- Email: [email protected]
Anatomic Reconstruction of the Anterior Cruciate Ligament Using Double-Bundle Hamstring Tendons: Surgical Techniques, Clinical Outcomes, and Complications
Anatomic Reconstruction of the Anterior Cruciate Ligament Using Double-Bundle Hamstring Tendons: Surgical Techniques, Clinical Outcomes, and Complications Hiroto Asagumo, M.D., Masashi Kimura, M.D., Ph.D., Yasukazu Kobayashi, M.D., Ph.D., Masanori Taki, M.D., and Kenji Takagishi, M.D., Ph.D.
Purpose: The objective of the study was to retrospectively compare the clinical outcomes of anatomic double-bundle anterior cruciate ligament reconstruction via hamstring tendons with singlebundle reconstruction between April 2002 and March 2004. Methods: We retrospectively reviewed 123 consecutive patients, 71 of whom underwent double-bundle reconstruction and 52 of whom underwent single-bundle reconstruction. The same postoperative rehabilitation protocol was used for all patients. The patients were followed up for a mean of 33 months. We evaluated manual knee laxity, anterior knee laxity as measured with the KT1000 arthrometer (MEDmetric, San Diego, CA), range of knee motion, isokinetic peak torque of knee extension and flexion strength adjusted for body weight as determined by Cybex testing (Lumex, Ronkonkoma, NY), and Lysholm score. Results: The Lachman test was negative in 64 cases (90%) and the pivot-shift test was negative in 62 cases (87%) in the double-bundle group. The Lachman test was negative in 45 cases (86%) and the pivot-shift test was negative in 42 cases (81%) in the single-bundle group. There was an extension deficit of greater than 5° in 19 cases (26%) in the double-bundle group and 6 cases (10%) in the single-bundle group (P ⬍ .05). The side-to-side difference in anterior tibial translation measured with the KT1000 arthrometer was 1.7 ⫾ 2.0 mm in the double-bundle group and 1.9 ⫾ 2.2 mm in the single-bundle group. The isokinetic peak torque of knee extension and flexion strength was 90% and 89%, respectively, in the double-bundle group and 87% and 86%, respectively, in the single-bundle group. The Lysholm score averaged 96.8 ⫾ 5.1 in the double-bundle group and 92.8 ⫾ 6.9 in the single-bundle group postoperatively. Conclusions: No significant difference was found between the 2 procedures with regard to manual knee laxity, anterior knee laxity measured by the KT1000 arthrometer, knee extension and flexion strength, and Lysholm score. In contrast, there was a significant difference in the range of knee motion between the 2 groups. The findings of our study do not support the routine adoption of double-bundle reconstruction. Level of Evidence: Level III, retrospective comparative study. Key Words: Anterior cruciate ligament—Double bundle—Anatomic reconstruction—Posterolateral bundle—Complications.
S
From Zenshukai Hospital, Gunma Sports Medicine Research Center (H.A., M.K., Y.K., M.T.), and Department of Orthopaedic Surgery, Gunma University Graduate School of Medicine (K.T.), Maebashi, Japan. The authors report no conflict of interest. Address correspondence and reprint requests to Hiroto Asagumo, M.D., Zenshukai Hospital, Gunma Sports Medicine Research Center, 1381 Ninomiya, Maebashi, Japan 379-2117. E-mail: [email protected] © 2007 by the Arthroscopy Association of North America 0749-8063/07/2306-6254$32.00/0 doi:10.1016/j.arthro.2007.01.009
602
everal authors have described double-bundle anterior cruciate ligament (ACL) reconstruction via hamstring tendons.1-12 Although normal knee laxity after ACL reconstruction is gained by use of bone– patellar tendon– bone graft, as well as hamstring tendons, the procedure using bone–patellar tendon– bone has some disadvantages, such as anterior knee pain and slow recovery of quadriceps muscle strength. There is concern regarding the current technique of ACL reconstruction in terms of residual laxity, especially in controlling rotation. To control rotatory instability, a number of authors have suggested reconstructing not just the anteromedial (AM) bundle but
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 23, No 6 (June), 2007: pp 602-609
ACL REPAIR VIA DOUBLE-BUNDLE HAMSTRING TENDON TABLE 1.
Age (yr) Sex (M/F) Mean preoperative period (mo) Mean follow-up period (mo) (range) Accompanying meniscal injury
603
Demographic Data
Single-Bundle Group
Double-Bundle Group
25.3 21/31 19 28 (24-39) MM in 10 (repair in 4 and meniscectomy in 2) and LM in 18 (repair in 1 and meniscectomy in 4)
24.2 33/38 17 29 (24-36) MM in 14 (repair in 7 and meniscectomy in 3) and LM in 35 (repair in 4 and meniscectomy in 7)
Abbreviations: MM, medial meniscus; LM, lateral meniscus.
also the posterolateral (PL) bundle.1-11 Adachi et al.5 and Hamada et al.2 indicated that there were no advantages for a double-bundle reconstruction as opposed to a single-bundle reconstruction in terms of anterior laxity. On the contrary, Muneta et al.11 and Yasuda et al.10 reported that the side-to-side anterior laxity of their double-bundle ACL reconstructions was significantly better than that of the single-bundle reconstruction. Since April 2000, we have performed double-bundle ACL reconstruction with the EndoButton device (Smith & Nephew Endoscopy, Andover, MA) for femoral-side fixation according to the method of Rosenberg and Graf13 to improve the results of reconstruction via semitendinosus (gracilis). In this study we compared the clinical outcome of anatomic doublebundle ACL reconstruction with single-bundle reconstruction in terms of manual knee laxity, range of knee motion, knee extension and flexion strength as determined by Cybex testing (Lumex, Ronkonkoma, NY), anterior knee laxity as measured by use of the KT1000 arthrometer (MEDmetric, San Diego, CA), and Lysholm score. Our hypothesis is that the anatomic double-bundle reconstruction is superior to the singlebundle procedure in terms of knee stability, especially with regard to rotatory instability. METHODS We retrospectively reviewed 123 consecutive patients with ACL-deficient knees who had undergone single- or double-bundle reconstruction between April 2002 and March 2004. Patients who underwent ACL reconstruction of both knees and those with an open physis were excluded from evaluation. Those with other ligamentous instability and those with articular cartilage lesions exceeding grade III were also excluded. The final study population consisted of 54 men and 69 women.
