The Prevalence of Tear Patterns and Their Effects on Tendon Healing After Arthroscopic Surgery in Patients With Full-Thickness Rotator Cuff Tears

Abstract

Background:

To achieve successful anatomic rotator cuff repair with minimal tension, both the tear pattern and tear size should be considered. However, little information is available concerning the frequency of tear patterns and their effects on tendon healing.

Purpose:

To evaluate the distribution of tear patterns in full-thickness rotator cuff tears and whether these patterns affect tendon healing after arthroscopic repair.

Study Design:

Case-control study; Level of evidence, 3.

Methods:

Between 2014 and 2021, patients who underwent arthroscopic surgery for symptomatic full-thickness rotator cuff tears with a minimum 2-year follow-up with postoperative magnetic resonance imaging or ultrasound were retrospectively reviewed. After the debridement of degenerative tendon tissue during arthroscopic surgery, the tear pattern was classified as crescent, U, or anterior or posterior L shaped. Intergroup differences in clinical and radiological characteristics were analyzed. In the subgroup analysis, patients were divided into 2 subgroups: small-to-medium or large-to-massive tears.

Results:

Among the 1037 patients with a full-thickness rotator cuff tear, the most common tear pattern was crescent shaped (39.6%), followed by posterior L, U, and anterior L shaped (26.0%, 21.4%, and 12.9%, respectively). In the subgroup analysis, 713 patients (68.8%) had small-to-medium tears, while 324 (31.2%) had large-to-massive tears. The proportion of large-to-massive tears was significantly higher for the anterior L-shaped tear pattern than for the other tear patterns (24.8%, 28.8%, 52.2%, and 32.6% for crescent, U, and anterior and posterior L shaped, respectively; P < .001). The anterior L-shaped tear pattern had a significantly higher retear rate than the other tear patterns in small-to-medium tears (7.8%, 13.0%, 28.0%, and 10.6% for crescent, U, and anterior and posterior L shaped, respectively; P < .001). The rate of revision surgery because of a symptomatic retear within 2 years after primary surgery was significantly higher for the anterior L-shaped tear pattern than for the other tear patterns (3.8%, 7.5%, 21.6%, and 0.0% for crescent, U, and anterior and posterior L shaped, respectively; P = .002).

Conclusion:

The prevalence of tear patterns varied depending on the tear size. In small-to-medium tears, the anterior L-shaped tear pattern had the lowest incidence among the tear patterns; however, it had a significantly higher retear rate. Furthermore, the anterior L-shaped tear pattern had a higher incidence of retears requiring early revision surgery than the other tear patterns.

Keywords

rotator cuff tear patterns arthroscopic rotator cuff repair retear

Various surgical techniques and options have been applied recently based on tear size to repair full-thickness rotator cuff tears successfully.²⁵ However, 3-dimensional tear patterns should be considered to achieve successful anatomic rotator cuff repair.^23,30 The geometric classification most commonly used for describing 3-dimensional patterns of full-thickness rotator cuff tears categorizes them into 4 primary patterns: crescent, U, and anterior and posterior L shaped.^16,26 Recognizing tear patterns is essential for anatomic repair with minimal tension and to prevent incorrect force transmission.⁴ The crescent-shaped tear pattern has medial-to-lateral mobility, allowing the torn tendon to be repaired directly on the greater tuberosity with minimal tension. The U-shaped tear pattern has anterior-to-posterior mobility with a sufficient remnant anterior supraspinatus tendon so that the torn tendon can be repaired in a side-to-side manner. The L-shaped tear pattern consists of a mobile edge on the anterior or posterior part of the torn supraspinatus tendon, and the mobile tendon edge can be reduced onto the footprint anterolaterally for the anterior L-shaped tear pattern and posterolaterally for the posterior L-shaped tear pattern. Therefore, surgical techniques to achieve anatomic reduction of torn tendons may vary according to tear patterns.^20,32 Although preoperative magnetic resonance imaging (MRI) could be used to predict some rotator cuff tear patterns and aid in surgical planning, minimal information related to clinical and radiological characteristics of tear patterns is available.

Rotator cuff tears are most likely to begin within the rotator crescent and propagate in both the anterior and posterior directions.^11,21 The stress distribution of the torn rotator cuff tendon and the risk of tear progression differ based on the tear pattern.^10,29,31 Therefore, the prevalence of tear patterns may vary according to the tear size, and the specific morphological characteristics of each tear pattern regarding the tear size could influence tendon healing. Several clinical studies have evaluated the effect of tear patterns on clinical outcomes and tendon integrity after arthroscopic rotator cuff repair.^16,26,33 However, the prevalence of rotator cuff tear patterns classified by tear size has not been documented, and minimal information exists about their effects on retears.

This study aimed to compare the distribution of tear patterns between small-to-medium and large-to-massive full-thickness rotator cuff tears and evaluate whether the tear pattern affects tendon healing after arthroscopic repair. The hypothesis was that the prevalence of tear patterns would vary depending on the tear size, and the retear rate would differ based on the tear pattern.

Methods

Patient Selection

We retrospectively reviewed the records of patients who underwent arthroscopic surgery with symptomatic full-thickness rotator cuff tears at one institution from March 2014 to June 2021. Patients were included in this study if they underwent primary arthroscopic surgery for a symptomatic full-thickness rotator cuff tear, MRI preoperatively, and imaging including MRI or ultrasonography at 6 months postoperatively and had been followed up for a minimum of 2 years after primary surgery. Patients were excluded if they underwent primary arthroscopic repair for a partial-thickness rotator cuff tear or an isolated subscapularis tear and had complex tear patterns including flap tears, tears at the musculotendinous junction, and tears not classified into specific tear patterns; irreparable tears requiring salvage procedures such as superior capsule reconstruction or tendon transfer; revision surgery; acute trauma; shoulder instability; glenohumeral osteoarthritis; a history of infections; neurological or systemic diseases influencing the shoulder joint; or incomplete medical records.

