Owing to the rarity of spontaneous resolution, most clinical cases involving tendon injury in dogs require surgical intervention to return patients to normal function.1 Factors that have been shown to improve the success of surgical intervention include the tensile strength of the repair and preventing gap formation at the tenorrhaphy site.2–5 Core suture patterns that have looped configurations, including locking and grasping suture patterns, have been demonstrated to have higher suture pullout strengths, compared with the strength of patterns without looped configurations.6,7 The locking-loop (LL) component of a Kessler suture is the transverse suture that is passed superficial to the longitudinal suture strands. When the construct is loaded, these loop lock around the longitudinally oriented bundles of collagen fibers within the tendon substance.8 In contrast to LLs, grasping loops do not tighten around but instead, pull through tendon fibers and distract with tension.9 Locking configurations used for the suture tenorrhaphy provide greater tensile strength and greater resistance to gap formation, compared with the grasping loops that are commonly used clinically.9
LL core suture patterns have been widely used for use for both human and canine tendon repair owing to their greater ability to resist deformation under tensile loading.2,10 Several modifications to the LL suture have been developed to increase the tensile strength of repairs in both in vivo and ex vivo models.11–14 Winters et al11 used an ex vivo canine flexor tendon model to evaluate looped suture configurations that incorporated an 8-strand repair technique and showed that 8-strand repairs had significantly higher tensile strength and greater stiffness, compared with similar 4-strand repairs. Tang et al12 demonstrated that increasing the length of suture purchase in a core LL pattern from 4 to 7 mm in cadaveric porcine tendons increased the strength of the repair and its resistance to 2-mm gap formation. Hatanaka and Manske15 used a modified locking core suture to repair cadaveric human flexor digitorum profundus tendons. These investigators showed that increasing the cross-sectional area (CSA) of the locking loop component from 10% to 30% to 50% was associated with a linear increase in the tensile strength of the repaired tenorrhaphy construct. In this same study, however, the tensile strength did not increase when the CSA engaged by the core suture was > 50%, even though there were more tendon fibers locked in the LL component of the construct.15 This same finding has also been observed when increasing the number of loops without increasing the volume of tissue engaged by the suture in cadaveric human flexor digitorum profundus tendons.9 A study by Peltz et al16 showed that use of a 4-mm, 4-strand cross-locked cruciate suture provided greater tensile strength and resistance to gap formation, compared with an equivalent 2-mm cross-locked pattern in porcine deep digital flexor tendons. A study by Duffy et al14 also demonstrated that increasing the suture size for a LL from 4-0 or 5-0 to size 0 or 2-0 polypropylene was an important factor, conferring significantly increased tensile strength to repaired canine tendons.
Despite these advances in treatment options, there is currently a paucity of information regarding the effect of loop diameter for LL suture pattens on tendon construct biomechanics. Increasing the diameter of the loop directly influences the number collagen fibrils that interact with the suture under conditions of axial loading. To the authors’ knowledge, the effect of loop diameter, while controlling for other extraneous variables, has not been evaluated to date.
The objective of the study reported here was to evaluate the effect of loop diameter of a LL suture on in vitro tensile strength and resistance to gap formation in a translational canine flexor tendon model. Our null hypothesis was loop diameter of a LL core suture would not affect the biomechanical properties of repaired tendons.
