SMJ 2003: 48(1) 21-23
Clifford J Eastmond MD FRCPE FRCP†, Elizabeth M Robertson FRCR DMRD*
†Department
of Rheumatology
*Department
of Clinical Radiology
Correspondence
address:
Dr
CJ Eastmond
Department
of Rheumatology
Aberdeen
Royal Infirmary
Foresterhill
Aberdeen
Email:
clifford.eastmond@arh.grampian.scot.nhs.uk
Key
words: Seronegative spondarthritis
Abstract
This
study was designed to assess the relative values current of locally available
investigations in the early diagnosis of inflammatory sacroiliitis.
Introduction
Established
diagnostic criteria for ankylosing spondylitis are acknowledged to be inadequate
for early diagnosis (1,2). They rely heavily on the detection of radiological
sacroiliitis (3,4,5) a manifestation with high specificity but which may take
several years to develop (6,7). A number of attempts have been made to identify
methods capable of making an early diagnosis and which retain a satisfactorily
high degree of specificity (8,9).
The
present study compares several of these in a prospective cohort of patients with
a five year follow up period estimated to provide an adequate interval for the
development of radiological sacroiliitis. At the commencement of this study MRI
(10) imaging access was not locally available for this and therefore not
included. Previous local experience with quantitative isotope scintiscanning
(11) had been disappointing (unpublished data) and therefore was not included in
the current study.
The
comparisons used were Calin’s clinical questionnaire (12), standard
The
New York criteria for ankylosing were established for epidemiological studies.
Materials and Methods
Thirteen
consecutive patients were identified and referred for this study in a 1 year
period by local consultant rheumatologists with clinical features that they
considered highly suspicious of inflammatory back disease in whom infection had
been excluded. All patients were
clinically assessed and followed up by one observer (CJE) at baseline, 2 ,4 and
5 years. At baseline demographic
data, the presence or absence of psoriasis, inflammatory bowel disease, history
of uveitis, answers to Calin’s questionnaire, measurement of spinal movements
(18), chest expansion, HLA-B27 status, postero anterior radiographs of
sacroiliac joints, CT scan of sacroiliac joints were undertaken and recorded.
Each patient was also scored against the spondyloarthropathy criteria. At
the follow up visits all these except HLA-B27, CT scanning and the
spondyloarthroapathy criteria were repeated.
This
study had the full approval of the local ethical and radiation protection
committees who advised on the frequency and extent of radiological
investigations.
A
final diagnosis of ankylosing spondylitis was established on the basis of the
New York Criteria (3). If these
were achieved prior to 5 year follow up the patient was not subjected to further
radiology and the subject only followed up thereafter for clinical purposes.
For
the purpose of this paper the baseline and five year data or the time of
diagnosis of ankylosing spondylitis if sooner are presented and analysed.
Statistical comparison are made using Fisher’s exact test.
All
radiographs and CT scans were anonymised, randomised and then read jointly by
both authors at a single sitting where a grading was agreed by comparison with
the atlas of standard radiographs (19).
Results
At
baseline (Table 1) all 13
patients gave positive responses to 4 or more of Calin’s five questions. This suggests that these questions equate with experienced
clinical opinion which was the primary criterion for entry but are not
sufficiently specific on their own for diagnostic purposes.
No patients fulfilled the New York Criteria though 2 patients had the
minimum radiological criteria (bilateral grade 2 sacroiliitis) lacking
sufficient clinical criteria for diagnosis.
In contrast 6 patients had bilateral grade 2 or greater sacroiliitis on
CT scanning. Eight patients satisfied the ESSG (17) and none satisfied ³6
Amor criteria (16) for spondylarthropathy.
At
5 year follow up (Table 2) 4
patients satisfied the New York Criteria for ankylosing spondylitis. This was achieved by these patients developing a higher grade
of sacroiliitis (grade 3) together with limitation of spinal movement.
No patients had less than 2.5 cm chest expansion either at baseline or
follow up. Five patients had normal sacroiliac joints on plain radiographs and
CT scanning and did not develop ankylosing spondylitis during follow up. Two of
these 5 were HLA-B27 positive. Table
3 summarises the individual radiological and clinical findings of the
remaining 8 patients all of whom were HLA-B27
positive.
Analysis
of individual baseline clinical, imaging and HLA-B27 status against a final
diagnosis of ankylosing spondylitis only shows a statistically significant
result for plain film radiography grade 2 (unilateral or bilateral) as a
predictor for diagnosis at 5 years (Table
4 and Table 5).
Neither of the composite spondylarthropathy criteria scores were
predictors for a final diagnosis of ankylosing spondylitis in this patient
group. Combination of imaging and HLA–B27 status do not add any additional
diagnostic prediction (Table 6).
