1 Introduction

Delamination is a horizontal and partial thickness split of the tendon substance between layers of a ruptured rotator cuff.^1,2 A widely ranging incidence of 38–92% has been reported, but the precise cause of delamination remains unknown.^1–3 In a histological study, Sonnabend et al.⁴ demonstrated that laminated tears of the rotator cuff generally occurred between two layers of different collagen fiber orientations. Nimura et al.⁵ suggested that a large portion of the deep layer observed in delaminated rotator cuff tears (RCTs) may be composed of an articular capsule. However, Cha et al.⁶ revealed that the deep layer reached the musculotendinous junction on magnetic resonance imaging (MRI) scans of patients with delaminated RCTs, and that the deep layer is considered to be a structure that serves as a tendon.

In 2008, Mochizuki et al.⁷ anatomically demonstrated an overlap of the distal supraspinatus (SS) tendon by the anterior part of the infraspinatus (IS), with the IS inserting more anteriorly than was previously thought.⁷ As a result, the anterior part of the IS tendon was found to overlap with the SS tendon to achieve an insertion that was quite anterior to the greater tuberosity of the humerus.⁸ In 2019, Yoo et al.⁹ described that articular-sided partial thickness RCTs were associated with the SS occupation ratio.

However, Mochizuki et al.⁷ reported that the mean anteroposterior width of the SS footprint was only 12.6 ± 2.0 mm. Furthermore, Han et al.¹⁰ and Tanaka et al.¹¹ described that the larger the size of the randomized controlled trial (RCT), the more frequent was the appearance of the torn site delamination in the posterior part compared with the anterior part. In 2016, Cha et al.⁶ demonstrated that 98.1% of the delaminated deep layer progressed in the posteromedial direction, leading to the assumption that the deep layer is most likely associated with the IS and functions as a tendon.

In 2012, Kato et al.⁵ revealed that the IS was composed of oblique and transverse parts in accordance with muscle fiber direction. They described that both parts had partially independent morphology and that the transverse part inserted into the main tendinous portion of the oblique part as a thin tendinous membrane.⁵ In 2017, Bacle et al.¹² revealed that the IS muscles were composed of three groups of fibers (cranial, central, and caudal) organized in two planes. They reported that the directions and distributions of the three groups of fibers were different (cranial and caudal parts in a superficial plane and central part in a deep plane).¹²

MRI is a valuable tool for the diagnosis and preoperative assessment of rotator cuff diseases.¹³ Rotator cuff muscle atrophy is typically assessed using the occupation ratio on oblique-sagittal MRI.¹⁴ Several studies have reported that the occupation ratio is associated with medial retraction of the rotator cuff.^15–17 Sasaki et al.¹⁵ demonstrated that the muscle volume did not change soon after surgery compared with the preoperative values, and in patients with moderate medial retraction or extended tearing in the transverse direction, the occupation ratio increased due to restoration of the SS muscle.

The purpose of the present study was to analyze the occupation ratios of the SS and IS muscles before and soon after delaminated rotator cuff repair. The authors hypothesized that the deep layers were related to the occupation ratios of the SS and deep plane of the IS.

2 Materials and methods

Following Institutional Review Board exempt approval (DKUH 2019-04-018), a total of 300 patients who underwent arthroscopic rotator cuff repair for delaminated tears were retrospectively enrolled in the study from June 2017 to December 2019. All procedures were performed by two surgeons (J.B.S. and J.S.Y.), using the suture bridge technique with an all-layer repair. Delamination was defined by a distinguishable interstitial horizontal gap or edge cleavage tearing between the superficial and deep surfaces of the torn tendon and was assessed intraoperatively by the two surgeons (J.B.S. and J.S.Y.). The subjects were divided into two groups: group A, non-delaminated tears; group B, deep layer, more medially retracted delaminated tears.

The inclusion criteria for this study were delaminated RCTs and arthroscopic en masse transosseous-equivalent suture bridge repair of all layers. The exclusion criteria were partial thickness RCTs, tears not amenable to complete repair, repair with biceps augmentation or rerouting, repair with footprint medialization, a superficial layer that was more retracted than the deep layer, previous surgery on the affected shoulder, and severe fatty infiltration precluding measurement of the occupation ratio.

