Methods and devices for graft fixation in anterior cruciate ligament reconstruction

  • Protocol
  • Intervention



This is the protocol for a review and there is no abstract. The objectives are as follows:

To assess the effects (benefits and harms) of different methods and devices for graft fixation in anterior cruciate ligament (ACL) reconstruction.

In setting out our comparisons, we will group those relating to femoral fixation separately from those relating to tibial fixation. For femoral fixation, we will compare devices based on different mechanisms of fixation, the same mechanism of devices, new versus old, hardware-free versus hardware, and hybrid versus single. For tibial fixation, we will compare intratunnel versus extratunnel fixation devices, and different commonly-used devices.

We will not include the comparison of bioabsorbable versus metallic interference screws, as this is covered in another Cochrane review (Debieux 2012).


Description of the condition

The anterior cruciate ligament (ACL), which is composed of tough fibrous material, is located within the knee joint. It spans from the medial surface of the lateral femoral condyle (part of the lower end of the thigh bone) to the front of the anterior tibial spine (part of the upper end of the shin bone). It acts primarily to maintain joint stability by restraining anterior translation (forward motion) of the tibia (shin bone) relative to the femur (thigh bone).

Anterior cruciate ligament injury, which usually comprises a complete rupture or tear of the ligament, is one of the most common knee injuries (Reinhardt 2010). It typically occurs in people engaging in high-risk sports activities such as football and skiing. Approximately 70% to 84% of ACL injuries are caused by a non-contact mechanism, such as a rapid deceleration, in a sports setting (Boden 2000; Faunø 2006; Griffin 2000; Hewett 2006; Noyes 1983). Thus these injuries usually occur without direct physical contact with other players or stationary objects at the time of injury (Agel 2005; Alentom-Geli 2009; Hohmann 2011).

Treatment of ACL rupture, hence deficiency, can be surgical or non-surgical. Surgery comprises ACL reconstruction, where the damaged ACL is replaced by either an autograft (tissue, such as part of the hamstrings tendon, extracted from the person's own body) or an allograft (a specially treated tendon or ligament extracted from a human cadaver) under arthroscopic control. Collins 2013 reported that 22.6% of diagnosed ACL lesions underwent ACL reconstruction in the three years after initial diagnosis; the majority of reconstructions took place within the first year after injury. It has been estimated that around 80,000 to 100,000 people in the United States undergo anterior cruciate ligament reconstruction each year (Griffin 2000; Grindstaff 2006). Both surgical and non-surgical treatment for ACL rupture are often associated with complications such as meniscal lesions, functional instability and early-onset osteoarthritis (Chhabra 2006; Lohmander 2007).

Description of the intervention

ACL reconstruction involves removal of the damaged ACL, harvesting of the graft (if an autograft is used), preparing of the graft, drilling of tibial and femoral tunnels, placing the graft in an anatomically similar or different position to the original ACL, and fixing the graft. Since the first arthroscopically assisted ACL reconstruction was performed in the UK in 1980 (Dandy 1982), numerous devices and methods for graft fixation have been developed and marketed.

Fixation methods mainly involve fixing soft tissue and fixing bone. The fixation site can be located in the distal femur or the proximal tibia, within a bony tunnel or at a distance from the knee joint. Thus, fixation methods can be classified mainly into four types: tissue fixation in the femoral site, tissue fixation in the tibial site, bone fixation in the femoral site, and bone fixation in the tibial site (Hapa 2009).

Devices identified from a recent literature search, conducted in May 2013, are described in Table 1. This is not an exhaustive list as new kinds of devices and methods are emerging on a regular basis. We have grouped devices by location in our summary below.

Table 1. List of ACL graft fixation devices identified from a recent literature search (22/05/2013)
  1. BTB = bone-patellar-tendon-bone
    ST = soft tissue