Double-bundle ACL reconstructions were performed in 71 consecutive patients (Table 1). The mean patient age was 24 years 2 months. The mean follow-up period was 2 years 5 months (range, 24 to 36 months). Single-bundle ACL reconstructions were performed in 52 consecutive patients. The mean patient age was 25 years 3 months. The mean follow-up period was 2 years 4 months (range, 24 to 39 months). At 1 year postoperatively, we performed secondlook arthroscopy, examining the status of the reconstructed ACL, meniscus, and articular cartilage and removing the staples from the tibia in all patients. At 1 year and 2 years postoperatively and at the final follow-up, we examined manual knee laxity, anterior tibial translation measured with the KT1000 knee arthrometer, range of knee motion, Lysholm score, and knee extension and flexion strength. Anterior tibial translation was measured with the KT1000 arthrometer with the knee in 20° of flexion, with an anterior force of 89 N being applied to the tibia. The maximum extension and flexion strength (adjusted for body weight) of both knees were measured by Cybex testing at 60°/s and are expressed as the percentage of the uninjured knee. Statistical analysis was performed with the unpaired t test, Mann-Whitney U test, and 2 test. The unpaired t test was used for anterior knee laxity measured with the KT1000 arthrometer and Lysholm score. The Mann-Whitney U test was used for manual knee laxity. The 2 test was used for range of knee motion, as well as knee extension and flexion strength. P ⬍ .05 was considered statistically significant. Surgical Procedure Double-Bundle Procedure: All surgeries were performed by use of the same procedure in each patient. After the knee was examined with the patient under general anesthesia, diagnostic arthroscopy was performed without an air tourniquet.
604
H. ASAGUMO ET AL.
FIGURE 1. Angle placement of guidewire: (A) axial view, (B) coronal view, and (C) sagittal view. The AM guidewire was set at 10° to 15° from the sagittal plane and 43° to 45° from the axial plane. The PL guidewire was set at 40° to 50° from the sagittal plane and 40° to 43° from the axial plane.
Semitendinosus tendons were harvested by use of a tendon harvester with a tourniquet after an approximately 4-cm-long longitudinal incision was made in the AM portion of the tibia 2 cm below the tibial tuberosity. If the diameter of either the AM or PL graft was smaller than 6 mm, we added the gracilis tendons. The angle of the tibial drill guide for the AM bundle was set at 43° to 45° from the axial plane and 10° to 15° from the sagittal plane. The angle of the tibial drill guide for the PL bundle was set at 40° to 43° from the axial plane and 40° to 50° from the sagittal plane (Fig 1). A Kirschner wire, 2.4 mm in diameter, was used as the
guidewire. The tibial guidewire for the AM bundle was positioned at the anterior portion of the natural ACL insertion, which is located 13 mm anterior to the anterior edge of the posterior cruciate ligament. The tibial guidewire for the PL bundle was placed 7 mm posterior and lateral to the AM guidewire (Fig 2). The tibial guidewire position was determined with an intraoperative 2-dimensional radiograph in the hyperextended position. The radiograph was checked according to anatomic landmarks to determine whether tibial guidewires were located within the normal ACL insertion. Two tibial tunnels were made with a cannulated drill corresponding to the measured diameter of
ACL REPAIR VIA DOUBLE-BUNDLE HAMSTRING TENDON
605
FIGURE 2. Tibial insertion of AM and PL bundles (right knee). The PL insertion is positioned at the posterior portion of the natural ACL insertion and immediately adjacent to the posterior cruciate ligament.
the prepared substitute. The diameter of the tibial tunnels was always 6 or 7 mm. To create the AM femoral tunnel in the lateral condyle, the femoral drill guide was positioned 5.5 mm anterior to the posterior edge of the intercondylar notch. The PL femoral guidewire was set lateral and posterior to the AM femoral tunnel with the knee in 90° of flexion (Fig 3). A Kirschner wire, 2.4 mm in diameter, and a cannulated drill (6 or 7 mm) were used in the same manner as for the tibial tunnels. The length of the femoral tunnels was always 25 to 30 mm, and the diameter of the femoral tunnels was always 6 or 7 mm. The length of each femoral drill hole was measured with a scaled probe (Depth Probe; Smith & Nephew Endoscopy). The AM bundle was positioned between 1:00 and 2:00 o’clock and the PL bundle around 3:00 o’clock on the lateral femoral condyle when reconstructing a left knee. The harvested semitendinosus tendon was cut in half and folded in a double-strand fashion. An EndoButton CL device (Smith & Nephew Endoscopy) was attached at the looped end, and a Telos artificial ligament (Telos, Marburg, Germany) was attached at the other end via a glove suture technique. The graft for the PL bundle was first introduced through the joint to the femoral drill hole by use of a Passing Pin (Smith & Nephew Endoscopy). The EndoButton was flipped and fixed on the femoral cortical surface. Then the graft for the AM bundle was introduced and fixed in the same manner. The staple was used for graft fixation on the tibial side at 30° of
FIGURE 3. Femoral insertion of AM and PL bundles (right knee). The PL insertion is positioned at the lowest point of the lateral femoral condyle.
knee flexion. Tibial fixation of the AM and PL bundles was carried out with 30 N of traction applied to each bundle (Figs 4-6). Postoperatively, a knee brace was used to immobilize the knee at 20° of flexion. Single-Bundle Procedure: Routine arthroscopic examination was performed by use of the same procedure as for the double-bundle reconstruction. The position of the tibial tunnel was at the center of the
FIGURE 4. Arthroscopic view showing grafted tendons for AM and PL bundles.
606
H. ASAGUMO ET AL. had recovered quadriceps strength to more than 60% of that in the uninjured knee. Sprinting and various competitive exercises were allowed 6 months postoperatively if the patient had recovered quadriceps strength to more than 80% of that in the uninjured knee. RESULTS Manual Knee Laxity In the double-bundle group the Lachman test was negative in 64 patients (90%), positive (⫹) in 5 (7%), and highly positive (⫹⫹) in 2 (3%) (Table 2). The pivot-shift test was negative in 62 patients (87%), positive (⫹) in 7 (10%), and highly positive (⫹⫹) in 2 (3%).
