Primary and Secondary Consequences of Rotator Cuff

Home , Rotator cuff tear

Hafizur Rahman Department of Mechanical Science and Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801 Primary and Secondary e-mail: [email protected] Consequences of Rotator Cuff Eric Currier Department of Mechanical Science and Engineering, Injury on Joint Stabilizing University of Illinois at Urbana-Champaign, Urbana, IL 61801 Tissues in the Shoulder e-mail: [email protected] Rotator cuff tears (RCTs) are one of the primary causes of shoulder pain and dysfunction Marshall Johnson in the upper extremity accounting over 4.5 million physician visits per year with 250,000 Department of Mechanical Engineering, rotator cuff repairs being performed annually in the U.S. While the tear is often consid- Georgia Institute of Technology, ered an injury to a specific tendon/tendons and consequently treated as such, there are Atlanta, GA 30332 secondary effects of RCTs that may have significant consequences for shoulder function. e-mail: [email protected] Specifically, RCTs have been shown to affect the joint cartilage, bone, the ligaments, as well as the remaining intact tendons of the shoulder joint. Injuries associated with the Rick Goding upper extremities account for the largest percent of workplace injuries. Unfortunately, Department of Orthopaedic, the variable success rate related to RCTs motivates the need for a better understanding Joint Preservation Institute of Iowa, of the biomechanical consequences associated with the shoulder injuries. Understanding West Des Moines, IA 50266 the timing of the injury and the secondary anatomic consequences that are likely to have e-mail: [email protected] occurred are also of great importance in treatment planning because the approach to the treatment algorithm is influenced by the functional and anatomic state of the rotator cuff Amy Wagoner Johnson and the shoulder complex in general. In this review, we summarized the contribution of Department of Mechanical Science RCTs to joint stability in terms of both primary (injured tendon) and secondary (remain- and Engineering, ing tissues) consequences including anatomic changes in the tissues surrounding the University of Illinois at Urbana-Champaign, affected tendon/tendons. The mechanical basis of normal shoulder joint function depends Urbana, IL 61801 on the balance between active muscle forces and passive stabilization from the joint e-mail: [email protected] surfaces, capsular ligaments, and labrum. Evaluating the role of all tissues working 1 together as a system for maintaining joint stability during function is important to under- Mariana E. Kersh stand the effects of RCT, specifically in the working population, and may provide insight Department of Mechanical Science into root causes of shoulder injury. [DOI: 10.1115/1.4037917] and Engineering, University of Illinois at Urbana-Champaign, Urbana, IL 61801 e-mail: [email protected]

1 Shoulder Function and Rotator Cuff Tears sports to occupational, and account for more work-related injuries (31%) than any other body region [4]. Among the shoulder inju- 1.1 The Shoulder Complex. The shoulder is among the most ries, rotator cuff tears (RCTs) warrant specific attention because mobile joints in the body allowing for significant range of motion of the high incidence among workers and the variable success rate in multiple planes. The shoulder complex is made of the scapula, of repairs. Rotator cuff problems account for over 4.5 million phy- clavicle, humerus, and the soft tissues that span the joint including sician visits per year [5], and rotator cuff repair is one of the most cartilage, capsular ligaments, the labrum, and surrounding common surgeries performed on the shoulder with 250,000 sur- muscle-tendon units (Fig. 1(a))[1,2]. The articulations in the geries performed annually in the U.S. [6,7]. Unfortunately, the shoulder complex include the glenohumeral joint, scapulothoracic success rate of rotator cuff repair is variable with many resulting articulation, and the acromioclavicular joint. These tissues work in retears. Revision surgeries can be as high as 30% for isolated in unison to complete a wide range of kinematic tasks. supraspinatus tendon tears [8]. Surprisingly, there is a dispropor- Often modeled as a ball-and-socket joint, the three shoulder tionately low amount of published research with regard to work- rotational degrees-of-freedom include flexion and extension, related injuries of the shoulder and rotator cuff tears (Fig. 2). This abduction and adduction, and internal and external rotation. paucity of data and the current revision rate suggests that the rela- Abduction and flexion account for the largest ranges of motion tionship between the injury mechanism, repair, and rehabilitation (170 6 10.8 deg and 164 6 10.2 deg, respectively) compared to with respect to rotator cuff tears is not well understood. extension (81 6 11.3 deg). The shoulder joint can rotate more A rotator cuff tear is described as a tear of one or more of the internally (86 6 4.6 deg) than externally (67 6 11.3 deg) [3]. rotator cuff tendons (supraspinatus, infraspinatus, teres minor, and subscapularis (Figs. 1(b) and 1(c))[2]) and is classified by the size 1.2 Rotator Cuff Tear and Treatment. Injuries to the upper of the tear. A full thickness tear indicates a through-thickness tear extremities can occur as a result of a wide range of activities from of the shoulder (Fig. 1(e)) while a partial thickness tear is described as the fraying of the tendon–bone connection (Fig. 1(f)) and can lead to a full tear if not treated properly [9–11]. RCTs 1Corresponding author. Manuscript received May 13, 2017; final manuscript received September 13, cause pain, depending on the severity of the tear, and can lead to 2017; published online September 29, 2017. Assoc. Editor: Kyle Allen. limited function in the affected shoulder, especially during

Journal of Biomechanical Engineering Copyright VC 2017 by ASME NOVEMBER 2017, Vol. 139 / 110801-1

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 10/14/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use Fig. 1 (a) Shoulder joint including bones, cartilage, and capsule, (b) posterior, and (c) anterior view of muscles that span the shoulder with rotator cuff muscles highlighted, (d) ligaments of the shoulder joint with coracoacromial, coracohumeral, and glenohumeral ligaments highlighted, (e) a full thickness tear in supraspinatus tendon, and (f) a partial thickness tear in supraspinatus tendon [1,2]

overhead activities [11,12]. Other symptoms include, but are not effect on the balance of muscle forces at the shoulder joint [19] limited to weakness, tenderness, and snapping sounds coming and are the subject of Sec. 2 of this review. In Sec. 3, we com- from the joint [11]. Individuals with RCTs have also reported dif- pared longitudinal changes in the mechanical properties of the ﬁculty sleeping on the effected side [12]. In contrast, RCTs can bone–tendon interface as a result of RCT. Next in Sec. 4, we eval- also be asymptomatic with little to no clinical symptoms [13]. uated the secondary consequences of rotator cuff tears on the

1.3 More Than Muscle: Evaluating the Consequences of RCTs on the Shoulder Complex. Due to the asymptomatic nature of many RCTs, it is not possible to know how many tears go unreported; however, it has been suggested that symptomatic RCTs accounted for 34.7% of all tears and asymptomatic tears for 65.3% [14]. While the tear is often considered to be an injury to the tendons, and is consequently treated as such, there has been evidence in the literature that the RCTs may have signiﬁcant effects on the remaining surrounding tissues. The mechanical basis of normal shoulder joint function depends on the balance between active muscle forces and passive stabilization from the joint surfaces, capsular ligaments, and labrum. Understanding the effect of rotator cuff tears on the mechanics of both the injured tendon and surrounding tissues is important for connecting and translating the results that arise from studies of in vivo shoulder kinematics, cadaveric studies using simulators, or in vivo muscle volume studies [15–18]. We suggest that an improved comprehen- sion of the mechanisms underlying shoulder function, before and Fig. 2 Number of articles found during the Pubmed search. after injury, can lead to improved diagnosis and treatment. For Pubmed search, we used the keywords as “A” and “B”, Therefore, we aimed to summarize the mechanical consequen- where “A” indicates either “Shoulder injury” or “RCT.” “B” indi- ces of rotator cuff tears on both the injured tendons and the sur- cates “Sports” or “Occupational” or “Work.” Graph shows that rounding tissues. Speciﬁcally, tears in the supraspinatus and there was higher number of papers published for “Sports” com- infraspinatus rotator cuff tendons have an immediate primary pared to work-related injuries.

110801-2 / Vol. 139, NOVEMBER 2017 Transactions of the ASME

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 10/14/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use mechanical properties of uninjured tendons, ligaments, and carti- bear. However, after 20 days of that detachment, modulus values lage. Finally, we identify opportunities for further study that may returned to pre-injury values. lead to provide better outcomes of rotator cuff surgeries. This trend was different when both supraspinatus and infraspinatus were injured [21]. Interestingly, supraspinatus area increased (33%) after 56 days of detachment and remained higher 2 Primary Effects of Rotator Cuff Tears (26%) compared to the control at 112 days (Fig. 3(c))[21]. How- We deﬁne the primary effects of rotator cuff tears as changes in ever, the modulus of elasticity of the supraspinatus tendon did not the mechanical or structural properties of the torn tendon. Rotator change for any time period following for multitendon tears (Fig. cuff tears predominantly occur in the supraspinatus tendon [12]. 3(d)), in contrast to the results of the single tear. Therefore, the Using rodent models, the elastic modulus of the supraspinatus mechanical change in supraspinatus seems to be dependent on decreased by (72%) after 14 days of detachment, and was likely whether or not it alone is torn, or whether there are multiple ten- associated with the increased area (200%) observed at the same don tears present. time period (Figs. 3(a) and 3(b))[20]. The thickening of the While the supraspinatus tends to be the most common tendon remaining tendons after injury is the physiological adaptive torn, the infraspinatus was more sensitive to multitendon tears response of the remaining tendons to the increased load that they than supraspinatus. When both infraspinatus and supraspinatus

Fig. 3 Change in supraspinatus tendon (a) area and (b) modulus of elasticity over time following its injury in rat (n 5 10 for each data point). Change in supraspinatus tendon (c) area and (d) modulus of elasticity over time following both supraspinatus and infraspinatus injuries (n 5 12 for each data point). Change in infraspinatus (e) area and (f) modulus of elasticity over time following both supraspinatus and infraspinatus injury (n 5 12 for each data point). The X-axis represents the time after injury. The Y-axis represents the properties. Closed and open symbols represent data for the control and injured tendons, respectively. * indicates statistically signiﬁcant difference between control (uninjured) and injured tendon [20,21].

Journal of Biomechanical Engineering NOVEMBER 2017, Vol. 139 / 110801-3

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 10/14/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use were torn, the modulus of elasticity of the infraspinatus tendon influence of age on immediate repair of supraspinatus tendon in changed: modulus of elasticity decreased at 28 days and increased rats compared to control (uninjured) conditions [24]. For the older at 112 days, while stiffness increased only after 112 days (Figs. group (24 months of age), failure stress and peak failure load sig- 3(e) and 3(f))[21]. The rationale for why infraspinatus is more nificantly decreased in the immediately repaired tendon at both 14 sensitive to multitendon tears than supraspinatus is not clear and and 56 days after repair compared to the control. In contrast, for a remains a point for further investigation. younger group (8 months of age), failure stress and peak failure Experiments have also been conducted to investigate the effects load of the immediate repair group only decreased after 14 days of tendon repairs on tendon mechanical properties [6,22–24]. The of repair. Therefore, properties of the repaired tendon depend not supraspinatus tendon in rabbit was repaired immediately after only on the timing of the surgical repair but also on the age. Ear- detachment, but both stiffness and peak load decreased after 7 lier repair providing better mechanical properties may lower the days compared to the uninjured supraspinatus tendon (52% for risk of the tendon retear. Finally, results also showed that aging stiffness and 60% for peak load) [22]. Another study compared has a negative impact on the healing of the tendons. different repair time periods (1, 2, 3 months delay repair) with Other species like canine and ovine also showed the change in control (uninjured) in rabbits [23]. Results showed that if the tendon properties after tears [25,26]. Studies have investigated the injured supraspinatus tendon was repaired after 1 month of disrup- change in the properties of the infraspinatus muscle and tendon tion, stiffness increased (19%) relative to control. However, if after disruption [25,26]. The stiffness and modulus of elasticity repaired after 2 or 3 months of disruption, stiffness did not change significantly increased in detached infraspinatus muscle after 84 paradoxically suggesting that waiting to repair the tendon restores days of detachment compared to uninjured muscle in canine [25]. pre-injury stiffness levels. Similarly, the modulus of elasticity of the infraspinatus tendon Galatz et al. compared the effect of immediate and delayed increased after 42 days and 126 days of detachment compared to repair (repaired after 3 weeks of detachment) in rat supraspinatus the uninjured tendon in ovine (60% for 42 days, and 70% for 126 tendon [6]. Area increased (46%) for delayed repair compared to days) [26]. immediate repair when measured after 28 days of repair. Maxi- Different animal species including rat, rabbit, canine, and sheep mum stress decreased (80%) for the delayed repair group while have been used to evaluate the effects of RCTs. However, Chaud- measured after 10 days of repair. Plate et al. also investigated the hury et al. used human biopsy samples to measure the effects of

Fig. 4 Change in insertion: (a) stiffness, (b) area, and (c) modulus of elasticity over time after operation. The X- axis represents the time after injury and repair. Y-axis represents the properties. Each unique symbol denotes a different study. All studies have been done using rodents and n indicates the number of the rodents used in different cases. Closed symbols represent the control data for each study. Studies 1, 2, 3, 4, and 5 represent Refs. [35], [32], [37], [34], and [36], respectively.

110801-4 / Vol. 139, NOVEMBER 2017 Transactions of the ASME

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 10/14/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use RCTs [27]. The storage modulus, calculated by dynamic shear tendons to injured and repaired supraspinatus tendons. The analysis, showed that the torn tendons had significantly lower repaired tendons were tested after several weeks to observe the modulus (20%) compared to normal tendons. effect of time on the healing of the insertion site. Studies also con- In summary, the increment or decrement of properties of the sidered the effect of different activity levels on the healing process injured tendons and muscles depends not only on the number of [32,34–37]. While the measured properties varied with time and tears but also on the time period after injury. Even after repair, the occasionally among activity levels, most properties changed from tendons may experience changes in properties compared to con- their corresponding control values (Fig. 4)[32,34–37]. It is impor- trol. These changes in tendon properties can further affect the bal- tant to note that the control measurements of Manning et al. [37] ance between muscle forces and passive stabilization and can lead differ from the control measurements of Thomopoulos et al. [34] to shoulder joint instability and abnormal joint kinematics. and Gimbel et al. [35] even though test methods were similar. The studies showed that the quality of the healing tissue differs for the 3 Effects of Rotator Cuff Tears on Tendon-Bone normal insertion site even after an extended test period. This is consistent with other studies that show that the normal four-zone Interface insertion site does not reform once damaged [6,28,34,38–49]. All The tendon to bone insertion site consists of functionally graded insertion studies that are reviewed in this paper used rat shoulders tissue whose function is to transfer load between the hard bone as test specimens. While rat shoulders have anatomy and repair and soft tendon. Without this transitional area, high stress concen- procedures comparable to the human shoulder, the conclusions trations would form at the interface of these two materials, leading made in these studies cannot be directly applied to the humans to an increased potential for failure [28–31]. The insertion site is [34,50]. Another limitation of these studies is that the supraspina- divided into four zones: tendon, fibrocartilage, mineralized fibro- tus tendons were “detached” or “transected” and not torn by an cartilage, and bone [32]. Each zone contributes to the overall gra- acute traumatic or chronic degenerative process. These tendons dient in cell phenotype, tissue organization, tissue composition, were completely separated from the humeral head while many and tissue mechanical properties [33]. human patients experience only partial tears. The studies on mechanical properties of insertion site that are These studies of the insertion site measured apparent properties reviewed in this paper compared normal, healthy supraspinatus and do not account for changes in properties along the tendon to

Fig. 5 Change in (a) infraspinatus and subscapularis properties due to supraspinatus tendon detachment (b) subscapularis properties after supraspinatus and infraspinatus tendons detachment, and (c) infraspinatus properties and supraspinatus and subscapularis tendons detachment. Properties are expressed as the percent change after the detachment compared to control (uninjured). 4 wks and 8 wks indicate properties measured at 4 weeks and 8 weeks after injury. The squiggly lines near the tendon insertion site represent tendon detachment. Proper- ties above and below the solid lines indicate percent increase and percent decrease from control, respectively. “nsd” indicates no signiﬁcant differences from control for that property [51].

