Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Femoral fixation methods for hamstring graft in anterior cruciate ligament reconstruction: A network meta-analysis of controlled clinical trials

  • Shixin Nie,

    Roles Conceptualization, Writing – original draft, Writing – review & editing

    Affiliations Department of Orthopedics, The First Affiliated Hospital of Chongqing Medical University, Chongqing, China, Orthopedic Laboratory of Chongqing Medical University, Chongqing, China

    ⨯
  • Shuqing Zhou,

    Roles Data curation, Formal analysis, Investigation, Methodology

    Affiliation Department of Orthopedics, The Centre Hospital of Jiangjin, Chongqing, China

    ⨯
  • Wei Huang

    Roles Conceptualization, Writing – review & editing

    huangwei68@263.net

    Affiliations Department of Orthopedics, The First Affiliated Hospital of Chongqing Medical University, Chongqing, China, Orthopedic Laboratory of Chongqing Medical University, Chongqing, China

    ⨯

Abstract

Objective

To compare the clinical effectiveness of cortical button (CB), cross-pin (CP) and compression with interference screws (IS) fixation techniques in anterior cruciate ligament (ACL) reconstruction using hamstring graft.

Methods

Studies were systematically retrieved from PubMed, Embase, Cochrane Library and Web of Science up to May 20, 2021. Primary outcomes were KT-1000 assessment, International Knee Documentation Committee (IKDC) score A or B, Lachman’s test, pivot-shift test, visual analogue scale (VAS) score, Lysholm score, Tegner score, and Cincinnati Knee Score. Secondary outcomes included reconstruction failures and synovitis. League tables, rank probabilities and forest plots were drawn for efficacy comparison.

Results

Twenty-six controlled clinical trials (CCTs) with 1,824 patients undergoing ACL reconstruction with hamstring graft were included. No significant differences were found among CB, CP and IS fixation methods regarding the 10 outcomes. For KT-1000 assessment, IKDC score A or B, Lachman’s test, VAS score and pivot-shift test, CP had the greatest probability of becoming the best method, and IS may be the suboptimal method in 4 out of these 5 outcomes except pivot-shift test.

Conclusions

CP, CB and IS fixations have comparable clinical performance, while CP fixation is most likely to be the optimum fixation technique for hamstring graft in ACL reconstruction. Future larger-sample studies of high quality comparing these techniques in more clinical outcomes are required.

Citation: Nie S, Zhou S, Huang W (2022) Femoral fixation methods for hamstring graft in anterior cruciate ligament reconstruction: A network meta-analysis of controlled clinical trials. PLoS ONE 17(9): e0275097. https://doi.org/10.1371/journal.pone.0275097

Editor: Osama Farouk, Assiut University Faculty of Medicine, EGYPT

Received: November 26, 2021; Accepted: September 11, 2022; Published: September 22, 2022

Copyright: © 2022 Nie et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: All relevant data are within the manuscript and its Supporting Information files.

Funding: The author(s) received no specific funding for this work.

Competing interests: The authors have declared that no competing interests exist.

Introduction

Anterior cruciate ligament (ACL) rupture is a common knee ligament injury, which occurs more in the physically active population than in the general population [1]. This ACL injury can lead to pain, functional limitations, osteoarthritis after knee trauma, and reduced quality of life [2, 3]. In the United States, around 400,000 ACL reconstructions are carried out yearly [4]. Autologous hamstring graft is in widespread use and considered as the gold standard of ACL reconstruction, minimizing donor site morbidity [5–8]. However, hamstring graft movement within the femoral tunnel may impede tendon-to-bone healing, so having stable fixation is of great concern [9, 10].

Currently, femoral fixation methods for ACL reconstruction cover three categories: cortical button (CB) fixation, cross-pin (CP) fixation, compression with interference screws (IS) [11–13]. As for an optimal fixation technique, Ibrahim et al. [14] proposed that CP femoral fixation brought greater knee laxity outcomes than CB fixation, while CB femoral fixation exhibited similar effects to CP fixation concerning clinical outcomes and postoperative knee laxity in autologous hamstring ACL reconstruction according to a meta-analysis of Jiang et al [15]. CP fixation was shown to have a smaller instrumented side-to-side anterior-posterior laxity difference than IS fixation, but these two techniques demonstrated comparable performance for hamstring autograft [12]. In Björkman et al.’s research, femoral fixation with CP and IS provided a similar clinical or radiographic result in ACL reconstruction [16]. Additionally, CB fixation was superior to IS fixation for double-bundle ACL reconstruction [17], whereas equivalent impacts were obtained with IS and CB fixation in regard to knee anteroposterior stability and other aspects for all-inside ACL allograft reconstruction [18]. Unfortunately, no studies have reported direct comparisons among CB, CP and IS fixation measures for ACL reconstruction with hamstring graft, and which technique is the best remains unclear. Although a network meta‑analysis from Yan et al. [19] revealed that IS femoral fixation may be the most preferred approach in ACL reconstruction, including different types of studies may lower the statistical power in this study. Thus, a latest network meta‑analysis is needed to further probe into the optimum fixation method.

This study aimed to explore a superior femoral fixation method by comparing the efficacy of CB, CP and IS techniques via a network meta‑analysis of controlled clinical trials (CCTs) in ACL reconstruction with hamstring graft, which may serve as a reference in choosing a fixation method for better rehabilitation.

