Supplement Article |
OP7. DEVELOPING DIAGNOSTIC CRITERIA IN PMR
Rheumatology, Southend Hospital, Westcliff, Essex, UK
Background: Polymyalgia rheumatica (PMR) represents the commonest indication for long term steroid therapy in the community. It is also subject to the widest variations of clinical practice since it may be managed in primary or secondary care by general practitioners, rheumatologists and non-rheumatologists. Many features of PMR, such as systemic symptoms, distal manifestations such as peripheral arthritis, and acute phase response, predispose the unwary clinician to diagnostic error. A response to steroids may encourage diagnostic error since steroids are potent anti-inflammatory agents that can mask symptoms from a host of serious and non-serious conditions.
Classification criteria for PMR are needed for major reasons: (1) to be able to order this clinical syndrome as a distinct disease entity separate from conditions such as seronegative and seropositive inflammatory arthritis and manifold other conditions sharing similar clinical characteristics (2) to be able to compare like groups of patients across populations of patients seen in different clinics and settings in different countries, and (3) to facilitate prediction of disease and treatment outcomes.
Methods: In our view, development of satisfactory classification criteria for PMR will require two stages. Stage 1 is development of consensus criteria and Stage 2 is a prospective validation study of patients presenting with proximal pain and stiffness.
An expected outcome of this process is the establishment of an infrastructure for ongoing international collaboration in the study of PMR.
Stage 1: Development of classification criteria in PMR: Currently there are no validated ACR classification criteria for PMR. Our first aim therefore is to develop ‘draft classification criteria’ based on a comprehensive review of literature and a consensus based judgment of an international collaborative group of experts. This exercise will develop these criteria, suggest an appropriate diagnostic process in polymyalgia, agree a typical steroid response and prioritize areas for future research in PMR.
Stage 2: Validation of the criteria in a prospective study of diagnostic outcomes in consecutive patients presenting with proximal pain and stiffness.
A retrospective case-control study will suffer from lack of adequate data on cases and controls at initial presentation. A reliance on the clinician's diagnosis in the absence of a ‘gold standard’ suffers from bias due to circularity. What is required is a prospective diagnostic evaluation of patients with polymyalgic ‘core symptoms’, i.e. proximal pain and stiffness and treated with uniform steroid schedule. The group will organize a multi-centre prospective (likely of 12 month duration) cohort study of patients presenting with the set of factors that must be present for the diagnosis of PMR. Stage 1 would have decided the inclusion/ exclusion features that will be tested, diagnostic algorithm that will be followed and initial dose and steroid taper schedule. Patients from rheumatology practices and primary care providers will be included.