Postoperative MRI scan is usually a nontraumatic evaluation method to assess the integrity of the tendon structure after rotator cuff repair [16, 17]. MRI assessment has proven to be a good standard for assessing the rotator cuff tendon integrity after repair . For full-thickness rotator cuff tears, MRI has higher sensitivity and superiority than other methods . The most significant finding of our study was that there were 5 type IV cases in SR group and 3 type IV cases in SB group. There was no significant difference of the postoperative re-tear rate between the SR group and the SB group.
In the past two decades, there have been various repair techniques for rotator cuff repair, such as single-row technology and double-row technology. So far, the suture bridge technology has proved to be equivalent or even better than the traditional double-row technology, because the suture bridge technology can achieve better load sharing between the inserted anchors [20–22]. In the present study, the postoperative re-tear rate of the SR group was 20%, and the re-tear rate of the SB group was 12%, and there was no difference in the post-operative re-tear rate between the SR group and the SB group. In contrast, Kim et al.  reported that the single row repair technique might have better tendon integrity than the double row suture bridge repair when remnant tendon was less than 10 mm. They stated that the double row repair might make the medial row to be placed more medially from the musculotendinous junction if the remnant tendon length was less than 10 mm, leading to too much tension on the weak medial muscular portion and subsequently tendon retear in the medial row.
In the present study, the retear sign (Sugaya type IV) was observed in both the SR group and SB group initially after surgery, which is mainly due to poor tissue quality of rotator cuff as well as configuration deformity after repair. Previously, Cho et al. analyzed the re-tear mode of single-row repair technique and double-row suture bridge repair technique using MRI postoperatively. They divided the re-tear mode into type I (repair of cuff tissue at the cuff insertion site was not observed on the greater tuberosity) and type II (remaining cuff tissue remaining at the insertion site despite re-tear). They observed that there were 14 cases of type I (73.7%) and 5 cases of type II (26.3%) in the single-row group, 7 cases of type I (25.9%) and 20 cases of type II (74.1%) in the double-row suture bridge group. The re-tear pattern was quite different between the single-row group and the double-row suture bridge group, which might be due to different repair configuration by different techniques.
Furthermore, tendon quality is considered to be a good prognostic factor for repair integrity . In this study, the SB group found 1 case with normal tendon coverage, and the other 2 cases showed poor rotator cuff tissue quality. Among the 5 type IV cases in the SR group, 3 cases were found to have normal tendon coverage, and the other 2 cases were found to have poor quality of the rotator cuff after arthroscopy. After surgical repair, the normal rotator cuff tendon on MRI may reveal regular or irregular, thickening or thinning, with moderate to low signal intensity . For rotator cuffs with poor tissue quality after repair, postoperative images need to be carefully analyzed, because there was some high signal intensity in the rotator cuff tendon due to inflammation, tendinopathy or granulation tissue . In cases with normal tendons, the re-tear sign on the rotator cuff tendon on MRI may be due to configuration deformities ((such as dog-ear deformity). Specifically, most deformities after rotator cuff repair were remodeled and rotator cuff tendon coverage may improve with time [14, 28]. Previously, Jin et al.  found that three patients who were classified as Sugaya type III at three months after surgery appeared as Sygaya type II at one year postoperatively. Such postoperative MRI findings should be interpreted with caution.
The current research has some limitations. First, the sample size of patients included and analyzed was small. The number of patients recruited was not large, which may introduce bias and inaccuracy. In addition, all images were evaluated by a clinician, although patient characteristics were ignored. In the future, two or more independent clinicians will be required to evaluate MRI to rule out the influence of normal and irregular rotator cuff tendons after surgery. Finally, there is a lack of long-term follow-up studies of MRI scans. The repaired tendon may undergo a healing process. After a period of recovery, it is still unknown whether the retear signs will disappear.