Postoperative Evaluation of Knees With Posterior Cruciate Ligament Injury Using Patient-Oriented Evaluation Method: Comparison of Single-Bundle and Anatomic Double-Bundle Techniques

Introduction Using the patient-based QOL evaluation scale SF-36 and conventional assessment methods, we evaluated the postoperative outcome of patients with posterior cruciate ligament (PCL) injury who underwent single-bundle or double-bundle reconstruction, and compared the two reconstruction techniques. Methods 37 male patients with isolated PCL injury who underwent reconstruction were randomized to receive single-bundle reconstruction (group S: n=20) or double-bundle reconstruction (group D: n=17). Before surgery and 6 and 24 months after surgery, patients were evaluated by SF-36 scores, Lysholm score, visual analog scale (VAS), posterior tibial displacement rate, and knee range of motion (ROM). Results For SF-36 evaluation at 6 months post-surgery, the scores of all the subscales improved to above the national standard values in group D, whereas none of the subscale scores reached the national standard values in group S, and three subscale scores were inferior in group S compared to group D. At 24 months post-surgery, improvement of all subscale scores to above the national standard values was achieved in both groups. Lysholm score, VAS score, and posterior tibial displacement rate improved after surgery in both groups, but no signicant intergroup differences were observed in all evaluation methods. For knee ROM, residual limitation of exion was signicantly more frequent in group S than in group D at 6 and 24 months post-surgery. Conclusion Arthroscopy-assisted single-bundle PCL reconstruction technique is considered to be a safe procedure with low invasiveness, but despite its widespread use, surgical result is not consistently good. This was attributed to the low reproducibility of the unique course and anatomy of PCL, and the anatomic double-bundle reconstruction technique was proposed aiming to improve treatment outcome. According to the present results, double-bundle reconstruction tended to achieve better restoration at an early stage compared to single-bundle reconstruction, with fewer patients having residual limitation of knee exion after surgery.


Introduction
The posterior cruciate ligament (PCL) is the thickest and strongest ligament in the knee joint and is responsible for 85-100% of posterior stabilization of the knee joint [1,2]. This ligament is composed of two bundles of bers: the anterolateral bundle (ALB) that tightens in knee exion, and the posteromedial bundle (PMB) that tightens in knee extension. The posterior stabilizing effect is obtained by the codominant relationship of these two bundles [3,4]. Ligament reconstruction is considered the rst choice of surgical treatment for PCL injury, and a modality that regains knee function at high activity level. In recent years, the double-bundle reconstruction method that allows more precise and anatomic reproduction of the PCL has been developed [5,6]. Both basic and clinical studies have reported the superiority of the double-bundle reconstruction technique over the conventional single-bundle reconstruction method [7,8]. On the other hand, there are also reports that showed no clear difference in treatment outcome between the single-bundle and double-bundle reconstruction techniques [9][10][11], or even when subjective evaluation showed superiority of the doublebundle reconstruction method, no signi cant difference in clinical outcome was found between the two techniques [12][13][14].
We have evaluated patients with knee ligament injuries using the Medical Outcome Study 36-item shortform health survey (SF-36) [15,16] that allows patient-oriented assessment of QOL subdivided into several health domains, and reported the importance of using not only doctor-based objective assessments but also patient-based subjective evaluation [17][18][19][20]. In the present study, we recruited patients who underwent anatomic double-bundle or single-bundle PCL reconstruction, and evaluated them over time by conventional assessment methods as well as SF-36, and compared the treatment outcome of the two surgical techniques.

Subjects and protocol
Of 113 men who were diagnosed with PCL injury at presentation to the Sports Medicine and Knee Center of Kofu National Hospital between January 2007 and February 2018, 94 patients with isolated PCL injury were followed prospectively, after excluding 19 patients with concurrent injuries of other ligaments, or severe meniscal or articular cartilage injuries. Of the 94 patients followed, 37 who opted for PCL reconstruction after continuously complaining of subjective symptoms for more than 3 months after injury, and underwent primary PCL reconstruction using autologous exor tendons were included in the present study. The patients were randomly assigned to receive one of two reconstruction techniques: 20 patients (aged 15 to 47 years, mean age 26.2 years) underwent single-bundle reconstruction (group S) and 17 patients (aged 16 to 52 years, mean age 29.4 years) underwent double-bundle reconstruction (group D). The subjects were followed until 24 months after surgery ( Figure 1). Only men were included as subjects in the present study, because double-bundle reconstruction was di cult to perform in women with small physical stature.