Materials and Methods
Specimen collection
Forty-eight superficial digital flexor tendons (SDFTs) were harvested from 24 healthy medium- to large-breed adult dogs weighing 24 to 29 kg. All dogs were euthanized for reasons unrelated to orthopedic disease and had been donated by a local shelter. Owing to the secondary use of cadaveric specimens, institutional care and use committee approval was not required by our institution for the purposes of this study. A focused orthopedic examination was performed by a single board-certified surgeon (DJD). Cadavers were excluded from the study if there was evidence of musculoskeletal disease or an angular limb deformity on examination. All specimens were harvested within 4 hours after euthanasia by a trained investigator (Y-JC) as described in detail elsewhere.17,18 Briefly, SDFTs were dissected from their origin on the medial humeral condyle to their insertion on the phalanges. The distal humeral metaphysis and pes were left intact to aid with specimen fixation. Surrounding soft tissues were removed without damaging the bone-muscle-tendon unit. Paired SDFTs were wrapped in saline soaked sponges and stored in sealable bags (Ziplock; SC Johnson & Son Inc) that were stored in a thermostatically controlled environment at –20°C in accordance with a validated technique.19 Specimens were thawed at room temperature (21°C) for 11 to 12 hours before testing.19
Surgical repair method
SDFTs were randomly assigned to 1 of 4 groups (n = 12/group) with a random number generator (Research Randomizer). The randomization scheme was controlled so that paired SDFTs (left and right forelimbs from the same cadavers) were not assigned to the same experimental group. For each specimen, a transverse tenotomy was performed with a No. 10 scalpel blade 2 cm distal to the myotendinous junction. The CSA of the distal tendon stump was photographed (iPhone SE Camera; Apple Inc) at a fixed distance of 10 cm adjacent to a surgical ruler (Medline; Northfield, IL) axially aligned with the tendon edge. CSAs were then measured with imaging software (ImageJ; National Institutes of Health) by a single investigator (Y-JC).
All specimens were repaired with a core LL suture pattern as previously described by Moores et al.10 Loops were positioned at a measured distance of 10 mm from the tenotomy site. Groups were randomly assigned to loop diameters of 1, 2, 3, and 4 mm. Loops were measured from the medial and lateral aspects of the SDFT with a digital caliper (stainless steel digital caliper; Neiko) rated to an accuracy of 0.02 mm (Figure 1). All patterns were performed with 2-0 polypropylene suture (Surgipro; Covidien Ltd) with a swaged-on V-20, 26-mm, half-circle, tapered needle. All constructs were repaired by a single trained investigator (Y-JC) under surgical lighting under the close supervision of a board-certified small animal surgeon (DJD) who assisted during the repair.
Illustration of an ex vivo model used to evaluate the effects of loop diameter of a modified Kessler locking-loop (LL) suture on in vitro tensile strength and gapping characteristics of canine flexor tendon repairs. Forty-eight cadaveric superficial digital flexor tendons from 24 adult medium- to large-breed dogs were obtained and transected 2 cm distal to the myotendinous junction. All specimens were then repaired with a core LL suture pattern of 2-0 polypropylene with loops that were 1 mm (A), 2 mm (B), 3 mm (C), or 4 mm (D) in diameter.
Citation: American Journal of Veterinary Research 83, 9; 10.2460/ajvr.21.12.0212
Biomechanical testing
A single load-to-failure test method was used to simulate acute overloading of repaired tendons during the immediate postoperative period. A uniaxial materials testing machine (Model 5967; Instron Inc) was used. For testing, a 4.5-mm stainless steel bolt was placed through the supratrochlear foramen of the humerus in a mediolateral orientation to allow specimens to be connected to the testing machine with a custom-built test jig. Distally, the pes was fixed securely with a bone clamp (SKU 1652-1; Sawbones) that was connected the load cell. Specimens were axially aligned with the materials testing machine to simulate the position of the canine forelimb during the stance phase of the gait cycle. The load transducer recorded the gathered data with customized computer software (Bluehill 3; Instron Inc) at a frequency of 100 Hz. A preload of 2 N was generated within the first 15 to 20 seconds of testing to ensure a consistent resting length and starting condition among repaired constructs. After the defined preload was reached, the system was automatically zeroed and calibrated. Repaired specimens were then distracted at a rate of 2 mm/min until failure, classified as construct failure or an acute decrease in the applied load of > 50%. A high-definition digital camera (Lumix DMC; Panasonic Corp) placed at a measured distance of 25 cm from the loaded specimen was used to film each test at 10 frames/s. Video data were synchronized with the load transducer data with an automatic triggering system. Mode of construct failure was recorded following the completion of each test as either suture breakage or suture pulling through the tendinous tissue.
