Outcomes after bone grafting in patients with and without ACL revision surgery: a retrospective study

Revision anterior cruciate ligament (ACL) reconstruction is continuously increasing with a current incidence of up to 15% [1‐3]. Younger age, secondary trauma, insufficient graft healing and technical factors such as graft fixation or position were associated with increased failure rate after primary ACL reconstruction [4].

In patients requiring revision ACL reconstruction, either a single-stage or a two-stage procedure is indicated based on tunnel position (especially partially anatomic tunnels) and width as well as other potential risk factors, such as lower limb malalignment and steep posterior tibial slope.

In contrast to a single-stage ACL revision, a two-stage procedure consists of 1) hardware removal and graft remnant debridement and bone tunnel grafting and 2) revision ACL reconstruction.

Although two-stage revision ACL reconstruction could achieve similar laxity measurement compared to primary ACL reconstruction, functional outcome parameters have generally been shown to be inferior [2, 5]. From the patient’s perspective especially a two-stage surgery requires high commitment the expectations should be set properly. Clinically, a subset of patients refuses to have further ACL revision surgery despite an already performed bone grafting.

Current literature is lacking of studies evaluating the clinical outcome in patients with and without revision ACL reconstruction following bone grafting. Therefore, the purpose of this study was to evaluate the clinical outcome in patients with and without revision ACL surgery following bone grafting. It was hypothesized that patient reported outcomes are comparable, while knee laxity measures are inferior in patients without revision ACL surgery.

In this retrospective study, patients who underwent bone tunnel grafting with autologous bone from iliac crest due to ACL re-injury between January 2011 and December 2014 were included. The decision to perform a two-stage revision was based on computer tomography (CT) scan analysis. Partially anatomic tunnel position combined with tunnel widening > 12 mm, significant tunnel widening (> 12 mm), suboptimal bone stock or inability to place new tunnel properly were indications for a two-stage procedure. Additional inclusion criteria were available, including preoperative radiographs (a.p. and lateral), CT scans of the involved knee and age > 18 years. Patients with multiple ligament instability requiring surgical treatment or missing data from primary ACL reconstruction were excluded. Demographic data, date and type of primary ACL surgery were extracted from the medical reports.

Fifty-nine patients with bone tunnel grafting were included. Five patients were excluded from the study, because they declined participating in the study. Of the remaining 54 patients, 9 declined to participate in follow-up examination, but all of them consented to complete the patient reported outcome scores. In 44 patients a two stage revision ACL reconstruction (RACL) was performed after bone grafting, whereas 10 patients did not undergo further operation after bone grafting (OBG). Primary ACL reconstructions were performed by several orthopedic surgeons and also at various hospitals. After ACL graft re-rupture all patients present with subjective instability and epsiodes of “giving way” in our outpatient clinic. ACL graft re-rupture was confirmed by clinical examination and magnetic resonance imaging (MRI).

In the RACL group consisting of 31 females and 13 males, the mean age at the time of bone grafting was 30,5 ± 8.5 years (range 16.7 to 49.4 years). For revision ACL reconstruction the contralateral hamstring tendons were used in 46.7% (21/45), ipsilateral hamstring tendons in 26.7% (12/45), ipsilateral quadriceps tendon in 22.2% (10/45) and ipsilateral BTB graft in 2.2% (1/45).

The group of patients without revision ACL reconstruction (OBG) after bone tunnel grafting included 7 females and 3 males with a mean patient age of 33.2 ± 10.3 years (range 18.2–47.7 years).

The main finding of the present study was that after bone tunnel grafting in case of ACL re-instability not every patient requires revision ACL reconstruction. However, there was also no significant difference in the objective clinical outcome between both groups.

Based on our results almost 20% refused a revision ACL reconstruction after bone tunnel grafting, although the initial plan was to perform a revision ACL reconstruction in all patients. Most of these patients have refused to a revision ACL reconstruction due to reduced symptoms after initial bone grafting. At final follow up, laxity measurements were clearly elevated when RACL reconstruction was not performed. However, no significant difference was observed in any of the other evaluated parameters.

