In vivo comparison of a fixed loop (EndoButton CL) with an adjustable loop (TightRope RT) device for femoral fixation of the graft in ACL reconstruction: A prospective randomized study and a literature review

Abstract

Introduction:

There is a lack of in vivo studies comparing the functional outcome and knee stability after anterior cruciate ligament reconstruction (ACLR) using fixed loop (EndoButton (EB) CL) and adjustable loop (TightRope (TR) RT) devices for femoral fixation of soft tissue grafts.

Materials and methods:

Functional outcomes were assessed in terms of the International Knee Documentation Committee (IKDC) and Lysholm scores, knee stability by anteroposterior laxity and side-to-side difference (SSD) using KT-1000 arthrometer. The evaluation was performed preoperatively and post-operatively at 6 months and 2 years.

Results:

Both groups were matched in terms of demographic, preoperative, intraoperative and post-operative covariates. EB (n = 52) appeared to have better IKDC and Lysholm scores at 6 months post-operative when compared to TR (n = 50). However, at a final follow-up of 2 years, the results were similar. The anterior tibial translation and SSD were statistically insignificant between the two groups at 6 months and 2 years.

Conclusion:

ACLR using EB or TR for femoral fixation gives substantially equivalent functional results and knee stability at mid-term follow-up.

Keywords

anterior cruciate ligament reconstruction EndoButton knee stability TightRope

Introduction

Arthroscopic anterior cruciate ligament reconstruction (ACLR) is currently the gold standard treatment for anterior cruciate ligament (ACL) tear, and the most commonly used free tendon grafts are autologous bone-patellar tendon-bone and hamstring tendons.¹ The objectives of ACLR are early rehabilitation and return to pre-injury activity.² Numerous graft fixation options are available for both tibia and femur, with continuous advancement in technology.¹ The weakest link in the early post-operative period are the fixation points of the graft in femoral and tibial tunnels.³ Cortical suspensory fixation is the current standard femoral fixation method for soft tissue grafts and is in widespread use.⁴ EndoButton (EB) CL (Smith & Nephew Inc, Andover, Massachusetts, USA) is one such device. But it is not without limitations, as it requires over-drilling of the femoral socket to allow for it to flip at lateral femoral cortex. As a result of this over-drilling, some part of the socket is not filled by the graft, hence leaving scope for graft motion within the tunnel, which can hamper graft incorporation and cause tunnel widening.⁴ Secondly, an error in measuring the length of the socket to be drilled can lead to either inability to flip the button or an insufficient graft length within the bone.⁵ Thirdly, as recent trend towards an anatomic tunnel placement can lead to short tunnel length, there is a concern for adequate graft length in the bone.⁵

In order to overcome these problems, intention to conserve bone and to achieve a better functional outcome, second generation suspensory fixation devices were introduced, for example, TightRope (TR) RT (Arthrex Inc, Naples, Florida, USA) and ToggleLoc with ZipLoop (Biomet Inc, Warsaw, Indiana, USA). These adjustable loop devices can be tightened intraoperatively, thus obviating the need for over-drilling.^4,6

There are a few in vitro studies comparing biomechanical properties of these two devices.^4,5,7

–10 Although the study designs and testing mechanisms are heterogeneous, they clearly demonstrate that fixed loop devices are biomechanically superior to adjustable loop devices. To the best of our knowledge, there are only three studies in the literature which compare the two devices in vivo.^11
–13 None of them have shown clinical superiority of one device over the other. We conducted this prospective study to find out which device fared better in terms of functional outcomes and laxity measurements at a final follow-up of 2 years.

Materials and methods

Patients undergoing primary arthroscopic ACLR using autologous hamstring graft, performed at a tertiary centre between November 2012 and August 2014, were included in the study. In EB group, the femoral fixation of quadrupled graft was done using EB CL Ultra (Smith & Nephew Inc, Andover, Massachusetts, USA) and in TR group using the TR (Arthrex Inc, Naples, Florida, USA). The inclusion criteria were a clinical and radiological evidence of ACL tear, age between 18 and 50 years with normal contralateral knee. Patients with posterior cruciate ligament tear, associated tibial plateau fracture, bilateral knee injury, suspected infection, and patients not consenting for inclusion in the study were excluded. Patients were allotted to either group using an online random number table generator. Institutional ethics committee clearance was obtained in accordance with the 1964 Helsinki declaration and its later amendments and informed consent was taken from all participants.

