One‐stage revision anterior cruciate ligament reconstruction: Preoperative evaluation, planning and surgical techniques. A review of current concepts

Abstract

The increased rate of anterior cruciate ligament (ACL) tears has led to a greater number of revisions. Revision surgery can be performed in one or two stages. Single‐stage revision ACL reconstruction (ssRACLR) may be performed when prior tunnels can be re‐used or bypassed whereas a two‐stage procedure is indicated when bone grafting of dilated tunnels prior to revision is necessary. While both approaches have shown similar functional outcomes and failure risk, ssRACLR is preferred, when possible, to avoid the increased morbidity, inconvenience and cost associated with two‐stage RACLR. In adequately planning for RACLR, a surgeon should investigate the mechanism and timing of injury as well as the previous graft selection, associated pathology and the tunnel placement and size. It is especially important to obtain radiographs and three‐dimensional imaging including magnetic resonance imaging (MRI) and computed tomography (CT), which allow the surgeon to accurately evaluate the entire tunnel architecture to determine surgical staging. Following a detailed assessment of the pathoanatomy, the surgeon may determine graft and hardware type, tunnel placement and utilization of lateral extra‐articular tenodesis (LET) and other procedures. In our experience, ssRACLR can be carried out for over 90% of revision cases with creative pre‐operative planning using autograft with bone plug(s), divergent tunnel creation on the femur (when necessary) and convergent tunnel creation on the tibia (when appropriate) and suspensory or interference fixation as needed. In revision scenarios, we believe that autografts with bone plugs provide the best opportunity for graft healing and incorporation and that LET can be a useful adjunct to reduce re‐tear rates. The purpose of this review is to report on the preoperative considerations and surgical techniques for performing ssRACLR, as well as the outcomes.

Level of Evidence

Level V expert opinion.

Abbreviations

3D

three dimensional

ACL

anterior cruciate ligament

ACLR

anterior cruciate ligament reconstruction

AP

anteroposterior

BTB

bone‐patellar tendon‐bone

CT

computed tomography

HS

hamstring

IS

interference screw

ITB

iliotibial band

LCL

lateral collateral ligament

LET

lateral extra‐articular tenodesis

MCL

medial collateral ligament

Metal IS

MIS

MRI

magnetic resonance imaging

PCL

posterior cruciate ligament

PFJ

patellofemoral joint

PLC

posterolateral corner

PRO

patient‐reported outcome

PT

physical therapy

QT‐B

quadriceps with bone plug

RACLR

revision ACL reconstruction

RCI

RetroConstruction Interference screw

ROM

range of motion

RTS

return to sport

ssRACLR

single‐stage revision ACL reconstruction

TKA

total knee arthroplasty

TOB

tie‐over‐button

INTRODUCTION

The increased incidence of anterior cruciate ligament (ACL) tears [33] has led to a ‘domino effect’ of increased rates of ACL reconstruction (ACLR) [42, 112] and subsequently, revision ACLR (RACLR). Revision rates are reported between 4.1% and 13.3% [68, 80, 102] and are significantly higher in adolescents than in adults [26, 42]. Failure of ACLR is a devastating event and is caused by a variety of factors including trauma, graft type, poor surgical technique, infection and associated knee pathology not addressed at prior surgery [7, 39, 91]. Patients with ruptured ACL grafts have a significantly increased risk of developing osteoarthritis [96].

Revision surgery is performed in one or two stages. Single‐stage revision ACLR (ssRACLR) may be done when the prior tunnels can be re‐used or bypassed and it is not necessary to perform bone grafting of dilated tunnels before revision. Most ACL revision surgeries are done in a single stage, with two‐stage procedures reported to be conducted in 8%–9% of patients [39]. Some studies have shown that there are no major differences in functional outcome and failure risk between one‐stage and two‐stage revision for ACLR [71, 80]. However, increased time between graft failure and revision is associated with osteoarthritis, meniscal lesions and decreased activity [18]. The two‐stage revision also carries the innate drawbacks of staged surgery: inconvenience, personal cost, morbidity, and longer overall rehabilitation [18, 28, 92, 103]. Thus, it is desirable to avoid two‐stage revision when possible.

In planning for ssRACLR, the surgeon must account for the previous graft, prior tunnel placement and size, the presence of hardware and the age and desired activity level of the patient [39]. Here, we report on our method for preoperative planning and surgical techniques for performing ssRACLR in almost all of our revision cases.