2.2

2.2 MRI evaluation

MRI was performed using a 3.0 T system (Ingenia 3.0 T; Philips, Houston, TX, USA) with the same protocol for each patient. The MRI protocol included obtaining proton density-weighted and T2-weighted oblique-coronal fat-saturated spin-echo images (3,300/14–95 [repetition time, ms/echo time, ms]; section thickness, 3 mm; intersection gap, 0.8 mm; field of view, 16 cm), T1-weighted oblique-coronal fat-saturated spin-echo images (777/12 [repetition time, ms/echo time, ms]; section thickness, 3 mm; intersection gap, 0.6 mm; field of view, 16 cm), T1-weighted oblique-sagittal spin-echo images (600/12 [repetition time, ms/echo time, ms]; section thickness, 3 mm; intersection gap, 1.2 mm; field of view, 16 cm), and T1-weighted transverse spin-echo images (600/12 [repetition time, ms/echo time, ms]; section thickness, 3 mm; intersection gap, 0.9 mm; field of view, 16 cm).

The size of the RCT was measured using the maximum diameter of the tear on oblique-sagittal T2-weighted images.¹⁸ The retraction of the RCT was measured using the maximum straight-line distance from the lateral edge of the torn layer to the most superior point of the greater tuberosity on oblique-coronal T2-weighted images.^18,19 In delaminated tears, the retraction length of the deep layers was defined as the lateral 2 mm from the margin of the articular cartilage.¹⁹

Muscle volume was evaluated by measuring the occupation ratios of the SS and IS in the most lateral view of the T1-weighted oblique-sagittal images in which the scapular spine remained in contact with the scapular body.²⁰ The occupation ratios of the SS to the fossa were measured using Thomazeau et al.’s method.²⁰ The number of pixels in the segmented area of the muscle belly was automatically measured using the INFINITT PACS software (INFINITT, Seoul, Republic of Korea), after which the cross-sectional area of the muscle was measured. In the same manner, the cross-sectional area of the SS fossa was also measured. The occupation ratio of the SS was calculated as the SS area divided by the SS fossa area. The occupation ratios of the IS were measured according to Kikukawa et al.’s method.²¹ The anatomic external rotation (aER) muscle was defined as the area surrounded by the posterior surface of the scapula, inferior margin of the teres minor (TM) muscle, and anterior surface of the posterior deltoid, including the areas of the IS and TM muscles.²¹ The occupation ratios of the IS were calculated as the IS area divided by the aER muscle area. Additionally, the IS was divided into two compartments (deep and superficial) based on modifications to the concept of Bacle et al.¹² and the method of Seo et al.¹⁸ (Fig. 1). MRI was obtained before surgery and shortly after surgery (mean, 5.3 days; range, 5–7 days). Restoration of the rotator cuff muscle was defined as the difference between the preoperative and postoperative occupation ratios.

Close

Fig. 1 T1-weighted oblique-sagittal image. White line surrounds the supraspinatus fossa and aER muscle; white dotted line surrounds the SS and IS muscle. The central part is classified as the deep plane, and the cranial and caudal parts were classified as the superficial plane. SS, supraspinatus; aER, anatomic external rotator; IS, infraspinatus.

Export to PPT

All MRI scans were measured by a consensus readout of two blinded observers (S.J.K. and J.S.Y.). Measurements were performed independently, and the results were not disclosed to the other surgeon. The mean value of the duplicate scores was used as the representative value. The intraobserver and interobserver correlations showed almost perfect agreements (Table 1).

Table 1 Reliability of grouping and measurement of occupation ratio.

Measurement position	ICC (JSY vs JSY)	ICC (SJK vs SJK)	ICC (JSY vs SJK)
Occupation ratio of supraspinatus	0.92	0.90	0.88
Occupation ratio of infraspinatus	0.90	0.87	0.85
Occupation ratio of infraspinatus deep plane	0.88	0.84	0.82
Occupation ratio of infraspinatus superficial plane	0.88	0.84	0.82

3 Results

3.1

3.1 Demographic data

A total of 150 patients who received the same 3.0-T system MRI and underwent arthroscopic en masse transosseous-equivalent suture bridge repair for RCTs were retrospectively enrolled in the study from June 2017 to April 2019. Patients with partial thickness RCTs (n = 68), tears not amenable to complete repair (n = 24), repair with biceps augmentation or rerouting (n = 18), repair with footprint medialization (n = 8), a superficial layer that was more retracted than the deep layer (n = 8), previous surgery on the affected shoulder (n = 14), and severe fatty infiltration precluding measurement of the occupation ratio (n = 16) were excluded. Of the remaining 144 cases, 132 patients who underwent postoperative MRI were included. Delaminated rotator cuff was found in 74 (56.1%) cases (Fig. 2). Finally, 132 patients with RCTs were divided and classified into the following groups: 58 in group A (non-delaminated tears) and 74 in group B (deep layer, more medially retracted delaminated tears). There were no significant differences in the demographic data, including age, sex, ratio of dominant arm involvement, weight, height, BMI, ratio of smoking, ASA scores, and symptom duration, between the groups (Table 2).

