Biomechanical analysis of single-row anchor, double-row anchor, and single-row knotless tape repair techniques for complete subscapularis tears

Abstract

Background

Repair techniques for complete subscapularis (SSC) tendon tears have not been evaluated and compared in detail. The purpose of this study was to evaluate the biomechanical performance of different repair techniques for complete SSC tendon tears.

Purpose

To evaluate the biomechanical performance of different repair techniques for complete SSC tendon tears.

Methods

In a total of eighteen fresh-frozen shoulders with an intact rotator cuff, complete SSC tendon tears were created. Six shoulders matched by age and gender were allocated to each of the three repair technique groups: (1) knotted mattress single-row repair (group SR), (2) knotted mattress double-row repair (group DR), and (3) knotless tape single-row repair (group SRT). The quality of reconstruction was evaluated by measuring the SSC tendon footprint coverage (% of pre-operative footprint), repair stiffness (N/mm), cyclic displacement (mm), and the ultimate load to failure (N).

Results

DR repair showed a significantly higher relative repair area compared to SR repair techniques (DR = 73.6% vs. SR = 45.6%, P = .009 vs. SRT = 49.4%, P = .006). Stiffness and cyclic displacement were comparable among all three groups (P = .292; P = .215). Ultimate failure load was greatest in the DR group at 305 N ±12 and showed a trend toward higher ultimate failure load compared to the SRT group (170 N ±42; P = .061). There was no significant difference between DR and SR groups (246 N ±71.4; P = .883).

Conclusion

DR repair technique for complete SSC tendon tears achieves a significantly greater footprint coverage and the highest load to ultimate failure of the repair compared to SR repair techniques.

The incidence of subscapularis (SSC) tendon tears is reported to be 30% in arthroscopically treated shoulder patients and makes up approximately 50% of all rotator cuff reconstructions.^5,8,33 As the strongest muscle of the rotator cuff, it plays an important role in ensuring the physiological function of the shoulder as the counterpart to the external rotators.³¹ While most SSC tendon tears involve the upper part of the muscle, the lower part may remain intact. However, despite their rarity, complete SSC tendon tears can occur and lead to severe disability.^{8,23,29,42,55,58} Isolated SSC tendon ruptures account for 20% of all SSC tendon ruptures and occur in 70% of cases due to trauma, particularly forced abduction and external rotation of the arm.¹³ Additional lesions of the long biceps tendon and involvement of the supraspinatus tendon are also common.^48,58

Repair of SSC tendon tears is an effective treatment with substantial and durable improvement in shoulder function.^2,20,42 In comparison to the remaining rotator muscles, the SSC tendon has the broadest footprint and is the most powerful muscle, contributing around 50% of the cuff moment.^12,31 The first arthroscopic repair was described by Burkhart et al^9,54 in 2002 and has since become widely accepted due to the advantages of arthroscopic treatment. In addition to improving shoulder function and reducing pain, the surgical treatment of SSC tendon ruptures aims to prevent retraction and fatty degeneration, thereby reducing the development of irreparable ruptures.³⁵

Single-row (SR) and double-row (DR) repairs have been considered the classic treatment techniques and therefore the biomechanical standard.^14,19,49,57 More recently, a knotless anchor technique using tapes instead of classic sutures (knotless tape repair) has been introduced.⁷ In biomechanical studies, this technique has shown comparable results to conventional methods in the upper third lesions of the SSC, with the advantages of time efficiency, simpler technical implementation, less tissue irritation due to lack of knots, and greater stability of the tapes compared to simple sutures.^6,7 In addition, potential positive aspects of tape repairs of the healing phase were discussed.⁴¹ However, there is a lack of biomechanical studies that directly compare different repair techniques in complete SSC tendon tears.

The aim of this study was therefore to analyze and compare the biomechanical stability of different refixation techniques of complete subscapularis tendon ruptures. The hypothesis of our study was that the DR repair technique for complete SSC tendon tears would have significantly better mechanical properties than SR repair techniques.

