ULTRASONIC EVALUATION OF ANTERIOR SHOULDER TRANSLATION IN NORMAL SHOULDERS

Anterior translation in the right shoulders of 23 volunteers was evaluated using ultrasound from an anterior approach with a 10 MHz, 6 cm wide linear transducer. A translatory force of 90 N was used to translate the humeral head in the adduction and internal rotation position, while a translatory force of 60 N was used in the more clinically relevant position of 90° abduction and external rotation position. The overall intra-observer coefficients of variation ranged from 0–13.0% (mean 3.8 ± 2.5%) for examiner I and 0.5–20.9% (mean 5.1 ± 3.9%) for examiner II. The overall inter-observer variation ranged from 0–29.8% (mean 9.3 ± 7.3%). The anterior translation of the humeral head in adduction and internal rotation following 90 N displacement force ranged from -2.6 to 12.9 mm (mean 2.1 ± 3.1 mm) for examiner I and from -4.1 to 4.7 mm (mean 1.1 ± 2.2 mm) for examiner II. The anterior translation of the humeral head in abduction and external rotation following 60 N displacement force ranged from -3.3 to 3.7 mm (mean 0.3 ± 1.9 mm) for examiner I and from -8.3 mm to 4.5 mm (mean -0.7 ± 2.6 mm) for examiner II. The intra-class correlation coefficients (r) for the measured anterior translation between the two examiners for the 2 positions were 0.029 and -0.058 respectively. We concluded that the inter-observer coefficient of variation remained excessive and the agreement in the measured anterior translation between the two examiners was poor. The finding of negative values in the measured anterior translation despite the use of 90 N and 60 N translatory force raises further concerns about the prospective clinical use of this technique at the present moment.