review article single-bundle versus double-bundle anterior ... · abstract: the aim of this...

Int J Clin Exp Med 2017;10(1):1-15www.ijcem.com /ISSN:1940-5901/IJCEM0040176

Review Article Single-bundle versus double-bundle anterior cruciate ligament reconstruction: a systematic review and meta-analysis

Lingde Kong, Zhao Liu, Fei Meng, Yong Shen

Department of Orthopedics, The Third Hospital of Hebei Medical University, Shijiazhuang 050051, Hebei, P. R. China

Received September 18, 2016; Accepted November 5, 2016; Epub January 15, 2017; Published January 30, 2017

Abstract: The aim of this analysis was to compare the functional outcomes, knee stability, and complications of patients after the arthroscopic double-bundle anterior cruciate ligament (ACL) reconstruction with those of single-bundle ACL reconstruction. An electronic search of the PubMed, EMBASE and Cochrane Library databases was performed. The analyzed outcomes were Lysholm scores, Tegner scores, subjective and objective International Knee Documentation Committee Scores (IKDC), Pivot-shift test, Lachman test, KT-1000/2000 measurement and complications. A total of 26 randomized controlled trials were available for this meta-analysis. The final results suggested that the double-bundle ACL reconstruction showed significant better outcomes in both functional out-comes and stability of knee joint, as assessed by the Lysholm scores, Tegner scores, objective IKDC, Pivot-shift test, Lachman test, and KT-1000/2000 (P<0.05). However, we found no evidence of a difference in subjective IKDC and complication rate. Based on current evidence, double-bundle ACL reconstruction appears to yield better outcomes when compared with single-bundle ACL reconstruction. However, the results should be interpreted with caution and further large-scale, well-designed RCTs on this topic are still needed.

Keywords: Single-bundle, double-bundle, anterior cruciate ligament, reconstruction, meta-analysis

Introduction

The anterior cruciate ligament (ACL) is an important structure that is located within the knee joint, and plays an important role in stabi-lizing the knee. Anatomical and biomechanical studies have demonstrated that normal ACL can be divided into anteromedial bundle and posterolateral bundle [1, 2], and it acts to pre-vent anterior translation of the tibia relative to the femur as well as to control rotational move-ments within the knee joint. The anteromedial bundle is almost vertically oriented in the coro-nal plane, which means that it is only able to withstand little rotational tibial load. The pos-terolateral bundle, on the other hand, is thought to control tibial rotation more effectively, as its orientation is almost horizontal. It seems that the two bundles work in a synergistic manner during the movement to stabilize the knee under both anteroposterior and rotational tibial torque [3].

ACL disruption, usually as a result of sports activities, is the most common cause of insta-bility of the knee joint [4]. Nowadays, surgical interventions have been developed to restore stability and allow patients to possess a more active life. Arthroscopic single-bundle ACL reconstruction with autologous tendons is a com-monly performed procedure. Normally, single-bundle ACL reconstruction primarily reproduces the anteromedial bundle. Nevertheless, with-out the posterolateral bundle, reconstructed ACL could not withstand rotational load [5]. In order to reproduce a ligament that is more simi-lar to the original ACL, double-bundle ACL reconstruction procedures have been developed.

Although both single- and double-bundle ACL reconstructions have been used, it remains debatable which technique has much more advantages based on current evidence [6, 7]. In recent years, several meta-analyses on this

http://www.ijcem.com

Single-bundle versus double-bundle ACL reconstruction

2 Int J Clin Exp Med 2017;10(1):1-15

topic have been published [8-11], but if the double-bundle is better than single-bundle technique still could not be determined. Be- sides, these studies did not distinguish short- and long-term outcomes clearly. Pooling these results together would absolutely increase the heterogeneity among studies. Furthermore, several randomized controlled trials (RCTs) have been published in recent years [12-15]. These studies might change some of the meta-analysis results because of the increased sam-ple size. In order to identify and summarize the best available evidence, we carried out this up-to-date meta-analysis, pooled relevant data from all available RCTs, and tried to compare the results of the double-bundle ACL recon-struction with those of single-bundle recon-struction. Our null hypothesis is that there is no difference in outcomes between the single-bundle and double-bundle ACL reconstruction.

Materials and methods

Literature search and inclusion criteria

We searched the PubMed, EMBASE and Co- chrane Library databases for studies that had been published before December 2015. The search strategies used the following format of search terms: (anterior cruciate ligament OR ACL) AND (surgery OR surgical OR reconstruc-tion OR reconstructed OR reconstructive). The

excluded if they (1) were abstracts, letters, reviews, or case reports; (2) had repeated data; and (3) did not report outcomes of interest.

Data extraction and outcome measures

We extracted data that included the general characteristics and the evaluated outcomes of each study. General characteristics referred to first author, year of publication, country, num-bers of patients in each groups, mean length of follow-up, and types of implant. The outcomes involved three aspects: functional outcomes, stability of knee and adverse events, which included Lysholm scores, Tegner scores, sub-jective and objective International Knee Do- cumentation Committee Scores (IKDC), Pivot-shift test, Lachman test, KT-1000/2000 mea-surement and complications.

Functional evaluation was performed with Lysholm scores, Tegner scores and IKDC. According to the objective IKDC scale, a nor-mally functioning knee is graded as A, a near-normally functioning knee is graded as B, a fairly functioning knee is graded as C and a poorly functioning knee is graded as D. This scale was treated as a dichotomous outcome: grade A was normal and other grades were abnormal.