The reasons why some patients refused a second ACL reconstruction might be multifactorial. Noyes et al. [6] divided patients with ACL injuries into three different groups depending on their behavior after the injury and the necessity of treatment. “Coper” could achieve recreational level of activity without reconstruction [7]. “Adapter” modified their level of activity until they could manage with the injury. “Non-coper” suffered from recurrent episodes of “giving-way” phenomena and subjective instability during activities of daily living. Therefore in non-coper ACL reconstruction is the treatment of choice [8]. On the other hand, conservative treatment in potential copers also leads to high rate of return to sport and significant reduced instability [9‐11]. Interestingly, Moksnes et al. [12] also described non-copers, with the potential to be converted to real copers by non-operative treatment [12]. Although 70% of non-copers were converted to real copers, this proportion might be too high [12]. Nevertheless it demonstrated that non-coper could be changed in to true coper by conservative treatment [11]. In addition, Fitzgerald et al. [9] achieved after standard physical therapy with perturbation training a high return to high-level activity and significant reduced episodes of “giving- way”.

After initial bone grafting, physical therapy with muscle strengthening and proprioceptive training was performed for at least 3 months. However, not only the improved muscle strength is the decisive factor, but the changed functional movement patterns after intensive physical therapy [11]. Therefore, one possible factor, why some patients in the current study rejected revision ACL reconstruction might be due to improved movement pattern. The comparable functional outcome score between both groups also confirm this.

Due to the retrospective character of the study a detailed interview, why the patients did not want to undergo the revision reconstruction right after bone grafting was not possible. But Feucht et al. [13] highlighted the inferior expectations concerning return to sport, pain and the overall condition in patients with revision ACL reconstruction or previous operations. Therefore, the reduced functional approach and the improved functional status after the bone grafting could be another explanation for the rejection.

Besides subjective instability, some patients also presented with deficits in motion or locking of the knee. Before the bone tunnel grafting a routine diagnostic arthroscopy and resection and debridement of the previous graft and scar tissue was performed. During this procedure additional pathologies, like meniscal tears or severe cartilage defects, were treated. Therefore, some of the complained symptoms might be reduced only by the first surgery. Nearly all of the patients (8 out of 10) without revision ACL reconstruction had additional treatment during the first stage of operation. Although these patients did not complain about subjective instability or recurrent episodes of “giving way” - phenomena, clinical instability was clearly evident without revision ACLR.

In case of revision surgery, a thorough analysis of already utilized tendons, muscle strength and preferred sports is necessary. A variety of possible grafts for revision ACL reconstruction are described [14, 15]. According to the results of the MARS group [16], autografts are favorable compared to allografts with regards to re-rupture rate, clinical outcome scores and sports function. Contralateral hamstrings were also the most common tendons for revision ACL reconstruction in our study population.

In general, ACL revision surgery is challenging and postoperative reported functional outcomes are reduced compared to primary reconstruction [2, 17]. Nevertheless, a recent meta-analysis demonstrated a significant improvement for Lysholm score by revision ACL reconstruction from average 61.5 to 87.5 points. Results for Lysholm scores in the present study are slightly below these values, but according to Briggs et al. [18] a normal to nearly normal overall knee function is associated with a Lysholm score of 73 ± 15 points. From previous literature, the minimum detectable difference for Lysholm score is reported to be between 8.9 and 10.1 points [18, 19]. However, in the current study, the difference between the two groups was 4.3 point. Therefore, no significant or clinically relevant difference was observed between the two study groups. Also, no significant difference between both groups for Tegner score was evident. After bone grafting patients with and without revision ACL reconstruction reached a mean Tegner activity score of 4. In case of revision a mean Tegner of 5 is reported in recent literature [20]. However, Tegner activity scale includes level of sport activity and occupational function [18]. A closer look what is included in a level 4 of Tegner score is necessary for a further interpretation. Patients with a score of 4 could do moderate heavy work and could at least run two times a week. Interestingly, the Tegner score in persons with a “normal knee function” is located at mean of 5.7 [21].

The present study has several limitations. First of all, it is a retrospective case series with known limitations. Furthermore, the number of included patients, especially in the group of patients without revision ACL surgery, is limited. In fact, the calculated power of the current study based on the Lysholm score was 0.33. As most of the patients regularly undergo revision ACL surgery following bone grafting, it is almost impossible to obtain sufficient statistical power. However, the current study highlights, that not every patient requires or is willing to undergo further ACL reconstruction without compromising patient reported outcomes. Follow- up in present study is short term and maybe difference between both groups only occur in mid and long term follow-up. On the other hand, to the authors’ knowledge, this is one of the first studies to compare clinical outcome in patients with and without revision ACL reconstruction after bone grafting.

Bone tunnel grafting due to revision ACL instability with or without second stage ACL revision surgery showed similar clinical results. Thus, in case of revision ACL instability precise patient selection for a renewed operative treatment is necessary.