Aperfix systemThe Aperfix (Cayenne Ltd) which was introduced in 2009 is a relatively new implant system. It is an all-inside system, requiring no cross pins or additional morbidity to the distal thigh, and aims to provide strong fixation of soft tissue grafts with circumferential aperture compression. This implant is made from polyetheretherketone (PEEK), which is a nonabsorbable, radiolucent material that causes no inflammatory response.
Bone mulch screwThe Bone mulch screw (Arthrotek Ltd) was introduced in 2003 for graft fixation in femoral side. It consists of an outer screw (including a threaded body and an outwardly extending nose portion which provides a mechanical fixation point by functioning as a post within the femoral tunnel) and an inner screw, introduced through the lateral femoral condyle. A hamstring graft is looped around this post, mechanically fixing one end of the graft.
Bone plugPress-fit bone plug fixation consists of a xenogenic spongiosa cylinder (Tubobone Ltd) (length: 30 mm, diameter: to match the diameter of the tibial tunnel) is inserted into the tibial tunnel, and a pusher is used to advance the cylinder to the proximal end of the tibial tunnel.
Button (Endobutton)The Endobutton (Smith & Nephew Ltd) was introduced in 1995 for graft fixation in anterior cruciate ligament reconstruction. The implant, which is widely used for fixation of quadruple hamstring grafts, is relatively cheap and consists of a suture loop and metal plate. The point of fixation is some distance from the joint. Theoretical advantages of this implant are due to a suspension mechanism and include reduced tunnel widening.
Button (Endobutton-CL)The Endobutton-CL (continuous loop) (Acufex or Smith & Nephew Ltd) was introduced in 2003 to solve the problems caused by the Endobutton, such as 'bungee effect' which could contribute to tunnel widening. The stiffer and slippage-free Endobutton-CL implant, therefore, seems to have created a more favourable biomechanical environment, resulting in less tunnel widening.
Button (Endobutton-CL-BTB)The Endobutton-CL-BTB (Smith & Nephew Ltd) was introduced in 2010 for graft fixation in anterior cruciate ligament reconstruction. The implant is composed of an Endobutton and 2 thin continuous polyester loops which are thinner than the loop installed in Endobutton-CL. A decrease in stiffness is considered to be one of the theoretical disadvantages of this system. A theoretical advantage of this implant, which can be used to fix the hamstring tendon graft, includes reducing operation time because the assistant surgeon can start fashioning the hybrid graft immediately after tendon harvest. In addition, the system comprises long and short loops, and can therefore be attached easily to the closed loop, which can also shorten operation time.
Button (flip-button or FlippTack)The flip-button (Karl Storz Ltd) was introduced in 2011 for graft fixation in cruciate ligament reconstruction. The tendons are fixed to this cortical fixation with a double-looped 1 mm Ethibond cord and 6 direction-changing knots.
Button (Retrobutton)The Retrobutton (Arthrex Ltd) was introduced in 2008 for graft fixation in anterior cruciate ligament reconstruction. This implant consists of a flat, 2.5 x 12 mm titanium button with an ultra high-molecular weight polyethylene loop. The loop is continuous, and its portion opposite the button is tripartite.
Button (TightRope RT)The TightRope RT (Arthrex Ltd) was introduced in 2013 for graft fixation in anterior cruciate ligament reconstruction. This implant has adjustable length loops that are tightened intraoperatively.
Button (ToggleLoc)The ToggleLoc (Biomet Ltd) was introduced in 2009 for graft fixation in anterior cruciate ligament reconstruction. The implant is a rectangular titanium rod that uses a suture sling (ZipLoop) to pass the tendon through the bone tunnel once the implant is in place on the cortical bone. This suspensory fixation device aims to provide adequate maximum pullout strength and minimise micromotion because of improved loop material characteristics.
Button (XO Button)The XO Button (ConMed Linvatec Ltd) was introduced in 2010 for fixation of the graft in anterior cruciate ligament reconstruction on the femoral side. This device consists of a metal component that sits on the lateral metaphyseal cortex and incorporates a prestretched continuous graft suspension loop.
CentraLocThe CentraLoc device (Arthrotek Ltd) was introduced in 2007 for soft tissue graft fixation in anterior cruciate ligament reconstruction on the tibial side. This device, which consists of a CentraLoc screw (bicortical screw) and a large “clover” washer with fins, aims to allow circumferential healing of the graft while enabling high cyclic loading.

Cross pin system

(Bilok ST)

The Bilok ST technique (Biocomposites or ArthroCare Ltd) was introduced in 2005 for graft fixation in anterior cruciate ligament reconstruction. This transverse suspension device (diameter: 9 mm; length; 35 mm) involves biodegradable poly-L-lactide/tri-calcium phosphate cross pin constructs whereby the tendon graft wraps 180°around it. Bilok ST is screwed across the tunnel which is different from other cross pin systems. A theoretical disadvantage of this implant is that a poor fit in the bone might result in a small amount of the graft slipping past the distal end of the screw.