The stability of the knee joint was evaluated by Pivot-shift test, Lachman test and KT1000/

Figure 1. Preferred reporting items for systematic reviews and meta-analyses (PRIS-MA) flow diagram outlining literature search results.

search was limited to English-language, human subjects, and RCTs. In addition, refer-ence lists of all the selected articles were hand-searched to identify other potentially eligible trials. This process was performed iteratively until no additional articles could be identified.

The following inclusive selec-tion criteria were applied: (1) study design: RCTs; (2) popu-lation: patients with a symp-tomatic rupture of the ACL; (3) intervention: anatomic dou-ble-bundle ACL reconstruc-tion; (4) comparison: single-bundle ACL reconstruction; (5) outcomes: functional out-comes, stability of knee joint and complications. Trials were


3 Int J Clin Exp Med 2017;10(1):1-15

Table 1. The basic characteristics of the included studies and participants

First author Publication year Country

Mean age

(years)

Total number

Single-bundle group

Double-bundle group

Mean follow-up (months) Implant Outcome

Adachi 2004 Japan 29.4 108 55 53 31.5 Semitendinosus tendons, gracilis tendons KT-2000 measurements

Jarvela 2007 Finland 33 55 25 30 14 (12-20) Semitendinosus tendons, gracilis tendons Objective IKDC scores, Lysholm scores, Pivot-shift test, complications

Muneta 2007 Japan 23.7 68 34 34 25.3 (18-41) Semitendinosus tendons Pivot-shift test, Lachman test, KT-1000 measurements, complications

Jarvela 2008 Finland 33 77 52 25 >24 (24-35) Semitendinosus tendons, gracilis tendons Objective IKDC scores, Lysholm scores, Pivot-shift test, KT-1000 measurements, complications

Siebold 2008 Germany 28.5 70 35 35 19 (13-24) Semitendinosus tendons, gracilis tendons Subjective IKDC scores, objective IKDC scores, Lysholm scores, Pivot-shift test, KT-1000 measurements, complications

Streich 2008 Germany 29.6 49 25 24 24 Semitendinosus tendons Subjective IKDC, Objective IKDC scores, Lysholm scores, Tegner scores, Pivot-shift test, KT-1000 measurements, complications

Zaffagnini 2008 Italy 26.5 72 35 37 36 Semitendinosus tendons, gracilis tendons Objective IKDC scores, Tegner scores, Pivot-shift test, KT-2000 measurements

Ibrahim 2009 Kuwait -- 200 150 50 29 (25-38) Semitendinosus tendons, gracilis tendons Objective IKDC scores, Pivot-shift test, Lachman test, complications

Wang 2009 China 25.5 64 32 32 16.1 (10-24) Semitendinosus tendons, gracilis tendons Subjective IKDC scores, Lysholm scores, Tegner scores, KT-2000 measurements

Aglietti 2010 Italy 28 70 35 35 24 Semitendinosus tendons, gracilis tendons Subjective IKDC scores, Pivot shift test, KT-1000 measurements, complications

Volpi 2010 Italy 28.5 40 20 20 29.5 Semitendinosus tendons, gracilis tendons, bone-patellar tendon-bone autograft

Objective IKDC scores

Araki 2011 Japan 25 20 10 10 12.7 Semitendinosus tendons, gracilis tendons Lysholm scores, Pivot shift test, Lach-man test, KT-1000 measurements

Fujita 2011 Japan 25.2 55 37 18 32.9 Semitendinosus tendons, gracilis tendons Lysholm scores, Pivot-shift test, KT-1000 measurements, complications

Hemmerich 2011 South Africa 30.2 32 17 15 5.2 Semitendinosus tendons, gracilis tendons Complications

Suomalainen 2011 Finland 32 121 60 61 27 Semitendinosus tendons, gracilis tendons Objective IKDC scores, Pivot-shift test, KT-1000 measurements, complications

Ochiai 2012 Japan 27.8 84 44 40 24 Semitendinosus tendons, gracilis tendons Lysholm scores, Pivot-shift test

Gobbi 2012 Italy 30.4 60 30 30 46.2 (36-60) Semitendinosus tendons Subjective IKDC scores, objective IKDC scores, Lysholm scores, Tegner scores, Pivot-shift test

Hussein 2012 Slovenia 32.9 209 78 131 51.1 (39-63) Semitendinosus tendons, gracilis tendons Subjective IKDC scores, objective IKDC scores, Lysholm scores, Pivot-shift test, KT-1000 measurement


4 Int J Clin Exp Med 2017;10(1):1-15

Lee 2012 Korea 30.3 37 18 19 24 Semitendinosus tendons, gracilis tendons Subjective IKDC scores, objective IKDC scores, Lysholm scores, Tegner scores, Pivot-shift test, Lachman test, KT-1000 measurements

Nunez 2012 Spain 30.7 52 23 29 24 Semitendinosus tendons, gracilis tendons Subjective IKDC scores

Suomalainen 2012 Finland 32.3 65 45 20 60 Semitendinosus tendons, gracilis tendons Objective IKDC scores, Lysholm scores, Pivot-shift test, KT-1000 measurement

Ahlden 2013 Sweden -- 98 48 50 26 (22-42) Semitendinosus tendons, gracilis tendons Lysholm scores, Pivot-shift test, Lach-man test, KT-1000 measurements

Xu 2014 China 31.7 66 32 34 16.3 (12-36) Semitendinosus tendons, gracilis tendons Subjective IKDC scores, Lysholm scores, Tegner scores, Pivot shift test, KT-1000 measurements

Zhang 2014 China -- 94 49 45 24 Semitendinosus tendons, gracilis tendons Lysholm scores, Tegner scores, compli-cations

Koga 2015 Japan 24.5 53 25 28 69 (36-140) Semitendinosus tendons KT measurements, Lysholm score, Tegner score, Pivot-shift test, Lachman test, complications

Mayr 2015 Germany 38.5 62 28 34 26 (23.3-32.7) Semitendinosus tendons, gracilis tendons Subjective IKDC scores, objective IKDC scores, complications

IKDC, International Knee Documentation Committee Scores.