Journal of Biomechanical Engineering NOVEMBER 2017, Vol. 139 / 110801-5

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 10/14/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use bone insertion site. The insertion site was categorized as bone or that the elastic modulus decreased in the midsubstance of the tendon compartments, and only viscoelastic measurements of the lower subscapularis and upper subscapularis, and insertion site of intact insertion site were considered [28]. While the current stud- lower subscapularis for overuse activity group [53]. Also, elastic ies focus on the properties of the insertion site after repair, no modulus increased in the insertion site of upper subscapularis. studies have tested the damaged insertion site. Due to insufficient However, area did not change for any region of these tendons. In knowledge about the natural healing process of the insertion site, contrast, for a single tear, no changes were observed for both little can be done to regenerate the normal tissue [32]. Therefore, modulus and area as mentioned earlier. Therefore, elastic moduli it is necessary to investigate the development of the normal four- are affected by overuse activity if the infraspinatus and supraspi- zone insertion site. natus are both detached. Comparisons were also performed between single tear (supraspinatus) and multitendon tears (supraspinatus and infraspinatus) 4 Secondary Effects of Rotator Cuff Tears to measure the contributions of the additional tear compared to The stability of the glenohumeral joint depends on the balance the single tear [54]. No area changes were observed for lower sub- between static and dynamic structures including the glenoid artic- scapularis and upper subscapularis for multitendon tears com- ular cartilage, glenoid labrum, ligaments, joint capsule, osseous pared to single tendon tear. Elastic modulus decreased for the structures, rotator cuff muscles, and other muscle structures sur- midsubstance region of the lower subscapularis and upper subsca- rounding the shoulder joint. In healthy shoulders, these structures pularis. However, the modulus increased for the insertion region allow for concentric rotation of the humeral head on the glenoid only in the upper subscapularis [54]. surface. However, the loss of muscle force due to RCTs likely leads to glenohumeral joint instability, and the articular surfaces 4.1.3 Multitendon Tear: Supraspinatus and Subscapularis. are exposed to abnormal joint mechanics. Therefore, in addition The structural and mechanical properties in the infraspinatus to the primary injured tendon, RCTs also have secondary effects changed due to tears in the supraspinatus and the subscapularis on the remaining intact tendons, cartilage, and ligaments of the (Fig. 5(c))[51]. For the infraspinatus tendon, the area increased shoulder. after 4 weeks and 8 weeks in a similar fashion as single tear. (For 4 weeks, 11% in double tear compared to 16% in single tear; for 8 weeks, 35% in double tear compared to 37% in single tear). Mod- 4.1 Intact Rotator Cuff Tendons. Due to the dependence ulus decreased and percent relaxation increased after 8 weeks of between glenohumeral joint structures, tears in any of rotator cuff detachment (for modulus, 22% in double tear relative to 26% in tendons will eventually affect the properties of other surrounding single tear; for percent relaxation, 14% for double tear relative to intact rotator cuff tendons. Several studies have shown that the 13% in single tear). Stiffness, peak load and equilibrium load mechanical properties of the infraspinatus and subscapularis ten- were not affected by multitendon tears, as was also reported for dons change due to tears in surrounding rotator cuff tendons. single tear. It is interesting that the degree to which these proper- Properties of the intact rotator cuff tendons were measured for ties are altered for multitendon tears are similar to single tear, and both control (uninjured, i.e., no tears in surrounding rotator cuff suggests that detachment of subscapularis in addition to supraspi- tendons) and injured tendons (at least one of surrounding rotator natus would not further change the infraspinatus properties. In cuff tendons is torn). We calculated the percent change of intact summary, the area of the intact tendons increased and modulus tendon properties after tearing in surrounding tendons relative to decreased after tears in surrounding tendons. the control condition and summarized these findings graphically in Fig. 5. 4.1.4 Comparison Between Single Tear and Multitendon Tears. While comparing between single tear and multitears, the 4.1.1 Single Tear of Supraspinatus. Detachment of the supra- area increased and modulus of elasticity decreased for infraspina- spinatus tendon caused changes in both structural and mechanical tus and subscapularis irrespective of the number of the tears. properties of the infraspinatus and subscapularis tendons (Fig. However, the degree to which these properties would change 5(a))[51]. The area of infraspinatus and subscapularis tendons depends on the number of the tears. For example, changes in prop- increased after 4 weeks and 8 weeks of detachment. However, the erties in subscapularis were more for multitears compared to sin- elastic modulus decreased and the percent relaxation increased gle tear. In contrast, the change in properties was identical for only after 8 weeks of detachment. No changes were observed for single tear and multitears in infraspinatus. Furthermore, different peak load and equilibrium load [51]. degrees of loadings have significant contributions on how these Experiments have been performed to compare between normal properties are changed. cage activity and overuse activity of rats due to supraspinatus tear. After supraspinatus tendon detachment, area and modulus did not change for the infraspinatus and subscapularis [52]. Although 4.2 Cartilage. The effect of supraspinatus and infraspinatus supraspinatus is the most frequent torn tendon among rotator tears on cartilage thickness and elastic modulus has been eval- cuffs, there are only two studies that measured the biomechanical uated using rat models. Glenoid cartilage thickness decreased in effects of supraspinatus torn tendons on surrounding intact rotator the antero-inferior region after the detachment of the supraspina- cuff tendons [51,52]. tus and the infraspinatus tendons while the elastic modulus decreased over a larger region of the glenoid (Fig. 6)[55]. Reuther 4.1.2 Multitendon Tear: Supraspinatus and Infraspinatus. et al. reported that after 8 weeks of supraspinatus tendon detach- Multitendon tears in both supraspinatus and infraspinatus also ment, within the overuse activity rat group, the equilibrium modu- affected the properties of subscapularis (Fig. 5(b))[51], but to dif- lus increased significantly in antero-inferior and superior regions ferent degrees compared to the single-tendon tear. The initial (4 of glenoid cartilage compared to normal cage activity group [52]. week) increase in area of the subscapularis was nearly twice as But no change in thickness was observed in any regions. How- high in the presence of both supraspinatus and infraspinatus tears ever, for detachment of supraspinatus and infraspinatus tendons compared to a supraspinatus tear only. By 8 weeks, the change together, the cartilage modulus decreased in the center and was less profound (68% in double tear compared to 50% in single posterior–superior regions in the overuse activity group [53]. tear). The modulus decrease in the double tear was nearly identi- The detachment of the biceps tendon in addition to multitendon cal to the single tear scenario. Stiffness only decreased after 8 tears (supraspinatus and infraspinatus) also reduced glenoid carti- weeks and no changes were seen for percent relaxation, peak load, lage thickness in the anterior–inferior region, with no change in and equilibrium load [51]. elastic modulus [56]. However, the elastic modulus decreased in Comparing between normal cage and overuse activity in rats the center of the glenoid for multitendon tears (supraspinatus and with multitendon tears (supraspinatus and infraspinatus) showed infraspinatus) compared to supraspinatus tendon tear [54]. Current

110801-6 / Vol. 139, NOVEMBER 2017 Transactions of the ASME

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 10/14/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use Table 1 Changes in ligament properties after RCT [63,64]

Age (years)

Ligament Measured property <60 61–87 >60

Length 17.68% Ø 21.43% Coracoacromial (medial band) Area, width, thickness Ø Ø Ø

Length 14.91% 21.76% 16.41% Area Ø þ50% þ36.67% Width, Thickness, Stiffness Ø Ø Ø Coracoacromial (lateral band) Failure Load and Displacement, Ø Ø Ø Total Failure Strain Ø Ø Ø Failure stress Ø 46.06% 32.06% Ligamentous failure strain — Ø — Total modulus — 58.74% — Ligamentous modulus — 43.94% —

decreased signiﬁcantly, þ increased signiﬁcantly.

studies only show the effect of rotator cuff injuries on glenoid car- the CAL properties changed for RCT for different age groups tilage mechanics for rats, no properties data available for human (Table 1)[63,64]. The thickness, width, and area of the medial glenohumeral cartilage. band of the CAL did not change for any age groups after RCT; Previous studies have suggested that there is a correlation however, the length of the medial band decreased 18% in younger between the glenohumeral joint arthrosis and tear/atrophy of the subjects. rotator cuff [57–61]. However, the exact pathogenesis of how the Whether or not the medial band length differs in older subjects articular cartilage degeneration occurs is not well understood. after RCT is less clear: one study reported decreased length [64] Therefore, further investigation is necessary to understand the car- while another showed no change in length [63]. The length and tilage degeneration process during the progression of RCTs. area of the lateral band of the CAL have been shown to be sensitive to RCTs. Similar to the medial band, the length of the lateral band consistently decreased after RCT. In contrast to the medial 4.3 Ligament. There are four ligaments surrounding the gle- band, area increased after RCT in subjects over 60 years of age. nohumeral joint: coracohumeral, superior glenohumeral, middle For older groups, the total and ligamentous modulus were signifi- glenohumeral, and inferior glenohumeral (Fig. 1(d))[2,62]. These cantly lowered after RCT [63]. There were no changes in width, ligaments are characterized as a thickening of the glenohumeral thickness, stiffness, failure load/displacement, total or ligamen- capsule. Yet, changes in mechanical and geometric properties due tous failure strain after RCT. to RCTs are only available for the coracoacromial ligament There are five types of CAL for the human shoulder joint: Y- (CAL), which extends between the coracoid process and acromion shaped (two bands), broad-band (one band), quadrangular (one of the scapula. No studies have evaluated the effects of RCTs on band), V-shaped (two bands), and multiple-banded [65]. Compar- the mechanical properties of superior, middle, and inferior gleno- ing these five types with RCT, no statistical relations were found humeral ligaments. Human cadaveric studies have reported how between types or geometric measurement of these CAL and RCT. However, if these five types were divided into two groups based on bundle numbers like unique bundle (broad-band and quadrangular) and more than one bundle (Y-shaped, V-shaped, and multiple-banded), CAL with more than one bundle showed a significant association with RCT with a longer lateral border and larger coracoid insertion [65]. In addition to mechanical testing, several other methods have been used to measure the elasticity of ligaments. Kijima et al. measured the strain ratio, defined as the ratio of strain of CAL to that of the RCT, as an index of the elasticity of CAL by ultrasound elastography. Here, the higher strain ratio indicates that the CAL is softer. The strain ratio of CAL with RCT (23.75 6 15.05) was higher than that of the older ligaments without rotator cuff tear (12.62 6 7.94) suggesting that the CAL softens in the presence of RCT [66]. Scanning acoustic microscopy has also been used to measure the speed of sound through the CAL, which is directly proportional to the square root of the Young’s modulus [67]. Using this method, the modulus of the CAL in those with the RCT was higher than without the RCT group.

5 Conclusions 5.1 Risk Factors Associated With RCTs. While RCTs can occur during traumatic events such as a fall or an accident, most Fig. 6 Thickness and elastic modulus change in glenoid cartilage after RCT. nsd indicates no statistically signiﬁcant differ- tears develop gradually [68]. The risk of rotator cuff tears ence after multitendon tears (supraspinatus and infraspinatus) increases with age [69–71], and more than half of the population compared to the control (uninjured). # indicates statistically sig- over 60 years of age is currently living with some degree of rota- niﬁcant decrease after multitendon tears compared to the con- tor cuff injury [71]. Arm dominance and gender have not been trol [55]. correlated with risk of RCT [13]; however, women have worse

Journal of Biomechanical Engineering NOVEMBER 2017, Vol. 139 / 110801-7

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 10/14/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use Table 2 Quantitative summary of odds ratios associated with tissue is important both from a clinical perspective and, in the risk factors shown in Fig. 7 (N 5 number of studies, cases of a Worker’s Compensation claim, the medical legal per- SD 5 standard deviation, SE 5 standard error) [73] spective. Localizing the time that the injury occurred can have the obvious benefit of elucidating causality in a Worker’s Compensa- Risk factor N Mean SD SE of mean tion claim. Repetition 10 4.22 2.32 0.73 Duration 7 4.01 3.43 1.30 Vibration 13 2.46 0.92 0.25 5.2 Improving the Treatment of RCTs. Rotator cuff tears Force 10 3.29 1.01 0.32 are one of the most common injuries affecting the upper extremity Posture 20 3.19 1.34 0.30 with variable success rate in repair surgeries. Animal species (rat, Heavy physical work 10 2.96 1.25 0.39 rabbit, sheep, canine) are considered as the most appropriate models and have been used in understanding the consequences of RCTs. However, none of them can exactly replicate the human rotator cuff repair outcomes over time than men based on postoper- shoulder. Although studies on the effect of RCTs have made con- ative pain and level of abduction [69]. Individuals who have a his- siderable progress in the last few decades providing the biome- tory of cigarette smoking or have higher body mass index have also chanical alterations in animals; still there are some limitations, been shown to be at increased risk for RCT [72]. Athletes who per- specifically, how the properties of the soft tissues (intact tendons, form high stress repetitive motions at the shoulder are more prone ligaments, and cartilages) will be altered for humans. to developing RCTs [68], but majority of injuries occur in the lay Many studies focus on RCTs from the point of injury forward population and often stem from work-related injuries. The distinc- (diagnosis, classifications, repair, etc.), and it is widely appreci- tion between sport and occupational tears becomes blurred when ated that the degree of return to pre-injury functional activities is the mechanism behind the tears is considered. related to both the rehabilitation protocol as well as the treatment A number of workplace risk factors have been identified for intervention [74]. Ideally, postintervention decisions are made in RCTs including large and sudden forces, heavy lifting, extensive concert between (1) the surgeon who initially prescribes which overhead activity, repetitive or long duration actions, and vibra- activities of daily living the patient can resume and (2) the physi- tion. A recent review paper [73] compiled an extensive list of cal therapist prescription of rehabilitation protocols. However, reported odds ratios and confidence intervals associated with rota- these recommendations are variable with respect to timing and tor cuff tears as a function of activity types. We extracted the odds progression [34,75–77], and in many cases can be based on clini- ratios and grouped them according to the following risk factors: cal experience rather than quantitative data [74]. repetition, duration, vibration, force, posture, and heavy physical The successful repair of RCTs requires understanding how the work (Fig. 7)[73]. A one-way analysis of variance with Tukey primary and secondary consequences might affect the shoulder posthoc analysis was conducted to determine if there was a statis- joint stability and function. Furthering the understanding of the tically significant difference in the risk factors (a ¼ 0.05) (Origin- tissue biomechanics of the shoulder girdle is essential to improv- Pro 2015, OriginLab Corporation, Northampton, MA). Repetition ing the care delivered to patients. The natural history of shoulder had the highest odds ratio (4.22) while vibration had the lowest pathology is very complex as changes in the anatomy and average odds ratio (2.46) (Table 2)[73]. There was a wide range dynamic stability of the shoulder, as a result of specific patholo- of odds ratios within any given risk factor, and comparisons of the gies, are an area where there is disagreement and at times contro- means across risk factors did not reveal significant differences versy within the clinical community. For instance, the role of the between risk factors. Differences in study methodologies and var- long head of Biceps in humeral head position is not well defined. iations in risk factors (such as high vibration versus low vibration Understanding this role would have a direct effect on surgical tool use) may partly explain this variability. decision making, such as whether performing a tenotomy during Limited studies have aimed to understand rotator cuff tears in rotator cuff repair is warranted. the populations that experience tears most: occupational workers Recently, the reconstruction of the superior capsule for com- in comparison to athletes, the latter of which have a unique anat- plete rotator cuff tear has received more attention. However, the omy and physiology. Understanding the compensatory effects of biomechanics of this surgery are not well defined and would assist injury to a rotator cuff tendon/tendons on the remaining uninjured in understanding the surgery further. Surgical planning for the shoulder is highly dependent on understanding the static and dynamic biomechanical issues in the normal and pathologic shoulder. Finally, in terms of rehabilitation, some patients with massive rotator cuff tears do well with treatment by physical ther- apy alone while others go on to a pseudoparalysis and have a nearly nonfunctional shoulder. These contrasting outcomes indicate that in some cases, the remaining tissues in the shoulder complex can work toward stabilizing the joint but the mechanisms behind this are not understood. Understanding the biomechanics including adaptive reconfiguration of the remaining soft tissues in this situation would be helpful in identifying those patients who may benefit most from specific postsurgery rehabilitation. Identifying changes in the mechanics after RCTs with development in tissue engineering will allow researchers to improve the current surgeries available to treat rotator cuff tears. Future research should aim to elucidate the effects of RCTs on human tissues to better the understanding of RCTs on human shoulder function and develop better treatment algorithms.

References [1] Blausen.com Staff, 2014, “Medical Gallery of Blausen Medical 2014,” Wiki- Journal Med., 1(2), p. 68. Fig. 7 Distribution of statistically signiﬁcant odds ratios with [2] Gray, H., 1918, Anatomy of the Human Body, 20th ed., Lea & Febiger, Philadel- RCTs as a function of risk factors [73] phia, PA.

