Summary

Surgeons should consider harvesting the central portion QT distal to the myotendinous junction to preserve donor site sensori-motor structures and include a bone block distally to preserve vascular structures that may contribute to the healing and maturation of the QT graft.

Abstract

Introduction

Quadriceps tendon autograft is still a less common graft choice for anterior cruciate ligament reconstruction (ACLR). However, high failure rates of hamstring ACLR especially in the pediatric population have prompted clinicians to explore other graft options. Quadriceps tendon (QT) autograft is an alternative to hamstrings. In the use of the QT, there remains controversy regarding the location of harvest and whether a bone block should be included. Therefore, the objective of the present study is to determine the distribution of nerve tracts and the location of the vascularity in the pediatric QT.

Methods

Six pediatric QTs (1 month to 3 years of age) from unpaired, fresh cadaveric knees were used. First, the QT was identified and cleared of the overlying skin and fat tissue. Then, the QT was harvested from the myotendinous junction to the patellar insertion including a proximal portion of the patella. After harvest, the samples were fixed in 10% formalin. Subsequently, the QT samples were transected into nine quadrants. In the sagittal plane, the distinction was made into medial, central and lateral portions. In the transverse plane the division was made into proximal (myotendinous junction QT-rectus muscle), mid (QT) and distal (patella-QT junction). This yielded a matrix of 9 areas for examination. After paraffin embedding, quadrants were sectioned at 6 µm thickness in the medial-lateral direction. Three sets of serial sections of each sample were then stained with Hematoxylin & Eosin, S100 (for nerve tissue and endings) and CD31 (for identification of blood vessels). The incidence of discrete S100 signal and CD31 were counted by three independent observers according to location within the QT, based on medial lateral and proximo-distal thirds. The number of S100 signals were then normalized to unit area.

Results

The histological and S-100 stained sections of the pediatric QT showed significantly lower number of nerve bundles in the mid portion of the QT. There was also a significantly larger number of nerve ending at the myotendinous junction in all segments. Vascular networks (as indicated by CD31 staining) was seen within 150 microns of the patellar insertion (patella-QT junction) in 3 of the 6 knees (2 mo, 3 mo and 26 mo). No such networks were observed in the others 3 samples.

Conclusion

In this study, we found that nerve tracts and endings were less numerous in the mid-portion of the pediatric QT, suggesting that this could be the preferred location for QT harvest. In addition, the density of nerve endings proximally at the myotendinous junction indicates that this area should be spared at the time of harvesting to avoid any potential complications related to postoperative pain or extensor mechanism weakness. Moreover, the fact that some patients have a significant vascular network closely associated with the patella suggests that perhaps a bone block may be beneficial to incorporate those vessel as part of the graft. Surgeons should consider harvesting the central portion QT distal to the myotendinous junction to preserve donor site sensori-motor structures and include a bone block distally to preserve vascular structures that may contribute to the healing and maturation of the QT graft.