Evolving evidence in the treatment of primary and recurrent posterior cruciate ligament injuries, part 2: surgical techniques, outcomes and rehabilitation

Posterior cruciate ligament reconstruction (PCL-R) techniques have been studied and evolved over the past decades, providing a solid and evidence-based foundation for the operative management of posterior cruciate ligament (PCL) injuries [1, 39, 41, 45, 77, 88]. Single-bundle (SB) and double-bundle (DB) PCL-R based on the transtibial or tibial inlay technique and implemented by an open, arthroscopically assisted, or all-arthroscopic approach, all offer certain advantages and disadvantages [1, 8, 15, 16, 34, 40, 45, 75, 77, 87, 89]. On the other hand, the PCL is characterized by a strong intrinsic healing capability, making the non-operative treatment approach a viable option, especially for partial PCL tears and tibial avulsion injuries of the PCL [2, 3, 26, 71, 72]. Good subjective and objective long-term outcomes after non-operative treatment with a fairly low prevalence of 11% of moderate to severe osteoarthritis (OA) after more than 14 years follow-up, keep the debate about the optimal treatment approach, operative vs. non-operative, in PCL injured patients ongoing [71]. As a result, a wide range of viable treatment options is available, enabling an individualized treatment approach based on the injury pattern and the patient’s compliance and demands.

In this review, we present the indications, techniques, and outcomes of operative and non-operative treatment of primary and recurrent PCL injuries. Moreover, risk factors associated with recurrent PCL deficiency and future perspectives are outlined.

Operative treatment of PCL injuries is indicated for patients with symptomatic grade III (complete) tears displaying inadequate functional improvement in response to non-operative treatment. Furthermore, patients with PCL injuries with high-grade knee laxity, or combined with intraarticular or capsuloligamentous injuries should be considered for operative treatment [55, 79]. A side-to-side difference in posterior tibial translation (PTT) greater than 8 mm revealed by stress-radiography indicates a complete PCL tear and presents an indication for operative treatment for symptomatic patients [29, 68, 69]. Additionally, the patient’s demands are essential in treatment decision-making, leading to the recommendation of PCL-R in an athletic population [39]. There is currently insufficient evidence in the literature to support a definitive treatment protocol; however, the historical preference towards non-operative treatment of isolated PCL injuries has recently displayed a shift in favor of operative treatment [17, 61, 79]. Indications for operative treatment may be based on increased translational and rotational tibial movement compared to contralateral PCL intact knees (as measured on in vivo kinematic analyses) [43]. An increasing number of intra-articular injuries (meniscus, cartilage), pathological changes of the anterior cruciate ligament (ACL), and an increased joint contact pressure (tibiofemoral and patellofemoral) have been observed in PCL deficient patients [20].

A failure rate of 1–25% after primary PCL-R is reported, mounting to 45% if unfavorable patient-reported outcomes (Knee Injury and Osteoarthritis Outcome Score < 40 points) are considered as subjective failure [4, 36, 41, 46, 81, 88]. However, inadequate reporting and varying definitions of failures require caution in the interpretation of failure rates. Debilitating pain and functional impairment during daily activities combined with a PTT of  ≥ 10 mm or confirmed graft failure based on magnetic resonance imaging (MRI) scans are accepted indications for revision PCL-R [58, 59]. The aim of revision PCL-R is to address the cause of initial failure and eliminate concurrent pathological changes causing pain and instability [41, 59]. Therefore, in case of revision PCL-R, extended diagnostic work-up including MRI, hip-knee-ankle radiographs, and computed tomography (CT) scans are required to assess concomitant injuries, lower limb alignment, and prior bone tunnel placement. The requirement of bone grafting due to prior semi-anatomic bone tunnels or the need for corrective osteotomies sometimes require staged revision PCL-R.

Owing to the anatomy of the PCL and the complex nature of PCL injuries, there is no consensus for a specific operative technique when considering PCL-R. Many biomechanical and clinical research efforts have enabled the development of techniques focusing on anatomic restoration of native knee kinematics [28, 66, 85]. Variations between the different reconstructive techniques depend primarily on tunnel placement, graft choice, graft positioning and fixation, and the choice of an arthroscopic or open surgical approach (Fig. 1). While studies have reported that both SB and DB PCL-R improve knee kinematics, biomechanical results have recently shown DB PCL-R to closer approximate the native state of the ligament [15, 23, 85].

The focus of anatomic SB PCL-R is to restore knee kinematics by aligning the path of the PCL graft with the native course of the stronger and more prominent anterolateral bundle (ALB). On the other hand, DB PCL-R aims to provide an alternative method by restoring the synergistic functionality of the ALB and posteromedial bundle (PMB) to closer approximate the native anatomy [5, 79].

