Anterior cruciate ligament (ACL) injuries are increasingly common, with 250 000 cases occurring annually in the United States.1,2 Typically, ACL injuries affect people in their prime working time of life; the greatest incidence occurring in 15- to 25-year-old participants of work and pivoting sports.3,4 Work-related injuries and accidents are also common.5 In either case, consequences include lost time from work and a lower quality of life.6
Anatomic ACL reconstruction surgery has evolved from the use of single-bundle to double-bundle (DB) tendon autografts. Double-bundle grafts, intended to address the complex 2-bundled anatomy and biomechanical function of the ACL, attempt to restore knee rotational stability, which arguably cannot be achieved consistently with single-bundle reconstructions.7
Randomized clinical trials (RCTs) comparing single-bundle and DB ACL reconstruction evaluate clinical stability and function, and some evaluate patient-reported outcomes8–14 in small sample sizes ranging from 40 patients13 to 108 patients.11 However, none of these trials include a disease-specific quality-of-life patient-reported outcome. Furthermore, none of the RCTs double blind the independent examiners and patients to avoid biases during outcome evaluation.9,15
Most primary ACL reconstruction studies are low-level evidence.16 Four published meta-analyses compared single-bundle and DB ACL reconstruction.17–20 Two separate meta-analyses of 19 RCTs concluded that, relative to single-bundle ACL reconstructions, the DB technique results in significantly better anterior and rotational stability and higher International Knee Documentation Committee (IKDC) Objective Scores, but no differences in Lysholm, Tegner activity and subjective IKDC outcomes.18,20 The third meta-analysis of 17 RCTs reported lower KT-1000 arthrometer measurements and extension deficits and higher subjective IKDC scores for DB reconstructions.17 A Cochrane systematic review, including 17 RCT and quasi-RCTs, concluded no statistically or clinically significant differences between single-bundle and DB reconstructions, based on subjective measures (IKDC, Tegner activity, and Lysholm scores) in the intermediate (6-month to 2-year) and long-term (2- to 5-year) post-operative periods.19 Long-term studies showed significant differences favoring the DB reconstruction for IKDC Objective Scores, KT-1000 arthrometer, and the pivot shift measurements.19 This review concluded that there was insufficient evidence to determine the superiority of single-bundle versus DB reconstructions and that “high quality, large and appropriately reported randomized controlled trials of double-bundle versus single-bundle reconstruction for anterior cruciate ligament rupture in adults appears justified.”19
This prospective, double-blind RCT is intended to provide this high-quality Level 1 evidence, comparing 3 anatomically positioned reconstructive surgical autograft options for primary ACL deficiency. Patient-reported, disease-specific quality-of-life outcome was measured at a minimum of 2-year postoperative follow-up. Secondary objectives assessed the 2-year clinical and functional outcomes following each type of reconstruction.
All patients (14-50 years old) presenting to the senior author at the University of Calgary Sport Medicine Centre from 2007 to 2010, with confirmed isolated ACL deficiency, were seen by the orthopedic research team. Clinical eligibility was determined using defined inclusion and exclusion criteria (Table 1). Ineligible patients were treated and followed accordingly. The principal investigator discussed the trial with each eligible patient and answered all study-related questions. Each patient was informed that, in addition to the independent examiner (who did not attend the surgery), they would be blinded to the reconstruction technique until the 2-year postoperative visit. The independent examiner obtained written informed consent.
Patients were evaluated preoperatively at baseline (≤3 months of surgery) and postoperatively at 3 and 6 months, 1 and 2 years. The independent trained examiner performed standardized bilateral knee examination and tests at each scheduled study visit: knee effusion and patella-femoral crepitus, anterior and posterior drawer, pivot and reverse pivot shift, valgus and varus stress, external rotation (30 and 90 degrees), McMurray, and knee range-of-motion using a long-arm goniometer.
