Effect of Suture Caliber and Number of Core Strands on Repair of Acute Achilles Ruptures: A Biomechanical Study

Abstract

Background:

Controversy exists regarding the ideal Achilles rupture treatment; however, operative treatment is considered for athletes and active patients. The ideal repair construct is evolving, and the effect of suture caliber or number of core strands has not been studied.

Methods:

Simulated mid-substance Achilles ruptures were performed in 24 cadavers. Specimens were randomized to three 6-core-strand style repair constructs: (1) 4 No. 2 sutures and two 2-mm tapes (2T); (2) 2 No. 2 sutures and four 2-mm tapes (4T); (3) 12 (double-6-strand) strand repair (12 No. 2-0 sutures [12S]). Repairs were subjected to a cyclic loading protocol representative of postoperative rehabilitation. These data were compared to a previously published standard open repair technique (6-core strands with No. 2 sutures) on 9 specimens tested under the same conditions.⁶

Results:

No significant elongation differences were observed between the repair groups and the previously published standard repair group in the first 2 stages of the simulated rehabilitation protocol. Both the 2T and 12S repairs survived a significantly greater number of cycles to failure (P = 0.0005, P = 0.0267, respectively) and had a significantly higher failure load (P = .0005, P = .0118, respectively) compared to the previously published data. These 2 constructs consistently survived the advanced stages of the simulated rehabilitation protocol. The majority of repairs failed at the knots.

Conclusions:

In this study, the 2T and 12S constructs survived the later stages of our simulated rehabilitation protocol, suggesting that they may be able to accommodate a more aggressive clinical rehabilitation protocol. Substituting suture-tape for 2 core strands or doubling the core strands with a smaller-caliber suture created a biomechanically stronger construct.

Clinical Relevance:

Achilles repair with an added nonabsorbable, high–tensile strength tape allowed for a stronger construct that may allow for a more aggressive, early rehabilitation protocol and earlier return to function.

Keywords

sports tendon disorders acute Achilles ruptures Achilles repair limited incision repair

Introduction

Despite being the largest tendon in the body, the Achilles is the third most ruptured, with an increasing incidence in certain populations.³⁵ Controversy exists regarding the optimal treatment of ruptures, with options ranging from nonoperative management to open operative repair.^8,13,35,36 Current literature suggests that operative repair reduces rerupture rates and accelerates return to work and return to sport.^5,15,27,35 Recent Level I literature also suggests that operative repair increases strength and improves clinical outcomes compared with nonoperative treatment at 6 and 18 months;²⁰ therefore, surgery is often recommended for athletes and active populations. Historically, open operative repair is associated with significant operative complications (27% in a recent meta-analysis¹⁵), including wound infection (3.9%¹⁵), sural nerve injury, and formation of adhesions. Yet newer, minimally invasive techniques have mitigated these complications and make operative repair more favorable in select populations.^11,25

A recent biomechanical cadaveric study found repair elongation to be significantly increased in minimally invasive techniques when compared to an open repair, but ultimate load was similar among both groups.⁶ Thus, improvement in repair technique is still needed to optimize Achilles rupture treatment. Recent literature suggests that clinical rehabilitation protocols consisting of both early weightbearing and early motion are critical to improving clinical outcomes following Achilles tendon repair.^3,4 Therefore, further improvement of the biomechanical strength of an open repair technique may allow patients to progress to an earlier and more aggressive postoperative rehabilitation protocol without risk of extensive gapping or elongation at the repair site.

Various factors are known to affect the repair strength of tendons in other anatomical locations and experimental models. The number of core strands and suture caliber has been found to significantly affect the strength of flexor tendon repairs of the forearm.²⁹ Suture number has also been reported to determine the strength in a bovine rotator cuff repair model.¹⁶ To our knowledge, the effect of suture caliber or number of core strands on the repair of acute Achilles rupture has not been studied. We hypothesized that varying the number of core strands or suture caliber used in a mini-open Achilles repair would significantly improve repair strength and decrease elongation in response to cyclic loading representative of an early rehabilitation protocol. Varying these components may create a construct that allows for a more aggressive rehabilitation protocol that could improve early return to function and clinical outcomes. This is particularly imperative in athletic populations where short- and long-term recovery are equally important.