Methods

Ethical approval was obtained for this cadaveric study by the local ethics committee and the institutional review board of the Balgrist University Hospital (Zürich, Switzerland).

Specimen preparation and setup

Eighteen fresh-frozen shoulders from ScienceCare (Phoenix, AZ, USA) were used in this study. The mean age of the specimens was 70.2 ± 4.1 years, and the specimens consisted of 13 female and 5 male shoulders. The age distribution in all groups (P = .994) and the gender distribution (P = .999) were equal with four female and two male specimens in two groups and five female and one male in 1 group. The specimens were assigned to three different groups: knotted mattress suture: single row (group SR); knotted mattress suture with an additional lateral, knotless anchor; double row group DR; and knotless tape repair with two anchors: single row (group SRT). There was no significant difference in age (group mattress [M]: 70.2 ± 4 years; group DR: 70.2 ± 5.2 years; group SRT 70.2 ± 4 years; P = .994) among all three groups. All specimens were inspected visually and by computed tomography scan for soft tissue or bone defects, or signs of moderate to advanced glenohumeral arthritis, and deemed to have no signs of glenohumeral arthritis or macroscopic rotator cuff tears. Further computed tomography scan evaluation included the bone mineral density (Hounsfield units (HU)) assessment in a 1 × 1 cm area of the posterior glenoid neck.¹⁸ There was no significant difference in mean bone density between the three groups (group SR: 25.3 ± 6.2 HU; group DR: 28.2 ± 10.2 HU; group SRT: 28.8 ± 8.6 HU; P = .76). All specimens were thawed for 24 hours at room temperature. All shoulders were dissected and cleaned of all soft tissue, leaving the rotator cuff muscles and tendons intact. Rotator cuff muscles were then elevated from the scapula. The glenohumeral joint was then disarticulated and the capsular tissue removed, leaving the proximal humerus with an intact SSC tendon inserting at the lesser tuberosity. The humeral shafts were osteotomized 15 cm distal to the inferior surface of the humeral head. The subscapularis muscle–tendon junction was subsequently reinforced with suture material to prevent the muscle from tearing during subsequent testing. Complete tears of the SSC tendon were created with a standard No. 15 blade. All SSC tendon repairs were performed with arthroscopic tools, but in an open setting. The native and repaired SSC tendon footprint areas were measured using a MicroScribe (Immersion, San Jose, CA, USA) by tracing the outline of the tendinous attachment. Tracing was performed using the digitizer plug-in included in the Rhino 6 Software (Robert McNeel & Associates, Seattle, WA, USA). The relative repair area was calculated as a percentage of the total area of the bony insertion. Specimens were kept moist with phosphate-buffered saline to prevent dehydration during specimen preparation, surgical repair, and testing.

Subscapularis repair

The SSC tendon was then repaired using three different techniques:

• (1)Knotted mattress suture: group SR
• (2)Knotted mattress suture with an additional lateral, knotless anchor: group DR
• (3)Knotless tape single-row repair with two anchors: group SRT

Group SR: knotted mattress suture, single row

Two 5.5-mm double-loaded anchors (Stryker Titanium 5.5 mm; Stryker Endoscopy, San Jose, CA, USA) were used to repair the defect (Fig. 1, A). The first anchor was placed in the superior third and the second anchor was placed in the inferior third of the lesser tuberosity approximately 5-10 mm medial to the bicipital groove. Each suture was passed separately through the tendon using an arthroscopic suture lasso. The superior sutures were fixed with a simple mattress configuration at the bottom and a modified lasso loop mattress stitch at the top to improve tissue grip as previously described by Lafosse et al.³⁴ The inferior sutures were fixed with two simple mattress stitches. Knots were tied in an arthroscopic fashion with alternating half-hitches with the help of a knot pusher.

Figure 1.