Randomization and Surgical Techniques
In the operating room under general anesthesia, an examination under anesthesia and a diagnostic arthroscopy were performed on all patients to confirm study eligibility (Table 1). Thirty-seven patients were excluded intraoperatively (Figure 1) because of grade 4 chondral damage (n = 15) or ACL preservation surgery (n = 22).21
Three hundred thirty eligible patients were intraoperatively and randomly allocated to 1 of the 3 treatment groups (110 patients per group; Figure 1). Computer-generated allocation was performed using varied block sizes and stratification by injury chronicity (acute: ≤3 months) to 1 ACL reconstruction technique with: (1) patellar tendon (PT), (2) quadruple-stranded hamstring tendon (HT), or (3) DB using HTs autograft.
The operative techniques for each group were identical except for the type of the autograft tissue and the utilization of 2 tunnels drilled in the tibia and femur for the DB technique. An endoscopic 1-incision technique using EndoButton (Smith and Nephew, Andover, Massachusetts) suspensory fixation on the femoral side and bioabsorbable interference screw (Bioscrew; Linvatec, Inc., Largo, Florida) fixation on the tibial side were performed. For all groups, a universal 5- to 7-cm anteromedial vertical incision was made for graft harvesting (Figure 2). The tibial and femoral tunnels were centered in the anatomic footprint of the ACL after removing the remaining ACL tissue. Standard femoral tunnel preparation was performed taking into account the size of the autografts. Patients were treated on an outpatient basis.22
Anterior Cruciate Ligament Reconstruction Using the Patellar Tendon
A standard bone-PT-bone autograft (9-11 mm) was harvested. The tibial tunnel was drilled with a standard endoscopic guide pre-set at 50 degrees, and the femoral tunnel was drilled using the trans-tibial approach with a standard femoral over-the-top guide. The femoral plug was undersized by 0.5 to 1.0 mm to allow for easy passage into the femoral tunnel. The femoral plug was attached to the EndoButton (Smith and Nephew) with 5.0-mm mersylene tape. The tibial plug was attached to nonabsorbable sutures. Tensioning of the graft was performed in 10 to 20 degrees of flexion.
Anterior Cruciate Ligament Reconstruction Using Quadruple Semitendinosus/Gracilis Tendons (HT)
The HTs were harvested, then looped, and sutured together, creating a 4-bundle hamstring graft. The looped end was attached to the EndoButton (Smith and Nephew) and the free ends to nonabsorbable sutures. The tibial tunnel was drilled with a standard guide pre-set at 50 degrees, and the femoral tunnel drilled using the trans-tibial approach and a standard femoral over-the-top guide. If it was not possible to place the femoral guide pin in the centre of the anatomic femoral footprint from the trans-tibial approach, the anteromedial portal was used. Tensioning of the graft was performed in 10 to 20 degrees of flexion.
Double-Bundle Anterior Cruciate Ligament Reconstruction Using 2-Stranded, Semitendinosus, and Gracilis Tendons (DB)
The semitendinosus and gracilis tendons were doubled separately to form the graft for the anteromedial (AM) and posterolateral (PL) bundles, respectively. The looped ends were attached to EndoButtons (Smith and Nephew) and the free ends to nonabsorbable sutures. The Smith and Nephew Acufex Director Set (Smith and Nephew) DB guide was used on both the tibia and femur to ensure separate tunnels were achieved and to minimize convergence. The anteromedial portal was used for femoral tunnel drilling in all cases. The posterolateral bundle was tensioned in 10-20 degrees of flexion, and the anteromedial bundle in 45 degrees of flexion.23,24 Bioscrews (Linvatec Inc.) were used in all AM tunnels for tibial fixation. A minority of PL tunnels had smaller diameters requiring Bio-Tenodesis screws (Athrex, Inc., Naples, Florida).
All treatment groups followed identical postoperative and rehabilitation protocols.25 Minor surgical deviations were reported within the context of re-injuries and failures in a separate manuscript (Mohtadi N, et al. Re-ruptures, re-injuries and revisions at a minimum two years follow-up, submitted CJSM July 2014).
In addition to the standardized knee examination, outcomes were measured at baseline, 3 and 6 months, 1 and 2 years postoperatively.
