Background The diagnosis of polymyalgia rheumatica (PMR) is often challenging as we still do not have a specific biomarker for PMR. Most of the patients are treated by the primary care physicians and never see a rheumatologist early in the disease course; and, the most of the referrals to the rheumatologist are for diagnostic purposes.
Objectives Primary objective: To elaborate the primary care referrals of patients with suspected PMR and their final diagnosis by a rheumatologist. Secondary objective: To assess the usefulness of the 2012 EULAR/ACR Classification Criteria for PMR and the PMR score (1).
Methods This is a prospective observational study of a cohort patients referred by the primary care physicians to our PMR Fast Track Clinic. All referred patients had a standardised set of screening blood tests for PMR and when clinically indicated, they underwent further investigations. We used the 2012 EULAR/ACR Classification Criteria for PMR and the PMR core was calculated in all patients at their first appointment. The sensitivity and specificity of the PMR scores were calculated in relation of the final diagnosis. The probability of GPs to make the right diagnosis of PMR was estimated using the one-sample binomial test.
Results Fifty patients were referred from the Primary Care with suspected PMR. In the PMR group, there were 29 patients (58%) who fulfilled the essential 2012 EULAR/ACR Classification Criteria for PMR and had confirmed diagnosis of PMR. Their PMR scores were in the range of 6/6 (23 patients, 79%), 5/6 (4 patients, 14%) and 4/6 (2 patients, 7%). In the non-PMR group, there were 21 patients (42%) in which PMR was excluded following the initial investigations. Their diagnosis varied grossly with 18 different diagnoses, including two solid and one haematological malignancy. Thirteen (62%) of these patients did not meet the required criteria for PMR, and therefore the PMR score was not applicable; 4 patients (19%) had PMR score of 4/6 and above, and 4 patients had score of 3/6. The probability of the primary care physicians to make the right diagnosis was estimated at 0.322 with p value of 0.16- which is non -significant. This support the null i.e. primary care physicians are referring by chance – what could be described as the same as “flipping a coin”. The sensitivity and specificity of PMR score was 100% for 4/6 or more, and specificity of 81%; for the score of 5/6 or more the sensitivity is 87% and the specificity 86%. Therefore, at 4 points or above there is a higher sensitivity but with 5 or more there is a slight better specificity but at the cost of sensitivity.
Conclusions The diagnosis of PMR remains challenging, not only for the general practitioners but also in the setting of the secondary and tertiary care. The 2012 EULAR/ACR Classification Criteria for PMR and the PMR score are useful tool when making the diagnosis. However, the most important is careful history and having a low threshold for further investigation when there is any doubt or atypical presentation of PMR
A European League Against Rheumatism/American College of Rheumatology Collaborative Initiative: 2012 Provisional Classification Criteria for Polymyalgia Rheumatica: Arthritis & Rheumatism, Vol 64, No. 4 April, 2012, pp. 943-954.
Acknowledgements Southend University Hospital
Disclosure of Interest None declared