Sonography of supraspinatus tendon abnormalities in the neutral versus Crass and modified Crass positions: a prospective study.

OBJECTIVES

The purpose of this study was to investigate how well the neutral arm position correlates with the Crass and modified Crass positions in sonographic evaluation of supraspinatus tendon abnormalities.

METHODS

A prospective study was performed on 32 shoulders in 31 patients with shoulder pain. Images were acquired in the long axis of the supraspinatus tendon in the neutral, Crass, and modified Crass positions and were reviewed independently of each other by 2 radiologists in consensus. The images were evaluated for visualized tendon length, tendinosis, partial- or full-thickness tears, calcific tendinosis, subdeltoid bursitis, and enthesopathic changes. Kappa values of agreement between neutral and Crass and neutral and modified Crass positions were determined.

RESULTS

The mean visualized tendon lengths were 1.78, 2.83, and 2.77 cm in the neutral, Crass, and modified Crass positions, respectively. Kappa agreements between neutral and Crass and neutral and modified Crass positions were as follows: perfect agreement for calcific tendinosis (neutral versus Crass, κ =1.00; neutral versus modified Crass, κ = 1.00), almost perfect agreement for full-thickness tears (neutral versus Crass, κ = 0.85; neutral versus modified Crass: κ= 0.92) and tendinosis (neutral versus Crass, κ = 0.93; neutral versus modified Crass, κ = 0.86), substantial agreement for enthesopathic changes (neutral versus Crass, κ= 0.80; neutral versus modified Crass, κ= 0.61) and subdeltoid bursitis (neutral versus Crass, κ = 0.63; neutral versus modified Crass, κ = 0.75), and moderate agreement for partial tears (neutral versus Crass, κ = 0.44; neutral versus modified Crass, κ = 0.47).

CONCLUSIONS

For patients who cannot tolerate the Crass or modified Crass position, the neutral position can show full-thickness tears, tendinosis, calcific tendinosis, enthesopathic changes, and bursitis.