PREOPERATIVE EVALUATION

This section's purpose is not to serve as an extensive guide on diagnosing ACLR failure, but to highlight details that can affect surgical planning.

History and physical

The goal is to avoid repeat surgery and enable the patient to return to their desired activity level without residual symptoms of pain or instability. A thorough preoperative workup is necessary to identify the cause of prior failure and to plan the revision. The primary concern is the history of instability. Many patients, especially those with multiple ipsilateral ACL revisions, report chronic instability without a definitive injury since their surgery. Upon further questioning, some patients who present following new trauma also claim to have never fully trusted their knee [39]. Understanding the mechanism of injury or absence of re‐injury provides insight into potential mechanical (graft size, strength, nature/material, irradiation, etc.), technical (tunnel location, graft impingement, tension, fixation, etc.) or biological (failure of graft incorporation and ligamentization) causes of failure [12, 77]. Timing of the injury in relation to the initial procedure is important: failures within 6 months of surgery often stem from technical errors or re‐injury, while later failures are more typically related to trauma [59]. Patient factors such as age, activity level, biological sex and body mass index should be considered as well [51]. It is helpful to ascertain prior drilling technique, graft choice, hardware and graft availability. A detailed history, physical examination and imaging can identify associated diagnoses including Osgood–Schlatter disease and patellar tendonitis that could affect graft selection [13, 57].

Physical examination should start with the patient standing and walking to evaluate for varus or valgus thrust due to collateral ligament laxity as well as alignment [76]. Inspection of scars from previous surgery is helpful as well. Note should be taken of any effusion visible during the acute phase or erythema that could suggest infection [113]. It is important to assess range of motion (ROM) including terminal extension and flexion [104]. Ligament tests are crucial for ACL re‐injury evaluation, with Lachman being the most sensitive and pivot shift the most specific for ACLR failure [58, 97]. However, recent injury may limit the value of these tests due to guarding. In addition to the ACL, the collaterals, posteromedial and posterolateral corners (PLCs), posterior cruciate ligament (PCL) and menisci should be examined with the uninjured knee always used as a reference [100].

Imaging

Radiographs should be part of every ACLR failure workup, including anteroposterior (AP), lateral, Merchant and Rosenberg (flexed view PA) view [29]. For suspected alignment abnormalities and/or when osteotomy is potentially indicated, hip‐to‐ankle radiographs are indicated [47]. Radiographs are useful to evaluate degenerative changes, joint space, tunnel location, tunnel size and hardware. Tunnel location is crucial for understanding the cause of failure and for planning revision surgery. Various methods accurately evaluate tunnel position without magnification bias [98]. Bernard et al. [10] described a method in which the distal femur is divided into a 4 × 4 grid on lateral X‐ray, using Blumensaat's line as the upper limit. Height is defined as the distance from the line to the most distal part of the condyle. The position of the femoral tunnel is then described relative to this grid. Ideally, the centre of the aperture should be in the inferior‐anterior part of the most superior‐posterior section (Figure 1a).

Figure 1.

Tunnel position assessment using lateral X‐ray views. (a) Lateral X‐ray of left knee showcasing the grid method [10]. The distal femur is divided into a 4 × 4 grid (by interrupted blue lines) on lateral X‐ray, using Blumensaat's line (solid blue line at the intercondylar notch) as the upper limit. Height (solid blue lines perpendicular to Blumensaat's line) is defined as the distance from the line to the most distal part of the condyle (solid blue line parallel to Blumensaat's line). The position of the femoral aperture centre (outer red circle is the aperture, inner red circle is its centre) should be in the inferior‐anterior part of the most superior‐posterior section (green dot). (b) Lateral X‐ray of the right knee showcasing tibial aperture placement assessment. The tibial plateau (solid blue rounded rectangle) is divided (interrupted blue lines) into quadrants. The ideal position of the tibial aperture centre (red dot) is considered to be about 40%–50% (green highlight) from anterior to posterior (posterior third of the second quadrant) [41]. Created in BioRender.com.

