techniques can easily be acquired, but for some validation remains to be completed, and the training of people in the skills required may need further development. While there has been a great emphasis on generic or transferable skills at undergraduate level, which is to be welcomed, the speci®c and practical skills relate more to postgraduate training. Finally, the advent of a new millennium is clearly a time to focus the mind on what is required in the future. WHAT CORE SKILLS ARE AN ESSENTIAL PART OF THE RHEUMATOLOGY CURRICULUM? The curriculum for higher specialist training in rheumatology has recently been the subject of a procedure to identify and de®ne more clearly a uni®ed approach across Europe. A core curriculum has been prepared in a document produced by the European Board of Rheumatology (UEMS). The approach by the Royal College of Physicians and Surgeons of Canada as part of their CanMEDS 2000 project took a broader interpretation of the curriculum. They analysed key competencies, which included skills relating to the roles of medical expert, communicator, collaborator, manager, health advocate, scholar and professional. This clearly covers the basic generic as well as speci®c skills required by doctors in general. The detail pertaining to the content of these curricula is described elsewhere in this book and will therefore not be covered further at this point. The area we will discuss, however, is that of the clinical skills required by the rheumatologist of the future. In the UEMS core curriculum, the skills are de®ned according to whether they are considered to be `core' or `optional' skills of the specialty (Table 1). The list of core skills is quite basic, and very few would argue that these skills should not be core. The list of optional skills, on the other hand, is lengthier and more controversial, probably re¯ecting the broad subspecialization across Europe. This curriculum does confer the advantage of o€ering broadly inclusive guidelines, which provides a uni®ed system while at the same time allowing for local variation. It gives a clear indication of the skills required according to whether they are essential in the clinical practice of rheumatology. This is not only important for trainees, but also, in the present climate, has far-reaching implications for continuous medical education (CME) for rheumatologists in the longer term. There are indeed two questions of relevance here: whether or not such optional skills should be attained by a rheumatologist, and whether the techniques are useful whoever performs them. It is apparent by the number performing a variety of optional skills that there is a lack of consensus, which probably re¯ects a lack of evidence. Many of the techniques included in the optional list are interventional, and their usefulness remains to be proved; we will discuss a number of these in more detail later in this chapter. In summary, the emphasis lies on those skills required for clinical practice and not for research. The level of knowledge required for the interpretation of their results and their application to the clinical problem has not been addressed and is more dicult to de®ne. In addition, while the UEMS curriculum describes the content, it fails to address issues in curriculum design such as methods of training. This is an aspect, however, that the General Medical Council in the UK and the UK Royal College of Physicians have

Practical skills for the future 637 Table 1. Skills required by a rheumatologist according to the UEMS curriculum.1 Core skills

Optional skills

To be able to aspirate and inject into joints and analyse synovial ¯uid To perform a needle biopsy of synovium To perform soft tissue injections To acquire counselling and communication skills To acquire management skills in running a rheumatology unit To recognize the value of audit methodology and speci®c outcome measures

To perform a biopsy of other structures, e.g. skin, fat, bone, muscle and salivary gland To perform and interpret bone densitometry To perform and interpret ultrasonographic examination for imaging the musculoskeletal system To perform and interpret electromyographic studies To perform arthroscopy To perform closed non-surgical synovectomy To perform epidural and regional nerve blocks To perform manipulation and mobilization techniques To perform intervertebral disc aspiration, injection or nucleolysis To perform and interpret capillaroscopy

emphasized in their approach.3 This has led to a renaissance of clinical methods in higher specialist training. WHAT DIFFERENCES EXIST IN TRAINING BETWEEN DIFFERENT COUNTRIES? The di€erence in undergraduate rheumatology education across countries world wide has been scrutinized in some depth at a meeting entitled `Undergraduate Education in Rheumatology', held in Singapore in 1997. In general, it can be seen in the UEMS curriculum that many of the di€erences that exist across Europe relate to the distinction of physical medicine and modern rheumatology. It can be seen, for example, that manipulation and mobilization techniques are included because in many central European cultures, including France, Germany and Italy, rheumatologists still practise physical medicine in addition to rheumatology. The trend in the UK and Ireland has been to separate rheumatology and physical or rehabilitation medicine into two distinct specialties. The di€erence at the postgraduate level is not as clearly de®ned in the literature. It is interesting to note that, in spite of the recent boom in Internet technology and the existence of developed web sites for rheumatology, including those of the American College of Rheumatology, the European League Against Rheumatism and the British Society for Rheumatology, training guidelines are not available. It is encouraging, however, that speci®c individuals and structured bodies are now being appointed by these groups to oversee the implementation of new initiatives to develop educational activities, which will hopefully address the spectrum from undergraduate to postgraduate education and on to CME, as well as involving patient education. WHAT IS THE ROLE OF ARTHROSCOPY, AND SHOULD THERAPEUTIC ARTHROSCOPY BE LEFT TO THE ORTHOPAEDIC SURGEONS? Phillip Bozzini designed the ®rst endoscope in 1806, but unfortunately, when he presented his paper to his colleagues, it was not well received, and his instrument was