Methods

Specimens

Thirty-six fresh-frozen human foot and ankle (mid-tibia to toe-tip) cadaveric specimens with no history of an Achilles injury, surgery, or other definitive foot and ankle pathology were used for the final analysis in this study. An equal number of adult male and female specimens under the age of 65 years were used. Twenty-four specimens were randomly assigned to one of 3 repair methods of varying suture caliber and number: (1) an open repair using a No. 2 braided polyethylene/polyester multifilament suture (FiberWire, Arthrex Inc) for 4 of the 6 core strands and 2-mm braided polyethylene/polyester multifilament sutures (FiberTape, Arthrex Inc) for the remaining 2 core strands (2T); (2) an open repair using No. 2 suture for 2 core strands and 2-mm suture tape for the remaining 4 core strands (4T); (3) an open repair using No. 2-0 sutures with 12 (double 6-strand) core strands (12S) (Figure 1). Similar biomechanical methods and testing were performed in a previous study at the same institution, and therefore, these data⁶ were compared to the results of this study. This group of specimens (n = 9) underwent an established 6-core-strand mini-open repair consisting of No. 2 sutures and an epitenon stitch and will be referred to as the “standard repair” when comparing to the 3 experimental groups investigated in this study. Additionally, 3 specimens (n = 3) were used for an assessment of the “intact” strength, stiffness, and elongation of the native Achilles tendon. The inclusion of these specimens brought the total specimen count for analysis in the current study to 36.

There is no standard of care in regard to suture type, number of core strands, or suture caliber in Achilles rupture repair. All repair constructs chosen for this study were modifications of an established 6-core-strand mini-open repair construct.⁶ The braided polyethylene/polyester multifilament suture tape repair constructs were chosen because they taper to a No. 2 suture. This suture can also fit through the PARS (Arthrex, Naples, FL) minimally invasive repair device and could therefore be tested in a minimally invasive model in future studies. The 12-core strand No. 2-0 (0.3-mm diameter) suture repair construct was chosen because it would most closely approximate the diameter of a No. 2 suture (0.5-mm) and minimize additional suture material in the repair.

Operative Technique

The Achilles tendon was exposed through a 3-cm, longitudinal incision on the posteromedial aspect of the ankle (only 1 cm longer than a horizontal percutaneous incision). Through this incision, a standardized, simulated midsubstance rupture was created 6 cm proximal to the center of the Achilles tendon calcaneal insertion. In an effort to re-create the mop-end of an Achilles rupture seen clinically, several asymmetric cuts (5-10 cuts, 1-2cm long) with a No. 15 blade were made to divide the native tendon. The Achilles tendon was then repaired according to the randomly assigned suture caliber and core-strand number and consisted of 3 modified Kessler sutures positioned in the anterior, posteromedial, and posterolateral portions with adequate suture purchase of each tendon stump (Figure 1). For all techniques, repair sutures were tied with 5 square knots while the foot was held in gentle plantarflexion to reapproximate the tendon ends (Figure 2). A running epitenon stitch was placed across the width of the tendon repair using 3-0 absorbable monofilament suture (Monocryl, Ethicon Inc) (Figure 3).

Specimen Preparation and Biomechanical Testing

Following repair, the Achilles tendons were isolated through meticulous dissection to preserve the integrity of the repair. Calcanei, including the Achilles insertion, were removed using a sagittal oscillating bone saw. Calcanei were potted in polymethyl methacrylate (PMMA) before biomechanical testing. Each repaired Achilles tendon and potted calcaneus were anatomically oriented and secured in an adjustable fixture and rigid clamp attached to the base of a dynamic testing machine (Instron ElectroPuls E10000; Instron Systems, Norwood, MA). A standardized 6-cm length of tendon proximal to the repair site was whip-stitched with No. 2 braided polyethylene/polyester multifilament looped suture (No. 2 FiberLoop; Arthrex Inc), wrapped helically with 24-gauge wire, and rigidly secured to the testing machine actuator in a custom steel clamp. This was the same setup as a previous Achilles tendon repair study⁶ (Figure 4).

All groups were subjected to a cyclic loading protocol representative of progressive early rehabilitation: 250 cycles at 1 Hz for each of the following stages; (1) 20-100 N, (2) 20-200 N, (3) 20-300 N, and (4) 20-400 N.⁶ Loads were selected to mimic a progressive, postoperative rehabilitation protocol and were based on previous literature describing load ranges experienced by the Achilles tendon during passive ankle flexion (20-100 N) and walking in a cam walker with (190 N) and without (369 N) a 1-inch heel lift.^1,7,22,28 Repairs that survived all 1000 cycles were pulled to failure. Cycles-to-failure, failure load, construct stiffness, and construct elongation were assessed to determine the strength and stability of each repair technique. Note that failure load was taken to be the maximum cyclic load corresponding to the respective 250 Hz loading stage for the repairs that failed during cyclic loading. Stiffness was calculated at the 20th cycle of the first loading block to allow for a “settling in” process whereby the repair constructs reached a steady elongation rate.