110801-8 / Vol. 139, NOVEMBER 2017 Transactions of the ASME

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 10/14/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use [3] Brown, L. P., Niehues, S. L., Harrah, A., Yavorsky, P., and Hirshman, H. P., [29] Thomopoulos, S., Marquez, J. P., Weinberger, B., Birman, V., and Genin, G. 1988, “Upper Extremity Range of Motion and Isokinetic Strength of the Inter- M., 2006, “Collagen Fiber Orientation at the Tendon to Bone Insertion and Its nal and External Shoulder Rotators in Major League Baseball Players,” Am. J. Influence on Stress Concentrations,” J. Biomech., 39(10), pp. 1842–1851. Sports Med., 16(6), pp. 577–585. [30] Genin, G. M., Kent, A., Birman, V., Wopenka, B., Pasteris, J. D., Marquez, P. [4] United States Bone and Joint Initiative, 2014, The Burden of Musculoskeletal J., and Thomopoulos, S., 2009, “Functional Grading of Mineral and Collagen in Diseases in the United States (BMUS), 3rd ed., United States Bone and Joint the Attachment of Tendon to Bone,” Biophys. J., 97(4), pp. 976–985. Initiative, Rosemont, IL. [31] Liu, Y., Birman, V., Chen, C., Thomopoulos, S., and Genin, G. M., 2011, [5] Oh, L. S., Wolf, B. R., Hall, M. P., Levy, B. A., and Marx, R. G., 2007, “Mechanisms of Bimaterial Attachment at the Interface of Tendon to Bone,” “Indications for Rotator Cuff Repair: A Systematic Review,” Clin. Orthop. ASME J. Eng. Mater. Technol., 29(1), pp. 997–1003. Relat. Res., 455(455), pp. 52–63. [32] Galatz, L. M., Sandell, L. J., Rothermich, S. Y., Das, R., Mastny, A., Havlioglu, [6] Galatz, L. M., Rothermich, S. Y., Zaegel, M., Silva, M. J., Havlioglu, N., and N., Silva, M. J., and Thomopoulos, S., 2006, “Characteristics of the Rat Supra- Thomopoulos, S., 2005, “Delayed Repair of Tendon to Bone Injuries Leads to spinatus Tendon During Tendon-to-Bone Healing After Acute Injury,” J. Decreased Biomechanical Properties and Bone Loss,” J. Orthop. Res., 23(6), Orthop. Res., 24(3), pp. 541–550. pp. 1441–1447. [33] Smith, L., Xia, Y., Galatz, L. M., Genin, G. M., and Thomopoulos, S., 2012, [7] Yeranosian, M. G., Terrell, R. D., Wang, J. C., McAllister, D. R., and Petrigliano, “Tissue-Engineering Strategies for the Tendon/Ligament-to-Bone Insertion,” F. A., 2013, “The Costs Associated With the Evaluation of Rotator Cuff Tears Connect. Tissue Res., 53(2), pp. 95–105. Before Surgical Repair,” J. Shoulder Elbow Surg., 22(12), pp. 1662–1666. [34] Thomopoulos, S., Williams, G. R., and Soslowsky, L. J., 2003, “Tendon to [8] Goutallier, D., Postel, J. M., Gleyze, P., Leguilloux, P., and Van Driessche, S., Bone Healing: Differences in Biomechanical, Structural, and Compositional 2003, “Influence of Cuff Muscle Fatty Degeneration on Anatomic and Func- Properties Due to a Range of Activity Levels,” ASME J. Biomech. Eng., tional Outcomes After Simple Suture of Full-Thickness Tears,” J. Shoulder 125(1), pp. 106–113. Elbow Surg., 12(6), pp. 550–554. [35] Gimbel, J. A., Van Kleunen, J. P., Williams, G. R., Thomopoulos, S., and [9] Kumar, R., and Jadhav, U., 2014, “Functional Evaluation of Patient After Soslowsky, L. J., 2007, “Long Durations of Immobilization in the Rat Result in Arthroscopic Repair of Rotator Cuff Tear,” J. Clin. Orthop. Trauma, 5(2), pp. Enhanced Mechanical Properties of the Healing Supraspinatus Tendon Insertion 84–90. Site,” ASME J. Biomech. Eng., 129(3), pp. 400–404. [10] Gumina, S., ed., 2017, Rotator Cuff Tear, 1st ed., Springer International [36] Peltz, C. D., Dourte, L. M., Kuntz, A. F., Sarver, J. J., Kim, S. Y., Williams, G. Publishing, Cham, Switzerland. R., and Soslowsky, L. J., 2009, “The Effect of Postoperative Passive Motion on [11] Pandey, V., and Jaap Willems, W., 2015, “Rotator Cuff Tear: A Detailed Rotator Cuff Healing in a Rat Model,” J. Bone Jt. Surg. Am., 91(10), Update,” Asia-Pac. J. Sports Med. Arthroscopy Rehabil. Technol., 2(1), pp. 2421–2429. pp. 1–14. [37] Manning, C. N., Kim, H. M., Sakiyama-Elbert, S., Galatz, L. M., Havlioglu, N., [12] Smith, K. L., Harryman, D. T., II, Antoniou, J., Campbell, B., Sidles, J. A., and and Thomopoulos, S., 2011, “Sustained Delivery of Transforming Growth Fac- Matsen, F. A., III, 2000, “A Prospective, Multipractice Study of Shoulder tor Beta Three Enhances Tendon-to-Bone Healing in a Rat Model,” J. Orthop. Function and Health Status in Patients With Documented Rotator Cuff Tears,” Res., 29(7), pp. 1099–1105. J. Shoulder Elbow Surg., 9(5), pp. 395–402. [38] Aoki, M., Oguma, H., Fukushima, S., Ishii, S., Ohtani, S., and Murakami, G., [13] Milgrom, C., Schaffler, M., Gilbert, S., and van Holsbeeck, M., 1995, “Rotator- 2001, “Fibrous Connection to Bone After Immediate Repair of the Canine Cuff Changes in Asymptomatic Adults. The Effect of Age, Hand Dominance Infraspinatus: The Most Effective Bony Surface for Tendon Attachment,” J. and Gender,” J. Bone Jt. Surg., 77(2), pp. 296–298. Shoulder Elbow Surg., 10(2), pp. 123–128. [14] Minagawa, H., Yamamoto, N., Abe, H., Fukuda, M., Seki, N., Kikuchi, K., [39] Fujioka, H., Thakur, R., Wang, G. J., Mizuno, K., Balian, G., and Hurwitz, S. Kijima, H., and Itoi, E., 2013, “Prevalence of Symptomatic and Asymptomatic R., 1998, “Comparison of Surgically Attached and Non-Attached Repair of the Rotator Cuff Tears in the General Population: From Mass-Screening in One Rat Achilles Tendon-Bone Interface. Cellular Organization and Type X Colla- Village,” J. Orthop., 10(1), pp. 8–12. gen Expression,” Connect. Tissue Res., 37(3–4), pp. 205–218. [15] Vidt, M. E., Santago, A. C., Marsh, A. P., Hegedus, E. J., Tuohy, C. J., Poeh- [40] Fujioka, H., Wang, G. J., Mizuno, K., Balian, G., and Hurwitz, S. R., 1997, ling, G. G., Freehill, M. T., Miller, M. E., and Saul, K. R., 2016, “The Effects “Changes in the Expression of Type-X Collagen in the Fibrocartilage of Rat of a Rotator Cuff Tear on Activities of Daily Living in Older Adults: A Kine- Achilles Tendon Attachment During Development,” J. Orthop. Res., 15(5), matic Analysis,” J. Biomech., 49(4), pp. 611–617. pp. 675–681. [16] Parsons, I., Apreleva, M., Fu, F. H., and Woo, S. L.-Y., 2002, “The Effect of [41] Rodeo, S. A., Arnoczky, S. P., Torzilli, P. A., Hidaka, C., and Warren, R. F., Rotator Cuff Tears on Reaction Forces at the Glenohumeral Joint,” J. Orthop. 1993, “Tendon-Healing in a Bone Tunnel. A Biomechanical and Histological Res., 20(3), pp. 439–446. Study in the Dog,” J. Bone Jt. Surg., 75(12), pp. 1795–1803. [17] Melis, B., Nemoz, C., and Walch, G., 2009, “Muscle Fatty Infiltration in Rota- [42] Silva, M. J., Boyer, M. I., Ditsios, K., Burns, M. E., Harwood, F. L., Amiel, D., tor Cuff Tears: Descriptive Analysis of 1688 Cases,” Orthop. Traumatol.: Surg. and Gelberman, R. H., 2002, “The Insertion Site of the Canine Flexor Digito- Res., 95(5), pp. 319–324. rum Profundus Tendon Heals Slowly Following Injury and Structure Repair,” J. [18] Miyazaki, A. N., Santos, P. D., Silva, L. A. D., Sella, G. D. V., Miranda, E. R. Orthop. Res., 20(3), pp. 447–453. D. A. B., and Zampieri, R., 2015, “Fatty Muscle Infiltration in Cuff Tear: Pre [43] St Pierre, P., Olson, E. N., Elliott, J. J., O’Hair, K. C., McKinney, L. A., and and Post Operative Evaluation by MRI,” Acta Ortop. Bras., 23(5), pp. 251–254. Ryan, J., 1995, “Tendon-Healing to Cortical Bone Compared With Healing to a [19] Burkhart, S. S., 1991, “Arthroscopic Treatment of Massive Rotator Cuff Tears: Cancellous Trough. A Biomechanical and Histological Evaluation in Goats,” J. Clinical Results and Biomechanical Rationale,” Clin. Orthop. Relat. Res., 267, Bone Jt. Surg., 77(12), pp. 1858–1866. pp. 45–56. [44] Thomopoulos, S., Hattersley, G., Rosen, V., Mertens, M., Galatz, L., Williams, [20] Gimbel, J. A., Van Kleunen, J. P., Mehta, S., Perry, S. M., Williams, G. R., and G. R., and Soslowsky, L. J., 2002, “The Localized Expression of Extracellular Soslowsky, L. J., 2004, “Supraspinatus Tendon Organizational and Mechanical Matrix Components in Healing Tendon Insertion Sites: An In Situ Hybridiza- Properties in a Chronic Rotator Cuff Tear Animal Model,” J. Biomech., 37(5), tion Study,” J. Orthop. Res., 20(3), pp. 454–463. pp. 739–749. [45] Benjamin, M., Kumai, T., Milz, S., Boszczyk, B. M., Boszczyk, A. A., and [21] Dourte, L. M., Perry, S. M., Getz, C. L., and Soslowsky, L. J., 2010, “Tendon Ralphs, J. R., 2002, “The Skeletal Attachment of Tendons–Tendon Properties Remain Altered in a Chronic Rat Rotator Cuff Model,” Clin. Orthop. “Entheses”,” Comp. Biochem. Physiol., Part A, 133(4), pp. 931–945. Relat. Res., 468(6), pp. 1485–1492. [46] Benjamin, M., Newell, R. L., Evans, E. J., Ralphs, J. R., and Pemberton, D. J., [22] Trudel, G., Ramachandran, N., Ryan, S. E., Rakhra, K., and Uhthoff, H. K., 1992, “The Structure of the Insertions of the Tendons of Biceps Brachii, Triceps 2010, “Supraspinatus Tendon Repair Into a Bony Trough in the Rabbit: and Brachialis in Elderly Dissecting Room Cadavers,” J. Anat., 180(Pt. 2), Mechanical Restoration and Correlative Imaging,” J. Orthop. Res., 28(6), pp. pp. 327–332. 710–715. [47] Benjamin, M., and Ralphs, J. R., 1998, “Fibrocartilage in Tendons and [23] Trudel, G., Ramachandran, N., Ryan, S. E., Rakhra, K., and Uhthoff, H. K., Ligaments—An Adaptation to Compressive Load,” J. Anat., 193(4), pp. 481–494. 2012, “Improved Strength of Early Versus Late Supraspinatus Tendon Repair: [48] Newsham-West, R., Nicholson, H., Walton, M., and Milburn, P., 2007, “Long- A Study in the Rabbit,” J. Shoulder Elbow Surg., 21(6), pp. 828–834. Term Morphology of a Healing Bone-Tendon Interface: A Histological Obser- [24] Plate, J. F., Brown, P. J., Walters, J., Clark, J. A., Smith, T. L., Freehill, M. T., vation in the Sheep Model,” J. Anat., 210(3), pp. 318–327. Tuohy, C. J., Stitzel, J. D., and Mannava, S., 2014, “Advanced Age Diminishes [49] Silva, M. J., Thomopoulos, S., Kusano, N., Zaegel, M. A., Harwood, F. L., Mat- Tendon-to-Bone Healing in a Rat Model of Rotator Cuff Repair,” Am. J. Sports suzaki, H., Havlioglu, N., Dovan, T. T., Amiel, D., and Gelberman, R. H., Med., 42(4), pp. 859–868. 2006, “Early Healing of Flexor Tendon Insertion Site Injuries: Tunnel Repair is [25] Safran, O., Derwin, K. A., Powell, K., and Iannotti, J. P., 2005, “Changes in Mechanically and Histologically Inferior to Surface Repair in a Canine Model,” Rotator Cuff Muscle Volume, fat Content, and Passive Mechanics After J. Orthop. Res., 24(5), pp. 990–1000. Chronic Detachment in a Canine Model,” J. Bone Jt. Surg., 87(12), pp. [50] Iannotti, J., Naranja, R., and Gartsman, G., 1994, “Surgical Treatment of 2662–2670. the Intact Cuff and Repairable Cuff Defect: Arthroscopic and Open [26] Coleman, S. H., Fealy, S., Ehteshami, J. R., MacGillivray, J. D., Altchek, D. Techniques,” Orthopaedic Knowledge Update: Shoulder and Elbow, Ameri- W., Warren, R. F., and Turner, A. S., 2003, “Chronic Rotator Cuff Injury and can Academy of Orthopaedic Surgeons, Rosemont, IL, pp. 151–156. Repair Model in Sheep,” J. Bone Jt. Surg., 85(12), pp. 2391–2402. [51] Perry, S. M., Getz, C. L., and Soslowsky, L. J., 2009, “After Rotator Cuff Tears, [27] Chaudhury, S., Holland, C., Vollrath, F., Carr, A. J., Holland, C., and Vollrath, the Remaining (Intact) Tendons are Mechanically Altered,” J. Shoulder Elbow F., 2011, “Comparing Normal and Torn Rotator Cuff Tendons Using Dynamic Surg., 18(1), pp. 52–57. Shear Analysis,” J. Bone Jt. Surg., 93(7), pp. 942–948. [52] Reuther, K. E., Thomas, S. J., Sarver, J. J., Tucker, J. J., Lee, C. S., Gray, C. F., [28] Thomopoulos, S., Williams, G. R., Gimbel, J. A., Favata, M., and Soslowsky, L. J., Glaser, D. L., and Soslowsky, L. J., 2013, “Effect of Return to Overuse Activity 2003, “Variation of Biomechanical, Structural, and Compositional Properties Following an Isolated Supraspinatus Tendon Tear on Adjacent Intact Tendons Along the Tendon to Bone Insertion Site,” J. Orthop. Res., 21(3), pp. 413–419. and Glenoid Cartilage in a Rat Model,” J. Orthop. Res., 31(5), pp. 710–715.