Illustration of the glenohumeral joint with the humeral head, the scapula, and the subscapularis repair. (A) Knotted mattress suture; group SR. (B) Knotted mattress suture with lateral, knotless anchor, group DR. (C) Knotless tape repair; single row (group SRT). Anchors illustrated as spiked circles and sutures in red, blue, and green lines. Created with BioRender.com. SR, single-row; DR, double-row.

Group DR: knotted mattress suture with an additional lateral, knotless anchor, double row

Two 5.5-mm double-loaded anchors (Stryker Titanium 5.5 mm; Stryker Endoscopy) were used to repair the SSC tendon tear as a medial row (Fig. 1, B). The repair was carried out in the same way as described previously for the SR repair. Both suture limbs from each medial anchor were then placed into the eyelet of a 4.5-mm lateral knotless anchor (4.5 -mm polyether ether ketone (PEEK) anchor ReelX anchors; Stryker, Kalamazoo, MI, USA) and inserted into the superior to middle zone of the bicipital groove (Fig. 1, B).

Group SRT: knotless tape repair with 2 anchors, single row

Two FiberTapes (Arthrex, Naples, Florida, USA) were used to repair the defect (Fig. 1, C). The superior FiberTape (Arthrex, Naples FL, USA) was shuttled through the upper border of the SSC tendon using an arthroscopic suture lasso as in a previously published figure-of-eight fashion.¹⁶ The inferior FiberTape was shuttled through the inferior third of the SSC tendon in an inverted mattress fashion using an arthroscopic suture lasso. Both limbs of the superior and inferior FibreTape were passed separately through the eyelet of two 4.5-mm knotless anchors (PEEK ReelX anchors; Stryker, Kalamazoo, MI, USA). The free tape limbs were tightened before fixation of the knotless anchors. The knotless anchors were then inserted about 5 mm medial to the entrance of the bicipital groove.

Biomechanical testing setup

The proximal humerus was fixed in a custom-made jig, which was adjustable through an X-Y table. The SSC musculotendinous junction tendon was reinforced with #2 Fiberwires (Arthrex), wrapped in gauze, and attached to a uniaxial material testing machine (Zwick-Roell) equipped with a 20-kN load cell. Similar test setups were previously established in a series of studies by Barber et al, Ma et al, and Hapa et al and represent the worst-case scenario for the pullout strength of suture anchors.^1,24,38,39 The humeral head was oriented to load along the course of the SSC tendon, according to previous studies with 0° abduction and 0° rotation.⁷ The force (N) and displacement (mm) data were recorded using the dedicated software (TestXpert v. 10.11; Zwick-Roell, Ulm, Germany). After specimen fixation, a pressure measurement was taken with the tendon in an unloaded state (0 N). Preconditioning was then performed (from 10 to 100 N for 10 cycles, at a speed of 1 mm/s). For the cyclic test, a preload of 10 N was applied, followed by 300 cycles of 10-100 N at a speed of 2 mm/s. Subsequently, a force-controlled ramped protocol was used for pressure measurements, in accordance with a previous biomechanical study.¹⁵ Measurements of pressure between the repaired SSC tendon and the bony insertion were performed with Tekscan pressure measurement system (sensor no. 4205; Tekscan Inc., South Boston, MA, USA). First, a preload of 10 N was applied, followed by a stepwise increase to 30 N at 2 mm/s. After unloading, the same procedure was repeated with a gradual increase to 50 N. For both conditions, 30 N and 50 N, the load was maintained for 30 s to allow for pressure recordings and was followed by unloading. These loading conditions should simulate post-operative physiological loads. Finally, the specimens were loaded at a speed of 0.5 mm/s until failure. Stress strain data and load were recorded for each repair (Fig. 2).

Figure 2.

Experimental setup for testing the knotted mattress repair with lateral knotless anchor (DR). DR, double-row; SSC, subscapularis; H, humerus; ISP, infraspinatus; SSP, supraspinatus.