The 32-item patient-reported anterior cruciate ligament quality-of-life (ACL-QOL) questionnaire assesses symptoms/physical complaints, work-related, sports/recreational, lifestyle, and social/emotional concerns. A higher score on the 0- to 100-mm visual analog scale response format represents better quality of life.6
Knee laxity measurements were determined by KT-1000 arthrometer (30 lbs/134 N), millimeter (mm) side-to-side differences (SSD),25,26 pivot shift grades (equal/0; glide/1; clunk/2; gross/3), effusion (none/mild/moderate/severe), passive flexion and extension range of motion (in degrees), IKDC Subjective scores (out of 100) and Objective Overall Group grades (A = normal, B = nearly normal, C = abnormal, D = severely abnormal),27,28 the Cincinnati Occupational Rating Scale (out of 100),29,30 Tegner Activity Levels,31 single-leg hop (percent of opposite side),32 and mean “skin-to-skin” surgical times for each procedure. Surgical times included the diagnostic arthroscopy and eligibility determinations. The incidences and risks of traumatic re-injuries, atraumatic graft failures, and contralateral ACL tears were also evaluated as part of this RCT but have been reported in a separate manuscript (Mohtadi N, et al. Re-ruptures, re-injuries and revisions at a minimum two years follow-up, submitted CJSM July 2014).
The independent examiner and the patients were unblinded to treatment allocation at 2 years postoperatively. The effectiveness of blinding to treatment allocation was measured as the proportion of correct guesses of graft type by the patients and the examiner before unblinding.
Sample Size Calculations
The sample size calculation was based on 88 ACL-deficient patients with surgical intervention and an ACL-QOL score of 74.5 (SD = 20.1) at a mean 39-month follow-up25; minimal clinically important difference of 10 points; power = 0.80; and P = 0.05. A Bonferroni adjustment for multiple comparisons resulted in 90 patients per group. The sample size, with a 20% lost-to-follow-up rate, was 108 patients per group, for a total of 324 patients. Because recruitment occurred in the clinic, with final eligibility determined intraoperatively, 330 patients were ultimately randomized.
All patients were analyzed on an “intention-to-treat” basis (ie, data analysis based on the initial treatment allocation and not on the treatment received) using a 5% significance level for all analyses. Adjusted Bonferroni comparisons and repeated-measures analyses, using a mixed-model analysis of variance for treatment group over time of assessment, were performed on the continuous variables. Chi-square analysis was used to compare the groups for categorical data. IBM-SPSS Statistics 20 software (SPSS, Chicago, Illinois) was used.
The last observation carried forward (LOCF) strategy was used to calculate missing 2-year IKDC Objective grades (based on KT-1000 arthrometer, range-of-motion, pivot shift, and effusion measurements) in 6 patients who had no documentation of injury throughout the 2-year follow-up period. One-year KT-1000 arthrometer and range-of-motion measurements were carried forward for the missing 2-year data. One-year pivot shift data were carried forward if the 6-month and 1-year grades were consistent. Effusion data were carried forward if the patient did not have positive effusion at 1 year. For positive effusion, data were carried forward if the 6-month and 1-year data were consistent.
Exploratory subgroup analyses were performed on the ACL-QOL, IKDC Objective Scores, pivot shift and KT-1000 Arthrometer, after excluding patients who had suffered traumatic re-injuries to the operative knee or contralateral ACL tears.
This RCT (NCT00529958) was approved by the University of Calgary Conjoint Health Research Ethics Board.
Demographics (Table 2), baseline meniscal, and chondral conditions between the groups (Table 3) were not different. Three hundred twenty-two patients (98%) completed a minimum of 2-year follow-up. Three PT patients withdrew after 6-months; 5 patients were lost to follow-up at 2 years (PT = 1, HT = 2, DB = 2).
Anterior cruciate ligament quality-of-life scores (Table 4) increased significantly from baseline to 2 years for all groups (P < 0.0001), but the change in scores over time was not statistically different between groups (P = 0.09; Figure 3). Mean 2-year ACL-QOL scores were not different between groups (P = 0.591; Table 4).