FIGURE 5. Postoperative radiograph. Two EndoButtons on the femoral cortex and four staples in the tibia are observed in the anteroposterior view.
ACL attachment and the position of the femoral tunnel was between 1:30 and 2:00 o’clock on the lateral femoral condyle when reconstructing a left knee. Quadrupled semitendinosus tendon was used for a grafted tendon, and gracilis was used if the semitendinosus tendon was insufficient. The tendon diameter was always 8 or 9 mm. Tibial fixation was carried out by use of staples with manual maximum traction at 30° of knee flexion. Postoperative Management The same postoperative rehabilitation protocol was used for all patients. Continuous passive motion was started 2 days postoperatively. Two weeks postoperatively, weight-bearing was started. Jogging was encouraged after 3 months postoperatively if the patient
FIGURE 6. Schema of anatomic reconstruction procedure of AM and PL bundles of ACL.
ACL REPAIR VIA DOUBLE-BUNDLE HAMSTRING TENDON TABLE 2.
Manual knee laxity Lachman test* ⫺ ⫹ ⫹⫹ Pivot-shift test* ⫺ ⫹ ⫹⫹ Anterior knee laxity (mm)* Range of knee motion Extension deficit† Flexion deficit* Isokinetic peak torque Extension strength* Flexion strength* Lysholm score* Preoperative Postoperative
Results
Single-Bundle Group
Double-Bundle Group
45 (86%) 6 (12%) 1 (2%)
64 (90%) 5 (7%) 2 (3%)
42 (81%) 7 (14%) 3 (5%) 1.9 ⫾ 2.2
62 (87%) 7 (10%) 2 (3%) 1.7 ⫾ 2.0
6 (10%) 3 (6%)
19 (26%) 5 (7%)
87% 86%
90% 89%
65.8 ⫾ 14.9 92.8 ⫾ 6.9
70.1 ⫾ 16.4 96.8 ⫾ 5.1
*P ⬎ .1. †P ⬍ .05.
607
Isokinetic Knee Extension and Flexion Strength by Cybex Testing The isokinetic peak torque of knee extension strength adjusted for body weight was 90% in the double-bundle group and 87% in the single-bundle group. The isokinetic peak torque of knee flexion strength adjusted for body weight was 89% in the double-bundle group and 86% in the single-bundle group. Lysholm Score The Lysholm score averaged 70.1 ⫾ 16.4 preoperatively and 96.8 ⫾ 5.1 postoperatively in the doublebundle group. In contrast, in the single-bundle group the score was 65.8 ⫾ 14.9 preoperatively and 92.8 ⫾ 6.9 postoperatively. Regarding the Lachman test, pivot-shift test, anterior knee laxity measured by use of the KT1000 arthrometer, knee extension and flexion strength, and Lysholm score, there was no significant difference between the single-bundle group and double-bundle group. In contrast, there was a significant difference in
In the single-bundle group the Lachman test was negative in 45 patients (86%), positive (⫹) in 6 (12%), and highly positive (⫹⫹) in 1 (2%). The pivot-shift test was negative in 42 patients (81%), positive (⫹) in 7 (14%), and highly positive (⫹⫹) in 3 (5%). Anterior Knee Laxity Measurements With KT-1000 Arthrometer In the double-bundle group the side-to-side difference in anterior tibial translation measured with the KT-1000 arthrometer was 1.7 ⫾ 2.0 mm. In the single-bundle group, the side-to-side difference was 1.9 ⫾ 2.2 mm. In 93% of patients in the double-bundle group and 90% of patients in the single-bundle group, the sideto-side difference ranged between 0 and 3 mm. Range of Knee Motion In the double-bundle group an extension deficit of more than 5° was found in 19 patients (26%) and a flexion deficit of more than 5° was found in 5 patients (7%). In the single-bundle group an extension deficit of more than 5° was found in 6 patients (10%) and a flexion deficit of more than 5° was found in 3 patients (6%).
FIGURE 7. Magnetic resonance image showing destruction of medial tibial plateau.
608
H. ASAGUMO ET AL.
FIGURE 8. Arthroscopic view showing damage to medial tibial plateau, as well as to PL bundle, caused by malpositioning of PL guidewire.