Outcome measures
In all patients, QOL was evaluated by SF-36 before surgery and 6 and 24 months after surgery, and the results were compared with the Japanese national standard (NBS; norm-based scoring: absolute scores of 0-100 were recalculated by standardizing each scale to have a mean score of 50 and standard deviation of 10 in the general Japanese population). In addition, the knee function in the same periods was assessed using the Visual Analog Scale (VAS, a clinical pain scale) [21], Lysholm scoring scale (minimum score 0, maximum score 100; scores below 65 are interpreted as poor function) [22], posterior tibial displacement rate measured from a stress plain radiograph taken while using a Telos SE device (Telos Japan, Tokyo, Japan) (Measurement was made with the knee exed to 90° and a force of 15 KPa applied to the anterior aspect of the center of tibia. Displacement was measured as the mid-point displacement rate. PCL impairment was diagnosed when the displacement rate was 45% or below) [23][24][25], and the difference in range of motion between the affected and unaffected knees. These outcome measures were compared between group S and group D.
SF-36 is composed of the following 8 subscales: physical functioning (PF) role-physical (RP), bodily pain (BP) and general health (GH), which constitute the physical health component; as well as vitality (VT), social functioning (SF), role-emotional (RE) and mental health (MH), which constitute the mental health component.

Surgical techniques
All the surgeries were performed by the rst author (S.O.) as the main operator.
To harvest and prepare the graft tendon, the patient was placed supine with the knee joint exed at approximately 90°. Through a skin incision of approximately 2.5 cm made on the medial side of the tibial tubercle, the semitendinosus tendon was elevated, together with the gracilis tendon if needed. Then the harvested tendon was bundled, and the two ends were attached to arti cial tendons (Endobutton Tape and Endobutton CL; Acufex; Smith & Nephew Mans eld, Massachusetts) to prepare the tendon graft. An arthroscope with 30° oblique view was used, and arthroscopic procedures were conducted via the anteromedial, antero-lateral and postero-medial portals.

Single-bundle reconstruction
Using the Pro-trac PCL Guide System (Acufex; Smith & Nephew), a specialized guide wire was inserted into the center of the PCL tibial footprint, and a cannulated drill and dilator were used to create a bone tibial tunnel with a diameter determined according to the width of the tendon graft. Next, using a Flip Cutter II (Arthrex), a bone femoral tunnel with a diameter depending on the width of the tendon graft was created by the outside-in method on the femoral side approximately 7 mm posterior to the margin of the articular cartilage at approximately 2 o'clock (right knee) or 10 o'clock (left knee) position. The tendon graft on the femoral side was xed using Endobutton (Acufex) and that on the tibial side using spike staples by double stapling method. The graft was xed at 90° knee exion with a tension of 60 N applied to the tendon graft. All the tendon grafts prepared had diameters of 8.5 mm or larger (8.5 to 9.0 mm) and lengths of 65 mm or longer (65 to 75 mm).

Double-bundle reconstruction
Using the Pro-trac ACL Guide System, a guide wire was inserted into the posteromedial part of the PCL tibial footprint, and then a cannulated drill and dilator were used to drill a bone tibial tunnel with the same diameter as the PMB of the graft tendon. Using the same method, another bone tibial tunnel with the same diameter as the ALB of the tendon graft was made in the anterolateral part of the PCL tibial footprint. Next, two femoral tunnels, each with the same diameter as the PMB or ALB, were created using the outside-in method. The tunnel for PMB reconstruction was located approximately 8 mm posterior to the articular cartilage margin at the anterior aspect of the intercondylar fossa at approximately 2:30 o'clock (right knee) or 10:30 o'clock (left knee) position. The tunnel for ALB reconstruction was located approximately 6 mm from the articular cartilage margin at approximately 1:00 o'clock (right knee) or 11:00 o'clock (left knee) position. For both bundles, the femoral side was xed with Endobutton and the tibial side with spike staples by double stapling method. The bundles were xed at 90° knee exion for the AMB and full knee extension for the PLB, while applying a tension of 40 N to the tendon graft. In all the tendon grafts harvested, both bundles had diameters of 6 mm or larger (6-7 mm) and lengths of 60 mm or longer (60-70 mm).

Postoperative management
The postoperative management protocol was the same for group S and group D. Range of motion training while wearing an orthosis with angle limitation was started from 2 weeks after surgery. Partial weight-bearing was permitted from 3 weeks, and full weight bearing from 6 weeks. Sports activities were restarted from around 9 months after surgery.

Statistical analysis
Mann-Whitney's U test and two-way ANOVA were used for statistical analyses of data. A p value less than 0.05 was considered signi cant. Statistics analyses were conducted using BellCurve for Excel (Social Survey Research Information Co., Ltd.).

Results
No serious postoperative complications such as re-rupture and deep wound infection occurred in both group S and group D, and none of the patients required re-operation. None of the patients deviated from the protocol after surgery.

Subjective evaluation by SF-36
The results of evaluation using SF-36 are shown in Figure 2. At 6 months after surgery, the scores of all the subscales improved to above the national standard values in group D, whereas none of the subscale scores reached the national standard values in group S. Furthermore, PF, RP and BP scores in group S were signi cantly worse than those in group D (p < 0.05).
At 24 months after surgery, both groups S and D achieved improvement of all subscale scores to above the national standard values. Moreover, PF, RP and BP scores improved signi cantly compared to before surgery in both groups (p < 0.05).
Signi cant improvement was observed at 6 and 24 months after surgery compared to before surgery in both group S and group D, but no signi cant intergroup differences were found.