Cross pin system


The Pinn-ACL cross pin system (ConMed Linvatec Ltd) is a cancellous suspension device which was introduced in 2010 for graft fixation in anterior cruciate ligament. Theoretical advantages (rigid and strong reconstruction) of this implant are probably due to the different fixation principle used: fixation of the cross pin within the metaphysis results in a relatively short graft length.
Cross pin system (Rigidfix)The Cross pin system (Rigidfix) (DePuy Mitek Ltd) was introduced in 2003 for graft fixation in anterior cruciate ligament reconstruction. The implant consists of 2 bioabsorbable (poly-L-lactide) Rigidfix cross pins which are inserted across the graft through 2 parallel drill holes from the lateral femoral condyle perpendicular to the femoral socket made by a Rigidfix cross pin guide. Theoretical advantages of this implant are due to the expansion mechanism and include good structural properties, maintaining anterior knee stability, higher failure loads, the leeway for graft tunnel mismatch, lower synovitic reactions rate, relatively easy revision surgery and low complication rate. However, the resorbable pins are more expensive than resorbable screws. In addition, the pins penetrate the tendon graft in line with the collagen fibres, so slippage may occur as the collagen fibres separate under tensile stresses. The diameter of the pin is 3.3 mm for soft tissue and 2.7 mm for bone-patellar-tendon-bone. The length is 42 mm.
Cross pin system (Slingshot)The Slingshot systems and Transfix (see next) use the same method for tendon fixation and are very similar, the difference being that a hammer is used for Transfix screws and a screwdriver for slingshot screws.
Cross pin system (Transfix)The Cross pin system (Transfix) (Arthrex Ltd) was introduced in 1998 for graft fixation in anterior cruciate ligament reconstruction. The implant consists of a 50 mm metal or bioabsorbable pin which is inserted through the lateral femoral condyle and the graft passed around the implant with the help of the Transfix guide. The point of fixation is closer to the joint. Theoretical advantages of this implant are due to suspension mechanism and include fewer tunnel enlargements, high failure load, less loss of tension during repetitive loading cycles, fixation closer to the joint line, greater stiffness and pull-out strength and low complication rate. However, this implant is more expensive than the Endobutton. This implant has a range of different lengths from 40 mm to 55 mm.
Double Spike Plate (DSP)The Double Spike Plate (Meira Corp Nagoya Ltd) was developed in 2002 to regulate graft tension by separating the step of tension from that of graft fixation in anterior cruciate ligament reconstruction. The DSP consists of a 1.5 mm thick, 17 mm long and 11 mm wide plate made of titanium alloy with 3 holes and 27 mm long spikes on the reverse side. This implant may facilitate graft fixation under a predetermined amount of tension.
EndoPearlThe EndoPearl (Linvatec Ltd) which was introduced in 2001 is secured with 2 No. 5 Ethibond sutures that are countersunk in the indentations of the device; thus, there is no need for increasing tunnel diameter. The spherical EndoPearl device (diameters 7, 8, 9 mm) is chosen corresponding to the graft diameter.
EvolgateThe Evolgate (Citieffe Ltd) which was introduced in 2003 is a tibial fixation device for soft tissues that consists of 3 parts, all made of a titanium alloy: a screw, a coil that is inserted inside the bone tunnel to reinforce the walls of the tunnel, and a washer for cortical fixation.
Expansion boltThe expansion bolt (Resofix Ltd) (5.8/8.7 × 10 × 35 mm) was introduced in 1999 for graft fixation. It includes a semicircular upper shell, a planar lower shell, and an expansion wedge with an overall height of 5 mm. It is made out of poly-D,L-lactide (Resomer Ltd).
Interference screw (bioabsorbable)The bioabsorbable interference screw was introduced in 1995 for graft fixation in anterior cruciate ligament reconstruction. The materials of this bioabsorbable implant include polyglycolic acid, poly-L-lactic acid, poly-D, L-lactic acid, polyglycolic acid with trimethylene carbonate, poly-L-lactic acid with hydroxyapatite and poly-L-lactic acid with β-tricalcium phosphate. This implant was developed to overcome the shortcomings of the metallic interference screw. The theoretical disadvantages include increased risk of breakage, and risks of infection and foreign body reactions.
Interference screw (metallic)The metallic interference screw was introduced in 1983 for graft fixation in anterior cruciate ligament reconstruction. Advantages of metallic interference screws are high initial fixation strength and the ease of use. The theoretical disadvantages include the laceration of the sutures and grafts during insertion that may complicate the reconstructive surgery. In addition, metallic screws would hinder subsequent magnetic resonance imaging examination. Difficulties may also exist with the removal of the metallic implant in revision surgery.
IntrafixThe Intrafix system (DePuy Mitek Ltd) was introduced in 2003 for graft fixation in anterior cruciate ligament reconstruction. The implant is composed of an expansion sheath made of plastic or absorbable material and a tapered expansion polyethylene screw (diameter: 7 to 10 mm). Theoretical advantages of this implant are due to this combination device and include clearly superior strength in the fixation.
Ligament plateThe Ligament plate (Solco Ltd) was introduced in 2009 for graft fixation. It is made of a titanium-based alloy and has a U-shaped body for graft suspension and 2 wings that contain screw holes to allow secure femoral fixation with the screw.
Linx-HTThe Linx-HT (Mitek Ltd) was introduced in 2005. It suspends the graft from a screw that is fixed into the cancellous bone of the femoral metaphysis.
Metallic set screw systemThe Metallic set screw system (Arthrex Ltd) consists of a 9 × 30 mm metallic screw augmented with a 2.4 × 45 mm metallic pin that can be used for transverse femoral graft fixation.
StapleThe staples (Smith & Nephew Ltd) are often used as a supplementary device with an interference screw for tibial fixation.
Swing bridgeThe Swing bridge (Citieffe Ltd) was introduced in 2003 for graft fixation. It has an eyelet, through which the tendons are directly looped, and a smooth metal half ring for cortical suspension fixation on the femoral side when the device is inserted and impacted using an out-in technique.
WasherThere are 3 Washer devices for soft tissue grafts fixation in tibial side. One is the Washerlock (Arthrotek Ltd) which was introduced in 1999. It is 20 mm in diameter, has 4 11 mm long peripheral spikes that straddle the graft, and 19.6 mm-long centrally placed spikes that penetrate the graft in multiple locations. The others are Tandem Washers (2 soft tissue washers, Synthes Ltd) and a 20 mm spiked metal Washer (Linvatec Ltd). A 20 mm diameter metal Washer with 12 x 1.3 mm length spikes is used to compress the graft to the tibial cortex with a 4.5 mm diameter bicortical screw. Fixation with the Tandem Washers requires 2 4.5 mm diameter bicortical screws and 2 13.5 mm diameter plastic spiked washers spaced 15 mm apart in tandem.