Patients were categorized into 4 groups based on the tear pattern that was classified by the senior author (S.-J.S.) during arthroscopic surgery after the debridement of degenerative tendon tissue: crescent-, U-, and anterior and posterior L-shaped tears. A crescent-shaped tear was defined as an anteroposterior (AP) tear size larger than the mediolateral (ML) tear size, with the torn tendon reduced directly onto the footprint at the greater tuberosity with minimal tension because of medial-to-lateral mobility. A U-shaped tear was defined as an ML tear size larger than the AP tear size, with the torn tendon reduced in the anterior-to-posterior direction because of anterior-to-posterior mobility and a sufficient remnant anterior supraspinatus tendon. An anterior L-shaped tear was defined as a tear that propagated through the rotator interval and anterior rotator cable, resulting in exposure of the long head of the biceps tendon and a rupture of the anterior attachment of the rotator cable and the cord of the supraspinatus tendon. The anterolateral tear apex is the mobile portion of the torn supraspinatus tendon and fits at the anterolateral corner of the footprint. However, a torn supraspinatus tendon is usually retracted medially, while the infraspinatus tendon is relatively preserved. A posterior L-shaped tear was defined as a tear that propagated along the interval between the supraspinatus and infraspinatus tendons, where the posterolateral tear apex of the torn supraspinatus tendon is the mobile portion of the torn tendon and fits at the posterolateral corner of the footprint. A posterior L-shaped tear was defined as a torn supraspinatus tendon less retracted medially than an anterior L-shaped tear. For the subgroup analysis, patients were divided into 2 subgroups according to the tear size: small-to-medium tears (<3 cm) and large-to-massive tears (≥3 cm).⁵ The institutional review board approved the study protocol.

Clinical Evaluation

All patients completed a questionnaire before surgery regarding arm dominance, occupation, symptom duration, and characteristics, such as age, sex, smoking status, and history of diabetes mellitus and dyslipidemia. Occupations were classified as heavy manual labor with high-demand shoulder activity (including farm, factory, and construction work) and nonmanual labor. Body mass index was calculated as the body weight divided by the square of the height. Preoperative parameters were assessed by 1 physician assistant with 5 years of orthopaedic experience, who was not involved in this study.

Radiological Evaluation

Preoperative MRI was performed to assess fatty infiltration of the rotator cuff muscles and supraspinatus muscle atrophy. Fatty infiltration was evaluated according to the Goutallier classification, as modified by Fuchs et al,⁶ for each rotator cuff muscle on sagittal MRI sections. Atrophy of the supraspinatus muscle was classified according to whether the tangent sign was present or absent on sagittal MRI. Postoperative MRI or ultrasound was performed to assess tendon integrity at 6 months after primary rotator cuff repair, regardless of symptoms. A rotator cuff retear was defined as complete discontinuity of the repaired rotator cuff tendon on postoperative imaging. A blinded musculoskeletal radiologist not involved in this study interpreted preoperative MRI and postoperative imaging scans.

Surgical Procedure and Intraoperative Evaluation

All surgical procedures were performed by the senior author, with patients under general anesthesia in the lateral position. Overall, 4 routine arthroscopic portals (anterior, posterior, lateral, and posterolateral) and 1 puncture for suture anchor placement were used to perform arthroscopic surgery. Intra-articular lesions were assessed through the posterior portal, including those involving the long head of the biceps tendon and the subscapularis tendon. Partial tears of the long head of the biceps tendon involving up to 50% of the tendon underwent debridement, and tears involving >50% or tendons with subluxations or dislocations were treated with tenodesis or tenotomy. The arthroscope was then inserted into the subacromial space from the posterior portal. After the torn end of the rotator cuff was debrided, the tear size of the rotator cuff was measured with a calibrated measuring device, and the tear pattern was classified. Crescent-shaped tears were repaired using the modified Mason-Allen single-row or suture bridge technique. U-shaped tears were repaired in a side-to-side manner using double-row suture anchors. Anterior L-shaped tears were treated using either the side-to-side or margin convergence technique between the anterior supraspinatus tendon and rotator interval tissue, partial repair, or anterior cable reconstruction (ACR). Partial repair was defined as a residual defect of the footprint larger than 1 × 1 cm² after primary repair.²⁴ ACR using a proximal part of the long head of the biceps tendon was performed if partial repair was expected because of retraction of the torn supraspinatus tendon with an anterior L-shaped tear and the biceps tendon was intact or had a partial tear of up to 50%.³⁴ The anterior rotator cable was reconstructed to restore superior stability and protect the adjacent tendon from progressive stress for an anterior L-shaped tear, which ruptured the anterior attachment of the rotator cable.²⁷ Posterior L-shaped tears were treated using a combined side-to-side technique in the posterior segment of the supraspinatus and infraspinatus tendons on the posterolateral corner of the footprint with modified Mason-Allen repair in the anterior segment of the supraspinatus tendon.

Postoperative Rehabilitation

All patients were required to wear an abduction brace with immobilization for 1 month. Passive shoulder range of motion (ROM) was allowed below the waist level on the first day of surgery when pain was tolerable. From 1 month postoperatively, patients were instructed to perform progressive active-assisted passive ROM exercises gradually, followed by active ROM exercises. At 3 months postoperatively, if the patients achieved over 80% of ROM compared with that of the opposite side, they could begin shoulder muscle strengthening exercises. All sports activities were permitted at 6 months after surgery.