Journal of Biomechanical Engineering NOVEMBER 2017, Vol. 139 / 110801-9

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 10/14/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use [53] Reuther, K., Thomas, S., Tucker, J., Sarver, J., Gray, C., Evans, E., Ilkhani- [66] Kijima, H., Minagawa, H., Tomioka, T., Yamada, S., and Nozaka, K., 2013, Pour, S., Glaser, D., and Soslowsky, L., 2013, “Returning to Overuse Activity “Elasticity of the Coracoacromial Ligament in Shoulders With Rotator Cuff Following a Supraspinatus and Infraspinatus Tear Leads to Joint Damage in a Tears: Measurement With Ultrasound Elastography,” Surg. Sci., 4(09), pp. 1–5. Rat Model,” J. Biomech., 46(11), pp. 1818–1824. [67] Kijima, H., Minagawa, H., Saijo, Y., Sano, H., Tomioka, T., Yamamoto, N., [54] Reuther, K. E., Thomas, S. J., Tucker, J. J., Sarver, J. J., Gray, C. F., Rooney, S. Shimada, Y., Okada, K., and Itoi, E., 2009, “Degenerated Coracoacromial Liga- I., Glaser, D. L., and Soslowsky, L. J., 2014, “Disruption of the Anterior- ment in Shoulders With Rotator Cuff Tears Shows Higher Elastic Modulus: Posterior Rotator Cuff Force Balance Alters Joint Function and Leads to Joint Measurement With Scanning Acoustic Microscopy,” J. Orthop. Sci., 14(1), pp. Damage in a Rat Model,” J. Orthop. Res., 32(5), pp. 638–644. 62–67. [55] Reuther, K. E., Sarver, J. J., Schultz, S. M., Lee, C. S., Sehgal, C. M., Glaser, [68] Hashimoto, T., Nobuhara, K., and Hamada, T., 2003, “Pathologic Evidence of D. L., and Soslowsky, L. J., 2012, “Glenoid Cartilage Mechanical Properties Degeneration as a Primary Cause of Rotator Cuff Tear,” Clin. Orthop. Relat. Decrease After Rotator Cuff Tears in a Rat Model,” J. Orthop. Res., 30(9), pp. Res., 415, pp. 111–120. 1435–1439. [69] Sommerich, C. M., McGlothlin, J. D., and Marras, W. S., 1993, “Occupational [56] Thomas, S. J., Reuther, K. E., Tucker, J. J., Sarver, J. J., Yannascoli, S. M., Risk Factors Associated With Soft Tissue Disorders of the Shoulder: A Review Caro, A. C., Voleti, P. B., Rooney, S. I., Glaser, D. L., and Soslowsky, L. J., of Recent Investigations in the Literature,” Ergonomics, 36(6), pp. 697–717. 2014, “Biceps Detachment Decreases Joint Damage in a Rotator Cuff Tear Rat [70] Sørensen, A. K. B., Bak, K., Krarup, A. L., Thune, C. H., Nygaard, M., Model,” Clin. Orthop. Relat. Res., 472(8), pp. 2404–2412. Jørgensen, U., Sloth, C., and Torp-Pedersen, S., 2007, “Acute Rotator Cuff [57] Hsu, H.-C., Luo, Z.-P., Stone, J. J., Huang, T.-H., and An, K.-N., 2003, Tear: Do We Miss the Early Diagnosis? A Prospective Study Showing a High “Correlation Between Rotator Cuff Tear and Glenohumeral Degeneration,” Incidence of Rotator Cuff Tears After Shoulder Trauma,” J. Shoulder Elbow Acta Orthop. Scand., 74(1), pp. 89–94. Surg., 16(2), pp. 174–180. [58] Umans, H. R., Pavlov, H., Berkowitz, M., and Warren, R. F., 2001, “Correlation [71] Itoi, E., Berglund, L. J., Grabowski, J. J., Schultz, F. M., Growney, E. S., Mor- of Radiographic and Arthroscopic Findings With Rotator Cuff Tears and rey, B. F., and An, K. N., 1995, “Tensile Properties of the Supraspinatus Degenerative Joint Disease,” J. Shoulder Elbow Surg., 10(5), pp. 428–433. Tendon,” J. Orthop. Res., 13(20), pp. 578–584. [59] Radke, S., Kenn, W., and Gohlke, F., 2001, “MRI of the Shoulder,” The Ortho- [72] Waehrer, G. M., Dong, X. S., Miller, T., Haile, E., and Men, Y., 2007, “Costs paedist, 30(8), pp. 484–491. of Occupational Injuries in Construction in the United States,” Accid. Anal. [60] Petersson, C. J., 1983, “Degeneration of the Gleno-Humeral Joint: An Anatomi- Prev., 39(6), pp. 1258–1266. cal Study,” Acta Orthop. Scand., 54(2), pp. 277–283. [73] Barry, M., and Huang, L., 2014, “Work-Related Risk Factors for Rotator Cuff Syn- [61] Kernwein, G. A., 1965, “Roentgenographic Diagnosis of Shoulder Dys- drome,” Accident Compensation Corporation, Wellington, New Zealand, accessed function,” JAMA, 194(10), pp. 1081–1085. Sept. 26, 2017, https://www.acc.co.nz/assets/research/work-risk-rotator-cuff.pdf [62] Lugo, R., Kung, P., and Ma, C. B., 2008, “Shoulder Biomechanics,” Eur. J. [74] van der Meijden, O. A., Westgard, P., Chandler, Z., Gaskill, T. R., Kokmeyer, Radiol., 68(1), pp. 16–24. D., and Millett, P. J., 2012, “Rehabilitation After Arthroscopic Rotator Cuff [63] Soslowsky, L. J., An, C. H., Johnston, S. P., and Carpenter, J. E., 1994, Repair: Current Concepts Review and Evidence-Based Guidelines,” Int. J. “Geometric and Mechanical Properties of the Coracoacromial Ligament and Sports Phys. Ther., 7(2), p. 197. Their Relationship to Rotator Cuff Disease,” Clin. Orthop. Relat. Res., (304), [75] Brislin, K. J., Field, L. D., and Savoie, F. H., 2007, “Complications After pp. 10–17. Arthroscopic Rotator Cuff Repair,” Arthroscopy, 23(2), pp. 124–128. [64] Fremerey, R., Bastian, L., and Siebert, W. E., 2000, “The Coracoacromial Liga- [76] Gerber, C., Schneeberger, A. G., Perren, S. M., and Nyffeler, R. W., 1999, ment: Anatomical and Biomechanical Properties With Respect to Age and “Experimental Rotator Cuff Repair. A Preliminary Study,” J. Bone Jt. Surg., Rotator Cuff Disease,” Knee Surg. Sports Traumatol. Arthroscopy, 8(5), pp. 81(9), pp. 1281–1290. 309–313. [77] Parsons, B. O., Gruson, K. I., Chen, D. D., Harrison, A. K., Gladstone, J., and [65] Kesmezacar, H., Akgun, I., Ogut, T., Gokay, S., and Uzun, I., 2008, “The Cora- Flatow, E. L., 2010, “Does Slower Rehabilitation After Arthroscopic Rotator coacromial Ligament: The Morphology and Relation to Rotator Cuff Patho- Cuff Repair Lead to Long-Term Stiffness?,” J. Shoulder Elbow Surg., 19(7), logy,” J. Shoulder Elbow Surg., 17(1), pp. 182–188. pp. 1034–1039.

110801-10 / Vol. 139, NOVEMBER 2017 Transactions of the ASME

Downloaded From: http://biomechanical.asmedigitalcollection.asme.org/ on 10/14/2017 Terms of Use: http://www.asme.org/about-asme/terms-of-use The effect of out-of-plane deformation on ligament surface strain measurements

Haﬁzur Rahman Department of Mechanical Science and Engineering University of Illinois at Urbana-Champaign Urbana, IL 61801 Email: [email protected]

Mariana E. Kersh∗ Department of Mechanical Science and Engineering University of Illinois at Urbana-Champaign Urbana, IL 61801 Email: [email protected]

ABSTRACT The characterization of biological tissues depends on accurate measurements of deformation and strain, but less attention has been given to the role of out-of-plane deformation in ligament strain. The objective of this study was to investigate the influence of out-of-plane deformation on surface strain measurements in healthy and damaged ligaments. Tensile tests on five porcine posterior cruciate ligaments (PCL) were performed before and after damage using the femur-PCL-tibia construct. Damage was simulated by loading the ligament to its maximum force capacity. Digitized surface dots were tracked using an optical motion capture system. The transverse strain (εxx), longitudinal strain (εyy), and shear strain (γxy) distributions on the ligament surface were obtained for the control and damaged states using two-dimensional (2d) strain and three-dimensional (3d) strain measurements. There was no significant difference between the 2d and 3d strains in the control state for all three strains. However, the value and location of the peak strain values (tensile and compressive) in ligament surfaces did change. The 2d peak tensile strain was both over and under-estimated, compared to 3d strain, when out of plane deformation was included for εxx and εyy; but consistently overestimated for positive γxy. The percentage of damaged regions, quantified as a loss in tensile strength, after damage was overpredicted by 2d strain for εyy. Care should be taken when using 2d surface strain as peak values and local damage is sensitive to out-of-plane deformation.

Keywords: Posterior cruciate ligament; Out-of-plane deformation; 2d strain; 3d strain.

∗Address all correspondence to this author.

1 1 Introduction Injuries to ligaments are a common occurrence specifically among the athletes and more physically active populations. Posterior cruciate ligament (PCL) rupture occurs in 5-20% of all acute ligament knee injuries and causes pain, swelling, instability, and functional disability of the knee joint [1,2]. The accurate measurement of displacement and strain is critical to characterize biological tissues, organs, and their interactions with biomedical devices [3]. In the case of ligaments, as well as several other musculoskeletal tissues, this characterization is challenging due to their inhomogeneous and anisotropic nature. A precise full field measurement of ligament strain is necessary to help identify local strain concentrations and regions susceptible to damage [3]. This information can then be used to evaluate structure-strength relationships at different length scales and inform the development of engineered materials designed to replace damaged ligaments. Several methods have been used to measure the ligament and capsule strain distribution such as photoelastic coating, high speed films, dye lines, marker bead etc [4–8]. In most cases, the strain measurement is based on in-plane deformation; neglecting the out-of-plane deformation and reports of strain is limited to the axis of loading [9]. Digital image correlation (DIC) is a technique that can measure both two-dimensional (2d) and three-dimensional (3d) surface strain [9] but care is needed to optimize the surface preparation, hardware, and software settings to obtain accurate and precise measurement of strain [3]. However, from the existing literature, it is not clear if the inclusion or exclusion of out-of- plane deformation has any effect on the ligament surface strain calculations and no comparisons have been made to determine the accuracy of a 2d approximation. Therefore, the goal of this study was to investigate the contribution of out-of-plane deformation on the measurement of PCL surface strains. We sought to identify if the out-of-plane deformation affects the surface strain calculations and assessment of local damage. A simple and non-invasive digitized surface dot marker method was used to measure the transverse, longitudinal, and shear strains on the ligament surface with and without out-of-plane deformation for both the control and damaged states.

2 Materials and methods 2.1 Specimen preparation Porcine knee specimens (n=5, six months old) were collected from the Meat Science Laboratory at the University of Illinois and stored at –20◦C. Specimens were thawed at room temperature overnight before dissection. All soft tissues were carefully dissected without disturbing any part of the PCL or its bony insertion sites leaving a bone (femur) – ligament (PCL) – bone (tibia) construct (Fig. 1A). The PCL was moistened with phosphate-buffered saline (PBS) solution during the dissection, specimen preparation, and mechanical testing to prevent dehydration.

2.2 Experimental setup Biomechanical testing was performed using a materials test machine (Instron Model 5967, Instron Corporation, Norwood, MA, USA). A custom ﬁxture was developed to mount the femur–PCL–tibia construct in the test machine. The femur and tibia were aligned as close to full extension as possible and embedded with a fast curing epoxy (Fig. 1B). Twenty-one surface dots were marked in a grid using permanent ink along the ligament surface (Fig. 1C). The dots were digitized using an optical motion capture system with an accuracy of 0.1 mm (Optotrak Certus, Northern Digital Inc., Waterloo, Canada) to record the three-dimensional coordinates (x, y, and z). The repeatability of digitizing landmarks was 0.147 mm; therefore displacements below 0.15 mm were excluded.

2 Fig. 1: (A) The posterior view of the knee joint after dissecting all soft tissues except the PCL and leaving a bone (femur) - ligament (PCL) - bone (tibia) construct, (B) experimental testing setup with potted knee specimen including the Optotrak markers to measure the strains, (C) twenty-one dots marked on the ligament surface to measure the local strain and then strain distribution maps, seven rows in the transverse direction and three columns in the longitudinal direction along the PCL, and (D) transverse and longitudinal local strains were calculated between the markers along the transverse and longitudinal directions as shown by black arrows. Local shear strain was calculated by the angular deformation of the plane formed by four adjacent surface dots from the longitudinal and transverse directions as shown by green arrows.

2.3 Biomechanical testing Testing for each specimen was performed in three steps: (1) pre-damage strain, (2) damage simulation, and (3) post-damage strain. The initial position of the construct was defined as the reference position, and all three steps were started from this reference position. First, the PCL insertion sites and surface dots of the pre-damage ligament (herein referred to as the control state) were digitized prior to testing. Specimens were preconditioned with 5 loading-unloading cycles of displacements from 0 to 1.5 mm at an extension rate of 50 mm/min. Next, the specimen was loaded to 100 N at 50 mm/min ensuring that the ligament stiffness remained in the elastic region. Immediately once the specimen reached 100 N, the PCL insertion sites and surface dots were digitized again to obtain the final positions. The specimen was then returned to the reference position. During the second step, we permanently deformed the ligament to induce damage. The specimen was again preconditioned and the construct was loaded at 50 mm/min until it reached the maximum load. The load-deformation curve was monitored during testing and the test was stopped as soon as the load began to decrease from maximum load. Finally, the damaged ligament was loaded again from the reference position using the same protocol as in the first step (loaded to 100 N). The PCL insertion sites and surface dots were re-digitized at the reference and loaded position.

3 2.4 Strain distribution map

We calculated the transverse strain (εxx), longitudinal strain (εyy), and shear strain (γxy) before and after simulated damage. The local transverse strain was calculated as the change in length between two surface dots in the transverse direction divided by their initial length (Fig. 1D). Transverse strain among all surface dots was calculated and a distribution map was obtained by interpolating the strain over the surface of the ligament using custom software (Matlab R2017, Mathworks Inc., Natick, MA, USA). Similarly, the longitudinal strain map was obtained between two surface dots along the longitudinal direction (Fig. 1D). In the case of shear strain, two direction vectors were identified using four adjacent surface dots from the longitudinal and transverse to form a plane (Fig. 1D). Local shear strain was calculated from the angular deformation of the plane and finally the shear strain distribution map was obtained. For all three types of strains (transverse, longitudinal, and shear), strain distribution maps were obtained for 2d strain and 3d strain for both control and damaged states. For 2d strain, the local strain between two surface dots was based on the ‘x’ and ‘y’ coordinates and the out-of-plane coordinate ‘z’ was omitted thus representing only in-plane deformation. The ‘z’ coordinate was included for 3d strain calculations between the two surface dots. A paired-sample t-test was used to determine differences between the 2d strain and 3d strain in control state for all three types of strains (εxx, εyy, γxy) (OriginPro 2018, OriginLab Corporation, Northampton, MA, USA). Significance was set at p <0.05 and trends at p <0.1.

2.5 Changes in peak strain and identiﬁcation of damaged regions To assess the sensitivity of peak strain to the inclusion of out-of-plane deformation we calculated the percent change of peak strain values from 2d strain to 3d strain in the control state. Only the tensile peak strain was considered for transverse and longitudinal strains. However, for peak shear strain, we calculated percentage changes for both positive and negative shear. To determine whether the assessment of damage varies when using 2d or 3d strain measurements, we compared the percentage of damaged regions in each ligament for longitudinal strain. Damage was deﬁned as regions with a loss in tensile strength; that is those areas initially in tension but observed to be in compression, indicating that those regions are no longer taking tensile loads. The percentage of damaged regions was calculated as the ratio of the number of these damaged regions to the original number of tensile regions in the control state.

3 Results There was no significant difference between the 2d and 3d strains in the control state for all three strains. However, two out of five specimens showed an increasing trend for εxx (specimen 3, p = 0.06 and specimen 4, p = 0.072) whereas one specimen had increasing trend for εyy (specimen 5, p = 0.098) when using 3d strain measurements. Overall the strain distribution maps (between 2d and 3d control; between 2d and 3d damaged) were similar except the values and locations of peak strains varied (Fig. 2). Change in peak location was more common for compressive strain compared to tensile strain. When out of plane deformation was included, the peak transverse tensile strain (εxx) increased in three specimens (mean increase = 71.17%) (Fig. 3A). For longitudinal strain (εyy), only one specimen resulted in increased strain (64%) (Fig. 3B). However, the peak positive shear (γxy) increased for all five specimens and peak negative shear decreased (Fig. 3C, 3D). In terms of damage, the 2d longitudinal strain measurements overestimated the percentage of damaged regions in four specimens and one specimen was underestimated (Fig. 4).

4 Fig. 2: (A) Transverse, (B)longitudinal, and (C) shear strain distirbution maps on the ligament surface of specimen 2 for 2d and 3d strains in both control and damaged stages. The location of the tensile and compressive peak for transverse and longitudinal strains was represented by solid arrow and dash arrow with magnitudes respectively. For shear strain, solid arrow and dash arrow represent the peak location for positive and negative strains respectively.