Devices for femoral fixation can be divided according to their underlying mechanisms: compression (producing compressive loads to the longitudinal axis of the graft), expansion (producing a bulging of the graft) or suspension (suspending the graft into the femoral tunnel). Typical examples of compression devices are interference screws (bioabsorbable or metallic) and bone plug. The cross pin (Rigidfix) system is a popular technique among expansion mechanisms. There are also some other devices which adopt a suspension mechanism and are fixed more or less far away from the knee joint, including three subdivisions according to the type of bone: cortical, cortical-cancellous or cancellous. Examples of cortical suspension devices are buttons, Swing Bridge, and Ligament Plate. Cortical-cancellous suspension devices consist of cross pin (Transfix), cross pin (Sling Shot) and bone mulch screw. Linx-HT is a cancellous suspension device.

Tibial fixation is considered to be the 'weak link' of initial graft fixation because of the lower bone mineral density of the proximal tibial compared with the distal femur and the angle at which the forces are applied to the graft on the tibial side (Brand 2000). Devices for tibial fixation can be divided according to the location of fixation: intratunnel fixation and extratunnel fixation. Intratunnel fixation methods primarily rely on the metallic or bioresorbable interference screw, a relatively novel approach called Intrafix, or a cross pin system.

How the intervention might work

The outcome of the reconstruction often depends on the outcome during the so-called 'weak link' time; in this period, the process of tendon-bone healing is incomplete. This is considered to last from six weeks to six months after surgery, while the bone or tissue graft is incorporated into the bony tunnel (Rodeo 1993). Since early post-surgical rehabilitation is considered important to facilitate full weight-bearing, full range of motion and early return to normal social and sporting life (Shelbourne 1990), there has been a focus on strong graft fixation. Clinical outcomes may be compromised, however, when using fixation methods with poor structural properties such as inadequate strength, stiffness or slippage.

Of the many graft fixation devices available, the anatomic (apertural, intratunnel) fixation, which is achieved with direct tendon-to-bone screw fixation, has gained wide popularity in the past decades. A typical example of anatomic fixation is the interference screw which was originally described in 1983 (Lambert 1983). This intratunnel fixation, which is more anatomically and biomechanically similar to the native ACL (Kurosaka 1987), may result in minimal graft elongation potentially leading to a more stable long-term result. However, possible disadvantages of this fixation method include: bone screw divergence; inappropriate insertion depth; presence of intra-articular hardware especially for metallic screws interfering with postoperative magnetic resonance imaging; difficulties in removal in case of revision surgery; damage of bone-tendon junction during screw placement; and violation of the posterior cortex.

