Ten-year outcomes of anterior cruciate ligament reconstruction with hamstring tendon autograft and femoral fixation with a cortico-cancellous screw suspension device

Anterior cruciate ligament (ACL) rupture is one of the most common knee injuries in the younger population. ACL reconstruction (ACLR) represents the treatment of choice especially in active patients that frequently desire to return to their pre-injury level of function and specific sport participation [1‐3]. It has been widely reported that the short-term results of ACLR are good to excellent in most patients, with high rate of return to sports and low percentage of failures [4‐6]. As regard to long-term outcomes, ACLR is thought to decrease the risk of osteoarthritis (OA) due to the re-established rotatory knee stability and subsequent decreased risk of meniscal and chondral damage [7]. However, several authors reported a high incidence of OA ranging between 21 and 50% after ACLR at 10- to 20-year follow-up [8, 9]. Moreover, a recent systematic review and meta-analysis found that transtibial ACLR resulted in a significantly greater incidence of OA compared with the anteromedial femoral tunnel drilling technique (53.7 vs 14.2%, p = 0.001) at 5- to 10-year follow-up [10]. Clinical outcomes of ACLR at minimum 10 years of follow-up have been assessed by several studies; however, these studies used bone–patellar tendon (BPT), allograft or Hamstring tendon (HT) combined to extra-articular reconstruction in all surgeries [11‐19]. Furthermore, available data are limited when assessing outcomes beyond 10 years. The aim of the present study was therefore to assess the clinical outcomes after ACLR with transtibial technique and HT autograft with femoral cortico-cancellous screw suspension device at minimum 10-year follow-up by recording subjective outcomes (patient-reported outcome measures [PROMs]). The secondary aim of this retrospective study was to determine whether ACL-reconstructed knees with HT autograft have a greater incidence of degenerative changes compared with the contralateral non-reconstructed knee. We hypothesized that the outcomes of ACLR were satisfactory at 10-year follow-up and that incidence of OA was similar to that reported in literature.

This study was performed after approval of the institutional review board. The surgery records and medical charts of consecutive patients who underwent a unilateral primary ACL reconstruction between January 2010 and December 2012 at our institution were eligible for inclusion (n = 480). Inclusion criteria: age > 18 years old, ACL reconstruction with HT autograft, transtibial technique with cortico-cancellous screw suspension device as femoral fixation. Exclusion criteria: previous contralateral surgeries or trauma, posterior cruciate ligament injuries, medial and/or lateral collateral ligament injuries in the ipsilateral knee, concomitant osteotomy procedures, previous ipsilateral knee surgery, patellar and/or quadriceps tendon graft, other systems of femoral ACL graft fixation, Kellgren–Lawrence grade ≥ 2 in the ipsilateral or contralateral knee at the time of surgery. Eligible patients (N = 102) were contacted and invited to visit our outpatient clinic again. Nine patients could not be contacted or were lost to follow-up. This resulted in a total of 93 patients who were included for the overall analysis. When other concomitant and/or contralateral knee surgeries occurred, patients were not included for the assessment of PROMs or OA. Reconstruction was performed once the knee had recovered from the acute trauma of the ACL injury and patients had a pain-free knee, no extension deficit and at least 90° of knee flexion.

The most important findings of the present study were that a relatively low graft rupture rate (5.4%) was reported after ACLR with transtibial technique, HT autograft and femoral cortico-cancellous screw suspension device. Although the Tegner activity level score significantly decreased compared to pre-surgery values, clinical and subjective outcome measures demonstrated that most patients were satisfied with the surgical outcome and quality of life. However, 41 (50%) patients had radiographic signs of OA in the ACL-reconstructed knee.

Analyzing the current literature, Leiter et al. [20] reported a graft rupture rate of 9% with HT graft reconstruction at 14.6 years of follow-up. Asik et al. [12] found a graft failure rate of HT of 1.5% in a cohort of 271 patients. However, the mean follow-up period was 82 (minimum 48 and maximum 108) months. A recent systematic review and meta-analysis that compares the clinical results between HT and patellar tendon reported a failure rate of HT graft from 6.3% to 18% at minimum 10 years of follow-up [21]. Our results showed a lower rate of graft failure (5.4%) compared to existing literature [22]. This difference could be explained by the fact that the mean age of their cohort at the time of surgery was 25 years, whereas the median age in our cohort was 28 (23–38) years at the time of surgery and several studies have shown that the failure rate of ACLR is higher in younger patients [23, 24]. Moreover, the decreased Tegner activity score at time of follow-up compared to pre-surgery levels could minimize the risk of graft failure after ACLR.

Our results were in line with the current literature concerning the incidence and severity of OA after ACLR. A recent systematic review and meta-analysis reported a high incidence of OA in ACL-reconstructed knee with HT graft after 20 years of follow-up [25]. Specifically, a grade of knee OA greater than or equal to 2 according to the Kellgren–Lawrence classification was found between 40% and 66% of cases. However, it is crucial to consider that the studies included in that review are dated to the 1990s and the current standard of practice could be more accurate in terms of tunnel preparation and femoral tunnel placement. However, these data were similar to those reported by Sollberger et al. [21]. In their systematic review, they reported a rate of knee OA ≥ 2 (KL classification) in 32–55.2% of cases after 10 years of follow-up. Although there is a large variation in the incidence of OA in ACL-reconstructed knees at long-term follow-up, meniscectomy has been shown to be a very strong predictor of OA [26]. However, these data were not confirmed by our study. A possible explanation could be that meniscal lesions have been found in a relatively low percentage of patients of our cohort; therefore, the sample size could not be enough to detect significance from multivariate analysis.

On the other hand, from our results, it has been reported that age at time of surgery represent a significant risk of developing knee OA after minimum 10 years of follow-up (odds ratio 1.14–95% CI 1.15–1.74, p = 0.001). Our results were in line with several studies. Hence, Curado et al reported that meniscectomy and age at time of surgery represent main risks for the development of subsequent knee OA [27].

The present study has several limitations that warrant disclosures. First of all, this is a retrospective study, and no control group was assessed. Then, even though consecutive patients have been recruited a selection bias may be present. Moreover, the limited sample size could make the study underpowered for multivariate analysis. Nine patients (8.8%) could not be contacted and were lost at follow-up.

The findings in the present study show that ACLR with HT graft and cortico-cancellous screw suspension device determined satisfied clinical results after 10 years of follow-up. From our cohort, a low rate of graft failure has been reported, even though almost 50% of patients present a knee OA greater or equal to grade II KL.