Pivot shift grades between the groups at all periods were not different (Table 5). The proportion of patients with a pivot shift ≥grade 2 at 2 years was not different between groups (PT = 14%; HT = 18%; DB = 19%; P = 0.573).
The IKDC Objective group grades were not statistically different between groups at all periods (Table 6). At 2 years, normal/nearly normal knees were not significantly different between groups: PT = 78%; HT = 73%; DB = 71% (P = 0.479). The IKDC Subjective scores (Table 7) increased significantly from baseline to 2 years for all groups (P < 0.0001). The HT group had consistently higher scores at all periods, but the scores were not different between groups at 2 years (P = 0.821; Table 7).
The SSD KT-1000 arthrometer measurements decreased significantly from baseline at 1 year (P = 0.0001) and 2 years (P = 0.0001) for all groups (Table 8). These changes were statistically different between the PT and HT (P = 0.005), and PT and DB (P = 0.04) groups only. At 1 and 2 years of follow-up, mean KT-1000 arthrometer SSD measurements were statistically significant between the PT and HT groups, and PT and DB groups (Table 8). There were no differences between groups for KT-1000 arthrometer as reported by IKDC Objective grades at all periods (Table 9). The proportion of patients with normal/nearly normal knees based on ≤5-mm KT-1000 arthrometer SSD measurements (PT = 93%; HT = 88%; DB = 85%) was not statistically different between groups (P = 0.173).
Passive range-of-motion measurements (Tables 10 and 11), Cincinnati Occupational Rating scores (Table 12), and Tegner Activity Levels (Tables 13 and 14) and the single-leg hop test (Table 15) were not different between groups.
Surgical times were statistically different between: PT and HT (P = 0.001), PT and DB (P = 0.001), and HT and DB (P = 0.001) reconstructions (Table 16).
In patients without traumatic re-injuries and/or contralateral injuries, there were no differences in ACL-QOL, pivot shift, IKDC Objective Scores or KT-1000 arthrometer measurements between groups (Table 17).
Upon unblinding, patients and the independent examiner correctly guessed the allocated graft type 51% and 46% of the time, respectively (Table 18).
This trial represents the largest RCT comparing ACL surgical techniques. Patients in all 3 groups improved from baseline for all outcomes. The PT reconstructions were less likely to have re-injuries and had statically tighter knees compared with HT and DB reconstructions (reported separately: Mohtadi N, et al. Re-ruptures, re-injuries and revisions at a minimum two years follow-up, submitted CJSM July 2014).
Trial strengths include the description of the patient population, a high volume of screened patients (n = 745), an explanation of the enrolment process, and definition of eligible and ineligible patients for the trial. The recruitment rate for this surgical trial is very high: 84% of eligible patients consented to the trial, with negligible loss-to-follow-up at 2 years (2%). Modern web-based, computer-generated randomization techniques for allocation concealment and the double blinding of the patients and independent examiner minimized bias. Additionally, nursing staff outside of the operating room, clinic staff, and physiotherapists involved in postoperative care were also blinded. Data collection was standardized with intraoperative and clinic-based report forms. The research assistant performing the randomization documented all intraoperative data. Randomization was independent of the surgeon and the independent examiner. Finally, a validated and disease-specific patient-reported outcome assessment was used with several secondary outcomes of specific interest to this surgery.
Trial limitations primarily relate to the technical aspects of performing differing techniques and whether the surgeon (N.M.) was able to perform each procedure with equal facility. However, these limitations were circumvented by having a surgeon with more than 20 years of previous experience with ACL reconstructive surgery. Every effort was made to ensure that technical expertise with the DB technique was addressed. This surgeon spent 2 years researching, learning, and subsequently reporting on the learning curve with DB ACL reconstructions before entering patients into the trial (oral communication: Mohtadi N. Double Bundle ACL reconstruction: The Learning Curve. Fowlers' Fellows Meeting 2008; Emerald Lake, BC). During the trial, this surgeon performed an average of 200 primary and revision ACL reconstructions per year with both HT and PT procedures. To ensure consistency, all grafts were prepared by the same surgical assistant. Before the trial, this surgical assistant prepared more than 5000 tendon grafts for ACL reconstruction throughout her 20 years of experience with multiple surgeons.