Close

Fig. 2 Study flow diagram. MRI: magnetic resonance image, RCTs: rotator cuff tears.

Export to PPT

Table 2 Demographic data.

	Group A (n = 58)	Group B (n = 74)	p- value
Age (Mean ± SD)	59.1 ± 7.7	57.2 ± 8.9	0.375
Sex (Male/Female)	34/24	48/26	0.604
Dominant arm/Non-dominant arm	26/32	42/32	0.262
Weight (Kg, Mean ± SD)	67.6 ± 9.2	65.7 ± 9.8	0.835
Height (cm, Mean ± SD)	163.6 ± 8.2	162.8 ± 6.8	0.468
Body mass index (Mean ± SD)	25.3 ± 3.5	24.8 ± 3.3	0.646
Smoking/Non-smoking	46/12	52/22	0.405
ASA score (1:2:3)	34:22:2	44:24:6	0.810
Symptom duration (months)	18.7 ± 36.8	15.3 ± 23.4	0.445

3.4

3.4 Analysis of the occupation ratios according to the size of the delaminated RCTs

For the evaluation of the SS and IS occupation ratios according to the size of the delaminated RCTS, 38 patients in group B were divided into two subgroups: group I (≤2-cm-sized delaminated RCTs, n = 17) and group II (>2-cm-sized delaminated RCTs, n = 20).

The mean RCT was larger in group II than in the other groups (p < 0.001), the mean superficial layer retraction was greater in group II than in the other groups (p = 0.004), and the mean deep layer retraction was greater in group II than in the other groups (p < 0.001, Table 4).

Table 4 Analysis of occupation ratios according to the size of delaminated rotator cuff tears.

	Non-delaminated RCTs (n = 58)	Deep layer more retracted delaminated RCTs (n = 74)		p- value
		Same or less than 2 cm sized delaminated RCTs (n = 34)	More than 2 cm sized delaminated RCTs (n = 40)
Size (mm)	14.8 ± 6.8	14.4 ± 3.2	26.8 ± 7.6	<0.001
Superficial layer Retraction (mm)	12.8 ± 7.9	11.2 ± 3.7	18.3 ± 6.5	0.004
Deep layer retraction (mm)	–	15.0 ± 6.1	22.0 ± 7.2	<0.001
Pre-operative occupation ratio (%)				–
Supraspinatus	59.9 ± 10.6	54.9 ± 8.0	45.8 ± 10.4	<0.001
Infraspinatus	62.8 ± 10.2	65.3 ± 5.9	59.7 ± 5.9	N.S.
Deep plane	31.6 ± 6.9	33.8 ± 4.7	27.8 ± 7.5	0.022
Superficial plane	31.2 ± 6.3	31.5 ± 4.7	31.9 ± 6.2	0.923
Post-operative occupation ratio (%)
Supraspinatus	66.9 ± 10.0	66.8 ± 7.0	62.9 ± 9.6	0.280
Infraspinatus	70.1 ± 9.1	70.0 ± 6.4	70.0 ± 7.8	0.999
Deep plane	34.4 ± 7.4	35.7 ± 5.0	35.4 ± 6.1	0.768
Superficial plane	35.7 ± 6.5	34.3 ± 4.8	34.6 ± 4.9	0.674
Restoration of occupation ratio (%)
Supraspinatus	7.0 ± 4.0	11.9 ± 6.3	17.1 ± 8.1	<0.001
Infraspinatus	7.3 ± 6.0	4.7 ± 3.4	10.3 ± 7.2	0.017
Deep plane	2.8 ± 3.7	1.9 ± 2.0	7.6 ± 5.1	<0.001
Superficial plane	4.5 ± 2.8	2.8 ± 3.2	2.7 ± 5.4	0.242

The mean preoperative SS occupation ratio was lower in group II than in the other groups (p < 0.001), and the mean preoperative IS deep plane occupation ratio was also lower in group II than in the other groups (p = 0.022); however, the mean preoperative total IS and IS superficial plane occupation ratios showed no difference between the groups (Fig. 3, Table 4).