Following test completion, load-displacement curves were derived with computerized software (Matlab R2018b; Mathworks). Yield load was defined as the first deviation from linearity of the load-displacement curve; peak load was defined as the highest load measured during each test; failure load was defined as the load at which the construct failed or the applied load decreased by > 50% during testing. Gap formation between the tendon ends was assessed with imaging software (ImageJ; National Institutes of Health) to identify the times and loads at which 1- and 3-mm gaps between the tendon ends occurred. Biomechanical and gap formation data were collected by a single investigator (Y-JC) and independently reviewed by another investigator (DJD).
Statistical analysis
An a priori power analysis was performed, which determined that ≥ 12 tendons/group would provide an 80% power to detect a mean ± SD difference of 28 ± 5 N between groups at a 95% confidence level. Data were assessed for normality with a Shapiro-Wilk Test. A χ2 test and mixed linear model were used to compare data between left and right limbs and control for CSA. Continuous variables were normally distributed and summarized as mean ± SD. Load data and gap formation were compared among groups with a mixed model. Proportional distributions of mode of specimen failure were compared with a Fisher exact test. All analyses were performed with commercial software (SAS, version 9.4; SAS Institute Inc), with values of P < .05 considered significant.
Results
For all 48 tendons, suture repair and biomechanical testing were conducted without any observed errors, and no specimens were excluded because of abnormalities involving the bone-muscle-tendon unit. Left and right limbs were equally distributed among experimental groups (P = .572). The mean ± SD CSA was 0.265 ± 0.07 cm2 and did not differ significantly (P = .353) among groups.
Biomechanical assessment
Yield load differed significantly (P = .004) among groups, with yield load for the 4-mm group significantly higher than yield loads for the 1-mm (P = .048) and 2-mm (P = .020) groups, but not significantly (P = .068) different from yield load for the 3-mm group (Table 1). Peak load also differed significantly (P < .001) among groups. Peak loads for the 4-mm (P < .001) and 3-mm (P = .014) groups were significantly higher than peak load for the 1-mm group, and peak load for the 4-mm group was significantly (P < .001) higher than peak load for the 2-mm group. Failure load differed significantly (P < .001) among groups, with failure load for the 4-mm group significantly greater than failure loads for the 1-mm (P < .001) and 2-mm (P = .001) groups.
Mean ± SD yield, peak, and failure loads for canine superficial deep flexor tendons (n = 12/group) that were transected 2 cm distal to the myotendinous junction and then repaired with a core locking loop suture pattern of 2-0 polypropylene with loops that were 1, 2, 3, or 4 mm in diameter.
| Loop diameter | Yield load (N) | Peak load (N) | Failure load (N) |
|---|---|---|---|
| 1 mm | 48.09 ± 12.90a | 54.32 ± 8.09a | 54.17 ± 7.94a |
| 2 mm | 45.90 ± 16.78a | 56.22 ± 7.46a | 56.22 ± 7.46a |
| 3 mm | 60.16 ± 8.73b | 62.80 ± 4.50b | 62.80 ± 4.50b |
| 4 mm | 63.28 ± 9.90b | 68.55 ± 4.57b | 68.49 ± 4.55b |
In each column, values with different superscript letters were significantly (P < .05) different.
Gap formation
Load required to create a 1-mm gap between the tendon ends differed significantly (P < .001) among groups, with load for the 4-mm significantly higher than loads for the 1-mm (P < .001), 2-mm (P < .001), and 3-mm (P < .001) groups (Table 2). Load required to create a 1-mm gap between the tendon ends was also significantly (P = .022) higher for the 3-mm loop than for the 1-mm group. All 48 specimens had evidence of 1-mm gap formation during mechanical testing.
Mean ± SD loads (N) required to create 1- and 3-mm gaps between the tendon ends.
| Loop diameter | 1-mm gap | 3-mm gap |
|---|---|---|
| 1 mm | 9.24 ± 4.97a | 15.05 ± 5.53a |
| 2 mm | 12.52 ± 4.41a | 18.89 ± 5.07a |
| 3 mm | 17.51 ± 8.03b | 24.45 ± 8.40b |
| 4 mm | 31.59 ± 10.13c | 38.55 ± 8.02c |
In each column, values with different superscript letters were significantly (P < .05) different.
