Diagnosis and treatment strategies of the multiligament injured knee: a scoping review

Descriptive analysis

The results of the search and study selection process are outlined in figure 1. Overall, 417 eligible studies relevant to the aims and research question were identified. In keeping with wider bibliometric trends in research of all aspects of MLKI, there was substantial chronological increase in the number of studies relating to diagnosis and treatment of MLKI (figure 2).

Figure 1Figure 1Figure 1

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses, extension for Scoping Reviews flow chart for study inclusion. MLKI, multiligament knee injury; MPFL, medial patellofemoral ligament; PCL, posterior cruciate ligament.

Figure 2Figure 2Figure 2

The number of multiligament knee injury studies published per year.

Research studies were identified from 37 countries, with 57% (239 studies) originating from the USA. A heatmap illustrating the countries with the greatest number of publications is shown in figure 3 (further details in online supplemental table 1).

Figure 3Figure 3Figure 3

Heatmap of countries by number of MLK-related publications. MLK, multiligament knee.

Study design and level of evidence

The studies varied considerably in their design and methodology (figure 4). As scoping reviews are intended to provide a map of what evidence has been produced as opposed to seeking only the best available evidence, no formal quality assessment was performed. Most studies were of lower order evidence, with over 90% being levels III–V. There were four randomised controlled trials (RCTs)—three assessed different aspects of rehabilitation following surgical treatment of MLKI18–20 and one assessed non-operative versus operative treatment of MLKI.21 Of the remaining studies, 31 were prospective, 189 were retrospective, 23 were systematic reviews, 30 were case reports/series and 128 comprised narrative reviews, technical notes and editorials. There were 12 preclinical studies.

Figure 4Figure 4Figure 4

The number of MLKI studies included by level of evidence. MLKI, multiligament knee injury.

The 23 systematic reviews that were included in this scoping review were evaluated to gain a representative insight into the estimated age, ethnicity and gender ratio of included participants in studies reporting on aspects of diagnosis and management of MLKI. Of the 23 included systematic reviews, 8 (35%) reported specifically on the gender of patients included. The overall percentage of included participants that were women in these systematic reviews was 22%, and the mean age of included participants was 31.8 years. No systematic reviews reported on the ethnicity of included participants.

Focus of studies

Of the included studies that presented original clinical data (n=254), 38 focused on aspects surrounding the diagnosis of MLKI (14.9%) (figure 5), including associated soft-tissue knee injuries, neurovascular injuries and the value of specific diagnostic modalities. Thirty-six studies were of level II or III evidence, consisting of 10 prospective and 26 retrospective studies; the remaining 2 were case reports.

Figure 5Figure 5Figure 5

The number of MLKI studies by focus of study. MLKI, multiligament knee injury.

The majority of included studies presenting original data reported on aspects of treatment of MLKI (n=207). These varied from assessing outcomes of surgical repair or reconstruction of MLKI, comparing timing of surgical intervention, non-operative management and predictors of poor outcome following surgical treatment of MLKI. Twenty-one studies (10.1%) were prospective, including one RCT. Seven studies (2.8%) representing original data focused on methods of rehabilitation following treatment of MLKI, and notably 42.9% of these were RCTs (n=3).

A small number of studies assessed MLKI in the context of high-level sport (n=15) or in a military setting (n=2); however, most studies did not specify such context.

We additionally reviewed the top authors by number of publications within the MLKI field, as first or senior author (online supplemental table 2). Most of these authors were from the USA, reflecting the geographical distribution of studies published. All were male, and all but one was white (n=22/23, 96%).