Methods

Search strategy and study selection

Studies concerning fixation methods in ACL reconstruction were systematically retrieved from PubMed, Embase, Cochrane Library and Web of Science up to May 20, 2021 by two investigators (SX Nie, SQ Zhou) independently. Search terms consisted of “Anterior Cruciate Ligament” OR “Anterior Cruciate Ligament Reconstruction” OR “Anterior Cruciate Ligament Injuries” OR “Anterior Cruciate Ligaments” OR “Cruciate Ligament, Anterior” OR “Cruciate Ligaments, Anterior” OR “Ligament, Anterior Cruciate” OR “Ligaments, Anterior Cruciate” OR “ACL” AND “Surgical Fixation Devices” OR “Orthopedic Fixation Devices” OR “Device, Fixation” OR “Devices, Fixation” OR “Fixation Device” OR “Fixation Devices” OR “Fasteners” OR “Fastener” OR “Fixator” OR “Fixators” OR “Bone Screws” OR “Screw” OR “Screws” OR “TransFix” OR “Intrafix” OR “Aperfix” OR “Arthrex” OR “Biotransfix” OR “Endobutton” OR “Rigidfix”. Then these studies were imported into Endnote X9 (Clarivate Analytics, Philadelphia, Pennsylvania, USA) for duplicate removal, and preliminary screening based on titles and abstracts was carried out, followed by full-text screening, so as to obtain qualified studies. Discussion was needed when opinions were divided.

Inclusion and exclusion criteria

Inclusion criteria were: (1) studies with patients undergoing ACL reconstruction with hamstring graft; (2) studies with interventions including ≥ 2 femoral fixation techniques; (3) studies exploring at least one of the following outcomes; (4) studies in English; (5) CCTs.

The interventions were divided into three categories: (1) CB (Endobutton, Ligament Anchor, Swing Bridge, Tightrope) fixation; (2) CP (Intrafix, Transfix, Rigid Fix, aperture fixation) fixation; (3) IS (Metal Interference Screw, Bioabsorbable Interference Screw).

Exclusion criteria were: (1) animal experiments; (2) publications that did not meet the research theme; (3) studies where valid data could not be extracted; (4) conference abstracts, case reports, editorial materials, reviews, and meta-analyses.

Outcome measures

Primary outcomes were KT-1000 (MEDmetric Corp, San Diego, CA, USA) assessment, International Knee Documentation Committee (IKDC) score A or B [20], Lachman’s test [21], pivot-shift test [22], visual analogue scale (VAS) score [23], Lysholm score [24], Tegner score [25], and Cincinnati Knee Score [26]. Secondary outcomes included reconstruction failures and synovitis.

Data extraction

Two independent researchers (SX Nie, SQ Zhou) extracted baseline information from the eligible studies. The information included author, year of publication, country, level of evidence, femoral tunnel placing, femoral fixation, graft type, tibial fixation, sample size, age, gender ratio, time from injury to surgery, follow-up time, and outcome measure. A consensus was reached through discussion with a third researcher (W Huang).

Risk of bias assessment

The risk of bias in each included CCT was evaluated applying the Cochrane Collaboration’s tool [27] by two reviewers separately (SX Nie, SQ Zhou). The domains for assessment included random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other bias. The risk of bias was categorized as low, unclear or high. Disagreements were resolved by a third researcher (W Huang).

Quality of evidence assessment

The quality of evidence in pairwise effect estimates and overall ranking of femoral fixation methods was evaluated with the approach proposed by Salanti et al. [28] which was based on methodology developed by the Grading of Recommendations Assessment, Development and Evaluation (GRADE) Working Groups. Five domains were assessed: study limitations, indirectness, inconsistency, imprecision, and publication bias. Then the quality of evidence was divided into four levels: high, moderate, low and very low.

Statistical analysis

R 4.0.3 software (R Foundation for Statistical Computing, Vienna, Austria) was employed for the network meta-analysis, and conventional meta-analysis was conducted with Stata 15.1 software (Stata, College Station, Texas, USA). Odds ratios (ORs) acted as the effect size of categorical outcomes, and standardized mean differences (SMDs) were used as the effect size of continuous outcomes. All estimates of these effect sizes reported were posterior medians with corresponding 95% credibility intervals (CrIs). When 95% CrIs excluded null values, significant effects of the femoral fixation methods on the different outcomes were identified. For each outcome measure, both fixed effects model and random effects model were initially fitted. Four Markov chains were adopted for every model to set initial values. The number of pre-iterations was set to 40,000, and the number of iteration operations was set to 200,000. The final model of each outcome was confirmed for subsequent analysis to attain the relative effects and ranking probabilities of different fixation measures in each outcome. In addition, the network plot, league table, rank probabilities and forest plot of each outcome measure were drawn. Node-split analysis was performed for consistency and inconsistency detection in direct and indirect comparisons when there was a closed loop. The strength of direct and indirect evidence was consistent if the difference between the deviance information criteria (DIC) of the consistency and inconsistency detection results was less than 5. P < 0.05 indicated a statistically significant difference.

Results

Characteristics of the included studies

Based on the search strategy, 4,522 studies were identified from the four databases. After duplicates were removed, 2,421 studies were left. Following that, screening was carried out by reading titles and abstracts, and then full texts. Finally, 26 CCTs [14, 16, 18, 29–51] with 1,824 patients were qualified for next analysis. Detailed search process is illustrated in Fig 1. These included trials were published between 2002 and 2020, with 9 trials of CB vs CP [14, 29, 36, 37, 42, 43, 46, 48, 51], 9 of CB vs IS [18, 30–33, 35, 44, 45, 50], and 8 of CP vs IS [16, 34, 38–41, 47, 49]. Follow-up time ranged from 6 months to 60 months. Table 1 exhibits the baseline data of the included trials. The major risk of bias was selection bias from random sequence generation. The overall risk of bias of these studies was low. Risk of bias assessment for the qualified studies is shown in Fig 2. The quality of evidence in pairwise effect estimates ranged from very low to high, and most of pairwise comparisons had high quality of evidence. The overall ranking of femoral fixation methods for KT-1000 assessment, IKDC score A or B, Lachman’ s test, Pivot-shift test, and VAS score had moderate, high, high, high, and low quality of evidence, respectively (Table 2).

thumbnail
Download:
Fig 1. Flow chart for search process.

https://doi.org/10.1371/journal.pone.0275097.g001

thumbnail
Download:
Fig 2. Risk of bias assessment for the included studies.

https://doi.org/10.1371/journal.pone.0275097.g002

thumbnail
Download:
Table 1. Baseline characteristics of the included studies.