After arthroscopic confirmation of a complete ACL tear, ipsilateral semitendinosus and gracilis were harvested and prepared. The femoral tunnel was drilled at 110—120° of knee flexion via the accessory anteromedial portal. The tibial tunnel was made using the tibial jig set at 55°. Diameter of the tunnels was decided based on the graft thickness. The graft fixation on the femoral side was done either with EB or TR. While in the EB group the femoral tunnel was over-drilled by 10 mm above the desired intraosseous graft length, the same was not required in case of TR. The fixation on the tibial side was done using interference screw (bioscrew) in either group.

Knee brace was applied post-operatively. Antibiotics were given according to the hospital protocol. Similar post-operative physiotherapy regime was followed for both the group. Clinical outcome was assessed by two independent observers who were blinded to the implant used, by using International Knee Documentation Committee (IKDC) and Lysholm scoring. Knee stability was evaluated by measuring anterior tibial translation (ATT) using KT-1000 arthrometer and by calculating side-to-side difference (SSD) of ATT. Assessment was done at 6 months and 2 years post-operatively. Statistical analysis was done using paired and unpaired Student’s t test for continuous data and Fisher’s exact test for categorical data.

Results

A total of 109 patients were initially included in the study. Five patients were lost to follow-up and two patients refused to give consent, hence for final analysis 102 patients were available (52 in EB group and 50 in TR group). Both the groups were comparable in terms of demographic details and preoperative scores and findings (Table 1).

Table 1.

Demographic parameters and preoperative scores.

Parameter	EB (n = 52)	TR (n = 50)	p Value
Age (years)	Mean 28.7 (SD 9.45) (range 18–50)	Mean 30.4 (SD 7.9) (range 18–50)	0.3311 (unpaired t test) (NS)
Sex (males:females)	41:11	40:10	1 (Fisher’s exact test) (NS)
Meniscal tear	12/52 (23%)	14/50 (28%)	0.8265 (Fisher’s exact test) (NS)
Injury–surgery interval (weeks)	Mean 15.9 (SD 11.02) (range 3–48)	Mean 17.7 (SD 10.11) (range 3–60)	0.3888 (unpaired t test) (NS)
IKDC score	Mean 37 (SD 6.55) (range 20.6–54.8)	Mean 38.5 (SD 4.92) (range 25.8–50.6)	0.1937 (unpaired t test) (NS)
Lysholm score	Mean 53.2 (SD 8.55) (range 35–71)	Mean 52 (SD 7.06) (range 38–67)	0.4603 (unpaired t test) (NS)
ATT (mm)	Mean 10.2 (SD 1.1) (range 8–12)	Mean 10.2 (SD 1.29) (range 8–12)	0.8970 (unpaired t test) (NS)
SSD (mm)	Mean 7.9 (SD 1.04) (range 6–10)	Mean 7.7 (SD 1.17) (range 6–10)	0.2703 (unpaired t test) (NS)

EB: EndoButton; TR: TightRope; IKDC: International Knee Documentation Committee; ATT: anterior tibial translation; SSD: side-to-side difference; NS: not significant.

The mean Lysholm scores and IKDC scores at 6 months and 2 years after surgery were significantly improved as compared with mean preoperative scores in both the groups. At 6 months’ follow-up, both scores were significantly better in the EB group (Table 2). However, this difference was not apparent at 2 years’ follow-up (Table 2). Hence, at final follow-up of 2 years, both groups were comparable with respect to Lysholm and IKDC scores.

Table 2.

Outcome analysis at 6 months and 24 months after surgery.