One 21-year-old female patient (body mass index = 23.4 kg/m2) crossed over from the DB allocated group to the HT group because of insufficient semitendinosus/gracilis tendon length and diameter for separate bundles in a DB reconstruction. The subsequent 4-standed HT graft diameter was 7.5 mm. Analyzed on an intention-to-treat basis, her 2-year ACL-QOL score was 86, indicating she was doing better than the 95% confidence interval for the DB group.
Subgroup analyses were performed to address the possible criticism that patients with re-injuries would negatively bias the outcomes. Excluding patients with re-injuries or contralateral tears resulted in no outcome differences between groups.
The LOCF strategy of carrying forward 1-year physical examination data for missing 2-year data was applied in only 6 patients (3 PT, 3DB) and only for the 2-year IKDC Objective outcome. All other reported outcomes were analyzed with the missing data. Because all 6 patients demonstrated consistency between their 6-month and 1-year physical examination measurements, were not lost to follow-up, had no record of re-injury or graft failure, and demonstrated improvement in ACL-QOL scores from 1- to 2-year follow-up, the authors agreed that it was valid to assume that the physical examination measurements remained consistent between 1- and 2-year follow-up. As a result of carrying the 1-year physical examination measurements forward, inferred 2-year IKDC Objective Scores were reported for these 6 patients (4 nearly normal knees, 2 abnormal knees) instead of having missing data. There were no measurable, statistical, or clinical differences between the 1-year and 2-year KT-1000 arthrometer data. Sensitivity analyses were performed for the IKDC Objective outcome at 2 years with and without the imputed missing data. The IKDC Objective Score at 2 years showed no statistically significant differences between study groups (P = 0.630) using the non-imputed data compared with the analysis using the LOCF method (P = 0.658), as shown in Table 6. Analysis of the data as 2 categories (ie, normal/nearly normal and abnormal/severely abnormal) showed no statistically significant differences between groups using the non-imputed data (P = 0.557) compared with the analysis using the LOCF method (P = 0.479).
It would be reasonable to conclude that the procedures are equivalent because the primary outcome (ACL-QOL) showed no significant differences between groups. However, considering the extra time, effort, required technical training, and cost differences, the DB reconstruction cannot be recommended over the HT or PT procedures. Although specific economic analyses were beyond the mandate of this trial, the DB technique has obvious incremental costs—twice the graft fixation costs and time-related costs of a longer procedure. With all other aspects of this trial being comparable between the 3 groups, the DB reconstruction may be similarly effective but is not recommended from an economic standpoint. Furthermore, actual cost-effectiveness should be determined based on long-term outcomes of a procedure.33,34 Therefore, the cost-effectiveness of single-bundle versus DB ACL reconstructions would be more appropriately determined at the 5- and 10-year follow-up of this trial.
This trial does present some differences compared with the literature. The most comparable information is a Cochrane review of 1433 patients from 17 randomized and quasi-randomized trials.19 This review found no differences in subjective functional knee scores between the single-bundle and DB reconstructions. Contrary to the current study, the review showed statistically improved static stability for DB reconstructions, as measured by KT-1000 arthrometer, pivot shift examination, and normal/nearly normal IKDC classifications.19 However, none of the studies in the review compared PT reconstruction in a single-bundle reconstruction context. Additionally, Cochrane reviews and the other meta-analyses are limited by studies that demonstrate high or unclear risks of bias, may be clinically or methodologically heterogeneous, and report different outcome measures. A recent RCT comparing single-bundle and DB hamstring autograft reconstructions, using a similar protocol, showed no differences in patient-reported outcomes, objective measures, or pivot shift.35 A lower percentage of grade 2 pivot shift was reported in both groups,35 which may be a reflection of the assessor performing the examination. The current trial eliminated this risk of bias by blinding the examiner to group allocation, ensuring independent examination of all patients, and demonstration of active clinical equipoise with no vested interest in the outcome of any group.
