Irradiated Hamstring Tendon Allograft Versus Autograft for Anatomic Double-Bundle Anterior Cruciate Ligament Reconstruction

Abstract

Background:

Most studies on grafts for anterior cruciate ligament (ACL) reconstruction (ACLR) have been of autografts or nonirradiated allografts with a single-bundle (SB) technique. Outcome reports evaluating anatomic double-bundle (DB) ACLR with a hamstring tendon autograft versus irradiated allograft are rare.

Purpose:

To compare the clinical outcomes of arthroscopic anatomic DB ACLR with a hamstring tendon autograft versus irradiated allograft.

Study Design:

Randomized controlled trial; Level of evidence, 2.

Methods:

Between 2008 and 2009, a total of 107 patients undergoing arthroscopic DB ACLR were prospectively randomized consecutively into 1 of 2 groups (autograft [Auto] group and irradiated allograft [Ir-Allo] group). All the surgical procedures were performed by the same senior surgeon using the DB reconstruction technique. All irradiated hamstring tendon allografts were sterilized with 2.5 Mrad of irradiation before distribution and were obtained from a single certified tissue bank. Graft fixation on the femoral side was by an Endobutton, and on the tibial side by a bioabsorbable interference screw augmented with a staple. The same rehabilitation protocol was applied to all patients. Before surgery and at a mean of 6.9 years of follow-up, patients were evaluated by the same observer according to objective and subjective clinical evaluations including detailed history, physical examination, radiography, functional knee ligament testing, KT-2000 arthrometer testing, Harner vertical jump and Daniel 1-legged hop tests, Lysholm score, Tegner score, International Knee Documentation Committee (IKDC) standard evaluation form, and Cincinnati knee score.

Results:

A total of 83 patients (Auto: n = 40 [mean age, 29.2 ± 6.9 years]; Ir-Allo: n = 43 [mean age, 28.6 ± 7.2 years]) fulfilled follow-up and clinical evaluations. No significant differences were found between the 2 groups according to the overall IKDC functional and subjective evaluations as well as testing of activity levels. Significant between-group differences were found when comparing the results at final follow-up according to the Lachman test, anterior drawer test, pivot-shift test, and KT-2000 arthrometer measurements (P < .001). Most importantly, 87.5% of patients in the Auto group and 34.9% in the Ir-Allo group had a side-to-side difference <3 mm. The rate of laxity (side-to-side difference >5 mm) with an irradiated allograft (30.2%) was higher than that with an autograft (7.5%) (P < .001). The failure rate in the Ir-Allo group (30.2%) was higher than that in the Auto group (7.5%) (P < .001). Anterior and rotational stability decreased significantly in the Ir-Allo group; patients in the Ir-Allo group also had a shorter operation time. There were 10.0% (4/40) of patients in the Auto group and 32.6% (19/43) of patients in the Ir-Allo group who had arthritic progression (P < .05).

Conclusion:

There were no significant differences in postoperative activity levels and functional outcomes between the Auto and Ir-Allo groups. However, a significant increase in anterior and rotational laxity in the Ir-Allo group was found according to evaluations. We do not advocate an irradiated hamstring tendon allograft for DB ACLR.

Trial Registration:

Clinical Trial Register System of The Affiliated Hospital of Qingdao University (qdfy-ky2008-12)

Keywords

anterior cruciate ligament reconstruction double bundle hamstring tendon irradiation allograft prospective randomized study

A rupture of the anterior cruciate ligament (ACL) is one of the most common ligamentous injuries of the knee, with an incidence of 35 of 100,000 and a 2 to 3 times higher risk of injuries for female patients.¹¹ Single-bundle (SB) ACL reconstruction (ACLR) as a gold standard⁹ has been performed for decades, but previously published studies have established that the ACL is a double-bundle (DB) ligament containing the anteromedial (AM) bundle and the posterolateral (PL) bundle, which have different functions.^17,25 Some studies have shown that reconstruction of both AM and PL bundles of the ACL restored anterior and rotational stability.^20,35 As a consequence, better clinical performance and a lower incidence of osteoarthritis (OA) should be observed in patients undergoing DB reconstruction.^10,16,18,37 In recent years, therefore, anatomic DB ACLR has gained popularity. The definition of anatomic ACLR is the functional restoration of the ACL to its native dimension, collagen orientation, and insertion site.³² Yet, to date, no consensus has developed as to whether DB reconstruction is better than SB reconstruction or vice versa. On one hand, part of the conclusion of DB reconstruction having better outcomes is biased by the insufficient sample size and relatively short follow-up. On the other hand, there are many arguments for DB reconstruction based on kinematic, cadaveric, in vivo, and clinical outcome studies.^10,16,18,37

