- Email: [email protected]
Abrasive Properties of Braided Polyblend Sutures in Cuff Tendon Repair: An in Vitro Biomechanical Study Exploring Regular and Tape Sutures
Abrasive Properties of Braided Polyblend Sutures in Cuff Tendon Repair: An in Vitro Biomechanical Study Exploring Regular and Tape Sutures Julien Deranlot, M.D., Nathalie Maurel, Ph.D., Amadou Diop, Ph.D., Nathalie Pratlong, M.Eng., Lucas Roche, M.Eng., Roch Tiemtore, M.D., and Geoffroy Nourissat, M.D., Ph.D.
Purpose: This study aimed to evaluate the abrasive properties of different suture materials (tape or regular) on the infraspinatus tendon of sheep. Methods: Four types of sutures were compared: FiberWire (Arthex, Naples FL), FiberTape (Arthrex), Orthocord (DePuy Mitek, Raynham, MA) and ForceFiber (Tornier, Bloomington, MN). Each suture (n ¼ 10) was cycled with a filxed load of 10 N and an alternating motion of the suture through sheep infraspinatus tendon, with an excursion of 30 mm. The migration of the suture as it cut through the tissue was measured at intervals of 5 cycles, up to failure or a total of 50 cycles, or a tendon tear greater than 13 mm. Results: ForceFiber and Orthocord sutures showed a significantly (P < .05) lower amount of abrasion compared with FiberWire and FiberTape: The mean cutting rate (defined as the size of the defect at the end of the test divided by 50 when this number of cycles was reached, or as 13 mm divided by the number of cycles to reach this value when the test was stopped before 50 cycles) was, respectively, 0.04 mm/cycle, 0.12 mm/cycle, 0.11 mm/cycle, 0.32 mm/cycle, and 0.25 mm/cycle. The defect size at 15 cycles was, respectively, 5.7 mm, 5.6 mm, 9.4 mm, 7.7 mm, and 7.4 mm. Although no statistical significance was found, sutures shaped in a tape form (FiberTape) were less aggressive on the tendon than the corresponding sutures in regular form (FiberWire). Conclusions: This study found increased abrasive effects of FiberWire and FiberTape compared with ForceFiber and Orthocord sutures. Clinical Relevance: Currently, surgeons have a large choice of suture materials. Knowledge of biomechanical characteristics of different braided polyblend suture materials could help surgeons decide which suture to use for rotator cuff tears.
T
he suture material used is an important factor in successful rotator cuff repair. Currently, surgeons have the choice between many types of sutures, and recently tape-shaped sutures were introduced based on the rationale that this kind of suture increases the surface of contact between cuff and sutures and reduces abrasive actions and thus decreases the cutting effect of sutures through the tendon. Many studies have compared different suture material regarding strength, abrasion
From Clinique Drouot (J.D.), Paris; Equipe Biomécanique et Remodelage Osseux (EPBRO) (N.M., A.D., N.P., L.R.), Arts et Métiers ParisTech, Paris; Clinique des Maussins (R.T.), Groupe Maussins, Paris; and UR4-UPMC (G.N.); Clinique des Maussins, Groupe Maussins, Paris, France. The authors report that they have no conflicts of interest in the authorship and publication of this article. Received September 23, 2013; accepted June 13, 2014. Address correspondence to Julien Deranlot M.D., Clinique Drouot, 20 Rue Laffitte, 75009 Paris, France. E-mail: [email protected] Ó 2014 by the Arthroscopy Association of North America 0749-8063/13692/$36.00 http://dx.doi.org/10.1016/j.arthro.2014.06.018
properties between sutures, and anchor eyelet,1-3 knot security,4 and failure of bone-tendon constructs, but few of them have focused only on the suture-tendon interface.5,6 To date, no study has compared the abrasive action of different braided polyblend suture materials on tendon. The current study evaluates the abrasive properties against tendon of different braided polyblend suture materials, in regular or tape design, through the infraspinatus tendon of sheep. We suspect that tape sutures are less aggressive inside the tendon and would decrease the cutting effect. Our hypothesis is that the abrasive properties (also called cutting effect) of different braided polyblend suture materials (tape or regular) are different.
Methods Four types of braided polyblend suture were studied: FiberTape (Arthrex, Naples, FL), FiberWire (Arthrex), Orthocord (DePuy Mitek, Raynham, MA), and Force Fiber (Tornier, Bloomington, MN). All sutures used in this study were No. 2.
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 30, No 12 (December), 2014: pp 1569-1573
1569
1570
J. DERANLOT ET AL.
Fig 1. The distal humerus was embedded on a fixation device before the testing setup (x ¼ diaphyseal humeral axis; y ¼ longitudinal axis of the tendon).