https://doi.org/10.1371/journal.pone.0275097.t001

thumbnail
Download:
Table 2. Summary of our confidence in effect estimates and ranking of femoral fixation methods.

https://doi.org/10.1371/journal.pone.0275097.t002

Network plots of fixation method comparisons

Network plots were depicted to reflect comparisons among CB, CP and IS fixation techniques (Fig 3). Five studies provided data for KT-1000 assessment. Direct comparisons were available between CP and IS, and between CP and CB, but there was no direct comparison between IS and CB. Most studies were performed on CP, followed by CB; direct comparison evidence for CP and CB was most abundant. As to IKDC score A or B, 16 CCTs were included. Direct evidence on the pairwise comparison of CP, CB and IS was displayed, constituting a closed-loop relationship. Most studies were done on both CP and CB, and on the direct comparison of these two methods. A closed loop was also formed for CP, CB and IS in terms of Lachman’s test which was explored in 5 studies. CP fixation was reported in the majority of the 5 studies, and most evidence for direct comparison was offered on CP and IS. Pivot-shift test was conducted in 9 trials. With the direct pairwise comparison of the three techniques, a closed-loop relation came into being. Most of the studies focused on CB, and on the head-to-head comparison of CB and IS. Besides, 3 trials were conducted on VAS score. Direct comparisons existed between CP and IS, and between IS and CB. Most trials investigated IS, followed by CB. Most direct evidence of comparison was available for IS and CB. According to the above, closed loops were formed for IKDC score A or B, Lachman’s test and pivot-shift test, and the results of node-split analysis indicated that the strength of the direct and indirect evidence was consistent (Table 3).

thumbnail
Download:
Fig 3. Network plots of fixation method comparisons for various outcomes.

3A: KT-1000 assessment; 3B: IKDC score A or B; 3C: Lachman’s test; 3D: Pivot-shift test; 3E: VAS score. IKDC: International Knee Documentation Committee; VAS: visual analogue scale; CB: cortical button; CP: cross-pin; IS: interference screw.

https://doi.org/10.1371/journal.pone.0275097.g003

thumbnail
Download:
Table 3. Consistency and inconsistency detection for the outcomes.

https://doi.org/10.1371/journal.pone.0275097.t003

League tables for fixation methods

The efficacies of CP, CB and IS techniques on different outcomes were compared in pairs, as shown in Table 4. As regards KT-1000 assessment, no significant differences were observed between CP and CB (pooled OR = 3.725, 95% CrI = 0.407–55.092), between IS and CB (pooled OR = 2.054, 95% CrI = 0.031–113.409), and between IS and CP (pooled OR = 0.551, 95% CrI = 0.015–10.848). Likewise, the effectivity of these 3 methods were similar in other 4 outcomes: IKDC score A or B (CP vs CB: pooled OR = 1.838, 95% CrI = 0.868–3.743; IS vs CB: pooled OR = 1.554, 95% CrI = 0.673–3.781; IS vs CP: pooled OR = 0.841, 95% CrI = 0.380–2.102), Lachman’s test (CP vs CB: pooled OR = 1.511, 95% CrI = 0.523–4.406; IS vs CB: pooled OR = 1.153, 95% CrI = 0.324–4.092; IS vs CP: pooled OR = 0.758, 95% CrI = 0.284–2.104), pivot-shift test (CP vs CB: pooled OR = 1.254, 95% CrI = 0.577–3.203; IS vs CB: pooled OR = 0.564, 95% CrI = 0.253–1.582; IS vs CP: pooled OR = 0.456, 95% CrI = 0.159–1.357), and VAS score (CP vs CB: pooled SMD = 1.135, 95% CrI = -3.438–6.773; IS vs CB: pooled SMD = 0.862, 95% CrI = -1.829–4.541; IS vs CP: pooled SMD = -0.298, 95% CrI = -4.298–3.715).

thumbnail
Download:
Table 4. League tables of fixation methods for various outcomes.

https://doi.org/10.1371/journal.pone.0275097.t004

Rank probabilities for fixation methods

Rank probabilities were illustrated for CP, CB and IS fixation approaches (Tables 5–9). As for KT-1000 assessment, IKDC score A or B, Lachman’s test, VAS score and pivot-shift test, CP had the greatest probability of becoming the best method, and IS may be the suboptimal method in 4 out of 5 outcomes except pivot-shift test.

thumbnail
Download:
Table 5. Rank probabilities of fixation methods for KT-1000 assessment.

https://doi.org/10.1371/journal.pone.0275097.t005

thumbnail
Download:
Table 6. Rank probabilities of fixation methods for IKDC score A or B.

https://doi.org/10.1371/journal.pone.0275097.t006

thumbnail
Download:
Table 7. Rank probabilities of fixation methods for Lachman’s test.

https://doi.org/10.1371/journal.pone.0275097.t007

thumbnail
Download:
Table 8. Rank probabilities of fixation methods for pivot-shift test.

https://doi.org/10.1371/journal.pone.0275097.t008

thumbnail
Download:
Table 9. Rank probabilities of fixation methods for VAS score.