To address these problems, various techniques with different mechanisms have been devised. For example, transverse femoral fixation, where the graft is fixed using either expansion (Rigidfix), lateral compression (setscrew), or a suspension (TransFix) mechanism. One potential benefit of these devices may be that they do not cause graft maceration. In addition, this technique may have applications in revision cases. Obvious disadvantages of this method are firstly that fixation of the bone block or soft tissue is central rather than along its entire length. Secondly, this mechanism reproduces a non-anatomic placement of the femoral insertion of the graft. Thirdly, graft fixation is not performed under direct visualisation.

Button fixation, at a distance from the joint line, is one of the most popular devices. This cortical suspensory graft fixation can fix both soft tissue (EndoButton, EndoButton-CL) and bone block (EndoButton-CL-BTB), and can also be used on the tibial side. Even though it has been criticised because of the so-called 'bungee-cord' (along-length graft-tunnel motion) and 'windshield-wiper' (across graft-tunnel motion) effects, this suspensory fixation has been favoured by many surgeons for its substantial initial fixation strength and the ease of use. Notably, tunnel widening is a problem that needs particular attention in terms of this technique.

Press-fit bone plug, which was first introduced in 1999, is a hardware-free method that may limit tunnel enlargement and thereby enhance tendon-to-bone healing. Satisfactory tensile strength and stiffness for this device have been shown by some biomechanical studies (Lee 2003; Pavlik 2003; Rupp 1997). Furthermore, this hardware-free technique avoids the problems associated with removal of the hardware in revision surgery. The Press-fit hardware-free technique is an example of this method.

Since each device has its possible benefits and potential harms, some hybrid fixations have been developed and may be of some value especially in the case of poor bone quality, such as revision surgery with tunnel widening, or in older people. A typical example of hybrid fixation is a bioabsorbable interference screw with some kind of button (such as Endobutton, Endobutton-CL, EndoPearl, or Flip-Button), which may help overcome concerns about low initial fixation strength and construct slippage when using bioabsorbable interference screw or button alone, while still allowing an all-inside ACL reconstruction. Some biomechanical studies have shown that the combined technique was able to withstand higher failure loads and was also stiffer than any one of them, and could potentially augment fixation in people with poor bone density or graft and tunnel diameter disparity (Walsh 2009; Weiler 2001).

Why it is important to do this review

ACL reconstruction is a common orthopaedic procedure that often successfully allows the person to return to their former activities after extensive rehabilitation. Nonetheless, early failure of the graft can occur and this can sometimes be directly linked with or partly related to the type of device used or poor methods, or both. The selection of the method or the device or both may also adversely affect revision ACL reconstruction or outcome in general. Furthermore, there is some evidence of later deterioration and complications such as meniscal lesions, functional instability and early-onset osteoarthritis (Chhabra 2006; Lohmander 2007). There are also great variations in practice and numerous devices on the market. Three recently published reviews (Colvin 2011; Han 2012; Ilahi 2009) compared different methods and devices in ACL reconstruction (except for comparison of bioabsorbable versus metallic interference screws). All three reviews are more limited in their scope than our planned review and used different suboptimal methods that may result in biased conclusions. For example, two reviews (Han 2012; Ilahi 2009) used search strategies restricted to MEDLINE and English-language publications, which could result in publication bias. In addition, none of the three reviews restricted their included studies to randomised controlled trials. We hope to inform practice in this area by conducting a systematic review of high quality, that includes a more comprehensive and up-to-date search of the literature and the best quality evidence from randomised controlled trials.


To assess the effects (benefits and harms) of different methods and devices for graft fixation in anterior cruciate ligament (ACL) reconstruction.

In setting out our comparisons, we will group those relating to femoral fixation separately from those relating to tibial fixation. For femoral fixation, we will compare devices based on different mechanisms of fixation, the same mechanism of devices, new versus old, hardware-free versus hardware, and hybrid versus single. For tibial fixation, we will compare intratunnel versus extratunnel fixation devices, and different commonly-used devices.

We will not include the comparison of bioabsorbable versus metallic interference screws, as this is covered in another Cochrane review (Debieux 2012).


Criteria for considering studies for this review

Types of studies

We will include all randomised or quasi-randomised (i.e. allocating participants to a treatment which is not strictly random: e.g. by hospital number) controlled trials comparing methods and devices for fixation in anterior cruciate ligament reconstruction tested within the clinical setting.

Types of participants

We will include trials of adults undergoing surgical reconstruction for a ruptured anterior cruciate ligament. Studies involving a policy of surgical treatment of other concomitant soft-tissue knee injuries, such as meniscal tear, in the same operation as anterior cruciate ligament reconstruction will be included, provided this applies to both groups.

Types of interventions

We will include comparisons of any type of device, apart from bioabsorbable with metallic interference screws (Debieux 2012), for graft fixation in various kinds of ACL reconstruction.