Humeral bones and attached infraspinatus tendons were harvested from 10 paired sheep shoulders (aged 6 months to 1 year); they were stored at 20 C and allowed to thaw before testing. They were then prepared by removing all muscle tissue and fixed to a custom-designed device for testing. The dimension and thickness of each infraspinatus tendon were obtained at the beginning of each trial using a Vernier caliper. The distal humerus was embedded (Fig 1) using a low-melting-point alloy (MCP 70, Mining and Chemical Products Ltd, Wellingborough, England). During the fixation procedure, we defined and materialized the referential axis system relative to the humerus using 2 pointing pieces at the medial and lateral humeral epicondyles and a V-shaped piece on the diaphysis, at 60% of the humerus total length. The mediolateral axis of the humerus was defined as the line joining the 2 epicondyles, and the diaphyseal axis was defined by the line joining the midpoint of the epicondyles and the center of the diaphyseal section at the “V” level.
Each suture was tested on each sheep, and the allocation to one side was randomized. Two suture materials were tested for each tendon. Suture was passed through the tendon using a medium-sized Mayo needle 2 mm proximal to the tendon footprint. The 2 sutures tested on the same tendon were symmetrically placed 5 mm apart. Intermittent irrigation with serum spray was performed during the trial to prevent dehydration of the specimen and provide adequate moisture. The embedded specimen was fixed on a mechanical testing machine (Instron 5565; Instron, Norwood, MA). It was placed so that the humeral bone axis was in a horizontal position and the longitudinal axis of the tendon was in a vertical position (Fig 1). The suture, already passed through the tendon, was then loaded vertically. A constant 10-N tensile preload was applied on one of its ends using a weight6 (by cable and pulley). The other end of the suture was linked to the crosshead of the testing machine, allowing the application of alternating motions. To avoid a collapse of the tendon during the test, its free edge was maintained by a thread fixed on the testing frame. The suture was cycled through the tendon at a mean speed of 15 mm/s, with an excursion of 30 mm per cycle resulting in a frequency of 0.5 Hz. The test was run up to 50 cycles or to reach a tendon tear greater than 13 mm. The size of the longitudinal defect (defect size) produced by the migration of the suture through the tendon was measured at intervals of 5 cycles. This measurement was performed optically using a chargecoupled device (CCD) camera placed 10 cm away from the tendon (0.03 mm/pixel). A pointer provided with 2 landmarks was used to manually identify the superior edge of the defect at different steps, and an image was taken for each of these steps: at the beginning of the test before any loading (zero defect), after the application of the 10-N preload, and at the end of each 5 cycles (Fig 2). From these images, the progression of the size of the defect was then calculated using custom-made software (MatLab; Mathworks, Natick, MA) previously validated in several studies4 (Fig 3).
Fig 2. View of the suture before loading (left) and during loading (right). The defect caused by the migration of the suture can be identified and was optically measured.
1571
BRAIDED POLYBLEND SUTURES IN CUFF TENDON REPAIR
Fig 3. Comparison of the different mechanical parameters calculated from optical measurements between suture materials (error bars represent 0.5 standard deviation). Significance of post hoc pairwise comparison tests is indicated: * ¼ P < .05 and ** ¼ P < .01. (A) Defect size after 15 cycles (in millimeters). (B) Number of cycles necessary to reach a 5 mm defect. (C) Mean cutting rate (millimeters/cycle). (D) Defect size after a 10-N preload (in millimeters) (FF, ForceFiber; FW, FiberWire).
Data and Statistical Analysis Four parameters were calculated from the optical measurements: The size of the defect after 15 cycles (minimal number of cycles reached in all cases) The mean cutting rate (defined as the size of the defect at the end of the test divided by 50 when this number of cycles was reached, or as 13 mm divided by the number of cycles to reach this value when the test was stopped before 50 cycles) The size of the defect after the application of the 10 N preload The number of cycles to reach a 5 mm defect Statistical analyses were performed with MedCalc software (MedCalc, Mariakerke, Belgium). Friedman nonparametric tests followed by post hoc pairwise comparison according to Conover7 were performed to compare the 4 sutures in the same procedure. Wilcoxon nonparametric tests were used to directly compare the FiberTape and the FiberWire, which had the same composition but different shapes, to analyze the effect of the shape. Values of P < .05 were considered statistically significant.
Results Tendon measurements before suture passing showed an average thickness of 3.0 mm and an average width of 20.3 mm. The mean tendon length was 18.2 mm. No statistically significant difference was noted between the different sutures for the defect size after the 10-N preload when considering the Friedman test globally comparing the 4 sutures (P ¼ .126). The highest value was obtained for the FiberWire and the lowest value was for the FiberTape. Values were close for the Orthocord and the ForceFiber (Table 1, Fig 3).