5 Int J Clin Exp Med 2017;10(1):1-15

2000 measurement. The Pivot-shift test was categorized as normal (negative), nearly normal (1+), abnormal (2+) and severely abnormal (3+). We recorded the number of individuals with negative pivot-shift test for meta-analysis.

The complications are the composite of menis-cus secondary tear, wound infections, graft fail-ures, re-rupture of ligament, Cyclops lesions as well as pain.

For continuous outcomes with no standard deviations reported, we tried to calculate stan-dard deviations from standard errors and sam-ple numbers; if a group was divided into several subgroups, the members, means and standard deviations were combined according to the methods described in the Cochrane Handbook for Systematic Reviews of Interventions [16]. Data were extracted independently by two authors. Any disagreements were resolved by discussion and consensus.

Assessment of methodological quality

The methodological quality of the studies was evaluated independently by two independent authors, without masking the trial names. The reviewers followed the instructions provided in the Cochrane Handbook for Systematic Re- views of Interventions [16]. The following do- mains were assessed: random sequence gen-eration, allocation concealment, blinding, incomplete data outcomes, revealing of selec-tive outcomes and any remaining biases.

Statistical analysis

Differences were expressed as risk ratios (RRs) with 95% confidence intervals (CIs) for dichoto-mous outcomes and mean differences (MDs)

adopted if statistically significant heterogeneity was present. Studies with an I2 statistic of 0%-25% were considered to have minor hetero-geneity, those with an I2 statistic of 25%-50% had low heterogeneity, those with an I2 statistic of 50%-75% had moderate heterogeneity, and those with an I2 statistic of >75% had high heterogeneity.

As the types of implant were not consistent between studies, we conducted sensitivity analysis to identify potential sources of hetero-geneity. Furthermore, to examine the influence of follow-up on the overall results, we further carried out subgroup analysis. If the mean length of follow-up was less than 24 months, it was considered as short-term; otherwise, it was long-term.

A funnel plot was created to assess for publica-tion bias, and bias can be seen if the plot is widely skewed. All statistical analyses were per-formed using Review Manager version 5.1 (The Cochrane Collaboration, Software Update, Oxford, UK). P value <0.05 was judged as sta-tistically significant, except where otherwise specified.

Results

Study identification and selection

A total of 328 records were identified by the ini-tial database search. One hundred and thirty-seven records were excluded because of dupli-cates and 161 were excluded for various rea-sons (reviews, non-randomized studies, or not relevant to our analysis) on the basis of the titles and abstracts. The remaining 30 were retrieved for full text review, and 4 were exclud-ed because they reported irrelevant outcomes.

Figure 2. Risk of bias graph: a review of the authors’ judgments regarding each risk of bias item, presented as percentages across all included studies.

with 95% CIs for continuous outcomes. Heterogeneity was analyzed with both the chi-square test and the I2 test. A P value of <0.10 for the chi-square test was interpreted as evidence of statistical het-erogeneity, and I2 was used to estimate total variation across the studies. A fixed-effect model was adopted if there was no statistical evi-dence of heterogeneity, and a random-effect model was


6 Int J Clin Exp Med 2017;10(1):1-15

Finally, 26 studies that met our inclusion crite-ria were included in the present meta-analysis [12-15, 17-38]. The selection process for RCTs included in the meta-analysis is shown in Figure 1.

Study characteristics

The main characteristics of the 26 citations included in the meta-analysis are presented in Table 1. These studies were published between 2004 and 2015. The sample sizes of the stud-ies ranged from 20 to 209 patients, and the mean length of follow-up ranged from 5.2 to 69 months. All studies reconstructed ACL with hamstring tendons (semitendinosus and graci-lis), with the exception of one study using both hamstring tendons and bone-patellar tendon-bone autograft [28]. Studies reported by Muneta et al. [36] and Koga et al. [13], as well as Jarvela et al. [35, 37] and Suomalainen et al. [18, 25] were considered to be the materials being reported in different ways or different follow-up time. We extract the most informative data to avoid duplication of information.

Assessment of risk of bias

The risk of bias is demonstrated graphically in Figure 2 and summarized in Figure 3. The ran-domization technique was mentioned in all tri-als. However, only 6 trials described the pro-cess of random sequence generation [15, 19, 23, 29, 32, 34], and 11 trials stated the meth-od of allocation concealment [12, 14, 17-19, 23, 25, 27, 31, 35, 37]. Blinding is rarely used in orthopedic surgery trials but 16 studies were blinded in the assessment of outcome [13, 17-19, 21, 23-25, 29, 31-36, 38]. Five trials were judged to be at high risk of attrition bias [25, 28, 32, 36, 38], but the other 14 were at low risk [12-14, 17, 19, 22-24, 27, 29-31, 33, 34].