638 A. Fraser and D. J. Veale

considered to have no clinical importance.4 The ®rst arthroscopy was not performed until 1918, when Professor Takagi of Tokyo examined a knee joint using a cystoscope 7.3 mm in diameter. The most important development in arthroscopy came in 1959 from Masaki Watanabe, when he introduced the Watanabe 21 arthroscope. The technological advances since this time have been remarkable, with ®bre-optic telescopes measuring as little as 1 mm in diameter allowing access to joints with as little invasiveness as an 18 gauge needle. The smallest-diameter rod lens telescope is 2.4 mm, which allows excellent resolution sucient for a full examination of the joint.5 Arthroscopy is therefore an established technique for assessing the intra-articular features of joints to evaluate synovitis and chondropathy6±11, to perform synovial membrane (SM) biopsy and joint lavage, and to administer therapy. It is an invasive procedure with the potential adverse e€ects of haemorrhage and infection. The major di€erence provided by modern technology is the size of the needle telescope, which enables the procedure to be performed safely under local anaesthesia and improves tolerability for the patient.5,6 In our experience, 88% of patients reported the arthroscopy to be as or more tolerable than expected; 68% experienced some pain (the mean pain on visual analogue scoring being 27/100 mm), related, in the majority, to the local anaesthetic. Reviews of a large number of studies have also reported arthroscopy to be acceptable to patients.9 Performing surgical procedures during arthroscopy is not feasible under these conditions; it is argued and widely accepted that unless the operator can extend the procedure to open surgery, more elaborate intervention cannot be justi®ed. Rheumatologists have taken great interest in arthroscopic examination from an early time, and this continues today in specialist centres in France, the UK, Sweden and the USA. All of these groups, in association with the International League Against Rheumatism, are currently involved in devising guidelines for training rheumatologists in arthroscopy. This is an issue of some controversy as there have been at least two attempts to produce guidelines for practice, one published by the Committee on Ethics and Standards and the Board of Directors of the Arthroscopy Association of North America12, the other by a group of French Societies involved in musculoskeletal medicine and radiology.13 The North American guidelines suggested a distinction between diagnostic and operative arthroscopy, while the French group suggested that any practitioner could perform arthroscopy as long as his or her training and continuing education were satisfactory and rigorous conditions of aseptic technique were applied. The French group also made the point that arthroscopy was ®rst performed by rheumatologists9, only subsequently being adopted by orthopaedic surgeons and physical therapists. It is important to remember that the type of patient undergoing arthroscopic examination by a rheumatologist is the arthritis patient, whereas orthopaedic surgeons predominantly scope post-traumatic injuries. This realization creates, in most clinical situations, a good collaboration between `medical' and `surgical' colleagues, which has led to a better understanding for all.

Clinical role of arthroscopy The diagnostic bene®t of arthroscopic examination by direct visualization is not universally accepted, but we have recently reported distinct vascular patterns that correlate to speci®c diagnoses.14 This has been supported by data from Fiocco et al and Kane et al (personal communications 1998).

Practical skills for the future 639

There are very few controlled studies that have examined the role of arthroscopic intervention as a treatment for rheumatic disease. One such study failed to show any signi®cant bene®t of arthroscopic surgery over tidal irrigation of the knee joint for osteo-arthritis.15 The number of patients included in these studies is small, which limits the extent to which the results can be interpreted. We are currently analysing the results of over 700 arthroscopic procedures performed at Leeds between 1995 and 1999, the main clinical indications for arthroscopy being osteo-arthritis of the knee, followed by persistent in¯ammation of the knee or wrist joint in rheumatoid arthritis (Table 2).

Table 2. Arthroscopic procedures performed at Leeds from 1995±99. Total no. of arthroscopies

Diagnoses

645

RA OA PsA Seroneg. Crystal Misc.

267 206 51 44 18 59

Joints examined

Primary indications

MCP/PIP Wrist Knee

Research 286 Therapeutic 134 Clinical study 144

37 34 571

PsA ˆ psoriatic arthritis; Seroneg. ˆ seronegative arthritis; Misc. ˆ uncon®rmed; MCP/PIP ˆ metacarpophalangeal/proximal interphalangeal joint.

