Introduction
Multiligament knee injuries (MLKIs) are a challenging clinical entity because of the increased number of ligaments involved, associated injuries, diagnosis, decision making about the optimal treatment and rehabilitation. These injuries are not as rare as previously reported; therefore a high level of suspicion and a thorough and systematic evaluation are mandatory to avoid missed or late diagnosis. Furthermore, some injuries may be misdiagnosed because about 50% of knee dislocations are reported to reduce before presentation. Historically, multiligament knee injuries were reported in association with knee dislocations ( Figs. 11.1 and 11.2 ); however, improved understanding of these injuries and better diagnostic tools have shown that not all MLKIs are caused by knee dislocations. True knee dislocations are rare and are associated with a higher risk of neurovascular injuries, , which can further complicate the injury and affect the prognosis. , It is important to diagnose and treat all injured structures because failure to address all injured structures can compromise knee function, leading to ligament reconstruction failure and poor outcomes.
Epidemiology and Injury Patterns
Previous studies reported that multiligament knee injuries caused by knee dislocations accounted for about 0.02% to 0.2% of all orthopaedic injuries. In a study based on registry data, Arom et al. estimated that the incidence of MLKI may be as high as 0.072 per 100 patient years.
Historically, knee dislocations have been associated with high-energy trauma such as motor vehicle accidents, falls from heights and farm or industry injuries. In the 19th and early 20th centuries the mechanisms of injury reported in the literature usually included more dramatic mechanisms such as a cart and horse falling on its owner, a man on horseback whose leg was pinned between a rail and the horse from which he was being thrown and, less often, some less dramatic injury mechanisms such as falls. With the advent of motorised vehicles, industrialisation and sports participation, the mechanisms of injury have changed.
Some studies have reported that 44% to 47% of MLKIs in their series were caused by sporting injuries. , With increasing sports participation, it is important to be aware of the possibility of complex knee injuries, despite an apparent low-energy mechanism, because these injuries can have devastating consequences if not properly diagnosed and managed. Furthermore, with the growing global burden of obesity, it is important to be aware of ultra-low-velocity knee dislocations and MLKI in obese individuals. Morbidly obese patients who sustain ultra-low-velocity MLKI have a high risk of neurovascular injuries and postoperative complications. The injury mechanism will differ depending on the clinical setting (trauma centre versus elective surgery centre), location (city, populations, country) and activities that are common in that area.
The medial side is the most commonly injured side of the knee, and up to 78% of high-grade (grade III) MCL injuries have an associated lesion, especially cruciate ligament and meniscus injuries. , Geeslin et al. reported that only 28% of posterolateral corner (PLC) knee injuries were isolated injuries, suggesting that most PLC injuries are combined ligament injuries or MLKI. Moatshe et al. reported that medial and lateral side injuries in knee dislocations (KDs) accounted for 52% and 28%, respectively, whereas Becker et al. reported a higher incidence of lateral injuries (knee dislocation grade III lateral (KDIII-L)), accounting for 43% of knee dislocations in a level 1 trauma centre.
Concomitant meniscal and cartilage injuries are common in multiligament injuries. Moatshe et al. reported that the incidence of meniscal injury in conjunction with multiligament knee injury was 37.3%, whereas the incidence of concomitant cartilage injury was 28.3%. Krych et al. reviewed a series of 122 knees and found that meniscal or chondral injury was present in 76% of cases. Richter et al. found that meniscal injury was only present in 15% of cases of multiligament knee injury, suggesting a lower prevalence. In a study on 194 sports-related MLKIs, LaPrade et al. found that 30% of the patients had chondral injuries and 55% had meniscal injuries. There has been increasing focus on meniscal root injuries; however, data on meniscal root injuries in the setting of MLKI are still lacking.