At 2 years, there was no difference in disease-specific quality-of-life outcome or IKDC grades between the 3 techniques for ACL reconstruction. Patellar tendon reconstructions had significantly lower side-to-side differences on static stability measures. Blinding of the patients and evaluator was achieved.
The authors acknowledge Kristie More for suggesting that double-blinding methodology be used for this study, Jocelyn Fredine for assisting with follow-up and data collection, Niko Lagumen for assisting with intraoperative randomization and data collection. The authors also acknowledge the surgical staff at the Peter Lougheed Hospital.
1. Giugliano DN, Solomon JL. ACL tears in female athletes. Phys Med Rehabil Clin N Am. 2007;18:417–438, viii.
2. Silvers HJ, Mandelbaum BR. Prevention of anterior cruciate ligament injury in the female athlete. Br J Sports Med. 2007;41(suppl 1):i52–i59.
3. Griffin LY, Agel J, Albohm MJ, et al.. Noncontact anterior cruciate ligament injuries: risk factors and prevention strategies. J Am Acad Orthop Surg. 2000;8:141–150.
4. Sutton KM, Bullock JM. Anterior cruciate ligament rupture: differences between males and females. J Am Acad Orthop Surg. 2013;21:41–50.
5. Feller JA, Webster KE. A randomized comparison of patellar tendon and hamstring tendon anterior cruciate ligament reconstruction. Am J Sports Med. 2003;31:564–573.
6. Mohtadi N. Development and validation of the quality of life outcome measure (questionnaire) for chronic anterior cruciate ligament deficiency. Am J Sports Med. 1998;26:350–359.
7. Steckel H, Starman JS, Baums MH, et al.. Anatomy of the anterior cruciate ligament double bundle structure: a macroscopic evaluation. Scand J Med Sci Sports. 2007;17:387–392.
8. Nunez M, Sastre S, Nunez E, et al.. Health-related quality of life and direct costs in patients with anterior cruciate ligament injury: single-bundle versus double-bundle reconstruction in a low-demand cohort—a randomized trial with 2 years of follow-up. Arthroscopy. 2012;28:929–935.
9. Suomalainen P, Jarvela T, Paakkala A, et al.. Double-bundle versus single-bundle anterior cruciate ligament reconstruction: a prospective randomized study with 5-year results. Am J Sports Med. 2012;40:1511–1518.
10. Xu Y, Ao YF, Wang JQ, et al.. Prospective randomized comparison of anatomic
single- and double-bundle anterior cruciate ligament reconstruction. Knee Surg Sports Traumatol Arthrosc. 2014;22:308–316.
11. Zhang Z, Gu B, Zhu W, et al.. Double-bundle versus single bundle anterior cruciate ligament reconstructions: a prospective, randomized study with 2-year follow-up. Eur J Orthop Surg Traumatol. 2013;24:559–565.
12. Streich NA, Friedrich K, Gotterbarm T, et al.. Reconstruction of the ACL with a semitendinosus tendon graft: a prospective randomized single blinded comparison of double-bundle versus single-bundle technique in male athletes. Knee Surg Sports Traumatol Arthrosc. 2008;16:232–238.
13. Volpi P, Cervellin M, Denti M, et al.. ACL reconstruction in sports active people: transtibial DB technique with ST/G vs. transtibial SB technique with BPTB: preliminary results. Injury. 2010;41:1168–1171.
14. Aglietti P, Giron F, Losco M, et al.. Comparison between single-and double-bundle anterior cruciate ligament reconstruction: a prospective, randomized, single-blinded clinical trial. Am J Sports Med. 2010;38:25–34.
15. Lui PP, Cheng YY, Yung SH, et al.. A randomized controlled trial comparing bone mineral density changes of three different ACL reconstruction techniques. Knee. 2012;19:779–785.
16. Samuelsson K, Desai N, McNair E, et al.. Level of evidence in anterior cruciate ligament reconstruction research: a systematic review. Am J Sports Med. 2013;41:924–934.
