| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Rheumatology 2001; 40: 1060-1063
© 2001 British Society for Rheumatology
Original Papers |
The utilization of synovial fluid analysis in the UK
SF Crystals Laboratory, University of Bristol, Department of Anatomy, School of Veterinary Science, Southwell Street, Bristol BS2 8EJ, UK
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
Objective. To ascertain what use is being made of synovial fluid analysis in the UK, who is carrying out polarized light microscopy (PLM), and what confidence clinicians have in the results.
Subjects and methods. A postal survey was developed, piloted, adjusted and then posted to 535 people, 90% of whom were senior rheumatologists and 10% orthopaedic surgeons, whose names had been obtained from professional lists.
Results. Three hundred and eleven replies (59%) were obtained after 1 month. Analysis of the replies showed that only microbiological tests and PLM are used regularly, that these are used mainly for the diagnosis of acute arthritis, and that the majority of respondents would like data from these assays to be available within 24 h. The majority of the respondents (95%) had access to PLM and 80% said that clinicians should be able to use it. However, PLM is currently being undertaken by a variety of people: non-specialist technicians (34% of respondents), specialist technicians (35%) and clinicians (31%). Respondents were confident in microbiological assays but not in cell counts or PLM, unless they were undertaking it themselves.
Conclusions. There is an urgent need for guidelines, standardization and education about the use of synovial fluid assays in the UK.
KEY WORDS: Synovial fluid, Assays, Microbiology, Polarized light microscopy, Crystals.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Synovial fluid analysis is one of the few laboratory tests that is exclusively used for the diagnosis or assessment of musculoskeletal diseases. It came into use for the diagnosis and assessment of arthritis in the late 1950s, largely as a result of the work of Hollander et al. [1], who recommended the use of macroscopic examination, cell counts, microbiology and biochemical tests to differentiate the various forms of arthritis. In the early 1960s, Hollander and McCarty introduced polarized light microscopy (PLM) of synovial fluid to allow the identification of urate and pyrophosphate crystals [2, 3], and this technique soon became established as the definitive way of diagnosing gout or pseudogout. Subsequently, all rheumatology texts have recommended the use of synovial fluid analysis in the diagnosis of all forms of acute arthritis, as well as when there is uncertainty as to the cause of chronic arthritis [4–6]. An investigation by Eisenberg et al. has indicated that synovial fluid results can lead to a change in the diagnosis and treatment in many cases [7], suggesting that it is an important adjunct to other diagnostic clues.
However, in the UK, as in many other countries, synovial fluid analysis (with the exception of microbiological assays) has not been established as a routine procedure within hospital pathology departments. As a result, different people carry out the various tests in different hospitals and departments, and synovial fluid tests have not been subjected to the same rigorous quality control and validation procedures as other forms of laboratory investigation.
As part of a programme of research on the value of synovial fluid analysis, we undertook a survey of rheumatologists and orthopaedic surgeons in the UK in order to ascertain the current utilization of this diagnostic procedure. We were particularly keen to learn which of the various types of available assay were being used regularly, what people wanted from the tests, and who was carrying them out.
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
A survey of the literature was undertaken in order to establish the main types of test being undertaken on synovial fluid and the potential value of these tests. On the basis of this survey, we divided the main assays into five groups: microbiological, PLM, total and differential white cell counts, cytology [i.e. special stained preparations for differentiating between cell types (mast cells, Reiders cells, etc.)], and biochemical assays.
With these data available, a simple questionnaire was designed and subjected to pilot testing on 10 local rheumatologists. The questionnaire was revised as a result of this pilot work. It contained questions relating to the utilization of the five main categories of assay: whether the respondent made use of each (never, sometimes or often), whether they used them in acute or chronic arthritis, and how quickly they wanted the results (immediately, within 24 h or within days). We also asked questions about the service available to each respondent for the three most important categories of test (microbiology, crystals and cell counts), including: access to PLM (yes or no) and who was doing the assays (non-specialist technicians, specialist technicians or clinicians). In this context, ‘clinician’ is a rheumatologist or orthopaedic surgeon and ‘specialist’ is someone who had been specially trained in the test and given responsibility for it. We also asked questions about the level of confidence that the respondent had in the various types of assay (high or low) and whether or not they felt that clinicians should learn how to examine synovial fluid. The questionnaire was then posted to 535 potential respondents, 90% of whom were rheumatologists (names being taken from the British Society of Rheumatology membership list) and 10% orthopaedic surgeons (names being taken from the British Orthopaedic Association list of academic orthopaedic surgeons), with a request for return within 1 month in a stamped addressed envelope.