Outcomes

Lysholm scores: Information on the Lysholm scores was provided in 13 studies. The test for heterogeneity was not significant, and the stud-ies have low heterogeneity (P for heterogeneity = 0.11; I2 = 35%). Using the fixed-effect model, the aggregated results suggested that the

Figure 3. Risk of bias summary: a review of the au-thors’ judgments regarding each risk of bias item, for each included study.


7 Int J Clin Exp Med 2017;10(1):1-15

Lysholm scores were lower in the single-bundle group in comparison to the double-bundle group at a statistically significant level (MD, -0.84; 95% CI, -1.67 to -0.02; P = 0.05) (Figure 4).

The subgroup analysis showed that there was no significant difference regarding the Lysholm scores between the two groups in a short-term follow-up (MD, 0.71; 95% CI, -2.20 to 3.62; P = 0.63); however, in the long-term follow-up,

Figure 4. Forest plot showing the comparison of the Lysholm scores between the single-bundle group and the double-bundle group. CI = confidence interval.

Table 2. Final results and subgroup analysisGroups No. of studies MD or RR (95% CI) P value Q-test (P) I2 (%)Lysholm scores 13 -0.84 (-1.67, -0.02) 0.05a 0.11 35 Short-term 6 0.71 (-2.20, 3.62) 0.63b 0.03 60 Long-term 9 -0.92 (-1.79, -0.05) 0.04b 0.76 0Tegner scores 6 -0.21 (-0.32, -0.10) <0.01a 0.96 0 Short-term 1 -0.39 (-0.98, 0.20) 0.20a -- -- Long-term 5 -0.25 (-0.35, -0.16) <0.01a 0.58 0Subjective IKDC 10 -1.10 (-2.33, 0.14) 0.08a 0.30 15 Short-term 4 0.61 (-2.13, 3.34) 0.66b 0.47 0 Long-term 8 -1.19 (-3.23, 0.85) 0.25b <0.01 65Objective IKDC 10 0.83 (0.75, 0.91) <0.01b <0.01 68 Short-term 2 0.50 (0.21, 1.21) 0.12b 0.03 79 Long-term 9 0.88 (0.77, 1.01) 0.07b 0.13 35Pivot-shift test 15 0.79 (0.74, 0.83) <0.01b <0.01 84 Short-term 5 0.85 (0.71, 1.03) 0.09b <0.01 75 Long-term 12 0.79 (0.67, 0.93) <0.01b <0.01 86Lachman test 5 0.82 (0.74, 0.92) <0.01a 0.28 22 Short-term 0 -- -- -- -- Long-term 5 0.82 (0.74, 0.92) <0.01a 0.28 22KT-1000/2000 measurement 14 0.29 (0.00, 0.59) 0.05b <0.01 60 Short-term 6 0.25 (-0.24, 0.74) 0.31b 0.01 67 Long-term 9 0.44 (0.07, 0.82) 0.02b 0.04 50Complications 10 1.46 (0.85, 2.51) 0.17a 0.42 3 Short-term 3 2.56 (0.72, 9.11) 0.15a 0.42 0 Long-term 9 1.22 (0.68, 2.19) 0.51a 0.38 7IKDC, International Knee Documentation Committee Scores. MD, mean difference, RR, risk ratio, afixed-effect model, brandom-effect model.


8 Int J Clin Exp Med 2017;10(1):1-15

patients in double-bundle group had a signifi-cantly higher Lysholm scores compared with these in single-bundle group (MD, -0.92; 95% CI, -1.79 to -0.05; P = 0.04) (Table 2).

Tegner scores: Six RCTs reported the Tegner scores in study patients. The test for heteroge-neity was not significant, and the studies have no heterogeneity (P for heterogeneity = 0.96; I2 = 0%). Using the fixed-effect model, the aggre-gated results suggested that the Tegner scores was greater in the double-bundle group in com-parison with the single-bundle group at a statis-tically significant level (MD, -0.21; 95% CI, -0.32 to -0.10; P<0.01) (Figure 5).

After subgroup analysis, we found that there was no significant difference regarding the Tegner scores between the two groups in a short-term follow-up (MD, -0.39; 95% CI, -0.98 to 0.20; P = 0.20); however, in a long-term fol-low-up, patients in double-bundle group had a significantly higher Tegner scores compared with these in single-bundle group (MD, -0.25; 95% CI, -0.35 to -0.16; P<0.01) (Table 2).

Subjective IKDC: Subjective IKDC was reported in 10 studies. The test for heterogeneity was

not significant, and the studies have minor het-erogeneity (P for heterogeneity = 0.08; I2 = 15%). Using the fixed-effect model, the aggre-gated results suggested that there was no sig-nificant difference in subjective IKDC between the single-bundle and double-bundle groups (MD, -1.10; 95% CI, -2.33 to 0.14; P = 0.08) (Figure 6).

The subgroup analysis showed that there was no significant difference between the two groups in neither short-term follow-up (MD, 0.61; 95% CI, -2.13 to 3.34; P = 0.66) nor long-term follow-up (MD, -1.19; 95% CI, -3.23 to 0.85; P = 0.25) (Table 2).

Objective IKDC: Objective IKDC was reported in 10 studies. The test for heterogeneity was sig-nificant, and the studies have moderate hetero-geneity (P for heterogeneity <0.01; I2 = 68%). Using the random-effect model, the aggregated results suggested that there was a higher rate of normally functioning knee in double-bundle group compared with single-bundle group (RR, 0.83; 95% CI, 0.75 to 0.91; P<0.01) (Figure 7).

The subgroup analysis showed that there was no significant difference in rate of normally

Figure 5. Forest plot showing the comparison of the Tegner scores between the single-bundle group and the double-bundle group. CI = confidence interval.