Nevertheless, when directly comparing the FiberWire and the FiberTape, which have similar structures but different shapes, using a Wilcoxon test, we found a significantly lower value for the FiberTape (P ¼ .027). Friedman tests found significant differences between sutures regarding the mean cutting rate and the defect size after 15 cycles (P ¼ .00074 and P ¼ .00008, respectively). The post hoc tests indicated that ForceFiber and Orthocord had significantly lower mean cutting rates and defect sizes after 15 cycles than did FiberTape and FiberWire. No significant difference was found between ForceFiber and Orthocord or between FiberWire and FiberTape (Table 1, Fig 3). Regarding the number of cycles to reach a 5-mm defect, the Friedman test found suture type to have a significant effect (P ¼ .0068). Post hoc tests indicated significantly higher values for the Orthocord compared with the FiberTape and the FiberWire (Table 1, Fig 3).
Discussion The suture-tendon interface represents a critical component of repair security in rotator cuff repair,1,6,8 especially in patients with weak tendons. One of the Table 1. Mean Values and Standard Deviation of the Calculated Parameters for the Different Sutures Variable FF O Defect size after a 10-N preload, Mean 2.0 1.8 mm SD 1.4 0.6 Number of cycles to reach a 5-mm Mean 15.3 27.7 defect SD 22.2 28.4 Defect size after 15 cycles, mm Mean 5.7 5.6 SD 1.4 2.6 Mean cutting rate, mm/cycle Mean 0.12 0.11 SD 0.07 0.05
T 1.2 0.5 7.4 4.6 7.4 2.1 0.25 0.13
FW 2.4 0.7 4.4 4.4 9.4 3.5 0.32 0.20
FF, ForceFiber; FW, FiberWire; O, Orthocord; SD, standard deviation; T, FiberTape.
1572
J. DERANLOT ET AL.
main causes of rotator cuff repair failure seems to be the cutting effect of the suture through the tendon. This cutting effect is probably multifactorial, involving inflammation, ischemia, mechanical stress, and abrasive properties of sutures. Increasing the strength of the suture must alert surgeons to the putative increase of the abrasive properties of those sutures. Although Kaplan et al.9 pointed out the fact that braided polyblend suture materials are abrasive on the surgeon’s gloves during arthroscopic shoulder surgery, very few studies have analyzed these abrasive effects on tendons. Wüst et al.1 showed that the abrasive effect is higher with braided sutures than with monofilament sutures, but these results were obtained using porcine cartilage and not tendon. Actually, most studies have focused on the mechanical properties of suture materials10-12 or have explored the failure mechanism of all the constructs (bone/tendon/ knot) under cyclic loading. Only 2 studies explored the suture-tendon interface, but none used the new suture materials. Bisson et al.5 analyzed the effect of the type of suture on the behavior of the suture-tendon interface in a bovine model. They observed no significant differences between any suture materials for ultimate tensile load but found that the most common failure mode during load-to-failure testing was suture breakage in polyester sutures and suture cutting through the tendon in the polyblend sutures. Kowalsky et al.6 also focused on the tendon-suture interface in an in vitro cadaveric study using human infraspinatus tendons. This study was the first to introduce a cyclic movement of the suture back and forth through the tendon during loading as done by Wüst et al.1 for cartilage. They showed that the braided polyblend suture material was significantly more resistant but also more abrasive on tendon and bone compared with monofilament sutures. The current study supports the hypothesis that the tested braided polyblend sutures have different abrasive properties. We focused our study on inner abrasive properties of suture materials against the tendon. Our primary end point was the defect inside the tendon. ForceFiber and Orthocord sutures were found to be significantly less abrasive than FiberWire and FiberTape when analyzing the mean cutting rate and the defect size at 15 cycles. The same trend was observed when considering the number of cycles necessary to reach a 5-mm defect, even if a significant difference was observed only between Orthocord and FiberWire and between Orthocord and FiberTape. Those different abrasive properties could be related to the composition and construction of the sutures. ForceFiber composition includes unique ultrahigh-molecular-weight polyethylene fibers. Orthocord is a composite suture material with absorbable polydioxanone (PDS) and