Statistical Analysis

All statistical analyses were conducted using SPSS (Version 25.0; IBM). Summary statistics are presented as means with standard deviations for continuous variables and as the number of patients with percentages for categorical variables. For continuous variables, the Student t test was used to compare mean differences between groups. One-way analysis of variance and the post hoc least significant difference test were used for the comparison of the 4 groups of tear patterns. For categorical variables, the chi-square test or Fisher exact test was used to assess differences in proportions between the subgroups of tear sizes. Logistic regression analysis was performed to evaluate the effect of multiple factors by calculating the odds ratio (OR) and 95% confidence interval (CI). The level of significance was set as 2-tailed P < .05 for all tests.

Results

A total of 1556 consecutive patients who underwent arthroscopic surgery with symptomatic full-thickness rotator cuff tears were retrospectively reviewed. Among these patients, 108 were excluded because of revision rotator cuff repair, 31 missed preoperative MRI, 43 had isolated subscapularis tendon tears, and 29 had combined glenohumeral joint instability. Of the 1345 patients with full-thickness tears, 126 (9.4%) had irreparable tears requiring salvage procedures such as superior capsule reconstruction or tendon transfer, 51 (3.8%) had complex tear patterns such as flap tears or musculotendinous junction tears, and 131 (9.7%) were lost to follow-up or missed postoperative imaging. Consequently, 1037 patients were included in this study.

The most common tear pattern was crescent-shaped tears with 411 patients (39.6%), followed by posterior L-shaped tears with 270 patients (26.0%), U-shaped tears with 222 patients (21.4%), and anterior L-shaped tears with 134 patients (12.9%). Preoperative factors are summarized in Table 1. No significant differences between groups were found in preoperative characteristics, except for age. Patients with crescent-shaped tears were significantly younger than those with the other tear patterns.

Table 1

Characteristics of All Patients^a

	Crescent (n = 411)	U (n = 222)	Anterior L (n = 134)	Posterior L (n = 270)	P Value
Age, y	58.6 ± 10.2	61.6 ± 8.5^b	62.0 ± 9.1^b	61.4 ± 10.4^b	<.001
Sex, male:female, n	189:222	98:124	59:75	120:150	.958
Dominant side, %	67.4	72.5	73.1	64.4	.152
Symptom duration, mo	11.4 ± 6.6	12.0 ± 6.5	11.6 ± 6.1	12.4 ± 7.3	.317
Body mass index, kg/m²	25.1 ± 3.1	24.7 ± 3.3	24.8 ± 3.5	25.3 ± 2.9	.227
Heavy labor occupation, %	17.0	17.6	8.2	14.8	.071
Diabetes mellitus, %	11.9	10.4	7.5	11.5	.525
Dyslipidemia, %	52.1	57.7	57.5	52.2	.423
Smoking, %	10.0	8.6	4.5	10.4	.216

Data are reported as mean ± SD unless otherwise indicated.

P < .05 versus crescent (Bonferroni post hoc test).

Anterior L-shaped tears showed significantly advanced fatty infiltration of the supraspinatus and infraspinatus compared with the other tear patterns (Table 2). Fatty infiltration of the subscapularis was more significantly advanced for anterior L-shaped tears than for crescent-shaped tears. U- and anterior L-shaped tears exhibited a significantly greater proportion of supraspinatus muscle atrophy than crescent- or posterior L-shaped tears. The proportion of large-to-massive tears was significantly greater for anterior L-shaped tears than for the other tear patterns. The AP tear size was significantly smaller for U-shaped tears than for the other tear patterns, while the ML tear size was significantly larger for anterior L-shaped tears and significantly smaller for crescent-shaped tears than for the other tear patterns. Regarding partial repair and additional ACR, 16 patients (3.9%) with crescent-shaped tears, 20 (9.0%) with U-shaped tears, 14 (10.4%) with anterior L-shaped tears, and 7 (2.6%) with posterior L-shaped tears underwent partial repair, while 43 patients (32.1%) with anterior L-shaped tears underwent ACR. Full-thickness subscapularis tears were significantly more frequent for anterior L- and U-shaped tears than for crescent- or posterior L-shaped tears (P < .001).

Table 2

Preoperative Radiological and Intraoperative Findings in All Patients^a

	Crescent (n = 411)	U (n = 222)	Anterior L (n = 134)	Posterior L (n = 270)	P Value
Preoperative radiological
Fatty infiltration
Supraspinatus	0.8 ± 0.8^b	0.9 ± 1.0^b	1.4 ± 0.8	0.9 ± 0.9^b	<.001
Infraspinatus	0.5 ± 0.7^b	0.5 ± 0.7^b	0.8 ± 0.9	0.6 ± 0.8^b	<.001
Subscapularis	0.2 ± 0.5^b	0.2 ± 0.4	0.3 ± 0.7	0.2 ± 0.5	.030
Atrophy of suprapinatus, %	8.8^b	13.1^b	31.3	7.8^b	<.001
Intraoperative
Large to massive tear, %	24.8^b	28.8^b	52.2	32.6^b	<.001
Anteroposterior tear size, mm	21.6 ± 7.3^c	17.7 ± 6.3	22.4 ± 6.0^c	21.9 ± 5.8^c	<.001
Mediolateral tear size, mm	15.9 ± 7.1	23.8 ± 5.8^{b , d}	25.3 ± 7.0^d	23.4 ± 5.7^{b , d}	<.001
Partial repair or ACR, %	3.9^{b , c}	9.0^b	42.5	2.6^{b , c}	<.001
Delamination, %	56.9	68.0	61.2	61.1	.059
Subscapularis tear, %	21.2^{b , c}	32.0	41.8	19.6^b,c	<.001

Data are reported as mean ± SD unless otherwise indicated. ACR, anterior cable reconstruction.