range of knee motion (extension deficit) between the 2 groups (P ⬍ .05). Meniscal lesions and articular cartilage lesions did not have an influence on clinical outcomes. Complications At second-look arthroscopic examination 1 year postoperatively, there were 5 cases showing PL bundle rupture only. Among 203 patients who underwent surgery between April 2002 and May 2005, we encountered 2 cases showing articular surface injuries of the medial tibial plateau, which was caused by malpositioning of the PL guidewire (Figs 7 and 8). When we create the tibial tunnel for the PL bundle via the transtibial tunnel technique, care is necessary to ensure that the angle of the Kirschner wire does not become more horizontal. There were no cases of artery or nerve injury and no signs of infection. DISCUSSION Several authors have reported on the double-bundle procedure via hamstring tendons.1-5,9,10,12,13 Recently, Yasuda et al.9 reported on anatomic reconstruction of the AM and PL bundles of the ACL by use of hamstring tendon grafts. Good clinical results were re-
ported with this method, and a new method of creating tibial and femoral tunnels for the PL bundle was introduced. The ACL can be anatomically divided into AM and PL portions, each of which shows different tension patterns.4,14-16 Theoretically, double-bundle reconstruction has several advantages over single-bundle reconstruction with a multistranded tendon with respect to regaining a structure that closely resembles the normal ACL. First, a double-bundle ACL has wider contact areas between the bone and grafted tendon than a singlebundle ACL via hamstring tendons, which means that only the margin of the tendon graft is anchored to the collagen fibers, resembling Sharpey’s fibers, and the tunnel wall in ACL reconstruction.7 Yagi et al.17 suggested biomechanical advantages of the double-bundle procedure compared with the single-bundle procedure in an in vitro study. Second, double-bundle ACL reconstruction allows AM and PL bundles to be created anatomically and independently. We compared the surgical outcomes of doublebundle ACL reconstruction with those of single-bundle ACL reconstruction performed between April 2002 and March 2004. We have found a significant number of flexion contractures 1 year postoperatively compared with the single-bundle procedure. In all cases flexion contractures were improved by performing notchplasty or removing Cyclops lesions during second-look arthroscopy. Our data showed that there was no significant difference in terms of knee stability, especially in terms of rotatory instability. Producing anatomic PL grafts did not affect the pivot-shift test in the double-bundle group. We encountered 2 cases showing articular surface injuries of the medial tibial plateau at the beginning of the anatomic double-bundle procedure. The complication of medial tibial plateau injury caused by malpositioning of the PL guidewire has not been described, to our knowledge, after the double-bundle procedure. Actually, we could not detect the articular surface injury of the medial tibial plateau immediately after overdrilling the Kirschner wire for the PL bundle. In both cases the patients have no clinical symptoms, such as pain, swelling, or giving way. We first detected this complication on follow-up magnetic resonance imaging 1 year postoperatively, and we confirmed the articular surface injury on second-look arthroscopy. We thought that, first, the subchondral bone was destroyed by overdrilling for the PL bundle and, second, the fragile articular surface was gradually
ACL REPAIR VIA DOUBLE-BUNDLE HAMSTRING TENDON punctured by the PL bundle. When we create the tibial tunnel for the PL bundle via the transtibial tunnel technique, care is necessary to prevent the angle of the Kirschner wire from becoming more horizontal. In this study we encountered 5 cases showing PL bundle rupture only on second-look arthroscopic examination 1 year postoperatively. It was thought that extension deficit and PL bundle rupture were related closely. The length pattern of the PL bundle is more tense in extension in an anatomic PL position than in the conventional PL position of the femoral tunnel. Tensioning the PL bundle near full extension makes theoretic sense because of the tension patterns in this bundle through a range of motion. The limitation of our study is that the follow-up period was only 2 years. To establish the efficacy of this double-bundle reconstruction, further prospective study with a longer follow-up period is definitely needed. CONCLUSIONS This study compared the results of double-bundle ACL reconstruction with those of single-bundle reconstruction and confirmed that there was no significant difference, with the exception of range of knee motion, between the 2 procedures. An extension deficit of greater than 5° was found in 19 cases (26%) in the double-bundle group and 6 cases (10%) in the single-bundle group (P ⬍ .05). The findings of our study do not support the routine adoption of the double-bundle procedure until the risk-benefit comparison to single-bundle reconstruction warrants the change.
3. 4.
5.
6.
7.
8. 9.
10.
11.
12.
13. 14. 15.
REFERENCES 16. 1. Bellier G, Christel P, Colombet P, Djian P, Franceschi JP, Sbihi A. Double-stranded hamstring graft for anterior cruciate ligament reconstruction. Arthroscopy 2004;20:890-894. 2. Hamada M, Shino K, Horibe S. Single- versus bi-socket anterior cruciate ligament reconstruction using autogenous multiple-stranded hamstrings tendons with EndoButton for
17.
609
femoral fixation: A prospective study. Arthroscopy 2001;17: 801-807. Mott HW. Semitendinosus anatomic reconstruction for cruciate ligament insufficiency. Clin Orthop Relat Res 1983:90-92. Muneta T, Sekiya I, Yagishita K, Ogiuchi T, Yamamoto H, Shinomiya K. Two-bundle anterior cruciate ligament reconstruction using semitendinosus tendon with endobuttons: Operative technique and preliminary results. Arthroscopy 1999; 15:618-623. Adachi N, Ochi M, Uchio Y, Iwasa J, Kuriwaka M, Ito Y. Reconstruction of the anterior cruciate ligament. Single- versus double-bundle multistranded hamstring tendons. J Bone Joint Surg Br 2004;86:515-520. Miura K, Woo SL-Y, Brinkley R, Fu Y-C, Noorani S. Effects of knee flexion angles for graft fixation on force distribution in double-bundle anterior cruciate ligament grafts. Am J Sports Med 2006;34:577-585. Tomita F, Yasuda K, Mikami S, Sakai T. Comparisons of intraosseous graft healing between the doubled flexor tendon graft and the bone-patellar tendon-bone graft in anterior cruciate ligament reconstruction. Arthroscopy 2001;17:461-476. Zelle BA, Brucker PU, Feng MT, Fu FH. Anatomical doublebundle anterior cruciate ligament reconstruction. Sports Med 2006;36:99-108. Yasuda K, Kondo E, Ichiyama H, et al. Anatomic reconstruction of the anteromedial and posterolateral bundles of the anterior cruciate ligament using hamstrings tendon grafts. Arthroscopy 2004;20:1015-1025. Yasuda K, Kondo E, Ichiyama H, Tanabe Y, Tohyama H. Clinical evaluation of anatomic double-bundle anterior cruciate ligament reconstruction procedure using hamstring tendon grafts: Comparisons among 3 different procedures. Arthroscopy 2006;22:240-251. Muneta T, Koga H, Morito T, Yagishita K. A retrospective study of the midterm outcome of two-bundle anterior cruciate ligament reconstruction using quadrupled semitendinosus tendon in comparison with one-bundle reconstruction. Arthroscopy 2006;22:252-258. Mae T, Shino K, Miyama T, et al. Single- versus two-femoral socket anterior cruciate ligament reconstruction technique: Biomechanical analysis using a robotic simulator. Arthroscopy 2001;17:708-716. Rosenberg TD, Graf B. Techniques for ACL reconstruction with Multi-Trac drill guide. Mansfield, MA: Acufex Microsurgical, 1994. Amis A, Dawkins G. Functional anatomy of anterior cruciate ligament. J Bone Joint Surg Br 1991;73:260-267. Girgis FG, Marshall JL, Monajem ARS. The cruciate ligament of the knee joint. Anatomical, functional and experimental analysis. Clin Orthop Relat Res 1975:216-231. Ishibashi Y, Tsuda E, Tazawa K, Sato H. Intraoperative evaluation of the anatomical double-bundle anterior cruciate ligament reconstruction with the OrthoPilot navigation system. Orthopedics 2005;28:s1277-s1282 (suppl). Yagi M, Wong EK, Kanamori A, Debski RE, Fu FH, Woo SL-Y. Biomechanical analysis of an anatomic anterior cruciate ligament reconstruction. Am J Sports Med 2002;30:660-666.