Load required to create a 3-mm gap between the tendon ends also differed significantly (P < .001) among groups, with the load for the 4-mm group significantly higher than loads for the 1-mm (P < .001), 2-mm (P < .001), and 3-mm (P < .001) groups (Table 2). During mechanical testing, all but 1 specimen (in the 4-mm group) developed a 3-mm gap between the tendon ends.
Failure mode
Failure mode did not differ significantly (P = 1.000) among experimental groups, with 46 of the 48 (96%) constructs failing as a result of suture breakage. One construct in the 1-mm group and another construct in the 2-mm group failed as a result of suture pulling through the tendinous tissue.
Discussion
The present study evaluated the effect of loop diameter of a LL suture on in vitro tensile strength and resistance to gap formation in a translational canine flexor tendon model. In contrast to our null hypothesis, we found that increasing the loop diameter increased the tensile strength and loads to cause both 1- and 3-mm gaps between the tendon ends. The results of this study show that loop diameter is an important component of the tenorrhaphy, with larger diameter loops conferring increased tensile strength and resistance to gap formation between tendon ends.
In this study, the yield, peak, and failure loads of the 4-mm group were approximately 1.3 times the comparable loads for the 1-mm group. The result of our study are in agreement with results of a human study by Hatanaka and Manske,15 who demonstrated that as the tendon CSA interacting with the suture increased from 10% to 50%, there was a 23% increase in the ultimate tensile strength of experimental repairs. However, in that study, the core pattern was a modified Pennington LL pattern, which is different from the LL core suture used in the study presented here. Although differences in suture pattern and suture size make direct comparison between studies challenging, the relative size of the loop component and its effect on tensile strength agrees with the findings of our study. A study by Xie et al20 evaluated the effects of circle locks of various diameters in a porcine tendon model. The results of that study illustrated that increasing the size of circle locks from 1 to 2 mm increased the ultimate strength; however, that same study found no difference between lock sizes of 2 and 3 mm. In the present study, there were no differences between the 1- and 2-mm loop diameter groups; however, the biomechanical loads were significantly higher for the 3- and 4-mm groups. Possible reasons for these observed differences between studies may relate to that fact that Xie et al20 only used 2 loops during tenorrhaphy repair, whereas the LL pattern in our study had 4 loops. We also recognize that these differences may represent a type-II error. We hypothesize that the number of loops in a LL pattern has a role in construct biomechanics, and this is an area for future investigation.
A study by Dona et al21 compared the size of locking sutures in a 4-strand cross-locked cruciate repair pattern and altered the relative percentage of tendon CSA (10%, 25%, 33%, and 50%) in an ovine translational research model. Investigators in that study illustrated that LLs engaging 25% of the relative CSA of the tendon had superior biomechanical properties, compared with loops engaging greater percentages of the CSA.21 The authors suggested that the larger loops were less efficient at locking and subsequently holding collagen bundles; however, it should be noted that they used 4-0 polypropylene for their core repairs,21 whereas we used 2-0 polypropylene for tenorrhaphy in our study. Prior investigators have found that both the size of the core suture14 and the use of peripheral epitendinous sutures17 play important roles in tenorrhaphy strength and likely account for the differences between studies. However, the differences in the type of tendon evaluated and the biomechanical testing methodology may also have contributed to the difference in results. In our study LLs with loop diameters of 3- and 4-mm diameter had greater tensile strength, compared with LLs with loop diameters of 1 and 2 mm. However, we recognize that different results may have been obtained if different sutures sizes, suture materials, or types of tendons (eg, flatter, thinner, or sheathed) had been evaluated.
