Investigational R

All investigatio

Laboratory, VA

J Shoulder Elbow Surg (2014) 23, 1813-1821

1058-2746/$ - s

http://dx.doi.org

www.elsevier.com/locate/ymse

Does transosseous-equivalent rotator cuff repairbiomechanically provide a ‘‘self-reinforcement’’effect compared with single-row repair?

Maxwell C. Park, MDa,*, Michelle H. McGarry, MSb, Robert C. Gunzenhauserb,Michael K. Benefiel, BAb, Chong J. Park, PhDc, Thay Q. Lee, PhDb,d

aDepartment of Orthopaedic Surgery, Southern California Permanente Medical Group, Woodland Hills, CA, USAbOrthopaedic Biomechanics Laboratory, Long Beach VA Healthcare System, Long Beach, CA, USAcDepartment of Mathematics and Statistics, San Diego State University, San Diego, CA, USAdDepartment of Orthopaedic Surgery, University of California, Irvine, CA, USA

Background: Transosseous-equivalent (TOE) rotator cuff repair has been theorized to be ‘‘self-reinforc-ing’’ against potentially destructive and increasing tendon loads. The goal of this study was to biomechan-ically verify and characterize the effect of increasing tendon load on frictional resistance over a repairedfootprint for single-row (SR) and TOE repair techniques.Methods: In 10 fresh frozen human shoulders, TOE and SR supraspinatus tendon repairs were performedin each specimen. For all repairs, a pressure sensor was secured at the tendon-footprint interface. Thesupraspinatus tendon was loaded with 0, 20, 40, 60, and 80 N at 0� and 30� abduction. Paired t testsand multivariate regression analyses were used for comparisons.Results: The SR repair had significant increases in footprint contact force, area, and pressure betweeneach and all tendon-loading conditions (P < .05). The TOE repair similarly demonstrated increases in foot-print contact force with increasing tendon load (P < .05). Comparing between repairs, TOE repair hadmore footprint contact force, area, pressure, and peak pressure at each load for both abduction angles(P < .05). With increasing load, the TOE repair had a significantly higher progression (slope) of footprintforce and pressure compared with the SR repair.Conclusions: Self-reinforcing capacity in rotator cuff repair has been biomechanically characterized andverified. The TOE repair, with tendon-bridging sutures fixed medially and spanning the footprint, providesdisproportionately more progressive footprint frictional resistance with increasing tendon loads comparedwith the SR repair secured over isolated fixation points. This self-reinforcing effect could help sustainstructural integrity and potentially improve healing biology.Level of evidence: Basic Science Study, Biomechanics.� 2014 Journal of Shoulder and Elbow Surgery Board of Trustees.

Keywords: Rotator cuff; transosseous-equivalent; self-reinforcing; single-row; biomechanics; shoulder

eview Board approval was not required for this study.

ns were performed at the Orthopaedic Biomechanics

Long Beach Healthcare System, Long Beach, CA.

*Reprint requests: Maxwell C. Park, MD, Woodland Hills Medical

Center, Kaiser Foundation Hospital, Department of Orthopaedic Surgery,

5601 De Soto Ave, Woodland Hills, CA 91365, USA.

E-mail address: maxwellpark1@yahoo.com (M.C. Park).

ee front matter � 2014 Journal of Shoulder and Elbow Surgery Board of Trustees.

/10.1016/j.jse.2014.03.008

Figure 1 Schematic representation shows how footprint contactincreases with increasing tendon load. As tendon load increasesfrom T to T0, the suture loop from a repair that is not fixedmedially elongates and narrows (double arrows) creating a ‘‘focalloop wedge.’’ This effect creates a compression vector over thefootprint laterally, and the exposed contact area (C) diminishes.With medial fixation and tendon-bridging sutures, the ‘‘wedge’’effect can include the entire medial footprint.

Figure 2 Custom testing apparatus.

1814 M.C. Park et al.

The concept of a ‘‘self-reinforcing’’ rotator cuff repairhas been described and theorized to provide a mechanismof resistance to structural failure in the face of otherwisedestructive forces.5 A repair construct that provides thismechanism has been described to mimic a Chinese fingertrap, ‘‘in which the system grips more tightly as the tensilelongitudinal disruptive force increases.’’5 This concept of‘‘self-reinforcement’’ in the setting of rotator cuff repair,although theorized, has not been biomechanically verified,quantified, or characterized.

The purpose of the current study was to determinewhethera transosseous-equivalent (TOE) repair could provide a self-reinforcing mechanism of preservation and be ‘‘self-protec-tive’’ against potentially destructive forces. Tendon-bridgingsutures provide a compression vector over the repairedtendon25,31 that may improve footprint contact force withprogressive tendon loading. These sutures can potentiallyprovide frictional resistance to increasing gap formation andultimate failure. However, similarly, with single-row (SR)repair, increasing tendon load will transmit some componentof the loading vector onto the repair site as well (Fig. 1). Wehypothesized that the TOE repair would provide dispropor-tionately more frictional force with the same progressivetendon-loading sequence compared with SR repair. Is itpossible that bridging sutures dissipate the compression forceover the tendon so that the theory of self-reinforcement issimply theoretical? If the progressive increase in footprintcompression force is not significantly different between bothrepairs, arguably, the theory of self-reinforcement does notapply to suture tendon-bridging constructs; in this case, tendonloadingwould simply be providing an obligatory transmissionof vector force to the footprint and no exceptional protectiveforce from the bridging sutures themselves.