nonabsorbable polyethylene. The partially absorbable suture is coated with a copolymer (union of 2 monomers). FiberWire has an ultrahigh-molecular-weight polyethylene core and a polyester braided sheath. FiberTape is composed of ultrahigh-molecular-weight polyethylene and polyester yarns braided over a core of FiberWire suture. FiberTape has a width of 2 mm, making it larger than any other tested material. Both include silicone elastomer, and a cyanoacrylate coating acts as a lubricant. The current study does not support the hypothesis that increasing the width of the suture, making it flat, decreases its abrasives properties. Nevertheless, it is to be noted that even if the Friedman test results comparing the 4 sutures were nonsignificant (P ¼ .126) regarding the defect size after the 10-N preload, this parameter was found to be significantly lower for the FiberTape than for the FiberWire in a direct comparison between them when using the Wilcoxon test (P ¼ .027). This can be an indicator of a lower abrasive effect of the FiberTape compared with that of the FiberWire (which have similar compositions but different shapes) in the early phase of the test before the application of alternating cutting motion. This difference was not present later in the final phase of the test. Limitations One study limitation included the use of sheep tendons in our biomechanical tests, which might not be similar to human tendons even though sheep infraspinatus tendons were shown to match the size, shape, thickness, and histologic aspects of human rotator cuff tendons.13 Moreover, the conditions of testing did not reflect the physiological situation of cuff repair and could emphasize the differences between the sutures, but our study was only interested in the abrasion of suture material on the tendon. Another limitation could be the loading conditions we used, inspired by Kowalsky et al.6 According to their work, we chose to apply cyclic loading combining a fixed tensile load and an alternating motion of the suture through the tendon, rather than a purely cyclic tensile load, as performed by Bisson et al.5 This kind of loading was also used by Wüst et al.1 for analyzing the abrasion effect of sutures on cartilage. Even if this did not reflect the exact physiological situation and could emphasize the differences between the sutures, we thought it more relevant to point out the abrasion effect of the sutures on tendon. In addition, the low number of tests is a statistical limitation, especially regarding the difference between FiberTape and FiberWire. A higher number of tests could have showed a difference between them. The current study reminds surgeons that it is not only the shape (flat or regular) or the design of the suture that
BRAIDED POLYBLEND SUTURES IN CUFF TENDON REPAIR
must be considered for choice in cuff repair but also the inner properties of the sutures.
Conclusions This biomechanical study investigated the abrasive properties of tape-shaped suture versus the other braided polyblend sutures. It found increased abrasive effects of FiberWire and FiberTape compared with ForceFiber and Orthocord sutures.
Acknowledgment The authors thank Thomas Lilin (Biomedical Research Centre, Alfort Veterinary School, MaisonsAlfort, France) and Josette Legagneux for their valuable assistance.
References 1. Wüst DM, Meyer DC, Favre P, Gerber C. Mechanical and handling properties of braided polyblend polyethylene sutures in comparison to braided polyester and monofilament polydioxanone sutures. Arthroscopy 2006;22: 1146-1153. 2. Meyer DC, Nyffeler RW, Fucentese SF, Gerber C. Failure of suture material at suture anchor eyelets. Arthroscopy 2002;18:1013-1019. 3. Diop A, Maurel N, Chang VK, Kany J, Duranthon L-D, Grimberg J. Tendon fixation in arthroscopic latissimus dorsi transfer for irreparable posterosuperior cuff tears: An in vitro biomechanical comparison of interference screw and suture anchors. Clin Biomech 2011;26: 904-909. 4. Rousseau B, Diop A, Atlan F, et al. Influence of prolonged immersion on the resistance of arthroscopy knots in biological media. Orthop Traumatol Surg Res 2013;99:138-144.
1573
5. Bisson LJ, Manohar LM, Wilkins RD, Gurske-Deperio J, Ehrensberger MT. Influence of suture material on the biomechanical behavior of suture-tendon specimens: A controlled study in bovine rotator cuff. Am J Sports Med 2008;36:907-912. 6. Kowalsky MS, Dellenbaugh SG, Erlichman DB, Gardner TR, Levine WN, Ahmad CS. Evaluation of suture abrasion against rotator cuff tendon and proximal humerus bone. Arthroscopy 2008;24:329-334. 7. Conover WJ. Rank tests for one sample, two samples, and k samples without the assumption of a continuous distribution function. Ann Statist 1973;6: 1105-1125. 8. Burkhart SS, Diaz Pagàn JL, Wirth MA, Athanasiou KA. Cyclic loading of anchor-based rotator cuff repairs: Confirmation of the tension overload phenomenon and comparison of suture anchor fixation with transosseous fixation. Arthroscopy 1997;13:720-724. 9. Kaplan KM, Gruson KI, Gorczynksi CT, Strauss EJ, Kummer FJ, Rokito AS. Glove tears during arthroscopic shoulder surgery using solid-core suture. Arthroscopy 2007;23:51-56. 10. Lo IKY, Burkhart SS, Athanasiou K. Abrasion resistance of two types of nonabsorbable braided suture. Arthroscopy 2004;20:407-413. 11. Reed SC, Glossop N, Ogilvie-Harris DJ. Full-thickness rotator cuff tears. A biomechanical comparison of suture versus bone anchor techniques. Am J Sports Med 1996;24: 46-48. 12. Barber FA, Feder SM, Burkhart SS, Ahrens J. The relationship of suture anchor failure and bone density to proximal humerus location: A cadaveric study. Arthroscopy 1997;13:340-345. 13. Longo UG, Forriol F, Campi S, Maffulli N, Denaro V. Animal models for translational research on shoulder pathologies: From bench to bedside. Sports Med Arthrosc 2011;19:184-193.