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Of the 535 questionnaires sent out, 311 (59%) were completed and returned within 1month. No reminders were sent out and we made no other attempt to increase this response rate.
The responses on the utilization of the five types of assay are shown in Table 1
. It is apparent that only microbiological tests and examination for crystals are undertaken often and that their major perceived value is in acute arthritis, although respondents varied greatly in their time requirements for the results. It should be noted that the figures in all columns are percentages of those responding to the question, and that many declined to give detailed answers to questions about cell assays, cytology or biochemical assays, presumably because they make little or no use of them.
|
Of the respondents who reported that they provided synovial fluid analysis, 34% said it was provided by non-specialist technicians [i.e. the service was provided by medical laboratory scientific officers (MLSOs)], 35% said it was provided by specialist technicians and 31% said it was provided by the clinician. Ninety-five per cent of respondents had access to PLM and 80% said that clinicians should be trained in synovial fluid examination for crystals and cell counts.
The results of the questions about the confidence that respondents had in the service available are shown in Table 2. It is clear that much more confidence was being expressed about microbiological assays than about cell counts or PLM.
|
We also analysed our data to see if there were some people who used all assays more than others. This appears to be the case. For example, significantly more of those who reported that they used crystal assays often also reported frequent use of microbiological tests and cell counts compared with those who looked for crystals occasionally (Table 3
).
|
We also found that very high confidence in the results of PLM and cell counts was more likely to be reported when the clinicians were doing it themselves (47 and 60% respectively) than when the results were obtained from non-specialist technicians (22 and 33% respectively, P<0.01)
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
This postal survey achieved a relatively high response rate (59%) after a single mailing, suggesting a good degree of interest in the subject. The fact that the sample was from the membership of specialist societies means that the results are probably fairly representative of the views of senior rheumatologists and orthopaedic surgeons in the UK.
One of our aims was to investigate the use of the various synovial fluid assays that are available. The data indicate that specialist clinicians make wide use of both microbiological assays (Gram staining and culture) and the search for synovial fluid crystals by PLM in order to aid in the diagnosis of acute arthritis.
Microbiology tests are promoted universally in the literature [8–10]. A positive Gram stain confirms infection and a positive culture gives a definitive diagnosis, identifying the causative organism and indicating the appropriate antibiotic to use [11]. Untreated or mistreated sepsis is life-threatening, and the tests, carried out by MLSOs in the pathology laboratory, are a priority among the provisions of the UK health service. Despite the known pitfalls, such as false negatives due to prior treatment with antibiotics, the confidence of specialist clinicians in hospital microbiology remains extremely high, much higher than for synovial fluid crystals or cell counts. Furthermore, there is no evidence that high-power microscopy is used regularly as an adjunct to traditional microbiology in the detection and identification of microorganisms [12].
Historically, the literature strongly supports the use of PLM for detecting and identifying crystals in synovial fluid [3, 13]. PLM for the detection and identification of synovial fluid monosodium urate monohydrate (MSUM) and calcium pyrophosphate dihydrate crystals provides a positive diagnosis of acute gout and pseudogout respectively and may lead to alterations in patient treatment [7]. Misdiagnosis and lack of appropriate treatment for gout can lead to long-term joint damage. The cost–effectiveness of this assay on a routine basis has een questioned recently [14], but the vast majority of our respondents had access to PLM, 70% of respondents used the test often and many used it for both acute and chronic arthritis. In the latter form of arthritis, detection of small numbers of MSUM crystals may be indicative of intercritical gout [15]. However, overall confidence in the results of the test is low. Several authors have questioned the reliability of results for synovial fluid crystal detection [16–19]; our results reinforce these findings and emphasize the need for improved standards in synovial fluid crystal identification [20].