Study Groups and Statistical Analysis

As a simplification of the full group comparison analysis, a statistical power calculation was made assuming nonparametric independent testing of central tendency. Assuming 2-tailed testing, an overall alpha level of 0.05 and a Bonferroni correction for all comparisons among 4 groups, 8 specimens per group was sufficient to detect an effect size of d = 2.06 with 80% power.

Displacement values after the 1st loading stage were compared using a non-parametric Kruskal-Wallis ANOVA test. Pairwise comparisons were then conducted with the Nemenyi post-hoc procedure, which is a nonparametric analog to the Tukey test. P-values less than .05 were considered significant. All analyses were conducted using the statistical programming language R³² with the packages PMCMR.³¹

Results

During biomechanical testing, all repairs evaluated in this study survived the first 2 loading blocks (20-100 N and 20-200 N, 250 cycles each). With a maximum of 1000 cycles, the 2T repair survived a median of 1000 cycles (range 507-1000). The 4T repair survived a median of 791 cycles (range 520-1000), and the 12S repair survived a median of 805 cycles (range 510-1000). The standard 6-core-strand repairs survived a median of 519 cycles (range 252-583). No significant differences were observed when comparing the cycle-to-failure among the 3 repair methods evaluated in this study. However, both the 2T repair and 12S repair survived a significantly greater number of cycles to failure when compared to the standard 6-core-strand repair reported in a previous study (P = .0005, P = .027, respectively).

The standard 6-core-strand repair had a 56% survival rate through the second loading block, and no specimens within this group survived the third loading block (501-750 cycles). The survival rates through the 4 cyclic loading blocks for the 3 repairs evaluated in this study are shown in Figure 5. All specimens that survived all 4 loading blocks (1000 cycles) were immediately pulled to failure. Both the 2T and 12S repair groups had a significantly higher failure load than the standard 6-core-strand repair (P = .0005, P = .0118, respectively). The median failure load for each repair was 2T (median 694 N, range 300-859), 4T (401 N, range 300-790), 12S (401 N, range 300-729, 6-core-strand (300 N, range 200-300) (Figure 6).

By the end of the first loading block (250 cycles), no repair group exhibited a significant difference (all P > .28) in measured elongation (displacement) when compared to the others. All repairs evaluated in this study exhibited similar elongation to the standard, mini-open 6-core-strand repairs (Figure 7). By the end of the first loading block, the median elongation was 5.3 mm (range 3.7-7.4) for this group. The median elongation of the remaining repairs was 2T (median 4.6 mm, range 3.8-6.0), 4T (median 4.3 mm, range 3.1-5.4), and 12S (median 5.1 mm, range 3.0-5.7). Failure was defined by an extreme drop in the measured load, resulting from failure at the knots or suture-tendon interface, or repair elongation that exceeded the maximum actuator travel distance of the testing machine (≥3 cm). The primary mechanism of failure was knot failure and a combination of suture cutout at the suture-tendon interface. The suture tapes were more prone to knot failure, whereas the No. 2-0 and No. 2 sutures frequently cut through the tendon.

Within the first loading block, no repair construct revealed a significant difference (all P > .20) in stiffness when compared to the others. When analyzing the 20th cycle of the first loading block, the median stiffness for each construct was standard repair (median 61 N/mm, range 52-68), 2T (54 N/mm, range 49-62), 4T (53 N/mm, range 45-60), and 12S (55 N/mm, range 43-64). No formal comparison was made between the repair constructs and the 3 specimens tested in the intact state because of the small sample size. However, the stiffness for the intact state was consistent (74, 79, and 79 N/mm) and 30% to 48% greater than the median stiffness for any of the repair constructs tested.

Discussion

In this study, a 12-core-strand repair with No. 2-0 and a repair where 2 of the 6 strands were substituted with suture tape, had significantly higher ultimate strength (cycles-to-failure) and ultimate failure load with similar gapping in the first 250 cycles when compared to an established 6-core-strand No. 2 suture construct. The 2 experimental constructs were not significantly different from each other, though there was a trend that favored the 2T construct in terms of ultimate strength (median 1000 cycles, range 507-1000) and maximum failure load (median 694 N, range 300-859). Because both of these repairs (12S and 2T) had good survival rates in the later stage of our simulated rehabilitations protocol, these constructs could potentially allow for immediate weightbearing in a neutral foot position.