Purpose: This study aimed to evaluate the abrasive properties of different suture materials (tape or regular) on the infraspinatus tendon of sheep. Methods: Four types of sutures were compared: FiberWire (Arthex, Naples FL), FiberTape (Arthrex), Orthocord (DePuy Mitek, Raynham, MA) and ForceFiber (Tornier, Bloomington, MN). Each suture (n ¼ 10) was cycled with a filxed load of 10 N and an alternating motion of the suture through sheep infraspinatus tendon, with an excursion of 30 mm. The migration of the suture as it cut through the tissue was measured at intervals of 5 cycles, up to failure or a total of 50 cycles, or a tendon tear greater than 13 mm. Results: ForceFiber and Orthocord sutures showed a significantly (P < .05) lower amount of abrasion compared with FiberWire and FiberTape: The mean cutting rate (defined as the size of the defect at the end of the test divided by 50 when this number of cycles was reached, or as 13 mm divided by the number of cycles to reach this value when the test was stopped before 50 cycles) was, respectively, 0.04 mm/cycle, 0.12 mm/cycle, 0.11 mm/cycle, 0.32 mm/cycle, and 0.25 mm/cycle. The defect size at 15 cycles was, respectively, 5.7 mm, 5.6 mm, 9.4 mm, 7.7 mm, and 7.4 mm. Although no statistical significance was found, sutures shaped in a tape form (FiberTape) were less aggressive on the tendon than the corresponding sutures in regular form (FiberWire). Conclusions: This study found increased abrasive effects of FiberWire and FiberTape compared with ForceFiber and Orthocord sutures. Clinical Relevance: Currently, surgeons have a large choice of suture materials. Knowledge of biomechanical characteristics of different braided polyblend suture materials could help surgeons decide which suture to use for rotator cuff tears.
T
he suture material used is an important factor in successful rotator cuff repair. Currently, surgeons have the choice between many types of sutures, and recently tape-shaped sutures were introduced based on the rationale that this kind of suture increases the surface of contact between cuff and sutures and reduces abrasive actions and thus decreases the cutting effect of sutures through the tendon. Many studies have compared different suture material regarding strength, abrasion
From Clinique Drouot (J.D.), Paris; Equipe Biomécanique et Remodelage Osseux (EPBRO) (N.M., A.D., N.P., L.R.), Arts et Métiers ParisTech, Paris; Clinique des Maussins (R.T.), Groupe Maussins, Paris; and UR4-UPMC (G.N.); Clinique des Maussins, Groupe Maussins, Paris, France. The authors report that they have no conflicts of interest in the authorship and publication of this article. Received September 23, 2013; accepted June 13, 2014. Address correspondence to Julien Deranlot M.D., Clinique Drouot, 20 Rue Laffitte, 75009 Paris, France. E-mail: [email protected] Ó 2014 by the Arthroscopy Association of North America 0749-8063/13692/$36.00 http://dx.doi.org/10.1016/j.arthro.2014.06.018
properties between sutures, and anchor eyelet,1-3 knot security,4 and failure of bone-tendon constructs, but few of them have focused only on the suture-tendon interface.5,6 To date, no study has compared the abrasive action of different braided polyblend suture materials on tendon. The current study evaluates the abrasive properties against tendon of different braided polyblend suture materials, in regular or tape design, through the infraspinatus tendon of sheep. We suspect that tape sutures are less aggressive inside the tendon and would decrease the cutting effect. Our hypothesis is that the abrasive properties (also called cutting effect) of different braided polyblend suture materials (tape or regular) are different.
Methods Four types of braided polyblend suture were studied: FiberTape (Arthrex, Naples, FL), FiberWire (Arthrex), Orthocord (DePuy Mitek, Raynham, MA), and Force Fiber (Tornier, Bloomington, MN). All sutures used in this study were No. 2.
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 30, No 12 (December), 2014: pp 1569-1573
1569
1570
J. DERANLOT ET AL.
Fig 1. The distal humerus was embedded on a fixation device before the testing setup (x ¼ diaphyseal humeral axis; y ¼ longitudinal axis of the tendon).