Purpose: The objective of the study was to retrospectively compare the clinical outcomes of anatomic double-bundle anterior cruciate ligament reconstruction via hamstring tendons with singlebundle reconstruction between April 2002 and March 2004. Methods: We retrospectively reviewed 123 consecutive patients, 71 of whom underwent double-bundle reconstruction and 52 of whom underwent single-bundle reconstruction. The same postoperative rehabilitation protocol was used for all patients. The patients were followed up for a mean of 33 months. We evaluated manual knee laxity, anterior knee laxity as measured with the KT1000 arthrometer (MEDmetric, San Diego, CA), range of knee motion, isokinetic peak torque of knee extension and flexion strength adjusted for body weight as determined by Cybex testing (Lumex, Ronkonkoma, NY), and Lysholm score. Results: The Lachman test was negative in 64 cases (90%) and the pivot-shift test was negative in 62 cases (87%) in the double-bundle group. The Lachman test was negative in 45 cases (86%) and the pivot-shift test was negative in 42 cases (81%) in the single-bundle group. There was an extension deficit of greater than 5° in 19 cases (26%) in the double-bundle group and 6 cases (10%) in the single-bundle group (P ⬍ .05). The side-to-side difference in anterior tibial translation measured with the KT1000 arthrometer was 1.7 ⫾ 2.0 mm in the double-bundle group and 1.9 ⫾ 2.2 mm in the single-bundle group. The isokinetic peak torque of knee extension and flexion strength was 90% and 89%, respectively, in the double-bundle group and 87% and 86%, respectively, in the single-bundle group. The Lysholm score averaged 96.8 ⫾ 5.1 in the double-bundle group and 92.8 ⫾ 6.9 in the single-bundle group postoperatively. Conclusions: No significant difference was found between the 2 procedures with regard to manual knee laxity, anterior knee laxity measured by the KT1000 arthrometer, knee extension and flexion strength, and Lysholm score. In contrast, there was a significant difference in the range of knee motion between the 2 groups. The findings of our study do not support the routine adoption of double-bundle reconstruction. Level of Evidence: Level III, retrospective comparative study. Key Words: Anterior cruciate ligament—Double bundle—Anatomic reconstruction—Posterolateral bundle—Complications.
S
From Zenshukai Hospital, Gunma Sports Medicine Research Center (H.A., M.K., Y.K., M.T.), and Department of Orthopaedic Surgery, Gunma University Graduate School of Medicine (K.T.), Maebashi, Japan. The authors report no conflict of interest. Address correspondence and reprint requests to Hiroto Asagumo, M.D., Zenshukai Hospital, Gunma Sports Medicine Research Center, 1381 Ninomiya, Maebashi, Japan 379-2117. E-mail: [email protected] © 2007 by the Arthroscopy Association of North America 0749-8063/07/2306-6254$32.00/0 doi:10.1016/j.arthro.2007.01.009
602
everal authors have described double-bundle anterior cruciate ligament (ACL) reconstruction via hamstring tendons.1-12 Although normal knee laxity after ACL reconstruction is gained by use of bone– patellar tendon– bone graft, as well as hamstring tendons, the procedure using bone–patellar tendon– bone has some disadvantages, such as anterior knee pain and slow recovery of quadriceps muscle strength. There is concern regarding the current technique of ACL reconstruction in terms of residual laxity, especially in controlling rotation. To control rotatory instability, a number of authors have suggested reconstructing not just the anteromedial (AM) bundle but
Arthroscopy: The Journal of Arthroscopic and Related Surgery, Vol 23, No 6 (June), 2007: pp 602-609
ACL REPAIR VIA DOUBLE-BUNDLE HAMSTRING TENDON TABLE 1.
Age (yr) Sex (M/F) Mean preoperative period (mo) Mean follow-up period (mo) (range) Accompanying meniscal injury
603
Demographic Data
Single-Bundle Group
Double-Bundle Group
25.3 21/31 19 28 (24-39) MM in 10 (repair in 4 and meniscectomy in 2) and LM in 18 (repair in 1 and meniscectomy in 4)
24.2 33/38 17 29 (24-36) MM in 14 (repair in 7 and meniscectomy in 3) and LM in 35 (repair in 4 and meniscectomy in 7)
Abbreviations: MM, medial meniscus; LM, lateral meniscus.
also the posterolateral (PL) bundle.1-11 Adachi et al.5 and Hamada et al.2 indicated that there were no advantages for a double-bundle reconstruction as opposed to a single-bundle reconstruction in terms of anterior laxity. On the contrary, Muneta et al.11 and Yasuda et al.10 reported that the side-to-side anterior laxity of their double-bundle ACL reconstructions was significantly better than that of the single-bundle reconstruction. Since April 2000, we have performed double-bundle ACL reconstruction with the EndoButton device (Smith & Nephew Endoscopy, Andover, MA) for femoral-side fixation according to the method of Rosenberg and Graf13 to improve the results of reconstruction via semitendinosus (gracilis). In this study we compared the clinical outcome of anatomic doublebundle ACL reconstruction with single-bundle reconstruction in terms of manual knee laxity, range of knee motion, knee extension and flexion strength as determined by Cybex testing (Lumex, Ronkonkoma, NY), anterior knee laxity as measured by use of the KT1000 arthrometer (MEDmetric, San Diego, CA), and Lysholm score. Our hypothesis is that the anatomic double-bundle reconstruction is superior to the singlebundle procedure in terms of knee stability, especially with regard to rotatory instability. METHODS We retrospectively reviewed 123 consecutive patients with ACL-deficient knees who had undergone single- or double-bundle reconstruction between April 2002 and March 2004. Patients who underwent ACL reconstruction of both knees and those with an open physis were excluded from evaluation. Those with other ligamentous instability and those with articular cartilage lesions exceeding grade III were also excluded. The final study population consisted of 54 men and 69 women.
