Sensitivity and Specificity of the CASPAR Criteria for Psoriatic Arthritis in a Family Medicine Clinic Setting
To the Editor:
Psoriatic arthritis (PsA) is an inflammatory arthritis associated with psoriasis with an estimated prevalence in the United States of 0.25%1. The prevalence of PsA in patients with psoriasis is 30%2. Since the prevalence of psoriasis in North America is 2%, the estimated population prevalence of PsA of 0.25% is probably an underestimate3. PsA is thus underdiagnosed. This may be due to the lack of recognition of the condition in family practice and dermatology practice.
The original diagnostic criteria of Moll and Wright are the simplest and the most frequently used4. In order to make the diagnostic criteria more specific a number of classification criteria were developed, but none were widely agreed upon or validated5,6. The ClASsification of Psoriatic ARthritis (CASPAR) study group published new criteria for the classification of PsA derived from patients attending rheumatology clinics5. The sensitivity and specificity of the CASPAR criteria in the original study were 91.4% and 98.7%, respectively5. Subsequently, we have shown that the criteria have a sensitivity of 99.1% in early PsA7. The performance of the CASPAR criteria outside specialty rheumatology clinics has not been investigated. The authors of the original study felt that it was not possible to apply the results of the study to the general population or to other clinic populations (e.g., dermatology clinic populations) as the study was conducted in patients with known inflammatory articular disease5. The authors recommended that the criteria be tested in the general population5. We therefore conducted the present study to evaluate the criteria in patients attending a Family Medicine clinic.
The CASPAR criteria were applied to consecutive consenting patients attending a Family Medicine clinic attached to the Toronto Western Hospital. All patients were assessed by a rheumatologist according to a standard protocol, based on which a diagnosis of PsA was made. The protocol was based on the University of Toronto PsA clinic protocol, and included questions about inflammatory joint symptoms and a complete physical examination including detailed musculoskeletal assessment8. All patients were tested for antinuclear antibodies and for rheumatoid factor. Other laboratory tests and radiographs were done when clinically indicated. The diagnosis was subsequently confirmed by a rheumatologist acknowledged to be an expert on PsA. The CASPAR criteria were applied to all patients participating in the study and in a subgroup with inflammatory musculoskeletal symptoms, and the sensitivity and specificity determined5.
A total of 175 subjects were evaluated. Their demographic and disease characteristics are given in Table 1. Thirty-seven (21%) had inflammatory musculoskeletal symptoms and/or signs at the time of evaluation. The disease characteristics of these 37 patients are also given in Table 1. Two subjects had PsA. Twenty-eight subjects had various musculoskeletal symptoms and/or signs, but a specific diagnosis was not made.
The CASPAR criteria for classification of PsA were applied to the entire group of 175 subjects. Both subjects with PsA satisfied the CASPAR criteria. Thus the sensitivity was 100%. Of the 173 subjects who did not have PsA, 171 did not satisfy the criteria. The specificity thus was 98.8%. The criteria were applied to the subgroup of subjects with inflammatory musculoskeletal symptoms/signs. Both patients with PsA satisfied the criteria (sensitivity 100%). Of the 35 subjects without PsA, 33 did not satisfy the criteria (specificity 94.3%). Two patients without PsA satisfied the CASPAR criteria. One had psoriasis and ankylosing spondylitis and the other had psoriasis and inflammatory back pain.
Thus, our study shows that the CASPAR criteria have a sensitivity of 100% and specificity of 98.8% (94.3% if restricted to patients with inflammatory musculoskeletal symptoms/sign) in a family practice setting. After further validation in primary care, early arthritis clinics, and general rheumatology clinics, the criteria may be used as both diagnostic and classification criteria. This will greatly help research on the epidemiology and genetics of PsA, and will also facilitate early diagnosis of this condition in clinical practice. The prevalence of inflammatory articular symptoms/signs in the study subjects was high (21%). The estimated prevalence of self-reported arthritis/rheumatism in Canada is 17.63%9. Our figures are higher, even though we have included patients with inflammatory musculoskeletal symptoms only. The higher prevalence could be explained by the fact that the clinic is attached to a major university teaching hospital dealing primarily with rheumatology and neurological sciences.
The results of our study show that CASPAR criteria are highly sensitive and specific when applied to subjects attending a family medicine clinic. The criteria thus have the potential to be used as diagnostic criteria for PsA and as a tool in epidemiological studies on the prevalence of PsA.
VINOD CHANDRAN, MBBS, MD, DM; CATHERINE T. SCHENTAG, MSc; DAFNA D. GLADMAN, MD, FRCPC, University of Toronto Psoriatic Arthritis Clinic, Centre for Prognosis Studies in the Rheumatic Diseases, Toronto Western Hospital, Toronto, Ontario, Canada. Address reprint requests to Dr. Gladman; E-mail:email@example.com
Vinod Chandran is supported by a Krembil Psoriatic Arthritis Fellowship. The University of Toronto Psoriatic Arthritis Program is partially supported by the Krembil Foundation.
3. Langley RG, Krueger GG, Griffiths CE. Psoriasis: epidemiology, clinical features, and quality of life. Ann Rheum Dis 2005;64 Suppl 2:ii18-23; discussion ii24-5.
4. Moll JM, Wright V. Psoriatic arthritis. Semin Arthritis Rheum 1973;3:55-78.
5. Taylor W, Gladman D, Helliwell P, Marchesoni A, Mease P, Mielants H; CASPAR Study Group. Classification criteria for psoriatic arthritis: development of new criteria from a large international study. Arthritis Rheum 2006;54:2665-73. [MEDLINE]