Research role of arthroscopy Recent technological developments as outlined above, linked with improved therapeutic potential, provide the rationale for the study of early arthritis, but this is still at the stage of research potential. Diagnostic technologies including needle arthroscopy, high-frequency ultrasonography (US) and magnetic resonance imaging (MRI) have developed rapidly, but a role in clinical practice remains unprove. A comparison of di€erent imaging modalities will provide useful information that may allow the `calibration' of techniques and a greater understanding of the pathogenesis. It could be argued that, for rheumatologists of the future to interpret fully results of such investigations, they require some ®rst-hand experience.

Synovial examination The examination of the synovium by direct visualization is only one aspect of our understanding of the numerous types of arthritis, microscopic examination being very much more complex. It may provide an even greater understanding in the future as new laboratory technologies allow analysis on a genetic as well as a molecular and cellular level. A great deal of e€ort has been invested in the standardization of arthroscopic technique of biopsy and the subsequent processing and analysis of synovial tissue. It has been shown that biopsies taken using a blind needle are quite comparable to arthroscopic biopsies, although some di€erences appeared when the site of tissue acquisition was the cartilage±pannus region.16 This is evidently a critical site in the pathological process in rheumatoid arthritis, so it may be necessary to be stringent with regard to this ®nding.

640 A. Fraser and D. J. Veale

Comparison with other imaging modalities How do arthroscopic ®ndings match other imaging modalities such as US and MRI? In the ®rst study of its kind, we compared same-day, sequential MRI and arthroscopy of the knee joint in rheumatoid arthritis patients17, demonstrating a good correlation at two di€erent time points between the techniques. A previous report of nonsequential arthroscopy and MR by Ostendorf et al18 suggested that arthroscopy might be superior in detecting synovitis in early disease. Some reports have suggested MRI to be 90% accurate, although in double-blind design studies, it has only achieved 75% and 78% accuracy.10 A comparison of arthroscopic assessment with high-resolution US is currently underway. Assessment of response to therapy The sequential imaging study demonstrated the ability of arthroscopy and MRI reliably to detect changes in response to therapy, but it is argued that arthroscopic lavage may itself be therapeutic. The dose±response e€ect seen in this study, the greatest improvement occurring in the high-dose patients and none in the placebo patients, suggests that the changes observed were real, re¯ecting a drug e€ect rather than a response to lavage. A number of studies of drug therapy using this approach are currently underway, which should allow a further validation of this approach. Finally, novel gene therapy techniques can be uniquely monitored in this way. In summary, rheumatologists now recognize the need to make a clear diagnosis, assess severity and introduce therapy at the earliest possible stage of the disease19, as well as to monitor the response to that therapy in the patient using the best available technology. This is especially relevant as new therapies are evaluated in the treatment of arthritis. Arthroscopy is not new4, and rheumatologists' interest in it is growing.5±11 New improved technology makes it less invasive and more informative, and the patients are not the same post-traumatic cases that the orthopaedic surgeon scopes. In the early phase of the disease, the in¯ammatory changes are often con®ned to the hand or wrist joints, while knee synovitis occurs late. It may thus prove essential to access small joints if we wish to examine the pathogenesis at the primary site of the disease. A further development in small-bore needle arthroscopes should improve access to small and large joints while minimizing invasiveness. Finally, the `calibration' of arthroscopy and SM histology with non-invasive imaging such as MRI and highresolution US will be crucial in the development of these techniques for the study of the pathogenesis of rheumatoid arthritis and psoriatic arthritis. IS THERE ANY PLACE FOR ELECTRODIAGNOSIS IN RHEUMATOLOGY TRAINING? Patients may present to the rheumatologist with a variety of neuromuscular symptoms such as pain, sti€ness, weakness and/or sensory symptoms. A detailed history and examination will in most cases lead to the diagnosis, but sometimes this does not occur and help may be required to de®ne an abnormality or localize a lesion. In such circumstances, the appropriate application of electrophysiological investigation may be a useful aid to diagnosis. However, do we as rheumatologists need speci®c training in this technology, or should we merely have an understanding of the procedure, the results and their relevance?