Double-bundle ACL reconstructions were performed in 71 consecutive patients (Table 1). The mean patient age was 24 years 2 months. The mean follow-up period was 2 years 5 months (range, 24 to 36 months). Single-bundle ACL reconstructions were performed in 52 consecutive patients. The mean patient age was 25 years 3 months. The mean follow-up period was 2 years 4 months (range, 24 to 39 months). At 1 year postoperatively, we performed secondlook arthroscopy, examining the status of the reconstructed ACL, meniscus, and articular cartilage and removing the staples from the tibia in all patients. At 1 year and 2 years postoperatively and at the final follow-up, we examined manual knee laxity, anterior tibial translation measured with the KT1000 knee arthrometer, range of knee motion, Lysholm score, and knee extension and flexion strength. Anterior tibial translation was measured with the KT1000 arthrometer with the knee in 20° of flexion, with an anterior force of 89 N being applied to the tibia. The maximum extension and flexion strength (adjusted for body weight) of both knees were measured by Cybex testing at 60°/s and are expressed as the percentage of the uninjured knee. Statistical analysis was performed with the unpaired t test, Mann-Whitney U test, and 2 test. The unpaired t test was used for anterior knee laxity measured with the KT1000 arthrometer and Lysholm score. The Mann-Whitney U test was used for manual knee laxity. The 2 test was used for range of knee motion, as well as knee extension and flexion strength. P ⬍ .05 was considered statistically significant. Surgical Procedure Double-Bundle Procedure: All surgeries were performed by use of the same procedure in each patient. After the knee was examined with the patient under general anesthesia, diagnostic arthroscopy was performed without an air tourniquet.
604
H. ASAGUMO ET AL.
FIGURE 1. Angle placement of guidewire: (A) axial view, (B) coronal view, and (C) sagittal view. The AM guidewire was set at 10° to 15° from the sagittal plane and 43° to 45° from the axial plane. The PL guidewire was set at 40° to 50° from the sagittal plane and 40° to 43° from the axial plane.
Semitendinosus tendons were harvested by use of a tendon harvester with a tourniquet after an approximately 4-cm-long longitudinal incision was made in the AM portion of the tibia 2 cm below the tibial tuberosity. If the diameter of either the AM or PL graft was smaller than 6 mm, we added the gracilis tendons. The angle of the tibial drill guide for the AM bundle was set at 43° to 45° from the axial plane and 10° to 15° from the sagittal plane. The angle of the tibial drill guide for the PL bundle was set at 40° to 43° from the axial plane and 40° to 50° from the sagittal plane (Fig 1). A Kirschner wire, 2.4 mm in diameter, was used as the
guidewire. The tibial guidewire for the AM bundle was positioned at the anterior portion of the natural ACL insertion, which is located 13 mm anterior to the anterior edge of the posterior cruciate ligament. The tibial guidewire for the PL bundle was placed 7 mm posterior and lateral to the AM guidewire (Fig 2). The tibial guidewire position was determined with an intraoperative 2-dimensional radiograph in the hyperextended position. The radiograph was checked according to anatomic landmarks to determine whether tibial guidewires were located within the normal ACL insertion. Two tibial tunnels were made with a cannulated drill corresponding to the measured diameter of
ACL REPAIR VIA DOUBLE-BUNDLE HAMSTRING TENDON
605
FIGURE 2. Tibial insertion of AM and PL bundles (right knee). The PL insertion is positioned at the posterior portion of the natural ACL insertion and immediately adjacent to the posterior cruciate ligament.
the prepared substitute. The diameter of the tibial tunnels was always 6 or 7 mm. To create the AM femoral tunnel in the lateral condyle, the femoral drill guide was positioned 5.5 mm anterior to the posterior edge of the intercondylar notch. The PL femoral guidewire was set lateral and posterior to the AM femoral tunnel with the knee in 90° of flexion (Fig 3). A Kirschner wire, 2.4 mm in diameter, and a cannulated drill (6 or 7 mm) were used in the same manner as for the tibial tunnels. The length of the femoral tunnels was always 25 to 30 mm, and the diameter of the femoral tunnels was always 6 or 7 mm. The length of each femoral drill hole was measured with a scaled probe (Depth Probe; Smith & Nephew Endoscopy). The AM bundle was positioned between 1:00 and 2:00 o’clock and the PL bundle around 3:00 o’clock on the lateral femoral condyle when reconstructing a left knee. The harvested semitendinosus tendon was cut in half and folded in a double-strand fashion. An EndoButton CL device (Smith & Nephew Endoscopy) was attached at the looped end, and a Telos artificial ligament (Telos, Marburg, Germany) was attached at the other end via a glove suture technique. The graft for the PL bundle was first introduced through the joint to the femoral drill hole by use of a Passing Pin (Smith & Nephew Endoscopy). The EndoButton was flipped and fixed on the femoral cortical surface. Then the graft for the AM bundle was introduced and fixed in the same manner. The staple was used for graft fixation on the tibial side at 30° of
FIGURE 3. Femoral insertion of AM and PL bundles (right knee). The PL insertion is positioned at the lowest point of the lateral femoral condyle.
knee flexion. Tibial fixation of the AM and PL bundles was carried out with 30 N of traction applied to each bundle (Figs 4-6). Postoperatively, a knee brace was used to immobilize the knee at 20° of flexion. Single-Bundle Procedure: Routine arthroscopic examination was performed by use of the same procedure as for the double-bundle reconstruction. The position of the tibial tunnel was at the center of the
FIGURE 4. Arthroscopic view showing grafted tendons for AM and PL bundles.
606
H. ASAGUMO ET AL. had recovered quadriceps strength to more than 60% of that in the uninjured knee. Sprinting and various competitive exercises were allowed 6 months postoperatively if the patient had recovered quadriceps strength to more than 80% of that in the uninjured knee. RESULTS Manual Knee Laxity In the double-bundle group the Lachman test was negative in 64 patients (90%), positive (⫹) in 5 (7%), and highly positive (⫹⫹) in 2 (3%) (Table 2). The pivot-shift test was negative in 62 patients (87%), positive (⫹) in 7 (10%), and highly positive (⫹⫹) in 2 (3%).