Materials and methods

Specimen preparation

The study used 10 fresh frozen human cadaveric male shoulderswith a mean age of 57.3 � 9.1 years (range, 46-71 years), withoutevidence of rotator cuff tear or pathology. Specimens were storedat �20�C and thawed 24 hours at room temperature beforedissection. Specimens were dissected of all soft tissue and dis-articulated at the glenohumeral joint to isolate the supraspinatusmuscle and its tendinous insertion on the greater tuberosity. Thehumerus was transected approximately 7 cm from the inferiorarticular margin and potted in a 1.5-inch-diameter polyvinylchloride pipe with plaster of Paris.

After the plaster was set, the supraspinatus tendon wassharply dissected from its insertion on the greater tuberosity,and the bony footprint of the supraspinatus was treated with afine rasp. The distal 10 mm end of the supraspinatus tendon wasresected from anterior-to-posterior to simulate a single-tendonrotator cuff tear.15,32 The proximal end of the supraspinatustendon was sutured using No. 2 nonabsorbable FiberWire su-ture (Arthrex, Naples, FL, USA) with an interlocking whip-stitch to use for applying various loads to the supraspinatus.

The humerus was inserted into a custom testing devicedesigned to allow for humeral abduction and supraspinatusmuscle loading (Fig. 2).31 A normal saline solution was used tokeep specimens moist during all phases of dissection, prepa-ration, and testing.

Pressure sensor preparation

Supraspinatus repair contact force, area, and pressure weremeasured using a Tekscan 4201 sensor (Tekscan Inc, South Bos-ton, MA, USA). Sensors were trimmed to fit the supraspinatusfootprint dimensions. The final dimension of the sensors was

Figure 3 Tekscan sensor secured to the rotator cuff footprintwith small screws.

Rotator cuff repair self-reinforcement 1815

8 mm (medial-lateral) by 20 mm (anterior-poster), resulting in atotal contact area of 160 mm2 for all specimens tested.31 Thesensors were then sealed between 2 layers of clear tape to preventmoisture from contacting the sensor, which would damage thesensing elements.31 The tape was trimmed close to the sensor inthe medial-lateral dimension to allow for anchor placement, whileleaving approximately 1 cm on the anterior and posterior ends toultimately allow for sensor attachment to the greater tuberosity.31

After trimming and sealing, each sensor was calibrated usingan Instron 4411 load cell (Instron, Norwood, MA, USA). Wedetermined from results of pilot testing of supraspinatus repaircontact using uncalibrated sensors that the highest sensitivitysetting could be used without saturating the sensor and that thecontact force range with varying supraspinatus loads was between5 N and 20 N. Before being calibrated, the sensors were loadedfrom 5 N to 20 N for 10 cycles using the Instron machine fittedwith a small plunger and foam pad cut to match the dimensions ofthe sensor. The sensors were then calibrated using a 2-pointcalibration at 5 N and 20 N. After being calibrated, the Tekscansensor was secured to the supraspinatus footprint using 2 smallscrews inserted through the anterior and posterior portion of thetape used to seal the sensors (Fig. 3).31 Immobilizing the Tekscansensor ensured that it would remain at the same locationthroughout all experimental trials. The supraspinatus tendon wasthen repaired, and all contact variables were measured.

Repair techniques

In 10 shoulder specimens, 2 repairs were performed for eachspecimen. One repair was a TOE repair including a medial pulley-mattress configuration (MP-TOE).30 The other repair was a SRrepair using double-loaded anchors and simple suture configura-tions. All measurements for each repair were carefully performedwith a ruler to standardize the technique for each specimen. Allknots were tied with a sliding double half-hitch knot first,

followed by alternating simple half-hitches, for a total of 5 throwsusing a knot-pusher. The MP-TOE repair was performed first,followed by the SR repair. Results of pilot testing and a previousstudy showed the suture-bridging technique did not affect thecontact measurements for the subsequent SR repair.31 For eachrepair, therefore, the same specimen was used to serve as areproducible internal control. The same surgeon (M.C.P.) per-formed all repairs.

MP-TOE repair

A previous study found the MP-TOE was biomechanicallyequivalent to the original TOE repair, even though the formerrepair required fewer suture passes through the repaired tendon,and the mode of failure for this repair was also more forgiving atthe musculotendinous junction.30 Two medial-row pilot holeswere punched at the medial edge of the greater tuberosity footprintjust lateral to the articular cartilage yet medial to the Tekscansensor; the anterior hole was placed 5 mm posterior to the bicipitalgroove. Two lateral pilot holes were punched 1 cm distal to thelateral edge of the supraspinatus footprint in line with the medialholes, and the distance between each suture anchor in the samerow was 15.0 mm anterior-to-posterior. All holes were placed at a45� angle relative to the bony surface. The medial row holesreceived 5.5-mm Bio-Corkscrew FT suture anchors (Arthrex),single-loaded with No. 2 nonabsorbable FiberWire suture(Arthrex). However, instead of passing the suture limbs throughthe supraspinatus tendon via 4 individual passes, anterior andposterior suture limb pairs were passed simultaneously, with only2 transtendon suture passes. A looped FiberSnare shuttle suture(Arthrex) was used to pass each suture limb pair 12 mm from thelateral tendon edge, simultaneously through the supraspinatustendon, centered above its respective anchor.