Thematic summaryNomenclature and definitions

MLKI has been traditionally defined as disruption to at least two of the four major knee ligaments comprising the ACL, PCL, MCL and LCL.14 More recently, there has been greater understanding of the significance of associated soft-tissue knee structures and their relative significance in conferring knee stability, leading to studies expanding this definition to include the posterolateral corner (PLC) and posteromedial corner (PMC) of the knee.22 Several recent systematic reviews assessing MLKI define the term specifically as ‘disruption of at least two of the four major knee ligaments, comprising the ACL, PCL, MCL (and PMC) and LCL (and PLC)’.1 4 15 23 Numerous original research studies and systematic reviews do not explicitly define their interpretation of MLKI, or do not define their interpretation of the ‘four primary knee ligaments’.24–30 One recent systematic review defined MLKI as ‘three or more ligaments injured and/or knee dislocation’ .31 Future studies should specifically state their definition of MLKI. We suggest a definition of the traumatic disruption of at least two of the major ligaments of the knee, comprising the ACL, PCL, PMC (superficial and deep MCL, posterior oblique ligament) or PLC (fibular collateral ligament, popliteus tendon, popliteofibular ligament (PFL)). Furthermore, ‘knee dislocation’ has often been used interchangeably with MLKI. This may be because historically, reported multiligament injuries were only observed in the context of knee dislocations (defined as total disruption of the tibiofemoral joint verified clinically or radiographically). In a landmark observational study by Wascher et al, 32 the authors reported that 50% of knee dislocations were reduced at time of presentation. Therefore, the definition of knee dislocation was expanded to include bicruciate knee ligament injuries with or without involvement of the collateral ligaments. Schenck developed a classification system for knee dislocations that was later modified by Wascher to include fracture dislocations.32 33 This classification did not cater for other MLKI that were not caused by knee dislocations. A greater understanding of knee anatomy, biomechanics and modern imaging have improved our understanding of both knee dislocations and MLKI. Importantly, it has become clear that although knee dislocations result in MLKI, not all MLKI are knee dislocations. This has created a lack of nomenclature for MLKI that are not classified as knee dislocations. Recent reports have demonstrated that MLKI not caused by knee dislocations are more common than those caused by knee dislocations.34 35 We suggest the term knee dislocation should be used with caution because it does not specify injured structures and might lead to misdiagnosis of relocated knee dislocations.

According to the Schenck classification, Schenck’s Knee Dislocation I (KD I) represents knee dislocations verified clinically or radiologically with one of the cruciate ligaments intact, which is a rare phenomenon.33 Some published studies have used the KD I classification to include MLKI involving a cruciate ligament and collateral ligament without verified dislocation36 37

Although the definitions may be equivocal or implied, there is a need for clarity to enable robust pooling and comparison of outcome data, and clear systematic searches to be performed. A universal consensus definition of MLKI and a requirement to state the author’s definition of such in any future study assessing MLKI may be warranted.

Clinical diagnosis

The value of clinical examination in the context of MLKI is unequivocal; however, the principles differ based on acute or chronically presenting injury. In acute injuries, MLKIs are often associated with high-energy trauma, and thus assessment using Advanced Trauma Life Support principles is advocated.38 Fifty per cent of knee dislocations reduce spontaneously prior to first assessment by a physician, and given the high correlation between knee dislocations and MLKIs, each MLKI should be treated as a true knee dislocation until proven otherwise.39 In subacute and chronic settings, the literature notes that physical examination of the knee is the initial method of choice for the diagnosis of MLKI, but appropriate examination in these settings relies on the patient’s ability to relax and the clinician’s ability to detect endpoints.40

In the subacute or chronic setting, patients with MLKIs may present with ligament injuries that initially go unrecognised, due to the extent of simultaneous injuries to other structures. Patients with chronic MLKIs may present with persistent pain, feelings of instability, especially during twisting and impact activities, and a mild knee effusion.41 Specific attention should be given to examining the integrity of the ACL, PCL, PLC, MCL and LCL.

The ACL is examined using the anterior drawer test, the Lachman test, and the Pivot Shift test. However, in the setting of chronic MLKI, isolating a positive ACL tear using anterior drawer is increasingly challenging. First, under normal circumstances, at 90° of knee flexion, the anteromedial tibia lies approximately 1 cm anterior to the distal femoral condyles. After disruption of both cruciate ligaments, this relationship may be altered, making it difficult to appreciate a true ACL injury via the anterior drawer test due to posterior subluxation of the tibia.41 Furthermore, although the pivot shift test is a useful clinical test to detect anterior knee instability, in the context of MLKI it may be limited if an associated tear or avulsion of the iliotibial band is present, which may not allow the shift to occur as the knee is progressively flexed during testing.41

The MCL and LCL are tested with a valgus and varus stress, respectively, with the knee held at 30° of flexion to isolate the collateral ligaments. Each test is repeated with the knee in full extension. Excessive lateral joint opening with varus stress in full knee extension suggests injury to other structures, which may include the ACL, PCL and PLC, in addition to an LCL injury.42 Similarly, laxity in valgus stress in full extension suggests an associated posteromedial capsular injury, ACL and/or PCL injury, in addition to an MCL injury.40 42 Thus, although sensitive to detect the presence of associated injuries, these tests are unspecific.