Some main comparisons are listed below.

Femoral fixation
1. Devices based on different mechanisms
  • Suspension versus expansion (e.g. button versus cross pin (Rigidfix))

  • Expansion versus compression (e.g. cross pin (Rigidfix) versus interference screw)

  • Suspension versus compression (e.g. button versus interference screw)

  • Transverse femoral screw versus interference screw

2. Devices based on the same mechanism
  • Different compression devices

  • Different expansion devices

  • Different suspension devices (including consideration of the type of bone)

3. New versus old
  • Endobutton-CL versus Endobutton

  • Transfix-II versus Transfix

  • Rigidfix versus Transfix

4. Hardware-free versus hardware
  • Press-fit bone plug versus button

  • Press-fit bone plug versus cross pin

  • Press-fit bone plug versus interference screw

5. Hybrid versus single
  • Button + interference screw versus button alone

  • Button + interference screw versus interference screw alone

Tibial fixation
  • Intratunnel versus extratunnel

  • Intrafix versus interference screw

  • Press-fit bone plug versus interference screw

  • Interference screw + staple versus interference screw alone

Types of outcome measures

Primary outcomes
  • Function and quality of life, preferably based on validated patient-reported outcome measures of knee function (e.g. the International Knee Documentation Committee (IKDC) (Hefti 1993), the Life Quality evaluation of ACL (Mohtadi 1998), the Lysholm score (Tegner 1985))

  • Surgery failure and adverse events (implant breakage, implant dislocation, graft rupture, deep vein thrombosis, superficial and deep infection, complex regional pain syndrome, need for revision surgery)

  • Activity level (Tegner activity level (Tegner 1985), time to recover normal gait, stair climbing, running gait and sprinting)

Secondary outcomes
  • Clinician-rated scores (IKDC objective part)

  • Knee stability (KT-1000 arthrometer (Balasch 1999), KT-2000 arthrometer (Myrer 1996))

  • Range of knee motion

  • Pain (visual analogue scale (VAS) score)

  • Clinical examination (Lachman, pivot-shift-test)

  • Objective function test (single hop test)

Timing of outcome assessment

Outcome assessment will be based on short-term (within six months after ACL reconstruction), intermediate-term (between six months and two years after ACL reconstruction) and long-term follow-up (more than two years after ACL reconstruction).

Search methods for identification of studies

Electronic searches

We will search the Cochrane Bone, Joint and Muscle Trauma Group Specialised Register (to the present), the Cochrane Central Register of Controlled Trials (CENTRAL) (The Cochrane Library current issue), MEDLINE (1946 to the present) and EMBASE (1980 to the present). We will also search the WHO International Clinical Trials Registry Platform, and Current Controlled Trials for ongoing trials. We will place no constraints based on language or publication status.

In MEDLINE (OVID Online), we will combine the sensitivity-maximising version of the Cochrane Highly Sensitive Search Strategy for identifying randomised trials (Lefebvre 2011) with the subject-specific search. The search strategies for The Cochrane Library (Wiley Online Library), MEDLINE and EMBASE (OVID Online) are shown in Appendix 1.

Searching other resources

We will check the bibliographies of relevant papers identified by the search strategies. We will also include the findings from handsearches of meetings of the European Society of Sports Traumatology Knee Surgery and Arthroscopy (ESSKA), American Orthopaedic Society for Sports Medicine (AOSSM), International Society of Arthroscopy, Knee Surgery and Orthopaedic Sports Medicine (ISAKOS), American Academy of Orthopaedic Surgeons (AAOS), World Congress on Orthopaedic Sports Trauma, and Arthroscopy Association of North America (AANA).

Where necessary we will contact the corresponding authors of studies identified by the search strategies for any unpublished results and data.

Data collection and analysis

Selection of studies

Two authors (GSG and LW) will independently assess and select potentially eligible trials identified by the search strategy. We will approach trial authors for further details when necessary. We will resolve any disagreement by discussion.

Data extraction and management

Two authors (ZC and LGH) will independently extract data using a data extraction form. We will resolve any differences by discussion. Two authors (GSG and LW) will enter data into Review Manager 5 (RevMan 2012). We will contact trial authors for additional information or data when necessary and practical.

Assessment of risk of bias in included studies

All four authors will independently assess the risk of bias in the included trials. We will appraise the following aspects according to The Cochrane Collaboration's 'Risk of bias' tool (Higgins 2011): sequence generation (selection bias), allocation concealment (selection bias), blinding of participants and personnel (performance bias), blinding of outcome assessors (ascertainment bias), incomplete outcome data (attrition bias), selective outcome reporting bias and other potential sources of bias (including sponsorship bias, differences in rehabilitation and performance bias relating to surgeon’s experience, especially with the devices). We will resolve any disagreements by discussion.

