Metacarpophalangeal joints at US in asymptomatic volunteers and cadaveric specimens

PURPOSE: To determine normal anatomy of metacarpophalangeal (MCP) joints at ultrasonography (US) and to compare findings with anatomic and histopathologic findings. MATERIALS AND METHODS: Right hands of five cadavers (two women, three men; age range at death, 46-96 years; mean age at death, 62 years) and dominant hands of 30 volunteers (15 men, 15 women; age range, 27-74 years; mean age, 43 years) were evaluated. Nonthumb MCP joints were scanned longitudinally and transversely by two musculoskeletal radiolologists working independently at US with a high-frequency linear-array transducer (frequency, 12 MHz). US appearances of intraarticular and periarticular structures were analyzed independently by these radiologists. Specimen sectioning provided anatomic and histopathologic comparisons. Interobserver precision was assessed with determination of coefficient of variation (CV). RESULTS: Sagittal US images enabled good visualization of the dorsal metacarpal synovial recess and the metacarpal head cartilage. In 37% of cases, a small depression (mean depth, 0.3 mm) was identified on the dorsal aspect of the metacarpal head, especially at the second MCP joint (19% of cases). Direct visualization of sagittal bands of the extensor hood was possible on dorsal transverse US scans, especially at the second and third MCP joints. In all cases, the first annular pulley was well seen on palmar transverse US scans. The radial collateral ligament of the second MCP joint was better recognized by the two observers than was the ulnar collateral ligament of the fifth MCP joint. Interobserver precision was high (CV range, 1.5%-6.5%) for intraarticular and periarticular structures. CONCLUSION: US provides detailed information about normal anatomy of nonthumb MCP joints. (C) RSNA, 2004.