The dial test is used to evaluate the structures that contribute to PLC stability, including the LCL, popliteus tendon and PFL.42 The test is performed with the knee held at both 30° and 90° of flexion. Increased external rotation at 30° but not at 90° is consistent with an isolated PLC injury, while increased external tibial rotation at both 30° and 90° suggests injury to both the PCL and PLC.41 43 An increase solely at 90° of flexion suggests a partial or complete tear of the PCL.41 The clinician must be aware that an increase in external rotation at both 30° and 90° during the dial test may also signify anteromedial rotatory instability instead of a PLC injury.44 Although useful, in the context of MLKI the dial test is again relatively unspecific in characterising individual injuries.45

Studies that have evaluated the sensitivity and specificity of individual clinical tests to detect ligamentous injury, such as those described here, are confined universally to single or two-structure injuries in MLKI,46–52 thus their sensitivity and specificity is largely unknown in the context of chronic MLKI. However, the sensitivity and specificity of these clinical tests is often reduced compared with their quoted accuracy in single or dual structure injuries due to the extent and severity of injury present in MLKI, and the number of structures affected.29 40 53–55 Therefore, although useful, it is essential that clinical examination is accompanied by sensitive and specific imaging.

Vascular injury and investigation

Nineteen studies were identified that specifically assessed vascular injury in the context of MLKI associated with knee dislocations (online supplemental table 3). Two systematic reviews5 8 have attempted to quantify the incidence of vascular injuries associated with knee dislocations; one reported an incidence of associated vascular injury ranging between 6% and 38%,8 and the other reporting an estimated overall incidence of 18% (171 of 862 patients from studies included).5 However, these studies included all classifications of knee dislocations and did not exclusively look at vascular injuries associated with MLKIs not caused by knee dislocations. Therefore, the prevalence and risk of vascular injuries in MLKIs not caused by knee dislocations is unknown, although there is low quality evidence suggesting that the risk of vascular injury is higher in cases of MLKI associated with knee dislocations, than in MLKIs involving two ligaments and not caused by knee dislocations.35

There is a significant body of low order evidence suggesting that clinical examination of pedal pulses alone is insufficient for the accurate diagnosis of vascular injury associated with MLKI,56–58 and there is general agreement that further investigation is required, most commonly in the form of Ankle Brachial Pressure Index (ABPI) measurement . However, controversy remains regarding the selective use of angiography based on combined pedal pulse and ABPI assessment, or routine angiography, in the context of knee dislocation.56 59–61 The majority of evidence assessing the relative value of investigative techniques for vascular injury in MLKI is composed of retrospective studies of relatively small case series.56 59 62–64 Currently, there is a tendency towards selective angiography due to associated risks and costs of routine invasive investigation, and CT angiography has been advocated due to its superior sensitivity and specificity in general orthopaedic trauma.65 Although attempts have been made to devise decision algorithms to aid selective angiography using risk profiles based on mechanism, physical examination and ABPI findings (ABPI <0.9),56 these are based on low order evidence or expert opinion and have not been independently validated for sensitivity and specificity. Nicandri et al 56 showed in a retrospective study the significant benefit of using an evidence-based standardised protocol for evaluation of suspected vascular injury in patients with MLKI; the use of an evidence-based protocol significantly reduced incidence of delay in diagnosis of vascular injury >8 hours, which translated to better clinical outcomes. However, such selective algorithms and evidence-based protocols have not been compared with routine CT angiography in high-quality randomised trials or prospective studies in the context of acute MLKI.