Our study demonstrated that increasing the diameter of the LL component of the loop from 1 to 4 mm in diameter increased the loads required to create 1- and 3-mm gaps by 242% and 156%, respectively. Preventing gap formation is important not only to allow earlier controlled loading but also because gap formation has been shown to increase the risk of peritendinous adhesion formation, resulting in decreased tendon gliding and excursion.5 A study by Gelberman et al5 demonstrated the importance of 3-mm gap formation between tendon ends in vivo. Those authors found that gaps ≥ 3 mm between tendon ends resulted in decreased tensile strength and increased risks of rerupture at the surgical site and of a need for reoperation. Similar to the present study, a study by Peltz et al16 found that larger diameter cross-locks (4 mm) had 13% greater resistance to gap formation at the repair site, compared with smaller (2 mm) cross-locks in deep digital flexor tendons from adult porcine forelimbs. In agreement with findings of our study, cross-locks engage a greater number of collagen fibrils that are consequentially tightened and compressed during loading to allow a greater degree of load sharing between the tendon and core suture pattern when load is applied to the construct. However, it should be noted that the association between the size of the loop and gap formation is still debated. Previous authors15,21 argue that the larger the locked component of the loop, the higher the propensity to create a gap between tendon ends. It is postulated that the larger the LL component, the more slack is created during conditions of incremental loading, leading to early separation and gapping at the repair site during axial distraction. However, in the present study, suture size (2-0 polypropylene) was larger than those used in previous studies (3-0 multifilament polyester15 and 4-0 polypropylene20). Larger-caliber suture materials are stiffer, thus requiring greater loads to deform during loading, because larger-caliber sutures have greater tensile strength than the tendon collagen fibers with which the suture interacts. Further studies are necessary to address the relationship between the biomechanical strength and tendency for gap formation with various types and sizes of sutures and different core suture patterns.
Most constructs (46/48 [96%]) in the present study failed as a result of suture breakage. Suture breakage occurs when the inherent strength of the suture or repair pattern is inferior to the tissue-suture interaction. Our results are in agreement with those of prior investigators.14,17 It has been postulated that the locking nature of these suture patterns may cause a stress concentration at the vertical component where the suture crosses the superficial transverse component within the tendon substance.15 The LL component can become kinked at the sharp angle that is created between the longitudinal and vertical components, with suture breakage seen at this location of localized stress.15 Previous studies15,16,20 evaluating the effect of suture size, proportion of the tendon engaged by the suture, and tendon CSA have reported similar failure modes and are in agreement with the findings of our study.
There are some inherent limitations of the present study. First, this was an ex vivo study using healthy tendons that underwent sharp transection, which differs from clinical cases in regard to potential underlying degenerative changes that may be present. Tendon blood supply is an important factor for intrinsic and extrinsic healing, and we were not able to assess the effect of the LL pattern on vascular compromise. The maximum loop diameter in this study was only 4 mm. It is plausible that in larger patients or larger tendons, larger loop diameters may lead to different results than those presented here. In this study, we did not cyclically test our repairs, instead choosing a single distraction-to-failure test to mimic an acute distractive force applied in the postoperative period due to overloading. We recognize that cyclical testing may be more representative of normal limb function during the gait cycle in dogs.22 Finally, we selected cadaveric tendons from medium- to large-breed dogs and used the forelimb SDFT. It should be noted that in most tendons, the 4-mm loop diameter was almost 40% of the tendon width. Thus, results of this study may not apply to thinner or flatter tendons, sheathed tendons, or tendons from smaller dogs.
In conclusion, the diameter of loops used for a LL core suture is an important biomechanical variable that influences construct tensile strength. Increasing the loop diameter increases the tensile strength and resistance to gap formation of the tenorrhaphy in this canine flexor tendon model. In dogs, loop diameters > 3 mm are recommended when the size of the tendon allows. Further studies are necessary to determine the clinical relevance of these findings and the role of loop diameter on tendon blood supply and healing in vivo.
Acknowledgments
No third-party funding or financial support was received in connection with this study or the writing and publication of the manuscript. The authors declare that there were no conflicts of interest related to the study.
References
- 1.↑
Stuart C, William GM. Muscle and tendon disorders. In: Johnston SA, Tobias KM, eds. Veterinary Surgery: Small Animal. Vol 2. 2nd ed. Elsevier; 2018:1316–1323.
- 2.↑
Strickland JW. Development of flexor tendon surgery: twenty-five years of progress. J Hand Surg Am. 2000;25(2):214–235.
- 3.
Silfverskiöld KL, May EJ. Gap formation after flexor tendon repair in zone II. Results with a new controlled motion programme. Scand J Plast Reconstr Surg Hand Surg. 1993;27(4):263–268.
