Enthesial fibrocartilage - bone interaction: a radiographic study of selected sites of nonsynovial peripheral enthesopathy.

OBJECTIVES: To assess radiological imaging and prevalence of pelvic (excluding sacroiliac joints), calcaneal, patellar and humeral enthesopathy (EN) in a cohort study.

MATERIALS-METHODS: Outpatients attending a state hospital rheumatology clinic for a continuous 4-year period, were consecutively screened for radiographic evidence of peripheral nonsynovial EN in pelvis, ankle, knee and shoulder regions and in particular sites within these regions regardless of symptoms. Imaging and prevalence were assessed in patients suffering from a variety of myoskeletal disorders by applying either of the following two plain X-ray criteria: a) tendon and/or ligament and/or fascia ossification, b) reactive bone proliferation resulting in excrescences and/or sclerosis and/or erosions.

RESULTS: A total of 3,670 patients were screened and a cohort of 585 patients (16%) with extraspinal peripheral EN was selected. Plain radiography provided good imaging of pelvic EN at iliac crests, greater trochanters, pubic symphysis and pubic rami, as well as of calcaneal, patellar and humeral head enthesopathic changes. Cohort recruitment by applying the two aforementioned criteria resulted in the formation of 2 groups: Group A, consisting of 169 patients (mean age in years 34 -/+ 8 SD) suffering from inflammatory myoskeletal disease represented by Seronegative Spondyloarthropathies (SSp); and Group B, including 416 patients (mean age 63 -/+ 7) suffering from degenerative/metabolic disorders classified as degenerative disease of the spine, hip or knee (70%), Diffuse Idiopathic Skeletal Hyperostosis (DISH) (11%) and rotator cuff (Rot/Cuff) syndromes (19%). Females were the predominant gender in the cohort and in Group B patients (both p<0.001 vs. males), while the opposite was true for the group of inflammatory diseases. Patients in Group A were younger and had shorter disease duration than those of group B (p<0.001 for both). Pelvic EN was the most frequent localization of EN within the cohort (46%, p<0.001) followed by both multiple site and patellar EN (24% and 22% respectively). Patients in Group A, had a significantly higher prevalence of pelvic EN compared to those in Group B (60% vs. 39%, p<0.001) and the former group was significantly associated with pelvic EN. On the contrary, although pelvis was also the predominant EN site in Group B, patellar and humeral head EN were significantly associated with noninflammatory diseases. In patients with SSps, pelvic EN predominance (60%) was followed by calcaneal involvement (p<0.01 vs. patellar and humeral head). These two were the skeletal sites that were significantly associated with individual diseases within Group A (pelvis with AS and Ps-Sp and calcaneus with RR). Within Group B, patients with knee OA, hip OA and Rot/Cuff showed EN site localization in absolute proximity with disease process, while in those with Deg/Sp or DISH pelvis was the predominant site involved.

CONCLUSIONS: Plain radiography provides good imaging of peripheral nonsynovial EN at well defined skeletal sites. Within a general rheumatic population, pelvic EN is the most prevalent localization followed by multiple site and patellar reactive bone lesions. Apart from seronegative spondyloarthopathies, degenerative and metabolic myoskeletal disorders contribute substantially to local induction of abnormal fibrous tissue/fibrocartilage-bone interactions resulting in radiographically detectable EN.