Imaging

Nineteen studies specifically assessed imaging modalities in MLKI (online supplemental table 4). Timely and appropriate imaging is essential in managing MLKI, particularly given that spontaneous reduction of knee dislocations occurs in up to 50% of acute dislocations. This can often lead to diagnosis of MLKI being missed acutely and presenting in the chronic setting.66 Although the use of MRI is widely advocated, literature assessing the sensitivity and specificity of MRI in patients with MLKI is limited.67–73 Original data regarding the value of MRI are of levels II-III evidence.56 67 70 73 74 However, studies report a wide range of sensitivity and specificity. Generally, the literature reports high sensitivity (97%–100%) but lower specificity (50%–67%) for MRI in the diagnosis of cruciate and collateral ligament tears.71 Few studies have assessed the diagnostic ability of MRI to identify injuries to the posteromedial and posterolateral structures, but two have noted reduced sensitivity of MRI in identifying injuries to these structures.67 71 Conversely, LaPrade et al 75 noted high sensitivity and specificity of MRI in identifying injuries to the posterolateral structures, ranging from 68% to 100%. Thus, modern MRI techniques may enable sensitive and specific identification of injuries to some soft-tissue knee structures; however, MRI has limitations particularly for PLC and PMC injuries, and there is no consensus describing the accepted sensitivity and specificity of MRI for diagnosing individual structural injuries in MLKI.

Although MRI is useful in diagnosing and characterising MLKI, it cannot demonstrate the functional consequences of ligament injuries as it is a static study.76 Furthermore, operative decision-making based on degree of instability cannot be made on MRI findings alone.77 Clinical examination risks subjective variation and error, as appropriate examination in these settings relies on the patient’s ability to relax and the clinician’s ability to detect endpoints during application of valgus and varus load at 30° of knee flexion and anteroposterior load.40 The presence of concurrent injuries, often associated with MLKI, can obscure subtle findings in clinical examination, such as injuries to the PMC or PLC. Stress radiographs have been advocated as a useful imaging modality to demonstrate the magnitude of knee instability in an objective and quantifiable way, which can aid preoperative decision making. They have also been advocated for postoperative follow-up of MLKI.78–80 However, there remains controversy regarding the technique of choice for such radiographs, with a variety being described.81–84 Although literature advocates the value of stress radiographs in isolated ligamentous injuries,40 82 85 the yield of such techniques in the setting of MLKI is not well studied, and most studies combine patients with both single ligament injuries and MLKIs in their case cohort. The literature assessing stress radiography in homogeneous cohorts of patients with MLKI is limited almost exclusively to small case series or technical notes,40 86–88 and incorporate a variety of techniques.89 There is a need for higher-quality evidence, perhaps through pooling of comparable data to more robustly inform on the value of stress radiography in MLKI . Furthermore, consensus is required regarding standardisation of stress radiography techniques to allow for pooling of comparable data.

Operative versus non-operative management

Of 11 studies included, 3 were systematic reviews of published evidence and two were narrative reviews (online supplemental table 5). Of remaining studies, one was an RCT published over 15 years ago, and five were retrospective. Over half of published studies were from the USA (n=6). Halinen et al 73 reported, in a prospective randomised study that compared non-operative and operative management of MLKI involving ACL and MCL rupture, equivalent functional results. However, all three systematic reviews favoured operative management of MLKI compared with non-operative management, reporting significantly higher rates of return to work, return to sport and functional outcome.1 12 All studies included in these systematic reviews were of low quality or were not formally assessed for quality. Furthermore, several critical aspects of management such as timing of intervention, technique and rehabilitation varied markedly, making objective comparisons of ‘operative’ and ‘non-operative’ management challenging. Operative techniques and rehabilitation protocols have also greatly evolved in recent years, and several of the studies included in these systematic reviews were over 20 years old. There is a distinct lack of high-quality pooled quantitative analyses of outcomes following operative versus non-operative management of MLKI, reflecting the heterogeneous nature of these injuries and treatment protocols.

