Forty seven studies were included in the review. Thirteen studies addressed cartilage pathologies; 10 studies addressed tendon and ligament pathologies; nine studies addressed cortical pathologies; and 26 studies addressed synovial pathologies. The number of participants in included studies ranged from two to 600; most studies had fewer than 50 patients. Definition of osteoarthritis varied and was unspecified in approximately half of the studies; where a definition was reported, American College of Rheumatology criteria were often used. Most studies described ultrasound technique and joint position during image acquisition; these varied between studies of the same joint area. A wide variety of pathologies were examined and definitions of the imaging appearance of the pathology imaged were only provided in approximately half of the studies, with no standard definition across studies.
Twenty seven studies assessed construct validity of ultrasonography (compared ultrasound with other examination techniques) and nine studies addressed criterion validity (compared ultrasound with the reference standard of either direct macroscopic or microscopic visualisation of the pathology (for example, by arthroscopy), examination during surgery or histopathological examination).
Cartilage: Comparisons were limited to cartilage thickness. Two studies found reasonable correlation between ultrasonography-detected cartilage thickness and histological cartilage thickness. One study demonstrated reasonable correlation between ultrasonography-detected cartilage thickness and magnetic resonance imaging.
Tendon and ligament: Ultrasound changes were usually compared with clinical examination. Results varied. There was poor correlation between ultrasound and clinical diagnosis of anserine tenobursitis (one study), but good correlation between ultrasound and clinical and radiographic changes of enthesitis at the shoulder and foot (two studies).
Cortical: Ultrasound was found to be more sensitive for detection of osteophytosis in the small joints of the hand than radiography (one study), but less sensitive to erosions (one study).
Synovial: Ultrasound was comparable to magnetic resonance imaging for detection of effusion, synovial hypertrophy and popliteal cysts. Ultrasound was more sensitive and specific than clinical examination in detecting effusion and synovial hypertrophy at the knee joint.
Overall, no consistent relationship between clinical symptoms and ultrasound-detected pathology was identified, although symptomatic joints tended to have more abnormalities on ultrasound than controls/healthy joints.
Eight studies assessed the ability of ultrasound to detect changes in joint pathology over time. The trend was towards a reduction in pathology with time after therapy.