Intra- and inter-observer reliability of ultrasound muscle thickness of gluteal and biceps femoris long head in individuals with and without SCI

PURPOSE: This study aimed to evaluate both inter- and intra-observer reliability of ultrasound-based muscle thickness measurements in able-bodied (AB) individuals, as well as intra-observer reliability in individuals with spinal cord injury (SCI).

METHODS: Ultrasound measurements of the gluteus maximus, medius, minimus and biceps femoris long head were performed on 31 AB participants and 30 participants with SCI. Each AB participant was scanned on two occasions by three observers, with three repetitions per muscle per occasion. The muscle thickness in participants with SCI was measured using three repetitions during a single test occasion, conducted by one observer. A generalizability (G) study was conducted to assess the reliability of the measurements.

RESULTS: In AB participants, intra-observer reliability for gluteal muscles ranged from G-coefficient: 0.57 to 0.89, and for biceps femoris long head from G-coefficient: 0.60 to 0.76. Inter-observer reliability in AB participants was G-coefficient:0.48-0.72 for the gluteal muscles and G-coefficient: 0.52 for the biceps femoris. In contrast, intra-observer reliability in participants with SCI was excellent across all muscles (G-coefficient: 0.95-0.99).

CONCLUSION: Ultrasound can assess muscle thickness with moderate to good intra-observer reliability in AB participants, but with only poor to moderate inter-observer reliability. In contrast, intra-observer reliability was excellent in participants with SCI. Reliability depends on observer experience and varies across muscles and populations.