Treatment of Large and Massive Rotator Cuff Tears: Does Infraspinatus Muscle Tear Affect Repair Integrity?

Article information

Clin Shoulder Elb. 2019;22(4):203-209

Publication date (electronic) : 2019 December 1

doi : https://doi.org/10.5397/cise.2019.22.4.203

Sungwook Choi, Hyunchul Yang, Hyunseong Kang, Gyeong Min Kim

Department of Orthopedic Surgery, Jeju National University School of Medicine, Jeju, Korea

Correspondence to: Hyunseong Kang Department of Orthopedic Surgery, Jeju National University Hospital, Jeju National University School of Medicine, 15 Aran 13-gil, Jeju 63241, Korea Tel: +82-64-717-1690, Fax: +82-64-717-1131, E-mail: oskanghs@gmail.com

Received 2019 September 6; Revised 2019 November 2; Accepted 2019 November 3.

Abstract

Introduction

Over the years, there have been numerous advances in surgical techniques of arthroscopic rotator cuff repair. However, the management of patients with large to massive rotator cuff tears remains a challenge for orthopedic surgeons. In the past, Goutallier et al. [1] devised a global fatty degeneration index (GFDI) and reported that a patient with grade 3 (as much muscle as fat) and grade 4 (less muscle than fat) fatty degeneration shows no improvement after rotator cuff repair. However, Burkhart et al. [2] reported that arthroscopic rotator cuff repair in patients with grade 3 or 4 fatty degeneration provides satisfactory functional improvement. As seen in previous reports, clinical outcomes and prognosis of large and massive rotator cuff tear are known to be unpredictable due to not only degeneration of rotator cuff, but also the high rate of retear.

Various factors influence the repair integrity, including patient age, tear size, and tendon quality [3]. In particular, higher preoperative fatty degeneration and larger tear sizes have a tendency to increase the retear rate [4]. It is critical to note that massive and irreparable rotator cuff tears should be considered as separate entities [5]. Massive tears are described as tears larger than 5 cm, and tears involving two or more tendons [6,7].

This study was undertaken to evaluate the clinical outcome and maintenance of the repair integrity or retear rate of arthroscopic rotator cuff repair in patients with large and massive sized rotator cuff tears. Furthermore, we investigated the retear patterns in cases with structural failure after rotator cuff repair, with use of follow-up magnetic resonance imaging (MRI, Achieva 3.0T; Philips, Amsterdam, the Netherlands).

Methods

Patient Assessment

The average patient age at the time of surgery was 65 years (range, 47–78 years), and postoperative follow-up duration was 36.1 months (range, 12–110 months). Patients enrolled included 28 male and 53 female, with the repairs being performed on 67 right and 14 left shoulders. Cuff tears in the dominant arm were seen in 73 cases, and in the nondominant arm in 8 cases. All patients had multiple tendon involvement.

History of trauma was reported by 17 patients; 49 patients were heavy workers, including farmers, carriers, and car mechanics, and 32 patients were required relatively low physical demands, such as housewives and office workers.

Arthroscopic suture bridge technique was applied in patients undergoing complete repair, whereas maximum partial repairs were performed in the remaining cases.

All patients were evaluated by the Constant score and the Shoulder Rating Scale of the University of California, Los Angeles (UCLA). Evaluations tear size, repair integrity, and fatty degeneration were performed preoperatively and at the last follow-up examination [8,9].

The initial size of rotator cuff tear was measured intraoperatively, based on the greatest dimension of tendon tear, and classified as follows: small, <1 cm; medium, 1 to 3 cm; large, 3 to 5 cm; or massive, >5 cm. Repair integrity of cuff muscles was evaluated with the classification devised by Sugaya et al. [10], through follow-up MRI: type I, sufficient thickness with homogeneous low intensity; type II, sufficient thickness with partial high intensity; type III, insufficient thickness without discontinuity; type IV, presence of a minor discontinuity; type V, presence of a major discontinuity. Types IV and V were defined as retears. Fatty degeneration of the rotator cuff was assessed by preoperative MRI using the GFDI proposed by Goutallier et al. [1]

The Collin’s classification [11] was employed to structurally analyze the rotator cuff tear patterns. Collin et al. [11] classified rotator cuff tear patterns into five components: type A, supraspinatus and superior subscapularis tears; type B, supraspinatus and entire subscapularis tears; type C, supraspinatus, superior subscapularis, and infraspinatus tears; type D, supraspinatus and infraspinatus tears; and type E, supraspinatus, infraspinatus, and teres minor tears (Fig. 1). None of the cases presented with entire tear subscapularis muscle and teres minor muscle tear on the preoperative MRIs; hence, types B and E were excluded. This study therefore included patients classified as type A, C, and D (Table 1).

Fig. 1.