Tibial graft fixation during anatomic SB and DB PCL-R can be performed using the transtibial or the tibial inlay technique by an all-arthroscopic, arthroscopic assisted, or an open approach [8, 42, 62, 76]. Positioning of a PCL drill guide medial to the tibial tubercle just proximal to the pes anserinus, aiming for a sagittal angle of 45°, allows subsequent insertion of a guidewire to reach the anatomic tibial PCL insertion zone when performing the transtibial technique (a more detailed description of radiographic landmarks for the native femoral and tibial insertion zones of the PCL is given in Part 1 of the evidence-based update on the management of primary and recurrent PCL injuries) [18, 74]. A frequently reported drawback of the transtibial technique is the formation of an acute angle by the PCL graft exiting the tibial tunnel, also known as the “killer turn” [53]. Graft degeneration, abrasion, and delayed or incomplete graft maturation are attributed to the killer-turn effect, which is believed to cause residual posterior laxity and increase the failure rate after transtibial PCL-R [8, 39, 84]. Protection of the posterior neurovascular bundle is of utmost importance. One study has shown that the mean distance between the popliteal artery and the posterior tibial cortex 5 mm distal to the joint line was significantly greater at 90° compared to 0° of knee flexion (7.7 ± 3.8 mm vs. 1.6 ± 1.3 mm) [33]. Accordingly, tibial tunnel drilling with fluoroscopic guidance is recommended at 90° knee flexion to prevent injury of the neurovascular structures.

The tibial inlay techniques have been developed to facilitate graft fixation by employing bone troughs for tibial graft insertion and thereby restoring the original PCL anatomy while diminishing PCL graft stress by avoiding the killer-turn effect [8]. The open or arthroscopically assisted approach to the tibial inlay technique requires the patient to be positioned in the lateral decubitus or prone position, or in supine position if the hip and knee are freely movable. After careful dissection, the bone block of the graft can be placed in the created trough and a screw is used for tibial graft fixation [8, 42].

Posterior cruciate ligament graft fixation is usually performed using interference screws, or suspensory fixation techniques [14, 21, 60, 79]. To optimize graft loads and to restore native knee kinematics, the ideal knee flexion angle during graft fixation is important [24, 31, 32]. Consequently, it has been demonstrated that graft fixation angles ranging from 75° to 105° of knee flexion restore knee kinematics to the same extent after SB PCL-R, albeit failing to restore laxity compared to the intact state [31]. In the case of DB PCL-R, PMB graft fixation at 0° and ALB graft fixation at 90° or 105° of knee flexion best restores native knee kinematics while avoiding excessive restriction of the tibiofemoral joint [32].

Graft choice is crucial in PCL-R and includes allografts and autografts as well as soft tissue-only, bone-tendon, and bone-tendon-bone grafts. While soft tissue-only grafts are often preferred for the transtibial technique, tendon-bone grafts are usually used when performing the tibial inlay technique [8, 42, 67]. Less residual PTT has been demonstrated when using autografts compared to allografts for PCL-R [6, 7]. Although a statistically significant difference between autograft and allograft use was demonstrated, the clinical relevance of a mean side-to-side difference in PTT of less than 1.5 mm has to be questioned [6, 7, 44, 80]. The use of allografts was associated with a shorter operation time, while autografts were associated with donor-site morbidity. With respect to patient-reported outcomes and graft failure rates, no difference between allografts and autografts for PCL-R could be demonstrated [6, 7]. Therefore, no appreciable difference seems to exist between the usage of allogenous or autologous tissue for PCL-R [25, 46, 82].

Unlike in ACL reconstructive procedures, less is known about risk factors and causes of failure in the treatment of PCL injuries [16, 41, 58, 81]. Failed operative treatment of isolated and combined PCL injuries is associated with severe limitations in daily, professional, and sports activities [58, 59, 81]. One study investigating 52 failed operative PCL procedures (including SB PCL-R, PCL repair, synthetic graft replacement, and PCL thermoplasty) reported that 71% of patients complained of moderate to severe pain during daily activities, 49% described their own knee condition as poor, and 75% have completely quit sports activities after a mean time of 42 months after the failed operative PCL treatment [58]. However, synthetic graft replacement and thermoplasty as a treatment for PCL injuries have been abandoned, which needs to be considered when interpreting the reported poor results. To avoid the detrimental effects of failed PCL-R, it is essential to be aware of the risk factors and understand the underlying causes of failure. In more than 50% of failed PCL-R, multiple factors for failure have been identified [16, 41, 58]. Posterolateral corner deficiency and femoral or tibial tunnel misplacement have been shown to be the most common causes of failure, accounting for 40–77% and 33–41%, respectively (Fig. 2) [41, 58]. An incorrect tunnel placement is characterized by a too proximal (deep) and posterior (low) position on the femoral site or a too anterior and proximal position on the tibial site, resulting in a vertical position of the graft [56, 58]. Further risk factors are varus malalignment, primary PCL suture repair, biological and technical failures, wrong surgical decision-making, a too early or too progressive rehabilitation protocol, and a low annual volume of PCL-Rs performed by the operating surgeon [16, 58, 81].