Lateral X‐ray is also important for evaluation of the tibial tunnel. The tibial plateau is divided into quadrants, and the tunnel's position is described relative to the anterior and posterior margins of the tibia. The ideal position of the tibial aperture centre is considered to be about 40%–50% (posterior third of the second quadrant) from the anterior to posterior [41] (Figure 1b). Impingement can also be evaluated on lateral radiographs with the knee in extension by extending a line from the anterior tibial tunnel to the femur; if this line projects anterior to the Blumensaat line, graft impingement is possible [44].

Three‐dimensional (3D) imaging is critical for surgical planning. Magnetic resonance imaging (MRI) evaluates graft integrity, menisci, cartilage and ligaments as well as arthrofibrosis and the presence of a cyclops lesion [88]. Computed tomography (CT) highlights bone and hardware to better evaluate tunnel widening and hardware position. Use of CT is advised when MRI and radiographs are inadequate to determine tunnel placement and size, inform fixation options or when radiographs show osteolysis or bone defects that are not well defined on the MRI images [100]. For those with less experience in revision ACL reconstruction, we would recommend obtaining a CT scan for every case. CT provides detailed information on bone quality, tunnel architecture, positioning and diameter and may be easier for the surgeon to interpret [70, 74].

In a similar fashion to the X‐ray, all CT planes should be assessed to get a comprehensive view of the tunnels. The addition of the axial plane allows for a more holistic view of the overall tunnel trajectory and the apertures, especially the tibial, which is perpendicular to the axis. Similar techniques as the ones described above for X‐ray can be used to assess tunnel position [85, 98, 109]. Tunnel size measurements should be taken at various points throughout the tunnel in all planes, as singular measurements, usually maximal diameter [56, 70, 87], are not representative of the true tunnel architecture. The first 10 mm near the apertures are of particular interest because [55, 62] they can affect the ability to arrive at an anatomic aperture through diverting, blending or entirely new drilling. The widest point and the exit of the tunnel [55, 87, 94] are crucial to whether there is adequate bone to host a revision tunnel with an acceptable trajectory. While various attempts have been made to reach a classification of tunnels based on position and size on CT [17, 65], it is still a very complex issue that needs a case‐specific and more sophisticated approach, including other factors such as tunnel morphology and bone quality. CT scan also allows for a more complex evaluation of hardware positioning, as radiopaque devices can be visualized (compared to X‐ray) and the devices' true position relative to the existing tunnel can be assessed clearly, facilitating decisions regarding revision tunnel placement and the need for hardware removal. 3D CT‐based reconstruction models can be useful to better understand the pathoanatomy of the knee. The best‐fit cylinder method of measuring tunnels, described by Crespo et al. [14], correlates strongly with drill size, aiding in planning RACLR by optimizing new tunnel size and trajectory. However, 2D CT has been described as more reliable for tunnel wall thickness [14].

PREOPERATIVE PLANNING

Decision‐making regarding single‐staged or two‐stage revision ACL reconstruction

Considerations regarding the stages needed for a successful RACLR can become very complex in certain cases. In our opinion, meticulous preoperative assessment and planning should allow for the final decision to be made before entering the operating room. Our goal is to avoid two‐stage RACLR, whenever possible, to avoid increasing the inconvenience and morbidity for the patient. Even though current evidence seems inconclusive [38, 83], we also want to avoid potential additional damage to intra‐articular structures, particularly the menisci and cartilage, secondary to a long ACL‐deficient interval. That being said, we always carry out a staged RACLR when the cause of failure is an active infection [73]. Additional situations where a two‐stage RACLR would be an absolute indication are summarized in Table 1 [24, 53].

Table 1.

Absolute indications for proceeding with a two‐stage RACLR. In these scenarios, a ssRACLR is not achievable without significantly compromising surgical outcomes [20, 24, 53].

Absolute indications for a two‐stage RACLR

ACL failure secondary to active infection. Failed double‐bundle ACL reconstruction with large footprint and substantial combined tunnel defect that would not allow for anatomic ssRACLR. Retained hardware would require removal for anatomic tunnel placement, and its removal would require creating a substantial bone defect based on pre‐operative assessment. Arthrofibrosis after ACLR requires arthroscopic release to restore full ROM.

Abbreviations: ACL, anterior cruciate ligament; ROM, range of motion; ssRACLR, single‐stage revision ACL reconstruction.