Figure 6. Forest plot showing the comparison of the subjective IKDC between the single-bundle group and the double-bundle group. CI = confidence interval.


9 Int J Clin Exp Med 2017;10(1):1-15

functioning knee between double-bundle group and single-bundle group in neither short-term follow-up (RR, 0.50; 95% CI, 0.21 to 1.21; P = 0.12) nor long-term follow-up (RR, 0.88; 95% CI, 0.77 to 1.01; P = 0.07) (Table 2).

Pivot-shift test: A total of 15 studies provided data on Pivot-shift test. The test for heteroge-neity was significant, and the studies have high heterogeneity (P for heterogeneity <0.01; I2 = 84%). Using the random-effect model, the rate of negative Pivot-shift test was significantly higher in double-bundle group compared with that in single-bundle group (RR, 0.79; 95% CI, 0.74 to 0.83; P<0.01) (Figure 8).

We further performed subgroup analysis, and the result showed that there was no significant

difference regarding the rate of negative Pivot-shift test between the two groups in a short-term follow-up (RR, 0.85; 95% CI, 0.71 to 1.03; P = 0.09); however, in a long-term follow-up, patients in double-bundle group had a signifi-cantly higher rate of negative Pivot-shift test compared with these in single-bundle group (RR, 0.79; 95% CI, 0.67 to 0.93; P<0.01) (Table 2).

Lachman test: There were 5 studies providing data on Lachman test. The test for heterogene-ity was not significant, and the studies have minor heterogeneity (P for heterogeneity = 0.28; I2 = 22%). Using the fixed-effect model, the aggregated results suggested that there was significant higher rate of negative Lachman test in double-bundle group in comparison with

Figure 7. Forest plot showing the comparison of the objective IKDC between the single-bundle group and the double-bundle group. CI = confidence interval.

Figure 8. Forest plot showing the comparison of the Pivot-shift test between the single-bundle group and the double-bundle group. CI = confidence interval.


10 Int J Clin Exp Med 2017;10(1):1-15

single-bundle groups (RR, 0.82; 95% CI, 0.74 to 0.92; P<0.01) (Figure 9).

As all data were in long-term follow-up, sub-group analysis did not provide any new information.

KT-1000/2000 measurement: A total of 14 RCTs reported the KT-1000/2000 measure-ment in study patients. The test for heterogene-ity was significant, and the studies have moder-ate heterogeneity (P for heterogeneity <0.01; I2 = 60%). Using the random-effect model, the aggregated results suggested that the KT- 1000/2000 measurement was significant lower in the double-bundle group in compari-son with the single-bundle group (MD, 0.29; 95% CI, 0.00 to 0.59; P = 0.05) (Figure 10).

After subgroup analysis, we found that there was no significant difference regarding the KT-1000/2000 measurement between the two groups in a short-term follow-up (MD, 0.25; 95% CI, -0.24 to 0.74; P = 0.31); however, in a

long-term follow-up, patients in double-bundle group had a significantly lower KT-1000/2000 measurement compared with these in single-bundle group (MD, 0.44; 95% CI, 0.07 to 0.82; P = 0.02) (Table 2).

Complications: Complications were reported in 10 studies. The test for heterogeneity was not significant, and the studies have minor statisti-cal evidence of heterogeneity (P for heteroge-neity = 0.42; I2 = 3%). Using the fixed-effect model, the aggregated results suggested that there was no significant difference in complica-tion rate between the single-bundle and dou-ble-bundle groups (RR, 1.46; 95% CI, 0.85 to 2.51; P = 0.17) (Figure 11).

The subgroup analysis showed that there was no significant difference between the two groups in neither short-term follow-up (RR, 2.56; 95% CI, 0.72 to 9.11; P = 0.15) nor long-term follow-up (RR, 1.22; 95% CI, 0.68 to 2.19; P = 0.51) (Table 2).

Figure 9. Forest plot showing the comparison of the Lachman test between the single-bundle group and the double-bundle group. CI = confidence interval.

Figure 10. Forest plot showing the comparison of the KT-1000/2000 measurement between the single-bundle group and the double-bundle group. CI = confidence interval.


11 Int J Clin Exp Med 2017;10(1):1-15

Sensitivity analysis: The exclusion of one study [28] using both hamstring tendons and bone-patellar tendon-bone autograft did not affect the statistical significance of any of the meta-analysis outcomes.

Publication bias

Figure 12 were a funnel plot of 13 studies pooled for Lysholm scores, and the plot showed mild asymmetry, but nearly all studies fall with-in the 95% CI axis for a given standard error. Thus, there is minimal evidence of publication bias.

Discussion

As an effective method for integrating valid information, meta-analysis can provide a founda-

ings may be useful for surgeons in clinical practice.

Usually, better functional outcomes are the pri-mary goal of surgeons. Although some studies have supported the evidence that the clinical outcomes of double-bundle ACL reconstruction are superior to those of single-bundle recon-struction [6, 13], some trials have not found significant differences in clinical outcomes between the two techniques [7, 39].

In a previous meta-analysis conducted by Li et al. [9], Lysholm scores and Tegner scores showed no significant difference between the two ACL reconstruction techniques. However, an apparent limitation of this meta-analysis is that it did not take the length of follow-up into

Figure 11. Forest plot showing the comparison of the complications between the single-bundle group and the double-bundle group. CI = confidence interval.