Two recent systematic reviews of clinical rehabilitation protocols for Achilles rupture following repair suggest that both early weightbearing and early motion are essential to improving clinical outcomes.^2-4 Both of these studies recommend protecting weightbearing patients in an equinus position with heel wedges or a plantar flexed ankle-foot orthosis for the first 3 weeks. A construct that skips the equinus protective stage of this protocol may allow for earlier return of motion and strength. Moreover, a stronger repair construct could help mitigate the risk of rerupture in the vulnerable inflammatory stages of tendon healing.³⁰ This could allow a patient to strengthen the muscle in a period where many clinicians caution patients during the rehabilitation protocol.^30,33 Additional clinical rehabilitation studies following repair and a better understanding of loading frequency of the repair are needed to prove this assumption.

In terms of suture repair constructs, it has been shown there is no significant difference in strength between a double Bunnell, double Kessler, and double Krackow stitch.²⁴ Modifications to the Krackow stitch, such as the “Gift Box,” have been shown to double the repair strength when compared to a normal Krackow suture (168 N vs 81 N).¹⁹ The “triple bundle” repair by Jaakkola et al was shown to have significantly greater strength when compared to a 4-strand Krackow repair (453 N vs 161 N).¹⁴ Yet, we have concerns that these repair constructs may strangulate the healthy tendon tissue and prolong healing. The results reported here show significantly stronger constructs, with mean failure loads of 625 N and 468 N for the 2T and 12S repairs, respectively. The standard 6-core-strand technique could not survive loads greater than 200 N.⁶ The authors believe the 2T repair construct was a biomechanically superior construct in terms of strength, with less potential to strangulate the tissue.

Furthermore, it has been reported that stress shielding, as in cases of immobilization, results in reduced structural properties¹⁷ and elasticity of tendons,¹² as well as overall loss of alignment in collagen fibers.¹² The 2T and 12S repairs in this study were ultimately stronger constructs and had similar stiffness to the other repairs; therefore, there is no increased risk of stress shielding at the repair site compared to other repair constructs. Early mobilization following repair has shown to accelerate the collagen remodeling process in an injured tendon.³⁰ Thus, a repair construct that could potentially allow for immediate mobilization and weight bearing (such as the 2T and 12S), may ultimately be replaced by more organized tendon.

In this study, the 4T construct seemed to fail earlier than the 2T construct, though these results were not statistically significant. We observed that many of the FiberTape constructs failed at the knots. However, in terms of the 4T constructs, it seemed that the sutures simply “sawed” through the tendon and failed at the suture-tendon interface. Gnandt et al recently performed a study comparing No. 2 FiberWire with FiberTape in a Mason-Allen, whip-stitch, and Krackow repair of simulated tendon ruptures in 15 matched cadaveric specimens.⁹ They found that pull-through was the most common mode of failure, with the FiberTape failing slightly more than the No 2. FiberWire (58.3% vs 55%). They also reported significantly greater failure load with the FiberTape in the whip-stitch and Krackow stitch models when compared with No. 2 FiberWire (428 N vs 331 N (P = .001) and 709 N vs 529 N (P = .015), respectively). It should be noted, however, that the whip-stitch model was performed in the quadriceps tendon and the Krackow model was performed in the Achilles. Nevertheless, these results seem to corroborate with what is reported here.

There are several limitations to this study that should be noted. First, it is a time zero, ex vivo biomechanical investigation that does not account for tendon healing that would aid the repair construct strength throughout rehabilitation. Second, the simulated tendon ruptures created in this study were not the same as the “mop-ended” ruptures seen clinically. We made several asymmetric cuts to simulate a shredded tendon tear, but it is still different from an actual Achilles rupture (Figure 2). Another limitation to this study includes the clinical practicality of bulky suture tape knots within the operative wound (Figure 3). We believe the wound could accommodate a 2T repair because this suture is located on the volar aspect of the construct. Moreover, releasing the flexor hallucis longus fascia would allow for more room to accommodate this larger suture. Lastly, the reported values of elongation included the entire tested construct, including contributions due to elongation of the Achilles tendon tissue itself and not repair site gapping directly. Testing of intact specimens should give the reader an idea of how much native tendon elongation contributes to this measurement.