4 Discussion The aim of this work was to understand the effect of out-of-plane deformation on ligament surface strain. While several methods exist to measure the surface strain of biological tissues, we utilized a simple and non-invasive digitized dot marker with an optical motion capture system to assess the 2d and 3d strains on ligament surface. Our results show that statistically there was no significant difference between the 2d and 3d strains in control states, but the value and position of the tensile and compressive peak strains changed. For shear strain, positive peak increased more than 300% for three specimens whereas negative peak decreased more than 100% for four specimens. Although the average strain distributions across the surfaces was not significantly different between 2d and 3d measurements, the change in peak values may increase in certain specimens causing higher local strain concentrations. Previous studies have shown that the shoulder and knee ligaments are often ruptured near the insertion sites [10–13]; typically, the common regions where the higher strain occurs in the ligaments compared to mid-substance [5, 6, 8, 11, 14]. Therefore, neglecting the out-of-plane deformation may mask the actual strain value and earlier failure may occur than expected based on 2d strain. Furthermore, all five specimens for εyy showed different percentages of damaged regions when compared between the 2d and 3d strains; indicating that the 2d strain may not accurately predict whether the local regions are in tension or compression before and after damage. The strain distribution maps suggest that transverse, longitudinal, and shear strains are nonuniformly distributed on the ligament surface. Most previous studies only report the longitudinal or transverse strain [9] and our results show that shear strain maps are similarly nonuniform. Although ligament shear strain receives less attention in literature, the shear deformation of the ligament surface was sensitive to damage; four out of five specimens had a four-fold increase in either positive/negative peak shear from control to injured states (for 3d shear strain) indicating angular deformation occurs on the ligament

5 Fig. 3: The percentage change in peak strain values from 2d strain to 3d strain in control states for (A) tensile peak for transverse strain, (B) tensile peak for longitudinal strain, (C) positive peak for shear strain, and (D) negative peak for shear strain. S1 to S5 represent the five specimens. The percentage change in tensile peak strain for transverse and longitudinal strains may increase or decrease from specimen to specimen; but positive peak in all five specimens for shear strain increased and negative peak decreased. surface during damage. Since ligament strain is heterogeneous with some regions stronger than others, care should be taken during the development of the ligament reconstruction grafts or even during the ligament repair so that the ligament can sustain the non-uniform load including non-tensile loading directions. These local mechanics are likely linked to more locally intrinsic micro-structural or compositional properties of ligaments. This variation may be due to differences in collagen fiber distribution, alignment, and cross- linking [15–18]. The variation in strain may also be a result of the compositional contributions of water, collagen, and glycosaminoglycan (GAG) [5, 19]. Further research to measure the local mechanical response and micro-structural analysis would be useful to explain the inhomogeneity and microstructure- function relationships. The current study has several limitations. First, we chose animal ligament (porcine PCL) for this biomechanical testing with a nominal sample size. We did not test the ligament at different flexion angles. Biomechanically, during flexion, the PCL experiences a different set of forces than those im- posed by this study and likely results in a different inhomogeneous force distribution. Furthermore, the method used to simulate damage is different than hypothesized physiological injury conditions. Future studies on ligaments with full volumetric displacement measurement and physiological boundary conditions are necessary to understand the through thickness surface strain distribution as it relates to specific mechanisms of ligament rupture. In conclusion, we have investigated the sensitivity of out-of-plane deformation on surface strain

6 Fig. 4: The percentage damaged regions from control to damaged states for longitudinal strain. S1 to S5 represent the ﬁve specimens. The grey bar represents the 2d strain whereas the white bar represents the 3d strain. The 2d strain overpredicts the percentage damaged regions in four specimens. measurement. The use of 2d surface measurements is likely sufﬁcient for elastic testing as evidenced by the similarity in the strain maps prior to damage. However, in some cases the peak strain may change and mask higher localized strain due to the omission of out-of-plane deformation. 2d strain in local regions also could be misleading while analyzing ligament damage.

References [1] Li, G., Papannagari, R., Li, M., Bingham, J., Nha, K. W., Allred, D., and Gill, T., 2008. “Effect of posterior cruciate ligament deﬁciency on in vivo translation and rotation of the knee during weightbearing ﬂexion.”. The American journal of sports medicine, 36(3), pp. 474–479. [2] Marx, R., Parker, R., Matava, M., and Sekiya, J., 2009. “Cruciate and Collateral Ligament In- juries”. In Chapter in AAOS: Comprehensive Orthopedic Review, J. Lieberman, ed. ch. 109, pp. 1113–1129. [3] Palanca, M., Tozzi, G., and Cristofolini, L., 2016. The use of digital image correlation in the biomechanical area: A review. [4] Hirokawa, S., Yamamoto, K., and Kawada, T., 2001. “Circumferential measurement and analysis of strain distribution in the human ACL using a photoelastic coating method”. Journal of Biomechanics, 34(9), pp. 1135–1143. [5] Lam, T. C., Shrive, N. G., and Frank, C. B., 1995. “Variations in rupture site and surface strains at failure in the maturing rabbit medial collateral ligament.”. Journal of biomechanical engineering, 117(4), pp. 455–61. [6] Noyes, F. R., Butler, D. L., Grood, E. S., Zernicke, R. F., and Hefzy, M. S., 1984. “Biomechanical Analysis of Human Knee Ligament Grafts used in Knee Ligament Repairs and Reconstructions”. J Bone Joint Surg AM, 66(3), pp. 344–352. [7] Rainis, C. A., Brown, A. J., McMahon, P. J., and Debski, R. E., 2012. “Effects of simulated injury on the anteroinferior glenohumeral capsule”. Medical and Biological Engineering and Computing, 50(12), pp. 1299–1307.

7 [8] Yamamoto, K., Hirokawa, S., and Kawada, T., 1998. “Strain distribution in the ligament using photoelasticity. A direct application to the human ACL”. Medical Engineering and Physics, 20(3), pp. 161–168. [9] Mallett, K. F., and Arruda, E. M., 2017. “Digital image correlation-aided mechanical characterization of the anteromedial and posterolateral bundles of the anterior cruciate ligament”. Acta Biomaterialia, 56, pp. 44–57. [10] Bey, M. J., Hunter, S. A., Kilambi, N., Butler, D. L., and Lindenfeld, T. N., 2005. “Structural and mechanical properties of the glenohumeral joint posterior capsule”. Journal of Shoulder and Elbow Surgery, 14(2), pp. 201–206. [11] Bigliani, L. U., Pollock, R. G., Soslowsky, L. J., Flatow, E. L., Pawluk, R. J., and Mow, V. C., 1992. “Tensile properties of the inferior glenohumeral ligament”. Journal of Orthopaedic Research, 10(2), pp. 187–197. [12] Lee, M., and Hyman, W., 2002. “Modeling of failure mode in knee ligaments depending on the strain rate”. BMC Musculoskeletal Disorders, 3(1), p. 3. [13] Noyes, F. R., DeLucas, J. L., and Torvik, P. J., 1974. “Biomechanics of anterior cruciate ligament failure: an analysis of strain-rate sensitivity and mechanisms of failure in primates.”. The Journal of bone and joint surgery. American volume, 56(2), pp. 236–253. [14] Butler, D. L., Sheh, M. Y., Stouffer, D. C., Samaranayake, V. A., and Levy, M. S., 1990. “Surface strain variation in human patellar tendon and knee cruciate ligaments”. J Biomech Eng, 112(1), pp. 38–45. [15] Castile, R. M., Skelley, N. W., Babaei, B., Brophy, R. H., and Lake, S. P., 2016. “Microstructural properties and mechanics vary between bundles of the human anterior cruciate ligament during stress-relaxation”. Journal of Biomechanics, 49(1), pp. 87–93. [16] Mommersteeg, T. J., Blankevoort, L., Kooloos, R. J., Hendriks, J. C., Kauer, J. M., and Huiskes, R., 1994. “Nonuniform distribution of collagen density in human knee ligaments”. Journal of Orthopaedic Research, 12(2), pp. 238–245. [17] Voycheck, C. A., Luu, K., McMahon, P. J., and Debski, R. E., 2014. “Collagen ﬁber alignment and maximum principal strain in the glenohumeral capsule predict location of failure during uniaxial extension”. Biomechanics and Modeling in Mechanobiology, 13(2), pp. 379–385. [18] Wright, J. O., Skelley, N. W., Schur, R. P., Castile, R. M., Lake, S. P., and Brophy, R. H., 2016. “Microstructural and Mechanical Properties of the Posterior Cruciate Ligament”. The Journal of Bone and Joint Surgery, 98(19), pp. 1656–1664. [19] Frank, C., McDonald, D., Lieber, R., and Sabiston, P., 1988. “Biochemical heterogeneity within the maturing rabbit medial collateral ligament.”. Clinical orthopaedics and related research(236), pp. 279–85.

8 Journal of Biomedical Optics 22(4), 046009 (April 2017)

Application of quantitative second-harmonic generation microscopy to posterior cruciate ligament for crimp analysis studies

Woowon Lee,a,b Hafizur Rahman,a Mariana E. Kersh,a and Kimani C. Toussaint Jr.a,b,c,d,* aUniversity of Illinois at Urbana-Champaign, Department of Mechanical Science and Engineering, Urbana, Illinois, United States bUniversity of Illinois at Urbana-Champaign, PROBE Lab, Urbana, Illinois, United States cUniversity of Illinois at Urbana-Champaign, Department of Electrical and Computer Engineering, Urbana, Illinois, United States dUniversity of Illinois at Urbana-Champaign, Department of Bioengineering, Urbana, Illinois, United States

Abstract. We use second-harmonic generation (SHG) microscopy to quantitatively characterize collagen fiber crimping in the posterior cruciate ligament (PCL). The obtained SHG images are utilized to define three distinct categories of crimp organization in the PCL. Using our previously published spatial-frequency analysis, we develop a simple algorithm to quantitatively distinguish the various crimp patterns. In addition, SHG microscopy reveals both the three-dimensional structural variation in some PCL crimp patterns as well as an underlying helicity in these patterns that have mainly been observed using electron microscopy. Our work highlights how SHG microscopy could potentially be used to link the fibrous structural information in the PCL to its mechanical properties. © 2017 Society of Photo-Optical Instrumentation Engineers (SPIE) [DOI: 10.1117/1.JBO.22.4.046009]

Keywords: second-harmonic generation imaging; posterior cruciate ligament; crimp analysis; Fourier transform analysis. Paper 170003RR received Jan. 2, 2017; accepted for publication Apr. 11, 2017; published online Apr. 28, 2017.

1 Introduction microscopy captures the birefringence of the sample by sand- A ligament is a band of soft fibrous tissue that connects bone to wiching the sample between a pair of crossed polarizers, and the birefringence intensity becomes maximum when the crossed bone. Its primary role is to passively control joint motion in polarizers are 45 deg with respect to the fiber.30 Occasionally, response to muscle forces while transmitting mechanical load – red first-order compensators are applied to show interference between bones.1 3 Routine daily activities apply repetitive and colors indicating fiber orientation, and the variation is observed sustained loads up to 25% of the maximum strength of knee – while rotating the polarizer.30 32 Under the PL microscope, ligaments,4,5 and excessive external force such as car accidents crimped collagen fibers appear to have extinction bands. or sports activities can reach up to four times of the daily applied Combined with tensile tests, researchers have extracted param- load.4 This can lead to ligament damage and injury,6 which is eters such as crimp angle and fiber orientation under different most common in knee injuries in the United Kingdom7 and loading conditions.27,28,33 Scanning and transmission electron has consistently been increasing among young athletes during 8 microscopies have been utilized to observe helical patterns the past two decades in the United States. Therefore, studying in ligaments, which refers to fiber undulation in different the mechanical properties of ligaments is crucial for developing planes.34–36 This helical fashion is also called “fibrillar effective injury prevention strategies and for providing crimp.”37,38 Furthermore, the regional difference between the improved surgical treatment. One of the main factors that affects periphery (anterior, posterior) and core portions within liga- the nonlinear mechanical and viscoelastic properties of liga- 9–13 ments has been analyzed using metrics such as the number of ments is the structural architecture. The collagen fiber crimps, fibril diameter, and fiber arrangement.38,39 However, arrangement, the primary load-bearing component in ligaments, these approaches have low-axial resolution and require rela- 14–17 has a structural hierarchy. From small to large, collagen tively invasive sample preparation stages including fixation, molecules aggregate to form microfibrils, which in turn form dehydration, and coating with gold.40,41 In contrast, quantitative subfibrils, then fibrils, fibers, fascicles, and eventually gross lig- second-harmonic generation (SHG) microscopy can provide ament. Approximately 90% of the collagen content in a ligament three-dimensional (3-D) quantitative analysis of type I collagen comprises type I collagen, which is less than what is found in a fiber organization and needs minimum sample preparation. 3,18 tendon. The collagen fibers are known to have complex wavy Previous studies have used SHG imaging on ligaments to mea- 19,20 undulation crimp patterns. Since crimping strongly affects sure parameters such as fiber density and orientation and have – the nonlinear mechanical response of a ligament,21 24 it is essen- correlated them to the computed mechanical properties to evalu- tial to use appropriate imaging techniques to examine the crimp ate the effects of smoking,42 laser damage,43 and chemicals.44 pattern. SHG is a second-order nonlinear optical process where the There have been numerous studies using polarized light (PL) signal generated by the sample is at half the wavelength of the – microscopy to image the crimp patterns in ligaments.25 29 PL incident optical (excitation) field.45,46 As a microscopy technique, SHG offers high-intrinsic contrast, submicron spatial

Journal of Biomedical Optics 046009-1 April 2017 • Vol. 22(4) Lee et al.: Application of quantitative second-harmonic generation microscopy. . . resolution, and 3-D optical sectioning.47 Collagen type I, being 2.2 Experimental Setup intrinsically noncentrosymmetric, is highly specific to SHG46,47 and provides an excellent opportunity for 3-D imaging of lig- Selected fibrous regions are imaged by SHG microscopy. aments. Our group has previously shown that SHG combined A tunable Ti:Sapphire laser producing 100-fs duration pulses with spatial Fourier transform (FT) analysis, FT-SHG, is a sim- spectrally centered at 780 nm illuminates the sample. The input × ple and powerful tool for analyzing collagen fiber organization beam is focused on the sample by a 40 0.65 NA objective lens in tendon,48 bone,49 and breast tissue.50 Moreover, we have (Olympus, PLAN N), and the backward SHG signal is collected shown how to extend quantitative SHG to analyze 3-D collagen by the same objective. A 390-nm bandpass filter separates the fiber organization.51,52 In this paper, we apply FT-SHG to analy- SHG signal from any generated autofluorescence. The beam is x y ∼ ∕ sis of the 3-D collagen fiber spatial arrangement in the posterior raster scanned by an - galvanometer scanner at 1.1 mm s, cruciate ligament (PCL). While Vidal and Mello32,53 have pre- and the epi-directed-SHG signal is collected by a photomulti- viously utilized FT analysis on tendons to detect crimp pattern plier tube (Hamamatsu, H10721-110). The focus drive attached variability, to the best of our knowledge, the work in this study is on the microscope (Olympus, IX81) moves in a step size of z the first approach to adopt FT-SHG to analyze 3-D crimp organi- 500 nm along the -axis, shifting the focus plane to generate zation in ligament tissue. We believe that our crimp pattern the 3-D stack, and for any single plane four images are tiled analysis could act as a bridge, providing additional information together to form a wider area. The size of the obtained SHG × × μ x y z in biomechanical studies that aim to predict damage or injury in 3-D image stacks are 200 200 30 m in the - - dimen- the PCL. sion. The average power of the laser on the sample plane is ∼10 mW. A more detailed description of the SHG microscopy 54 2 Method setup can be found elsewhere.