Severe tunnel widening, indicated by a 16 mm widening perpendicular to the tunnel axis, has been described as requiring a two‐stage reconstruction [32, 64]. However, because tunnels are a three‐dimensional structure, we do not strictly adhere to this rule. Rather, we focus on a tunnel that can accommodate a graft with interference fixation securely. Previous tunnel placement is equally as important as size, and when the combination of size and location allows for the creation of a secure, divergent tunnel, we opt for a ssRACLR. We also try to account for the size of the tunnel relative to that of the distal femur which accommodates it. Alternatively, we use suspensory fixation on the femoral side when greater than 50% of the tunnel wall is fresh bone to allow for graft healing and incorporation. While it has been described as a relative indication for a two‐stage procedure, we do not typically perform meniscus transplantation in our revision cases because, in our experience, the methods described below are adequate without transplant, and the durability of meniscus transplantation for patients returning to sport is unpredictable [1]. Finally, although technically not a two‐stage RACLR, corrective osteotomy for malalignment may be carried out in a different stage than the actual revision [20, 32, 53].

Graft selection and use of lateral extra‐articular tenodesis

Clinical considerations for graft type include previous grafts used, available options, type and location of hardware, ipsilateral and contralateral patellofemoral joint (PFJ) condition, tunnel size and patient characteristics and preference. Autografts, if available, are preferred due to the decreased risk of graft failure as compared to allografts [15, 35, 46, 54, 61, 66, 92]. However, autograft options are sometimes limited in revision settings, especially in patients with multiple ACLRs and patient reluctance. For these cases, allografts may be considered [67], allowing the surgeon to minimize donor site morbidity and reduce surgical time [54, 72].

Regarding graft choice, ipsilateral bone‐patellar tendon‐bone (BTB) autograft is preferred due to the low rate of re‐tear [31, 61, 84]. For example, BTB is ideal for active, skeletally mature patients who had primary reconstruction during adolescence with hamstring (HS) autograft, quadriceps autograft or allograft. Repeat harvesting of BTB is associated with poor results and is not indicated [49, 50, 60]. If ipsilateral BTB is unavailable, contralateral BTB is a viable option with excellent results, reported in the literature [25]. Contralateral BTB may also be considered over ipsilateral BTB for cases with a large Osgood Schlatter‐related ossicle on the ipsilateral tibial tubercle. When BTB is unavailable, the quadriceps with patella bone plug (QT‐B) is an exceptional option. We prefer QT with bone in the revision setting as it allows for the bone to be positioned on the femoral side for bone‐to‐bone healing with either interference screw or suspensory fixation [16, 95]. Ipsilateral QT‐B is also a good option for patients wanting to avoid contralateral procedures after reconstruction with ipsilateral BTB. Planning for ipsilateral QT‐B after ipsilateral BTB includes evaluation of patellar integrity with CT to make sure adequate bone will be left in place between the proximal and distal harvesting sites [43]. While we prefer grafts with bone plugs for revision surgery, HS may be considered for older patients who are less active, especially if the new femoral tunnel is through virgin bone, as it allows for improved graft incorporation and healing compared to a scarred and/or enlarged tunnel [36]. Allografts are avoided and considered only when either autograft is not an option or the patient insists on allograft and is willing to accept the higher failure rate [35]. We present a simplified version of our decision‐making process (Figure 2).

Figure 2.

Simplified binary decision tree depicting our decision‐making regarding graft selection in ssRACLR. ‘Potential Concerns’ address factors mentioned in the text, such as tunnels, availability, quality, patient preference and so on. ACL, anterior cruciate ligament; BTB, bone‐patellar tendon‐bone; ssRACLR, single‐stage revision ACL reconstruction. Created in BioRender.com.

Consideration of lateral extra‐articular tenodesis (LET) to improve rotational stability and decrease failure rate is also important [82]. Patients who are at increased risk of graft rupture have the most benefit from LET [79]. Thus, a low threshold should be maintained for high‐level athletes whose goal is to return to the same level after RACLR [86]. Regardless of the level of competition, we consider the goal of returning to a cutting and pivoting sport after RACLR an indication for the surgery [81]. Patients who have had multiple ipsilateral ACL failures without a clear cause (e.g., high tibial slope, increased depth of lateral femoral condyle) can also benefit from LET [22, 34]. A case has also been made for the utility of LET in patients undergoing RACLR with soft tissue grafts [79]. Apart from restoring rotational stability [27], LET has been described to reduce average graft load by as much as 43% [23], thus ‘protecting’ the graft during incorporation and revascularization [19]. When considering the use of LET in RACLR, we prefer the combination of an autograft with bone plug for RACLR and the modified Lemaire technique [30] for LET, as it has shown favourable results in all outcomes [11]. As well, an alternative that has been described as having similar efficacy is the use of an HT autograft combined with anterolateral ligament (ALL) reconstruction [89]. PLC instability should be excluded before proceeding with LET to avoid fixating the tibia in a posterolateral subluxated position [79, 101].