FIGURE 5. Postoperative radiograph. Two EndoButtons on the femoral cortex and four staples in the tibia are observed in the anteroposterior view.
ACL attachment and the position of the femoral tunnel was between 1:30 and 2:00 o’clock on the lateral femoral condyle when reconstructing a left knee. Quadrupled semitendinosus tendon was used for a grafted tendon, and gracilis was used if the semitendinosus tendon was insufficient. The tendon diameter was always 8 or 9 mm. Tibial fixation was carried out by use of staples with manual maximum traction at 30° of knee flexion. Postoperative Management The same postoperative rehabilitation protocol was used for all patients. Continuous passive motion was started 2 days postoperatively. Two weeks postoperatively, weight-bearing was started. Jogging was encouraged after 3 months postoperatively if the patient
FIGURE 6. Schema of anatomic reconstruction procedure of AM and PL bundles of ACL.
ACL REPAIR VIA DOUBLE-BUNDLE HAMSTRING TENDON TABLE 2.
Manual knee laxity Lachman test* ⫺ ⫹ ⫹⫹ Pivot-shift test* ⫺ ⫹ ⫹⫹ Anterior knee laxity (mm)* Range of knee motion Extension deficit† Flexion deficit* Isokinetic peak torque Extension strength* Flexion strength* Lysholm score* Preoperative Postoperative
Results
Single-Bundle Group
Double-Bundle Group
45 (86%) 6 (12%) 1 (2%)
64 (90%) 5 (7%) 2 (3%)
42 (81%) 7 (14%) 3 (5%) 1.9 ⫾ 2.2
62 (87%) 7 (10%) 2 (3%) 1.7 ⫾ 2.0
6 (10%) 3 (6%)
19 (26%) 5 (7%)
87% 86%
90% 89%
65.8 ⫾ 14.9 92.8 ⫾ 6.9
70.1 ⫾ 16.4 96.8 ⫾ 5.1
*P ⬎ .1. †P ⬍ .05.
607
Isokinetic Knee Extension and Flexion Strength by Cybex Testing The isokinetic peak torque of knee extension strength adjusted for body weight was 90% in the double-bundle group and 87% in the single-bundle group. The isokinetic peak torque of knee flexion strength adjusted for body weight was 89% in the double-bundle group and 86% in the single-bundle group. Lysholm Score The Lysholm score averaged 70.1 ⫾ 16.4 preoperatively and 96.8 ⫾ 5.1 postoperatively in the doublebundle group. In contrast, in the single-bundle group the score was 65.8 ⫾ 14.9 preoperatively and 92.8 ⫾ 6.9 postoperatively. Regarding the Lachman test, pivot-shift test, anterior knee laxity measured by use of the KT1000 arthrometer, knee extension and flexion strength, and Lysholm score, there was no significant difference between the single-bundle group and double-bundle group. In contrast, there was a significant difference in
In the single-bundle group the Lachman test was negative in 45 patients (86%), positive (⫹) in 6 (12%), and highly positive (⫹⫹) in 1 (2%). The pivot-shift test was negative in 42 patients (81%), positive (⫹) in 7 (14%), and highly positive (⫹⫹) in 3 (5%). Anterior Knee Laxity Measurements With KT-1000 Arthrometer In the double-bundle group the side-to-side difference in anterior tibial translation measured with the KT-1000 arthrometer was 1.7 ⫾ 2.0 mm. In the single-bundle group, the side-to-side difference was 1.9 ⫾ 2.2 mm. In 93% of patients in the double-bundle group and 90% of patients in the single-bundle group, the sideto-side difference ranged between 0 and 3 mm. Range of Knee Motion In the double-bundle group an extension deficit of more than 5° was found in 19 patients (26%) and a flexion deficit of more than 5° was found in 5 patients (7%). In the single-bundle group an extension deficit of more than 5° was found in 6 patients (10%) and a flexion deficit of more than 5° was found in 3 patients (6%).
FIGURE 7. Magnetic resonance image showing destruction of medial tibial plateau.
608
H. ASAGUMO ET AL.
FIGURE 8. Arthroscopic view showing damage to medial tibial plateau, as well as to PL bundle, caused by malpositioning of PL guidewire.
range of knee motion (extension deficit) between the 2 groups (P ⬍ .05). Meniscal lesions and articular cartilage lesions did not have an influence on clinical outcomes. Complications At second-look arthroscopic examination 1 year postoperatively, there were 5 cases showing PL bundle rupture only. Among 203 patients who underwent surgery between April 2002 and May 2005, we encountered 2 cases showing articular surface injuries of the medial tibial plateau, which was caused by malpositioning of the PL guidewire (Figs 7 and 8). When we create the tibial tunnel for the PL bundle via the transtibial tunnel technique, care is necessary to ensure that the angle of the Kirschner wire does not become more horizontal. There were no cases of artery or nerve injury and no signs of infection. DISCUSSION Several authors have reported on the double-bundle procedure via hamstring tendons.1-5,9,10,12,13 Recently, Yasuda et al.9 reported on anatomic reconstruction of the AM and PL bundles of the ACL by use of hamstring tendon grafts. Good clinical results were re-
ported with this method, and a new method of creating tibial and femoral tunnels for the PL bundle was introduced. The ACL can be anatomically divided into AM and PL portions, each of which shows different tension patterns.4,14-16 Theoretically, double-bundle reconstruction has several advantages over single-bundle reconstruction with a multistranded tendon with respect to regaining a structure that closely resembles the normal ACL. First, a double-bundle ACL has wider contact areas between the bone and grafted tendon than a singlebundle ACL via hamstring tendons, which means that only the margin of the tendon graft is anchored to the collagen fibers, resembling Sharpey’s fibers, and the tunnel wall in ACL reconstruction.7 Yagi et al.17 suggested biomechanical advantages of the double-bundle procedure compared with the single-bundle procedure in an in vitro study. Second, double-bundle ACL reconstruction allows AM and PL bundles to be created anatomically and independently. We compared the surgical outcomes of doublebundle ACL reconstruction with those of single-bundle ACL reconstruction performed between April 2002 and March 2004. We have found a significant number of flexion contractures 1 year postoperatively compared with the single-bundle procedure. In all cases flexion contractures were improved