2.1 Sample Preparation 2.3 Image Analysis Five porcine knee PCL specimens (of age six months) are cut The obtained SHG images are analyzed using a customized into thirds along the proximal-distal ligament direction and MATLAB code to label the crimp patterns in each region embedded in an optimal cutting temperature compound (Fig. 1). (Fig. 2). The type discerning algorithm initially calculates the The middle region between the proximal and distal portions percentage of the dark pixels in the entire image. Those possess- is sectioned at 100-μm thickness by a cryostat (Leica, ing a lower intensity than the dark threshold are identified as CM3050S). The thin sections are collected from between the dark pixels, and the dark threshold is a single value applied on anterior–posterior regions, thereby neglecting the sheath. The the entire image, which segments the background to the SHG tissue sections are then placed on glass microscope slides signal.41 If the number of dark pixels in the image is less and #1.5 coverslips are mounted on top with the aqueous mount- than 10%, the image is assigned as category A (CAT A) or cating media. Tweezers are used to gently lower the coverslip on egory B (CAT B); conversely, when the number of dark pixels the microscope slide to avoid creating any air bubbles. After comprises more than 10% of the region, the image is identified the samples dry, nail polish is applied on the corners to seal as category C (CAT C). Next, spatial FT analysis is carried out the samples. All ligaments are collected from the Meat Science on the image, generating a spatial-frequency domain image, and Laboratory of the University of Illinois at Urbana-Champaign. the corresponding two-dimensional magnitude spectrum is inte- This study is exempt from the Illinois Institutional Animal Care grated radially from 0 deg to 360 deg with a step size of 1 deg. and Use Committee (IACUC). Radial integration values lower than 30% of the maximum value

Fig. 1 (a) Digital image of PCL anatomy, (b) cutting of the PCL into thirds along the proximal-distal direction, and (c) subsequent sectioning of the middle piece into 100-μm thick slices. (d) Mounting of thin sections on a microscope slide with a coverslip placed on top.

Journal of Biomedical Optics 046009-2 April 2017 • Vol. 22(4) Lee et al.: Application of quantitative second-harmonic generation microscopy. . .

Fig. 2 Flowchart of image analysis for differentiating crimp types. Initially, the pixels having a lower intensity than the dark threshold are counted and are used to separate CAT A, B from CAT C. FT analysis is applied followed by radial integration to identify CAT A, B, C-1, and C-2. See text for details. are removed, while those above this threshold are plotted in neighboring SHG bands in CAT C-2 images. Initially, each a polar angular plot for each angle. If the plotted curve fits bright band is isolated, and then FT analysis is applied to the strongly with a wrapped Gaussian distribution55,56 (r2 > 0.95), isolated images to calculate the orientation. The same process it is defined as CATA. When the correlation is weak (r2 < 0.95), is implemented on each band, and the corresponding angle is then it is defined as CAT B. For CAT C, the identical radial inte- computed. gration process is applied. In the range of 0 deg to 180 deg, the polar plot with two distinct peaks is further classified as CAT C- 3 Results and Discussion 2, whereas a single peak plot is CAT C-1. The two peaks can be Based on observation, the SHG images are categorized as CAT verified by fitting a wrapped two-term Gaussian distribution and A, B, or C as shown in Fig. 3. CATA is characterized as samples observing the distance of the two means of each peak. The dif- with little or no observed crimps in the fibers. Crimps that are ference of means for CAT C-1 is narrower than 35 deg and wider confined in-plane are defined as CAT B. Crimps that are out-of- than 70 deg for CAT C-2. This classification process is applied plane, along the third spatial dimension, are defined as CAT C. to all SHG images in the stack. Once the images have a clear We consider the fibrous features on the SHG images as “fibers” tendency toward one type of crimping throughout the whole due to the diameter coinciding with previous studies,19 and the stack, the entire region is assigned that type. A polar plot probability of each crimp type to be detected in the fibrous that represents the volume can be generated by adding all the regions of the PCL is comparable. We note that conventional data points from each slice. sample preparation stages such as dissection and cutting into To look at the effect of helicity, the crimp angle is obtained slices could alter the original, natural fiber structure of ligament. by measuring the fiber orientation differences of the two However, structure analysis performed on ex vivo condition is

Fig. 3 Three types of collagen fiber structure in the PCL. CAT A has almost no waviness of fibers (crimps) and the majority of them are in a single orientation. CAT B has in-plane crimps spreading the range of possible fiber orientations. CAT C has repeating dark and bright bands perpendicular to the fiber direction illustrating crimps out-of-plane. The scale bar is 20 μm.

Journal of Biomedical Optics 046009-3 April 2017 • Vol. 22(4) Lee et al.: Application of quantitative second-harmonic generation microscopy. . .

Fig. 4 Representative (a) SHG images and the corresponding (b) FT images for each crimp type (CAT A, B, C-1, C-2). The insets represent a magnified view of low spatial-frequency components (center pixels in the black box). These images are integrated radially across different angles and (c) the resulting integrated values (blue) are plotted in a polar plot format. CAT A and B are fitted with a wrapped Gaussian distribution (red), and the correlation coefficient is calculated. CAT C-1 and C-2 are fitted with a two-term wrapped Gaussian distribution (red), and the number of peaks is detected by measuring the distance between the means of the two peaks. For visualization purposes, the FT images are converted to log scale. The scale bar is 40 μm. a common methodology conducted by many researchers that spatial-frequency data) from each slice within the volume of has been shown to be of value.25,28,37,38,43 The primary focus of images per crimp category. Note that the actual summed inten- this study is to employ SHG imaging to quantify collagen fiber sities for the stack, which would typically be labeled along the crimping of ligament under ex vivo conditions. radius of the plots, are not shown in Fig. 4(c). The straight fibers Figure 4 shows the results of our FT-SHG analysis on rep- in CAT A are characteristic of a single orientation, which cor- resentative crimp patterns for CAT A, B, and C. In Fig. 4(a),we responds to a wrapped Gaussian distribution55,56 in the polar plot observe the typical SHG images for each type. We confirm that (red curve) with r2 > 0.95. The majority of fibers in CAT A are CATA visually appears to have little to no crimps, whereas CAT oriented along ∼95 deg [observed in Fig. 4(a)] as confirmed by B shows an in-plane crimp pattern. We observe in Fig. 4(a) that the peak of the corresponding plot in Fig. 4(c) being marginally CAT C images also have crimp patterns but with significantly over 90 deg. Conversely, fibers in CAT B are curled in-plane. In more dark SHG areas compared to CATA and B. The dark areas this case, the in-plane crimping results in neighboring angles are generated because of the out-of-plane crimps. This is due to being added to the primary fiber orientation and resulting in the SHG intensity reducing once the fibers go out of the image a broader peak in the polar plot. It can be observed that the radial plane and the signal dropping significantly when the fibers are integration distribution regarding CAT B deviates from a perpendicular with the image plane.47,57 For further analysis, wrapped Gaussian distribution and has a lower correlation coef- CAT C is divided into C-1 and C-2, which exhibit irregular ficient with r2 < 0.95. In the case of CAT C, the circular plot and regular crimp patterns, respectively. from 0 deg to 180 deg has two peaks for CAT C-2. One of The corresponding spatial-frequency maps (on a log scale) the peaks represents the orientation of the fibers in the CAT for the aforementioned crimp patterns are shown in Fig. 4(b), C-2 SHG image, which is along ∼110 deg. The other peak where the insets highlight the low spatial-frequency components results from the direction of the regular out-of-plane crimp pat- in the center of the spectrum. For CAT A and CAT B, the dom- tern and appears almost orthogonal from the fiber orientation. inant preferred fiber orientations observed in Fig. 4(a) images This distance between the peaks is calculated by fitting a are readily picked up in the spectral data. In the case of CAT wrapped two-term Gaussian distribution and measuring the dif- C-1, we observe more spatial isotropy regarding collagen fiber ference between the means of each term. In terms of CAT C-1, preferred orientation and crimp direction. However, the spatial- the out-of-plane crimps are not clear enough to appear as frequency patterns for CAT C-2 reveal the preferred orientation a peak on the circular plot, so there is usually one single of the collagen fibers (similar to the cases of CATA and CAT B), broad peak. The irregularity of the dark bands in CAT C-1 as well as the approximately orthogonal dark bands observed in could be caused by the disorder of the crimps37,58 and an oblique the CAT C-2 SHG image. The low spatial-frequency compo- cut from a regular crimp pattern. nents in CAT C-2 are affected by the dark bands. The samples are also imaged by conventional imaging Figure 4(c) shows the radial integration polar plots, which techniques such as bright-field [Fig. 5(a)] and PL microscopy are the sum of the radial integration (obtained from the [Fig. 5(b)]. The low contrast in Fig. 5(a) results from the

Journal of Biomedical Optics 046009-4 April 2017 • Vol. 22(4) Lee et al.: Application of quantitative second-harmonic generation microscopy. . .

Fig. 5 Images of the entire ligament sample using (a) bright-field and (b) PL microscopy (in yellow pseu- docolor). Selected fibrous regions on each sample are imaged with SHG in 3-D stacks with the step size of 500 nm. There is (c) a structure variation along the z-axis shown by SHG images. All slices in each stack are distinguished by the flowchart and (d) assigned a crimp type. Scale bar is 2 mm and 40 μm, respectively. unstained biological samples having inherently little contrast.59 It is important to note that we focus on the aforementioned These instruments are described elsewhere in more detail54 and advantages of SHG microscopy, which make it suitable for are used for the purpose of mapping and selecting areas for sub- revealing collagen fiber structural information. While an in sequent SHG imaging. Due to the tissue size (1.2 × 1.2 cm), depth one-to-one comparison between imaging modalities is ∼450 images are stitched to generate images with larger not our primary aim, a brief qualitative comparison of images fields-of-view. By observing the bright-field images, we choose obtained using SHG microscopy to other modalities is illus- three fibrous regions per sample and obtain SHG images on the trated in the Appendix. selected regions. As shown in Fig. 5(c), SHG microscopy illus- In the case of CAT C-2 images, we observe a repetitive pat- trates the collagen fiber structure variation along the depth tern of alternating fiber orientations [Figs. 6(a)–6(b)]. In other (z-plane) and the corresponding assigned type. This structure words, the fiber bands in the SHG images have a similar orien- variation also differs by region. For example, in Fig. 5(d), region tation in every other band. For example, in Figs. 6(a)–6(b), the 2 seems to have irregular out-of-plane crimps on slice 1 (CAT C- first red arrow from the top along with the third and fifth arrows 1), but 12.5 μm below the plane the pattern appears to be more have comparable directions. This also applies to the second, organized and is categorized as type CAT C-2. 25 μm below fourth, and sixth arrows. Thus, we suggest that the fibers are slice 1, the out-of-plane crimps disappear, and the image is not only out-of-plane crimped but also in a helical crimp pattern assigned to CAT B. Other regions have less feature variance, [Fig. 6(c)]. This is because helical crimps naturally have an out- and the assigned type does not change along the z-axis. of-plane region once the fibers twist and change directions. The The optical sectioning capability of SHG imaging46,51,60 direction shift is repetitive, which leads to dark bands and bright allows us to capture the overall 3-D structural variations in lig- bands alternating orientation for CAT C-2 crimps. These dark aments. SHG imaging can also be applied to thick samples by and bright regions are indicated as blue and yellow boxes, collecting the back-scattered SHG signal.61,62 There have been respectively, in Figs. 6(a)–6(b). However, since the low-SHG- studies using PL microscopy quantifying crimp organization by signal areas are relatively large, it can be thought that the measuring the linear birefringence;31 however, low-axial resolu- twist areas have a planar crimp that makes the out-of-plane tion diminishes its ability to observe 3-D structural variation. area extended. We utilize solder wires to replicate the crimp

Journal of Biomedical Optics 046009-5 April 2017 • Vol. 22(4) Lee et al.: Application of quantitative second-harmonic generation microscopy. . .

Fig. 6 In (a) and (b) CAT C-2 images, fiber orientation shifts in a repetitive pattern. Fibers have a similar orientation in every other bright band. Regions in the yellow rectangle have a high SHG intensity, indicating that the fibers are parallel with the image plane. Regions in the blue rectangle have a low intensity, suggesting that the fibers are oblique or perpendicular with the image plane. (c) Solder wires are used to describe the helical crimp pattern with an out-of-plane crimp. (d) Schematic demonstrating how crimped angle is calculated. The scale bar is 20 μm. pattern in Fig. 6(c), which also illustrates each individual 4 Conclusion crimped fiber aligned next to each other forming bundles. In conclusion, FT-SHG was applied for the first time on the PCL These bundles of fibers are stacked in layers maintaining its to quantitatively assess collagen fiber organization. The results structure. This can be explained due to no particular change of our work could be summarized in three major points. First, z in the crimp pattern throughout the -stack. Features in CAT we were able to identify three types of crimping patterns in PCL C-2, such as the bright and dark areas and fiber directions, samples. A customized code, along with FT analysis, was used remained constant even though the depth of focus changes to quantitatively distinguish each type. Second, unlike tradi- μ for roughly 20 m. tional imaging modalities used to analyze the PCL, we demon- 34–36 Previous studies conducted using electron microscopy strated the 3-D variation of crimp structure using SHG imaging. 30,31,63 and PL microscopy show similar results of the helical Third, we verified the helical crimp format of collagen fibers in crimp patterns. Electron microscopes were able to image twisted ligament previously observed by electron and PL microscopes. fibers directly from highly magnified images, and PL micros- Our work takes advantage of the label-free, high-contrast, and copy revealed the helicity of fibers measuring densitometric optical sectioning capabilities of SHG microscopy to highlight features from the birefringence intensity. From the helical struc- its potential to be used to link the fibrous structural information ture, researchers have suggested their mechanical roles act like in the PCL with its underlying mechanical properties. a buffer system-absorbing load to prevent damage that could occur in fibers during elongation.38 The helical model has also been applied to simulation models64 and analyzed for applications such as grafts.35,36 Our results from SHG imaging Appendix brings another modality for imaging the helicity of ligament Figure 7 shows a comparison of ligament images taken by SHG fibers and demonstrates the potential to provide image data microscopy in comparison to the more conventional approaches for 3-D computational simulations of fiber-based anisotropic of bright-field and PL microscopy. It is important to note that the materials. relative image locations are not identical because of the use of The crimp angle is defined as the angle of the crimp with different imaging platforms. However, the regions are chosen respect to the horizontal [Fig. 6(d)]; thus, intuitively crimp angle where clear image fiducials are present to facilitate coregistra- in our case is equivalent to one half of the angle differences tion with a spatial uncertainty of less than 150 μm. Bright-field between each band. The measured crimp angle (data not shown) images normally show fibrous areas, whereas PL images all matches well with previously published data.38,64,65 show out-of-plane crimps (CAT C). For the SHG images, a

Journal of Biomedical Optics 046009-6 April 2017 • Vol. 22(4) Lee et al.: Application of quantitative second-harmonic generation microscopy. . .

Fig. 7 Comparison of different imaging modalities on the selected fibrous regions in the PCL. For the SHG images, one image per region is selected as a representative from the entire 3-D stack. The scale bar is 40 μm.

single image per region from the stack was chosen, and the 8. C. R. LaBella, W. Hennrikus, and T. E. Hewett, “Anterior cruciate ligament image features vary by region. injuries: diagnosis, treatment, and prevention,” Pediatrics 133, e1437– e1450 (2014). Disclosures 9. J. Callaghan, The Adult Knee, Lippincott Williams & Wilkins, Philadelphia (2003). We have no relevant financial interests in this manuscript. There 10. S. M. Full et al., “Effect of fiber orientation of collagen-based electro- is no conflict of interest. spun meshes on human fibroblasts for ligament tissue engineering applications,” J. Biomed. Mater. Res. Part B Appl. Biomater. 103, Acknowledgments 39–46 (2015). 11. S. Pal, Design of Artificial Human Joints & Organs, pp. 23–40, This work was partially supported by the University of Illinois at Springer Science & Business Media, New York (2014). Urbana-Champaign. 12. S. Woo, P. Renstrom, and S. Arnoczky, The Encyclopaedia of Sports Medicine: An IOC Medical Commission Publication, Tendinopathy in Athletes, John Wiley & Sons, Oxford (2008). References 13. P. Netti et al., “Structure-mechanical properties relationship of natural tendons and ligaments,” J. Mater. Sci. Mater. Med. 7,525–530 1. J. Dvorak et al., “Biomechanics of the craniocervical region: the alar and (1996). transverse ligaments,” J. Orthop. Res. 6, 452–461 (1988). 14. A. Viidik, “Functional properties of collagenous tissues,” Int. Rev. 2. J. Goodfellow and J. O’Connor, “The mechanics of the knee and pros- Connect. Tissue Res. 6, 127–215 (1973). thesis design,” J. Bone Jt. Surg. 60B, 358–369 (1978). 15. J. Kastelic, A. Galeski, and E. Baer, “The multicomposite structure of 3. U. Meyer et al., Fundamentals of Tissue Engineering and Regenerative tendon,” Connect. Tissue Res. 6,11–23 (2009). Medicine, Springer Science & Business Media, Berlin, Germany 16. D. Elliott, “Structure and function of mammalian tendon,” Biol. Rev. (2009). 40 – 4. L. Yahia, Ligament and Ligamentoplasties, Springer Science & , 392 421 (1965). 17. W. C. Dale et al., “On the ultrastructure of mammalian tendon,” Business Media, Berlin, Germany (2012). 28 – 5. F. Noyes et al., “Biomechanical analysis of human ligament grafts used Experientia , 1293 1295 (1972). in knee-ligament repairs and reconstructions,” J. Bone Jt. Surg. 66, 344– 18. M. Nordin, T. Lorenz, and M. Campello, Basic Biomechanics of the – 352 (1984). Musculoskeletal System, pp. 102 125, Lippincott Williams & Wilkins, 6. S. Yamamoto et al., “The strain-rate dependence of mechanical proper- Berlin, Germany (2000). “ ” ties of rabbit knee ligament,” in Int. Technical Conf. on the Enhanced 19. D. L. Butler et al., Biomechanics of ligaments and tendons, Exerc. 6 – Safety of Vehicles, pp. 1–9 (2003). Sport Sci. Rev. , 125 182 (1978). “ ” 7. S. Bollen, “Epidemiology of knee injuries: diagnosis and triage,” Br. J. 20. C. B. Frank, Ligament structure, physiology and function, 4 – Sports Med. 34, 227–228 (2000). J. Musculoskelet. Neuronal Interact. , 199 201 (2004).