Handling tunnels and hardware from prior surgery

Well‐placed tunnels should be redrilled in the revision procedure before or after hardware removal, if necessary. Preoperative evaluation is important for planning proper tunnel placement for revision reconstruction. Wide tunnels are an issue due to the risk of compromised fixation and healing. Dimensions and tunnel location based on pre‐operative imaging can guide options such as divergent tunnel placement, screw stacking, secondary fixation or a two‐staged procedure if single‐stage surgery is not appropriate [8].

Poorly placed femoral tunnels present challenges and are a frequent cause of ACLR failure [63]. Tunnels are typically too far anterior and/or vertical, most commonly when the index procedure is performed with the transtibial approach [2]. We prefer to create a new anatomic tunnel, completely avoiding the previous tunnel and fixation device, if possible. Prior fixation devices are removed only if they are in the way, and they are left in place if not. If a prior femoral tunnel is slightly off the anatomic position, we usually opt for a ‘blended’ tunnel aperture, ideally with a bone plug on the graft. We prefer a bone plug on the femoral side for better fixation and bone‐to‐bone healing.

For malpositioned tibial tunnels, our philosophy is similar. A new divergent tunnel can be created if it does not interfere with the desired placement, with or without hardware removal. We will sometimes blend the tunnel to arrive at an anatomic aperture. Prior hardware can also be partially removed and/or reamed, depending on the situation (Figure 3). Decision‐making is more complicated when the new tunnel is adjacent to keep the aperture in the anatomic position and to avoid a ‘figure‐of‐eight’ defect [69]. Adequate fixation is challenging in cases where the aperture is wider or when the bone between divergent tunnels is not adequate to support interference screws (IS) [108]. Allograft bone grafts can be used to fill a defect at the time of ssRACLR with varying bone sizes depending on the tunnel defect, although we very rarely use this option [78, 106]. The anatomy of the tibia also allows access to the tibial ACL footprint from different angles, and if necessary, an anterolateral tibial tunnel can be drilled to avoid a two‐stage revision. This has been described as a reliable method with comparable tunnel quality, functional outcomes and complication rates to the standard technique [52]. In these cases, the tibial tunnel must be created adjacent to the tibial tubercle and a small amount of anterior compartment muscle must be elevated. Also, the graft must make a sharp turn in the knee, and we use a relatively shorter bone plug to facilitate graft passage. When using this technique, we make sure we have bone grafted the prior medial tunnel at the time of revision to restore tibial osseous integrity.

Figure 3.

Arthroscopic view of the tibial tunnel aperture during right knee ssRACLR. (a) The tip of the plastic interference screw (red arrow) can be visualized approximately 5 mm above the tibial tunnel aperture inside the joint space. (b) The plastic interference screw remnant can be observed at the primary ACLR tibial aperture after the protruding intra‐articular part was removed with the shaver. The remaining screw tip can also be observed (red arrow). (c) The new tibial aperture drilled for the revision graft can be visualized. Part of the screw was drilled through to reach an anatomic tibial aperture, which was just posterior to the previous one. The remnant of the plastic interference screw can be observed at the anterior border of the tunnel (red arrow). ACLR, anterior cruciate ligament reconstruction; ssRACLR, single‐stage revision ACL reconstruction. Created in BioRender.com.

Hardware removal should ideally be performed only if anatomic tunnel creation and graft fixation cannot be accomplished without doing so. However, optimal tunnel position should not be compromised to retain hardware. In cases where hardware removal is problematic, particularly in the femur, the divergent tunnel concept and alternative techniques like over‐the‐top could potentially yield similar outcomes [48, 114]. However, we have not used this technique in skeletally mature patients since it is non‐anatomic. Finally, the availability of implant‐specific removal instrumentation should be arranged preoperatively. We have an ACL hardware removal set that contains a wide variety of screwdrivers, awls, osteotomes and coring reamers.