Close

Fig. 3 (A) Comparison of the preoperative occupation ratios of the SS. (B) Comparison of the preoperative occupation ratios of the total IS. (C) Comparison of the preoperative occupation ratios of the IS deep plane. (D) Comparison of the preoperative occupation ratios of the IS superficial plane. Statistically significant differences are indicated by brackets and an asterisk. SS; supraspinatus, IS; infraspinatus.

Export to PPT

The mean values of the postoperative SS occupation ratio were 66.8 in group I and 62.9 in group II, and the mean values of the postoperative IS occupation ratio were 70.0 in group I and 70.0 in group II. The mean values of the postoperative IS deep plane occupation ratio were 35.7 in group I and 35.4 in group II, and the mean values of the postoperative IS superficial plane occupation ratio were 34.3 in group I and 34.6 in group II. There was no statistically significant difference between the groups (Table 4).

The mean values of the SS occupation ratio restoration were higher in group II than in the other groups (p < 0.001), and the mean values of the IS occupation ratio restoration were higher in group II than in the other groups (p = 0.017). The mean values of the IS deep plane occupation ratio restoration were higher in group II than in the other groups (p < 0.001); however, the mean values of the IS superficial plane occupation ratio restoration showed no difference between the two groups (Fig. 4, Table 4).

Close

Fig. 4 (A) Comparison of the restorations of the SS occupation ratio. (B) Comparison of the restorations of the total IS occupation ratio. (C) Comparison of the restorations of the IS deep plane occupation ratio. (D) Comparison of the restorations of the IS superficial plane occupation ratio. Statistically significant differences are indicated by brackets and an asterisk. SS, supraspinatus; IS, infraspinatus.

Export to PPT

4 Discussion

The major findings of the present study were that the restoration of the SS occupation ratio was greater in the delaminated tears than in the non-delaminated tears. In addition, the restoration of the SS and IS deep plane occupation ratios was greater in >2-cm-sized delaminated tears than in ≤2-cm-sized delaminated and non-delaminated tears.

Muscle atrophy of the SS has previously been assessed using the occupation ratio along the oblique-sagittal plane on MRI. When this method is used, the cutting plane is defined by osseous landmarks, such as the scapular spine, and is fixed in any series of examinations.^16,17 The cutting plane of the muscle belly where the cross-sectional area is measured will change because the tendons are repaired by lateral traction toward the facet of the greater tuberosity. Increases in the occupation ratio may simply indicate the change in the cutting plane of the muscle belly, owing to structural changes caused by the operation, rather than the recovery of the muscle volume.^16,17 Furthermore, Lhee et al.¹⁷ reported that the increase in the occupation ratio observed soon after surgery depended on the degree of the preoperative SS tendon medial retraction. Chung et al.²² also described that both the cross-sectional area and the volume of SS were increased immediately postoperatively through combined two-dimensional (2D) and three-dimensional (3D) evaluations. However, Sasaki et al.¹⁵ reported that while the muscle volume did not change soon after surgery compared with the preoperative values based on a 3D evaluation, the occupation ratio did increase, probably due to lateral traction of the SS muscle. Although the occupation ratio may not reflect the muscle volume of the rotator cuff, it is related to retraction of the torn tendon. Therefore, the occupation ratios were used for the evaluation of medial retraction and postoperative restoration according to the delamination pattern in the present study.

Nimura et al.⁵ conducted an anatomic study on the attachment of the articular capsule on the greater tuberosity by removing the rotator cuff. The width of the articular capsule was measured as between 3.5 and 9.1 mm, with a thicker footprint than previously believed.⁵ Their study suggests that a large portion of the deep layer observed in delaminated RCTs may be composed of an articular capsule.⁵ However, Cha et al.⁶ revealed that the deep layer reached the musculotendinous junction on the MRI scans of patients with delaminated RCTs and that the deep layer can be considered a structure that serves as a tendon connected to the IS and SS, rather than a static structure consisting of the capsule alone. They demonstrated that the retraction pattern of the delaminated deep layer may be affected by the IS and SS.⁶ The SS muscle consists of an anterior (SSa) and a posterior (SSp) part. The SSa muscular fibers constitute a fusiform muscle by inserting onto a long intramuscular tendinous core laterally continued by a cordlike tendon.^23,24 The SSp muscle is smaller and unipennate, and its muscular fibers run laterally into a short and flat strap-like tendon at the undersurface of the muscle.^8,23,24 The SSp tendon and its footprint both thin posteriorly, and the tendon inserts just posterior to the SSa tendon into the medial margin of the anterior third of the greater tuberosity.^8,23,24 The authors contend that this structure is the reason why delaminated tears are common in larger than medium-sized RCTs. Tanaka et al.¹¹ reported the prevalence of delaminated RCTs according to the size of the tear (small: 20.5%, medium: 53.3%, large: 67.6%, massive: 60.0%), and the present study also showed that the delaminated tears were larger than the non-delaminated tears.