To secure the medial tendon, 1 suture limb from the anterioranchor and 1 limb from the posterior anchor were tied with thestandard knot over the knot-pusher shaft, and the posterior freesuture limb was pulled to bring the knot to the posterior anchor,creating a broad mattress configuration between anchors. Theremaining 2 suture limbs were tied with a non-sliding knot overthe anterior anchor. This technique creates 2 broad overlappinganterior-posterior suture bridges at the medial aspect of thesupraspinatus footprint. The sutures tails were not cut. Two of theposterior sutures from the tied mattresses, 1 from each anchor,were incorporated into a 5.5-mm knotless SwiveLock anchor(Arthrex) and placed posteriorly at the distal-lateral position. Witha constant application of tension, the screw was placed at a 45�

angle relative to the proximal lateral humerus. This process wasrepeated with the remaining 2 anterior suture limbs to completethe repair, thus creating an additional 4 suture tendon-bridgingconstruct criss-crossing centrally (Fig. 4). After advancing thelateral anchors, the free suture limbs were cut.

SR suture anchor repair

Two lateral holes were punched as far lateral as possible while stillremaining on top of the footprint, thus maximizing the potentialtendon contact area on the bony insertion.31 Each anchor wascentered 15.0 mm from the other anterior to posterior. Two 5.5-mm Bio-Corkscrew FT anchors were used, which were double-loaded with No. 2 nonabsorbable FiberWire suture.32 Simple

Figure 4 Medial pulley transosseous-equivalent repair with abroad medial interimplant mattress requiring only 2 tendon suturepasses.

Figure 5 Single-row repair using 2 double-loaded suture an-chors with simple suture configurations.

1816 M.C. Park et al.

suture configurations were used for the lateral row. The suturepasses were approximately 7 mm apart from one another for agiven anchor and 10 mm medial to the lateral edge of the simu-lated tear (Fig. 5), similar to another study.29 Although middlefootprint25 and medial SR repairs37,40 have been described, thislaterally based repair configuration was used to maximize foot-print coverage3,10,29 without compromising the sensor, thusminimizing bias against this repair.

Biomechanical testing

Fixed within the custom testing apparatus, the humerus was lockedin neutral rotation by aligning the supraspinatus footprint dimension(anterior-to-posterior) orthogonal to the loading vector so that theapproximate supraspinatus centroid was slightly posterior to thecenter of the humeral head.29 The loading vector line of action wasdefined by aNo. 2whip-stitched interlocking suture alignedwith thecentroid of themuscle and parallel to the baseplate.31 The suturewaspassed through the pulley system from which the loads wereapplied.31 To evaluate the effect of increasing supraspinatus load onfootprint frictional force, contact measurements were recorded withthe following supraspinatus loads: 0, 20, 40, 60, and 80 N. Mea-surements were performed at 0� and 30� of glenohumeral abduc-tion.31 The abduction angle was defined relative to lines along the

length of the humerus and the supraspinatus muscle and tendon, aspreviously described.31 The pulley component was adjusted verti-cally when the abduction angle was changed to keep the loadedsuture parallel with the baseplate.

Before testing, specimens were sequentially ramp loaded using20-N increments, from 0 to 80 N. The loads applied simulatephysiologic loads that may be experienced postoperatively.1,29,34

Measurements were then performed in the same step-wise orderfor the first trial from 0 to 80 N, and then followed with a secondtrial; 2 separate trials were performed to confirm repeatability, andthe values were averaged.

After testingwas completed at both abduction angles for theMP-TOE repair, the repair was carefully removed and the SR repair wasperformed using the same specimen. Pilot studies demonstrated nogross effect of repair order on contact variables. But we noted thatthe SR repair would potentially affect the tendon and bone relativelymore for a subsequent MP-TOE repair based on where the sutureswere passed through the tendon and where the suture anchors wereplaced on the tuberosity; for the SR repair, the MP-TOE bridgingsutureswould cross the affected tendon and anchor sites from the SRrepair, potentially adding more variability between testing condi-tions. The Tekscan pad was left attached to the greater tuberositywhile the second repair was performed.Allmeasurementswere thenrepeated with the SR repair. For each repair, the footprint contact

Table I Supraspinatus repair contact characteristics for a single-row repair and medial-pulley transosseous equivalent repair withincreasing supraspinatus load at 0� abduction.

Supraspinatus load Area (mm2) Force (N) Pressure (kPa) Peak pressure (kPa)

Mean (SE) Mean (SE) Mean (SE) Mean (SE)

Single-row0 N 31.6 (6.4) 1.3 (0.5) 33.7 (7.3) 47.5 (14.3)20 N 58.7 (9.0) 3.0 (0.9) 44.1 (7.0) 69.0 (14.9)40 N 68.0 (8.7) 4.7 (1.4) 58.1 (10.9) 97.5 (19.5)60 N 76.2 (8.3) 6.2 (1.9) 69.2 (14.4) 123.0 (29.7)80 N 80.8 (8.0) 7.5 (2.4) 78.5 (17.4) 149.0 (37.2)

MP-TOE0 N 84.3 (11.4) 5.3 (1.0) 59.2 (7.0) 144.5 (37.4)20 N 110.0 (6.1) 10.5 (1.5) 92.4 (10.8) 211.5 (34.7)40 N 114.8 (4.9) 14.0 (2.1) 119.0 (15.1) 246.5 (28.9)60 N 118.6 (4.2) 16.9 (2.6) 139.4 (19.8) 285.5 (33.2)80 N 121.2 (3.8) 19.0 (3.1) 154.3 (24.1) 305.5 (39.1)

Single-row vs MP-TOE P values0 N .004 .013 .023 .04020 N .001 .002 .004 .00240 N <.001 .003 .008 .00260 N .001 .005 .012 .00480 N .001 .008 .018 .007

MP-TOE, medial-pulley transosseous equivalent repair; SE, standard error.