SURGICAL TECHNIQUE

A combination of neuraxial anaesthesia and peripheral nerve block is typically preferred. After induction, the patient is placed supine with a Surgical Knee Holder to allow for knee hyperflexion. The leg is prepped and draped, with the contralateral leg prepped if needed for harvesting. A pivot shift test is performed preoperatively under anaesthesia.

Graft harvesting and preparation are as follows. For BTB, a longitudinal incision on the patellar tendon's lateral border is made to harvest the central third of the patellar tendon with bone plugs from the proximal tibia and distal patella sized 10 × 18 mm² and 10 × 20 mm², respectively. We insert the tibial graft plug in the femur because the tendon insertion allows more space on the graft for intra‐articular screw insertion. We use an 18 mm long plug to insert into the femur for ease of graft passage. Longer bone plugs are more challenging to pass through the knee and into the femur. Each bone plug is fitted with #5 Ethibond (©Ethicon US) sutures through drill holes. For QT‐B, a longitudinal incision over the quadriceps tendon is made to harvest typically 1 cm wide of tendon and a total graft length of 8 cm, with bone plug sizes either 9 or 10 mm. After the final preparation of grafts with bone plug, the remaining bone is saved in a sterile container and is used to graft the patellar bone defect at the end of the procedure. For HS, a longitudinal incision over the pes anserinus tendon insertion is made to harvest the semitendinosus and gracilis tendons, which are folded into a six‐strand graft using an adjustable loop with a button.

Diagnostic arthroscopy is performed next with standard anteromedial and anterolateral portals to examine the suprapatellar pouch, medial and lateral gutters, PFJ, trochlea and medial and lateral compartments for cartilage defects, synovitis, loose bodies and meniscal lesions, which are addressed as appropriate. The intercondylar notch is then visualized, the previous graft is assessed and resected and intra‐articular apertures and tunnels are evaluated to confirm the preoperative findings and plan. We do not perform a notchplasty unless there is overgrowth of femoral notch bone that precludes anatomic graft placement due to impingement.

If necessary, hardware removal follows. This process involves the excision of soft tissue and bone that obstruct device access, using curettes, shavers and osteotomes. Screws are removed with manufacturer‐specific screwdrivers, while plastic screws can be drilled through, if needed (Figure 3). All debris from removal or drilling should be meticulously removed. Tibial tunnel drilling is performed with the knee flexed to 90 degrees. For the femoral tunnel, the accessory anteromedial inferior portal technique is used to locate the desired aperture position and drill a new or ‘blended’ tunnel, or re‐ream the old tunnel, usually to 10 mm. We create the femoral tunnel in deeper knee flexion, typically approximately 120°, to avoid the femoral posterior cortex and to leave a solid posterior back wall for the tunnel. Tibial tunnels are created in a similar fashion based on planning and evaluation, either through the previous incision or a new one, also usually to 10 mm. We typically re‐use the same start point on the tibia, but depending on the tibial cortical integrity, we may select a new start point on the tibia and either blend the intra‐articular aperture with the previous one or make a new tunnel entirely. Sometimes, holes left from hardware removal can be used as the new femoral or tibial tunnel, or part of the new tunnel. In cases where we create an anterolateral tibial tunnel, we do so by removing the most proximal and medial aspect of the anterior compartment musculature with electrocautery and periosteal elevator, and by drilling our tunnel just lateral to the tibial tuberosity. We meticulously remove old graft tissue from inside the tunnels and smooth the tunnels with a shaver to ensure easy graft passage.

The graft is then passed and fixed on the femoral side. If LET is planned, we defer final ACL fixation on the tibia until the iliotibial band (ITB) graft is in place and then fix the ACL definitively before final fixation of the LET. We cycle the knee with tension on the graft and then perform tibial fixation of the ACL graft in full extension, but never in hyperextension.