P < .05 versus anterior L (Bonferroni post hoc test).

P < .05 versus U (Bonferroni post hoc test).

P < .05 versus crescent (Bonferroni post hoc test).

Postoperative imaging was conducted at a mean of 6.5 months, with 85.8% of patients undergoing MRI and 14.2% undergoing ultrasound. Among the 937 patients who underwent complete repair without augmentation, retears were observed in 128 patients (13.7%), and anterior L-shaped tears (31.2%) had a significantly higher retear rate than the other tear patterns (9.9%, 15.3%, and 12.9% for crescent, U, and posterior L shaped, respectively; P < .001). The rate of revision surgery because of a symptomatic retear within 2 years after primary surgery was significantly higher for anterior L-shaped tears than for the other tear patterns (3.8%, 7.5%, 21.6%, and 0.0% for crescent, U, and anterior and posterior L shaped, respectively; P = .002).

In the subgroup analysis, 713 patients (68.8%) had small-to-medium tears, and 324 (31.2%) had large-to-massive tears. For small-to-medium tears, the most common tear pattern was crescent-shaped tears with 309 patients (43.3%), followed by posterior L-, U-, and anterior L-shaped tears with 182, 158, and 64 patients, respectively (25.5%, 22.2%, and 9.0%, respectively). Patients with crescent-shaped tears were significantly younger than patients with U- and posterior L-shaped tears. No significant differences were present in other preoperative factors among the 4 groups (Table 3).

Table 3

Characteristics of Patients With Small-to-Medium Tears^a

	Crescent (n = 309)	U (n = 158)	Anterior L (n = 64)	Posterior L (n = 182)	P Value
Age, y	57.4 ± 10.2	60.0 ± 9.4^b	60.5 ± 9.8	59.8 ± 10.1^b	.008
Sex, male:female, n	131:178	64:94	26:38	78:104	.966
Dominant side, %	67.3	73.4	75.0	68.7	.422
Symptom duration, mo	11.5 ± 6.5	11.8 ± 6.7	11.3 ± 5.3	12.3 ± 7.1	.577
Body mass index, kg/m²	25.2 ± 3.0	24.7 ± 3.4	24.7 ± 3.6	25.3 ± 3.0	.261
Heavy labor occupation, %	16.2	17.1	14.1	15.4	.945
Diabetes mellitus, %	12.0	10.8	12.5	11.5	.978
Dyslipidemia, %	53.1	62.7	57.8	59.3	.219
Smoking, %	9.7	7.6	9.4	11.0	.766

Data are reported as mean ± SD unless otherwise indicated.

P < .05 versus crescent (Bonferroni post hoc test).

Fatty infiltration of the supraspinatus and the proportion of supraspinatus muscle atrophy were significantly greater for anterior L-shaped tears than for the other tear patterns (Table 4). The AP tear size was significantly smaller for U-shaped tears than for the other tear patterns, while the ML tear size was significantly smaller for crescent-shaped tears than for the other tear patterns. Partial repair or ACR was performed in a significantly larger proportion of anterior L-shaped tears than in the other tear patterns. Full-thickness subscapularis tears were significantly more frequent with anterior L- and U-shaped tears than with posterior L- or crescent-shaped tears.

Table 4

Preoperative Radiological and Intraoperative Findings in Patients With Small-to-Medium Tears^a

	Crescent (n = 309)	U (n = 158)	Anterior L (n = 64)	Posterior L (n = 182)	P Value
Preoperative radiological
Fatty infiltration
Suprapinatus	0.5 ± 0.6^b	0.5 ± 0.8^b	1.0 ± 0.8	0.5 ± 0.7^b	<.001
Infraspinatus	0.3 ± 0.5	0.3 ± 0.5	0.3 ± 0.5	0.3 ± 0.6	.713
Subscapularis	0.1 ± 0.2	0.1 ± 0.3	0.1 ± 0.5	0.1 ± 0.4	.214
Atrophy of subscapularis, %	1.6^b	5.7^b	20.3	6.0^b	<.001
Intraoperative
Anteroposterior tear size, mm	18.6 ± 5.4^c	15.7 ± 4.5	18.4 ± 3.7^c	19.7 ± 4.8^c	<.001
Mediolateral tear size, mm	12.6 ± 3.7	21.8 ± 5.4^d	21.8 ± 5.1^d	21.0 ± 4.6^d	<.001
Partial repair or ACR, %	1.0^b	2.5^b	21.9	1.1^b	<.001
Delamination, %	52.9	63.0	58.0	58.9	.204
Subscapularis tear, %	18.4^b,c	29.7	35.9	17.0^b,c	<.001

Data are reported as mean ± SD unless otherwise indicated. ACR, anterior cable reconstruction.

P < .05 versus anterior L (Bonferroni post hoc test).

P < .05 versus U (Bonferroni post hoc test).

P < .05 versus crescent (Bonferroni post hoc test).

For large-to-massive tears, the most common tear pattern was crescent-shaped tears with 102 patients (31.5%), followed by posterior L-shaped tears with 88 (27.2%), anterior L-shaped tears with 70 (21.6%), and U-shaped tears with 64 (19.8%). Preoperative factors are summarized in Table 5. No significant differences were present in preoperative characteristics among the 4 groups.