Rotator cuff repair self-reinforcement 1817

area, contact force, mean contact pressure, and peak contact pres-sure using a 2� 2-pixel area were determined and averaged for bothtrials, and finally, averaged across the specimens. After measure-ments were completed, each Tekscan pad was removed and thencalibrated and tested for functionality and loss of sensitivity usingthe Instron 4411 materials testing machine. Accuracy after testingwaswithin a range of� 1.5N,with repeatability at� 0.7N. Inherentloss of sensitivity31 and decreased sensitivity with testing duration20

were both considered when using the Tekscan technology. Pilotstudies demonstrated relatively insignificant sensitivity losses,likely related to the number of testing conditions evaluated, and thestatic (vs cyclic) loading conditions necessitated by the studydesign.

Statistical analyses

A paired t test was used to statistically compare the values be-tween the MP-TOE repair and the SR repair. A paired t test wasalso used to compare between abduction angles and also betweensupraspinatus loading conditions for a given repair. The slopes ofcontact force vs supraspinatus force and contact pressure vssupraspinatus force were calculated to evaluate the effect ofincreasing muscle load. Multivariate regression analysis wasperformed to compare the differences in slopes between the 2repair groups. A P value of <.05 was used as the level of sig-nificance for all comparisons.

In a previous study using the same testing apparatus andsimilar methods, a power analysis was performed for contact areameasurements for a TOE repair comparing positions of abductionat 0�, 30�, and 60�.31 On the basis of analyses in our laboratorythat involved measuring contact area by use of our Tekscan sensorand software, the standard deviation was approximately 7 mm2.

To detect a minimum 14 mm2 (8.75% of the sensor) difference incontact area between abduction positions with a 5% level of sig-nificance (a ¼ .05), at least 6 specimens were determined to benecessary for achieving a power of 95%.

Results

The MP-TOE repair had significantly larger footprint force,contact area, mean contact pressure, and peak contactpressure compared with the SR repair at each supraspinatusload for 0� (Table I) and 30� (Table II) abduction (P < .05).The MP-TOE repair resulted in an average 86.2% increasein contact force, 219.7% increase in contact area, 93.5%increase in contact pressure, and 143.7% increase in peakcontact pressure compared with the SR repair when aver-aged across supraspinatus loads and abduction angles.

For a given repair, there were no significant differencesin contact characteristics between the 2 abduction angles.The SR repair had significant (P < .05) increases in foot-print force (N) between every tendon-loading condition(1.4 � 0.5, 2.9 � 0.8, 4.6 � 1.3, 6.0 � 1.6, and 7.6 � 2.1 Nfor tendon loads of 0, 20, 40, 60, and 80 N, respectively;Fig. 6). For the SR repair group, this same relationshipbetween all measurements was seen with contact area atboth abduction angles and at 0� abduction for pressure(Fig. 7). The MP-TOE repair demonstrated the same sig-nificant relationships (P < .05) for footprint force withincreasing tendon load (5.9 � 1.2, 10.6 � 1.7, 14.1 � 2.2,16.6 � 2.6, and 18.4 � 3.1 N for each progressive tendonload, respectively; Fig. 6). For the MP-TOE repair group,

Table II Supraspinatus repair contact characteristics for a single-row repair and medial-pulley transosseous equivalent repair withincreasing supraspinatus load at 30� abduction

Supraspinatus load Area (mm2) Force (N) Pressure (kPa) Peak pressure (kPa)

Mean (SE) Mean (SE) Mean (SE) Mean (SE)

Single-row0 N 33.0 (6.8) 1.4 (0.5) 33.1 (7.2) 46.0 (14.3)20 N 56.5 (9.4) 2.9 (0.8) 46.9 (6.4) 77.0 (14.6)40 N 69.3 (9.9) 4.6 (1.3) 59.2 (9.0) 102.5 (22.3)60 N 77.8 (9.4) 6.0 (1.6) 71.9 (11.7) 128.0 (29.2)80 N 84.2 (8.8) 7.6 (2.1) 84.0 (14.4) 156.5 (33.7)

MP-TOE0 N 89.0 (13.7) 5.9 (1.2) 61.5 (9.0) 120.5 (22.3)20 N 109.8 (8.6) 10.6 (1.7) 93.2 (11.2) 186.5 (27.5)40 N 118.4 (6.5) 14.1 (2.2) 116.5 (15.1) 234.0 (33.6)60 N 120.8 (6.0) 16.6 (2.6) 135.2 (19.0) 273.5 (39.1)80 N 120.7 (6.0) 18.4 (3.1) 149.6 (22.1) 298.0 (43.5)

Single-row vs MP-TOE P values0 N .005 .005 .005 .01120 N .001 .001 .004 .00540 N .002 .001 .006 .00460 N .003 .002 .012 .00680 N .002 .005 .020 .010

MP-TOE, medial-pulley transosseous equivalent repair; SE, standard error.

Figure 6 Footprint contact force and slope with increasingtendon load. With increasing supraspinatus loads, the medial

1818 M.C. Park et al.

this same relationship between all measurements was alsoseen for contact pressure (Fig. 7) at both abduction anglesand at 30� abduction for area.