Recently, a biomechanical study has shown an inverse correlation between the graft force after PCL-R and the posterior tibial slope (PTS). Irrespective of the loading condition and the knee flexion angle, a flatter (reduced) PTS leads to increasing PCL graft forces [10]. Clinically, a significant negative correlation between PTS and residual PTT measured on posterior stress radiographs and a significant positive correlation between PTS and the reduction of PTT from pre- to postoperative has been shown for SB PCL-R [22]. However, this observation has not been demonstrated for DB PCL-R [11]. Additionally, a significantly lower PTS has been observed in patients undergoing primary PCL-R compared to sex- and age-matched controls without ligamentous injury (6° vs. 9°) [12]. Analysis of the injury mechanism revealed that the PTS was significantly lower in non-contact PCL injuries compared to contact injuries (5° vs. 6°) [12], highlighting the impact of the PTS on AP laxity.

At 15 years follow-up with graft failure as the endpoint (need for revision PCL-R, high tibial osteotomy, arthroplasty, complete graft tear based on MRI, or > 10 mm side-to-side difference in PTT based on posterior stress radiographs), the survival rates have been reported to be approximately 82% and 84% for SB (n = 28) and DB (n = 36) Achilles tendon allograft PCL-R, respectively [89].

Complications in isolated and combined PCL-R have been reported to occur in up to 53% of surgeries and can be divided into complications generally associated with operative procedures and in complications which are inherently related to the different techniques in PCL-R [15, 34, 36, 38, 40, 45, 59, 67, 70, 77, 81, 88, 90]. Well-known complications in knee surgery and thus also described in PCL-R are postoperative hematoma, surgical site infections, arthrofibrosis, reflex sympathetic dystrophy syndrome, anterior knee pain, paresthesia, neurovascular injuries, deep vein thrombosis, and graft failure with recurrent pain and instability [15, 16, 36, 59, 67, 70, 77, 81, 87, 90]. Iatrogenic injuries to the neurovascular structures of the posterior part of the knee represent the most dreaded complications in PCL-R. The occurrence is rare, yet several case reports about popliteal artery lacerations, occlusions, and also popliteal vein injuries have been published [51, 57, 86]. Precise anatomic knowledge and fluoroscopically guided tibial tunnel drilling in 90° knee flexion may help to prevent iatrogenic neurovascular injuries [33, 54]. In the setting of revision PCL-R, scar formation has to be considered, which may alter the natural course of the neurovascular structures.

A meta-analysis reported that the incidence of perioperative complications is 1.7 times higher for the tibial inlay compared to the transtibial technique. However, no statistically significant difference was noted [40]. The open tibial inlay technique requires careful dissection of the popliteal fossa and poses the risk of neurovascular injuries [8, 42, 81]. Consequently, all-arthroscopic tibial inlay techniques have evolved and have been shown to be biomechanically comparable to an open tibial inlay technique [91]. Additional reported complications related to the tibial inlay technique include fractures of the tibial bone plug during graft fixation [34]. Double-bundle PCL-R is surgically challenging and requires precise knowledge of anatomy and bony landmarks for accurate tunnel placement [5, 56]. One study reported a fracture of the separating bony bridge between the ALB and PMB femoral tunnel in DB PCL-R [87]. Furthermore, the tibial inlay and DB techniques are associated with a significantly longer operation time compared to the transtibial and SB techniques, increasing the overall risk of peri- and postoperative complications [45, 70].