Table 5

Characteristics of Patients With Large-to-Massive Tears^a

	Crescent (n = 102)	U (n = 64)	Anterior L (n = 70)	Posterior L (n = 88)	P Value
Age, y	62.4 ± 9.2	65.3 ± 3.7	63.2 ± 8.2	64.7 ± 9.6	.097
Sex, male:female, n	58:44	34:30	33:37	42:46	.517
Dominant side, %	67.6	70.3	71.4	55.7	.126
Symptom duration, mo	11.4 ± 7.0	12.4 ± 6.0	11.9 ± 6.7	12.7 ± 7.8	.617
Body mass index, kg/m²	24.6 ± 3.4	24.6 ± 3.3	25.0 ± 3.4	25.2 ± 2.7	.494
Heavy labor occupation, %	19.6	18.8	12.9	13.6	.254
Diabetes mellitus, %	11.8	9.4	2.9	11.4	.200
Dyslipidemia, %	52.9	67.2	62.9	55.7	.248
Smoking, %	10.8	10.9	0.0	9.1	.054

Data are reported as mean ± SD unless otherwise indicated.

The proportion of supraspinatus muscle atrophy was significantly greater with anterior L-shaped tears than with posterior L-shaped tears (Table 6). The AP tear size was significantly larger with crescent-shaped tears and significantly smaller with U-shaped tears than with the other tear patterns. The ML tear size was significantly smaller with crescent-shaped tears than with the other tear patterns. Anterior L-shaped tears had a significantly greater proportion of partial repair or ACR than the other tear patterns. Full-thickness subscapularis tears were significantly more frequent with anterior L-shaped tears than with crescent- or posterior L-shaped tears.

Table 6

Preoperative Radiological and Intraoperative Findings in Patients With Large-to-Massive Tears^a

	Crescent (n = 102)	U (n = 64)	Anterior L (n = 70)	Posterior L (n = 88)	P Value
Preoperative radiological
Fatty infiltration
Supraspintus	1.7 ± 0.6	1.8 ± 0.8	1.7 ± 0.7	1.8 ± 0.6	.670
Infraspinatus	1.2 ± 0.9	1.1 ± 0.6	1.3 ± 0.8	1.1 ± 0.8	.271
Subscapularis	0.5 ± 0.9	0.4 ± 0.5	0.4 ± 0.8	0.4 ± 0.8	.745
Atrophy of subscapularis, %	30.4	31.3	41.4	27.8^b	<.001
Intraoperative
Anteroposterior tear size, mm	30.7 ± 3.9	22.5 ± 7.4^c	26.1 ± 5.2^{c , d}	26.5 ± 5.0^{c , d}	<.001
Mediolateral tear size, mm	26.1 ± 4.5	28.9 ± 3.0^c	28.5 ± 7.0^c	28.3 ± 4.6^c	<.001
Partial repair or ACR, %	12.7^b	25.0^b	61.4	5.7^b,d	<.001
Delamination, %	68.6	81.3	64.3	64.8	.113
Subscapularis tear, %	29.4^b	37.5	47.1	25.0^b	.019

Data are reported as mean ± SD unless otherwise indicated. ACR, anterior cable reconstruction.

P < .05 versus anterior L (Bonferroni post hoc test).

P < .05 versus crescent (Bonferroni post hoc test).

P < .05 versus U (Bonferroni post hoc test).

In the patients with large-to-massive tears, no significant differences were present in the retear rate based on the tear pattern (16.9%, 22.9%, 37.0%, and 18.1% for crescent, U, and anterior and posterior L shaped, respectively; P = .127). However, with small-to-medium tears, anterior L-shaped tears had a significantly higher retear rate than the other tear patterns (7.8%, 13.0%, 28.0%, and 10.6% for crescent, U, and anterior and posterior L shaped, respectively; P < .001). In the small-to-medium tear subgroup, logistic regression analysis was performed to evaluate the effect of tear patterns on rotator cuff retears. Anterior L-shaped tear (OR, 2.83 [95% CI, 1.24-6.44]; P = .013), fatty infiltration of the supraspinatus (OR, 2.14 [95% CI, 1.55-2.95]; P < .001), and age (OR, 1.06 [95% CI, 1.03-1.09]; P < .001) were significantly associated with rotator cuff retears. Anterior L-shaped tears had a 1.32-fold higher risk compared with fatty infiltration of the supraspinatus and a 2.67-fold higher risk compared with age.

Discussion

The most important finding of this study was that anterior L-shaped tears had a significantly higher retear rate, despite having the lowest incidence among the tear patterns in small-to-medium tears. Furthermore, anterior L-shaped tears had a higher incidence of retears requiring early revision surgery than the other tear patterns. In large-to-massive tears, the retear rate did not significantly differ among the tear patterns.

Most studies regarding the prevalence of rotator cuff tear patterns have reported that the crescent-shaped tear is the most common, whereas the anterior L-shaped tear is less frequently observed.^16,30,33 However, in our analysis of a large number of patients, the distribution of tear sizes in the cohort affected the distribution of tear patterns. Anterior L-shaped tears were the least frequent tear pattern, whereas crescent-shaped tears were the most frequent among small-to-medium tears. The prevalence of large-to-massive tears was 21.6% in patients with anterior L-shaped tears, and the proportion of crescent-shaped tears was relatively decreased compared with that in small-to-medium tears. Consistent with our results, Watson et al³³ reported that 7.9% of patients with full-thickness rotator cuff tears had anterior L-shaped tears, which was the least frequent tear pattern. The proportion of small-to-medium tears in their cohort was 82.9%. However, a recent clinical study analyzing tear patterns in large-to-massive tears reported a prevalence of 27.5% for anterior L-shaped tears.³⁵ The differences in the frequency of tear patterns according to the tear size can be attributed to the varied directions in which tear propagation and enlargement occur.