A total and differential synovial fluid white cell count is recommended as helpful in distinguishing between disease categories (inflammatory vs non-inflammatory disease) [4]. However, the real value of cell counts in arthritis has never been validated, and our survey shows that cell counts were used less widely than either microbiological assays or PLM for crystals, and that confidence in the results is poor. Again, evidence in the literature supports the need for standardization of the procedure as well as further work to establish how useful it might be as an adjunct to diagnosis [11].
Despite the claims made in the literature for the usefulness of synovial fluid cytology [21–25], the test is used regularly by only 10% of clinicians. The provision of a dedicated service with expert specialist technicians is required for this work, and may not be considered cost–effective by the majority. Biochemistry assays, either those promulgated in the early literature, such as synovial fluid glucose and lactate dehydrogenase [26] or the more recently developed markers of disease progression [27, 28], are used very rarely
.
The second aim of the survey was to investigate the provision of non-microbiological tests in the UK. The data show that this is quite evenly split between non-specialist MLSOs, specialist technicians and the clinicians themselves. Our survey showed that when light microscopy for crystals, and cell counts, were carried out by clinicians, confidence in the results was much higher than when non-specialist MLSOs or specialist technicians were doing the tests. Yet the majority of the respondents considered that specialist, trained technicians should be providing this service. Not surprisingly, there are some clinicians who make more use of all the assays than others. Those who use the crystal assay are likely to use microbiology more often. This difference is unrelated to confidence in the results of the tests, as confidence in microbiology tests is high across the board. The trend may reflect differences in clinical practice between hospitals that differ in their reliance on clinical vs laboratory data.
The synovial fluid crystal assay is extremely important diagnostically: the synovial fluid cell count is a useful aid to diagnosis. It would be most unfortunate if these tests were to fall into disuse solely because clinicians have no confidence in the results, when straightforward steps could be taken to improve this situation. Shmerling pointed out in 1994 [11] that there was no consensus on what is a routine synovial fluid analysis, and this observation remains true today. There is an urgent need for guidelines, standardization and education regarding the use of these assays.
|
Acknowledgments |
|---|
We are grateful for the financial support of the Arthritis Research Campaign and the Medical Research Council. The Department of Social Medicine at the University of Bristol is the lead centre of the MRC Health Services Research Collaboration.
|
Notes |
|---|
Correspondence to: H. Amer.
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
- Hollander JL, Reginato A, Torralba TP. Examination of synovial fluid as a diagnostic aid in arthritis. Med Clin N Am1996;50:1281–93.
- McCarty DJ, Hollander JL. Identification of urate crystals in gouty synovial fluid. Ann Intern Med1961;54:452.
- McCarty DJ. Calcium pyrophosphate dihydrate deposition disease. Arthritis Rheum1976;19:275–85.
- Gatter RA, Schumacher HR. Microscopic findings under compensated polarized light and phase light. In: Practical handbook of joint fluid analysis, 2nd edn. Philadelphia: Lea & Febiger, 1991:45–58.
- Dieppe PA, Calvert P. Crystals and joint disease. London: Chapman and Hall, 1983.
- Guidelines and a proposed audit protocol for the initial management of an acute hot joint. In: report of a joint working group of the British Society for Rheumatology and the Research Unit of the Royal College of Physicians. J R Coll Phys Lond1992;26:83–5.[Medline]
- Eisenberg JM, Schumacher HR, Davidson PK, Kaufman AL. Usefulness of synovial fluid analysis in the evaluation of joint effusions: use of threshold analysis and likelihood ratios to assess a diagnostic test. Arch Intern Med1984;144:715–9.[Abstract]
- Ma DT, Carroll GJ. Monoarthropathy. Could this be infection? Review. Aust Fam Phys1998;271:28–31.