Early versus delayed surgery

Fourteen studies reported on timing of surgery for MLKI (online supplemental table 6), although most focused on bicruciate ligament injuries. Of these, eight (60%) were from the USA. All studies were of levels III–V evidence, and 10 were appraisals of published evidence—8 were systematic reviews and 2 were narrative reviews. Therefore, only 5/15 studies (33%) comprised original research assessing early versus delayed surgery. There is consensus for the demarcation in time point between ‘early’ and ‘late’ surgery for MLKI1 23 25 90–94; early being <3 weeks and late being >3 weeks; however, the evidence for this demarcation is unclear. Levy et al 1 noted that 3 weeks had been considered a critical time period following injury, when tissue planes can be identified and are of sufficient integrity to allow reapproximation and suture placement. Such an arbitrary cut-off may oversimplify the concept of optimal time for surgical intervention in MLKI.

The number of appraisals of published evidence compared with original data regarding timing leads to multiple pooled analyses of similar datasets. Eight systematic reviews have drawn varying conclusions regarding relative advantages of early versus delayed intervention in MLKI. Marder et al, 25 Barfield et al 95 and Jiang et al 28 found there was insufficient evidence to advocate one approach over the other, but Marder et al 25 did note better functional outcomes in the delayed intervention cohort. This contrasts with pooled analyses by Levy et al,1 Vicenti et al, 23 Hohmann et al 30 and Mook et al 14 who reported higher functional outcome scores with early intervention, although some have identified greater rates of stiffness.1 14 Ultimately, these studies all suffer from the same flaws: the overwhelming majority comprise retrospective case series or small cohorts, there is inconsistent reporting of a variety of differing outcomes, functional outcome scores used are not validated for MLKI, and separate reviews assess heterogeneous patient populations.

There is a need for consistent outcome reporting in future studies to allow for pooling of data, robust comparisons of interventions and further analyses of more specific time intervals for intervention. Specifically, we require robust evidence that assesses whether bicruciate ligament injuries act similarly, and therefore, should be managed similarly, to MLKI involving one cruciate and one corner. Given that most existing evidence assessing early versus delayed surgery in MLKI assesses bicruciate MLKI, it is still unclear whether this data can be extrapolated to other patterns of injury. Multicentre prospective trials and a registry dataset with common reporting variables may enable high-quality and sufficiently powered studies to be performed to achieve consensus in this regard.

Staged versus single stage surgery

Of those studies that assessed single versus staged approaches to surgery for MLKI, three were systematic reviews (online supplemental table 7)14 25 95; two found that neither approach was superior, and both concluded that the evidence was currently insufficient in both number and quality to draw clear conclusions. Mook et al noted better functional outcomes and less stiffness with staged procedures.14 There appears to be crossover in nomenclature regarding ‘early versus late’ intervention and ‘single versus staged’ intervention, with some studies using these terms interchangeably, incorporating both repair and reconstruction in both strategies.25 95 Furthermore, it remains unclear whether all patterns of MLKI act similarly, and therefore, should be managed singularly. For example, concomitant fracture or extensor injury in MLKI is not an uncommon occurrence, particularly in the context of knee dislocation, and can influence management choice of single versus staged surgery. There is a need for specific evidence and consensus regarding the choice of single versus staged intervention depending on pattern of injury and associated injuries.

Reconstruction versus repair

Seventeen studies explicitly evaluated surgical reconstruction versus repair in MLKI (online supplemental table 8). Fifteen were of level III evidence: two were systematic reviews and the remainder were retrospective. The main controversies lie in whether both extra-articular and intra-articular ligament injuries should be entirely reconstructed or repaired, or whether a combination of these strategies is most appropriate and should be tailored to the pattern of MLKI.

Several studies have approached the subject of repair and reconstruction by assessing individual ligaments separately, and the authors recommend this approach. Vicenti et al 23 evaluated reconstruction versus repair for cruciate ligament rupture, PLC and PMC in the context of MLKI in a systematic review. They described higher-quality studies by Stannard et al 96 and Levy et al 97 which identified significantly lower rates of failure when the PLC was reconstructed, with higher rates of stability. Although retrospective case series have been described advocating repair of the PLC and lateral structures,98 99 no high-quality prospective evidence has yet been identified showing significant advantages over reconstruction.