by performing notchplasty or removing Cyclops lesions during second-look arthroscopy. Our data showed that there was no significant difference in terms of knee stability, especially in terms of rotatory instability. Producing anatomic PL grafts did not affect the pivot-shift test in the double-bundle group. We encountered 2 cases showing articular surface injuries of the medial tibial plateau at the beginning of the anatomic double-bundle procedure. The complication of medial tibial plateau injury caused by malpositioning of the PL guidewire has not been described, to our knowledge, after the double-bundle procedure. Actually, we could not detect the articular surface injury of the medial tibial plateau immediately after overdrilling the Kirschner wire for the PL bundle. In both cases the patients have no clinical symptoms, such as pain, swelling, or giving way. We first detected this complication on follow-up magnetic resonance imaging 1 year postoperatively, and we confirmed the articular surface injury on second-look arthroscopy. We thought that, first, the subchondral bone was destroyed by overdrilling for the PL bundle and, second, the fragile articular surface was gradually
ACL REPAIR VIA DOUBLE-BUNDLE HAMSTRING TENDON punctured by the PL bundle. When we create the tibial tunnel for the PL bundle via the transtibial tunnel technique, care is necessary to prevent the angle of the Kirschner wire from becoming more horizontal. In this study we encountered 5 cases showing PL bundle rupture only on second-look arthroscopic examination 1 year postoperatively. It was thought that extension deficit and PL bundle rupture were related closely. The length pattern of the PL bundle is more tense in extension in an anatomic PL position than in the conventional PL position of the femoral tunnel. Tensioning the PL bundle near full extension makes theoretic sense because of the tension patterns in this bundle through a range of motion. The limitation of our study is that the follow-up period was only 2 years. To establish the efficacy of this double-bundle reconstruction, further prospective study with a longer follow-up period is definitely needed. CONCLUSIONS This study compared the results of double-bundle ACL reconstruction with those of single-bundle reconstruction and confirmed that there was no significant difference, with the exception of range of knee motion, between the 2 procedures. An extension deficit of greater than 5° was found in 19 cases (26%) in the double-bundle group and 6 cases (10%) in the single-bundle group (P ⬍ .05). The findings of our study do not support the routine adoption of the double-bundle procedure until the risk-benefit comparison to single-bundle reconstruction warrants the change.
3. 4.
5.
6.
7.
8. 9.
10.
11.
12.
13. 14. 15.
REFERENCES 16. 1. Bellier G, Christel P, Colombet P, Djian P, Franceschi JP, Sbihi A. Double-stranded hamstring graft for anterior cruciate ligament reconstruction. Arthroscopy 2004;20:890-894. 2. Hamada M, Shino K, Horibe S. Single- versus bi-socket anterior cruciate ligament reconstruction using autogenous multiple-stranded hamstrings tendons with EndoButton for
17.
609
femoral fixation: A prospective study. Arthroscopy 2001;17: 801-807. Mott HW. Semitendinosus anatomic reconstruction for cruciate ligament insufficiency. Clin Orthop Relat Res 1983:90-92. Muneta T, Sekiya I, Yagishita K, Ogiuchi T, Yamamoto H, Shinomiya K. Two-bundle anterior cruciate ligament reconstruction using semitendinosus tendon with endobuttons: Operative technique and preliminary results. Arthroscopy 1999; 15:618-623. Adachi N, Ochi M, Uchio Y, Iwasa J, Kuriwaka M, Ito Y. Reconstruction of the anterior cruciate ligament. Single- versus double-bundle multistranded hamstring tendons. J Bone Joint Surg Br 2004;86:515-520. Miura K, Woo SL-Y, Brinkley R, Fu Y-C, Noorani S. Effects of knee flexion angles for graft fixation on force distribution in double-bundle anterior cruciate ligament grafts. Am J Sports Med 2006;34:577-585. Tomita F, Yasuda K, Mikami S, Sakai T. Comparisons of intraosseous graft healing between the doubled flexor tendon graft and the bone-patellar tendon-bone graft in anterior cruciate ligament reconstruction. Arthroscopy 2001;17:461-476. Zelle BA, Brucker PU, Feng MT, Fu FH. Anatomical doublebundle anterior cruciate ligament reconstruction. Sports Med 2006;36:99-108. Yasuda K, Kondo E, Ichiyama H, et al. Anatomic reconstruction of the anteromedial and posterolateral bundles of the anterior cruciate ligament using hamstrings tendon grafts. Arthroscopy 2004;20:1015-1025. Yasuda K, Kondo E, Ichiyama H, Tanabe Y, Tohyama H. Clinical evaluation of anatomic double-bundle anterior cruciate ligament reconstruction procedure using hamstring tendon grafts: Comparisons among 3 different procedures. Arthroscopy 2006;22:240-251. Muneta T, Koga H, Morito T, Yagishita K. A retrospective study of the midterm outcome of two-bundle anterior cruciate ligament reconstruction using quadrupled semitendinosus tendon in comparison with one-bundle reconstruction. Arthroscopy 2006;22:252-258. Mae T, Shino K, Miyama T, et al. Single- versus two-femoral socket anterior cruciate ligament reconstruction technique: Biomechanical analysis using a robotic simulator. Arthroscopy 2001;17:708-716. Rosenberg TD, Graf B. Techniques for ACL reconstruction with Multi-Trac drill guide. Mansfield, MA: Acufex Microsurgical, 1994. Amis A, Dawkins G. Functional anatomy of anterior cruciate ligament. J Bone Joint Surg Br 1991;73:260-267. Girgis FG, Marshall JL, Monajem ARS. The cruciate ligament of the knee joint. Anatomical, functional and experimental analysis. Clin Orthop Relat Res 1975:216-231. Ishibashi Y, Tsuda E, Tazawa K, Sato H. Intraoperative evaluation of the anatomical double-bundle anterior cruciate ligament reconstruction with the OrthoPilot navigation system. Orthopedics 2005;28:s1277-s1282 (suppl). Yagi M, Wong EK, Kanamori A, Debski RE, Fu FH, Woo SL-Y. Biomechanical analysis of an anatomic anterior cruciate ligament reconstruction. Am J Sports Med 2002;30:660-666.