In the current study, as the tear size increased, the proportion of crescent- and U-shaped tears decreased, while the proportion of L-shaped tears increased. Therefore, L-shaped tears may develop over time from crescent- or U-shaped tears because of the medial-to-lateral direction of tear propagation at the anterior portion of crescent-shaped tears and the anterior-to-lateral direction of tear propagation at the anterior portion of U-shaped tears. In a study analyzing the stress distribution in rotator cuff tendons based on tear shape, tear propagation was found to vary with the shape of the tears, and stress concentrations were identified as prominent at the anterior and posterior tendon edges and the innermost corner of longitudinal tears.³¹ In addition, small-to-medium crescent-shaped tears have been found in younger patients, which could be attributed to the progressive nature of degenerative rotator cuff disease.¹³ Rotator cuff tears are considered to begin at the supraspinatus tendon, and this portion is considered to transmit the contractile load, causing tear enlargement in anterior and posterior directions.^8,9,11 Therefore, a crescent-shaped tear may be the initial stage of full-thickness rotator cuff tears, and the initially dominant distribution of crescent-shaped tears in small-to-medium tears may transition as the tear progresses in a combined transverse and longitudinal direction, causing various tear shapes with increasing tear sizes, which likely influences the tear prevalence.

The supraspinatus exhibits asymmetrical anatomy between anterior and posterior regions, with an anterior cord region and a posterior strap region.²⁹ The anterior supraspinatus has a fusiform structure with an intramuscular tendinous core, which bears 3 times more stress than the posterior strap region.²⁷ As the core transitions distally, it becomes a cord of the supraspinatus tendon, interdigitating with the anterior rotator cable and attaching to the footprint of the anterior greater tuberosity. Force transmission through both structures is disproportionately high, which demonstrates their primary load-bearing role.^17,27,29 Therefore, a rupture is related to increased fatty infiltration in the rotator cuff muscle, and more severe tendon degeneration can result from impaired muscle force transmission.^1,3,22,23 Restoration failure of the cord of the supraspinatus tendon could lead to stress concentrations and uneven force transmission, resulting in retears and functional impairment. Consistently, in this study, anterior L-shaped tears involving the cord and anterior rotator cable disruption were associated with advanced fatty infiltration and atrophy of the supraspinatus muscle. Poor tendon quality of anterior L-shaped tears increased tendon stiffness and retraction, resulting in a higher rate of partial repair and the need for additional augmentation procedures.

Even with complete repair, anterior L-shaped tears had a significantly higher retear rate than the other tear patterns. Cho et al³ also reported that anterior rotator cable integrity had a significant influence on the retear rate. A rotator cable rupture has been shown to increase gapping, decrease stiffness, and cause abnormal strain magnitudes and distributions within the rotator cuff in a cadaveric model.³⁶ Disruption of the rotator cable may affect shoulder biomechanics and the biological environment for rotator cuff healing. In previous studies evaluating retears using postoperative MRI, ACR reduced rotator cuff healing failure after repair.^7,34 In retracted anterior L-shaped tears, the retear rate without ACR was 39%,⁷ while that with ACR using the long head of the biceps tendon was 18.7%.³⁴ A reconstructed anterior rotator cable carries the preponderance of the load on the repaired rotator cuff,^17,28 and restoring the rotator cable is crucial for its stress-shield capability and tendon-to-bone healing.¹² Therefore, the condition of the anterior rotator cable and the cord of the supraspinatus tendon in the repaired construct should be examined intraoperatively, and additional procedures could be considered to ensure appropriate force transmission and postoperative tendon integrity, especially for anterior L-shaped tears.

Recognizing rotator cuff tear patterns has been emphasized to achieve anatomic reduction, which is crucial for attaining low-tension repair and preventing improper force transmission of a repaired tendon.^{2,4,20,23,26,30} Each rotator cuff tear pattern has a different direction of tendon end mobility and varies in the reduction technique. In the current study, L-shaped tears exhibited a larger AP tear size than U-shaped tears and a larger ML tear size than crescent-shaped tears, which are complex tear patterns containing longitudinal and transverse tear components with large tendon defect areas. Therefore, considering the potential challenges in achieving anatomic reduction with such tear patterns, identifying which margin is more mobile than the other for tension-free reduction may be increasingly valuable in promoting the healing rate.^4,20 Consequently, appropriate repair techniques with anatomic reduction and minimal tension should be applied based on different rotator cuff tear patterns to improve postoperative outcomes.

Retear rates for various rotator cuff tear patterns differed according to the tear size. The retear rate for anterior L-shaped tears in small-to-medium tears was higher than that for the other tear patterns, whereas no significant difference was observed in large-to-massive tears. In small-to-medium tears, anterior L-shaped tears showed advanced fatty infiltration and atrophy of the supraspinatus muscle. Namdari et al²¹ demonstrated that a rupture of the anterior rotator cable was associated with more advanced supraspinatus muscle degeneration in patients with small-to-medium rotator cuff tears. Biomechanical studies have demonstrated that the rotator cable is the primary load-bearing structure within the supraspinatus for force transmission to the proximal humerus.^19,27 Impaired muscle force transmission may have consequences on rapid muscle and tendon degeneration of the supraspinatus in small-to-medium tears, and the difference in postoperative tendon healing was emphasized because of the low retear rate of small-to-medium tears. However, in large-to-massive tears, tendon healing could be influenced by multiple factors because of their chronicity.¹⁵ Various factors, including age, diabetes, hyperlipidemia, smoking, tendon quality, and fatty infiltration of the rotator cuff, have been reported to affect tendon healing independently.¹⁴ The combination of such factors associated with a retear contributed to similar retear rates across different tear patterns, and the healing process is multifactorial.