Practical skills for the future 641

The two main tools employed in the electrophysiological investigation of musculoskeletal disease are nerve conduction studies (NCS) and needle electromyography (EMG). NCS involve the application of a brief electrical stimulus to a peripheral nerve and the recording of the subsequent motor or sensory response at surface electrodes. EMG is the recording of the electrical activity of a contracting muscle by the insertion of a needle electrode into the body of that muscle. Normal resting muscle is electrically silent, but when they contract, activated muscle units produce an electrical signal. The recorded activities of motor units may be analysed individually or summated. Nerve conduction studies The most common nerve entrapment syndrome is median nerve compression at the wrist, resulting in the clinical entity of carpal tunnel syndrome. This disorder is commonly investigated and characterized using electrophysiology. The use of NCS in the study of carpal tunnel syndrome has been criticized for lacking sensitivity. Newer techniques such as high-resolution US are being investigated for applications in this ®eld20, but NCS remains the investigation of choice for nerve entrapment phenomena of not only the median nerve, but also the ulnar and common peroneal nerves. Peripheral neuropathies of diverse cause may be characterized by NCS (Table 3), and more subtle changes may be helpful in de®ning the diagnosis, aetiology and occasionally prognosis. Radiculopathies are frequently encountered in the rheumatology clinic, primarily in relation to cervical or lumbar intervertebral disc herniation, presenting clinically as referred pain, sensory symptoms and weakness. The majority of patients will not require electrophysiological assessment as the diagnosis may be clinically evident or be demonstrated by appropriate imaging. There are, however, occasions on which NCS and/or EMG may contribute to the diagnosis and to the development of an e€ective treatment programme.21 Table 3. Demyelinating and axonal peripheral neuropathies. Demyelinating

Axonal

Guillain±Barre syndrome Hereditary neuropathies Paraproteinaemic neuropathy Leukodystrophy

Toxic (alcohol, drugs) Diabetic neuropathy

Electromyography The diagnosis of primary muscle disease is often complex, depending upon the pattern of numerous diagnostic criteria.22,23 Needle EMG is a useful tool in this respect and frequently distinguishes myopathy from other causes of muscle weakness, possibly even identifying the distribution and severity of the disease. EMG is not, however, useful for the elucidation of a speci®c diagnosis, nor is it useful in the longitudinal assessment of disease progression or treatment response. Furthermore, acquired and inherited muscle diseases show similar EMG changes.21 The appropriate application of electrophysiological investigation can be useful, but how much does the practising clinician need to know about it, and does it need to be an integral part of training?

642 A. Fraser and D. J. Veale

Electrophysiology and the rheumatologist The practice and interpretation of electrophysiological studies is a time-consuming and highly specialized area. The level of skill required can only be obtained from an in-depth training and many years of experience to provide reliable, repeatable and worthwhile results. Therefore, for rheumatologists routinely to undertake a training in its practice would be neither useful nor ecient. An appreciation, however, of the basic principles, the areas in which it may aid in diagnosis and, more importantly, its limitations, would appear to be sucient for the investigating rheumatologist. Muscle biopsy Muscle disease is a common feature of many disorders seen in rheumatology practice. This may be primary (e.g. polymyositis), inherited (e.g. McArdle's syndrome) or most commonly secondary (e.g. rheumatoid arthritis and systemic lupus erythematosus). The muscle disease is not always a priority in the light of other potentially life-threatening end-organ involvement. Muscle biopsy for histological, histochemical and biochemical analysis often but not invariably de®nes the nature and extent of disease as lesions of the muscle may be patchy. Traditional open biopsy is, however, invasive, painful and expensive, as well as requiring the co-operation of an interested and capable surgeon. Therefore, the limitations of EMG and an understandable reluctance to progress to open muscle biopsy do mean that a de®nitive diagnosis is occasionally not made despite the fact that treatment may be prolonged and aggressive. Percutaneous muscle biopsy has been used in the clinical and research setting for many years but is still not widely accepted. Several studies have assessed the performance of this method in comparison to open biopsy, and their ®ndings are consistent24±26 (Table 4). In conclusion, we believe that electrophysiological studies require highly specialized skills that put them beyond the routine practice of rheumatologists. Percutaneous muscle biopsy is, however, a simple, safe technique with considerable advantages over an open biopsy, so we suggest that all rheumatologists should be routinely trained to perform this procedure. Table 4. Advantages of percutaneous muscle biopsy. . . . . . . .

