Comparison of clinical results of anteromedial and transtibial femoral tunnel drilling in ACL reconstruction

Three hundred patients with anterior cruciate ligament injuries were treated with an arthroscopic reconstruction of the ACL. The patients for this study were chosen retrospectively: 300 consecutive ACL reconstruction patients from our database at the Orton Orthopaedic Hospital, Helsinki, Finland. Sixty of the patients had been included in a previous RCT for comparing fixation methods [8]. The reconstructions took place from January 2006 to August 2011. During that time period, the practice of the clinic changed; therefore, at first the TT drilling technique was used, and afterwards the AM drilling technique was employed. According to the drilling technique, the patients were divided into two groups: 150 ACL reconstructions using the AM drilling technique and 150 reconstructions using the TT drilling technique. Side-to-side laxity measurements were excluded for the patients who had a bilateral ACL tear, whether they were operated on or not. Revision ACL reconstruction was performed on 18 patients in the AM group and on 17 in the TT group. These patients were excluded from the final evaluation, leaving 132 patients in the AM group and 133 patients in the TT group. Ninety patients (68.2%) in the AM group and 86 (64.7%) patients in the TT group attended the 1-year follow-up. At the 2-year follow-up, there were 60 (45.5%) and 58 (43.6%) patients, respectively.

In the AM group, the reconstructions were done using either the TLS™ technique and a semitendinosus graft (n = 87) or Retrobutton™ in the femoral tunnel and BioScrew™ in the tibial tunnel with a semitendinosus-gracilis graft (n = 63). In the TT group, the fixation methods used were Rigidfix™ in the femoral tunnel and Intrafix™ in the tibial tunnel. The graft used was a semitendinosus-gracilis graft. The results did not reveal any statistically significant differences between the two different fixation methods used in the AM group. All the reconstructions were performed by two experienced knee surgeons (AH, JS).

The TLS™ reconstruction technique has been described by Collette and Cassard [9]. Normally, it is sufficient to use only the semitendinosus tendon as a graft. Bone sockets 10–15 mm deep are made with a hand-powered retrodrill taken through 4.5 mm transtibial and from outside to inside constructed transfemoral drill tunnels. The bone sockets are created doing 360 degree turns while pulling the retrodrill outwards. The graft is pulled inside the knee, the tapes slipped through the tunnels, and the graft is pulled to the right tension. A 10 mm titanium screw is inserted to lock the tapes; the length of the screw is 20 mm in the femur and 25 mm in the tibia.

When using Retrobutton™, the tunnel diameter is the same as the graft’s. A Retrobutton™ loop and implant are attached to the graft, and the graft-implant complex is pulled in to the femoral tunnel. The Retrobutton™ plate is flipped outside the femoral cortex, and the graft is tightened, followed with the tibial fixation by a BioScrew™ of 30 mm length. The screw, the diameter of which is equal to the graft, is inserted eccentrically, compressing the graft against the bony tunnel wall.

The graft in the Rigidfix™ fixation is constructed according to the manufacturer’s instructions: with whipstiches of No. 1 Vicryl (Ethicon Inc., Johnson and Johnson, Somerville, New Jersey) to join the doubled limbs of the semitendinosus and gracilis tendons together. Using drill guides, the depth of the transtibially drilled femoral tunnel is 30–40 mm. With Rigidfix™ instrumentation, two transverse tunnels are drilled for the fixation devices. After the graft is passed in to the femoral tunnel, two Rigidfix™ implants are tapped through the drill guide sleeves transfixing the graft. For the Intrafix™ tibial fixation, No. 1 absorbable whipstitch is used on the graft ends, and the graft is spread across four quadrants between the sleeve and the drill tunnel. After cycling the knee 10 to 15 times, the graft is tightened, and the expansion sleeve and the screw are introduced concentrically to compress the four limbs between the bony tunnel and the device. Three different screw sizes are used: for a graft size up to 8 mm, a 6–8 mm screw is used (7–9 mm screw if the bone quality is suboptimal), and in a graft larger than 8 mm, a 8–10 mm screw is used.

Post-operative care was identical for both groups. Immediate mobilization was allowed, and no knee braces were used. Partial weight bearing was allowed immediately and full weight bearing was begun at 2 weeks. Light motion, including on an exercise bicycle, was allowed at 3 weeks. At 6 weeks, active knee extension, deep water running, proprioceptive exercises and weight training was begun with physiotherapists. At 3 months, jumping and jogging were allowed, for which patients were actively trained. From 6 to 12 months after surgery, a gradual return to sports was allowed. After 1 year, there were no restrictions.

To evaluate the results, the subjective International Knee Documentation Committee (IKDC) score (0–100), Lysholm knee score (0–100) and Tegner activity level (0–10) were used. The clinical tests employed were the Lachman and anterior drawer tests to determine anterior laxity and the pivot shift to determine rotatory stability. The Lachman and anterior drawer tests and the pivot shift were graded negative, slightly positive and clearly positive. The anterior-posterior laxity was measured (side-to-side difference with manual maximum force; KT-2000 arthrometer, MEDmetric Corporation, San Diego California) by comparing the injured knee to the control one. These tests and scores were collected before the operation and at the 1- and 2-year follow-ups. Patients also evaluated their activity level before the trauma (Tegner).

The descriptive statistics are presented as means with standard deviation (SD) or counts with percentages. The AM and TT groups were compared preoperatively with a t-test for continuous variables and a Pearson’s chi-squared test for categorical variables. Repeated measures of the changes in outcomes (Tegner, Lysholm, and IKDC scores and side-to side laxity difference evaluations) were compared between the groups with mixed-effects models and an unstructured covariance structure (i.e. the Kenward-Roger method for calculating the degrees of freedom). Fixed effects included the group, the time and group × time interactions. We used age, gender and baseline values as covariates. The repeated measurements were taken at different time points, 1 and 2 years. The mixed models allowed analyses of unbalanced datasets without imputation; therefore, we analysed all available data with the full analysis set. Repeated measures of the pivot shift test were compared between groups with a random-effects logit model, which included age and gender as covariates. A bootstrap method was used when the theoretical distribution of the test statistics was unknown or in the case of a violation of the assumptions (e.g. non-normality). Hommel’s adjustment will be applied to correct levels of significance for multiple testing. Normal distributions were evaluated graphically and with the Shapiro-Wilk W test.

Basic statistical analysis was done using the BMDP Statistical Package (Statistical Solutions Ltd., Cork, Ireland); for more advanced analysis, Stata 16.1 (StataCorp LP; College Station, TX, USA) was used.