With increasing supraspinatus loads, the MP-TOE repairhad a significantly higher progression (slope) of footprintcontact force compared with the SR repair at 0� (P ¼ .025)and 30� (P ¼ .014) abduction (Fig. 6). The MP-TOE repairalso had a significantly higher progression (slope) of foot-print contact pressure compared with the SR repair at 30�

abduction (P ¼ .034) (Fig. 7). For each abduction angle, theslopes for force and pressure were averaged over the 10measurements (obtained from each of 10 specimens) for boththe MP-TOE and SR repairs; the ratio of the repair averages(MP-TOE slope average/SR slope average) for a givenabduction position were 1.99 (footprint force at 0� abduc-tion), 1.72 (footprint pressure at 0� abduction), 2.20 (foot-print force at 30� abduction), and 2.07 (footprint pressure at30� abduction). For a given repair, there were no significantdifferences in slope comparing 0� and 30� of abduction.

pulley transosseous-equivalent (MP-TOE) repair had a signifi-cantly higher progression (slope) of footprint contact forcecompared with the single-row (SR) repair at 0� (P ¼ .0249) and30� (P ¼ .01404) abduction. For a given repair, there were nosignificant differences in slope comparing 0� and 30� of abduc-tion. ‘‘Repair #,’’ where the # (0� or 30�) represents the abductionangle, followed by the linear equation quantifying the slope.

Discussion

The concept of ‘‘self-reinforcement’’ in rotator cuff repairhas been theorized to apply to tendon-bridging constructsbetween medial and lateral rows.5 The theory postulates thatincreasing footprint frictional forces can be generatedwith progressive and yet counterproductive tendon-loadingforces. The purpose of the current study was to biomechan-ically validate this theory and characterize frictional force at

a repaired rotator cuff footprint as it relates to progressivelyincreasing failure loads. Burkhart et al5 described the self-reinforcement concept involving a ‘‘wedge’’ effect createdby tendon-bridging sutures that are fixed between both

Figure 7 Footprint contact pressure and slope with increasingtendon load. The medial pulley transosseous-equivalent (MP-TOE)repair had a significantly higher progression (slope) of footprintcontact pressure compared with the single-row (SR) repair at 30�

abduction (P¼ .03425). For a given repair, therewere no significantdifferences in slope comparing 0� and 30� of abduction. ‘‘Repair #’’,where the # (0� or 30�) represents the abduction angle, followed bythe linear equation quantifying the slope.

Rotator cuff repair self-reinforcement 1819

medial and lateral bony anchors: the tendon is increasingly‘‘wedged’’ between the sutures and bony footprint asprogressive tendon loading occurs. Baseline footprintcompression comes fromdistal-lateral anchor placement thatcreates an obligatory normalizing force that is perpendicularto the axis of tendon loading26,31; the current study has shownthat tendon loading actually enhances this footprintcompression effect rather than promote construct failure, asintuition may suggest.

With respect to SR repair, the anchor is not distal-lateralto the footprint, and compression to bone is more likelyarising from the suture loop knot itself.25 With tendonloading, this loop elongates, with the superior limb creatinga relative compression vector and a ‘‘focal loop wedge’’effect. This elongation is obligatory with tendon loading forany suture loop (Fig. 1). Because there is no medial fixationfor this loop and more complete footprint ‘‘self-protection’’cannot occur with relatively less footprint contact areainvolved (and therefore less frictional resistance to tendondistraction), tendon loading could coincidentally contributedirectly to construct failure (eg, gapping, decreased yieldloads) in the SR repair setting.15

The current study has shown that increasing tendon loadsdo in fact increase footprint compression and frictional force;in addition, area and pressure significantly increases as wellfor a given repair. Thiswas seen for both TOEand SR repairs.A prior study showed that footprint contact significantly in-creases with the TOE repair compared with double-row andSR repairs with both abduction and rotation31; in that study,the load was kept constant. Arguably, the most significantfinding in the current study is that the progressive footprint

force increases (slopes of measurements) were significantlydifferent between TOE and SR repairs, highlighting theadditional contribution tendon-bridging sutures conveyto the self-reinforcing mechanism. If the slopes of theincreasing footprint forces were not different between re-pairs, the effect would be considered obligatory and notnovel; so beyond the absolute contact differences betweenrepairs, which was characterized in a prior study,31 the dif-ferences in slopes verify that disproportionate frictionalforces are created with TOE repair, even in the face ofincreasing tendon-loading forces. In fact, for both 0� and 30�

abduction, average footprint force ratios between MP-TOEand SR repairs (MP-TOE:SR) were 1.99 and 2.20, respec-tively, suggesting TOE repair can provide approximatelytwice as much ‘‘self-reinforcing’’ effect.

Abduction, as a kinematic variable, did not statisticallyaffect footprint contact forces for a given repair technique.In general, increasing abduction has been shown todecrease footprint contact forces.31 Statistically significantdifferences were seen for 0� and 60�, not 0� and 30�, whichis consistent with the current study. It is possible that atabduction angles of 60� or higher, SR repair woulddemonstrate less footprint force transmission withincreasing tendon loads compared with the 0� and 30�

conditions.The current study can be considered ‘‘part 2’’ of a prior

study that analyzed the failure load characteristics of a tech-nically optimizedTOErepairdanalogous to theoriginal 2-partbiomechanical characterization of the TOE repair.27,32 ‘‘Part1’’ of the current study highlighted the benefit of a technicallyoptimizedTOErepair requiringonly 2 suture passesvs 4 for theoriginal description of the TOE repair. Yet, both repairs werebiomechanically equivalent.30 The current study verifies thetheory of self-reinforcement using this simplified 2-suture passTOE repair that uses a single, broad medial mattress configu-ration, which is relatively technically efficient24 yet potentiallyless traumatic at the musculotendinous junction.30 Medialmattress fixation has been shown to improve biomechanicalperformance,6,14,19 and whether knotless repair constructs areclinically sufficient is debatable.