Graft options for ACLR include bone–patellar tendon–bone (BPTB) autografts, hamstring tendon autografts, quadriceps autografts, and various allografts.³⁴ For many years, the BPTB autograft was regarded as the gold standard for ACLR. During the past few decades, the use of hamstring tendons has increased dramatically. Studies have shown that hamstring autografts and BPTB autografts have similar outcomes.^3,15 Some available evidence suggested that hamstring autografts might cause less donor site morbidity and faster quadriceps recovery postoperatively compared with BPTB autografts.^3,15 However, the hamstring autograft also has its problems. The harvest of hamstring tendons may result in saphenous nerve damage and postoperative knee flexion weakness.^13,14,22,29 Thus, a desire to avoid the sacrifice of autologous tissue and to minimize surgical trauma and postoperative donor site morbidity has promoted the consideration of alternative graft sources.^3,4,15,34 Because of these advantages, the use of allografts in ACLR has increased in recent years. Possible disadvantages of allograft use are disease transmission, delayed graft incorporation, graft laxity, and failure with prolonged use.² Gamma irradiation, which has known bactericidal and virucidal properties, is currently the most popular option for the sterilization of allografts. However, some studies^5,8,12 have demonstrated that irradiation has deleterious effects on biomechanical properties in a dose-dependent manner. Some studies also report that irradiated allografts have a higher failure rate.^24,30 As most adopted the SB reconstruction technique for the injured ACL, studies about the clinical outcomes of anatomic DB ACLR with irradiated allografts were never reported.

In this prospective randomized clinical study, we adopted the anatomic DB ACLR technique using an irradiated hamstring tendon allograft and an autograft for ACL ruptures. The purpose of the study was to analyze the clinical outcomes of arthroscopic anatomic DB ACLR with an irradiated hamstring tendon allograft compared with an autograft. We hypothesized that the clinical outcomes of ACLR with an irradiated hamstring tendon allograft would have significant differences compared with those with a hamstring tendon autograft in terms of objective and subjective evaluations.

Methods

Patients and Inclusion Criteria

Approval for this study was obtained from the institutional review board of our hospital. From 2008 to 2009, a total of 121 patients with acute or chronic ACL ruptures underwent ACLR, and 107 patients who met the inclusion and exclusion criteria gave their written informed consent to participate in the study. Patients were excluded from the study if they had a previous injury or surgery on the affected knee, had open physes present, had severe arthritic changes in the knee, had multiple ligamentous injuries, had malalignment, or lacked the ability to complete the study protocol. Patients with revision reconstruction, associated injuries of the PL corner, and deficiency or reconstruction of the ACL in the contralateral knee were also excluded. Only primary unilateral reconstructions of the ACL were included in the study. Patients with a tibial insertion site of <14 mm and a notch width of <12 mm as measured during surgery were also excluded. However, patients with minor medial collateral ligament sprains (grade <2), meniscal tears, or previous diagnostic arthroscopic surgery were not excluded from the study. To meet the inclusion criteria, all patients were examined carefully in clinics and also preoperatively under anesthesia. All patients underwent preoperative magnetic resonance imaging (MRI) to exclude combined complicated ligament injuries to their knees. MRI was also used to measure the size of the native ACL insertion to help in preoperative planning. Patients were randomized on the day of surgery to either the hamstring tendon autograft group (Auto group; n = 54) or the irradiated allograft group (Ir-Allo group; n = 53). A computer software program was used to generate the random allocation sequence. To minimize the effect of bias, the random allocation sequence remained concealed from those enrolling patients into the study.

