RA is a progressive disease that, if left unchecked, results in irreversible joint damage. Early diagnosis is often difficult due to the non-specific symptoms and signs that develop relatively slowly. In particular, the conventional diagnostic criteria developed by the American College of Rheumatology (ACR) among patients, the slow development of signs and symptoms, and ambiguous symptoms similar to other inflammatory polyarthropathies. Furthermore, conventional diagnostic criteria developed by the ACR were derived from patients with established disease [4
]. These criteria were not designed to detect early clinical symptoms of the disease process. In fact, recent data suggest that, in the first 3 months during disease onset, ACR criteria may well be misleading rather than helpful in making a definitive diagnosis of RA [5
]. The natural history of RA is such that the early months of disease are a critical period during which irreversible joint damage occurs. The advantages of an accurate, early clinical RA diagnosis thus emphasize the need for more accurate methods of detection [6
]. It should be noted, however, that currently the majority of patients do not present with a disease duration of less than 3 months, therefore ACR criteria would be valid. However, as indicated later, the majority of joint erosions are not detected by radiographic analysis in patients at presentation.
The current ACR recommendations for diagnosing RA include radiographic analysis of selected joints. The sensitivity of radiography, however, is very limited. For instance, at presentation, the characteristic or pathognomonic radiographic-identified joint erosion was detected in only 15% of patients with RA [7
]. In contrast, when imaged by MRI, abnormalities were detected in 70% of patients, involving 35% of the patients' joints. In another study, MRI was performed on 42 patients with early RA (median disease duration, 4 months) at baseline and at 1 year. Synovitis, measured by MRI, was predictive for the development of joint erosions at 1 year [8
]. The use of imaging has allowed the distinction to be made between entheseal-based disease and primary intrasynovial disease [9
]. For example, MRI has been shown to be useful in differentiating polymyalgia rheumatica from RA by imaging extracapsular enhancement in the latter [10
]. MRI has been used to determine whether recent-onset knee synovitis differed in RA patients compared with those who had spondyloarthropathy. Data showed that the non-RA patients had a periarticular location of entheseal inflammation and bone edema, while RA patients did not [12
]. In another study, RA patients classified with a good prognosis (defined as acute onset [~24 hours] of disease) were shown to have the same characteristic entheseal pathology as opposed to the intrasynovial inflammatory pathology of RA and, importantly, MRI detection of enthesitis correlated closely with favorable clinical outcomes (not requiring disease-modifying therapy) in these patients [4
]. The characteristic feature of the good-prognosis patients (all of whom fulfilled the ACR diagnostic criteria for RA) was that they lacked the intra-articular 'bare area' region lesions, whereas such a change was a universal feature of poor prognosis in other RA patients studied. This finding suggests that many RA patients with a good prognosis exhibit a unique pathology, an important clinical consideration. Unfortunately, these differences between diseases have been difficult to detect during routine clinical examination, providing further rationale for the use of MRI as a diagnostic tool in this subset of patients.