Elsevier

Journal of Biomechanics

Volume 36, Issue 11, November 2003, Pages 1641-1647
Journal of Biomechanics

In vitro evaluation of combined graft deformation in anterior cruciate ligament reconstructions

https://doi.org/10.1016/S0021-9290(03)00187-8Get rights and content

Abstract

In this paper, we present a simple method of calculating deformation of an anterior cruciate ligament graft in combined elongation, bending and total twisting. We also report our results on these deformations for three types of ligament reconstruction in cadaver knees: two isometric reconstructions, using either a biological bone-patellar tendon-bone autograft or an artificial Trevira® prosthesis, and modified over-the-top reconstruction with the prosthesis. The data show that the modified OTT technique produced elongation of the graft equivalent to that of the synthetic isometric technique, but significantly less than isometric reconstruction with the biological graft (p<0.05, ANOVA). Moreover, the grafts were subjected to bending and twisting.

Introduction

Anterior cruciate ligament (ACL) reconstructions are performed to reestablish knee stability after ACL rupture. To preserve normal knee kinematics, graft placement must be precise (Cooper et al., 1998). If not, graft failure may occur, leading either to loss of motion or to constrained knee movement and further intra-articular damage (Cooper et al., 1998; Sidles et al., 1988). Therefore, surgeons generally aim at placing the graft in a position where it will undergo minimal elongation at each flexion–extension cycle of the knee (Odensten and Gillquist, 1985; Sapega et al., 1990). Generally, less than 2 mm elongation is referred to in the literature as “isometric placement”. However, no information are given on maximal bending and total twisting values that would be acceptable to avoid early graft failure.

Numerous studies have proposed different “ideal” insertion points to achieve isometry (Cooper et al., 1998; Hefzy and Grood, 1986; Hefzy et al., 1989; Sidles et al., 1988). The literature on the subject has, however, yielded conflicting results (Cooper et al., 1998; Fleming et al., 1994; Sidles et al., 1988). This can be explained by the diversity of laboratory testing conditions and, more specifically, by the different ways used to load the knee (Hollis et al., 1991; Sapega et al., 1990). The concept of isometry has, therefore, lost popularity, and anatomic placement is now increasingly practiced (Fu and Musahl, 2002). To our mind, ideal graft placement nevertheless remains an important issue to ensure minimal graft deformation and avoid its premature rupture.

Graft deformation depends on joint displacements and the material properties of the graft. In vitro studies have shown that during knee flexion–extension, the role of the graft is to guide, together with other passive structures, the knee's movement along a unique three-dimensional (3D) pathway of motion (Feikes et al., 1998; Hagemeister et al., 2002b). Knowing the insertion sites of the ligaments, it is possible to estimate their displacement during the complex 3D movement of bony surfaces in space, by tracking knee kinematics via movement sensors firmly attached to the bones of cadaver knees. Once a ligament has been resected and replaced by a biological or synthetic graft, new knee kinematics are generated, depending on many factors, such as graft type, graft stiffness, positioning and pretensioning (Hagemeister et al (2002a), Hagemeister et al (2002b)). When these new kinematics are recorded, conversely, graft deformation can be estimated by tracking tunnel entrances and exits of the reconstruction procedures. Since knee movement is 3D, the displacement of one bony surface (femur) with respect to another (tibia) will not only create graft elongation, but also twisting of the intra-articular part about its long axis and bending at the femoral and tibial tunnel entrances. Both twisting and bending will depend on tunnel orientation, whereas elongation depends on the placement of insertion points in the intra-articular space of the knee (Hefzy et al., 1989; Hefzy and Grood, 1986; Sati et al., 1997).

The aim of this project was to compare combined graft deformation (elongation, bending and twisting) for different types of ACL reconstructions on cadaver knees. We performed

  • (1)

    bone-patellar tendon-bone (BPTB) isometric reconstruction,

  • (2)

    synthetic isometric reconstruction with a polyethyleneterephtalate (TREVIRA®-Hochfest, Telos, Marburg, Germany) ligament, and

  • (3)

    modified over-the-top (OTT) reconstruction with the same artificial ligament. OTT reconstruction was a modification of the classical technique described by Macintosh (1974) in that the tibial tunnel entrance was placed posterior and medial to it usual insertion point on the tibial plateau (dorso-medial placement). This method had been proposed by Krudwig (1997) to minimize elongation of OTT reconstructions.

No cadaver studies have yet shown that graft elongation with an OTT method is equivalent to that for isometric placement, and there have been no cadaver investigations comparing the two methods using biological and synthetic grafts. This is a first attempt to estimate also graft twisting and bending at tunnel entrances for different ACL reconstructions.

In this paper, we present first a simplified method of calculating these combined graft deformations, and then the results obtained with the three above-mentioned reconstruction methods on cadaver knees.

Section snippets

Specimen preparation

Tests were performed on 10 fresh-frozen cadaver knees aged 64–88 yr (mean=79.8 yr). All knees were free from ligamentous pathologies and devoid of any other gross deformities. They were stored at −20°C in a freezer, and kept moist with saline during the entire experiment.

The femur was fixed on a bench with the tibia hanging down, free to move. In the beginning of the experiment, the knee was dissected, leaving only the ligaments, the capsule and the quadriceps/patellar tendon unit. The middle

Validation of estimation with our method compared to Gely's method

We validated the similarity of our method to Gely's (Gely et al., 1984) numerically in a Matlab environment. We programmed both methods to estimate combined ligament deformations. Initial entrance and exit points of the femoral and tibial tunnels were selected randomly for two situations (initial and final state, assessed), and flexion, torsion angles and elongation were calculated with both methods. To compare flexion angles, we added to Gely's calculations the value of the initial flexion

Discussion

The experimental set-up used here was based on the estimation of combined graft deformation during passive flexion–extension movement of cadaver knees obtained by pulling on the quadriceps tendon. As already noted, depending on loading of the knee, the results on graft elongation can vary (Sapega et al., 1990). The data presented should not be taken as absolute values, but should be considered in comparison. In fact, the same well-controlled experimental conditions were used for three different

Conclusion

To achieve a longer lifetime of artificial ligaments, or to avoid cartilage damage due to improper graft placement, the precise positioning of a graft replacing the ACL and tunnel orientation are of prime importance. They result not only in elongation, but also in graft bending at bony tunnel entrance and graft twisting during 3D knee movement. This work has shown that OTT placement with dorso-medial tibial positioning can reduce graft elongation, so that it is nearly isometric. Depending on

Acknowledgements

We thank Gerald Parent and Benoit Godbout for their technical support, and Ovid Da Silva for his editorial assistance. Grant support from the Natural Sciences and Engineering Research Council of Canada is acknowledged.

References (25)

  • M.S. Hefzy et al.

    Sensitivity of insertion locations on length patterns of anterior cruciate ligament fibres

    Journal of Biomechanical Engineering

    (1986)
  • M.S. Hefzy et al.

    Factors affecting the region of most isometric femoral attachments. Part IIThe anterior cruciate ligament

    The American Journal of Sports Medicine

    (1989)
  • Cited by (0)

    View full text