Of the 107 participants, 14 (13.1%) were excluded at the time of surgery because the tibial insertion site was <14 mm or the width of the notch was <12 mm as measured during the operation. Most of the patients excluded were female (13/14). Five patients (4.7%) failed to undergo anatomic DB ACLR during the operation; the cause of failure was improper positioning of the bone tunnels. The tunnel placements could interfere with each other, and these patients were also excluded from the study. Among the remaining 88 patients, 83 (94.3%) (Auto: n = 40; Ir-Allo: n = 43) were available for full evaluation. Five patients (5.7%) (Auto: n = 4; Ir-Allo: n = 1) were lost during follow-up because of migration (Auto: n = 1; Ir-Allo: n = 1) or lack of interest (Auto: n = 3). Three patients lost in the Auto group were known to have satisfactory clinical outcomes of ACLR. The 1 patient lost to follow-up in the Ir-Allo group refused to continue to participate in the clinical evaluation because of poor clinical outcomes of ACLR. A study flow diagram is provided in Figure 1.

Harvest and Preparation of Grafts

For anatomic ACLR, the patient was placed supine on the operating table. The injured knee was placed in a leg holder with an unsterile tourniquet around the upper thigh, allowing greater than 120° of knee flexion. The autogenous gracilis and semitendinosus tendons were harvested through an approximately 2-cm oblique incision over the pes anserinus by using a tendon stripper (Arthrex).³⁰ On the Graftmaster board (Smith & Nephew), the tendons were cleaned of soft tissue, and their tapered parts were cut off. Both the semitendinosus tendon, prepared for the AM bundle, and the gracilis tendon, prepared for the PL bundle, were folded in half, and each bundle was looped over single Endobuttons (Smith & Nephew). The distal free ends of the tendons were armed with No. 6 Ethibond sutures (Ethicon) using a whipstitch technique, and the grafts were pretensioned under 20 lb for 20 minutes and ready for passing through bone tunnels.³⁰ The diameter of the doubled semitendinosus graft was approximately 6.5 to 7.5 mm, and the diameter of the doubled gracilis graft was approximately 5.5 to 6.5 mm.

All the hamstring tendon allogenic constructs (semitendinosus and gracilis) were irradiated fresh-frozen grafts that were supplied by a certified tissue bank, which had policies for serological and microbiological testing in accordance with guidelines set forth by the American Association of Tissue Banks and the Food and Drug Administration. The irradiated allografts received a radiation dose of 2.5 Mrad before distribution. On the day of the operation, the graft was thawed in sterile normal saline at room temperature before preparation. All allograft constructs were prepared in an identical fashion with a similar diameter as the autograft constructs. The semitendinosus tendon allograft was also prepared for the AM bundle, and the gracilis tendon allograft was prepared for the PL bundle. Allografts were preconditioned using the Graftmaster board at 20 lb of tension for 20 minutes.³⁰

Surgical Technique

All ACLR procedures were performed by the same senior arthroscopic surgeon. All patients underwent an examination under anesthesia to confirm the preoperative diagnosis. A 3-portal technique was used with a high anterolateral portal, central portal, and accessory medial portal.^1,5 After establishing the arthroscopic portals, arthroscopic surgery was performed to determine the extent of the ACL injury and possible concomitant injuries and decide if anatomic DB reconstruction should be performed. The tibial and femoral insertion sites were marked and measured with an arthroscopic ruler (Smith & Nephew).^14,21,33 Then, the width and height of the notch were also measured. After identifying the lateral intercondylar ridge, bifurcate ridge, and tibial footprint, the tibial and femoral tunnels were placed in the center of the tibial and femoral AM and PL bundle insertion sites as previously reported.^14,21,33 The sizes of the tunnels were drilled according to the precise diameter of the graft. For acute ACL injuries, the tibial attachments of the remainder of the native ACL were preserved as much as possible to serve as a landmark for tibial guide pin placement. For tibial tunnel placement, the tibial guide (Smith & Nephew) was set at 45° and placed in the center of the PL bundle for guide pin passage. The tibial guide was adjusted to 55° and centered within the AM footprint for passage of the AM guide pin. For preparation of the femoral tunnel, the central and accessory medial portals were used to achieve accurate tunnel placement. The femoral tunnels were prepared with a 2-mm bony bridge between them. After the femoral and tibial tunnels were prepared, the grafts were put in place, and the femoral side graft was fixed with an Endobutton. Preconditioning of the grafts was performed by flexing and extending the knee through a range of motion (ROM) from 0° to 120° approximately 25 times with firm traction on the tibial extremity. The Endobutton was thus held against the femoral cortex and the graft stretched as much as possible, avoiding possible secondary slackening. Then, the tibial side grafts were fixed with a bioabsorbable interference screw (Smith & Nephew) augmented with a staple. The graft was fixed with an interference screw with the knee at 45° of flexion for the AM bundle and in full extension for the PL bundle. The screw was the same diameter as the drilled tunnel. An intraoperative radiograph was obtained to verify correct placement of the graft, the Endobutton, and the screw. Then, a final arthroscopic inspection was performed to confirm again appropriate positioning and tensioning of the grafts and the absence of graft impingement. No notchplasty was needed. Before reconstruction, meniscal injuries were addressed. Tears in the red-red or red-white zone were repaired using an inside-out technique. Partial meniscectomy was performed for irreparable tears. Because there were strict inclusion and exclusion criteria, no patients included in the study had severe chondral damage (eg, OA). Most of the chondral lesions were from instability of the knee after an ACL rupture. Thus, for chondral lesions, we only performed debridement. No patients needed concomitant surgery for medial or lateral collateral ligament injuries. Routine closure of all wounds was performed. The knee was placed in extension after surgery, and the rehabilitation protocol was applied.³⁰