Parameter	EB (n = 52)	TR (n = 50)	p Value (unpaired t test)
IKDC	0: 37 (SD 6.55)	0: 38.5 (SD 4.92)	0.1937 (NS)
	6 months: 79.5 (SD 6.12)	6 months: 74.9 (SD 2.77)	<0.0001 (SS)
	24 months: 85.2 (SD 3.66)	24 months: 84.3 (SD 1.52)	0.1173 (NS)
Lysholm score	0: 53.2 (SD 8.55)	0: 52 (SD 7.06)	0.4603 (NS)
	6 months: 88.5 (SD 3.48)	6 months: 83.4 (SD 2.52)	<0.0001 (SS)
	24 months: 91.8 (SD 2.45)	24 months: 91.8 (SD 1.94)	0.9553 (NS)
ATT (mm)	0: 10.2 (SD 1.1)	0: 10.2 (SD 1.29)	0.8970 (NS)
	6 months: 2.9 (SD 1.15)	6 months: 2.84 (SD 1)	0.8348 (NS)
	24 months: 2.4 (SD 0.82)	24 months: 2.7 (SD 0.96)	0.1490 (NS)
SSD (mm)	0: 7.9 (SD 1.04)	0: 7.7 (SD 1.17)	0.2703 (NS)
	6 months: 0.6 (SD 1)	6 months: 0.4 (SD 1.26)	0.3358 (NS)
	24 months: 0.12 (SD 0.92)	24 months: 0.16 (SD 1.33)	0.8438 (NS)

EB: EndoButton; TR: TightRope; IKDC: International Knee Documentation Committee; ATT: anterior tibial translation; SSD: side-to-side difference; SS: statistically significant; NS: not significant.

ATT also improved significantly at 6 months and 2 years post-operatively when compared with the preoperative value in both the groups. Also, ATT was similar in both the groups at 6 months and 24 months follow-up. Similar trend was also observed for the SSD (Table 2). Failure of ACLR was observed in two patients in EB group and one patient in TR group. All cases of failure were due to significant re-injury.

Discussion

In suspensory fixation, the recent shift from EB to TR has been due to the ease of use and variability of length with the shorter femoral tunnel, but post-operative laxity (due to loosening over time) has been an issue for some surgeons.^4,5 The rationale behind TR design is that a shorter vacant tunnel length between button of TR and lateral most end of hamstring graft will potentially reduce the ‘bungee cord’ effect, hence decreasing tunnel widening.¹³ However, biomechanical studies have shown that as time progresses, adjustable loop devices lengthen, which may result in loosening of the graft and clinical failure.^4,5,7

–10 It is a well-known observation that biomechanical testing data do not always translate into clinical and functional results, because in vitro conditions can be very different from in vivo conditions. Only a handful of clinical studies comparing fixed loop to adjustable loop devices have been done.^11
–13 None of these studies have shown a significant difference in terms of clinical scores, tunnel widening and KT-1000 readings. Our study also corroborates these findings. Table 3 summarizes the findings of previous clinical and biomechanical studies.

Table 3.

A review of biomechanical and clinical studies comparing EB and TR devices.