Limitations

The present study has several limitations. First, a single surgeon classified the rotator cuff tear pattern during arthroscopic surgery; thus, an error in judgment could have been made. However, the tear was classified objectively by its morphology and the optimal technique that could achieve anatomic reduction while minimizing tendon tension. Second, minor complex tear patterns such as flap tears, tears at the musculotendinous junction, and tears that could not be classified into specific tear patterns were excluded from this study. Flap tears or musculotendinous junction tears are specific tear patterns that are significant enough to be studied independently. However, complex tear patterns were not analyzed as a separate group because of their small sample size, which would have increased the complexity of statistical analysis without yielding clinically meaningful results. Third, retears were confirmed using MRI or ultrasound. However, diagnosing retears with MRI is time consuming and costly, while ultrasound has been reported to have comparable accuracy to MRI in diagnosing retears after rotator cuff repair. Fourth, postoperative imaging was conducted at 6 months after surgery, which differs from the timing of final clinical outcomes assessed at 24 months after surgery. Nevertheless, a previous report demonstrated no significant difference in retear rates between the 6- and 24-month time points.¹⁸ Fifth, we did not analyze the difference in pain and shoulder function among the tear patterns. However, we evaluated critical postoperative outcomes such as the retear rate and the number of patients undergoing revision surgery for symptomatic retears according to tear patterns. Finally, we retrospectively reviewed medical records, and there was some loss of data, including missing imaging evaluations, along with an inherent bias issue.

Conclusion

The prevalence of full-thickness rotator cuff tear patterns varied depending on the tear size. In small-to-medium tears, anterior L-shaped tears had a significantly higher retear rate, despite having the lowest incidence among the tear patterns. Furthermore, anterior L-shaped tears had a higher incidence of retears requiring early revision surgery than the other tear patterns. In large-to-massive tears, the retear rate did not significantly differ among the tear patterns. To improve postoperative tendon integrity, it is recommended to consider the tear pattern and the functional anatomy related to the cord and anterior rotator cable.

Footnotes

Acknowledgements

The authors appreciate statistical advice from Professor Hye Ah Lee (Clinical Trial Center, Mokdong Hospital, Ewha Womans University College of Medicine).

Submitted October 21, 2024; accepted November 14, 2024.

One or more of the authors has declared the following potential conflict of interest or source of funding: This research was supported by the Basic Science Research Program through the National Research Foundation of Korea, funded by the Ministry of Education (NRF-2022R1A2C1009316). AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.

ORCID iDs

Sang-Jin Shin

Sanghyeon Lee

References

Burkhart

Esch

Jolson

. The rotator crescent and rotator cable: an anatomic description of the shoulder’s “suspension bridge.” Arthroscopy. 1993;9(6):611-616.

Burkhart

IK.

Arthroscopic rotator cuff repair. J Am Acad Orthop Surg. 2006;14(6):333-346.

Cho

Moon

Hong

Bae

Rhee

YG.

Comparison of clinical and radiological results in the arthroscopic repair of full-thickness rotator cuff tears with and without the anterior attachment of the rotator cable. Am J Sports Med. 2017;45(11):2532-2539.

Davidson

Burkhart

SS.

The geometric classification of rotator cuff tears: a system linking tear pattern to treatment and prognosis. Arthroscopy. 2010;26(3):417-424.

DeOrio

Cofield

RH.

Results of a second attempt at surgical repair of a failed initial rotator-cuff repair. J Bone Joint Surg Am. 1984;66(4):563-567.

Fuchs

Weishaupt

Zanetti

Hodler

Gerber

Fatty degeneration of the muscles of the rotator cuff: assessment by computed tomography versus magnetic resonance imaging. J Shoulder Elbow Surg. 1999;8(6):599-605.

Jeon

Lee

Kim

Shin

SJ.

Does additional biceps augmentation improve rotator cuff healing and clinical outcomes in anterior L-shaped rotator cuff tears? Clinical comparisons with arthroscopic partial repair. Am J Sports Med. 2017;45(13):2982-2988.

Jeong

Min

Park

Han

Yoo

JC.

Location of rotator cuff tear initiation: a magnetic resonance imaging study of 191 shoulders. Am J Sports Med. 2018;46(3):649-655.

Keener

Galatz

Teefey

, et al. A prospective evaluation of survivorship of asymptomatic degenerative rotator cuff tears. J Bone Joint Surg Am. 2015;97(2):89-98.

10.

Keener

Hsu

Steger-May

Teefey

Chamberlain

Yamaguchi

Patterns of tear progression for asymptomatic degenerative rotator cuff tears. J Shoulder Elbow Surg. 2015;24(12):1845-1851.

11.

Kim

Dahiya

Teefey

, et al. Location and initiation of degenerative rotator cuff tears: an analysis of three hundred and sixty shoulders. J Bone Joint Surg Am. 2010;92(5):1088-1096.

12.

Kim

Shin

SJ.