Humeral bones and attached infraspinatus tendons were harvested from 10 paired sheep shoulders (aged 6 months to 1 year); they were stored at 20 C and allowed to thaw before testing. They were then prepared by removing all muscle tissue and fixed to a custom-designed device for testing. The dimension and thickness of each infraspinatus tendon were obtained at the beginning of each trial using a Vernier caliper. The distal humerus was embedded (Fig 1) using a low-melting-point alloy (MCP 70, Mining and Chemical Products Ltd, Wellingborough, England). During the fixation procedure, we defined and materialized the referential axis system relative to the humerus using 2 pointing pieces at the medial and lateral humeral epicondyles and a V-shaped piece on the diaphysis, at 60% of the humerus total length. The mediolateral axis of the humerus was defined as the line joining the 2 epicondyles, and the diaphyseal axis was defined by the line joining the midpoint of the epicondyles and the center of the diaphyseal section at the “V” level.
Each suture was tested on each sheep, and the allocation to one side was randomized. Two suture materials were tested for each tendon. Suture was passed through the tendon using a medium-sized Mayo needle 2 mm proximal to the tendon footprint. The 2 sutures tested on the same tendon were symmetrically placed 5 mm apart. Intermittent irrigation with serum spray was performed during the trial to prevent dehydration of the specimen and provide adequate moisture. The embedded specimen was fixed on a mechanical testing machine (Instron 5565; Instron, Norwood, MA). It was placed so that the humeral bone axis was in a horizontal position and the longitudinal axis of the tendon was in a vertical position (Fig 1). The suture, already passed through the tendon, was then loaded vertically. A constant 10-N tensile preload was applied on one of its ends using a weight6 (by cable and pulley). The other end of the suture was linked to the crosshead of the testing machine, allowing the application of alternating motions. To avoid a collapse of the tendon during the test, its free edge was maintained by a thread fixed on the testing frame. The suture was cycled through the tendon at a mean speed of 15 mm/s, with an excursion of 30 mm per cycle resulting in a frequency of 0.5 Hz. The test was run up to 50 cycles or to reach a tendon tear greater than 13 mm. The size of the longitudinal defect (defect size) produced by the migration of the suture through the tendon was measured at intervals of 5 cycles. This measurement was performed optically using a chargecoupled device (CCD) camera placed 10 cm away from the tendon (0.03 mm/pixel). A pointer provided with 2 landmarks was used to manually identify the superior edge of the defect at different steps, and an image was taken for each of these steps: at the beginning of the test before any loading (zero defect), after the application of the 10-N preload, and at the end of each 5 cycles (Fig 2). From these images, the progression of the size of the defect was then calculated using custom-made software (MatLab; Mathworks, Natick, MA) previously validated in several studies4 (Fig 3).
Fig 2. View of the suture before loading (left) and during loading (right). The defect caused by the migration of the suture can be identified and was optically measured.
1571
BRAIDED POLYBLEND SUTURES IN CUFF TENDON REPAIR
Fig 3. Comparison of the different mechanical parameters calculated from optical measurements between suture materials (error bars represent 0.5 standard deviation). Significance of post hoc pairwise comparison tests is indicated: * ¼ P < .05 and ** ¼ P < .01. (A) Defect size after 15 cycles (in millimeters). (B) Number of cycles necessary to reach a 5 mm defect. (C) Mean cutting rate (millimeters/cycle). (D) Defect size after a 10-N preload (in millimeters) (FF, ForceFiber; FW, FiberWire).
Data and Statistical Analysis Four parameters were calculated from the optical measurements: The size of the defect after 15 cycles (minimal number of cycles reached in all cases) The mean cutting rate (defined as the size of the defect at the end of the test divided by 50 when this number of cycles was reached, or as 13 mm divided by the number of cycles to reach this value when the test was stopped before 50 cycles) The size of the defect after the application of the 10 N preload The number of cycles to reach a 5 mm defect Statistical analyses were performed with MedCalc software (MedCalc, Mariakerke, Belgium). Friedman nonparametric tests followed by post hoc pairwise comparison according to Conover7 were performed to compare the 4 sutures in the same procedure. Wilcoxon nonparametric tests were used to directly compare the FiberTape and the FiberWire, which had the same composition but different shapes, to analyze the effect of the shape. Values of P < .05 were considered statistically significant.
Results Tendon measurements before suture passing showed an average thickness of 3.0 mm and an average width of 20.3 mm. The mean tendon length was 18.2 mm. No statistically significant difference was noted between the different sutures for the defect size after the 10-N preload when considering the Friedman test globally comparing the 4 sutures (P ¼ .126). The highest value was obtained for the FiberWire and the lowest value was for the FiberTape. Values were close for the Orthocord and the ForceFiber (Table 1, Fig 3).