Authors	Study design	Devices tested	Result	Conclusion
Petre et al.⁸ (biomechanical study)	Four cortical soft tissue graft fixation devices tested under cyclic and pull to failure loading in porcine models (both as isolated devices and as porcine femur constructs)	Fixed loop – EB CL, XOB. Adjustable loop – TR RT, TL with ZipLoop.	Cyclic displacement after 1000 cycles for isolated device testing was <1 mm for all devices but for porcine constructs it was: EB-1.88 mm, XOB-1.82 mm, TR-2.74 mm, TL-3.34 mm. Ultimate failure load was: EB-1456 N, XOB-1748 N, TR-859 N, TL-1334 N.	EB, XOB and TR RT were biomechanically sound for initial fixation. TL had sufficient ultimate failure strength but showed clinical failure (> 3 mm displacement) during cyclical loads.
Barrow et al.⁵ (biomechanical study)	Three cortical soft tissue fixation devices tested under extended cyclic loading and pull to failure loading. Adjustable loop devices were also tested with tied suture ends.	Fixed loop – EB CL Ultra. Adjustable loop – TR RT, TL with ZipLoop.	Mean lengthening after 4500 cycles was EB-1.34 mm, TR unknotted-42.45 mm, TR knotted-13.36 mm, TL unknotted-5.76 mm, TL knotted-5.63 mm. Ultimate load to failure was EB-1529 N, TR unknotted-809.11 N, TR knotted-1074.50 N, TL unknotted-1652.13 N, TL knotted-1419 N.	Adjustable loop devices showed clinically significant increase in loop length (>3 mm) during cyclic testing because of pulling of free suture ends into the adjustable loop. Knotting free threads of TR RT resulted in a significant decrease in lengthening (from 42.45 mm to 13.36 mm), but it was still >3 mm.
Eguchi et al.⁴ (biomechanical study)	Two cortical soft tissue fixation devices tested under extended cyclic loading and pull to failure loading (both as isolated devices and as porcine femur constructs).	Fixed Loop – EB CL Ultra. Adjustable loop – TR RT.	Isolated device testing showed higher ultimate tensile strength for EB (1430 N) than TR (866 N). Displacement testing of isolated devices showed significant difference between the two (EB-2.03 mm, TR-4.05 mm), whereas testing of construct did not show any significant difference between the two.	EB is mechanically superior to TR. In construct testing, tendon breakage was the common failure mode for EB, whereas loop breakage was the common failure mode for TR. As specimen testing did not show any significant difference, this study stated that with careful preconditioning of the graft implant complex, both devices should work well in vivo.
Johnson et al.⁷ (biomechanical study)	Five cortical soft tissue fixation devices tested under cyclic loading and pull to failure loading. Adjustable loop devices were also tested after retensioning.	Fixed loop – EB CL Ultra, XOB, RL. Adjustable loop – TR RT, TL with ZipLoop.	Peak displacements (mean) were EB-1.05 mm, RL-1.09 mm, XOB-1.65 mm, TR-2.20 mm, TR retensioned-1.81 mm, TL-3.69 mm, TL retensioned-3.22 mm. Ultimate strength was highest for TL retensioned (2231 N) followed by XOB (2218 N) and lowest for TR (784 N).	Both fixed loop devices were superior to adjustable loop devices, with EB showing least displacement. Retensioning did not significantly alter the biomechanical properties of adjustable loop devices.
Mayr et al.⁹ (biomechanical study)	Tibial side fixations with adjustable loop cortical button and interference screw were compared in calf tibiae. Femoral fixation was adjustable loop cortical button in both.	Adjustable loop device – TR RT. Interference screw – BioComposite BS.	Initial (cycles 1–5), secondary (cycles 6– 1000) and total elongation was higher for TR (2.47 m, 3.56 mm and 6.03 mm, respectively) than for BS (1.46 mm, 1.87 mm and 3.33 mm, respectively). Mean pull-out stiffness was TR – 185 N and BS – 309 N.	Adjustable loop button fixation showed higher graft elongation on cyclic loading, higher ultimate graft failure loads, and a lower mean pull-out stiffness compared to interference screw fixation at time zero. This study concluded that interference screw is a biomechanically superior fixation method at tibial side.
Boyle et al.¹¹ (clinical study) (2-year follow-up)	Retrospective cohort study of 188 patients who underwent ACLR with autologous hamstrings (73 with TR RT and 115 with RB on femoral side). The two groups were compared with respect to KT-1000 arthrometer readings and graft failure rates.	Fixed loop – RB. Adjustable loop – TR RT.	Maximum SSD in KT-1000 testing at 2 years was 1.14 mm for TR and 1.07 mm for RB (p = 0.90). Graft failure rate was 10% for TR and 11% for RB (p = 0.71)	There was no significant difference between knee stability and graft failure rates at final follow-up of 2 years. Hence, both devices work equally well in vivo.
Noonan et al.¹⁰ (biomechanical study conducted in three phases)	Phase 1 – Device only testing with 50 N-250 N protocol. Phase 2 – Device only testing under low load conditions (10 N). Phase 3 – Porcine femur construct testing similar to in vivo conditions. In all phases, TR was tested again after retensioning, after knot tying and after both.	Fixed loop – EB CL. Adjustable loop – TR RT.	In phases 1 and 2, the displacement was higher for TR, and this difference was eliminated after retensioning and knot tying (TR-0.96 mm to 0.38 mm in phase 1, 4.22 mm to 0.051 mm in phase 2). In phase 3, baseline elongation was comparable in two (TR-2.7 mm, EB-3.0 mm), but after retensioning and knot tying the displacement of TR reduced to 1.5 mm, which was significantly lower than EB.	TR demonstrated a statistically significant increase in elongation as compared to EB. But, retensioning and knot tying resulted in a marked reduction in elongation in TR constructs. Clinically, this is important as it demonstrates that retensioning and knot tying can effectively reduce the loop slippage and elongation during rehabilitation.
Lanzetti et al.¹² (clinical and radiological study)	Prospective study involving 44 patients (22 in each group). Clinical examination, functional knee scoring, KT-1000 measurements and femoral tunnel diameter measurement on CT scan at a 12 month follow-up.	Fixed loop – EB. Adjustable loop – TR.	KT-1000 readings were similar, SSD being 2.1 mm (TR) and 2.3 mm (EB). All functional scores (IKDC, Lysholm, Tegner) were similar in both groups. Femoral tunnel diameters in four cuts were also similar (TR-10.56 to 11.15 mm; EB-10.43 to 11.77 mm).	Both devices give comparable clinical results and similar tunnel widening, hence both are equally good.
Choi NH et al. ¹³ (clinical and radiological study)	Retrospective cohort study comparing two groups (67 patients with fixed loop device femoral fixation and 50 patients with adjustable loop femoral fixation). Clinical examination, functional knee scoring, KT-1000 measurements and femoral tunnel diameter measurement on radiographs at a 12 month follow-up.	Fixed loop – EB. Adjustable loop – TR.	Mean KT-1000 measurement was 1.5 mm (EB) and 1.2 mm (TR) (p = 0.530). No significant difference was seen on knee scoring (Lysholm, Tegner). Femoral tunnel diameters in AP and lateral views were EB-42.2 mm and 38.1 mm, TR-43 mm and 35.8 mm (p = 0.801 for AP and 0.422 for lateral views)	There was no significant difference between two groups with respect to clinical examination, KT-1000 measurements and femoral and tibial tunnel widening on radiographs.