No difference in clinical outcomes following repair of large retracted anterior rotator cuff tears using patch augmentation with human dermal allograft versus anterior cable reconstruction with biceps tendon autograft. Arthroscopy. 2024;40(2):294-302.

13.

Kim

Bae

Lee

Jee

Park

CK.

Tear progression of symptomatic full-thickness and partial-thickness rotator cuff tears as measured by repeated MRI. Knee Surg Sports Traumatol Arthrosc. 2017;25(7):2073-2080.

14.

Kwon

Kim

Lee

Kim

JH.

The rotator cuff healing index: a new scoring system to predict rotator cuff healing after surgical repair. Am J Sports Med. 2019;47(1):173-180.

15.

Lam

Murrell

GA.

Factors predicting rotator cuff retears: an analysis of 1000 consecutive rotator cuff repairs. Am J Sports Med. 2014;42(5):1134-1142.

16.

Lee

Kim

Choi

Kim

SH.

Characteristics and outcomes of L-shaped and reverse L-shaped rotator cuff tears. Bone Joint J. 2022;104(3):394-400.

17.

Mesiha

Derwin

Sibole

Erdemir

McCarron

JA.

The biomechanical relevance of anterior rotator cuff cable tears in a cadaveric shoulder model. J Bone Joint Surg Am. 2013;95(20):1817-1824.

18.

Miller

Downie

Kohen

, et al. When do rotator cuff repairs fail? Serial ultrasound examination after arthroscopic repair of large and massive rotator cuff tears. Am J Sports Med. 2011;39(10):2064-2070.

19.

Miller

Fujimaki

Araki

Musahl

Debski

RE.

Strain distribution due to propagation of tears in the anterior supraspinatus tendon. J Orthop Res. 2014;32(10):1283-1289.

20.

Millett

Warth

RJ.

Posterosuperior rotator cuff tears: classification, pattern recognition, and treatment. J Am Acad Orthop Surg. 2014;22(8):521-534.

21.

Namdari

Donegan

Dahiya

Galatz

Yamaguchi

Keener

JD.

Characteristics of small to medium-sized rotator cuff tears with and without disruption of the anterior supraspinatus tendon. J Shoulder Elbow Surg. 2014;23(1):20-27.

22.

Nguyen

Quigley

Galle

, et al. Margin convergence anchorage to bone for reconstruction of the anterior attachment of the rotator cable. Arthroscopy. 2012;28(9):1237-1245.

23.

Nho

Ghodadra

Provencher

Reiff

Romeo

AA.

Anatomic reduction and next-generation fixation constructs for arthroscopic repair of crescent, L-shaped, and U-shaped rotator cuff tears. Arthroscopy. 2009;25(5):553-559.

24.

Park

Kang

Lee

Shin

SJ.

A novel reparability assessment scoring system for full-thickness rotator cuff tears. Orthop J Sports Med. 2020;8(8):2325967120940979.

25.

Park

J-Y

Lee

J-H

K-S

, et al. Rotator cuff retear after repair surgery: comparison between experienced and inexperienced surgeons. Clin Shoulder Elb. 2021;24(3):135-140.

26.

Park

Jung

Jeon

Cho

Choi

KS.

Arthroscopic repair of large U-shaped rotator cuff tears without margin convergence versus repair of crescent- or L-shaped tears. Am J Sports Med. 2014;42(1):103-111.

27.

Park

Detoc

Lee

TQ.

Anterior cable reconstruction: prioritize rotator cable and tendon cord when considering superior capsular reconstruction. Arthroscopy. 2022;38(5):1705-1713.

28.

Park

Hung

DeGiacomo

McGarry

Adamson

Lee

TQ.

Anterior cable reconstruction of the superior capsule using semitendinosus allograft for large rotator cuff defects limits superior migration and subacromial contact without inhibiting range of motion: a biomechanical analysis. Arthroscopy. 2021;37(5):1400-1410.

29.

Roh

Wang

April

Pollock

Bigliani

Flatow

. Anterior and posterior musculotendinous anatomy of the supraspinatus. J Shoulder Elbow Surg. 2000;9(5):436-440.

30.

Sallay

Hunker

Lim

JK.

Frequency of various tear patterns in full-thickness tears of the rotator cuff. Arthroscopy. 2007;23(10):1052-1059.

31.

Sano

Hatta

Yamamoto

Itoi

Stress distribution within rotator cuff tendons with a crescent-shaped and an L-shaped tear. Am J Sports Med. 2013;41(10):2262-2269.

32.

Shin

Kim

Park

Lee

Kim

MS.

Anterior cable reconstruction using the biceps tendon in retracted anterior L-shaped rotator cuff tears. Arthrosc Tech. 2021;10(1):e55-e60.

33.

Watson

Allen

Robbins

Bedi

Gagnier

Miller

Does the rotator cuff tear pattern influence clinical outcomes after surgical repair?

Orthop J Sports Med. 2018;6(3):2325967118763107.

34.

Yang

Lee

Shin

SJ.

Satisfactory functional and structural outcomes of anterior cable reconstruction using the proximal biceps tendon for large retracted rotator cuff tears. Knee Surg Sports Traumatol Arthrosc. 2023;31(5):1910-1918.

35.

Yeo

MHX

Seah

SJS

Lee

Lie

DTT

. Different types of rotator cuff tear morphology do not affect post-repair clinical outcomes in large to massive tears. J ISAKOS. 2024;9(1):25-33.

36.

Yuri

Prado

Hoshikawa

Trevino

3rd Giambini

Tension distribution in articular surfaces of the rotator cable and crescent: a cadaveric study. J Bone Joint Surg Am. 2024;106(12):1100-1107.