https://doi.org/10.1371/journal.pone.0275097.t009

Forest plots for fixation methods

According to the results of forest plots, CP and CB (CP vs CB: pooled OR = 3.800, 95% CrI = 0.410–58.000) as well as IS and CP (IS vs CP: pooled OR = 0.520, 95% CrI = 0.014–10.000) exhibited comparable effects concerning KT-1000 assessment. For IKDC score A or B, no statistically significant differences existed between CP and CB (CP vs CB: pooled OR = 2.200, 95% CrI = 0.930–5.900), IS and CB (IS vs CB: pooled OR = 1.000, 95% CrI = 0.280–3.500), and IS and CP (IS vs CP: pooled OR = 1.200, 95% CrI = 0.440–6.200). Consistently, the equivalent effectiveness of CP and CB (CP vs CB: pooled OR = 1.600, 95% CrI = 0.490–5.700), IS and CB (IS vs CB: pooled OR = 0.760, 95% CrI = 0.076–5.900), and IS and CP (IS vs CP: pooled OR = 0.830, 95% CrI = 0.280–2.400) was demonstrated in Lachman’s test. CP and CB (CP vs CB: pooled OR = 1.200, 95% CrI = 0.390–3.300), IS and CB (IS vs CB: pooled OR = 0.670, 95% CrI = 0.250–2.700), and IS and CP (IS vs CP: pooled OR = 0.310, 95% CrI = 0.036–1.700) also had similar impacts on pivot-shift test. In terms of VAS score, IS did not significantly differ from CB (IS vs CB: pooled SMD = 0.860, 95% CrI = -1.800–4.500) and CP (IS vs CP: pooled SMD = -0.300, 95% CrI = -4.300–3.700). Three studies reported on Lysholm score, and combined analysis revealed that CB presented similar efficacy to CP (CB vs CP: pooled OR = 1.220, 95% confidence interval (CI) = 0.460–3.240, P = 0.686). When reconstruction failures were taken into consideration, the comprehensive analysis of 2 studies indicated that there was no statistical difference between CB and CP (CB vs CP: pooled OR = 1.010, 95% CI = 0.390–2.670, P = 0.977).

Fixation methods for single or double bundle ACL reconstruction

For KT-1000 assessment, only one study compared CP and IS fixation methods in single bundle ACL reconstruction, and no significant difference was found between CP and IS (OR = 0.833, 95% CI = 0.211–3.294, P = 0.795); there was no report on double bundle ACL reconstruction. Regarding IKDC score A or B, CP was shown to have a similar effect to IS in single bundle ACL reconstruction according to a single study (OR = 2.000, 95% CI = 0.525–7.621, P = 0.310); network meta-analysis was performed for double bundle ACL reconstruction based on 2 studies, and revealed that IS was significantly more effective than CB (pooled OR = 1.307, 95% CrI = 3.695, 62.929) and CP (pooled OR = 1.180, 95% CrI = 3.254–54.762), and CP was likely to be better than CB according to rank probabilities. Concerning Lachman’s test, 2 studies provided direct evidence for the comparison between CP and IS in single bundle ACL reconstruction, and meta-analysis demonstrated no significant difference between CP and IS (pooled OR = 1.175, 95% CI = 0.463–2.986, P = 0.734). Network meta-analysis with data from 2 trials for double bundle ACL reconstruction illustrated that CP, CB and IS had comparable impacts on Lachman’s test, and rank probabilities indicated that CP had the highest probability of becoming the optimal method (64.77% probability), and CB was most likely to be the suboptimal method (52.89% probability). As to pivot-shift test, direct evidence from 2 studies on the comparison between CP and IS in single bundle ACL reconstruction exhibited equivalent effectiveness of CP and IS (pooled OR = 2.645, 95% CI = 0.637–10.984, P = 0.181); for double bundle ACL reconstruction, network meta-analysis of 3 trials showed no significant difference among CP, CB and IS, while CP was most likely to be the best fixation method (77.41% probability), and CB was most likely to be the second best method (53.75% probability). Besides, no studies about VAS score reported the technique used for ACL reconstruction (single bundle or double bundle).

Fixation methods when placing the femoral tunnel via transtibial drilling

Twelve studies reported the technique used for placing the femoral tunnel, and all of them used transtibial drilling. Among these 12 studies, 3 had KT-1000 assessment, and network meta-analysis exhibited that CP, CB and IS had similar influences on KT-1000 assessment, while CP was most likely to be the optimum fixation method (47.60% probability), and IS was most likely to be the suboptimum method (35.52% probability). Concerning IKDC score A or B, 8 trials provided data for comparisons among CP, CB and IS. It was found that IS was significantly more effective than CB (pooled OR = 1.323, 95% CrI = 1.005–3.732), and IS had the greatest probability of becoming the best method (93.35% probability), and the second best method was most likely CP (85.49% probability). As regards Lachman’s test, 3 studies were included for network meta-analysis. No significant differences were observed among CP, CB and IS; IS was most likely to be the optimal method (53.31% probability), and CP was most likely to be the suboptimal method (51.90% probability). With respect to pivot-shift test, 4 studies were qualified. Consequently, IS was significantly less effective than CB (pooled OR = 0.001, 95% CrI = 4.887×10−18–0.464) and CP (pooled OR = 0.001, 95% CrI = 4.887×10−18–0.441) for pivot-shift test, and CP had the highest likelihood of being the best method (58.87% probability). Of the 12 studies, none assessed VAS score.

Discussion

The current network meta-analysis found with 26 CCTs of 1,824 patients that CP, CB and IS displayed similar effects on different clinical outcomes in ACL reconstruction with hamstring graft, which was consistent with the findings of previous meta-analyses [52–54] and network meta-analyses [19, 55]. Nevertheless, CP may be more effective than CB and IS for hamstring graft fixation in ACL reconstruction according to rank probabilities analysis; based on this, CP may be prioritized in the femoral fixation of hamstring grafts for ACL reconstruction, so that more satisfactory recovery could be expected.