Journal of Biomedical Optics 046009-7 April 2017 • Vol. 22(4) Lee et al.: Application of quantitative second-harmonic generation microscopy. . .

21. J. Diamant et al., “Collagen: ultrastructure and its relations to mechani- 47. F. Pavone and P. Campagnola, Second Harmonic Generation Imaging, cal properties as a function of ageing,” Proc. R. Soc. London Ser. B CRC Press, Boca Raton, Florida (2013). 180, 293–315 (1972). 48. M. Sivaguru et al., “Quantitative analysis of collagen fiber organization 22. K. A. Hansen, J. A. Weiss, and J. K. Barton, “Recruitment of tendon in injured tendons using Fourier transform-second harmonic generation crimp with applied tensile strain,” J. Biomech. Eng. 124,72–77 imaging,” Opt. Express 18, 24983 (2010). (2002). 49. R. Ambekar et al., “Quantitative second-harmonic generation micros- 23. L. U. Bigliani et al., “Tensile properties of the inferior glenohumeral copy for imaging porcine cortical bone: comparison to SEM and ligament,” J. Orthop. Res. 10, 187–197 (1992). its potential to investigate age-related changes,” Bone 50,643–650 24. Y. Lanir, “Structure-strength relations in mammalian tendon,” Biophys. (2012). J. 24, 541–554 (1978). 50. R. Ambekar et al., “Quantifying collagen structure in breast biopsies 25. L. J. Gathercole and A. Keller, “Crimp morphology in the fibre-forming using second-harmonic generation imaging,” Biomed. Opt. Express collagens,” Matrix 11, 214–234 (1991). 3, 2021–2035 (2012). 26. K. S. Miller et al., “Characterizing local collagen fiber re-alignment 51. T. Y. Lau, R. Ambekar, and K. C. Toussaint, “Quantification of collagen and crimp behavior throughout mechanical testing in a mature mouse fiber organization using three-dimensional Fourier transform-second- supraspinatus tendon model,” J. Biomech. 45, 2061–2065 (2012). harmonic generation imaging,” Opt. Express 20, 21821 (2012). 27. T. York et al., “Real-time high-resolution measurement of collagen 52. T. Y. Lau et al., “Application of Fourier transform-second-harmonic alignment in dynamically loaded soft tissue,” J. Biomed. Opt. 19, generation imaging to the rat cervix,” J. Microsc. 251,77–83 066011 (2014). (2013). 28. N. W. Skelley et al., “Differences in the microstructural properties of 53. B. de C. Vidal and M. L. S. Mello, “Optical anisotropy of collagen fibers the anteromedial and posterolateral bundles of the anterior cruciate of rat calcaneal tendons: an approach to spatially resolved supramolecu- ligament,” Am. J. Sports Med. 43, 928–936 (2015). lar organization,” Acta Histochem. 112,53–61 (2010). 29. K. Hayashi and P. Muir, Advances in the Canine Cranial Cruciate 54. W. Lee et al., “Third-harmonic generation imaging of breast tissue Ligament, pp. 43–92, John Wiley & Sons, Hoboken, New Jersey biopsies,” J. Microsc. 264, 175–181 (2016). (2011). 55. N. Fisher, Statistical Analysis of Ciruclar Data, Cambridge University 30. B. C. Vidal, “Form birefringence as applied to biopolymer and in- Press, Cambridge, United Kingdom (1993). organic material supraorganization,” Biotech. Histochem. 85, 365–378 56. T. Diethe, N. Twomey, and P. Flach, “Bayesian modeling of the tem- (2010). poral aspects of smart home activity with circular statistics,” in Joint 31. B. de Campos Vidal, “Image analysis of tendon helical superstructure European Conf. on Machine Learning and Knowledge Discovery in using interference and polarized light microscopy,” Micron 34, 423–432 Databases, pp. 279–294 (2015). (2003). 57. E. Yew and C. Sheppard, “Effects of axial field components on second 32. B. de Campos Vidal and M. L. S. Mello, “Structural organization of harmonic generation microscopy,” Opt. Express 14, 1167–1174 collagen fibers in chordae tendineae as assessed by optical anisotropic (2006). properties and Fast Fourier transform,” J. Struct. Biol. 167, 166–175 58. W. Maurel et al., Biomechanical Models for Soft Tissue Simulation, (2009). pp. 1–21, Springer Science & Business Media, Berlin, Germany (2013). 33. M. Franchi et al., “Crimp morphology in relaxed and stretched rat 59. C. Krafft and J. Popp, Comprehensive Biomedical Physics, pp. 189– Achilles tendon,” J. Anat. 210,1–7 (2007). 208, Newnes, Amsterdam, Netherlands (2014). 34. L. H. Yahia and G. Drouin, “Collagen structure in human anterior cruciate 60. M. Shribak and R. Oldenbourg, “Techniques for fast and sensitive mea- ligament and patellar tendon,” J. Mater. Sci. 23,3750–3755 (1988). surements of two-dimensional birefringence distributions,” Appl. Opt. 35. L. H. Yahia and G. Drouin, “Microscopical investigation of canine ante- 42, 3009–3017 (2003). rior cruciate ligament and patellar tendon: collagen fascicle morphology 61. A. H. Reshak, “Second harmonic generation from thick leaves using and architecture,” J. Orthop. Res. 7, 243–251 (1989). the two-photon laser scanning microscope,” Micron 40,455–462 36. J. Zhu et al., “Ultrastructural and morphological characteristics of (2009). human anterior cruciate ligament and hamstring tendons,” Anat. Rec. 62. X. Han and E. Brown, “Measurement of the ratio of forward-propagat- 295, 1430–1436 (2012). ing to back- propagating second harmonic signal using a single objec- 37. M. Franchi et al., “Tendon and ligament fibrillar crimps give rise to left- tive,” Opt. Express 18, 10538 (2010). handed helices of collagen fibrils in both planar and helical crimps,” 63. B. de C. Vidal, “Crimp as part of a helical structure,” C. R. Acad. Sci. III. J. Anat. 216, 301–309 (2010). 318, 173–178 (1995). 38. M. Franchi et al., “Collagen fibre arrangement and functional crimping 64. S. P. Reese, S. A. Maas, and J. A. Weiss, “Micromechanical models of pattern of the medial collateral ligament in the rat knee,” Knee Surg. helical superstructures in ligament and tendon fibers predict large Sport. Traumatol. Arthrosc. 18, 1671–1678 (2010). Poisson’s ratios,” J. Biomech. 43, 1394–1400 (2010). 39. S. Zaffagnini et al., “Collagen fibre and fibril ultrastructural arrange- 65. A. K. Harvey, M. S. Thompson, and S. M. Brady, “Helical crimp model ment of the superficial medial collateral ligament in the human predicts material properties from tendon microsctructure,” in Medical knee,” Knee Surg. Sport. Traumatol. Arthrosc. 23, 3674–3682 Image Understanding and Analysis, Warwick, United Kingdom, (2015). pp. 1–5 (2010). 40. S. M. Wilson and A. Bacic, “Preparation of plant cells for transmission electron microscopy to optimize immunogold labeling of carbohydrate Woowon Lee received his bachelor’s degree in mechanical engi- and protein epitopes,” Nat. Protoc. 7, 1716–1727 (2012). neering from Hanyang University, Republic of Korea, and his mas- 41. P. K. Chu and X. Liu, Biomaterials Fabrication and Processing ter’s degree from the University of Illinois at Urbana-Champaign Handbook, CRC Press, Boca Raton (2008). (UIUC). He is a graduate research assistant in the Department of 42. J. E. Kelleher, T. Siegmund, and R. W. Chan, “Collagen microstructure Mechanical Science and Engineering at UIUC. His research interest in the vocal ligament: initial results on the potential effects of smoking,” is nonlinear imaging techniques such as SHG and THG analyzed Laryngoscope 124, E361–E367 (2014). by a quantitative metric applied on biological samples. He is also 43. N. Y. Ignatieva et al., “Laser-induced modification of the patellar interested in visualizing material properties by nonlinear optical ligament tissue: comparative study of structural and optical changes,” methods. Lasers Med. Sci. 26, 401–413 (2011). Hafizur Rahman received his bachelor’s degree in mechanical 44. H. Chen et al., “Microstructure and mechanical property of engineering from Bangladesh University of Engineering and Glutaraidehyde-treated porcine pulmonary ligament,” J. Biomech. Eng. 138 Technology, Bangladesh, in 2009, and his MS degree in ocean engi- , 061003 (2016). neering from Florida Atlantic University, USA, in 2013. Currently, he “ ” 45. P. A. Franken et al., Generation of optical harmonics, Phys. Rev. Lett. is a PhD candidate at the University of Illinois at Urbana-Champaign, 7 – , 118 119 (1961). USA. His research focuses on understanding the effects of 46. B. Masters and P. So, Handbook of Biomedical Nonlinear Optical rotator cuff tears on shoulder kinematics and glenohumeral cartilage Microscopy, Oxford University Press, Oxford (2008). pressure.

Journal of Biomedical Optics 046009-8 April 2017 • Vol. 22(4) Lee et al.: Application of quantitative second-harmonic generation microscopy. . .

Mariana E. Kersh received her BA degree in English, her MS degree Kimani C. Toussaint is an associate professor in the Department of in mechanical engineering, and her PhD in materials science. She Mechanical Science and Engineering and an affiliate in the Depart- was a postdoctoral McKenzie fellow at the University of Melbourne. ments of Electrical and Computer Engineering, and Bioengineering She is an assistant professor in the Department of Mechanical at the University of Illinois at Urbana-Champaign. He directs the Science and Engineering at the University of Illinois at Urbana- PROBE lab, which focuses on developing optical techniques for Champaign and is head of the Tissue Biomechanics Laboratory. quantitatively imaging collagen-based tissues and investigating the Her research focuses on structure–function relationships in ortho- properties of plasmonic nanostructures for control of near-field optical pedic tissues to investigate musculoskeletal diseases. forces. He is a senior member in SPIE, OSA, and IEEE.

Journal of Biomedical Optics 046009-9 April 2017 • Vol. 22(4) A Retrospective Analysis of the Modified Intervastus Approach

Muthana Sartawi1, Hafizur Rahman2, James Kohlmann1, Ross Leighton3, Mariana E. Kersh2

1. Christie Clinic, 2110 Fox Drive, Suite B, Champaign, IL 61820 2. Department of Mechanical Science and Engineering, University of Illinois at Urbana Champaign, 1206 W Green Street, Urbana, IL 61801

ABSTRACT

The subvastus (SV) approach is a well-known muscle and tendon sparing approach for total knee arthroplasty (TKA) which has been shown in some studies to provide better outcomes in visual analog pain score, knee range of motion (ROM), straight leg raise as well as faster rehabilitation compared to standard medial parapatellar (MP) approach. We previously described a new knee replacement technique known as Modified Intervastus (MIV) approach. The MIV approach is a muscle and tendon sparing approach that is extensile and simple to perform. It may be used in the majority of complex primary cases and revisions. Here we describe the surgical technique for performing the MIV approach and provide functional outcome measures. A total of 127 patients (mean age = 66.75 years) underwent TKA using the MIV approach with one year follow up. Clinical outcomes were assessed by recording duration of hospital stay, visual analog score (VAS), knee range of motion preoperatively, and at several postoperative time points and the length of time required to ambulate independently (without assisted devices). The average length of hospital stay was 2.22 days. Visual analog pain score decreased significantly from preoperative (3.69 ± 2.22) to postoperative day 1 (3.17 ± 1.97) (p < 0.05). Although knee ROM decreased after 1 week of surgery, the ROM increased by 6 weeks after surgery compared to the preoperative ROM and the trend continued over the 1 year follow up. A third of the patients were able to walk independently (without assisted devices) at 2 weeks and 78% at 8 weeks. The MIV approach to the knee is a muscle and tendon sparing approach that is extensile, and reproducible with an accelerated recovery time. The MIV approach offers advantages over the SV approach and may be used for complex primary and revision total knee cases.

Keywords: Total knee arthroplasty; modified intervastus approach; subvastus approach

______This paper has been conditionally accepted to the American Journal of Orthopedics at the time of submission of this dissertation.

1. Introduction Total knee arthroplasty (TKA) is one of the most common orthopedic surgery procedures with more than 600,000 TKA performed annually in the United States and by 2030 the number is expected to reach 3.48 million per year1. Several approaches have been described for total TKA, and the medial parapatellar (MP) approach to the knee is considered the workhorse of total knee replacements. It is an extensile approach that is easy to perform, but may delay active knee extension and straight leg raise after surgery2. Alternative approaches such as the subvastus (SV) and midvastus approach to the knee (Figure 1A) typically allow a more rapid straight leg raise, but may be more challenging to perform and time consuming on morbidly obese patients, muscular patients, and patients with severe deformities3. The intervastus approach (Figure 1A) described by Whiteside and others utilizes the interval between the quadriceps tendon and vastus medialis4. Although it is a simple approach to perform that is extensile, it is not considered tendon sparing since the vastus medialis inserts onto the medial aspect of the quadriceps tendon (Figure 2)5. Therefore, dissecting through this interval, without elevation of the vastus medialis damages the quadriceps tendon and strangulation of the muscle occurs with repair of the arthrotomy site (Figure 3B). This is even more likely in patients with a low lying vastus medialis6.

Figure 1: Illustration of the knee anatomy demonstrating the (A) intervastus approach along with the other known approaches to the knee and the (B) modified intervastus approach to the knee with elevation of the medial edge of the vastus medialis to expose the underlying capsule.

The modified intervastus (MIV) approach described previously7, may be used in the majority of patients undergoing total knee arthroplasty. The advantages of this approach include its extensile nature, similar to a medial parapatellar approach, and preservation of the extensor mechanism and the vastus medialis, leading to a more rapid return to active knee extension than

Figure 2: Intraoperative image of the modified intervastus approach to the knee during repair of the vastus medialis fascia illustrating the attachment of the vastus medialis onto the medial aspect of the quadriceps tendon (arrowheads). is traditionally observed2,7. The approach is simple to perform, easy to close, and is compatible with more extensile approaches such as a quadriceps snip if required in revision scenarios7. However, the functional outcomes of the MIV approach have not been quantified yet. It is unknown whether these outcomes will offer any advantages compared to the SV approach. Therefore, the objective of this study was to measure the functional outcomes of the MIV approach and to compare the results with the previously published SV approach. We hypothesized that utilizing the MIV approach for total knee arthroplasty surgery would lead to early straight leg raise and increase in knee range of motion postoperatively.

Figure 3: Cross sectional anatomy of the knee illustrating the Intervastus approach. (A) Dissection through the intervastus interval preserves vastus medialis but cuts through the quadriceps tendon. (B) arthrotomy site repair with a watertight seal type of closure causing strangulation of the vastus medialis.

2. Surgical Technique The MIV approach preserves the quadriceps tendon and vastus medialis. After exposure of the vastus medialis muscle, the interval between the quadriceps tendon and vastus medialis is identified. The fascia overlying the lateral edge of the vastus medialis is incised where it meets the quadriceps tendon (Figure 4A). The muscle is then bluntly elevated off of the underlying capsule just enough to allow for a capsular repair later (Figure 4B). An arthrotomy is then performed from cephalad to caudal (Figure 4C). This interval may be extended proximally between the vastus intermedius and vastus medialis to expose the distal femur if needed. Closure is then performed by repairing the capsule with absorbable suture and the vastus medialis fascia is repaired back the medial edge of the quadriceps tendon restoring the anatomy (Figure 4D).