This raises the question: Do focal medial mattress knotstenodese the rotator cuff such that the self-reinforcing effectis actually limited or neutralized? From the results of thecurrent study we know the MP-TOE repair with a broadinterimplant mattress does not have this tenodesis effect. Incontrast, prior studies have shown the importance of focalmedial knots.2,6,14,18,19,33 However, clinically, knotlessrepair medially has shown promising results.4,38 It is possiblethat knotless repair that uses tendon-bridging sutures createsa self-reinforcing environment that supplants the need forfocal medial knots. Furthermore, the anterior ‘‘cord’’ of thesupraspinatus tendon, which also represents the anteriorportion of the rotator cable, has been shown to be important,particularly in external rotation.13,21,28,29,35 Additionalanterior fixation has been theorized to improve bio-mechanical performance.21,28,29 The question of whether

1820 M.C. Park et al.

this additional fixation potentially neutralizes the ‘‘self-reinforcement’’ effect requires further study as well.

Insofar as the SR repair does create increased footprintcontact forces and pressure with increasing physiologictendon loads, it could be argued that SR repair is also self-reinforcing, with some degree of structural self-preservationoccurring. SR repair would not be self-reinforcing if tendonloading primarily and coincidentally created loss of struc-tural integrity with suture loop elongation (and more gapformation). In contrast, progressive tendon loadingwith TOErepair disproportionately (P < .05) increases repair sitefrictional resistance to failure compared with SR repair.However, other SR repairs have been described withimproved loading and contact variables.16,17 These repairsthat lack linked footprint-spanning sutures anchored medi-ally are not optimizing footprint contact area restoration.16

Therefore, a mechanism for optimal self-reinforcement inthese more technically involved SR repairs would not exist.

The limitations to this study include those inherent to allcadaveric studies, with only time-zero information available.In addition, biological and healing factors cannot beincluded; for example, tendon vascularity cannot be tested.However, intratendinous blood flow has been shown to bepreserved after suture tendon-bridging repair,9 and bloodsupply has been shown to come from the peribursal tissue andbony anchor sites.12 Despite this, there has been concernabout medial tendon strangulation and necrosis with failureafter TOE repair fixed medially.7,8 The MP-TOE repair wasdesigned to create a broad zone of fixation vs multiple focalspot-welds of fixation, demonstrating less tendon cutout withimproved load-sharing capacity.30 This medial interimplantmattress suture-bridge may help to maintain vascularity andstructural integrity, without compromising load-to-failurebiomechanics.30 Although cadaveric studies are limited inassessing healing potential, laboratory studies can providerelevant information from highly controlled variables,limiting confounding factors and bias.

Conclusions

The results from the current study show that the forcetransmission from muscle to footprint appears to beobligatory, with tendon loading for both SR and TOErepair. However, theTOE repair demonstrated significantlymore progressive frictional force and pressure transmissionbased on the differences in slopes (Figs. 6 and 7); arguably,given the slope ratios between repairs, the TOE repair canprovide approximately double the reinforcing effect.Because the progressive increase in footprint compressionforce is significantly different between both repairs, argu-ably, the theory of self-reinforcement does apply to suturetendon-bridging constructs with relatively exceptionalprotective forces arising from the bridging sutures them-selves. This characterization helps to verify the theory of

self-reinforcement and provides another reason whytransosseous-equivalent repair performs well biomechani-cally. This may explain the favorable outcomes seen clin-ically with interconnected, tendon-bridging constructs thatrestore the entire footprint.4,11,22,23,36,38,39 The SR repair,with a limited contact area and suture loops that are notanchored to bone medially, demonstrates a ‘‘focal loopwedge’’ effect (Fig. 1) and may not provide an optimizedself-protective environmentwith tendon loading and sutureloop elongation.

Disclaimer

This study received funding from a research grant fromArthrex Inc., Naples, FL, USA.

The authors, their immediate families, and anyresearch foundations with which they are affiliated havenot received any financial payments or other benefits fromany commercial entity related to the subject of this article.

References

1. Ahmad CS, Stewart AM, Izquierdo R, Bigliani LU. Tendon-bone

interface motion in transosseous suture and suture anchor rotator cuff

repair techniques. Am J Sports Med 2005;33:1667-71. http://dx.doi.

org/10.1177/0363546505278252

2. Anderl W, Heuberer PR, Laky B, Kriegleder B, Reihsner R,

Eberhardsteiner J. Superiority of bridging techniques with medial

fixation on initial strength. Knee Surg Sports Traumatol Arthrosc

2012;20:2559-66. http://dx.doi.org/10.1007/s00167-012-1922-9

3. Boileau P, Brassart N, Watkinson DJ, Carles M, Hatzidakis AM,

Krishnan SG. Arthroscopic repair of Full-thickness tears of the

supraspinatus: does the tendon really heal? J Bone Joint Surg Am

2005;87:1229-40. http://dx.doi.org/10.2106/JBJS.D.02035

4. Boyer P, Bouthors C, Delcourt T, Stewart O, Hamida F, Mylle G, et al.

Arthroscopic double-row cuff repair with suture-bridging: a structural

and functional comparison of two techniques. Knee Surg Sports Trau-

matol Arthrosc 2013;. http://dx.doi.org/10.1007/s00167-013-2401-7

5. Burkhart SS, Adams CR, Burkhart SS, Schoolfield JD. A biome-

chanical comparison of 2 techniques of footprint reconstruction for

rotator cuff repair: the SwiveLock-FiberChain construct versus stan-

dard double-row repair. Arthroscopy 2009;25:274-81. http://dx.doi.

org/10.1016/j.arthro.2008.09.024

6. Busfield BT, Glousman RE, McGarry MH, Tibone JE, Lee TQ. A

biomechanical comparison of 2 technical variations of double-row ro-

tator cuff fixation: the importance ofmedial-rowknots.AmJSportsMed

2008;36:901-6. http://dx.doi.org/10.1177/0363546507312640

7. Cho NS, Lee BG, Rhee YG. Arthroscopic rotator cuff repair using a

suture bridge technique: is the repair integrity actually maintained?