Neurovascular Injury
The rate of popliteal artery injury in knee dislocations has been reported to range between 7% and 48%, , , and as high as 64% in cases of concomitant fractures, depending on the individual hospital reporting. Medina et al. reported that the incidence of neural and vascular injury in patients who sustained knee dislocation was 25% and 18%, respectively. Becker et al. reported a similar incidence of neural and vascular injury at 25% and 21%, respectively, in a series of 106 patients. Moatshe et al. evaluated 303 patients with knee dislocations and reported that vascular injuries occurred in 5% of the cases, whereas common peroneal nerve injury occurred in 19%. Interestingly, Moatshe et al. also found that patients who had a PLC injury had 42 times greater odds of experiencing a peroneal nerve injury and 9.2 times greater odds of experiencing a popliteal artery injury than patients without a PLC injury. Furthermore, patients with peroneal nerve injury had 20 times greater odds of experiencing concomitant vascular injury. Together these studies highlight the importance of evaluating for simultaneous neurovascular involvement when evaluating patients with MLKI, and the surgeon should not hesitate to pursue additional angiographic imaging in patients who suffer associated peroneal nerve injury. Level 1 trauma centres, where the majority of patients with knee dislocations have high-energy trauma, usually report a higher prevalence of concomitant vascular injuries compared with elective clinics treating low-energy injuries. The risk of vascular injury is correlated with the degree of energy, concomitant fractures and type of dislocation. Timely diagnosis of vascular injuries is imperative to minimise the risk of amputation as a result of limb ischemia.
Classification
The most widely used classification was described by Schenck in 1994 and later modified by Wascher. This classification system is based on the anatomical pattern of ligament injury and reports associated injuries, including neurovascular injuries and fractures ( Table 11.1 ).
| KDI | Injury to single cruciate plus collaterals |
| KDII | Injury to ACL and PCL with intact collaterals |
| KDIII-M | Injury to ACL, PCL, MCL |
| KDIII-L | Injury to ACL, PCL, LCL |
| KDIV | Injury to ACL, PCL, MCL, LCL |
| KDV | Dislocation plus fracture |
Diagnosis
History
A thorough history is mandatory to understand the injury mechanism and the energy involved. A video demonstrating the injury may also help elucidate the mechanism. High-energy injuries are usually associated with other organ injuries, including fractures and head, thorax and abdominal injuries. All patients with high-energy trauma mechanism should be evaluated for a potential knee injury, and patients with an MLKI should be evaluated for concomitant injuries, including fractures and neurovascular injuries. Low-energy injury mechanism usually results in an injury confined to the knee.
Physical Examination
Physical examination should be thorough and systematic to avoid missing some injuries. An examination focussing on swelling, alignment, passive and active range of motion and knee stability should be performed when possible; however, this may be limited by pain in the acute phase. Neurovascular evaluation should be performed in all patients with suspected MLKI because missed vascular injuries can have catastrophic consequences.
Vascular assessment should include palpation of the posterior tibial and dorsalis pedis arterial pulses bilaterally in addition to assessing potential differences in skin colour and capillary filling. Pulse symmetry is important because the presence of pulses does not exclude vascular injury. Serial assessment is recommended because the vascular changes may not be immediately apparent on first examination. Stannard et al. found a strong correlation between the results of a serial physical examination and a need for an arteriography in a series of 126 patients with knee dislocations and concluded that serial physical examination was a safe and prudent policy in evaluating patients with knee dislocations. Furthermore, Stannard et al. reported a positive predictive value (PPV) of 90% for the clinical examination, negative predictive value (NPV) of 100%, sensitivity of 100% and specificity of 99%, which were comparable to those reported by Hollis et al. who reported a sensitivity and specificity of 100% for the physical examination.
Ankle–brachial index (ABI) is an important adjuvant, especially when physical examination is equivocal or there is concomitant neurological injury. An ABI less than 0.9 warrants angiography. , Posterolateral corner injuries and peroneal nerve injuries in the setting of knee dislocations have been reported to be associated with a high risk of vascular injuries, and patients with such injuries should be evaluated closely. Wascher et al. reported that patients with bicruciate injury had an equivalent risk of neurovascular injuries as patients with frank knee dislocations. In cases where serial examinations are not possible, or where the utility of measuring ABI may be limited such as in ipsilateral limb fracture or hypovolemic shock, computed tomographic (CT) angiography is recommended.
Imaging
Diagnostic imaging also plays an essential role in the evaluation of a multiligament injured knee because physical examination alone has been demonstrated to be less reliable in the diagnosis of knee ligament injuries. Conventional radiographs are the first imaging modalities performed to exclude fractures. The advent of magnetic resonance imaging (MRI) has led to improved diagnosis of soft tissue injuries about the knee. Ligament injuries; meniscal tears, including meniscal root tears; and chondral lesions are more accurately diagnosed with the use of MRI. MRI is therefore routinely used in the diagnosis of MLKI ( Fig. 11.3 ). Stress radiographs play an important role in the diagnosis and classification of ligamentous injury, particularly in PCL, posteromedial corner (PMC) and PLC injuries. Obtaining stress radiographs may be difficult in the acute phase because of pain and patient guarding; , , therefore using a mini C-arm with examination under anaesthesia during surgery may be helpful.