Evaluation of pain by VAS
The mean VAS scores before surgery and 6 and 24 months after surgery were respectively 45.5 ± 25.8, 19.4 ± 20.3 and 16.1 ± 23.5 in group S; and 54.5 ± 24.4, 16.4 ± 19.2 and 11.8 ± 15.5 in group D. Although signi cant improvement was achieved at 6 and 24 months after surgery compared to before surgery in both groups, there were no signi cant differences between the two groups.

Evaluation of knee instability by posterior tibial displacement rate
The mean posterior tibial displacement rates (%) before surgery and 6 and 24 months after surgery were respectively 44.6 ± 9.1, 51 ± 7.7 and 53.7 ± 6.8 in group S; and 43.9 ± 4.2, 52.3 ± 7 and 51.6 ± 5.9 in group D. Although signi cant improvement was obtained after surgery compared to before surgery in both groups, no signi cant intergroup differences were observed.

Evaluation of knee range of motion
The results of knee range of motion are shown in Table 1. Limitation of exion of 5 degree or more remained detectable at 6 and 24 months after surgery in 9 patients (45%) and 6 patients (30%), respectively, in group S; and in 2 patients (30%) and 1 patient (5%) in group D. The proportions of patients with limitation of exion were signi cantly higher in group S than in group D (6 months after surgery: p = 0.021, 24 months after surgery: p = 0.041).

Discussion
PCL has a strong innate healing capacity, and many patients with PCL injury attain good improvement with conservative therapy [26][27][28][29], but surgery is selected by patients in whom severe posterior instability remains and subjective symptoms persist [30]. Among the surgical modalities, arthroscopic single-bundle PCL reconstruction is widely used in view of its low invasiveness and safety. However, according to a systematic review reported by Kim et al. [31], arthroscopically assisted single-bundle PCL reconstruction for high-grade PCL injuries provides some improvement of instability, but does not restore normal knee stability or prevent the development of degenerative osteoarthritis. In our previous studies, we found that persistent limitation of exion accompanied by pain deteriorated the treatment result of single-bundle PCL reconstruction [17,18].
Low reproducibility of the unique course and anatomy of the PCL was considered to be the cause of unsuccessful PCL reconstruction [32]. To overcome these issues, the anatomic double-bundle reconstruction method was developed [5,6]. In this study, we compared the relative merits and demerits of the single-bundle and double-bundle reconstruction techniques using the patient-based SF-36 healthrelated QOL scale with scienti cally proven reliability and validity [33,34] together with the conventional objective clinical measures.
In the present study, improvements in Lysholm score, VAS score, and posterior tibial displacement rate after surgery compared to before surgery were achieved by both single-bundle and double-bundle reconstruction techniques, with no signi cant intergroup differences in all three objective assessment methods. On the other hand, evaluation using SF-36 showed improvement of all subscale scores to above the national standard values in group D from the early post-surgical period of 6 months, whereas none of the subscale scores reached the national standard values in group S, and signi cant intergroup differences in three subscales belonging to the physical health component were observed. At 24 months after surgery, improvement of all subscale scores to above the national standard was attained in both groups, and all the subscale scores were apparently higher in group D than in group S, although there were no signi cant differences. Regarding range of motion of the knee, signi cantly higher proportions of knees in group S had residual limitation of exion compared to group D, both at 6 and 24 months after surgery.
By reconstructing the ALB and PMB separately, the double-bundle reconstruction technique is considered capable of mimicking the native PCL both anatomically and functionally [7,[35][36][37]. We speculate that in the double-bundle reconstruction, the morphology of the tendon graft divided into two bundles more closely reproduces the at structure of the native PCL and reduces the interference in the popliteal region during exion, which may have decreased the limitation of exion after reconstruction as observed in this study. Smooth knee motion relieves the physical pain from the early period after surgical, which probably contributes to favorable subjective evaluation of the double-bundle reconstruction technique by patients.
At the last evaluation of treatment outcome, overall improvement was observed in both subjective and objective evaluations for both surgical techniques, with no clear differences. However, we believe that anatomic double-bundle reconstruction, which confers bene ts of smooth knee motion early after surgery and low rate of residual limitation of exion, should be recommended. We carried out a prospective clinical trial Ethical approval was obtained from the institutional review board of the National Hospital Organization, Kofu National Hospital. Informed consent was obtained from all the patients.

Consent for publication
Not applicable Availability of data and materials All data and materials were in the compliance with the journal's policy fully.
Competing interests Figure 1 Flowchart of prospective study of male patients with posterior cruciate ligament injury treated at our center. of all subscale scores to above the national standard values was achieved in both groups, and signi cant improvement of PF, RP and BP compared to pre-surgery scores was observed in both groups.