Statistical analysis

Normal distribution was assessed with the Shapiro–Wilk test. Categorical variables were analyzed between groups using the chi-square test. Comparison of data of all groups was carried out using analysis of variance (parametric data) and the Kruskal–Wallis test (nonparametric data). The Bonferroni correction (analysis of variance) and the Dunn–Bonferroni correction (Kruskal–Wallis) were used to set the significance at P < .05, and all P values were 2-tailed.

Results

The mean size of the total SSC tendon footprint at the lesser tuberosity was 590.7 ± 96.4 mm², and no significant differences were found between the groups (P = .860). In one female shoulder (69 years) of the SR group, an anchor pullout occurred in the loading cycles, and therefore, the specimen was excluded from further analysis.

The largest repair area was found in the DR repair (DR group) with 429 ± 82.7 mm², which was 73.6% ± 6.3 of the native SSC tendon footprint. There was a significantly larger repair area of the DR compared to the SR and SRT groups (284 mm² P = .008; 286 mm² P = .006). However, there was no significant difference between the SR group and the SRT group (P = .999). Relative repair area (%) was significantly smaller in the SR group compared to the DR group and in the SRT group compared to the DR group (SR vs. DR, P = .003; DR vs. SRT P = .009). There was no significant difference between the SR group and the SRT group (P = .679).

The maximum cyclic displacement was recorded in group SRT with 11 ± 6.7 mm, which was not significantly different from group SR (7.6 ± 1.2 mm) and group DR (7.3 ± 0.9 mm) (P = .781). Stiffness was not significantly different between group SRT 17.8 ± 1.9 N/mm, group DR 17.4 ± 2.6 N/mm, and group SR 15.6 ± 2.3 N/mm (P = .292). Ultimate failure load (N) in the DR group with 305 ± 12 N was not significantly different compared to the SR group (246 ± 71.4 N; P = .883) but showed a trend toward a higher ultimate failure load compared to the SRT group (170 ± 41.7 N; P = .061) (Table I).

Table I.

Repair area and biomechanical testing of single-row anchor mattress suture, double-row anchor mattress, and single-row knotless tape repair.

Variable∗,†	Group SR	Group DR	Group SRT	P value‡
Age (yr)	70.4 ± 4.5	70.2 ± 5.2	70.2 ± 4	.95
Footprint repair area (mm2)	284 ± 53.4	429 ± 82.7	286 ± 54.3	.003
Footprint repair area (%)	45.6 ± 9.3	73.6 ± 6.3	49.4 ± 10.8	.005
Maximum cyclical displacement (mm)	7.6 ± 1.2	7.3 ± 0.9	11 ± 6.7	.781
Maximum load to failure (N)	246 ± 71.4	305 ± 12	170 ± 41.7	.061
Stiffness (N/mm)	15.6 ± 2.3	17.4 ± 2.6	17.8 ± 1.9	.292

SR, single-row; DR, double-row; SRT, knotless tape repair; H, humerus; ISP, infraspinatus; SSP, supraspinatus.

Values in boldface indicate significance at a level of P < .05.

^∗Data shown as mean value ± standard deviation.

^†One shoulder was excluded from analysis because of anchor pullout in the loading cycles.

^‡The groups were compared using analysis of variance (parametric data).

The most common failure mode was anchor pullout (16/18; 89%), followed by suture cut-through at the suture–tendon interface (2/18; 11%). No failures at the musculotendinous junction or fracture of the lesser tuberosity were detected.

Detailed distribution of failure modes is listed in Table II. Due to technical problems, the measurements of the pressure between the repaired SSC tendon and the bony insertion were not recorded correctly, showing inconsistent values during testing. Therefore, they were excluded from further analysis.

Table II.

Failure mechanism.