EB: EndoButton; TR: TightRope; TL: ToggleLoc; BS: biodegradable screw; RB: RetroButton; XOB: XO Button; IKDC: International Knee Documentation Committee; SSD: side-to-side difference.

EB is currently the most accepted device for the fixation of soft tissue grafts in arthroscopic ACLR, despite of having potential disadvantage that it requires 6–10 mm of over-drilling of the femoral tunnel in order to flip the button on the lateral femoral cortex.⁶ Theoretically, this may lead to excessive graft motion in the tunnel, tunnel widening and delay in graft tunnel healing. It has been shown that new bone ingrowth around the tendon graft is inversely related to the magnitude of graft motion within the tunnel.^14,15 As over-drilling is not needed in TR, it has a potential of improving graft healing within the tunnel, possibly resulting in a better clinical outcome.⁶ Another advantage of TR is the intraoperative flexibility it provides to the surgeon, as it can be tightened even after flipping.

An interesting finding in the present study was better functional results at 6 months in EB group as compared to TR group; however, the results at the end of 2 years were similar. Although the explanation for this discrepancy is unclear, it may be due to small sample size or difference in rehabilitation capacity of patients. In contrast to the IKDC and Lysholm scores, the ATT and SSD measurements were similar in both the groups at 6 months and 2 years’ follow-up.

A comprehensive review of previous biomechanical and clinical studies is presented in Table 3. Advantages and disadvantages of both fixation devices are present in Table 4.

Table 4.

A comparison between EB and TR.

EB		TR
Advantages	Disadvantages	Advantages	Disadvantages
Long track record.	Potentially greater graft motion in tunnel due to over-drilling.	Does not require any intraoperative calculations.	Biomechanically proven to be inferior to EB.
Biomechanically proven to be superior.	Needs exact tunnel measurement and precise depth of socket.	Can be used in shorter tunnels also.	Potential for elongation during cyclic loading, leading to loss of graft tension.
Less elongation on cyclic loading.	Difficult to use in shorter tunnels due to short available graft length in bony socket.	Graft can be tensioned even after flipping.
	Cannot be used for fixation on tibial side.	Whole of the socket filled with graft, potentially eliminating graft motion.
		Can be used for tibial side fixation also.