Despite significant rehabilitation, many patients may not regain full calf strength even after several years following Achilles rupture.^10,23 Predisposing factors to strength deficits include tendon elongation, older age, rerupture, and prolonged immobilization.^18,21,26,34 The goal of this study was to create a biomechanically superior repair construct that could allow for an early and more aggressive early rehabilitation program and potentially eliminate or minimize the postoperative immobilization period. We believe the 2T repair construct may accommodate a rehabilitation program with early weightbearing without heel lifts and early strengthening that could help mitigate strength deficits seen after Achilles repair. We hypothesize that this repair construct would facilitate earlier return to preinjury function in athletic patients. Future directions for this study include developing a clinical rehabilitation protocol that will allow us to test this hypothesis.

Footnotes

Acknowledgements

The authors would like to thank Grant J. Dornan, MS, for his help with statistical analysis. The authors acknowledge Arthrex, Inc, for in-kind donations of operative supplies.

Declaration of Conflicting Interests

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The authors report nonfinancial support from Arthrex Inc during the conduct of the study. Thomas O. Clanton, MD, is a paid consultant for and receives from Arthrex, Inc.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Akizuki

Gartman

Nisonson

Ben-Avi

McHugh

MP.

The relative stress on the Achilles tendon during ambulation in an ankle immobiliser: implications for rehabilitation after Achilles tendon repair. Br J Sports Med. 2001;35(5):329-333; discussion 333-324.

Beskin

Sanders

Hunter

Hughston

JC.

Surgical repair of Achilles tendon ruptures. Am J Sports Med. 1987;15(1):1-8.

Braunstein

Baumbach

Boecker

Carmont

Polzer

Development of an accelerated functional rehabilitation protocol following minimal invasive Achilles tendon repair [published online ahead of print September 26, 2015]. Knee Surg Sports Traumatol Arthrosc. doi:10.1007/s00167-015-3795-1.

Brumann

Baumbach

Mutschler

Polzer

Accelerated rehabilitation following Achilles tendon repair after acute rupture—development of an evidence-based treatment protocol. Injury. 2014;45(11):1782-1790.

Cetti

Christensen

Ejsted

Jensen

Jorgensen

Operative versus nonoperative treatment of Achilles tendon rupture. A prospective randomized study and review of the literature. Am J Sports Med. 1993;21(6):791-799.

Clanton

Haytmanek

Williams

. A biomechanical comparison of an open repair and 3 minimally invasive percutaneous Achilles tendon repair techniques during a simulated, progressive rehabilitation protocol. Am J Sports Med. 2015;43(8):1957-1964.

Demetracopoulos

Gilbert

Young

Baxter

Deland

JT.

Limited-open Achilles tendon repair using locking sutures versus nonlocking sutures: an in vitro model. Foot Ankle Int. 2014;35(6):612-618.

Ebinesan

Sarai

Walley

Maffulli

Conservative, open or percutaneous repair for acute rupture of the Achilles tendon. Disabil Rehabil. 2008;30(20-22):1721-1725.

Gnandt

Smith

Nguyen-Ta

McDonald

LeClere

LE.

High-tensile strength tape versus high-tensile strength suture: a biomechanical study. Arthroscopy. 2016;32(2):356-363.

10.

Horstmann

Lukas

Merk

Brauner

Mundermann

Deficits 10-years after Achilles tendon repair. Int J Sports Med. 2012;33(6):474-479.

11.

Hsu

Jones

Cohen

Davis

Ellington

Anderson

RB.

Clinical outcomes and complications of percutaneous Achilles repair system versus open technique for acute Achilles tendon ruptures. Foot Ankle Int. 2015;36(11):1279-1286.

12.

Ikoma

Kido

Nagae

. Effects of stress-shielding on the dynamic viscoelasticity and ordering of the collagen fibers in rabbit Achilles tendon. J Orthop Res. 2013;31(11):1708-1712.

13.

Inglis

Scott

Sculco

Patterson

AH.

Ruptures of the tendo achillis. An objective assessment of surgical and non-surgical treatment. J Bone Joint Surg Am. 1976;58(7):990-993.

14.

Jaakkola

Hutton

Beskin

Lee

GP.

Achilles tendon rupture repair: biomechanical comparison of the triple bundle technique versus the Krakow locking loop technique. Foot Ankle Int. 2000;21(1):14-17.

15.

Jones

Khan

Carey Smith

RL.

