For SSM atrophy US SE had a positive correlation (r = 0.829,
p =<0.001),
without SE (r = 0.679,
p =<0.05) with reference to MRI.
For FD grades US SE also had a positive correlation (r = 0.873,
p =<0.001),
US alone (r = 0.713,
p =<0.05) with reference to MRI.
Conventional grey scale US could not clearly differentiate muscle from the fat in 6 cases of 9 partial tears,
because the echogenicity was equal.
US SE improved the diagnosis of fatty degeneration and atrophy in comparance to US alone.
SE helps to delineate the SSM from the fatty tissue.
Fatty tissue coloured red,
yellow and green (low stiffness areas) in comparance to the muscle (high stiffness object),
which coloured blue Fig. 10
SE was more sensitive in RC full tears than in partial RC tears. We had 100% sensitivity for prediction of SS MA and FD in full thickness tears and only 66% of sensitivity in partial RCTs.
SE provided more accurate measurements of occupational ratio of SSM atrophy by highlighting area of FD Fig. 11
We also had limitations of the study:
SE is a subjective modality and strongly depends on the quality factor and expirience of the examiner.
We had ‘‘almost perfect interobserver agreement’’ of SE data with a kappa value of 0.83 (for all pathology).
This agreement turned to ‘‘substantial’’,
0.66 only for partial tears.
Limited FOV during SE study could not sometimes carefully occupy the whole SSM with fatty tissue for comparance.
There was difficulty in estimation caused by the posterior acoustic shadowing of the superior border and the spine of the scapula.
We had an artifact of the region close to the scapula,
which decreased the measurement of the real size of the muscle Fig. 12
Fusion US/MRI helped to overcome this limitations of US.
A limited field of view of US was enlarged by adding Fusion with MRI,
that allows an evaluation of the whole cross-sectional area of the region Fig. 13 Therefore,
the combination of evaluating the same structure with two imaging modalities simultaneously improves the understanding of anatomy and at the same time pathologic conditions.
We had a group of consecutive patients with different RC pathologies.
We had a small number of patients with RCTs.
The intraobserver reliability of SE was not investigated in this study.
SE was not sensitive in patients with increased body mass index that were excluded from the study.