The development of CP for femoral fixation in ACL reconstruction intends to deal with underlying problems linked to IS and CB techniques, with less anteroposterior laxity and sufficient mechanical strength [56, 57]. An instrumented side-to-side anterior-posterior laxity difference was prominently reduced by CP versus IS, as reported by Hu and others [12]. This supports our revelation to a certain extent that CP had a higher probability of being better to IS in fixing hamstring graft for ACL reconstruction, in terms of knee stability, pain, function and physical activities. The possible superiority of CP to IS concerning knee stability may be due to the fact that CP fixation is performed strictly because anchorage looseness is not allowed, while for IS, the looseness relevant to the tunnel wall occurs [49]. A systematic review reported that the failure rates of bioabsorbable IS, metallic IS and CP were 6.1, 3.3 and 1.7%, separately [58], indicating that CP with a higher success rate can be selected before IS. Additionally, compared with CP femoral fixation, IS was associated with a markedly higher risk of ACL revision for patients receiving ACL reconstruction [59]. Furthermore, CP may be more efficacious than CB in femoral fixation for hamstring ACL reconstruction in this paper. A meta-analysis by Lee et al. [60] demonstrated more femoral tunnel widening after applying CB fixation than CP fixation to reconstruct ACL. CB fixation was also in association with more laxity compared with CP [30]. Tunnel widening possibly links to knee laxity and graft failure [61], and consequently requires a staged revision through bone grafting [62], which is not conducive to the recovery of patients suffering from torn ACL. The above findings reinforce the possibility of CP as the optimum fixation. Apart from the afore-mentioned studies [19, 52–55], the studies of Hu et al [12] and Jiang et al [15] also showed that IS and CB femoral fixations had equivalent impacts on clinical performance to CP femoral fixation for ACL reconstruction with hamstring graft. Similar effects of CP, IS and CB are confirmed using different analytical methods, and more investigations on the comparisons of these three techniques are necessary to validate that CP is most likely to be the best fixation technique.

Among the included studies, regarding Tegner score, Kuskucu et al. [43] discovered that among 24 patients receiving CB fixation, 17 improved from level 4 to level 6 or 7, and the other patients remained at level 4 or 5. After CP fixation, 25 of 32 patients improved from level 4 to level 6 or 7, and the rest of patients remained at level 4 or 5. The Tegner score in the CB and CP groups was reported by Zehir et al. [51] to be comparable. Frosch et al. [38] showed that the average Tegner score was 5.83 points (±2.00) for IS fixation and 5.83 points (±1.24) for CP fixation, and no significant difference was found between the two groups. Since the data from the above studies cannot be synthesized, we only described the results of these studies. Besides, merely Buelow et al. [33] studied the Cincinnati Knee Score of patients in the CB and IS groups during a 2-year follow-up period. The CB group increased from preoperative 44 ± 9.8 to 87 ± 8.9, and the IS group elevated from preoperative 46 ± 10.2 to 86 ± 8.5, without a significant difference between the two groups. Of note, Stengel et al. [49] reported 1 out of 28 patients undergoing CP fixation developed synovitis, and 4 out of 26 patients having IS fixation suffered from synovitis, suggesting that patients receiving CP fixation might have a lower rate of adverse events than those with IS fixation in ACL reconstruction. This necessitates more research into adverse event occurrences after the three fixation methods. Given the highest likelihood of CP femoral fixation being the optimal method in hamstring graft for ACL reconstruction as regards clinical efficacy, together with safety, surgeons may give priority to CP fixation when performing ACL reconstruction, combined with their experience and proficiency as well as the cost of surgery, so that patients could get better rehabilitation under their timely and effective decision-making.

With respect to the technique used for ACL reconstruction (single bundle or double bundle) and the technique used for placing the femoral tunnel (transtibial or transportal or outside-in), 8 studies reported the technique used for ACL reconstruction; 12 studies reported the technique used for placing the femoral tunnel, and all of them applied transtibial drilling. Based on the above information, we have assessed the effect of the femoral fixation methods on the outcomes under these reported ACL reconstruction and femoral tunnel placing techniques. For pivot-shift test under double bundle ACL reconstruction and KT-1000 assessment under the transtibial drilling technique, CP, CB and IS exerted similar influences, while CP was most likely to be the optimum fixation method, which was consistent with our main findings, indicating that these techniques might have no effects on pivot-shift test under double bundle ACL reconstruction and KT-1000 assessment under transtibial drilling. However, this is just a conjecture, and we cannot determine whether these techniques have effects on the outcomes, because most studies did not report on these techniques. Relevant studies should provide complete information on ACL reconstruction and femoral tunnel placing techniques, so that the impact of these techniques on the outcomes can be assessed and a better femoral fixation method can be offered to patients undergoing ACL reconstruction with hamstring graft for better recovery.

Through this network meta-analysis of CCTs, CP fixation was recommended as the first choice to fix hamstring grafts in ACL reconstruction. Nonetheless, certain limitations cannot be ignored. First, the included studies did not provide direct evidence for comparisons among CP, CB and IS on some outcome measures, and there was subjectivity in the outcome evaluation. Besides, this analysis could not determine whether the technique used for ACL reconstruction (single bundle or double bundle) and the technique used for placing the femoral tunnel (transtibial or transportal or outside-in) have effects on the outcomes, since most studies did not report on these techniques. Second, heterogeneity probably from different fixation devices, surgical methods and follow-up time was not addressed. Further, studies in other languages were not included. Third, for KT-1000 assessment, the comparison of CB and IS was based on indirect evidence of low confidence, and for VAS score, the comparison of CB and CP was based on indirect evidence of very low confidence, which may affect the reliability of the comparison results that IS may be better than CB for KT-1000 assessment and CP may be better than CB for VAS score. Future high-quality evidence is warranted to verify these results.

Conclusion

CP, CB and IS fixations exhibit similar clinical performance, whereas CP fixation has the greatest probability of being more effective than CB and IS for hamstring graft in ACL reconstruction. This study underscores the need for further larger-sample studies of high quality to compare the impacts of these techniques on more clinical outcomes.