For the evaluation of posterior knee stability, posterior stress radiographs are used ( Fig. 11.4 ). The PCL has an intrinsic ability to heal and may regain continuity of its fibres after an injury; , however, if not treated, the PCL may heal in an elongated position, , resulting in persistent posterior subluxation of the tibia and instability during loading. In the setting of chronic PCL tears and PCL reconstruction graft tears, MRI may demonstrate continuity of the PCL fibres; however, the functionality of this tissue is difficult to assess. Posterior stress radiographs have been reported to provide a more objective method to assess the structural integrity of the PCL. , They have also been demonstrated to have good interobserver and intraobserver reliability in the diagnosis of PCL tears. , A posterior translation side-to-side difference of 0 to 7 mm is usually found in patients with partial PCL tears and in patients who are too sore to put sufficient weight on the knee; an 8 to 11 mm side-to-side difference is associated with a complete isolated PCL tear; 12 mm or greater is usually observed in patients with a complete PCL tear with an additional ligament injury, usually the PLC or PMC, but can also be seen in patients with decreased sagittal plane tibial slope.
To assess the integrity of the medial structures, bilateral valgus stress radiographs are performed with the knees at 20 to 30 degrees of flexion. Medial gapping is assessed by measuring the shortest distance between the subchondral bone surface of the most distal aspect of the medial femoral condyle and the corresponding tibial plateau. Overall, a 3.2 mm side-to-side difference of gapping represents a complete superficial medial collateral ligament tear, and a 9.8 mm difference represents injury to all medial structures.
To evaluate the integrity of the PLC structures, bilateral varus stress radiographs are performed with the knees flexed to 20 to 30 degrees. Lateral gapping is assessed by measuring the shortest distance between the subchondral bone surface of the most distal aspect of the lateral femoral condyle and the corresponding tibial plateau. A side-to-side lateral gapping difference of 2.7 to 4.0 mm in varus stress radiographs represents a complete fibular collateral ligament (FCL) tear, whereas a side-to-side difference greater than 4 mm represents a complete PLC injury ( Fig. 11.5 ).
Treatment
The treatment of MLKI has evolved over the years. In the 19th century closed reduction was commonly used, and amputation was considered in patients with irreducible knee dislocations and vascular injuries. In the late 19th century open reduction of irreducible knee dislocations was described as an option that could potentially salvage the injured limb. Immobilisation after reduction was commonly used and offered improvement in function despite risk of joint stiffness. After the introduction of aseptic surgical techniques, surgical treatment of the dislocated knees with repair of the ligaments replaced conservative treatment with immobilisation.
Improved understanding of knee anatomy and biomechanics, coupled with modern technology for both diagnosis and surgical treatment, has brought other options in the treatment of these injuries. However, there is still controversy on the treatment of MLKI. Some areas of controversy include repair versus reconstruction, augmenting repair and reconstruction with internal bracing, single-stage versus staged reconstruction, anatomical reconstructions versus nonanatomical reconstructions, active rehabilitation versus conservative protocols and standardised patient outcome measures. In addition, there are several surgical aspects that are challenging in the treatment of these patients, including the risk of tunnel convergence in the limited bone stock of the distal femur and proximal tibia and the effect of the ligament graft tensioning sequence on tibiofemoral orientation and knee kinematics. These are important factors because they can affect the outcome of surgery and the longevity of the knee joint.
Operative Treatment
Some studies have demonstrated that surgical treatment of MLKI yields superior outcomes compared with nonoperative treatment; , , therefore operative treatment is considered the standard of care. Nonoperative treatment can be considered in patients with other life-threatening injuries or where surgery is contraindicated because of comorbidities.