Within the past 10 years (2010s), studies have consistently supported non-operative treatment for isolated grade I, grade II, and nondisplaced tibial avulsion PCL injuries [52, 63, 65, 71, 72]. There continues to be a debate regarding the management of isolated grade III injuries as there is limited data on the outcomes following non-operative treatment. A prospective cohort study in high-level athletes with grade II (n = 25) and grade III (n = 21) isolated acute PCL injuries showed that approximately 83% of athletes were able to participate at a competitive sports level (mean Tegner Activity Scale, 9) after non-operative treatment at an average follow-up period of 5 years [2]. In addition, an epidemiological study demonstrated a median lay-off time of 31 days after PCL injury for professional male soccer players. However, these prospectively collected data in men’s professional soccer included all grades of PCL injuries as well as operatively and non-operatively treated athletes [48]. Accordingly, initial non-operative management based on functional bracing and rehabilitation with optional delayed PCL-R seems to be reasonable for isolated acute PCL injuries, even for high-level athletes with grade III PCL injuries [2]. Although the PCL has a strong intrinsic healing capability, residual posterior laxity is a serious and frequently observed disadvantage of non-operative management [52, 71, 72]. However, the subjective and objective outcomes after non-operative treatment are promising [3, 26, 30, 63, 71, 72]. One prospective study demonstrated increased knee laxity based on manual testing in 9% of patients following non-operative treatment after a mean follow-up of 14 years. Additionally, instrumented laxity testing (KT-1000) revealed a mean side-to-side difference of 3 mm [71]. Nevertheless, the majority of patients were able to regain functional range-of-motion (ROM) and sufficient quadriceps strength to return to activities of daily living, with 45% participating in jumping and pivoting activities [71]. Furthermore, no correlation between functional outcomes and grade of laxity could be observed [71]. While non-operative management remains an integral part of the management of isolated PCL injures, it is important to acknowledge that unsatisfactory outcomes may occur. One study showed that patients undergoing non-operative treatment of isolated PCL injuries occasionally experienced pain and swelling in 81% and 56% of patients, respectively [13]. Additionally, a considerable number of PCL deficient patients developed subsequent meniscal injuries requiring subsequent surgery as well as a deterioration of the articular cartilage on average 13 years after the injury, indicating residual knee laxity [13]. This is also supported by the development of moderate to severe OA in approximately 11% of patients at long-term follow-up [71]. There is a paucity of studies comparing operative and non-operative treatment in PCL deficient patients. However, it has been shown that non-operative treatment leads to significantly more subsequent meniscal injuries as well as a higher rate of OA and a higher conversion rate to total knee arthroplasty compared to operative treatment [83].

Rehabilitation protocols whether for non-operative treatment or postoperative care, are inconsistently reported in the literature [65]. Agreement exists in the combination of temporary immobilization/bracing and exercise therapy. Accordingly, appropriate stabilization by initial static and later functional bracing accompanied by progressive exercise therapy is important, whether post-injury or postoperatively, to support the healing process of the PCL [3, 26, 30]. A dynamic anterior drawer brace facilitates end-to-end contact between the torn PCL fibers by applying an anteriorly directed force along the proximal tibia [37]. Studies demonstrated a reduction of PTT based on instrumented laxity measurement following non-operative treatment using static and dynamic braces with posterior tibial support [26, 30]. Initially, partial weight-bearing is recommended and ROM exercises are performed in the prone position to minimize hamstring activity and to counteract the gravity-induced posterior tibial sag [2, 35, 65]. The following weeks are accompanied by advancement to full weight-bearing with strong emphasis on quadriceps strengthening. Jogging and sport-specific exercises are often initiated in the sixth postoperative month. Full ROM, quadriceps strength, and a firm endpoint in the posterior drawer test are required before return to cutting and pivoting sports [35, 65]. This can take up to 12 months, however, quicker recovery with return to sports at 16 weeks has been reported in high-level athletes [2].

Following PCL-R, weight-bearing as tolerated with a knee brace providing posterior tibial support and locked in full extension is recommended for the first 3–6 weeks, followed by functional bracing for up to 6 months, to promote healing and prevent a fixed posterior tibial subluxation [19, 35, 65]. The authors’ recommendation for non-operative treatment and postoperative rehabilitation is illustrated in Fig. 3.

It is unknown which demographic, surgical, or patient-related factors reliably predict clinical and functional outcomes for both operative and non-operative treatment. Since non-operative treatment may yield satisfactory outcomes, an initial non-operative treatment of PCL injuries with an optional delayed PCL-R is often recommended. However, the optimal timing of operative PCL-R (early vs. delayed surgery) as well as the most appropriate timing of post-injury/postoperative rehabilitation (early vs. delayed) is not sufficiently supported by high-quality evidence, which is even more pronounced in combined PCL injuries. Therefore, a prospective randomized multi-center clinical trial—Surgical Timing and Rehabilitation (STaR) Trial for Multiple Ligament Knee Injuries—is currently ongoing to provide evidence for the optimal timing of operative treatment and non-operative/postoperative rehabilitation [49].

As demonstrated in this two-part review PCL injuries are complex and commonly associated with neurovascular compromise and multiple ligament knee injuries involving the PLC and less commonly MCL. Clinicians must have a thorough understanding of anatomy and biomechanics to aid in decision-making, analysis of concomitant injuries, and operative treatment. Treatment algorithms following history and physical examination involve advanced imaging including stress radiographs and MRI. Exact treatment decisions are finalized and often revised after examination under anesthesia and afford availability of multiple surgical tools and graft choices as well as the flexibility of the experienced surgeon. Good to excellent outcomes with high patient satisfaction can be achieved. High-quality and large-scale studies are needed to provide further evidence for an individualized treatment approach pursuing the ultimate goal of restoring native knee kinematics and facilitating a return to daily, professional, and sports activities.