Surgical interventions for treating acute Achilles tendon rupture: key findings from a recent cochrane review. J Bone Joint Surg Am. 2012;94(12):e88.

16.

Jost

Khair

Chen

. Suture number determines strength of rotator cuff repair. J Bone Joint Surg Am. 2012;94(14):e100.

17.

Jung

Fisher

Woo

SL.

Role of biomechanics in the understanding of normal, injured, and healing ligaments and tendons. Sports Med Arthrosc Rehabil Ther Technol. 2009;1(1):9. doi:10.1186/1758-2555-1-9.

18.

Kangas

Pajala

Ohtonen

Leppilahti

Achilles tendon elongation after rupture repair: a randomized comparison of 2 postoperative regimens. Am J Sports Med. 2007;35(1):59-64.

19.

Labib

Rolf

Dacus

Hutton

WC.

The “Giftbox” repair of the Achilles tendon: a modification of the Krackow technique. Foot Ankle Int. 2009;30(5):410-414.

20.

Lantto

Heikkinen

Flinkkila

. A prospective randomized trial comparing surgical and nonsurgical treatments of acute Achilles tendon ruptures. Am J Sports Med. 2016;44(9):2406-2414.

21.

Lee

Goldsmith

Nicholas

. Optimizing Achilles tendon repair: effect of epitendinous suture augmentation on the strength of Achilles tendon repairs. Foot Ankle Int. 2008;29(4):427-432.

22.

Lee

Sileo

Kremenic

. Cyclic loading of 3 Achilles tendon repairs simulating early postoperative forces. Am J Sports Med. 2009;37(4):786-790.

23.

Leppilahti

Lahde

Forsman

. Relationship between calf muscle size and strength after Achilles rupture repair. Foot Ankle Int. 2000;21(4):330-335.

24.

McCoy

Haddad

SL.

The strength of Achilles tendon repair: a comparison of three suture techniques in human cadaver tendons. Foot Ankle Int. 2010;31(8):701-705.

25.

McMahon

Smith

Hing

CB.

A meta-analysis of randomised controlled trials comparing conventional to minimally invasive approaches for repair of an Achilles tendon rupture. Foot Ankle Surg. 2011;17(4):211-217.

26.

Metz

van der Heijden

Verleisdonk

Andrlik

van der Werken

Persistent disability despite sufficient calf muscle strength after rerupture of surgically treated acute Achilles tendon ruptures. Foot Ankle Spec. 2011;4(2):77-81.

27.

Metz

Verleisdonk

van der Heijden

. Acute Achilles tendon rupture: minimally invasive surgery versus nonoperative treatment with immediate full weightbearing—a randomized controlled trial. Am J Sports Med. 2008;36(9):1688-1694.

28.

Orishimo

Burstein

Mullaney

. Effect of knee flexion angle on Achilles tendon force and ankle joint plantarflexion moment during passive dorsiflexion. J Foot Ankle Surg. 2008;47(1):34-39.

29.

Osei

Stepan

Calfee

. The effect of suture caliber and number of core suture strands on zone II flexor tendon repair: a study in human cadavers. J Hand Surg Am. 2014;39(2):262-268.

30.

Pneumaticos

McGarvey

Mody

Trevino

SG.

The effects of early mobilization in the healing of Achilles tendon repair. Foot Ankle Int. 2000;21(7):551-557.

31.

Pohlert

The Pairwise Multiple Comparison of Mean Ranks Package (PMCMR). R Package. Vienna, Austria: R Foundation for Statistical Computing; 2014.

32.

R Development Core Team. R: A language and enviroment for statistical computing [computer program]. Vienna, Austria: R Foundation for Statistical Computing; 2015.

33.

Rettig

Liotta

Klootwyk

Porter

Mieling

Potential risk of rerupture in primary Achilles tendon repair in athletes younger than 30 years of age. Am J Sports Med. 2005;33(1):119-123.

34.

Silbernagel

Steele

Manal

Deficits in heel-rise height and Achilles tendon elongation occur in patients recovering from an Achilles tendon rupture. Am J Sports Med. 2012;40(7):1564-1571. Computing; 2015.

35.

Wilkins

Bisson

LJ.

Operative versus nonoperative management of acute Achilles tendon ruptures: a quantitative systematic review of randomized controlled trials. Am J Sports Med. 2012;40(9):2154-2160.

36.

Wong

Barrass

Maffulli

Quantitative review of operative and nonoperative management of Achilles tendon ruptures. Am J Sports Med. 2002;30(4):565-575.