(A) Five muscles that constitute the rotator cuff. (B) Collin’s classification [11] classifies rotator cuff tear into five groups: type A, supraspinatus and superior subscapularis tears; type B, supraspinatus and entire subscapularis tears; type C, supraspinatus, superior subscapularis, and infraspinatus tears; type D, supraspinatus and infraspinatus tears; and type E, supraspinatus, infraspinatus, and teres minor tears.

Table 1.

Patient and Tear Demographics

Surgical Technique

All surgical procedures were performed under general anesthesia, by a single surgeon. The position was not a complete lateral direction, but a semi-lateral position tilted 45 degrees, such that the patient’s front face was diagonal. A diagnostic arthroscopic procedure was performed through posterior and anterior portals. The glenohumeral joint was evaluated for any arthritic changes and other accompanying intra-articular lesions. During surgery, the tear size, pattern, tendon quality, and presence of delamination were identified. Acromioplasty was completed for all patients. After evaluating mobility of the torn cuff, complete repair, wherever possible, was performed using the double-row suture technique.

In the double-row suture technique, medial row anchors (3.7 mm and 4.5 mm Bio-Corkscrew suture anchor; Arthrex, Naples, FL, USA) were first placed in a location just lateral to the articular surface of the humeral head, considered as the medial edge of the footprint. Next, either a Scorpion (Arthrex) or suture hook (Linvatec, Largo, FL, USA) was used to pass the suture through the supraspinatus tendon, near the musculotendinous junction. The knots were tied on the medial row, and the tendon was reduced into the bone. Lateral row anchors (3.5 mm and 4.5 mm Bio-PushLock anchor or SwiveLock; Arthrex) were placed 5 to 10 mm lateral to the greater tuberosity after preparing the bone sockets. The number of anchors and selection of anchor size depended on the tear size, bone quality of the footprint, and anchor stability. The delamination tendon was repaired after passing the thread through the suture hook as much as possible.

In case of incomplete coverage of the foot print, margin convergence sutures were applied first and the torn tendon was pulled to cover the foot print to the maximum possible; suture anchors were subsequently placed to the bone by the single-row fixation technique. Biceps tenotomy or tenodesis was performed in patients with severe degeneration of the biceps tendon. Biceps tenotomy was performed for patients older than age 65 years, whereas tenodesis was performed for patients younger than age 65 years. Patients with biceps partial tear underwent only debridement (Fig. 2).

Fig. 2.

(A) Preoperative coronal T2-weighted image showing a full-thickness rotator cuff tear with substantial muscle retraction. (B) The arthroscopic view showing massive rotator cuff tear. Degeneration of rotator cuff is severe and glenoid is exposed. (C) Suture anchors were placed to repair the rotator cuff on the footprint. (D) A repair configuration after arthroscopic rotator cuff repair. Due to the extensive rupture of the rotator cuff, the footprint was not completely covered.

Results

All functional scores showed significant improvement over the preoperative scores at the most recent follow-up. The mean UCLA score increased from a preoperative mean of 15.17 ± 4.09 points to 30.26 ± 3.50 points (p<0.001), and the Constant score improved from a preoperative mean of 54.68 ± 12.48 points to 82.49 ± 10.32 points (p<0.001) at the last follow-up (Table 2).

Table 2.

Clinical and Radiologic Outcomes

Repair integrity evaluation by the Sugaya classification with MRI revealed type I healing in 6 cases (7.4%), type II in 30 cases (37.0%), type III in 12 cases (14.8%), type IV in 8 cases (9.9%), and type V in 25 cases (30.9%). The overall number of retears (types IV and V) was 33 cases (40.7%). The retear rate was 22.9% (11 of 48) in large tears, and 66.7% (22 of 33) in massive tears. A larger intraoperative tear size was associated with a statistically significant higher rate of retear (p<0.001; Fig. 3).

Fig. 3.

Correlation between preoperative tear size and repair integrity (p<0.001).

The average follow-up period of patients with maintained repair integrity was 36.2 months, and for patients with retear was 36.1 months, which were statistically not different (p <0.988).

Preoperative GFDI was confirmed by MRI: 2 cases were determined to be grade 1 and had 1 retear; 29 cases showed grade 2 with 3 retears (10.3%). A retear was found in 14 of 32 grade 3 cases (43.8%), and 15 retears were observed in 18 grade 4 cases (83.3%). Higher preoperative fatty degeneration grades were associated with increased incidence of retear (p<0.001; Fig. 4).

Fig. 4.

Correlation between global fatty degeneration index (GFDI) and repair integrity (p<0.001).

Collin’s classification was also confirmed by MRI; 15 cases were determined as type A with 2 retears (13.3%), 26 type C cases had 17 retears (65.4%), and 40 type D cases had 14 retears (35.0%). To evaluate whether infraspinatus tear affected repair integrity, we compared two groups, type A with type C and D. Although there were few type A cases, a statistically significant difference was obtained in the repair integrity between the two groups (p=0.036; Fig. 5). For each comparative analysis using ANOVA test, type C had more retears than A (p<0.001) and D (p<0.03). However, there was no statistically significant difference between types A and D, although D tended to retear more than A (p=0.28; Table 3).