Measures of treatment effect

We will calculate risk ratios (RRs) with accompanying 95% confidence intervals (95% CIs) for dichotomous outcomes, and mean differences (MD) with 95% CIs for continuous outcomes. When pooling continuous data from outcomes measured in different ways, we will use standardised mean differences (SMDs) and 95% CIs.

Unit of analysis issues

We anticipate that the unit of randomisation in the included studies will usually be the individual participant. However, where trials include adults with bilateral injuries, results may be presented for knees or limbs rather than for individual participants. Where such unit of analysis issues arise and appropriate corrections have not been made, we will consider presenting the data for such trials only where the disparity between the units of analysis and randomisation is small. We will avoid unit of analysis issues relating to repeated observations of the same outcome, such as results presented for several periods of follow-up.

Dealing with missing data

We will calculate missing data, such as standard deviation from standard error, 95% confidence intervals or exact P values. If this is not feasible, we will try to contact the trial authors for missing data. Where possible and appropriate, we will conduct intention-to-treat analyses to include all participants randomised to the intervention groups. If it is impossible to acquire missing data, such as the standard deviations, we will list the original results rather than imputing data into analyses. Furthermore, we will consider the potential impact of missing data on the findings of the review. For example, we will investigate the effects of dropouts and exclusions by conducting 'worst-case' and 'best-case' scenario analyses.

Assessment of heterogeneity

We will evaluate heterogeneity by visual inspection of the confidence intervals overlap and by using the standard Chi² test, with additional consideration of the I² statistic. We will base our interpretation of the I² results as suggested in Higgins 2011: 0% to 40% might not be important; 30% to 60% may represent moderate heterogeneity; 50% to 90% may represent substantial heterogeneity; and 75% to 100% may represent considerable heterogeneity.

Assessment of reporting biases

If sufficient studies (at least 10) are available, we will assess publication bias by examining funnel plot asymmetry.

Data synthesis

When we consider it appropriate, we will pool results of comparable groups of trials using both fixed-effect and random-effects models. The choice of the model to report will be guided by a careful consideration of the extent of heterogeneity and whether it can be explained, in addition to other factors such as the number and size of studies that are included. We will use 95% confidence intervals throughout. We will consider not pooling data where there is considerable heterogeneity (I² > 75%) that cannot be explained by the diversity of methodological or clinical features among the trials. Where it is not appropriate to pool data, we will still present trial data in the analyses or tables for illustrative purposes and report these in the text.

Subgroup analysis and investigation of heterogeneity

Where sufficient data are available, we will conduct subgroup analysis to explore the following aspects.

  • Different types of surgical techniques (double-bundle versus single-bundle)

  • Different types of graft (patellar versus hamstring tendon; autograft versus allograft)

  • Different locations of devices (tibial versus femoral fixation of the graft, or the same device used in both the tibial and femoral tunnel)

  • Different devices on the other side of the graft in both the experimental and control group

  • Different time interval between ACL injury and reconstruction surgery (acute (less than two months) versus chronic (more than two months) (Harilainen 2009; Mariani 2001))

We will investigate whether the results of subgroups are significantly different by inspecting the overlap of confidence intervals and performing the test for subgroup differences available in RevMan.

Sensitivity analysis

We plan to conduct sensitivity analyses to investigate the influence of the different aspects of trial and review methodology, including the effects of including trials at high or unclear risk of selection bias from lack of allocation concealment or performance bias relating to differences in surgeons' experience in the devices, the effects of missing dichotomous data and the selection of statistical model (fixed-effect versus random-effects) for pooling.

'Summary of findings' tables

Where there are sufficient data, we will summarise the results for the main comparisons described in Types of interventions in 'Summary of findings' tables. We shall use the GRADE approach to assess the quality of evidence related to each of the primary outcomes listed in Types of outcome measures (Higgins 2011; see section 12.2).


We would like to thank Mario Lenza and Nikolaos Paschos for helpful comments at editorial and external review of this protocol. We would also like to thank Lindsey Elstub and Diane Horsley for editorial assistance, and Joanne Elliott for developing the search strategies.