Rehabilitation

All patients were rehabilitated according to the same protocol, which began at the time of the initial diagnosis.³⁰ In this period before surgery, patients participated in physical therapy to restore full knee ROM and a normal gait and to eliminate knee swelling. After reconstruction, all patients of the 2 groups followed the same postoperative rehabilitation protocol. At the first postoperative visit at 14 days after surgery, the sutures were removed, and ROM exercises were initiated. The goal of physical therapy at this stage was to obtain full extension as compared with the contralateral side. Once ROM was improving, closed chain exercises were encouraged immediately to work on strengthening. Active, nonweightbearing straight-leg raises were started to strengthen the quadriceps immediately after surgery to prevent extension lag as good active quadriceps function and strength protect the knee from instability. Continuous passive motion began the day after surgery for 2 hours twice a day. Patients would start between 0° and 45° and increased to 10° per day as tolerated to a maximum of 120°. Then, it was discontinued. The progression of weightbearing with crutches or canes was on an as-tolerated basis, being guided by the presence and degree of pain and swelling. Crutch or cane use could be discontinued when gait was normalized. Proprioception activities, such as slide boarding and the use of balls and racquets with balance activities, among others, as well as aquatic programs including pool running and swimming, were allowed 8 weeks postoperatively and extended through approximately 4 to 6 months. Functional activities including low-speed treadmill jogging were permitted at 4 to 6 months after surgery, and the patient could start to perform low-level agility and jumping exercises over the next 2 months. Contact sports were allowed at 6 months at the earliest, provided that the patient had regained full functional stability in terms of strength, coordination, and balance as compared with the contralateral leg. Usually, 9 to 12 months was needed for patients to return to full sports activity. Appropriate modifications to the ROM limits and weightbearing status were made for concomitant meniscal repair and chondral treatment. A functional brace was recommended for use during sports activities for the first 1 to 2 years after surgery.