PMB can be performed in the outpatient clinic or day case unit The patient may be discharged as early as an hour after the procedure The procedure is well tolerated under local anaesthesia Cost is consequently kept to a minimum It is more ecient and safer than open biopsy It rarely fails to supply a tissue sample sucient for analysis It is a simple technique that is easily taught

SHOULD RHEUMATOLOGISTS BE CAPABLE OF USING US IMAGING? Technological advance has been staggering in the past century and at the turn of the new century, and this has profound implications for the practice of clinical rheumatology. High-resolution ultrasound (HRUS) for musculoskeletal disease may

Practical skills for the future 643

potentially alter the practice and structure of clinical rheumatology for the future. Should rheumatologists perform HRUS in the clinic, or should we leave this to radiologists? US has been utilized for the diagnosis and treatment of soft tissue and bony disorders since the early 1970s, but its resolution and imaging quality have been variable. Recent improvements in technology have led to a gain in its popularity in the rheumatological and radiological specialties as a research tool and as an extension of clinical examination in routine practice. The explosion of interest in the use of musculoskeletal US is a function of advances in technology allied with reduced cost.27±29 These factors together have resulted in a greater availability of this technology to the practising rheumatologist. The advent of high-resolution transducers and real-time imaging has exposed the true potential of musculoskeletal US. Technological advances (e.g. digital beam formers, two-dimensional arrays and non-linear techniques) continue and will undoubtedly contribute to improved resolution and consequently a greater utility of US in the future.28 The proposed bene®ts of US are currently (a) that it o€ers a good resolution of super®cial musculoskeletal structures; (b) that it is inexpensive compared with other imaging modalities; and (c) that it is generally available and causes no known incidental injury or damage. US does therefore appear to have clinical applications over and above those o€ered by traditional plain radiography and MRI, although its place is still a topic of hot debate. There are, however, limitations in its use; as an emerging technique, many of its applications continue to be de®ned in specialist centres, and comparison with other new imaging techniques is ongoing. In addition, US provides poor access to certain joint surfaces, such as the retropatellar area and deep interdigital web spaces, as a result of blind spots. The biggest problem facing the widespread deployment of US in the clinical setting is, however, training. The acquisition of sucient imaging and diagnostic skills by the clinician to allow interpretation and intervention to be safe and worthwhile must be ensured. US is highly operator dependent, and there is a steep learning curve in its use, certain minimal skills being required before it can become useful. Rheumatological applications of musculoskeletal US The potential applications of US in the ®eld of musculoskeletal medicine are protean, but the current main areas of interest are outlined in Table 5. The evidence supporting these applications remains limited but is expanding at an exponential rate. The `rheumatologist's extended ®nger' is a recently coined phrase stressing the role of US as a bedside tool for use by the practising clinician.30 In obstetric medicine, the acknowledged superior sensitivity of US over clinical examination for the de®nition of certain soft tissue structures has meant that it has become an intrinsic part of intrauterine fetal assessment. There is increasing evidence to suggest that there are similar

Table 5. Current uses of ultrasound in musculoskeletal disease . . . . . .

As an extension to clinical examination in the evaluation of articular and peri-articular disease For the detection and objective quanti®cation of synovitis in in¯ammatory arthritis In the assessment of ligament, tendon, muscle and nerve integrity For the isolation and aspiration of ¯uid collections In the detection of loose bodies For guiding procedures, e.g. injections

644 A. Fraser and D. J. Veale

bene®ts to clinical sensitivity in the ®eld of rheumatology, which promotes the concept of the addition of this tool to the clinician's existing armamentarium. Work in progress in our unit has been aimed at determining the impact of US on clinical diagnosis, including disease and anatomical site, and its e€ect, if any, on management. Patients have undergone standard clinical assessment, and a diagnosis and management plan has been proposed. Following this, a US scan of the clinically a€ected area has been undertaken in the outpatient department and its e€ect on diagnosis and management recorded. To date, 52 patients have been studied. In 28 (54%), the diagnosis has been altered by US (2 in terms of disease, and 26 in terms of anatomical site), and the proposed management has been altered following US in 29 (56%). This preliminary work suggests that US has an impact on diagnosis and management in the majority of patients scanned. However, a larger number of patients needs to be studied, and recruitment is ongoing, with more speci®c outcome measures being employed. Recent studies of US for the detection of synovitis have demonstrated increased sensitivity over clinical examination, synovitis being detected in over 10 times more metatarsophalangeal joints using US.31 Further studies have suggested that US is sensitive enough to detect a change in synovial activity with good intra- and inter-observer variability values. In the clinical setting, a not uncommon dilemma occurs when intraarticular ¯uid and synovitis cannot be distinguished by examination. Ultrasonographically, these two entities are distinct, and the application of graded compression helps in their di€erentiation. Tendons, ligaments and nerves represent structures that are widely accepted as being readily assessed using US, and high-frequency probes depict intricate tendonous, ligamentous and neural structures, furthermore o€ering the opportunity for their dynamic assessment and for a quick and easy comparison between sides. Fluid has distinct US properties that make it easily located, be it in tendons, bursae, muscles or joints, but the di€erentiation between ¯uid types (serous, blood or septic) requires aspiration and analysis. Intra-articular loose bodies are a common clinical problem, presenting with locking, giving away, limitation of movement, premature osteo-arthritis or e€usion. The clinical indicators of loose bodies are non-speci®c so their diagnosis is frequently based on imaging ®ndings. Computed tomography scanning and MRI are commonly employed for this purpose, but US o€ers cheaper, more accessible imaging, together with the previously mentioned added bene®ts of dynamic assessment and ready symmetrical comparison. The use of US in the guidance of therapeutic and diagnostic procedures has, not surprisingly, generated a great deal of interest. Arthrocentesis and other diagnostic aspiration procedures may be accurately guided by US. Fluoroscopic techniques have traditionally been used in this role but o€er less visualization of needle location, an increased exposure to ionizing radiation and poorer soft tissue assessment.32 The guidance of anaesthetic injections to larger joints such as the hip has been found to be useful, and US has also been used to guide the biopsy of erosions and in¯amed entheseal sites.33 Therapeutic procedures may also be enhanced by the use of US. It has been demonstrated that injections guided by clinical parameters are inaccurate, irrespective of the experience of the practitioner.34 Accuracy can be improved by the use of US, and this improvement in placement has been shown to translate into an improved clinical outcome.35 Bene®ts have also been noted in the injection of in¯ammatory heel pain, rotator cu€ calci®cation, plantar fasciitis and retrocalcaneal bursae.35,36