Ligament repair
Mayo Robson in 1895 was the first surgeon credited to have performed both an anterior crucial ligament (ACL) and PCL repair, in a 41-year-old miner who was treated 9 months after the index injury. Repair of knee ligamentous injuries was also performed in the first half of the 20th century, as it was reported by Stellhorn in 1934. In 1955 O’Donoghue reported good outcomes after operative repair of 80 knee ligament injuries, and early repair within 2 weeks was associated with better outcomes. After the report by O’Donoghue, more studies on the repair of knee ligaments were published from the 1960s to the 1980s, reporting improved outcomes superior to nonoperative treatment. , ,
Mariani et al. evaluated outcomes in a cohort of patients with multiligament injuries, 52 patients treated with repair of the ligaments versus 28 treated with reconstructions. Patients with repair of cruciate ligaments had higher rates of flexion deficit, higher rates of posterior instability (with a 100% recurrence of a posterior sag sign) and lower rates of return to preinjury activity levels. Other reports in the literature have demonstrated that repair of PLC injuries is associated with higher failure rates and reoperation rates compared with ligament reconstructions. , Stannard et al. performed a prospective study with 57 knees; failure rates were 37% and 9% in the repair and the reconstruction groups, respectively, at a minimum 24 months of follow-up. Levy et al. reported a significantly higher rate of failure for repair (40%) compared with reconstruction of the FCL/PLC (6%). However, there was no significant difference in the International Knee Documentation Committee (IKDC) subjective scores and Lysholm scores between the repair group and the reconstruction group. King et al. compared clinical and functional outcomes of surgically treated medial and lateral knee dislocations in a retrospective study that included 56 patients (24 with the KDIII medial (KDIII-M) injury pattern (43%) and 32 with the KDIII-L injury pattern (57%)), with a mean follow-up of 6.5 years. Patients who underwent medial repairs had significantly worse outcomes for both Lysholm and IKDC scores. These reported high failure and reoperation rates have led some authors to advocate for reconstruction of all torn knee ligaments in the setting of an MLKI. Repair of the collaterals is usually reserved for bony avulsions that are large enough to be fixed with hardware or suture anchors.
There have been reports of good functional outcomes after open ligament repair. Furthermore, the advent of internal bracing has led to resurgence of ligament repair; however, long-term outcomes and complications after ligament repair with internal bracing are still lacking.
Ligament reconstructions
Early reports on knee ligament reconstruction were on open surgical techniques. Hesse is reported to have introduced knee ligament reconstruction in a brief report published in 1914, where he described a technique of reconstructing the ACL and PCL using free grafts of fascia lata pulled through tunnels. Later, Dr Ernest William Hey Groves, who was one of the eminent orthopaedic surgeons of his era, described a technique of reconstructing the ACL using the iliotibial band and a PCL reconstruction using a semitendinosus tendon. Despite these early reports, cruciate ligament reconstruction only became popular in the latter half of the 20th century. From the late 1980s and 1990s, reconstruction of torn ligaments was increasingly performed and accepted as the preferred method.
The arthroscopic techniques used in isolated ACL and PCL reconstruction became available for multiligament-injured knees and have become the standard of care in most places. , , , Fanelli et al. found improved patient-reported outcomes after arthroscopic-assisted PCL and ACL reconstructions and open reconstruction of the PLC. Several studies have reported improved outcomes after ligament reconstruction of MLKI. , , It is worth noting that most of the techniques in the 1990s were not anatomical, despite reported improved subjective patient outcomes. The PLC was reconstructed using different techniques, including biceps femoris tendon tenodesis, split biceps tendon tenodesis, semitendinosus free graft or allograft tissue in primarily nonanatomical techniques.
Most surgical techniques involve the use of autografts, including bone–patella tendon–bone (BTB), hamstring (semitendinosus and gracilis) and quadriceps (with or without bone block) tendons. Techniques using allografts have been reported for the reconstruction of knee ligaments. The use of allografts provides more graft options, especially when three of four ligaments are torn and in revision cases; however, allografts are expensive and not readily available in all hospitals. Where allografts are not available, harvesting allografts from the contralateral side is a viable option; however, donor site morbidity is a concern.
Anatomical reconstructions
Since the 2010s there has been an increasing focus on anatomical reconstructions of all torn knee ligaments. Studies on the anatomy and biomechanics of knee ligaments have led to a better understanding of the surgically pertinent anatomy of knee ligaments and the development of biomechanically validated anatomical reconstruction techniques. , , Anatomical reconstruction of the injured structures using biomechanically validated techniques restores knee kinematics to near normal and yields improved patient outcomes. , , Therefore, in the setting of multiligament injuries, reconstruction of all the torn ligaments is recommended.
Timing of surgery
Timing of surgery during MLKI is a topic of debate, and there is still no consensus on the point of demarcation between acute and chronic. Some authors have used 3 weeks as the critical time to better identify and treat the structures before scar tissue forms, making dissection and identification of the structures difficult, and tissue necrosis affects outcomes. , , , However, some authors have used a 6-week timeline to demarcate acute and chronic injuries.