Figure 12. The funnel plot of 13 studies pooled for Lysholm scores, and the plot showed mild asymmetry, but nearly all studies fall within the 95% CI axis for a given standard error.

tion for making clinical deci-sions. In this analysis, we col-lected evidence from out-comes of RCTs that compared single-bundle ACL reconstruc-tion with double-bundle reconstruction. The final results of our review suggests that the double-bundle ACL recon-struction could have signifi-cant better outcomes in both functional outcomes and sta-bility of knee joint, as assessed by the Lysholm scores, Tegner scores, objective IKDC, Pivot-shift test, Lach- man test, and KT-1000/2000 measurement. However, we found no evidence of a differ-ence in subjective IKDC and complication rate. These find-


12 Int J Clin Exp Med 2017;10(1):1-15

consideration, which was important to the assessment of clinical outcomes. In this study, we included some new RCTs and divided these trials into the short-term group (<24 months) and long-term group (≥24 months) according to the length of follow-up. In the short-term sub-group, there was no significant difference in Lysholm scores and Tegner scores between the single-bundle and double-bundle groups. However, in the long-term subgroup, double-bundle technique results in significantly higher scores when compared with the single-bundle technique. From the above results, we can make a consideration that possibly because of the reconstruction of the posterolateral bundle, the double-bundle technique yields a better clinical outcome when compared with the SB technique in a long-term follow-up.

Although the result showed a significant differ-ence in the objective IKDC grading, we could not detect the significant difference in the sub-jective IKDC score between the two techniques. This could be due to the fact that objective and subjective IKDC scores are different measure-ments, and the subjective score may not be sensitive enough to detect the discrepancy resulting from the restoration of anatomical structure and the stability of knee.

Pivot-shift test, Lachman test and KT-1000/ 2000 measurement were used in this study to evaluate the stability for ACL reconstruction. Among them, Pivot-shift test was designed to evaluate knee rotational instability, and side-to-side anteroposterior stabilization was mea-sured with KT-1000/2000 and Lachman test. Previous cadaveric and biomechanical studies had proved that single-bundle anteromedial or posterolateral reconstruction alone was insuf-ficient in controlling the combined rotatory load and valgus torque, and double-bundle ACL reconstruction was better for control of knee rotational instability [40, 41]. However, Desai et al. [8] pooled 15 studies and concluded that anatomic double-bundle ACL reconstruction did not lead to significant improvements in Pivot-shift test, Lachman test, or anterior draw-er test compared with single-bundle ACL recon-struction. In contrast to their findings, our study supported the other previously published stud-ies and demonstrated that double-bundle reconstruction is superior to the single-bundle technique in restoring anterior and rotational stability, as evidenced by the Pivot shift test,

Lachman test, and KT-1000/2000 measure-ments. Based on the current studies, we assumed that double-bundle reconstruction of ACL is a better technique, which had an advan-tage of preventing anterior translation of the tibia relative to the femur as well as controlling rotational movements within the knee joint.

As graft failures were few among either group, we analyzed some adverse events after sur-gery. In the current meta-analysis, we found results pointing to a trend favoring double-bun-dle reconstruction when it comes to adverse events without being able to show statistically significant difference, and our data concur with a prior study [9]. However, more studies are still needed to draw a definite conclusion.

We acknowledge that this study has several limitations. First, the general lack of random sequence generation and allocation conceal-ment methods in some included RCTs de- creased the methodological quality, and there were the potential to overestimate the effect. Second, it is usually impossible to blind sur-geons regarding surgical procedures, so perfor-mance bias is inevitable in the present meta-analysis. Third, the length of follow-up and study quality varied across the studies, which was a definite source of heterogeneity and would impact the final results. Finally, we only included English-language studies. Despite our best efforts to use multiple literature search approaches, we may have omitted some non-English-language studies with results that may have been applicable to our meta-analysis.

Despite these limitations, this study has clinical value. Overall, our meta-analysis demonstrated the superiority of double-bundle over single-bundle ACL reconstruction. Patients with dou-ble-bundle reconstruction of ACL could have better functional recovery and more stabilized knee, without increased complications. There- fore, we still recommend the use of double-bun-dle technique in the process of ACL reconstruc-tion. Anyhow, the results should be interpreted with caution and further large-scale, well-designed RCTs on this topic are still in need.

Disclosure of conflict of interest

None.

Address correspondence to: Yong Shen, Depart- ment of Orthopedics, The Third Hospital of Hebei


13 Int J Clin Exp Med 2017;10(1):1-15

Medical University, 139 Ziqiang Road, Shijiazhuang 050051, Hebei, P. R. China. Tel: 86-0311-88602316; Fax: 86-0311-88602016; E-mail: [email protected]

References

[1] Girgis FG, Marshall JL, Monajem A. The cruci-ate ligaments of the knee joint. Anatomical, functional and experimental analysis. Clin Orthop Relat Res 1975: 216-231.

[2] Amis AA, Dawkins GP. Functional anatomy of the anterior cruciate ligament. Fibre bundle ac-tions related to ligament replacements and in-juries. J Bone Joint Surg Br 1991; 73: 260-267.

[3] Zantop T, Herbort M, Raschke MJ, Fu FH, Petersen W. The role of the anteromedial and posterolateral bundles of the anterior cruciate ligament in anterior tibial translation and inter-nal rotation. Am J Sports Med 2007; 35: 223-227.

[4] Gianotti SM, Marshall SW, Hume PA, Bunt L. Incidence of anterior cruciate ligament injury and other knee ligament injuries: a national population-based study. J Sci Med Sport 2009; 12: 622-627.