Nevertheless, when directly comparing the FiberWire and the FiberTape, which have similar structures but different shapes, using a Wilcoxon test, we found a significantly lower value for the FiberTape (P ¼ .027). Friedman tests found significant differences between sutures regarding the mean cutting rate and the defect size after 15 cycles (P ¼ .00074 and P ¼ .00008, respectively). The post hoc tests indicated that ForceFiber and Orthocord had significantly lower mean cutting rates and defect sizes after 15 cycles than did FiberTape and FiberWire. No significant difference was found between ForceFiber and Orthocord or between FiberWire and FiberTape (Table 1, Fig 3). Regarding the number of cycles to reach a 5-mm defect, the Friedman test found suture type to have a significant effect (P ¼ .0068). Post hoc tests indicated significantly higher values for the Orthocord compared with the FiberTape and the FiberWire (Table 1, Fig 3).
Discussion The suture-tendon interface represents a critical component of repair security in rotator cuff repair,1,6,8 especially in patients with weak tendons. One of the Table 1. Mean Values and Standard Deviation of the Calculated Parameters for the Different Sutures Variable FF O Defect size after a 10-N preload, Mean 2.0 1.8 mm SD 1.4 0.6 Number of cycles to reach a 5-mm Mean 15.3 27.7 defect SD 22.2 28.4 Defect size after 15 cycles, mm Mean 5.7 5.6 SD 1.4 2.6 Mean cutting rate, mm/cycle Mean 0.12 0.11 SD 0.07 0.05
T 1.2 0.5 7.4 4.6 7.4 2.1 0.25 0.13
FW 2.4 0.7 4.4 4.4 9.4 3.5 0.32 0.20
FF, ForceFiber; FW, FiberWire; O, Orthocord; SD, standard deviation; T, FiberTape.
1572
J. DERANLOT ET AL.
main causes of rotator cuff repair failure seems to be the cutting effect of the suture through the tendon. This cutting effect is probably multifactorial, involving inflammation, ischemia, mechanical stress, and abrasive properties of sutures. Increasing the strength of the suture must alert surgeons to the putative increase of the abrasive properties of those sutures. Although Kaplan et al.9 pointed out the fact that braided polyblend suture materials are abrasive on the surgeon’s gloves during arthroscopic shoulder surgery, very few studies have analyzed these abrasive effects on tendons. Wüst et al.1 showed that the abrasive effect is higher with braided sutures than with monofilament sutures, but these results were obtained using porcine cartilage and not tendon. Actually, most studies have focused on the mechanical properties of suture materials10-12 or have explored the failure mechanism of all the constructs (bone/tendon/ knot) under cyclic loading. Only 2 studies explored the suture-tendon interface, but none used the new suture materials. Bisson et al.5 analyzed the effect of the type of suture on the behavior of the suture-tendon interface in a bovine model. They observed no significant differences between any suture materials for ultimate tensile load but found that the most common failure mode during load-to-failure testing was suture breakage in polyester sutures and suture cutting through the tendon in the polyblend sutures. Kowalsky et al.6 also focused on the tendon-suture interface in an in vitro cadaveric study using human infraspinatus tendons. This study was the first to introduce a cyclic movement of the suture back and forth through the tendon during loading as done by Wüst et al.1 for cartilage. They showed that the braided polyblend suture material was significantly more resistant but also more abrasive on tendon and bone compared with monofilament sutures. The current study supports the hypothesis that the tested braided polyblend sutures have different abrasive properties. We focused our study on inner abrasive properties of suture materials against the tendon. Our primary end point was the defect inside the tendon. ForceFiber and Orthocord sutures were found to be significantly less abrasive than FiberWire and FiberTape when analyzing the mean cutting rate and the defect size at 15 cycles. The same trend was observed when considering the number of cycles necessary to reach a 5-mm defect, even if a significant difference was observed only between Orthocord and FiberWire and between Orthocord and FiberTape. Those different abrasive properties could be related to the composition and construction of the sutures. ForceFiber composition includes unique ultrahigh-molecular-weight polyethylene fibers. Orthocord is a composite suture material with absorbable polydioxanone (PDS) and
nonabsorbable polyethylene. The partially absorbable suture is coated with a copolymer (union of 2 monomers). FiberWire has an ultrahigh-molecular-weight polyethylene core and a polyester braided sheath. FiberTape is composed of ultrahigh-molecular-weight polyethylene and polyester yarns braided over a core of FiberWire suture. FiberTape has a width of 2 mm, making it larger than any other tested material. Both include silicone elastomer, and a cyanoacrylate coating acts as a lubricant. The current study does not support the hypothesis that increasing the width of the suture, making it flat, decreases its abrasives properties. Nevertheless, it is to be noted that even if the Friedman test results comparing the 4 sutures were nonsignificant (P ¼ .126) regarding the defect size after the 10-N preload, this parameter was found to be significantly lower for the FiberTape than for the FiberWire in a direct comparison between them when using the Wilcoxon test (P ¼ .027). This can be an indicator of a lower abrasive effect of the FiberTape compared with that of the FiberWire (which have similar compositions but different shapes) in the early phase of the test before the application of alternating cutting motion. This difference was not present later in the final phase of the test. Limitations One study limitation included the use of sheep tendons in our biomechanical tests, which might not be similar to human tendons even though sheep infraspinatus tendons were shown to match the size, shape, thickness, and histologic aspects of human rotator cuff tendons.13 Moreover, the conditions of testing did not reflect the physiological situation of cuff repair and could emphasize the differences between the sutures, but our study was only interested in the abrasion of suture material on the tendon. Another limitation could be the loading conditions we used, inspired by Kowalsky et al.6 According to their work, we chose to apply cyclic loading combining a fixed tensile load and an alternating motion of the suture through the tendon, rather than a purely cyclic tensile load, as performed by Bisson et al.5 This kind of loading was also used by Wüst et al.1 for analyzing the abrasion effect of sutures on cartilage. Even if this did not reflect the exact physiological situation and could emphasize the differences between the sutures, we thought it more relevant to point out the abrasion effect of the sutures on tendon. In addition, the low number of tests is a statistical limitation, especially regarding the difference between FiberTape and FiberWire. A higher number of tests could have showed a difference between them. The current study reminds surgeons that it is not only the shape (flat or regular) or the design of the suture that
BRAIDED POLYBLEND SUTURES IN CUFF TENDON REPAIR
must be considered for choice in cuff repair but also the inner properties of the sutures.
