Morphological analysis of failed double bundle ACL grafts.

Introduction Recent clinical and biomechanical studies have demonstrated better results with double-bundle anterior cruciate ligament (ACL) reconstruction compared to single-bundle reconstruction. Little is known about the healing process after double bundle ACL reconstruction and nothing in case of failure. The purpose of this study was to analyze ACL double bundle graft remodeling process by evaluation of cell population and quantification of collagen fibril profiles in four traumatic failed grafts. These morphological appearances were then correlated to normal tendon allograft and ACL. The hypothesis of the study was that a different maturation process occurs between the two bundles. Materials and Methods Four ruptured ACL reconstructed remnants were obtained from 4 patients undergoing revision of the ACL surgery. In all cases the graft used for the primary reconstruction was tibialis anterior allograft. In 2 cases, both bundles were torn, with only the AM bundle in the remaining 2 cases. The average time between the primary surgery and the second trauma was 11 (9-14) months. Three intact ligaments (positive control group) was obtained from 3 patients (average age 64.3 years; 62-66) undergoing total knee arthroplasty. Biopsy specimens of 3 tibialis anterior tendon allografts were used as negative control group. Each biopsy was sheared longitudinally in two identical parts. The first sample was stained with Masson Trichrome and the cellular density of each specimen was estimated with light microscopy by counting the number of cells per mm2. The second sample was analyzed using transmission electron microscopy to measure the diameter of the collagen fibrils, the density (number of collagen fibrils per square micrometer), and the percentage area of collagen fibrils (% area occupied by total collagen fibrils relative to the cross-sectional area). Results There was a statistical difference between AM and PL groups and AM and allografts groups for all the variables analyzed (p < 0.05). No difference was observed between AM and ACL groups with exception of collagen fibrils density (p < 0.05). There were differences between PL and allograft groups exclusively for density and area of collagen fibrils and between PL and ACL groups for diameter, density and area of the collagen fibrils (p < 0.05). Discussion The presence of cells, the density and the percentage of area occupied by the fibrils of the two bundles when compared to control groups demonstrate an evolution of the grafts from tendon to ligament. The different maturation stage of the two bundles is suggested by the differences in all the variables considered. The AM portion appears more mature than the PL one. Our results suggest that the grafts are still immature at an average of 11 month postoperatively. This study has two limitations: we used as positive control group ACL from elderly patients and we didn’t compare our results to a single-bundle control group. Age-related changes occur on collagen fibrils, especially diameter size. It’s difficult to compare our results to single-bundle control group due to different variables such as surgical technique, fixation device, rehabilitation protocol and follow-up. Long term prospective randomized studies between single and double bundle reconstruction will likely demonstrate if an anatomic reconstruction is able to restore the normal structure and ultrastructure of the native AM and PL bundles.