- Sack K. Monoarthritis: differential diagnosis. Am J Med1997;102(Suppl. 1A):30S–4S.[Medline]
-
Weston VC, Jones AC, Bradbury N et al. Clinical features and outcome of septic arthritis in a single UK Health District, 1982–1991. Ann Rheum Dis:1999;58:214–219.
[Abstract/Free Full Text] - Shmerling RH. Synovial fluid analysis: A critical reappraisal. Rheum Dis Clin N Am1994;20:503–12.[ISI][Medline]
- Freemont AJ, Denton J. Cytocentrifuge preparations. In: Atlas of synovial fluid cytopathology. Current histopathology, Vol. 18. Chap. 9: section 1. Dordrecht, The Netherlands: Kluwer Academic, 1991:57–60.
- Mandel NS, Mandel GS. Monosodium urate monohydrate, the gout culprit. J Am Chem Soc1976;98:2319–21.[Medline]
- Pal B, Nash J, Oppenheim B et al. Is routine synovial fluid analysis necessary? Lessons and recommendations from an audit. Rheumatol Int1999;18:181–2.[Medline]
-
Pascual E, Batile-Gualda E, Martinez A et al. Synovial fluid analysis for diagnosis of intercritical gout. Ann Intern Med1999;131:756–9.
[Abstract/Free Full Text] - Hasselbacher P. Variation in synovial fluid analyses by hospital laboratories. Arthritis Rheum1987;30:637–42.[ISI][Medline]
- Schumacher HR, Sieck MS, Rothfuss C et al. Reproducibility of synovial fluid analyses, a study among four laboratories. Arthritis Rheum1986;29:770–4.[Medline]
- McGill NW, York HF. Reproducibility of synovial fluid examination for crystals. Aust NZ J Med1991;34:710–3.
- Von Essen R, Hollta AMH. Quality control of the laboratory diagnosis of gout by synovial fluid microscopy. Scand J Rheumatol1990;19:232–4.[Medline]
- Swan AJ, Price G, Dieppe P. Practical steps towards improving the accuracy of identification of urate and pyrophosphate crystals in synovial fluids. Rheumatol Eur1999;28:111–2.
- Villaneuva TG, Schumacher HR. Cytological examination of synovial fluid. Diagn Cytopathol1987;4:141–7.
-
Broderick P, Coruese N, Pienik M et al. Exfoliative cytology interpretation of synovial fluid in joint disease. J Bone Joint Surg1976;58:396–9.
[Abstract/Free Full Text] - Naib ZM. Cytology of synovial fluids. Acta Cytol1973;17:299–309.[Medline]
- Peoc'h M, Dutet ML, Toscan du Plantier N et al. Cytology of fluids of articular puncture. Ann Pathol1998;18:213–7.[Medline]
- Freemont AJ. Synovial fluid analysis—its place, usefulness, indications and potential relevant findings. Rheumatol Eur1995;24:69–71.
-
Gobelet C, Gerster JC. Synovial fluid lactate levels in septic and non-septic arthritides. Ann Rheum Dis1984;43:742–5.
[Abstract/Free Full Text] -
Aman S, Risteli J, Luukkkainan R et al. The value of synovial fluid analysis in the assessment of knee joint destruction in arthritis in a 3-year follow up study. Ann Rheum Dis1999;58:559–62.
[Abstract/Free Full Text] - Myers SL. Synovial fluid markers of osteoarthritis. Rheum Dis Clin N Am1999;25:433–49.[Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  R. de Jonge, R. Brouwer, M. Smit, M. de Frankrijker-Merkestijn, R. J. E. M. Dolhain, J. M. W. Hazes, A. W. van Toorenenbergen, and J. Lindemans Automated counting of white blood cells in synovial fluid Rheumatology, February 1, 2004; 43(2): 170 - 173. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A Swan, H Amer, and P Dieppe The value of synovial fluid assays in the diagnosis of joint disease: a literature survey Ann Rheum Dis, June 1, 2002; 61(6): 493 - 498. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||