ABSTRACT
STUDY DESIGN: Single-cohort laboratory-based study. OBJECTIVES: To measure thoracic spine extension motion during bilateral arm elevation using functional radiography and photographic image analysis. BACKGROUND: Impairment of thoracic spine extension motion may impact shoulder girdle function. Motion of the thoracic spine during arm movement has not been directly measured using functional radiographic analysis. METHODS: In 21 asymptomatic men, thoracic kyphosis was measured in neutral standing and in end-range bilateral arm elevation, using lateral radiographs and photographic image analysis. Using both measurement techniques, the difference in thoracic kyphosis between the 2 body positions was used to quantify the range of extension motion of the thoracic spine. Bland-Altman plots were used to examine the agreement between measurement techniques. The relationship between the amount of thoracic kyphosis in neutral standing and kyphosis in full bilateral arm elevation was also examined. RESULTS: The mean ± SD increase in thoracic extension with bilateral arm elevation was 12.8° ± 7.6° and 10.5° ± 4.4°, when measured from the radiographs and photographs, respectively. There was a significant correlation between the radiographic and photographic measurements of the amount of thoracic kyphosis measured in neutral posture (r = 0.71, P<.01) and for the kyphosis measured in full bilateral arm elevation (r = 0.79, P<.001). The mean difference between the 2 measurement techniques was 2.1° for kyphosis measured in neutral posture and 0.5° when measured in full bilateral arm elevation. The thoracic kyphosis angle measured in neutral posture was strongly correlated with the thoracic kyphosis angle measured in full bilateral arm elevation when measured with both radiographic (r = 0.80, P<.001) and photographic (r = 0.84, P<.001) techniques. CONCLUSION: In asymptomatic men, bilateral arm elevation is associated with movement of the thoracic spine toward extension, but the amount of movement is variable among individuals.