Michelin et al.⁸ reported that the overlapping of the SSp and the IS tendons was distinguishable on MRI. They also revealed that a linear area of increased signal was observed between the SSp and the IS tendons and demonstrated that the anterior part of the IS tendon overlapped with the SS tendon to reach an insertion anterior to the greater tuberosity of the humerus.⁸ Rahu et al.²⁵ demonstrated that the SS tendon consists of superficial and deep sections. Moreover, the superficial portion extends directly to the humerus, and the deep layer is tightly connected to the rotator cable and rotator crescent area.²⁵ In addition, Yoo et al.⁹ reported that articular-sided partial thickness RCTs are associated with the occupation ratio of the SS. Therefore, we hypothesized that the deep layer in delaminated RCTs is related to the SSp tendons. However, division of the SS muscle in the T1-weighted oblique-sagittal images was not performed because it was not our objective, although the tendinous portions of the SSa and SSp were distinguishable on MRI (Fig. 5). In line with our hypothesis, this study showed that the SS occupation ratio on preoperative MRI was significantly lower and that the restoration of the SS after surgery was greater in the delaminated tears than in the non-delaminated tears (Fig. 6).

Close

Fig. 5 Tracing serial oblique sagittal images of 54-years old female with delaminated rotator cuff tears. Division of the SS muscle it was not objective, although tendinous portions of anterior and posterior part of SS were distinguishable. The tendons of IS showed two layers, deep tendon is located in the central fiber and superficial tendon is located between central fiber and cranial fiber Asterisk: tendinous portions of anterior and posterior part of SS, Arrow: deep tendon of IS, arrowhead: superficial tendon of IS, SS; supraspinatus, IS; infraspinatus.

Export to PPT

Close

Fig. 6 MRI findings of 61-years old male with more than 2 cm sized delaminated rotator cuff tears (A) Oblique-sagittal image before surgery. (Preoperative SS occupation ratio; 50.4, total IS occupation ratio; 46.4, IS deep plane; 15.9, IS superficial plane; 30.5) (B) Oblique-sagittal image 5 days after surgery. (Postperative SS occupation ratio; 73.7, total IS occupation ratio; 72.8, IS deep plane; 33.5, IS superficial plane; 39.2, Restoration of SS; 10.8, total IS; 23.4, IS deep plane; 17.7, IS superficial plane 8.7) MRI; magnetic resonance imaging, SS; supraspinatus, IS; infraspinatus.

Export to PPT

However, delaminated RCTS are common in larger than medium-sized RCTs,¹¹ and Cha et al.⁶ revealed that 98.1% of the delaminated deep layer progressed in the posteromedial direction, leading to the assumption that the deep layer is most likely associated with the IS and functions as a tendon. Mochizuki et al.⁷ reported that the mean anteroposterior width of the SS footprint was 12.6 ± 2.0 mm. However, Rahu et al.²⁵ revealed that the SS tendon inserts more posteriorly and the IS tendon insert more anteriorly than previously reported. In 2012, Kato et al.⁵ reported that the IS was composed of two groups (oblique and transverse) in accordance with muscle fiber direction. They conducted a histological study showing that almost the entire tendinous portion of the IS is derived from the inferior part and that the tendinous portion of the superior part is comprised of thin membrane-like tissues, which attach to the tendinous portion of the oblique part.⁵ In 2017, Bacle et al.¹² demonstrated that the IS muscles were composed of three groups of fibers (cranial, central, and caudal) organized in two planes. In this study, the central group was classified as the deep plane and the cranial and caudal parts were classified as the superficial plane. In almost all cases, the IS tendons showed two layers, with the deep tendon located in the central fiber and the superficial tendon located between the central and cranial fibers (Fig. 5). In line with our hypothesis, the restoration of the IS deep plane occupation ratio after surgery was higher in the group with delaminated tears than in the group with non-delaminated tears, although the total IS and IS superficial plane occupation ratio showed no statistical differences between the two groups. Furthermore, the restoration of the IS deep plane occupation ratio after surgery was greater in ≥2-cm-sized delaminated tears than in <2-cm-sized delaminated tears in the analysis of occupation ratios according to the size of the delaminated RCT (Fig. 6).