Calin’s questionnaire results have not been analysed further since all
patients gave positive responses at baseline.
Discussion
The
small numbers included in this study reflects the expressed intention to only
include patients in whom the referring consultant would have strongly considered
the diagnosis of inflammatory back disease and therefore have accordingly
offered both an appropriate diagnosis and treatment. Hence this reflects current
clinical practice and the need to undertake definitive confirmatory
investigations. This study has evaluated those routinely available at the time
and for which there was some validity from the published literature at the
commencement of the study.
Most
comparisons of diagnostic imaging methods in sacroiliitis have been undertaken
at a single time point (10, 13, 20-25). Whilst
this may establish that some methods may have a higher sensitivity it does not
establish that they also retain sufficient specificity to be satisfactory
predictors for the subsequent development of established disease.
Such specificity is important both from the aspect of basic scientific
study of disease pathogenesis, therapy and changes in prognosis from
interventions but also from the personal impact on individual patients.
Early accurate diagnosis will help to ensure patients receive appropriate
therapeutic interventions and avoid unnecessary and harmful ones.
However, early inaccurate diagnosis may result in individuals being
erroneously accorded a diagnosis of a chronic potentially disabling disease with
the potential for adverse psychological consequences, difficulties with health
and life insurance and with employment. It is therefore important to exert some
caution in documenting a diagnosis before classical criteria have been met.
The
present study demonstrates that plain film radiography accurately compared with
standard radiographs in conjunction with experienced clinical assessment is the
best current predictor of subsequent development of ankylosing spondylitis at 5
years diagnosed against established criteria.
It may be safe to use unilateral grade 2 sacroiliitis in clinical
practice as a marker for the diagnosis of ankylosing spondylitis. However, CT
scanning and HLA–B27 typing alone or in combination with each other or plain
radiographs do not add to early diagnostic certainty and should not be used to
define ankylosing spondylitis in routine clinical practice. Whilst criteria for
spondylarthropthy have clinical validity in defining a clinical syndrome they do
not predict the subset of patients who will develop ankylosing spondylitis. They
should therefore be recognised as having a clinical and research purpose in
defining the syndrome of spondylarthropathy and not any particular subset of
that syndrome.
The
high prevalence of HLA B27 in this group emphasises the caution that must be
applied to interpreting the presence of a combination of clinical inflammatory
back pain and HLA B27 with the inevitable development of ankylosing spondylitis.
This study is not able to determine if individuals who had not satisfied the
criteria for the diagnosis of ankylosing spondylitis at five years of follow up
would do so after longer follow up or whether they would develop other features
of spondylarthropathy in the future. However, this study does confirm that
caution needs to be applied to such patients when offering a definitive
diagnosis for which there may be personal adverse consequences of a diagnosis of
ankylosing spondylitis in the absence of classical criteria as defined by the
New York criteria. This should not mean that such individuals should not be
given appropriate advice and treatment but rather that the diagnosis should be
appropriately interpreted and qualified by using terms such as inflammatory back
pain or clinical sacroiliitis.
It
is unfortunate that MRI scanning could not be included in this longitudinal
study for technical reasons. The status of MRI scanning in the assessment of
sacroiliac disease is unresolved. It
is clear that disco vertebral disease and bone oedema are sensitively detected
by MRI but the position is less clear in relation to sacroiliac disease(26, 27,
28). It will be important for future evaluation of the role of MRI scanning in
early diagnosis of ankylosing spondylitis that similar longitudinal and
comparative studies to the present one are undertaken.
ACKNOWLEDGMENTS
We wish to acknowledge the assistance of our colleagues who allowed us to study patients under their care and also those patients for their cooperation with this five year study.
References
1.
Rigby AS, and Wood PHN,
Observations on
diagnostic criteria for ankylosing spondylitis
Clinical and experimental
Rheumatology 1993;11: 5-12.
2.
Gran JT, and Husby G.
The Epidemiology of
Ankylosing Spondylitis
Seminars in arthritis and
Rheumatism 1993; 22:319-334.
3.
Bennett PH and Wood PHN
In: Population Studies of
the Rheumatic Diseases. Proceeds of the third International Symposium
Excerpta Medica Foundation,
Amsterdam 1965:456-457.
4.
Cats A, Van der Linden SJ, Goei The HS, et al
Proposals for diagnostic
criteria of ankylosing spondylitis and allied disorders
Clinical and Experimental
Rheumatology 1987;5:167-171.
5.
Bollow M, Loreck D, Braun J, et al
Sacroliitis: the key
symptom of spondylarthropathies 2: morphological aspects (Review).
Rofo. Fortschritte auf dem
Gebiete der Rontgenstrahlen und der Nuen Bildgebenden Verfahren 1997;166:275-89.