Appendix 1. Search strategies

The Cochrane Library (Wiley Online Library)

#1 MeSH descriptor: [Anterior Cruciate Ligament] this term only  
#2 MeSH descriptor: [Anterior Cruciate Ligament Reconstruction] this term only 
#3 (anterior near/2 cruciate* near/2 ligament*) or ACL:ti,ab,kw (Word variations have been searched)  
#4 #1 or #2 or #3
#5 MeSH descriptor: [Orthopedic Fixation Devices] this term only 
#6 (pin* or nail* or screw* or plate* or fix* or endobutton* or bone plug or transfix* of bio-transfix* or biotransfix* or staple* or bolt* or slingshot* or washer* or ligament plate* or swing bridge* or intrafix* or biofix* or bioscrew* or linx-ht or endopearl or evolgate or suture*):ti,ab,kw (Word variations have been searched) 
#7 MeSH descriptor: [Bone Nails] this term only 
#8 MeSH descriptor: [Bone Plates] this term only 
#9 MeSH descriptor: [Bone Screws] this term only 
#10 MeSH descriptor: [External Fixators] this term only 
#11 MeSH descriptor: [Internal Fixators] this term only 
#12 MeSH descriptor: [Sutures] this term only    
#13 #5 or #6 or #7 or #8 or #9 or #10 or #11 or #12 
#14 #4 and #13 


1 Anterior Cruciate Ligament/
2 Anterior Cruciate Ligament Reconstruction/
3 ((anterior adj2 cruciate$ adj2 ligament$) or ACL).tw.
4 1 or 2 or 3
5 Orthopedic Fixation Devices/
6 (pin$1 or nail$ or screw$1 or plate$ or fix$ or endobutton$ or bone plug or transfix$ of bio-transfix$ or biotransfix$ or staple$ or bolt$ or slingshot$ or washer$ or ligament plate$ or swing bridge$ or intrafix$ or biofix$ or bioscrew$ or linx-ht or endopearl or evolgate or suture$).tw.
7 Bone Nails/ or Bone plates/ or Bone screws/ or External Fixators/ or Internal Fixators/ or Suture/
8 5 or 7
9 4 and 8
10 Randomized controlled
11 Controlled clinical
12 randomized.ab.
13 placebo.ab.
14 Drug Therapy.fs.
15 randomly.ab.
16 trial.ab.
17 groups.ab.
18 or/10-17
19 exp Animals/ not Humans/
20 18 not 19
21 9 and 20


1 Anterior Cruciate Ligament/ or Anterior Cruciate Ligament Reconstruction/ or Anterior Cruciate Ligament Rupture/
2 ((anterior adj2 cruciate$ adj2 ligament$) or ACL).tw.
3 1 or 2
4 Orthopedic Equipment/
5 (pin$1 or nail$ or screw$1 or plate$ or fix$ or endobutton$ or bone plug or transfix$ of bio-transfix$ or biotransfix$ or staple$ or bolt$ or slingshot$ or washer$ or ligament plate$ or swing bridge$ or intrafix$ or biofix$ or bioscrew$ or linx-ht or endopearl or evolgate or suture$).tw.
6 External Fixator/ or Internal Fixator/ or Bone Nail/ or Bone Plate/ or Bone Screw/ or Fixation Device/ or Suture Anchor/
7 4 or 5 or 6
8 3 and 7
9 Randomized Controlled Trial/
10 Clinical Trial/
11 Controlled Clinical Trial/
12 Randomization/
13 Single Blind Procedure/
14 Double Blind Procedure/
15 crossover procedure/
16 Placebo/
17 Prospective Study/
18 ((clinical or controlled or comparative or placebo or prospective$ or randomi#ed) adj3 (trial or study)).tw.
19 (random$ adj7 (allocat$ or allot$ or assign$ or basis$ or divid$ or order$)).tw.
20 ((singl$ or doubl$ or trebl$ or tripl$) adj7 (blind$ or mask$)).tw.
21 (cross?over$ or (cross adj1 over$)).tw.
22 ((allocat$ or allot$ or assign$ or divid$) adj3 (condition$ or experiment$ or intervention$ or treatment$ or therap$ or control$ or group$)).tw.
24 or/9-23
25 Case Study/ or Abstract Report/ or Letter/
26 24 not 25
27 8 and 26
28 Animal/ or Animal Experiment/ or Nonhuman/
29 Human/
30 and/28-29
31 28 not 30
32 27 not 31

Contributions of authors

ZC, WJ and LGH were involved in the conception and design of the review.
ZC and WJ drafted the protocol.
LGH edited the protocol.
YTB, GSG and LW helped find articles related to the protocol.
LGH is the guarantor of the review.

Declarations of interest

Chao Zeng - none known
Guanghua Lei - none known
Shuguang Gao - none known
Wei Luo - none known

Sources of support

Internal sources

  • Central South University, China.

External sources

  • No sources of support supplied