Group SR	Group DR	Group SRT
Anchor pullout (n = 6)	Anchor pullout (n = 5)	Anchor pullout (n = 5)
	Suture cut-through (n = 1)	Suture cut-through (n = 1)

SR, single-row; DR, double-row; SRT, knotless tape repair.

Discussion

The main finding of the study is that the DR repair technique for SSC tendon tears results in a significantly larger repair area of the original footprint compared to SR repair techniques. Secondarily, the greatest load to failure and lowest cyclic displacement were observed in the DR technique but did not reach statistical significance when compared to the SR repair technique.

Suture anchor repair of the SSC tendon is an established treatment option for SSC tendon rupture with good to excellent clinical results.^25,33,51 There are several techniques for repairing the SSC tendon, such as simple SR and DR techniques, but there is no clear consensus on the DR vs. SR repairs.¹¹ A meta-analysis by Sobhy et al⁵³ reported the superiority of DR over SR rotator cuff repair techniques in terms of functional outcomes and structural cuff integrity, especially at long-term follow-up. However, the included studies were heterogeneous, with isolated SSC tendon repairs in a minority of all cases. Another meta-analysis by Millet et al⁴⁵ found no differences in long-term clinical outcomes but reported significantly higher retear rates with SR vs. DR repairs. Although some studies have found no biomechanical advantage in different techniques,⁴⁰ there is evidence that DR techniques are superior to SR ones. Baums et al^3,4 investigated different fixation techniques in biomechanical studies and reported that DR fixation resulted in improved initial fixation strength and ultimate tensile load. However, this study only examined the infraspinatus tendon in a sheep model; thus, the application to the SSC tendon of the human shoulder remains unclear. Smith et al⁵² conducted another biomechanical study investigating the arthroscopic repair of SR and DR fixation in supraspinatus repair using human cadaveric shoulders. The authors concluded that DR reconstruction of the supraspinatus tendon insertion may provide a more reliable construct than an SR repair with a greater load to failure of 320 N for the DR compared to 224 N for the SR repairs. However, they did not evaluate the DR fixation for the SSC tendon. The results of the current study confirm these findings for the SSC tendon repair, showing greater load to failure and cyclic displacement for a DR repair technique. This may be especially beneficial for SSC tendon repairs, as it is the strongest muscle in the rotator cuff.³¹

From a clinical perspective, the ultimate biomechanical load to failure is an important measure as it resembles the failure strength of the repair in the early post-operative period. There is no clear consensus on the required minimum failure load for rotator cuff repair. Active and passive tension contribute to the force on the tendon of the rotator cuff. Passive properties of the rotator cuff consist of connective tissue such as collagen fibers, aponeurosis, and other elements. In an in vivo study by Gerber et al,²⁶ passive forces of 10 to 15 N were measured during reconstruction of supraspinatus tendons in a 60° abduction position, which rose up to 60 N when the arm was brought in neutral position, especially in long-standing ruptures. In a study by Hughes et al,²⁸ an SSC force of approximately 280 N was measured during abduction activities, and forces of approximately 100 N were measured during static arm elevation with the elbow extended. Moreover, Lorbach et al³⁷ calculated that a 20 N belly-press test resulted in a force of 200 N on the SSC attachment. Another proposed minimum construct strength by Mazzocca et al⁴³ determined that a minimum failure load of 250 N is required after rotator cuff repair. While transosseous sutures can fail at low tensile loads, suture augmentation and anchor repair improve the failure strength up to 400 N.^3,10,21 In the current study, only the DR repair achieved the proposed minimum failure load of 250 N. The DR repair technique demonstrated the highest failure strength of almost 400 N, which is consistent with the previously mentioned studies.