Studies have reported superior outcomes in acutely treated patients compared with treatment of patients with chronic injury. , Even though some surgeons are concerned about the risk of joint stiffness in acutely treated injuries, Levy et al. reported no difference in range of motion after surgery in patients with acute and chronic injuries in a systematic review of the literature that included five studies. Some authors have reported better subjective outcomes after staged surgery compared with acute single-stage surgery. , However, staging the reconstruction can potentially alter joint kinematics and increase the risk of graft failure.
Clinical studies have demonstrated chronic instability caused by a deficient ACL to be a risk factor for meniscal tears and chondral degeneration secondary to altered changed knee loading and biomechanics, and this may also be true for chronic instability caused by MLKI. In addition, chronic instability puts more load on the intact structures and can potentially lead to general knee joint laxity and poor outcomes after surgery in the chronic phase.
Furthermore, staging reconstruction can delay rehabilitation. This has led to some authors advocating for concurrent reconstruction of the ACL and PCL in a single setting. LaPrade et al. , demonstrated in biomechanical studies that in knees with deficient PLC the graft forces on both the ACL and PCL were significantly increased when loading the knee. Therefore, if the PLC is not reconstructed or fails, the graft forces on the cruciate ligament grafts could eventually lead to graft failure. Furthermore, knees with a deficient PLC have been demonstrated to result in significantly increased forces on the medial compartment, which can lead to medial joint overload and early osteoarthritis. , Early surgery should be coupled with early functional rehabilitation to minimise the risk of joint stiffness. LaPrade et al. reported approximately a 9% rate of joint stiffness after surgical treatment of MLKI in 194 sports-related injuries.
In high-energy trauma, surgery may be delayed because of injuries to the soft tissue around the knee and concomitant injuries to other vital organs. Furthermore, prolonged operative time because of the treatment of concomitant injuries in a multitraumatised patient with multiligament injuries is an argument against single-stage surgery. However, stiffness in these patients may be easier to treat than recurrent instability. Therefore it is important when treating patients with MLKI to adapt the treatment to the patient, conditions and setup of the hospital. However, single-stage surgery should be the goal whenever possible.
Intraoperative Challenges
Tunnel convergence
During multiple knee ligament reconstruction surgery, several tunnels and sockets are usually created for the reconstruction grafts, and there is an increased risk of tunnel collision because of limited bone stock of the distal femur and proximal tibia. Tunnel length and diameter and the number of ligaments reconstructed are correlated with the risk of tunnel convergence; however, socket length and diameter should not be compromised because this can increase the risk of graft failure. Tunnel convergence increases the risk of reconstruction graft failure because of the potential damage to reconstruction grafts and fixation devices and risk of not having sufficient bone stock between the grafts for fixation and graft incorporation. Drilling all of the guide pins that have a risk of converging before reaming the sockets can give an opportunity to adjust the pins and avoid convergence. After reaming the sockets for cruciate ligaments in the femur, an arthroscope can be used to visualise the collateral ligament guide pins and ensure that they are clear of the cruciate ligament sockets. The use of intraoperative fluoroscopy may be helpful.
Moatshe et al. reported a 67% tunnel convergence rate between the posterior oblique ligament (POL) tunnel and the PCL tunnel in the tibia when the POL tunnel was aimed at Gerdy’s tubercle. They recommended that the POL tunnels be aimed to a point 15 mm medial to Gerdy’s tubercle to reduce risk of convergence with the PCL and that the superficial medial collateral ligament (sMCL) tunnel be aimed 30 degrees distally in the tibia to avoid convergence with the PCL tunnel.
On the lateral femoral side, Moatshe et al. reported a high risk of tunnel convergence between the ACL and FCL tunnels (sockets) if the FCL is drilled at 0 degrees in the axial and coronal planes. A 35- to 40-degree angulation in the axial plane and 0-degree angulation in the coronal plane (35 to 40 degrees anteriorly) was safe and avoided tunnel convergence ( Fig. 11.6 ). Carmada et al. reported a high risk of tunnel convergence between the ACL and the FCL (69% to 75% depending on the length of the tunnel) and recommended aiming the FCL tunnel 0 degrees in the coronal plane and 20 to 40 degrees in the axial plane. On the medial side, aiming the sMCL tunnel 40 degrees in the axial and coronal planes (40 degrees anteriorly and proximally) and the POL tunnel 20 degrees in the axial and coronal planes (20 degrees anteriorly and proximally) was safe to avoid convergence with the double bundle PCL tunnels ( Fig. 11.7 ). Gelber et al. reported that angulations of 30 degrees in the axial plane and coronal plane reduced the risk of convergence with the PCL tunnels; however, the tunnel diameter for the PCL was smaller than those used by Moatshe et al.