Clinical Evaluation

Assessments of the involved knee were performed preoperatively and at the follow-up point (2nd week; 1st, 3rd, 6th, 9th, and 12th month; and every year thereafter postoperatively) to obtain objective and subjective measures of the clinical outcomes of ACLR.³⁰ All patients were examined by an orthopaedic surgeon who was not involved in the surgical procedure. The examination of knee laxity included the Lachman test, anterior drawer test (ADT), pivot-shift test, and varus/valgus stress test. The classifications used for Lachman and ADT were grade 0 (–1 to 2 mm), grade 1 (3-5 mm), grade 2 (6-10 mm), and grade 3 (>10 mm); for pivot-shift test they were grade 0 (equal), grade 1 (glide), grade 2 (clunk), and grade 3 (gross). Instrumented anteroposterior laxity was measured using the KT-2000 arthrometer (Medmetric) by manual maximum testing with the knee positioned in 15° of flexion. Functional tests included ROM of the knee, the Harner vertical jump test, the Daniel 1-legged hop test, and the standard knee ligament evaluation form of the International Knee Documentation Committee (IKDC). The Vertec unit (Sports Imports) was used to perform the Harner vertical jump test, and a tape measure affixed to the floor was used to measure the Daniel 1-legged hop. The patient performed 3 trials on each leg. The average of the 3 trials was used to determine the vertical jump and hop indices by dividing the involved leg by the noninvolved leg and multiplying by 100%. A quotient (%) was used to determine the limb symmetry index (normal, ≥90%; nearly normal, 76%-89%; abnormal, 50%-75%; or severely abnormal, <50%). The subjective evaluation included the Cincinnati knee score and IKDC subjective knee form, which consists of a questionnaire rating symptoms of pain, swelling, instability, and so on. According to the subjective IKDC form, higher scores reflected fewer symptoms and better knee function. The Tegner activity score and the modified Lysholm knee scoring scale were also used to assess the patient’s activity level and knee function preoperatively and at the final follow-up. Weightbearing anteroposterior, lateral, and femoral-patellar radiographs in 30° of flexion were taken of both knees (ipsilateral and contralateral) preoperatively and at the final follow-up. The radiographs were taken under standardized conditions to obtain reproducible images. The grade of OA was evaluated by 2 independent, unbiased blinded radiologists according to the classifications of Kellgren-Lawrence (1, doubtful: minute osteophytes, doubtful significance; 2, minimal: definite osteophytes, unimpaired joint space; 3, moderate: moderate diminution of joint space; and 4, severe: joint space greatly impaired with sclerosis of subchondral bone).

Statistical Analysis

We used SPSS for Windows (version 18.0; SPSS Inc) for statistical analysis. The independent t test was used for the comparison of continuous variables, and the chi-square test was used for the categorical variables. Before initiation of the study, a power analysis was performed by a statistician to ensure enrollment of an adequate sample size. As one of our main outcome measurements was the KT-2000 arthrometer, the required sample size was calculated considering the minimum clinical significance of anteroposterior knee laxity (KT-2000 arthrometer side-to-side difference), which was assumed to be 5 mm. The minimum sample size of 40 patients per group would be needed to find a significant difference between groups at an alpha level of 0.05 and power of 0.80. In addition to the required 80 participants, 27 more patients were recruited to allow for possible attrition.

Results

Patients

The 83 study patients had a mean 6.9 years (range, 5.5-8.0 years) of follow-up after anatomic DB ACLR. Sporting activities were the main cause of injury of the patients. Playing football and basketball were the most common injury-causing sports. Patients’ demographics and characteristics are outlined in Table 1. There was no statistical significance between the 2 groups.

Table 1

Demographic Data and Characteristics of Study Sample (N = 83 Patients)^a

	Auto Group (n = 40)	Ir-Allo Group (n = 43)	P Value
Age at surgery, y	29.2 ± 6.9 (18-55)	28.6 ± 7.2 (18-50)	.700
Sex, female:male, n	8:32	9:34	.916
Side, right:left, n	21:19	21:22	.739
Time from injury to surgery, mo	1.5 ± 1.1 (0.3-6.5)	1.6 ± 1.2 (0.5-7.0)	.694
Follow-up interval, y	7.0 ± 0.7 (5.5-8.0)	6.8 ± 0.8 (5.5-8.0)	.231
Type of injury, n (%)			.845
Spontaneous rupture	0 (0.0)	0 (0.0)
Sports	35 (87.5)	37 (86.0)
Work	1 (2.5)	2 (4.7)
Traffic accident	3 (7.5)	3 (7.0)
Other/missing	1 (2.5)	1 (2.3)

Data are reported as mean ± SD (range) unless otherwise indicated. There were no statistically significant between-group differences. Auto, autograft; Ir-Allo, irradiated allograft.

General Results

The mean duration of the autograft procedure (mean, 115 minutes; range, 90-135 minutes) was 9 minutes longer compared with the allograft procedure (mean, 106 minutes; range, 80-130 minutes). A statistically significant difference was found between the 2 groups (P < .001).

Complications

There were 2 patients with a superficial wound infection in the Ir-Allo group at the incision area that needed antibiotic treatment and healed well soon after. There were no complications that required reoperation or readmission. In the Auto group, hypoesthesia of the medial saphenous nerve territory was experienced in 3 patients. None was disabling. There were no cases of deep venous thrombosis, deep infection, arthrofibrosis, pain on kneeling, anterior knee pain, failure of fixation, or blowout fracture in either treatment group during the study.