[5] Gabriel MT, Wong EK, Woo SL, Yagi M, Debski RE. Distribution of in situ forces in the anterior cruciate ligament in response to rotatory loads. J Orthop Res 2004; 22: 85-89.

[6] Morey VM, Nag HL, Chowdhury B, Sankineani SR, Naranje SM. A prospective comparative study of clinical and functional outcomes be-tween anatomic double bundle and single bundle hamstring grafts for arthroscopic ante-rior cruciate ligament reconstruction. Int J Surg 2015; 21: 162-167.

[7] Park SY, Oh H, Park SW, Lee JH, Lee SH, Yoon KH. Clinical outcomes of remnant-preserving augmentation versus double-bundle recon-struction in the anterior cruciate ligament re-construction. Arthroscopy 2012; 28: 1833-1841.

[8] Desai N, Bjornsson H, Musahl V, Bhandari M, Petzold M, Fu FH, Samuelsson K. Anatomic single-versus double-bundle ACL reconstruc-tion: a meta-analysis. Knee Surg Sports Traumatol Arthrosc 2014; 22: 1009-1023.

[9] Li YL, Ning GZ, Wu Q, Wu QL, Li Y, Hao Y, Feng SQ. Single-bundle or double-bundle for anteri-or cruciate ligament reconstruction: a meta-analysis. Knee 2014; 21: 28-37.

[10] Meredick RB, Vance KJ, Appleby D, Lubowitz JH. Outcome of single-bundle versus double-bundle reconstruction of the anterior cruciate ligament: a meta-analysis. Am J Sports Med 2008; 36: 1414-1421.

[11] Zhu Y, Tang RK, Zhao P, Zhu SS, Li YG, Li JB. Double-bundle reconstruction results in supe-rior clinical outcome than single-bundle recon-struction. Knee Surg Sports Traumatol Arthrosc 2013; 21: 1085-1096.

[12] Mayr HO, Benecke P, Hoell A, Schmitt-Sody M, Bernstein A, Suedkamp NP, Stoehr A. Single-bundle versus double-bundle anterior cruciate ligament reconstruction: a compara-tive 2-year follow-up. Arthroscopy 2016; 32: 34-42.

[13] Koga H, Muneta T, Yagishita K, Watanabe T, Mochizuki T, Horie M, Nakamura T, Otabe K, Sekiya I. Mid- to long-term results of single-bundle versus double-bundle anterior cruciate ligament reconstruction: randomized con-trolled trial. Arthroscopy 2015; 31: 69-76.

[14] Zhang Z, Gu B, Zhu W, Zhu L. Double-bundle versus single-bundle anterior cruciate liga-ment reconstructions: a prospective, random-ized study with 2-year follow-up. Eur J Orthop Surg Traumatol 2014; 24: 559-565.

[15] Xu Y, Ao YF, Wang JQ, Cui GQ. Prospective randomized comparison of anatomic single- and double-bundle anterior cruciate ligament reconstruction. Knee Surg Sports Traumatol Arthrosc 2014; 22: 308-316.

[16] In: Higgins JP, Green S, editors. Cochrane handbook for systematic reviews of interventions. Version 5.1.0 [updated March 2011]. The Cochrane Collaboration; 2011. Available from www.cochrane-handbook.org.

[17] Ahlden M, Sernert N, Karlsson J, Kartus J. A prospective randomized study comparing dou-ble- and single-bundle techniques for anterior cruciate ligament reconstruction. Am J Sports Med 2013; 41: 2484-2491.

[18] Suomalainen P, Jarvela T, Paakkala A, Kannus P, Jarvinen M. Double-bundle versus single-bundle anterior cruciate ligament reconstruc-tion: a prospective randomized study with 5-year results. Am J Sports Med 2012; 40: 1511-1518.

[19] Nunez M, Sastre S, Nunez E, Lozano L, Nicodemo C, Segur JM. Health-related quality of life and direct costs in patients with anterior cruciate ligament injury: single-bundle versus double-bundle reconstruction in a low-demand cohort--a randomized trial with 2 years of fol-low-up. Arthroscopy 2012; 28: 929-935.

[20] Lee S, Kim H, Jang J, Seong SC, Lee MC. Comparison of anterior and rotatory laxity us-ing navigation between single- and double-bundle ACL reconstruction: prospective ran-domized trial. Knee Surg Sports Traumatol Arthrosc 2012; 20: 752-761.

[21] Gobbi A, Mahajan V, Karnatzikos G, Nakamura N. Single-versus double-bundle ACL recon-

mailto:[email protected]

mailto:[email protected]


14 Int J Clin Exp Med 2017;10(1):1-15

struction: is there any difference in stability and function at 3-year followup? Clin Orthop Relat Res 2012; 470: 824-834.

[22] Ochiai S, Hagino T, Senga S, Saito M, Haro H. Prospective evaluation of patients with anteri-or cruciate ligament reconstruction using a patient-based health-related survey: compari-son of single-bundle and anatomical double-bundle techniques. Arch Orthop Trauma Surg 2012; 132: 393-398.

[23] Hussein M, van Eck CF, Cretnik A, Dinevski D, Fu FH. Prospective randomized clinical evalua-tion of conventional single-bundle, anatomic single-bundle, and anatomic double-bundle anterior cruciate ligament reconstruction: 281 cases with 3- to 5-year follow-up. Am J Sports Med 2012; 40: 512-520.

[24] Hemmerich A, van der Merwe W, Batterham M, Vaughan CL. Knee rotational laxity in a ran-domized comparison of single-versus double-bundle anterior cruciate ligament reconstruc-tion. Am J Sports Med 2011; 39: 48-56.