More recent techniques have described the use of knotless repair for both partial- and full-thickness SSC tendon ruptures.^16,17,22 Advantages of knotless repair include reduced surgical time, the ability to use a single portal technique, and a shorter learning curve. However, repair of larger tendon tears and late repairs can be difficult with a single-portal technique.²² In a biomechanical study by Borbas et al,⁷ the knotless tape repair showed equivalent stability to a knotted lasso-loop mattress and superior repair strength compared to a knotted horizontal mattress technique for upper-third SSC tendon tears. The present findings are consistent with this observation for full-thickness SSC tendon repairs, showing superior performance of the knotless tape repair compared with the knotted mattress repair, whereas the DR technique yielded the highest maximal failure load. An obvious advantage of the SR techniques is the simpler repair technique, which is further simplified with knotless tape techniques. In addition, the costs for both SR options are lower due to shorter operating times and less material required. In fact, in a biomechanical analysis by Lui et al,³⁶ suture tape had superior biomechanical properties with greater footprint contact pressure and higher failure rate compared to No. 2 sutures. Furthermore, there are reports on improved perfusion due to reduced local pressure with knotless tape repair in several studies.⁴¹

In addition to biomechanical strength, the reinserted footprint is another factor in achieving satisfactory stability. In a study by Koh et al,³² a larger footprint coverage of the repair resulted in an increased tendon healing area at the bone surface. However, both complete and incomplete footprint coverage in rotator cuff repair showed no differences in clinical scores and range of motion at short-term follow-up. Another study by Joeng et al³⁰ also found no difference whether the rotator cuff footprint was partially or completely covered. In contrast, Heuberer et al²⁷ prospectively evaluated the outcomes of massive rotator cuff tears treated with arthroscopic débridement, partial, and complete repair, with the most favorable short-term improvements seen after complete repair. Therefore, to achieve greater footprint coverage, DR suture anchor fixation can be used, which restores a greater percentage of the anatomic footprint compared to a SR technique.⁴⁶ In another study by Mazzoca et al,⁴³ the DR anchor repairs consistently restored a larger footprint than the SR method. The current study showed similar results, with the largest repair area found in the DR knotted mattress repair technique. This is particularly interesting for the subscapularis, which has the largest footprint compared to the other rotator cuff muscles.¹²

There are certain limitations associated with this biomechanical study. First, the bone and soft tissue quality of the specimens used may differ from in vivo conditions. By far, the most common failure mechanism in our study was anchor pullout, likely related to reduced cadaveric bone quality due to postmortem changes such as dehydration and collagen degradation, which may result in increased brittleness and reduced toughness.⁵⁶ Secondly, the study does not consider the healing phase and the effect of the different techniques. Therefore, the results help to make an approximate statement of the forces that the repair can withstand immediately post-operatively. Despite biomechanical differences in repair techniques, the effect on clinical outcome remains with studies reporting similar clinical outcomes regardless of repair technique.^19,44,47,50 Third, muscle traction on the bone anchors was simulated. For reasons of reproducibility, complicated motion patterns were not used, and the load on the SSC tendon anchorage was simulated by unidirectional traction. The humerus was fixed, so no change in the angle between the humerus and subscapularis was considered. Finally, in the methods of this study, the detailed gap formation in relation to failure load was not analyzed. Therefore, unacceptable gap formation may have occurred before the reported ultimate failure loads. A larger repair area does not mean sufficient healing when gap formation may be clinically insufficient. Further studies with a detailed analysis of gap formation could help to better evaluate different repair constructs in the future.

Despite these limitations, the present study is an important contribution to the current literature, particularly with regard to complete SSC tendon tears, evaluating SR and DR repair.

Conclusion

DR repair technique for complete SSC tendon tears achieves significantly greater footprint coverage and the highest load to ultimate failure of the repair compared to SR repair techniques. This study suggests that DR repair techniques should be considered in the clinical setting of complete SSC tendon repairs.

Acknowledgment

Figures were created with BioRender.com.

Disclaimers

Funding: No funding was disclosed by the authors.

Conflicts of interest: The authors, their immediate families, and any research foundation with which they are affiliated have not received any financial payments or other benefits from any commercial entity related to the subject of this article.