Am J Sports Med 2011;39:2108-16. http://dx.doi.org/10.1177/

0363546510397171

8. Cho NS, Yi JW, Lee BG, Rhee YG. Retear patterns after arthroscopic

rotator cuff repair: single-rowversus suture bridge technique.AmJSports

Med 2010;38:664-71. http://dx.doi.org/10.1177/0363546509350081

9. Christoforetti JJ, Krupp RJ, Singleton SB, Kissenberth MJ, Cook C,

Hawkins RJ. Arthroscopic suture bridge transosseous equivalent fix-

ation of rotator cuff tendon preserves intratendinous blood flow at the

Rotator cuff repair self-reinforcement 1821

time of initial fixation. J Shoulder Elbow Surg 2012;21:523-30. http://

dx.doi.org/10.1016/j.jse.2011.02.012

10. Deutsch A, Kroll DG, Hasapes J, Staewon SR, Pham C, Tait C. Repair

integrity and clinical outcome after arthroscopic rotator cuff repair

using single-row anchor fixation: a prospective study of single-tendon

and two-tendon tears. J Shoulder Elbow Surg 2008;17:845-52. http://

dx.doi.org/10.1016/j.jse.2008.04.004

11. Frank JB, ElAttrache NS, Dines JS, Blackburn A, Crues J, Tibone JE.

Repair site integrity after arthroscopic transosseous-equivalent/suture-

bridge rotator cuff repair. Am J Sports Med 2008;36:1496-503. http://

dx.doi.org/10.1177/0363546507313574

12. Gamradt SC, Gallo RA, Adler RS, Maderazo A, Altchek DW,

Warren RF, et al. Vascularity of the supraspinatus tendon three months

after repair: characterization using contrast-enhanced ultrasound. J

Shoulder Elbow Surg 2010;19:73-80. http://dx.doi.org/10.1016/j.jse.

2009.04.004

13. Gates JJ, Gilliland J, McGarry MH, Park MC, Acevedo D,

Fitzpatrick MJ, et al. The influence of distinct anatomic subregions of

the supraspinatus on humeral rotation. J Orthop Res 2010;28:12-7.

http://dx.doi.org/10.1002/jor.20947

14. Kaplan K, ElAttrache NS, Vazquez O, Chen YJ, Lee TQ. Knotless

rotator cuff repair in an external rotation model: the importance of

medial-row horizontal mattress sutures. Arthroscopy 2011;27:471-8.

http://dx.doi.org/10.1016/j.arthro.2010.11.006

15. KimDH, ElAttrache NS, Tibone JE, Jun BJ, DeLaMora SN, Kvitne RS,

et al. Biomechanical comparison of a single-row versus double-row

suture anchor technique for rotator cuff repair. Am J Sports Med

2006;34:407-14. http://dx.doi.org/10.1177/0363546505281238

16. Lorbach O, Kieb M, Raber F, Busch LC, Kohn D, Pape D. Comparable

biomechanical results for a modified single-row rotator cuff recon-

struction using triple-loaded suture anchors versus a suture-bridging

double-row repair. Arthroscopy 2012;28:178-87. http://dx.doi.org/10.

1016/j.arthro.2011.08.298

17. Lorbach O, Kieb M, Raber F, Busch LC, Kohn DM, Pape D.

Three-dimensional evaluation of cyclic displacement in single-row

and double-row rotator cuff reconstruction under static external rota-

tion. Am J Sports Med 2013;41:153-62. http://dx.doi.org/10.1177/

0363546512466652

18. Maguire M, Goldberg J, Bokor D, Bertollo N, Pelletier MH, Harper W,

et al. Biomechanical evaluation of four different transosseous-equivalent/

suture bridge rotator cuff repairs. Knee Surg Sports Traumatol Arthrosc

2011;19:1582-7. http://dx.doi.org/10.1007/s00167-011-1436-x

19. Mall NA, Lee AS, Jaskamdip C, Van Thiel GS, Romeo AA, Verma NN,

et al. Transosseous-equivalent rotator cuff repair: a systematic review on

the biomechanical importance of tying the medial row. Arthroscopy

2013;29:377-86. http://dx.doi.org/10.1016/j.arthro.2012.11.008

20. Mazzocca AD, Bollier MJ, Ciminiello AM, Obopilwe E,

DeAngelis JP, Burkhart SS, et al. Biomechanical evaluation of

arthroscopic rotator cuff repairs over time. Arthroscopy 2010;26:592-

9. http://dx.doi.org/10.1016/j.arthro.2010.02.009

21. Nguyen ML, Quigley RJ, McGarry MH, Jun BJ, Gupta R, Burkhart SS,

et al. Margin convergence anchorage to bone for reconstruction of the

anterior attachment of the rotator cable. Arthroscopy 2012;28:1237-45.

http://dx.doi.org/10.1016/j.arthro.2012.02.016

22. Park JY, Lee SY, Chung SW, Zulkifli H, Cho JH, Oh KS. Clinical

comparison between double-row and transosseous-equivalent repairs

for medium to large size rotator cuff tears. Arch Orthop Trauma Surg

2013;133:1727-34. http://dx.doi.org/10.1007/s00402-013-1872-9

23. Park JY, Lhee SH, Oh KS, Moon SG, Hwang JT. Clinical and ultra-

sonographic outcomes of arthroscopic suture bridge repair for massive

rotator cuff tear. Arthroscopy 2013;29:280-9. http://dx.doi.org/10.