Supporting information

S1 Checklist. PRISMA 2020 checklist.

https://doi.org/10.1371/journal.pone.0275097.s001

(DOCX)

References

  1. 1. Moses B, Orchard J, Orchard J. Systematic review: Annual incidence of ACL injury and surgery in various populations. Res Sports Med. 2012;20(3–4):157–79. pmid:22742074
  2. 2. Whittaker JL, Woodhouse LJ, Nettel-Aguirre A, Emery CA. Outcomes associated with early post-traumatic osteoarthritis and other negative health consequences 3–10 years following knee joint injury in youth sport. Osteoarthritis Cartilage. 2015;23(7):1122–9. pmid:25725392
  3. 3. Mouton C, Moksnes H, Janssen R, Fink C, Zaffagnini S, Monllau JC, et al. Preliminary experience of an international orthopaedic registry: the ESSKA Paediatric Anterior Cruciate Ligament Initiative (PAMI) registry. J Exp Orthop. 2021;8(1):45. pmid:34173077
  4. 4. Murray MM. Optimizing outcomes of ACL surgery-Is autograft reconstruction the only reasonable option? J Orthop Res. 2021. pmid:34191344
  5. 5. Pinczewski L, Roe J, Salmon L. Why autologous hamstring tendon reconstruction should now be considered the gold standard for anterior cruciate ligament reconstruction in athletes. Br J Sports Med. 2009;43(5):325–7. pmid:19224910
  6. 6. Domnick C, Garcia P, Raschke MJ, Glasbrenner J, Lodde G, Fink C, et al. Trends and incidences of ligament-surgeries and osteotomies of the knee: an analysis of German inpatient records 2005–2013. Arch Orthop Trauma Surg. 2017;137(7):989–95. pmid:28466182
  7. 7. Petersen W, Zantop T. Return to play following ACL reconstruction: survey among experienced arthroscopic surgeons (AGA instructors). Arch Orthop Trauma Surg. 2013;133(7):969–77. pmid:23604790
  8. 8. Yasuda K, Tsujino J, Ohkoshi Y, Tanabe Y, Kaneda K. Graft site morbidity with autogenous semitendinosus and gracilis tendons. Am J Sports Med. 1995;23(6):706–14. pmid:8600739
  9. 9. Weimann A, Rodieck M, Zantop T, Hassenpflug J, Petersen W. Primary stability of hamstring graft fixation with biodegradable suspension versus interference screws. Arthroscopy. 2005;21(3):266–74. pmid:15756178
  10. 10. Höher J, Livesay GA, Ma CB, Withrow JD, Fu FH, Woo SL. Hamstring graft motion in the femoral bone tunnel when using titanium button/polyester tape fixation. Knee Surg Sports Traumatol Arthrosc. 1999;7(4):215–9. pmid:10462210
  11. 11. Hagemans FJA, Jonkers FJ, van Dam MJJ, von Gerhardt AL, van der List JP. Clinical and Radiographic Outcomes of Anterior Cruciate Ligament Reconstruction With Hamstring Tendon Graft and Femoral Cortical Button Fixation at Minimum 20-Year Follow-up. Am J Sports Med. 2020;48(12):2962–9. pmid:32941081
  12. 12. Hu B, Shen W, Zhou C, Meng J, Wu H, Yan S. Cross Pin Versus Interference Screw for Femoral Graft Fixation in Hamstring Anterior Cruciate Ligament Reconstruction: A Systematic Review and Meta-analysis of Clinical Outcomes. Arthroscopy. 2018;34(2):615–23. pmid:29066266
  13. 13. Meyer DC, Stalder M, Koch PP, Snedeker JG, Farshad M. Contact pressure on ACL hamstring grafts in the bone tunnel with interference screw fixation—dynamic adaptation under load. Knee. 2012;19(5):676–9. pmid:22197631
  14. 14. Ibrahim SA, Abdul Ghafar S, Marwan Y, Mahgoub AM, Al Misfer A, Farouk H, et al. Intratunnel versus extratunnel autologous hamstring double-bundle graft for anterior cruciate ligament reconstruction: a comparison of 2 femoral fixation procedures. Am J Sports Med. 2015;43(1):161–8. pmid:25349264
  15. 15. Jiang H, Ma G, Li Q, Hu Y, Li J, Tang X. Cortical Button Versus Cross-pin Femoral Fixation for Hamstring Anterior Cruciate Ligament Reconstruction: A Meta-analysis of Randomized Controlled Trials. Am J Sports Med. 2018;46(9):2277–84. pmid:28753392
  16. 16. Björkman P, Sandelin J, Harilainen A. A randomized prospective controlled study with 5-year follow-up of cross-pin femoral fixation versus metal interference screw fixation in anterior cruciate ligament reconstruction. Knee Surg Sports Traumatol Arthrosc. 2015;23(8):2353–9. pmid:24832696
  17. 17. Lehmann AK, Osada N, Zantop T, Raschke MJ, Petersen W. Femoral bridge stability in double-bundle ACL reconstruction: impact of bridge width and different fixation techniques on the structural properties of the graft/femur complex. Arch Orthop Trauma Surg. 2009;129(8):1127–32. pmid:19357859
  18. 18. Lubowitz JH, Schwartzberg R, Smith P. Cortical Suspensory Button Versus Aperture Interference Screw Fixation for Knee Anterior Cruciate Ligament Soft-Tissue Allograft: A Prospective, Randomized Controlled Trial. Arthroscopy. 2015;31(9):1733–9. pmid:25911394
  19. 19. Yan L, Li JJ, Zhu Y, Liu H, Liu R, Zhao B, et al. Interference screws are more likely to perform better than cortical button and cross-pin fixation for hamstring autograft in ACL reconstruction: a Bayesian network meta-analysis. Knee Surg Sports Traumatol Arthrosc. 2021;29(6):1850–61. pmid:32813044
  20. 20. Irrgang JJ, Anderson AF, Boland AL, Harner CD, Kurosaka M, Neyret P, et al. Development and validation of the international knee documentation committee subjective knee form. Am J Sports Med. 2001;29(5):600–13. pmid:11573919
  21. 21. Kim SJ, Kim HK. Reliability of the anterior drawer test, the pivot shift test, and the Lachman test. Clin Orthop Relat Res. 1995;(317):237–42.