Practical skills for the future 645

Learning musculoskeletal ultrasound There are, however, certain basic requirements for the useful application of musculoskeletal US: 1. 2. 3. 4.

a sound appreciation of the technical foundations of US; an extensive knowledge of the normal anatomy of the human skeleton; an ability to recognize the most commonly encountered pathological conditions; an ability safely and accurately to determine when the parameters of normal architecture have been breached.

The probability is that the current trend towards the wider clinical application of this tool will continue, and at an accelerated pace. If this is to be the case, however, certain questions need now be addressed regarding its safe and ecient introduction. The expanding list of clinical applications for US in musculoskeletal disease, and the emerging realization that its dynamic application in the outpatient department maximizes its potential, has meant not only a shift from radiology to rheumatology outpatient department, but also from radiologist to rheumatologist. Rheumatology and radiology have traditionally been close allies in the diagnosis and treatment of disease. This new dawn for US requires that this relationship is still closer, and the recent advances we have discussed have come only as a reuslt of this medical symbiosis; its future is similarly dependent. That high-resolution US may become as important to clinical rheumatology as the echocardiograph has become to cardiologists is probable, but this will only occur with the co-operation of the radiology fraternity, particularly in the task of teaching clinicians to a safe and useful level of competence. It may well be that only when this technique has become an integral part of clinical diagnosis and treatment will its full potential be achieved. The clinical application of musculoskeletal US is currently gaining a foothold in the teaching hospital research unit. The next step will be its dissemination to smaller clinical units, and from there to individual clinicians' consulting rooms. One of the main driving forces behind these developments may be our patients themselves. Patients are increasingly better educated about their health, about services and about what treatment is potentially available to them. It may be that patient expectations require clinicians to provide the best tools available for diagnosis and treatment, and such pressure will require us to provide this service. In summary, the steep learning curve required to obtain sucient pro®ciency in this technique and to prevent diagnostic errors means that a great deal of application is required from trainee and teacher alike. It may be argued that the subtle nuances of minor tendon and joint pathology are dicult to detect using US without years of experience. Conversely, however, it may be argued that the purpose of any training programme would be to impart a basic knowledge. This would include an understanding of the technical foundations of US, an ability to recognize gross structural abnormalities and to locate ¯uid collections and areas of in¯ammation and, ®nally, the ability to employ US in the guidance of certain frequently employed therapeutic procedures. Armed with these basic skills, clinicians, fully aware of their limitations, would be in a position to practise US safely in appropriate circumstances and with time to gain experience and to develop their skills. It would appear intuitive that the potential bene®ts of musculoskeletal US for the rheumatologist are set to increase with time. Therefore the inevitable arrival of the US machine in the outpatient department needs to be well planned in order to avoid some of the well-publicized diculties encountered with the advent of new techniques

646 A. Fraser and D. J. Veale

such as endoscopic surgery, which require the adequate education and training of sta€ prior to their clinical application.

