The tear of a rotator cuff is a common manifestation of RCT. However, to date, the characteristics and causes of tears remain unclear. Harryman et al.  reported that the deep structure of a rotator cuff not only comprises of muscle and tendon tissues, but also the tissue of the joint capsule. Zbigniew et al.  also reported similar results. Therefore, from an anatomical point of view, the two layers of the tear rotator cuff are not the same. The upper layer comprises of muscle and tendon tissues, while the lower layer comprises of the woven structure of the muscle tendon and joint capsule tissue.
Therefore, when the rotator cuff is teared, the upper and lower layers are not sufficiently combined to cause delamination. Nimura et al.  further investigated the attachment of the joint capsule. The attachment of the articular capsule at different positions on the greater tuberosity of the humerus also differs. The joint capsule after the supraspinatus muscle stop point is weak at the greater tuberosity, and can easily tear. This may be one of the reasons for the higher incidence of delamination, when accompanied by an infraspinatus tear. Sang et al.  also found that it is easier to retract backward and inward after tearing of the lower layer. The reason may be that the muscle fibers that constitute the lower layer mainly come from the subscapularis muscle. In the present study, patients with larger tears had a significantly increased probability of delamination, and this is similar to the reported results.
Other factors correlated to rotator cuff delamination have also been reported, but these results were not consistent. MacDougal et al.  reported that delamination was correlated to age and gender, while Matsuki et al. [12, 13] did not reach the same conclusion. Satoshi  reported that there was no correlation between delamination and the history and duration of trauma, which is similar to the results of the present study. In addition, in the present study, the correlation between subacromial impact and spallation was analyzed. The results revealed that there was a certain correlation between impact and delamination, especially for patients with partial tears on the bursa side (type IIb), and the probability significantly increased. The reason may be that under repeated impacts of the acromion, the upper rotator cuff stops were broken, while the lower rotator cuff was not involved. The tissue degeneration and relaxation between these two layers causes the upper layer to retract.
It has been reported [1, 2] that the therapeutic effect of rotator cuff injury with spallation is not very good. However, this report did not describe the treatment in detail. Zilber et al.  reported that the treatment effect was not significantly different from that of patients without stratification. However, a higher proportion of patients with subscapular muscle steatosis and decreased external rotation function could be observed in that report. Indeed, this result may be correlated to the treatment method for subscapular muscle injury. Sugaya et al.  proposed that the torn layers should be repaired, in order to restore the damaged rotator cuff to a complete structure. Furthermore, the double-layer double-row (DLDR) or double-layer suture bridge (DLSB) technique should be used for completely torn cuffs, in order to obtain very good results. Sonnabend  noted that when suturing the upper and lower layers, attention should be given to the freshness between these two layers, and that the synovial and other tissues between these two layers should be cleaned to facilitate the healing between layers
In the present group of patients, DLSB technology was used for patients with full-thickness full-layer tear and spallation. These two layers were cleaned and freshened. Furthermore, the inner row suture was knotted, while the outside row suture was pressed by an external anchor to strengthen the contact between these two layers. Good results were obtained. For patients with upper tears and retraction on the bursal side, the suture bridge technique or modified Mason-Allen technique was adopted according to the size of the tear, and the postoperative results were satisfactory.