17. Li X, Xu CP, Song JQ, et al.. Single-bundle versus double-bundle anterior cruciate ligament reconstruction: an up-to-date meta-analysis. Int Orthop. 2013;37:213–226.
18. Li YL, Ning GZ, Wu Q, et al.. Single-bundle or double-bundle for anterior cruciate ligament reconstruction: A meta-analysis. Knee. 2014;21:28–37.
19. Tiamklang T, Sumanont S, Foocharoen T, et al.. Double-bundle versus single-bundle reconstruction for anterior cruciate ligament rupture in adults. Cochrane Database Syst Rev. 2012;11:CD008413.
20. Xu M, Gao S, Zeng C, et al.. Outcomes of anterior cruciate ligament reconstruction using single-bundle versus double-bundle technique: meta-analysis of 19 randomized controlled trials. Arthroscopy. 2013;29:357–365.
21. Naylor A, Mohtadi N, Chan D, et al.. Anterior cruciate ligament preservation during reconstructive surgery: does the extra surgical effort improve patient outcomes at one year? Sport Orthop Traumatol. 2013;29:29–39.
22. Krywulak SA, Mohtadi NG, Russell ML, et al.. Patient satisfaction with inpatient versus outpatient reconstruction of the anterior cruciate ligament: a randomized clinical trial. Can J Surg. 2005;48:201–206.
23. Bellier G, Christel P, Colombet P, et al.. Double-stranded hamstring graft for anterior cruciate ligament reconstruction. Arthroscopy. 2004;20:890–894.
24. Yasuda K, Kondo E, Ichiyama H, et al.. 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.
25. Grant JA, Mohtadi NG, Maitland ME, et al.. Comparison of home versus physical therapy-supervised rehabilitation programs after anterior cruciate ligament reconstruction: a randomized clinical trial. Am J Sports Med. 2005;33:1288–1297.
26. Maitland ME, Bell GD, Mohtadi NG, et al.. Quantitative analysis of anterior cruciate ligament instability. Clin Biomech (Bristol, Avon). 1995;10:93–97.
27. Hefti F, Muller W, Jakob RP, et al.. Evaluation of knee ligament injuries with the IKDC form. Knee Surg Sports Traumatol Arthrosc. 1993;1:226–234.
28. Irrgang JJ, Anderson AF, Boland AL, et al.. Development and validation of the international knee documentation committee subjective knee form. Am J Sports Med. 2001;29:600–613.
29. Barber-Westin SD, Noyes FR, McCloskey JW. Rigorous statistical reliability, validity, and responsiveness testing of the Cincinnati knee rating system in 350 subjects with uninjured, injured, or anterior cruciate ligament-reconstructed knees. Am J Sports Med. 1999;27:402–416.
30. Noyes FR, Barber SD, Mooar LA. A rationale for assessing sports activity levels and limitations in knee disorders. Clin Orthop Relat Res. 1989:238–249.
31. Tegner Y, Lysholm J. Rating systems in the evaluation of knee ligament injuries. Clin Orthop Relat Res. 1985:43–49.
32. Irrgang JJ, Anderson AF, Boland AL, et al.. Responsiveness of the international Knee Documentation Committee Subjective Knee Form. Am J Sports Med. 2006;34:1567–1573.
33. Fu FH, Kowalchuk D. Cost analysis comparing single-bundle and double-bundle anterior cruciate ligament (ACL) reconstruction. Am J Sports Med. 2009;37:e1; author reply e2.
34. Paxton ES, Kymes SM, Brophy RH. Cost-effectiveness of anterior cruciate ligament reconstruction: a preliminary comparison of single-bundle and double-bundle techniques. Am J Sports Med. 2010;38:2417–2425.
35. Ahlden M, Sernert N, Karlsson J, et al.. A prospective randomized study comparing double- and single-bundle techniques for anterior cruciate ligament reconstruction. Am J Sports Med. 2013;41:2484–2491.
36. International Cartilage Repair Society (ICRS). ICRS Cartilage Injury Evaluation Package. [ICRS web site]. April 28, 2000. http://www.cartilage.org
. Accessed April 22, 2014.