SUMMARY In a relatively short period of time, the specialty of rheumatology has grown rapidly, and there has been an explosion of technology and unprecedented political and social change in both the UK and Europe. The area of education is being scrutinized at undergraduate and postgraduate levels, while CME has become a real issue. This provides a unique opportunity to develop a substantive curriculum for rheumatology training across countries. The role of arthroscopy as a treatment for rheumatic disease remains unclear, further controlled studies being necessary. Its role in research is clear, it is well tolerated by patients, and it allows direct access to joints and synovium for genetic studies, including gene therapy. Initial comparative studies of arthroscopy with MRI and US suggest that their results are similar, but it would appear that quanti®able changes may prove more complex to measure. HRUS o€ers good imaging of super®cial musculoskeletal structures and is inexpensive, with no adverse e€ects. The usefulness of HRUS is not in question, nor is the feasibility of rheumatologists performing it in the clinic, but it has not yet been established whether it is either ecient or cost-e€ective for them to do so. The technological development of US is not complete: machines will get smaller, more computerized and therefore more user-friendly, and this, along with patient demand, will clearly drive their use in the clinic. The practical skills needed by the rheumatologist of the future may thus be diverse; while arthroscopy and US are feasible, highly specialised skills such as electrophysiology are not realistic.

Practice points . rheumatologists arthroscope patients with arthritis whereas orthopaedic surgeons predominantly scope patients with post-traumatic injuries . few controlled studies have examined the role of arthroscopic intervention as a treatment for rheumatic diseases . 88% of patients reported the arthroscopy to be as or more tolerable than expected nerve conduction studies are primarily employed in the evaluation of: . nerve entrapment syndromes . peripheral neuropathies . lower motor neurone disorders ultrasound: . o€ers a good resolution of super®cial musculoskeletal structures . is inexpensive compared with other imaging modalities . is generally available and causes no known incidental injury or damage

Practical skills for the future 647

Research agenda research needs to address: . a comparison of di€erent imaging modalities . the genetic analysis of synovium . the monitoring of new therapy, e.g. gene technology

Acknowledgements We are grateful to Dr Richard Wake®eld, Professor Wayne Gibbon, Mr Michael Walsh (Leeds), Dr Sta€an Linblad (Stockholm), Dr Xavier Aryal (Paris), Dr Bob Ike (Minneapolis) and Dr JA Twomey. We also thank the ARC, Storz (UK) and ATL (Seattle, USA) for the equipment provided for the programme. The Rheumatology Rehabilitation Research Unit at Leeds is funded by the Arthritis Research Campaign.

REFERENCES 1. Woolf AD. The EULAR standing committee for Education and Training (EURORITS). Rheumatology in Europe 1995; 24(1): 31. 2. Jerome D & Tugwell P. Skills development in rheumatology undergraduate and postgraduate curricula. Journal of Rheumatology 1999; 26 (S55): 38±40. 3. Education Committee of the General Medical Council. Tomorrow's Doctors. London: Kiek & Read, 1993. 4. Dandy DJ (ed.). Arthroscopic management of the knee. In Current Problems in Orthopaedics, 2nd edn, pp 1±2. Edinburgh: Churchill Livingstone, 1987. * 5. Veale DJ. The role of arthroscopy in early arthritis. Clinical and Experimental Rheumatology 1999; 17: 37±38. 6. Lindblad S & Hedfors E. Intra-articular variation in synovitis: local macroscopic and microscopic signs of in¯ammatory activity are signi®cantly correlated. Arthritis and Rheumatism 1985; 28: 977±987. 7. Jayson MIV & St J Dixon A. Arthroscopy of the knee in rheumatic diseases. Annals of the Rheumatic Diseases 1968; 27: 503±511. * 8. Ayral X, Dougados M, Listrat V et al. Chondroscopy: a new method for scoring chondropathy. Seminars of Arthritis and Rheumatism 1993; 22: 289±297. 9. Ayral X & Dougados M. Rheumatological arthroscopy or research arthroscopy in rheumatology. British Journal of Rheumatology 1998; 37: 1039±1041. *10. Chang RW & Sharma L. Why a rheumatologist should be interested in arthroscopy. Arthritis and Rheumatism 1994; 37: 1573±1576. 11. Ike RW. Arthroscopy in rheumatology: a tool in search of a job. Journal of Rheumatology 1994; 21: 1987±1988. 12. Committee on Ethics and Standards, Board of Directors of the Arthroscopy Association of North America. Suggested guidelines for the practice of arthroscopic surgery. Arthroscopy 1994; 10. 13. SocieÂte Francaise de Chirurgie Orthopedique et Traumatologique, SocieÂte Francaise d'Arthroscopie, SocieÂte Francaise de Radiologie, SocieÂte Francaise de Rhumatologie, SocieÂte Francaise de Traumatologie du Sport, avec le participation de l'ANDEM. Conference de consensus: l'arthroscopie du genou. Texte du consensus, 1994. Paris: ANDEM. *14. Reece R, Canete J, Parsons W et al. Distinct vascular patterns of early synovitis in psoriatic, reactive and rheumatoid arthritis. Arthritis and Rheumatism 1999; 42(7): 1481±1484. 15. Chang RW, Falconer J, Stulberg SD et al. A randomized, controlled trial of arthroscopic surgery versus closed-needle joint lavage for patients with osteoarthritis of the knee. Arthritis and Rheumatism 1993; 36: 289±296. 16. Youssef P, Breedveld FC, Bresniham B et al. Quantitative microscopic analysis of in¯ammation in rheumatoid arthritis synovial membrane samples selected at arthroscopy compared with samples taken blindly by needle biopsy. Arthritis and Rheumatism 1998; 41(4): 663±669. *17. Veale DJ, Reece RJ, Orgles CS et al. Intra-articular therapy with anti-CD4 monoclonal antibody in RA: a magnetic resonance and arthroscopic study. Annals of the Rheumatic Diseases 1999; 58: 342±349.