Figure 4: Cross sectional anatomy of the knee illustrating the modified intervastus approach. (A) Vastus medialis fascia incision followed by (B) Blunt elevation of the vastus medialis to expose the underlying capsule. The capsule is then incised ensuring to leave a lateral capsular flap for closure. (C) Illustration of the arthotomy in the modified intervastus approach to the knee after elevation of the vastus medialis preserving the quadriceps tendon and vastus medialis. (D) Cross sectional anatomy and magnified view of the knee illustrating closure of the modified intervastus approach. A double layer closure starting with a watertight repair of the capsule followed by repair of the vastus medialis fascia back to the medial edge of the quadriceps tendon restoring the anatomy with no soft tissue tension.

3. Patients and Methods A retrospective review of functional outcomes after TKA using the MIV approach was conducted. The study was approved by the University of Illinois Institutional Review Board. A total of 127 patients of mean age 66.75 years (range 48 – 86 years) with primary osteoarthritis of the knee who were indicated for a total knee replacement with 1 year follow up were included. The patient demographics were shown in Table 1. All the patients underwent TKA using the MIV approach described above by two experienced orthopedic surgeons at the same institution. Patients received spinal anesthesia along with a periarticular pain block intraoperatively. A measured resection technique was used by one surgeon and a gap balancing technique by the other. Patellar resurfacing was done in all cases. Patellar tracking was checked intra-operatively using the ‘no- touch’ technique and the need for a lateral release was noted. Drains were removed at postoperative day 1. Oral opioids were given as needed. Intravenous antibiotics were continued for 24 hours. Oral anticoagulants were used for thromboembolism prophylaxis for 3 weeks. Patients were mobilized on the day of surgery with full weight bearing under the supervision of an experienced physical therapist. Static and dynamic quadriceps exercises were started on the same day of surgery along with active knee range of motion exercises. Pain score, extensor lag, range of motion, walking ability and complications were recorded in all the patients.

Table 1: Patient Demographics

Total no. of patients 127

Male 44 Gender Female 83

Mean ± Standard deviation 66.75 ± 9.12 Age (years) Range 48 – 86

Mean ± Standard deviation 218.38 ± 54.47 Weight (lb) Range 125 – 364

Mean ± Standard deviation 34.10 ± 7.22 BMI (kg/m2) Range 21.1 – 62.5

Duration of hospital stay and visual analog score (VAS) at preoperative and postoperative day 1 were recorded. Patient walking distance with assistance was measured on the day of surgery, after surgery, and on the day of hospital discharge. The patients were assessed

5 preoperatively and postoperatively at 1 week, 2 weeks, 6 weeks, 3 months, 6 months, and 1 year for knee range of motion (ROM). A one-way ANOVA was conducted to compare the preoperative and postoperative day 1 VAS with significance set at p < 0.05 (OriginPro 2015, OriginLab Corporation, Northampton, MA, USA). To identify differences in knee ROM between preoperative and postoperative follow-up periods (1 week, 2 weeks, 6 weeks, 3 months, 6 months, and 1 year), a one-way ANOVA with post hoc Tukey test was used. Significance was set at p < 0.05.

4. Results All patients were able to fully straight leg raise and demonstrate functional knee range of motion by postoperative day 1. The patella tracked centrally in all patients and none required a lateral retinacular release. The majority of patients were discharged in the first 48 hours after surgery on oral narcotics. None required IV narcotics during their hospital stay or a blood transfusion. Two cases were complicated by severe knee skin blistering postoperatively due to a reaction to an adhesive dressing; one of which, was complicated by skin necrosis leading to a flap reconstruction that became infected requiring a two-stage revision. A separate case had an acute postoperative infection that required irrigation and debridement with polyethylene exchange. After a 12 week course of antibiotics the infection was eradicated. All patients reported a high satisfaction rate during their acute postoperative phase. Postoperatively, all patients were able to walk on the day of surgery either independently or with some assistance. On the day of surgery, 10% of patients were able to walk more than 200 feet and this increased to 65% of patients able to walk more than 200 feet on the day of discharge (compare Figure 5A and 5B). Within 2 weeks of surgery, 30% of patients could walk independently (without assisted devices) and increased to 78% by 8 weeks of surgery (Figure 5C).

Figure 5: Walking distance in feet as a percentage of the patients measured on (A) surgery day after the surgery, and (B) discharge day, (C) The percentage of patients able to walk independently reported in terms of number of weeks from surgery.

The average length of hospital stay was 2.22 days with standard deviation of 1.21 (Table 2). Pain assessed using a visual analog score was lower on postoperative day 1 (3.17 ± 1.97) than the preoperative score (3.69 ± 2.22, p < 0.05) (Table 2). Overall, knee ROM significantly increased during the follow up after the surgery. Initially, the ROM decreased after 1 week of surgery (90.82 ± 10.28) compared to preoperative ROM (101.04 ± 19.48, p < 0.001) (Figure 6). At two weeks after surgery, knee ROM returned to the preoperative value (100.70 ± 13.36). By 6 weeks after surgery, knee range of motion was 17 degrees greater than the preoperative range of motion (118.45 ± 11.89, p < 0.001). Knee range of motion remained stable at three and six month assessments, and showed further improvement by one year (126.62 ± 9.81, p < 0.001) compared to the preoperative state (Figure 6). The net improvement in knee range of motion was 25 degrees of increased knee flexion by one year.

Table 2: Outcomes of the Modified intervastus (MIV) approach (VAS: Visual analog score). * indicates significant differences in VAS between preoperative and postoperative day 1 Mean ± Standard deviation

Length of hospital stay (days) 2.22 ± 1.21

Pain level: VAS Preoperative 3.69 ± 2.22 *

Postoperative day 1 3.17 ± 1.97 *

5. Discussion TKA is a successful procedure that restores knee function with pain relief in osteoarthritis patients. While the SV approach for TKA has better outcomes in terms of visual analog pain score, ROM, straight leg raise with faster rehabilitation compared to the standard MP approach8– 10; however, it can be challenging and time consuming when used in morbidly obese and muscular patients3. The SV approach can also increase the risk of complications such as patellar tendon avulsion or medial collateral injury due to the difficulty in exposure specifically for knees with limited ROM11. Here we introduce a new TKA method, the MIV approach, as an alternative to the SV approach overcoming most of these difficulties. With the prevalence of morbid obesity and the market demand for minimally invasive techniques, we believe the MIV approach represents a good approach for surgeons since it is easy to perform, does not require specialized instrumentation and is a reproducible approach even on the most complex deformities. The minimal time added to ensure blunt elevation of the vastus medialis muscle and an anatomic repair of the underlying knee capsule and vastus medialis fascia to the medial edge of the quadriceps tendon allows restoration of the anatomy and a robust double-layered watertight seal closure with no strangulation of the soft tissues. We believe this reproducible muscle and tendon sparing approach, that allows gentle soft tissue

7 handling even on the most complex primary total knee cases, may lead to less soft tissue swelling and therefore less postoperative pain leading to an accelerated recovery.

Figure 6: Knee range of motion (ROM) in degrees for preoperative and postoperative at 1 week, 2 weeks, 6 weeks, 3 months, 6 months, and 1 year. * indicates significant differences between the time periods (p < 0.001). Although knee ROM decreased significantly after postoperative 1 week, but started to increase after 6 weeks of surgery; continued the trend over the follow up of 1 year.

The pain level in this group of patients was reduced after the MIV approach as indicated by the VAS. VAS was significantly decreased at postoperative day 1 compared to preoperative (p < 0.05) indicating patients feeling less pain at the very next day after the surgery. The average VAS at postoperative day 1 from other studies for SV approach was 3.3312, 3.89, 5.0213 which is similar to our MIV approach value of 3.17. Periarticular blocks were available for this study group, and no peripheral nerve blocks were used. Some studies of the SV approach mention the use of peripheral nerve blocks while the others did not describe the method used for treatment or control of postoperative pain. The decreased length of stay, decreased reported pain levels, and the observed increased knee range of motion seen in the MIV and SV approach study groups may in part be attributable to the treatment of postoperative pain. Compared to the SV approach, the MIV approach results in a decrease in average length of hospital stay. Previous studies have reported average length of hospital stays for the SV approach ranging from 3.4 days14 to 8.3 days15, with an average stay across the four reported studies of more than double that of the MIV

8 approach16,17. Decreasing hospital stay length may improve patient outcomes by decreasing the chances of nosocomial infections, medication errors and motivates early patient ambulation. Patient ambulation also increased from the day of surgery to discharge day for MIV approach. Only 10% patients were able to walk with assistance more than 200 feet on the day of surgery; however, this percentage increased to 65% on the discharge day showing an excellent recovery of walking ability within the two days of surgery. Mehta et al. reported that 95% patients undergoing subvastus/midvastus approaches could walk greater than 10 blocks during 6 months follow up18. For the MIV approach, 78% patients were able to walk independently within 8 weeks of surgery. The MIV approach shows improvement in functional outcome in terms of knee ROM from preoperative to postoperative. Although knee ROM decreased significantly at postoperative 1 week which is most likely due to a peak in postoperative pain and swelling; the ROM started to increase at postoperative 6 weeks significantly. After 1 year, the knee ROM increased by 25% compared to preoperative ROM for MIV approach. We found the average knee ROM at postoperative 3 months using the MIV approach was 120.4 degrees. Knee ROM at postoperative 3 months using the SV approach reported by other studies ranged from 87.119 to 12012. The average range of motion reported for the SV approach was 103.5 degrees2,10,18. Similarly, after 1 year, the average ROM by MIV approach was 126.62 degrees whereas average knee ROM from other studies using the SV approach was 114.1 degrees2,10,12,19. Achieving knee range of motion of 120 degrees allows patients to return to their baseline function and perform activities of daily living without limitation. With the prevalence of knee osteoarthritis in the younger population, the importance of increased knee range of motion post TKA becomes more demanding in order for patients to return to work and function at their baseline. The advantages of the MIV approach, in addition to it being a muscle and tendon sparing approach is that it is simple to perform and gives similar exposure to the medial parapatellar approach even on difficult primary cases. It is also as extensile and compatible with maneuvers such as quadriceps snip if required. There is also minimal, if any, muscle retraction preventing muscle injury that may be seen with some of the other muscle sparing approaches to the knee especially in the more difficult primary total knee cases. Another advantage to this approach, as opposed to the other approaches to the knee described is that it allows a double layered closure decreasing the possibility of an arthrotomy dehiscence. Because of this double layered closure, there is no muscle strangulation with repair of the arthotomy site as seen in some of the other approaches to the knee and the vascularity and innervation to the vastus medialis is preserved. One limitation of this study is that it is a retrospective analysis and our results thus far, in terms of functional outcomes, are compared to data from the literature. While our functional measures indicate that there is improvement in range of motion and capacity to walk independently, quantitative assessment of maintenance of muscle function via electromyography or dynamometer measurements remains to be performed. Finally, the long-term results beyond one-year post-operative range of motion have yet to be quantified. Next steps for further

9 validating the benefits of this approach would include quantitative assessment of muscle function and the incorporation of a randomized control trial. In summary, the new MIV approach is easy to perform and is compatible with more extensile approaches such as quadriceps snip if required in revision scenarios. The MIV approach preserved the extensor mechanism along with the vascularity and innervation to the vastus medialis and has similar advantages for functional outcomes as compared to SV approach. Therefore, the MIV approach may be used in the majority of patients undergoing the total knee arthroplasty.

References 1. Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am. 2007;89(4):780-785. doi:10.2106/JBJS.F.00222 2. Dutka J, Skowronek M, Sosin P, Skowronek P. Subvastus and Medial Parapatellar Approaches in TKA: Comparison of Functional Results. Orthopedics. 2011;34(6):148- 153. doi:10.3928/01477447-20110427-05 3. Hofmann AA, Plaster RL, Murdock LE. Subvastus (Southern) approach for primary total knee arthroplasty. Clin Orthop Relat Res. 1991;(269):70-77. http://www.ncbi.nlm.nih.gov/pubmed/1864059. 4. Jojima H, Whiteside L a, Ogata K. Anatomic consideration of nerve supply to the vastus medialis in knee surgery. Clin Orthop Relat Res. 2004;(423):157-160. doi:10.1097/01.blo.0000128642.61260.b3 5. Chavan SK, Wabale RN. Reviewing morphology of Quadriceps femoris muscle. J Morphol Sci. 2016;33(2):112-117. doi:10.4322/jms.053513 6. Holt G, Nunn T, Allen RA, Forrester AW, Gregori A. Variation of the Vastus Medialis Obliquus Insertion and its Relevance to Minimally Invasive Total Knee Arthroplasty. J Arthroplasty. 2008;23(4):600-604. doi:10.1016/j.arth.2007.05.053 7. Sartawi M, Kohlman J, Valle C Della. Modified Intervastus Approach to the Knee. Journal of Knee Surgery. 2017. 8. Hu X, Wang G, Pei F, et al. A meta-analysis of the sub-vastus approach and medial parapatellar approach in total knee arthroplasty. Knee Surgery, Sport Traumatol Arthrosc. 2013;21(10):2398-2404. doi:10.1007/s00167-012-2080-9 9. Jain S, Wasnik S, Mittal A, Hegde C. Outcome of subvastus approach in elderly nonobese patients undergoing bilateral simultaneous total knee arthroplasty: A randomized controlled study. Indian J Orthop. 2013;47(1):45-49. doi:10.4103/0019-5413.106900 10. Kholeif AM, Radwan YA, Mansour AM, Almoalem HA. Early Functional Results of the Subvastus and Medial Parapatellar Approaches in Total Knee Arthroplasty. Med J Cairo Univ. 2017;85(3):849-854. 11. Shah NA, Patil HG, Vaishnav VO, Savale A. Total knee arthroplasty using subvastus approach in stiff knee: A retrospective analysis of 110 cases. Indian J Orthop.

2016;50(2):166-171. doi:10.4103/0019-5413.177582 12. Koh IJ, Kim MW, Kim MS, Jang SW, Park DC, In Y. The Patient’s Perception Does Not Differ Following Subvastus and Medial Parapatellar Approaches in Total Knee Arthroplasty: A Simultaneous Bilateral Randomized Study. J Arthroplasty. 2016;31(1):112-117. doi:10.1016/j.arth.2015.08.004 13. van Hemert WLW, Senden R, Grimm B, van der Linde MJA, Lataster A, Heyligers IC. Early functional outcome after subvastus or parapatellar approach in knee arthroplasty is comparable. Knee Surgery, Sport Traumatol Arthrosc. 2011;19(6):943-951. doi:10.1007/s00167-010-1292-0 14. Schroer WC, Diesfeld PJ, Reedy ME, LeMarr AR. Mini-Subvastus Approach for Total Knee Arthroplasty. J Arthroplasty. 2008;23(1):19-25. doi:10.1016/j.arth.2006.12.100 15. Varela-Egocheaga JR, Suarez-Suarez MA, Fernandez-Villan M, Gonzalez-Sastre V, Varela-Gomez JR, Rodriguez-Merchan C. Minimally invasive subvastus approach: improving the results of total knee arthroplasty: a prospective, randomized trial. Clin Orthop Relat Res. 2010;468(5):1200-1208. doi:10.1007/s11999-009-1160-8 16. Bourke MG, Jull G a, Buttrum PJ, Fitzpatrick PL, Dalton P a, Russell TG. Comparing Outcomes of Medial Parapatellar and Subvastus Approaches in Total Knee Arthroplasty A Randomized Controlled Trial. J Arthroplasty. 2011;27(3):347-353.e1. doi:10.1016/j.arth.2011.06.005 17. Cameron B, Galatz L, Williams GRJ. Factors affecting the outcome of total shoulder arthroplasty. Am J Orthop (Belle Mead NJ). 2001;30(8):613-623. http://ovidsp.ovid.com/ovidweb.cgi?T=JS&PAGE=reference&D=med4&NEWS=N&AN =11520017. 18. Mehta N, Bhat MS, Goyal A, Mishra P, Joshi D, Chaudhary D. Quadriceps sparing (subvastus/midvastus) approach versus the conventional medial parapatellar approach in primary knee arthroplasty. J Arthrosc Jt Surg. 2017;4:15-20. 19. Bridgman SA, Walley G, MacKenzie G, Clement D, Griffiths D, Maffulli N. Sub-vastus approach is more effective than a medial parapatellar approach in primary total knee arthroplasty: A randomized controlled trial. Knee. 2009;16(3):216-222. doi:10.1016/j.knee.2008.11.012