Recently, Ishibashi et al 100 advocated the acute repair of extra-articular ligaments and a delayed reconstruction of cruciate ligaments, noting that there was no difference in functional outcome between those who had primary extra-articular ligament repair and intra-articular ligament reconstruction, or extraarticular ligament repair only. A more recent retrospective study by Gan et al has also supported this protocol of repair of extra-articular ligaments and reconstruction of cruciate ligaments for Schenck’s KD III (KD III) and IV (KD IV) MLKI.101

Debate continues regarding the treatment of individual ligament injuries with repair or reconstruction in the context of MLKI. The highest quality evidence currently supports reconstruction. The authors advocate a tailored approach with surgical strategy planned separately for individual ligaments. This should be based on the available evidence for outcomes of reconstruction and repair specific to the single ligament in question, the operating surgeon’s preference based on their own skill-set and experience, and patient factors including comorbidity profile, functional level and their own priorities relating to timing of recovery for the purposes of return to work or sport. There is a need for high-quality prospective evidence to assess these operative strategies and achieve consensus.

Early versus delayed range of movement for rehabilitation

Fifteen studies assessed aspects of rehabilitation following MLKI (online supplemental table 9). The two areas of investigation within MLKI rehabilitation that were subject to the highest quality evidence were (1) Early versus delayed mobilisation following surgical intervention for MLKI and (2) the method of bracing postoperatively. Interestingly, postoperative rehabilitation following MLKI has been subject to the highest quality of research within all fields of MLKI research, with three of the four RCTs identified i focused on rehabilitation.18–20

RCTs investigating the benefit of bracing have largely compared two approaches, hinged external fixation or hinged knee bracing.18 19 Angelini et al 18 compared a hinged external fixator to rigid casting in extension (for 3 weeks), followed by gradual progressive passive range of movement exercises, with weightbearing only permitted after 6 weeks. Functional outcomes were significantly higher in the external fixator group. They concluded that the use of an external fixator following MLKI reconstruction is beneficial; however, the fact that this was compared with an alternative strategy of casting in extension for a period of 3 weeks was not emphasised. Stannard et al 19 compared hinged external fixation with hinged knee brace following MLKI reconstruction in an RCT. The trial was not sufficiently powered; however, no significant difference was observed in rate of failures between brace or external fixator, although 28% of MLKI reconstructions failed in brace compared with 15% in fixator. Functional outcome, pain, return to work and overall ROM were comparable.

Hoit et al 20 compared early (day 1) versus late (following 3 weeks of full-time extension splint immobilisation) physiotherapy in an RCT. No statistical difference was found between the need for postoperative MUA between groups in the first 6 months post-procedure, however, the study was underpowered. No differences were found in ROM, stability or patient-reported quality of life at 1 year.

A recent meta-analysis14 of early immobilisation versus early mobilisation for patients who had acute surgical intervention for MLKI concluded that a strategy of early mobilisation was significantly better tfor stability, ROM and functional outcome. This has since been supported by a more recent systematic review.15 However, the authors acknowledged the widely varying protocols for weightbearing, bracing, timing of initiation and types of physical therapy in these studies. Despite the variation in rehabilitation strategies discussed, Keeling et al identified that specific rehabilitation protocols were consistently referenced, described by Edson and Fanelli.91 102

Consensus has not been reached for a specific rehabilitation strategy following surgery for MLKI; the consistent themes for debate are postoperative weightbearing status, optimal type of bracing required, duration of bracing, rehabilitation protocols and early versus delayed physiotherapy. Evidence suggests early physiotherapy achieves better outcomes in stability, ROM and functional outcome,14 15 103 however, significant variation in the rehabilitation protocol employed limits the applicability of this conclusion. Although attempts have been made to conduct high-quality RCTs, most have not been sufficiently powered to provide definitive answers to these questions.

Diversity in research

Although gender was reported, ethnicity was not reported for patients in the systematic reviews and we are therefore unable to comment on potential influence of ethnicity on outcomes following MLKI. Gender has been reported to have an influence on risk of ACL injury and may also be associated with risk for MLKI.104 There was noted to be a lack of gender diversity within author groups, with the 20 most published lead or senior authors being men (online supplemental table 2). A lack of gender and ethnic diversity has been highlighted by multiple sources including the American Orthopaedic Association and British Orthopaedic Association as a critical issue.105 Diversity in healthcare has been shown to be beneficial multilaterally, including in achieving improved patient communication, education and outcomes.106

留言 (0)

沒有登入
gif