6.
Mau W, Zeidler H, Mau R, et al
Clinical Features and
Prognosis of Patients with Possible Ankylosing Spondylitis. Results of a 10-Year
Followup.
Journal of Rheumatology
1988;15:1109-1114.
7.
Khan MA.
Delay in the radiographic
detection of sacroiliitis in ankylosing spondylitis
Arthritis and Rheumatism
1992;35:727.
8.
Dougdos M.
How to diagnose
spondyloarthropathy
Clinical and Experimental
Rheumatology 1993;11: 1-3.
9.
Collantes-Estevez E, Del Mazo AC and Munoz-Gomariz E.
Assessment of 2 Systems of
Spondyloathropathy Diagnostic and Classification
Criteria (Amor and ESSG) by
a Spanish Multicenter Study
Journal of Rheumatology
1995;22:246-251.
10.
Blum U, Buitrago-Tellez C, Mundinger A, et al
Magnetic resonance imaging
(MRI) for detection of active sacroiliitis – a prospective study comparing
conventional radiographjy, scintigraphy, and contrast enhanced MRI
Journal of Rheumatology
1996;23:2017-15.
11.
Peh WC, Ho WY and Luk KD
Applications of bone
scintigraphy in ankylosing spondylitis
Clinical Imaging
1997;21:54-62.
12.
Calin A, Porta J, Fries JF and Schurman DJ.
Clinical History as a
Screening Test for Ankylosing Spondylitis
JAMA 1977;237:2613-2614.
13.
Geijer M, Sihlbom H, Gothlin JH, et al
The role of CT in the
diagnosis of sacro-iliitis
Acta Radiologica
1998;39:265-8.
14.
Gran JT and Husby G
HLA-B27 and
spondyloarthropathy: value for early diagnosis ? (Review) (23 refs)
Journal of Medical Genetics
1995;32:479-501.
15.
Olajos A and Suranyi P
The value of HLA-B27 typing
in the diagnosis of early, oligosymptomatic spondylarthropathies (letter
comment).
British Journal of
Rheumatology 1996;35:192.
16.
Amor B, Dougados M and Mijiiyawa M.
Critère diagnostique des
spondylaarthropthies.
Revue de Rheumatisme et
Maladie Osteoarticularies 1990; 57: 85-89
17.
17
Dougados M, Vander Linden S, Juhlin R, et al
The Europian
spondylarthropathy study group preliminary criteria for the classification of
spondylarthropathy.
Arthritis and Rheumatism
1991; 34: 1218-1227.
18.
Moll JMH and Wright V
The pattern of chest
expansion and spinal mobility in ankylosing spondylitis. An objective clinical
study of 106 patients
Rheumatology and
Rehabilitation 1973;12:115-134.
19.
The epidemiology of chronic rheumatism vol 2. Atlas of standard
radiographs of arthritis
Blackwell, Oxford. 1963.
20.
Kozin F, Carrera GF, Ryan LM, et al
Computed tomography in the
diagnosis of sacroiliitis
Arthtritis and Rheumatism
1981; 24: 1479-85.
21.
Docherty P, Mitchell MJ, MacMillan L, et al
Magnetric resonance imaging
in the detection of sacroiliitis
The Journal of Rheumatology
1992; 19:393-401.
22.
Bollow M, Bierdermann T, Kannenberg J, et al
Use of dynamic magnetic
resonance imaging to detect sacroiliitis in HLA-B27 positive and negative
children with juvenile arthritides
Journal of Rheumatology
1998; 25:556-64.
23.
Jevtic V, Rozman B, Kos-Golja M, et al
MR imagiing in seronegative
spondyloarthritis
Radiologe 1996;36 : 624-31.
24.
Remy M, Bouillet P, Bertin P, et al
Evaluation of magnetic
resonance imaging for the detection of sacroiliitis in patients with early
seronegativek spondylarthropathy
Revue Du Rhumatisme,
English Edition 1996;63:577-93.
25.
Bollow M, Braun J, Hamm B, et al
Early sacroiliitis in
patients with spondyloarthropathy: evaluation with dynamic gadolinium-enhanced
MR imaging
Radiology 1995;194:529-36.
26.
Ahlstrom H, Felteluis N, Nyman R
Magnetic Resonance Imaging
of Sacroiliac joint inflammation
Arthritis and Rheumatism
1990: 33: 1763
27.
Murphy MD, Welzel LH, Bramble JM
Sacroilitis MR Imaging and
Findings
Radiology 1991, 180: 239
28.
Braun J, Bollow M,
Imaging of inflammatory
rheumatic diseases of the skeleton (sacroiliitis). When and how should we use
MRI?
Journal of Clinical Rheumatology 2000; 6: 339-349.