  22. 22. Ayeni OR, Chahal M, Tran MN, Sprague S. Pivot shift as an outcome measure for ACL reconstruction: a systematic review. Knee Surg Sports Traumatol Arthrosc. 2012;20(4):767–77. pmid:22218828
  23. 23. Kersten P, White PJ, Tennant A. Is the pain visual analogue scale linear and responsive to change? An exploration using Rasch analysis. PloS One. 2014;9(6):e99485. pmid:24921952
  24. 24. Briggs KK, Lysholm J, Tegner Y, Rodkey WG, Kocher MS, Steadman JR. The reliability, validity, and responsiveness of the Lysholm score and Tegner activity scale for anterior cruciate ligament injuries of the knee: 25 years later. Am J Sports Med. 2009;37(5):890–7. pmid:19261899
  25. 25. Tegner Y, Lysholm J. Rating systems in the evaluation of knee ligament injuries. Clin Orthop Relat Res. 1985;(198):43–9. pmid:4028566
  26. 26. Risberg MA, Holm I, Steen H, Beynnon BD. Sensitivity to changes over time for the IKDC form, the Lysholm score, and the Cincinnati knee score. A prospective study of 120 ACL reconstructed patients with a 2-year follow-up. Knee Surg Sports Traumatol Arthrosc. 1999;7(3):152–9. pmid:10401651
  27. 27. Higgins JP, Altman DG, Gøtzsche PC, Jüni P, Moher D, Oxman AD, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ. 2011;343:d5928. pmid:22008217
  28. 28. Salanti G, Del Giovane C, Chaimani A, Caldwell DM, Higgins JP. Evaluating the quality of evidence from a network meta-analysis. PLoS One. 2014 Jul 3;9(7):e99682. pmid:24992266
  29. 29. Baumfeld JA, Diduch DR, Rubino LJ, Hart JA, Miller MD, Barr MS, et al. Tunnel widening following anterior cruciate ligament reconstruction using hamstring autograft: a comparison between double cross-pin and suspensory graft fixation. Knee Surg Sports Traumatol Arthrosc. 2008;16(12):1108–13. pmid:18791702
  30. 30. Benea H, d’Astorg H, Klouche S, Bauer T, Tomoaia G, Hardy P. Pain evaluation after all-inside anterior cruciate ligament reconstruction and short term functional results of a prospective randomized study. Knee. 2014;21(1):102–6. pmid:24269603
  31. 31. Ma CB, Francis K, Towers J, Irrgang J, Fu FH, Harner CH. Hamstring anterior cruciate ligament reconstruction: a comparison of bioabsorbable interference screw and endobutton-post fixation. Arthroscopy. 2004;20(2):122–8. pmid:14760343
  32. 32. Boutsiadis A, Panisset JC, Devitt BM, Mauris F, Barthelemy R, Barth J. Anterior Laxity at 2 Years After Anterior Cruciate Ligament Reconstruction Is Comparable When Using Adjustable-Loop Suspensory Fixation and Interference Screw Fixation. Am J Sports Med. 2018;46(10):2366–75. pmid:30015501
  33. 33. Buelow JU, Siebold R, Ellermann A. A prospective evaluation of tunnel enlargement in anterior cruciate ligament reconstruction with hamstrings: extracortical versus anatomical fixation. Knee Surg Sports Traumatol Arthrosc. 2002;10(2):80–5. pmid:11914764
  34. 34. Capuano L, Hardy P, Longo UG, Denaro V, Maffulli N. No difference in clinical results between femoral transfixation and bio-interference screw fixation in hamstring tendon ACL reconstruction. A preliminary study. Knee. 2008;15(3):174–9.
  35. 35. Chiang ER, Chen KH, Chih-Chang Lin A, Wang ST, Wu HT, Ma HL, et al. Comparison of Tunnel Enlargement and Clinical Outcome Between Bioabsorbable Interference Screws and Cortical Button-Post Fixation in Arthroscopic Double-Bundle Anterior Cruciate Ligament Reconstruction: A Prospective, Randomized Study With a Minimum Follow-Up of 2 Years. Arthroscopy. 2019;35(2):544–51. pmid:30712629
  36. 36. Eajazi A, Madadi F, Madadi F, Boreiri M. Comparison of different methods of femoral fixation anterior cruciate ligament reconstruction. Acta medica Iranica. 2013;51(7):444–8. pmid:23945887
  37. 37. Fauno P, Kaalund S. Tunnel widening after hamstring anterior cruciate ligament reconstruction is influenced by the type of graft fixation used: a prospective randomized study. Arthroscopy. 2005;21(11):1337–41. pmid:16325084
  38. 38. Frosch S, Rittstieg A, Balcarek P, Walde TA, Schüttrumpf JP, Wachowski MM, et al. Bioabsorbable interference screw versus bioabsorbable cross pins: influence of femoral graft fixation on the clinical outcome after ACL reconstruction. Knee Surg Sports Traumatol Arthrosc. 2012;20(11):2251–6. pmid:22290125
  39. 39. Gifstad T, Drogset JO, Grøntvedt T, Hortemo GS. Femoral fixation of hamstring tendon grafts in ACL reconstructions: the 2-year follow-up results of a prospective randomized controlled study. Knee Surg Sports Traumatol Arthrosc. 2014;22(9):2153–62. pmid:24005332
  40. 40. Harilainen A, Sandelin J, Jansson KA. Cross-pin femoral fixation versus metal interference screw fixation in anterior cruciate ligament reconstruction with hamstring tendons: results of a controlled prospective randomized study with 2-year follow-up. Arthroscopy. 2005;21(1):25–33. pmid:15650663
  41. 41. Harilainen A, Sandelin J. A prospective comparison of 3 hamstring ACL fixation devices—Rigidfix, BioScrew, and Intrafix—randomized into 4 groups with 2 years of follow-up. Am J Sports Med. 2009;37(4):699–706. pmid:19188561
  42. 42. Ibrahim SA, Hamido F, Al Misfer AK, Mahgoob A, Ghafar SA, Alhran H. Anterior cruciate ligament reconstruction using autologous hamstring double bundle graft compared with single bundle procedures. J Bone Joint Surg Br. 2009;91(10):1310–5. pmid:19794165