Intraoperative Findings

Lesions to the meniscus, cartilage, and ligament, as well as arthroscopic therapy to the meniscus and cartilage, are outlined in Table 2. There was no statistical difference between groups. No treatment was performed for medial ligament injuries.

Table 2

Arthroscopic Findings and Treatment at Anterior Cruciate Ligament Reconstruction^a

	Auto Group (n = 40)	Ir-Allo Group (n = 43)	P Value
Medial meniscal tears			.794
None	24 (60.0)	27 (62.8)
Partial	14 (35.0)	15 (34.9)
Complete	2 (5.0)	1 (2.3)
Lateral meniscal tears			.685
None	25 (62.5)	25 (58.1)
Partial	12 (30.0)	16 (37.2)
Complete	3 (7.5)	2 (4.7)
Treatment of meniscal tears			.865
NA	16 (40.0)	18 (41.9)
Repair	7 (17.5)	9 (20.9)
Partial meniscectomy	17 (42.5)	16 (37.2)
Cartilage damage (Outerbridge grade)			.896
Normal	33 (82.5)	35 (81.4)
1	4 (10.0)	5 (11.6)
2	3 (7.5)	3 (7.0)
3	00 (0.0)	00 (0.0)
4	00 (0.0)	00 (0.0)
Treatment for cartilage damage			.845
NA	35 (87.5)	37 (86.0)
Debridement	5 (12.5)	6 (14.0)
Medial ligament injuries			.999
Normal	38 (95.0)	40 (93.0)
Grade 1	2 (5.0)	3 (7.0)
Grade 2	00 (0.0)	00 (0.0)
Grade 3	00 (0.0)	00 (0.0)

Data are reported as n (%). The chi-square test was used for the categorical variables. There were no statistically significant between-group differences. Auto, autograft; Ir-Allo, irradiated allograft; NA, not applicable as the meniscus or cartilage was normal.

Objective Clinical Results

Patients of the Auto group showed no significantly better rating according to the chi-square test for the overall IKDC score compared with the Ir-Allo group (Table 3). When comparing the side-to-side difference in anterior tibial displacement, with the percentage of the side-to-side difference <3 mm and >5 mm between the 2 groups according to the manual maximum KT-2000 arthrometer test, significant differences were found. The failure rate in the Ir-Allo group (30.2%) was higher than that in the Auto group (7.5%) (P < .001). The mean anterior tibial displacement of patients in the Ir-Allo group was also significantly higher than that in the other group (Table 4). According to the assessments evaluating the laxity rate of the pivot-shift test, ADT, and Lachman test, significant differences were found comparing the Ir-Allo group with the Auto group (Table 5). Concerning ROM, the vertical jump test, and the 1-legged hop test, all data were collected and comparable between the groups at follow-up (Table 3). No significant differences were found between groups.

Table 3

Knee Functional Assessment at Final Follow-up^a

	Auto Group (n = 40)	Ir-Allo Group (n = 43)	P Value
Overall IKDC			.435
Normal	14 (35.0)	13 (30.2)
Nearly normal	24 (60.0)	25 (58.1)
Abnormal	2 (5.0)	4 (9.3)
Severely abnormal	00 (0.0)	1 (2.3)
Range of motion			.999
Normal	21 (52.5)	22 (51.2)
Nearly normal	18 (45.0)	19 (44.2)
Abnormal	1 (2.5)	2 (4.7)
Severely abnormal	00 (0.0)	00 (0.0)
Harner vertical jump test			.677
Normal	15 (37.5)	13 (30.2)
Nearly normal	23 (57.5)	26 (60.5)
Abnormal	2 (5.0)	4 (9.3)
Severely abnormal	00 (0.0)	00 (0.0)
Daniel 1-legged hop test			.677
Normal	16 (40.0)	14 (32.6)
Nearly normal	22 (55.0)	25 (58.1)
Abnormal	2 (5.0)	4 (9.3)
Severely abnormal	00 (0.0)	00 (0.0)

Data are reported as n (%). The chi-square test was used for the categorical variables. There were no statistically significant between-group differences. Auto, autograft; IKDC, International Knee Documentation Committee; Ir-Allo, irradiated allograft.