The association between delamination and outcomes remains unclear. Boileau et al.¹ and Flurin et al.² have reported that delamination was a negative prognostic factor for the anatomic results of rotator cuff repair. On the other hand, Kwon et al.²⁶ described that delamination was not an independent prognostic factor for rotator cuff healing in a retrospective analysis of 1043 consecutive cases. There are various surgical techniques for delaminated RCTs. In 2016, Kim et al.²⁷ reported that both conventional en masse repair and separate double-layer double-row repair were effective in improving clinical outcomes in the treatment of delaminated RCTs in a randomized controlled study. However, when delamination occurs, the deep layer is typically more retracted than the superficial layer and is likely to be repaired under tension when reduced into the footprint as compared with the superficial layer.^3,4,27 As a result, Kim et al.²⁷ hypothesized that better outcomes would be seen in bursal layer–only repair, as the articular layer is under tension when repaired into its footprint during the all-layers repair, although the outcomes of the two methods showed no differences. However, Kim et al.²⁸ reported that patients with an incompletely repaired articular-side rotator cuff revealed a higher postoperative re-tear rate than that of patients with a complete repair. In 2019, Kim et al.¹⁹ recommended emphasizing the reduction of tension within the deep layer during the repair procedure, because the retraction length of the deep layer was identified as the only significant factor influencing repair integrity in delaminated RCTs.

Understanding the anatomy of the overlap of tendons and division of rotator cuff muscles is necessary to guide decision-making regarding surgical methods in the treatment of delaminated RCTs. The present study showed that the deep layer is associated with the SS and the deep plane of the IS. Therefore, in the surgical treatment of RCTs, restoration of the retracted deep layer should also be considered for the restoration of medially retracted rotator cuff muscles.

There were several limitations in the present study. First, despite the excellent interobserver agreement, measurement errors for the division of the IS groups were still possible. To reduce such errors, measurements were performed independently, and the results were not disclosed to the other surgeon. The mean value of the duplicate scores was used as the representative value. Second, the occupation ratio evaluation with 2D MRI scans was performed without 3D evaluation. Vidt et al.²⁹ revealed that there were no significant associations between single-image assessments and 3D measurements of fatty infiltration of the SS and IS. However, muscle evaluation using oblique-sagittal MRI is still widely used, due to its simplicity and relevance.^30,31 Moreover, comparison of the preoperative and shortly postoperative occupation ratios is very useful for evaluating the medial retraction of the rotator cuff muscles.¹⁵ Third, although muscle atrophy is affected not only by medial retraction of the tendons but also age, sex, activity, general condition, symptom duration, and trauma history, this study only evaluated the relationship between the pattern of the RCTs and the occupation ratios due to medial retraction of the tendons using MRI scans without addressing the other factors. Fourth, complete repair was intraoperatively assessed by the footprint coverage using bursal layer repair after surgery, and the deep layer repair status was not assessed using glenohumeral joint exploration according to the method of Kim et al.,²⁸ although the deep layer was always checked when passing the sutures through in all of the cases. Finally, this study involved only a time-zero evaluation of the postoperative occupation ratio changes. We believe that the significant differences in the preoperative findings and similarity of the postoperative occupation ratios led to clinical outcomes similar to those reported in previous studies.^27,32 However, whether there was healing of the reduced deep layer along with the superficial layer or a re-tear without healing was not evaluated. Therefore, further MRI follow-up studies to determine the correlations between clinical outcomes are necessary.

5 Conclusion

Restoration of the SS occupation ratio was higher in delaminated tears than in non-delaminated tears. In addition, restoration of the SS and IS deep plane occupation ratios were higher in >2-cm-sized delaminated tears than in ≤2-cm-sized delaminated and non-delaminated tears.

Funding

There is no funding source.

Ethical approval

This article does not contain any studies with human participants or animals performed by any of the authors.

“All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.”

Informed consent

“Informed consent was obtained from all individual participants included in the study.”

Authors' contributions

SHY, JBS designed this study, participated in the sequence alignment, and drafted the manuscript.

JWP gathered the retrospective demographic data of the cases.

SJK, JSY measured the radiologic findings, divided the groups, and wrote the manuscript.

All authors read and approved the final manuscript.