1016/j.arthro.2012.09.008

24. Park MC. Biomechanical validation of rotator cuff repair techniques

and considerations for a ‘‘technical efficiency ratio’’. Arthroscopy

2013;29:1230-4. http://dx.doi.org/10.1016/j.arthro.2013.03.079

25. Park MC, Cadet ER, Levine WN, Bigliani LU, Ahmad CS. Tendon-to-

bone pressure distributions at a repaired rotator cuff footprint using

transosseous suture and suture anchor fixation techniques. Am J Sports

Med 2005;33:1154-9. http://dx.doi.org/10.1177/0363546504273053

26. Park MC, ElAttrache NS, Ahmad CS, Tibone JE. ‘‘Transosseous-

equivalent’’ rotator cuff repair technique. Arthroscopy 2006;22(1360):

e1-5. http://dx.doi.org/10.1016/j.arthro.2006.07.017

27. Park MC, ElAttrache NS, Tibone JE, Ahmad CS, Jun BJ, Lee TQ. Part

I: footprint contact characteristics for an arthroscopic transosseous-

equivalent rotator cuff repair technique. J Shoulder Elbow Surg

2007;16:461-8. http://dx.doi.org/10.1016/j.jse.2006.09.010

28. Park MC, Idjadi JA, ElAttrache NS, Tibone JE, McGarry MH,

Lee TQ. The effect of dynamic external rotation comparing 2

footprint-restoring rotator cuff repair techniques. Am J Sports Med

2008;36:893-900. http://dx.doi.org/10.1177/0363546507313092

29. Park MC, Jun BJ, Park CJ, Ahmad CS, ElAttrache NS, Lee TQ.

The biomechanical effects of dynamic external rotation on rotator cuff

repair compared to testing with the humerus fixed. Am J Sports Med

2007;35:1931-9. http://dx.doi.org/10.1177/0363546507304139

30. Park MC, Peterson A, Patton J, McGarry MH, Park CJ, Lee TQ.

Biomechanical effects of a 2 suture-pass medial inter-implant mattress

on transosseous-equivalent rotator cuff repair and considerations for a

‘‘technical efficiency ratio’’. J Shoulder Elbow Surg 2013;23:361-8.

http://dx.doi.org/10.1016/j.jse.2013.06.019

31. Park MC, Pirolo JM, Park CJ, McGarry MH, Tibone JE, Lee TQ. The

effect of abduction and rotation on footprint contact for single-row,

double-row, and transosseous-equivalent rotator cuff repair tech-

niques. Am J Sports Med 2009;37:1599-608. http://dx.doi.org/10.

1177/0363546509332506

32. Park MC, Tibone JE, ElAttrache NS, Ahmad CS, Jun BJ, Lee TQ. Part

II: biomechanical assessment for a footprint-restoring arthroscopic

transosseous-equivalent rotator cuff repair technique compared to a

double-row technique. J Shoulder Elbow Surg 2007;16:469-76. http://

dx.doi.org/10.1016/j.jse.2006.09.011

33. Pauly S, Fiebig D, Kieser B, Albrecht B, Schill A, Scheibel M.

Biomechanical comparison of four double-row speed-bridging rotator

cuff repair techniques with or without medial or lateral row

enhancement. Knee Surg Sports Traumatol Arthrosc 2011;19:2090-7.

http://dx.doi.org/10.1007/s00167-011-1517-x

34. Reilly P, Bull AMJ, Amis AA, Wallace AL, Richards A, Hill AM, et al.

Passive tension and gap formation of rotator cuff repairs. J Shoulder

ElbowSurg 2004;13:664-7. http://dx.doi.org/10.1016/j.jse.2004.03.011

35. Roh MS, Wang VM, April EW, Pollock RG, Bigliani LU, Flatow EL.

Anterior and posterior musculotendinous anatomy of the supra-

spinatus. J Shoulder Elbow Surg 2000;9:436-40.

36. Sethi PM, Noonan BC, Cunningham J, Shreck E, Miller S. Repair

results of 2-tendon rotator cuff tears utilizing the transosseous

equivalent technique. J Shoulder Elbow Surg 2010;19:1210-7. http://

dx.doi.org/10.1016/j.jse.2010.03.018

37. Takeda H, Urata S, Matsuura M, Nakayama A, Yonemitsu H. The

influence of medial reattachment of the torn cuff for retracted rotator

cuff tears. J Shoulder Elbow Surg 2007;16:316-20. http://dx.doi.org/

10.1016/j.jse.2006.10.001

38. Vaishnav S, Millett PJ. Arthroscopic rotator cuff repair: scientific

rationale, surgical technique, and early clinical and functional results

of a knotless self-reinforcing double-row rotator cuff repair system. J

Shoulder Elbow Surg 2010;19:83-90. http://dx.doi.org/10.1016/j.jse.

2009.12.012

39. Voigt C, Bosse C, Vosshenrich R, Schulz AP, Lill H. Arthroscopic

supraspinatus tendon repairwith suture-bridging technique.AmJ Sports

Med 2010;38:983-91. http://dx.doi.org/10.1177/0363546509359063

40. Yamamoto N, Itoi E, Tuoheti Y, Seki N, Abe H, Minagawa H, et al.

Glenohumeral joint motion after medial shift of the attachment site of

the supraspinatus tendon: a cadaveric study. J Shoulder Elbow Surg

2007;16:373-8. http://dx.doi.org/10.1016/j.jse.2006.06.016