Fixation options

Fixation method considerations include tunnel characteristics and graft choice. For soft tissue grafts, cortical‐suspensory fixation on the femur is an excellent option [45]. Also, if the femoral aperture is compromised at all, we consider tying over a button when using a graft with a bone plug, which is facilitated by the open approach required for lateral tenodesis. On the tibia, we generally prefer an interference screw made of metal or polyetheretherketone (PEEK). We prefer using backup fixation with a suture anchor for soft tissue grafts on the tibia to prevent potential graft slippage through the softer tibial bone [75, 93, 105]. Fixation approach for grafts with a bone plug varies based on tunnel quality. We prefer metal interference screws in general due to fixation, cost, visibility on X‐ray and lack of bone lysis and synovitis associated with resorbable screws [110]. We use backup fixation with a button if necessary [9, 37]. Patients who have had staple fixation on the tibia at prior surgery could have larger residual bone defects if the staple must be removed.

Lateral extra‐articular tenodesis

We use the modified Lemaire technique [30] when performing LET. A lateral longitudinal incision is made to harvest a central strip of the ITB, 1 cm wide and 8 cm long, with the distal attachment being left intact. A locked non‐absorbable suture is placed in it and the ITB strip is then tunnelled under the lateral collateral ligament (LCL) adjacent to its origin at the lateral epicondyle. A guide pin is placed just anterior and proximal to the epicondyle. Then, the graft is confirmed to be isometric from 0‐30 degrees, and looser in flexion greater than 30 degrees where the ACL is not important for AP knee stability. A socket is reamed to 6 mm, and the ITB graft is pulled into the socket and fixed with an interference screw after definitive ACL fixation. A demonstration of our technique can be found in the video (Supporting Information S1: Online Resource 1). Final testing for fixation is done before copious irrigation and closure. All other ligamentous injuries should be addressed in the revision setting.

DISCUSSION

Good outcomes have been associated with ssRACLR. Re‐revision rates ranged from 2.2%–2.79% among cohorts, with overall failure reported at 2.1%–4.4% [83, 107]. The total knee arthroplasty (TKA) rate was 1.1% in a study with a mean postoperative follow‐up of 4.3 years [107]. A report on 409 patients undergoing ssRACLR with minimum 2 years of follow‐up showed a significant reduction in side‐to‐side AP laxity and improvement in patient‐reported outcome (PTOs) scores. No difference in outcomes was observed between patients with maximum AP tunnel diameter of <12 mm versus >12 mm [87]. ‘Blended’/redrilled femoral tunnels showed no difference compared to new tunnels in a 93‐patient study with mean follow‐up of 3.9 years. All tunnels were planned based on standard evidence‐based recommendations [3]. Regarding RTS, a study involving 40 elite athletes had 35 (87.5%) RTS: at mean of 4.4 years postoperatively, 29 (72.5%) were playing again professionally, but only 15 (37.5%) had returned to the same level [107].

Higher rates of same‐level RTS are reported when RACLR is combined with LET. A study on 17 patients undergoing RACLR with BTB and LET via a MacIntosh‐modified Arnold‐Coker Technique [4] reports a 58.8% RTS to the same level, with only one failure at mean follow‐up of 2.5 years. However, no information on competition level was provided [111]. In 19 patients with preoperative Tegner Activity Scale ≥ 6 who underwent ssRACLR with BTB and LET via the Lemaire technique, all patients achieved RTS with 52.6% achieving same‐level RTS with no re‐tears at minimum 2‐year follow‐up [11]. Finally, a systematic review of eight studies comparing ssRACLR with LET to ssRACLR alone found lower failure rates, better side‐to‐side AP laxity difference, less high‐grade pivot shift and no difference in RTS and PROs [90].

Available literature has limited evidence on which autograft option has more favourable outcomes in RACLR. Allograft tissue has been found to have higher failure rates and we avoid using allograft unless other options are precluded, or the patient has a strong preference and is willing to accept the risk. Setliff et al. [95] compared patients undergoing RACLR with QT‐B after primary BTB or BTB after primary QT‐B to those with HS after primary BTB or QT‐B. All grafts were ipsilateral to the ACL injuries. They found no statistically significant differences in RTS, same‐level RTS and PROs at one‐year follow‐up. However, they included only 15 matched pairs, and when looking at absolute differences, RTS and same‐level RTS were 31% and 11% higher in extensor‐to‐extensor patients, respectively. Consequently, lack of statistical significance could very well be attributed to lack of power. A recent meta‐analysis comparing RACLR with QT, HS and BTB found no statistically significant difference in all outcomes. Yet, most of the forest plots suggested the superiority of QT over HS, with moderate heterogeneity at worst, but broad confidence intervals in individual studies [5]. These can be attributed mostly to the small sample size, but other factors such as differences in collecting and reporting PROs and lack of standardized ACL failure definition among studies [6, 21] could have contributed to the high variability observed [5]. Confounding factors, for instance, primary ACLR graft choice, are also not consistently reported throughout the studies [6, 21, 40, 99]. Finally, as radiologic advancements and machine learning applications increase, studies should be conducted to develop surgeon‐friendly techniques for better tunnel analysis and preoperative planning to further reduce unnecessary two‐stage procedures and improve patient outcomes.