- 4.
May EJ, Silfverskiöld KL. Rate of recovery after flexor tendon repair in zone II. A prospective longitudinal study of 145 digits. Scand J Plast Reconstr Surg Hand Surg. 1993;27(2):89–94.
- 5.↑
Gelberman RH, Boyer MI, Brodt MD, Winters SC, Silva MJ. The effect of gap formation at the repair site on the strength and excursion of intrasynovial flexor tendons. an experimental study on the early stages of tendon-healing in dogs. J Bone Joint Surg. 1999;81(7):975–982.
- 6.↑
Savage R. In vitro studies of a new method of flexor tendon repair. J Hand Surg Br. 1985;10(2):135–141.
- 7.↑
Kessler I, Nissim F. Primary repair without immobilization of flexor tendon division within the digital sheath. An experimental and clinical study. Acta Orthop Scand. 1969;40(5):587–601.
- 9.↑
Hotokezaka S, Manske PR. Differences between locking loops and grasping loops: effects on 2-strand core suture. J Hand Surg Am. 1997;22(6):995–1003.
- 10.↑
Moores AP, Owen MR, Tarlton JF. The three-loop pulley suture versus two locking-loop sutures for the repair of canine achilles tendons. Vet Surg. 2004;33(2):131–137.
- 11.↑
Winters SC, Gelberman RH, Woo SL, Chan SS, Grewal R, Seiler JG. The effects of multiple-strand suture methods on the strength and excursion of repaired intrasynovial flexor tendons: a biomechanical study in dogs. J Hand Surg Am. 1998;23(1):97–104.
- 12.↑
Tang JB, Zhang Y, Cao Y, Xie RG. Core suture purchase affects strength of tendon repairs. J Hand Surg Am. 2005;30(6):1262–1266.
- 13.
Kim HM, Nelson G, Thomopoulos S, Silva MJ, Das R, Gelberman RH. Technical and biological modifications for enhanced flexor tendon repair. J Hand Surg Am. 2010;35(6):1031–1037.
- 14.↑
Duffy DJ, Curcillo CJ, Chang YJ, Moore GE. Effect of suture caliber on the tensile strength of tenorrhaphies in cadaveric canine tendons. Am J Vet Res. 2020;81(9):714–719.
- 15.↑
Hatanaka H, Manske PR. Effect of the cross-sectional area of locking loops in flexor tendon repair. J Hand Surg Am. 1999;24(4):751–760.
- 16.↑
Peltz TS, Haddad R, Scougall PJ, Nicklin S, Gianoutsos MP, Walsh WR. Influence of locking stitch size in a four-strand cross-locked cruciate flexor tendon repair. J Hand Surg Am. 2011;36(3):450–455.
- 17.↑
Duffy DJ, Chang YJ, Fisher MB, Chambers AR, Moore GE. Effect of epitendinous suture caliber on the tensile strength of repaired canine flexor tendons. Am J Vet Res. 2021;82(6):510–515.
- 18.↑
Duffy DJ, Chang YJ, Fisher MB, Moore GE. Effect of partial vs complete circumferential epitendinous suture placement on the biomechanical properties and gap formation of canine cadaveric tendons. Vet Surg. 2020;49(8):1571–1579.
- 19.↑
Goh KL, Chen Y, Chou SM, Listrat A, Bechet D, Wess TJ. Effects of frozen storage temperature on the elasticity of tendons from a small murine model. Animal. 2010;4(9):1613–1617.
- 20.↑
Xie RG, Xue HG, Gu JH, Tan J, Tang JB. Effects of locking area on strength of 2- and 4-strand locking tendon repairs. J Hand Surg Am. 2005;30(3):455–460.
- 21.↑
Dona E, Gianoutsos MP, Walsh WR. Optimizing biomechanical performance of the 4-strand cruciate flexor tendon repair. J Hand Surg Am. 2004;29(4):571–580.
- 22.↑
Pruitt DL, Manske PR, Fink B. Cyclic stress analysis of flexor tendon repair. J Hand Surg Am. 1991;16(4):701–707.