[25] Suomalainen P, Moisala AS, Paakkala A, Kannus P, Jarvela T. Double-bundle versus single-bundle anterior cruciate ligament recon-struction: randomized clinical and magnetic resonance imaging study with 2-year follow-up. Am J Sports Med 2011; 39: 1615-1622.

[26] Fujita N, Kuroda R, Matsumoto T, Yamaguchi M, Yagi M, Matsumoto A, Kubo S, Matsushita T, Hoshino Y, Nishimoto K, Araki D, Kurosaka M. Comparison of the clinical outcome of double-bundle, anteromedial single-bundle, and pos-terolateral single-bundle anterior cruciate liga-ment reconstruction using hamstring tendon graft with minimum 2-year follow-up. Arth- roscopy 2011; 27: 906-913.

[27] Araki D, Kuroda R, Kubo S, Fujita N, Tei K, Nishimoto K, Hoshino Y, Matsushita T, Matsumoto T, Nagamune K, Kurosaka M. A prospective randomised study of anatomical single-bundle versus double-bundle anterior cruciate ligament reconstruction: quantitative evaluation using an electromagnetic measure-ment system. Int Orthop 2011; 35: 439-446.

[28] Volpi P, Cervellin M, Denti M, Bait C, Melegati G, Quaglia A, de Girolamo L. ACL reconstruc-tion in sports active people: transtibial DB technique with ST/G vs. transtibial SB tech-nique with BPTB: preliminary results. Injury 2010; 41: 1168-1171.

[29] Aglietti P, Giron F, Losco M, Cuomo P, Ciardullo A, Mondanelli N. Comparison between single-and double-bundle anterior cruciate ligament reconstruction: a prospective, randomized, single-blinded clinical trial. Am J Sports Med 2010; 38: 25-34.

[30] Wang JQ, Ao YF, Yu CL, Liu P, Xu Y, Chen LX. Clinical evaluation of double-bundle anterior cruciate ligament reconstruction procedure using hamstring tendon grafts: a prospective, randomized and controlled study. Chin Med J (Engl) 2009; 122: 706-711.

[31] Ibrahim SA, Hamido F, Al MA, Mahgoob A, Ghafar SA, Alhran H. Anterior cruciate ligament reconstruction using autologous ham-string double bundle graft compared with sin-gle bundle procedures. J Bone Joint Surg Br 2009; 91: 1310-1315.

[32] Zaffagnini S, Bruni D, Russo A, Takazawa Y, Lo PM, Giordano G, Marcacci M. ST/G ACL recon-struction: double strand plus extra-articular sling vs double bundle, randomized study at 3-year follow-up. Scand J Med Sci Sports 2008; 18: 573-581.

[33] Streich NA, Friedrich K, Gotterbarm T, Schmitt H. Reconstruction of the ACL with a semitendi-nosus tendon graft: a prospective randomized single blinded comparison of double-bundle versus single-bundle technique in male ath-letes. Knee Surg Sports Traumatol Arthrosc 2008; 16: 232-238.

[34] Siebold R, Dehler C, Ellert T. Prospective ran-domized comparison of double-bundle versus single-bundle anterior cruciate ligament recon-struction. Arthroscopy 2008; 24: 137-145.

[35] Jarvela T, Moisala AS, Sihvonen R, Jarvela S, Kannus P, Jarvinen M. Double-bundle anterior cruciate ligament reconstruction using ham-string autografts and bioabsorbable interfer-ence screw fixation: prospective, randomized, clinical study with 2-year results. Am J Sports Med 2008; 36: 290-297.

[36] Muneta T, Koga H, Mochizuki T, Ju YJ, Hara K, Nimura A, Yagishita K, Sekiya I. A prospective randomized study of 4-strand semitendinosus tendon anterior cruciate ligament reconstruc-tion comparing single-bundle and double-bun-dle techniques. Arthroscopy 2007; 23: 618-628.

[37] Jarvela T. Double-bundle versus single-bundle anterior cruciate ligament reconstruction: a prospective, randomize clinical study. Knee Surg Sports Traumatol Arthrosc 2007; 15: 500-507.

[38] Adachi N, Ochi M, Uchio Y, Iwasa J, Kuriwaka M, Ito Y. Reconstruction of the anterior cruciate ligament. Single-versus double-bundle multi-stranded hamstring tendons. J Bone Joint Surg Br 2004; 86: 515-520.

[39] Nunez M, Sastre S, Nunez E, Lozano L, Nicodemo C, Segur JM. Health-related quality of life and direct costs in patients with anterior cruciate ligament injury: single-bundle versus


15 Int J Clin Exp Med 2017;10(1):1-15

double-bundle reconstruction in a low-demand cohort--a randomized trial with 2 years of fol-low-up. Arthroscopy 2012; 28: 929-935.

[40] Yagi M, Wong EK, Kanamori A, Debski RE, Fu FH, Woo SL. Biomechanical analysis of an ana-tomic anterior cruciate ligament reconstruc-tion. Am J Sports Med 2002; 30: 660-666.

[41] Kondo E, Merican AM, Yasuda K, Amis AA. Biomechanical comparison of anatomic dou-ble-bundle, anatomic single-bundle, and non-anatomic single-bundle anterior cruciate liga-ment reconstructions. Am J Sports Med 2011; 39: 279-288.

review article single-bundle versus double-bundle anterior ... · abstract: the aim of this...

Documents