648 A. Fraser and D. J. Veale 18. Ostendorf B, Dann P, Kocot D et al. Micro-arthroscopy versus MRI of MCP-joints in rheumatoid arthritis: comparing assessment of joint pathology detection. Arthritis and Rheumatism 1997; 40: S151 (supplement 9). 19. Emery P. The optimal management of early rheumatoid disease: the key to preventing disability. British Journal of Rheumatology 1994; 33: 765±768. *20. Lee D, van Holsbeeck MT, Janevski PK et al. Diagnosis of carpal tunnel syndrome: ultrasound versus electromyography. Radiology Clinics of North America 1999; 37(4): 859±873. *21. Twomey JA. Neurophysiology in musculoskeletal disease. Reports on Rheumatic Diseases 1998; 15: 1±4. 22. Payan J. Electromyography in polymyositis and some related disorders. Clinics of Rheumatic Diseases 1984; 10: 75±83. 23. Mechler F. In¯ammatory muscle diseases: electrodiagnosis. Rheumatology in Europe 1996; 25(2): 52±54. 24. Mubarak SJ, Chambers HG & Wenger DR. Percutaneous muscle biopsy in the diagnosis of neuromuscular disease. Journal of Pediatric Orthopedics 1992; 12(2): 191±196. *25. Haddad MG, West RL, Treadwell EL & Fraser DD. Diagnosis of in¯ammatory myopathy by percutaneous needle biopsy with demonstration of the focal nature of myositis. American Journal of Clinical Pathology 1994; 101(5): 661±664. 26. Dietrichson P, Coakley J, Smith PE et al. Conchotome and needle percutaneous biopsy of skeletal muscle. Journal of Neurology, Neurosurgery and Psychiatry 1987; 50(11): 1461±1467. 27. Gibbon WW. Musculoskeletal ultrasound. BaillieÁre's Clinical Rheumatology 1996; 10(4): 561±588. *28. Adler RS. Future and new developments in musculoskeletal ultrasound. Radiology Clinics of North America 1999; 37(4): 623±633. *29. Gibbon WW & Wake®eld RJ. Ultrasound in in¯ammatory disease. Radiology Clinics of North America 1999; 37(4): 633±651. 30. Manger B & Kalden J. Joint and connective tissue ultrasonography ± a rheumatological bedside procedure? Arthritis and Rheumatism 1995; 38: 736±742. 31. Wake®eld RJ, Karim Z, Conaghan PG et al. Sonography is more sensitive at detecting synovitis in the metatarsophalangeal joints than clinical examination. Arthritis and Rheumatism 1999; 42: S352. 32. Jonnes A, Regan M, Ledingham J et al. Importance of placement of intra-articular steroid injections. British Medical Journal 1993; 307: 1329±1330. 33. McGonagle D, Gibbon WW, O'Connor P et al. A preliminary study of ultrasound of bone erosion in early rheumatic arthritis. Rheumatology 1999; 38: 329±331. 34. Eustace JA, Brophy DP, Gibney RP et al. Comparison of the accuracy of steroid placement with clinical outcome in patients with shoulder symptoms. Annals of the Rheumatic Diseases 1997; 56: 59±63. 35. Cunnane G, Brophy DP, Gibney RG & Fitzgerald O. Diagnosis and treatment of heel pain in chronic in¯ammatory arthritis using ultrasound. Seminars in Arthritis and Rheumatism 1996; 25: 383±389. 36. Farin PU, Rasanen H, Jaroma H & Harju A. Rotator cu€ calci®cations: treatment with ultrasound-guided percutaneous needle aspiration and lavage. Skeletal Radiology 1996; 25: 551±554.