Conclusions This biomechanical study investigated the abrasive properties of tape-shaped suture versus the other braided polyblend sutures. It found increased abrasive effects of FiberWire and FiberTape compared with ForceFiber and Orthocord sutures.
Acknowledgment The authors thank Thomas Lilin (Biomedical Research Centre, Alfort Veterinary School, MaisonsAlfort, France) and Josette Legagneux for their valuable assistance.
References 1. Wüst DM, Meyer DC, Favre P, Gerber C. Mechanical and handling properties of braided polyblend polyethylene sutures in comparison to braided polyester and monofilament polydioxanone sutures. Arthroscopy 2006;22: 1146-1153. 2. Meyer DC, Nyffeler RW, Fucentese SF, Gerber C. Failure of suture material at suture anchor eyelets. Arthroscopy 2002;18:1013-1019. 3. Diop A, Maurel N, Chang VK, Kany J, Duranthon L-D, Grimberg J. Tendon fixation in arthroscopic latissimus dorsi transfer for irreparable posterosuperior cuff tears: An in vitro biomechanical comparison of interference screw and suture anchors. Clin Biomech 2011;26: 904-909. 4. Rousseau B, Diop A, Atlan F, et al. Influence of prolonged immersion on the resistance of arthroscopy knots in biological media. Orthop Traumatol Surg Res 2013;99:138-144.
1573
5. Bisson LJ, Manohar LM, Wilkins RD, Gurske-Deperio J, Ehrensberger MT. Influence of suture material on the biomechanical behavior of suture-tendon specimens: A controlled study in bovine rotator cuff. Am J Sports Med 2008;36:907-912. 6. Kowalsky MS, Dellenbaugh SG, Erlichman DB, Gardner TR, Levine WN, Ahmad CS. Evaluation of suture abrasion against rotator cuff tendon and proximal humerus bone. Arthroscopy 2008;24:329-334. 7. Conover WJ. Rank tests for one sample, two samples, and k samples without the assumption of a continuous distribution function. Ann Statist 1973;6: 1105-1125. 8. Burkhart SS, Diaz Pagàn JL, Wirth MA, Athanasiou KA. Cyclic loading of anchor-based rotator cuff repairs: Confirmation of the tension overload phenomenon and comparison of suture anchor fixation with transosseous fixation. Arthroscopy 1997;13:720-724. 9. Kaplan KM, Gruson KI, Gorczynksi CT, Strauss EJ, Kummer FJ, Rokito AS. Glove tears during arthroscopic shoulder surgery using solid-core suture. Arthroscopy 2007;23:51-56. 10. Lo IKY, Burkhart SS, Athanasiou K. Abrasion resistance of two types of nonabsorbable braided suture. Arthroscopy 2004;20:407-413. 11. Reed SC, Glossop N, Ogilvie-Harris DJ. Full-thickness rotator cuff tears. A biomechanical comparison of suture versus bone anchor techniques. Am J Sports Med 1996;24: 46-48. 12. Barber FA, Feder SM, Burkhart SS, Ahrens J. The relationship of suture anchor failure and bone density to proximal humerus location: A cadaveric study. Arthroscopy 1997;13:340-345. 13. Longo UG, Forriol F, Campi S, Maffulli N, Denaro V. Animal models for translational research on shoulder pathologies: From bench to bedside. Sports Med Arthrosc 2011;19:184-193.