This study, even though comprehensive, does not come without limitations inherent to its nature. Our literature review has potential selection bias in the included studies. Additionally, this review is limited by the level of evidence of the included studies. Some of the opinions expressed in this article are based on our experience, which may allow room for a more flexible and perhaps ‘real‐world’ contextualization but might also reduce the objectivity of some of the findings.

CONCLUSION

In our experience, ssRACLR can be carried out for over 90% of revision cases with creative pre‐operative planning using autograft with bone plug(s), divergent tunnel creation on the femur (when necessary) and convergent tunnel creation on the tibia (when appropriate) and suspensory or interference fixation as needed. We believe autografts with bone plugs provide the best opportunity for graft healing and incorporation in what is often a suboptimal situation due to enlarged tunnels with graft scarred in them from the prior ACL reconstruction. Finally, LET has been a useful adjunct for ssRACLR to reduce re‐tear rates.

AUTHOR CONTRIBUTIONS

The idea for this article was primarily conceived by Robert G. Marx. The literature search was mainly performed by Periklis Giannakis, Sophia T. Zhuang and Jacob L. Rosenstadt, with additional suggestions by Robert G. Marx. Figures were created using BioRender.com by Periklis Giannakis and Sophia T. Zhuang, under Periklis Giannakis's license, and edits were proposed by Robert G. Marx. Surgical technique showcased in ‘Supporting Information S1: Online Resource 1’ is performed by Robert G. Marx. Manuscript writing, including original and subsequent draft preparation, was carried out by Periklis Giannakis, Sophia T. Zhuang and Jacob L. Rosenstadt. Supervision, major and minor revision suggestions before submission and overall project administration were conducted by Robert G. Marx. All authors approved the final version to be published and agreed to be accountable for all aspects of the work.

CONFLICT OF INTEREST STATEMENT

Robert G. Marx reports relationships including board membership with the International Society of Arthroscopy, Knee Surgery, and Orthopaedic Sports Medicine, equity and stock compensation from Mend Nutrition Inc. for a science advisory board seat, and royalties from books published with Springer and Demos Health while declaring that none of these could have influenced the work reported in this paper. The remaining authors declare no conflicts of interest.

ETHICS STATEMENT

This is a review study, where no subjects were involved in any part of the process. Thus, obtaining an ethics approval statement was not necessary. This study includes images such as x‐rays and arthroscopic images, as well as video capture of open surgery. No information that could potentially lead to the identification of subjects is included in the aforementioned multimedia files, and as a result, informed consent was waived.

Supporting information

Online Resource 1 Video demonstration of the modified Lemaire technique[30] for lateral extra‐articular tenodesis (LET). A lateral longitudinal incision is made to harvest a central strip of the iliotibial band (ITB), 1 cm wide and 8 cm long, with the distal attachment being left intact. A locked non‐absorbable suture is placed in it and the ITB strip is then tunneled under the lateral collateral ligament (LCL) adjacent to its origin at the lateral epicondyle. A guide pin is placed just anterior and proximal to the epicondyle. Then, the graft is confirmed to be isometric from 0‐30 degrees, and looser in flexion greater than 30 degrees where the ACL is not important for AP knee stability. A socket is reamed to 6 mm, and the ITB graft is pulled into the socket and fixed with an interference screw after definitive anterior cruciate ligament (ACL) fixation.

Giannakis, P. , Zhuang, S.T. , Rosenstadt, J.L. & Marx, R.G. (2025) One‐stage revision anterior cruciate ligament reconstruction: preoperative evaluation, planning and surgical techniques. A review of current concepts. Journal of Experimental Orthopaedics, 12, e70111. 10.1002/jeo2.70111

DATA AVAILABILITY STATEMENT

No data set was generated or analyzed for the conduction of this study.