EB: EndoButton; TR: TightRope.

Limitations of our study include a small sample size, and lack of measurement of tunnel widening at final follow-up.

Conclusions

Final follow-up of 2 years revealed similar knee scores and laxity in the two groups. Although many in vitro studies have shown that adjustable loop devices lengthen with cyclic displacement and are biomechanically inferior to fixed loop devices, previous clinical studies as well as our study fail to corroborate this. Hence, the biomechanical data must be interpreted with caution. The logical step forward would be to conduct well-designed Randomized Control Trials comparing the two devices. Until further evidence clearly shows the superiority of one device over the other, both can be expected to yield similar results.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

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

References

Bennett

. Current trends in anterior cruciate ligament reconstruction: part II. Operative procedures and clinical correlations. Am J Sports Med 2000; 28(1): 124–130.

Shelbourne

Nitz

. Accelerated rehabilitation after anterior cruciate ligament reconstruction. Am J Sports Med 1990; 18(3): 292–299.

Bedi

Rodeo

. Biology of anterior cruciate ligament graft healing. In: Noyes

(ed) Noyes knee disorders: surgery, rehabilitation, clinical outcomes. 1st ed. Philadelphia, PA: Saunders, Inc, 2010, pp. 117–129.

Eguchi

Ochi

Adachi

. Mechanical properties of suspensory fixation devices for anterior cruciate ligament reconstruction: comparison of the fixed-length loop device versus the adjustable-length loop device. Knee 2014; 21(3): 743–748.

Barrow

Pilia

Guda

. Femoral suspension devices for anterior cruciate ligament reconstruction: Do adjustable loops lengthen? Am J Sports Med 2014; 42(2): 343–349.

Lubowitz

Amhad

Anderson

. All-inside anterior cruciate ligament graft-link technique: second-generation, no-incision anterior cruciate ligament reconstruction. Arthroscopy 2011; 27(5): 717–727.

Johnson

Smith

LaPrade

. A biomechanical comparison of femoral cortical suspension devices for soft tissue anterior cruciate ligament reconstruction under high loads. Am J Sports Med 2015; 43(1): 154–160.

Petre

Smith

Jansson

. Femoral cortical suspension devices for soft tissue anterior cruciate ligament reconstruction a comparative biomechanical study. Am J Sports Med 2013; 41(2): 416–422.

Mayr

Heinrichs

Eichinger

. Biomechanical comparison of 2 anterior cruciate ligament graft preparation techniques for tibial fixation: adjustable-length loop cortical button or interference screw. Am J Sports Med 2015; 43(6): 1380–1385.

10.

Noonan

Dines

Allen

. Biomechanical evaluation of an adjustable loop suspensory anterior cruciate ligament reconstruction fixation device: the value of retensioning and knot tying. Arthroscopy 2016; 32(10): 2050–2059.

11.

Boyle

Vovos

Walker

. Does adjustable-loop femoral cortical suspension loosen after anterior cruciate ligament reconstruction? A retrospective comparative study. Knee 2015; 22(4): 304–308.

12.

Lanzetti

Monaco

De Carli

. Can an adjustable-loop length suspensory fixation device reduce femoral tunnel enlargement in anterior cruciate ligament reconstruction? A prospective computer tomography study. Knee 2016; 23(5): 837–841.

13.

Choi

Yang

Victoroff

. Clinical and radiological outcomes after hamstring anterior cruciate ligament reconstructions: comparison between fixed-loop and adjustable-loop cortical suspension devices. Am J Sports Med 2017; 45(4): 826–831.

14.

Berg

Pollard

Kang

. Interarticular bone tunnel healing. Arthroscopy 2001; 17(2): 189–195.

15.

Rodeo

Arnoczky

Torzilli

. Tendon-healing in a bone tunnel. A biomechanical and histological study in the dog. J Bone Joint Surg Am 1993; 75(12): 1795–1803.