  43. 43. Kuskucu SM. Comparison of short-term results of bone tunnel enlargement between EndoButton CL and cross-pin fixation systems after chronic anterior cruciate ligament reconstruction with autologous quadrupled hamstring tendons. J Int Med Res. 2008;36(1):23–30. pmid:18230264
  44. 44. Mayr R, Smekal V, Koidl C, Coppola C, Eichinger M, Rudisch A, et al. ACL reconstruction with adjustable-length loop cortical button fixation results in less tibial tunnel widening compared with interference screw fixation. Knee Surg Sports Traumatol Arthrosc. 2020;28(4):1036–44. pmid:31372680
  45. 45. Ping LW, Bin S, Rui Y, Yang S, Zheng ZZ, Yue D. Arthroscopic ACL reconstruction with reverse "Y"-plasty grafts and fixation in the femur with either a bioabsorbable interference screw or an Endobutton. Knee. 2012;19(2):78–83. pmid:21159514
  46. 46. Price R, Stoney J, Brown G. Prospective randomized comparison of endobutton versus cross-pin femoral fixation in hamstring anterior cruciate ligament reconstruction with 2-year follow-up. ANZ J Surg. 2010;80(3):162–5. pmid:20575918
  47. 47. Rose T, Hepp P, Venus J, Stockmar C, Josten C, Lill H. Prospective randomized clinical comparison of femoral transfixation versus bioscrew fixation in hamstring tendon ACL reconstruction—a preliminary report. Knee Surg Sports Traumatol Arthrosc. 2006;14(8):730–8. pmid:16465537
  48. 48. Sabat D, Kundu K, Arora S, Kumar V. Tunnel widening after anterior cruciate ligament reconstruction: a prospective randomized computed tomography—based study comparing 2 different femoral fixation methods for hamstring graft. Arthroscopy. 2011;27(6):776–83. pmid:21624672
  49. 49. Stengel D, Casper D, Bauwens K, Ekkernkamp A, Wich M. Bioresorbable pins and interference screws for fixation of hamstring tendon grafts in anterior cruciate ligament reconstruction surgery: a randomized controlled trial. Am J Sports Med. 2009;37(9):1692–8. pmid:19483077
  50. 50. Yari SS, El Naga AN, Patel A, Qadeer AA, Shah A. TightRope Versus Biocomposite Interference Screw for Fixation in Allograft ACL Reconstruction: Prospective Evaluation of Osseous Integration and Patient Outcomes. JB JS Open Access. 2020;5(2):e0057. pmid:33123662
  51. 51. Zehir S, Zehir R. Suspensory fixation versus novel transverse crosspin for femoral fixation in anterior cruciate ligament reconstruction. Arch Orthop Trauma Surg. 2014;134(11):1579–85. pmid:25047162
  52. 52. Colvin A, Sharma C, Parides M, Glashow J. What is the best femoral fixation of hamstring autografts in anterior cruciate ligament reconstruction?: a meta-analysis. Clin Orthop Relat Res. 2011;469(4):1075–81. pmid:21063817
  53. 53. Browning WM 3rd, Kluczynski MA, Curatolo C, Marzo JM. Suspensory Versus Aperture Fixation of a Quadrupled Hamstring Tendon Autograft in Anterior Cruciate Ligament Reconstruction: A Meta-analysis. Am J Sports Med. 2017;45(10):2418–27. pmid:28068159
  54. 54. Ilahi OA, Nolla JM, Ho DM. Intra-tunnel fixation versus extra-tunnel fixation of hamstring anterior cruciate ligament reconstruction: a meta-analysis. J Knee Surg. 2009;22(2):120–9. pmid:19476176
  55. 55. Hurley ET, Gianakos AL, Anil U, Strauss EJ, Gonzalez-Lomas G. No difference in outcomes between femoral fixation methods with hamstring autograft in anterior cruciate ligament reconstruction—A network meta-analysis. Knee. 2019;26(2):292–301. pmid:30773253
  56. 56. Marx RG, Spock CR. Complications following hamstring anterior cruciate ligament reconstruction with femoral cross-pin fixation. Arthroscopy. 2005;21(6):762. pmid:15944640
  57. 57. Qi W, Liu Y, Xue J, Li H, Wang J, Qu F. Applying Cross-Pin System in Both Femoral and Tibial Fixation in Anterior Cruciate Ligament Reconstruction Using Hamstring Tendons. Arthrosc Tech. 2015;4(5):e397–402. pmid:26697293
  58. 58. Speziali A, Delcogliano M, Tei M, Placella G, Bartoli M, Menghi A, et al. Fixation techniques for the anterior cruciate ligament reconstruction: early follow-up. A systematic review of level I and II therapeutic studies. Musculoskelet Surg. 2014;98(3):179–87. pmid:25269758
  59. 59. Snaebjörnsson T, Hamrin Senorski E, Svantesson E, Westin O, Persson A, Karlsson J, et al. Graft Fixation and Timing of Surgery Are Predictors of Early Anterior Cruciate Ligament Revision: A Cohort Study from the Swedish and Norwegian Knee Ligament Registries Based on 18,425 Patients. JB JS Open Access. 2019;4(4):e0037. pmid:32043061
  60. 60. Lee DH, Son DW, Seo YR, Lee IG. Comparison of femoral tunnel widening after anterior cruciate ligament reconstruction using cortical button fixation versus transfemoral cross-pin fixation: a systematic review and meta-analysis. Knee Surg Relat Res. 2020;32(1):11. pmid:32660647
  61. 61. Srinivas DK, Kanthila M, Saya RP, Vidyasagar J. Femoral and Tibial Tunnel Widening following Anterior Cruciate Ligament Reconstruction using Various Modalities of Fixation: A Prospective Observational Study. J Clin Diagn Res. 2016;10(11):Rc09–rc11. pmid:28050456
  62. 62. Kraeutler MJ, Welton KL, McCarty EC, Bravman JT. Revision Anterior Cruciate Ligament Reconstruction. J Bone Joint Surg Am. 2017;99(19):1689–96. pmid:28976434