Fig. 5.

Correlation between Collin’s classification and repair integrity (p=0.036).

Table 3.

Multiple Comparisons by Collin’s Classification

Discussion

The goal of arthroscopic rotator cuff repair is to maintain the mechanical strength to avoid retears due to repeated stress in daily life. Therefore, it is important to maintain the integrity of the repaired rotator cuff until complete healing is achieved. Arthroscopic rotator cuff repair results in significant improvement of shoulder pain and function [4,12,13]. It is equally critical to plan repair strategies before surgery, and MRI is the most useful tool for assessing severity of the tear preoperatively. Yoo et al. [14] identified that sagittal tear size of >32 mm and coronal tear size of >31 mm on MRI are associated with the inability to obtain an anatomical repair. As mentioned previously, severe fatty degeneration correlates with unsatisfactory surgical outcomes and higher retear rate [1,15]. However, with advances in surgical techniques, numerous researches report a significant improvement in the clinical outcomes of patients with large to massive rotator cuff tears [2,16-18]. Since most large to massive rotator cuff tears encompass the supraspinatus muscle, the fatty degeneration of supraspinatus muscle is associated with the surgical outcome.

Park et al. [12] reported that repair of large to massive rotator cuff tears using the double-row technique is superior to the single-row method. We previously reported that retear rates are higher in cases of larger intraoperative tear size and higher preoperative grade of fatty degeneration [4]. We obtained similar results in the current study for patients with large to massive rotator cuff tears.

Although the repair integrity was not maintained, patients were satisfied with surgical outcomes and displayed improved functional scores. These results are similar to the report of Burkhart et al. [2], but the previous study lacked analysis of the postoperative tendon status. Paxton et al. [19] reported that clinical improvement and pain relief after arthroscopic rotator cuff repair of large and massive tears are durable at the time of longterm follow-up (10 years), despite early retears. These results remained unchanged in spite of radiographic progression of arthropathy.

Furthermore, Cho et al. [20,21] reported that medial cuff failure is a frequent retear pattern and mentioned various causes of retears, including poor-quality tendon tissue and suture breakage. Similarly, retears at the musculocutaneous junction were commonly observed in our study.

Recently, Miller et al. [22] proposed that better characterization of timing the structural failure of rotator cuff repairs would help identify weaknesses in repair strategies. He reported that early retears were due to mechanical failure, intraoperative fixation deficiencies, poor compliance with postoperative immobilization, and excessive tension at the repair site; conversely, late retears were due to biologic failure, alterations in the initial biologic healing environment, and medical comorbidities. In our study, it was difficult to identify actual timing of the retear of rotator cuff retear. Further researches to identify timing of retears would help plan better repair strategies.

Park et al. [23] reported that grade II and higher infraspinatus fatty degeneration correlated with a higher failure rate in small and medium tear patients. We used the Collin’s classification to classify the patient group to determine whether torn infraspinatus muscles affect the outcome of surgery. It is more likely to involve multiple tendons rather than just one tendon in a large to massive rotator cuff tear. Therefore, the authors figured that comparison among a group of affected tendons was clinically more significant than identification of outcomes of a single tendon tear.

In our study, types C and D involving infraspinatus muscles had a higher retear rate than type A (comprising supraspinatus muscle and subscapularis muscle). However, when type A, C, and D were compared using ANOVA, type C had a higher retear rate than A and D, but there was no significant difference between A and D. All three muscles are involved in type C, which might affect the high retear rate. Although type D had a higher retear rate than type A, there was no statistically significant difference.

There were several limitations to this study. First, the experimental design was a retrospective review with no control group, and was not a randomized controlled blind trial. Second, postoperative repair integrity evaluation by MRI was not conducted in all cases, which may have resulted in a subtle selection bias. Third, the surgery was not performed only with the double-row suture technique, but also with single row suture technique, marginal convergence, biceps tenodesis or tenotomy, depending on the tear size and quality. It is possible that these differences in surgical methods may have influenced the outcomes. Fourth, the follow-up periods in some patients were relatively short. Lastly, multivariate logistic regression analysis needs to be considered for identifying other factors in addition to the involved tendon; however, this study has a small number of patients.

Conclusion

The retear rate was 40.7% for patients with large and massive tear size, but the clinical symptoms of the patients with or without retear showed good results. Arthroscopic rotator cuff repair yields improved clinical outcomes and a relatively high degree of patient satisfaction, despite difficulty in maintaining the repair integrity. The degree of fatty degeneration and higher number of involved tendons influences the high retear rate. There was no difference in the effect of subscapularis muscle or infraspinatus muscle involvement on the retear rate.

Article information Continued