Table 4

Preoperative and Follow-up Results of Knee Stability According to KT-2000 Arthrometer^a

	Preoperative		Follow-up
	Auto Group	Ir-Allo Group	Auto Group	Ir-Allo Group	P Value^b
Anterior tibial displacement (all patients), mm	10.9 ± 1.8 (8-18)	11.0 ± 1.7 (7-19)	5.0 ± 0.8 (1-10)	8.7 ± 1.1 (1-14)	<.001
Side-to-side difference, mm	—	—	2.4 ± 0.7 (0-7)	5.5 ± 1.0 (0-10)	<.001
Side-to-side difference <3 mm	—	—	35 (87.5)	15 (34.9)	<.001
Side-to-side difference >5 mm	—	—	3 (7.5)	13 (30.2)

Data are reported as mean ± SD (range) or n (%). Preoperative data for side-to-side difference were not available. Significant differences were found between groups. Auto, autograft; Ir-Allo, irradiated allograft.

Comparison between Auto group versus Ir-Allo group.

Table 5

Comparison of Rotational and Anterior Stability of Operated Knees at Final Follow-up^a

	Auto Group (n = 40)	Ir-Allo Group (n = 43)	P Value
Pivot-shift test			.004
0	37 (92.5)	27 (62.8)
1	3 (7.5)	11 (25.6)
2	00 (0.0)	5 (11.6)
3	00 (0.0)	00 (0.0)
Anterior drawer test			<.001
0	34 (85.0)	16 (37.2)
1	4 (10.0)	15 (34.9)
2	2 (5.0)	12 (27.9)
3	00 (0.0)	00 (0.0)
Lachman test			<.001
0	33 (82.5)	15 (34.9)
1	4 (10.0)	16 (37.2)
2	3 (7.5)	12 (27.9)
3	00 (0.0)	00 (0.0)

Data are reported as n (%). The chi-square test was used for the categorical variables. Significant differences were found between groups for all 3 tests. Auto, autograft; Ir-Allo, irradiated allograft.

Subjective Clinical Results

According to the subjective IKDC, Lysholm, and Tegner scores, there were no significant differences between the 2 groups (P = .0748, .0727, and .0730, respectively). Most patients were satisfied with their performance in sporting activity. For the Cincinnati knee score, no significant differences were found between groups. The data are shown in Table 6.

Table 6

Subjective Evaluations and Activity Level at Final Follow-up^a

	Auto Group (n = 40)	Ir-Allo Group (n = 43)	P Value
Subjective IKDC	89 ± 9 (65-100)	85 ± 11 (60-100)	.0748
Cincinnati knee score	90 ± 10 (50-100)	87 ± 12 (45-100)	.2214
Lysholm score	90 ± 11 (65-100)	86 ± 9 (50-100)	.0727
Tegner score	7.8 ± 1.2 (3-9)	7.3 ± 1.3 (3-9)	.0730

Data are reported as mean ± SD (range). There were no statistically significant between-group differences. Auto, autograft; IKDC, International Knee Documentation Committee; Ir-Allo, irradiated allograft.

Radiographic Findings

As to the OA rate of the operated knee, when compared with the contralateral side according to the Kellgren-Lawrence classification, in the Auto group, 3 patients (7.5%) decreased 1 grade, 1 patient (2.5%) decreased 2 grades, and in 90.0% of the patients (36/40), there was no change. In the Ir-Allo group, 9 patients (20.9%) decreased 1 grade, 5 patients (11.6%) decreased 2 grades, and in 67.4% of the patients (29/43), there was no change. For the Auto group, there was no significant difference in the development of OA between the operated knee in comparison to the contralateral knee at the final follow-up, while for the Ir-Allo group, a significant difference (P < .05) was found in the development of OA between the operated knee in comparison to the contralateral knee. There was a statistical difference (P < .05) when comparing the development of OA between the Auto and Ir-Allo groups at the final follow-up (Table 7).

Table 7

Radiological Results According to the Kellgren-Lawrence Classification^a

	Auto Group (n = 40)			Ir-Allo Group (n = 43)
	Ipsilateral	Contralateral	Difference in Percentage Points	Ipsilateral	Contralateral	Difference in Percentage Points
Kellgren-Lawrence grade
0	34 (85.0)	36 (90.0)	−5.0	28 (65.1)	38 (88.4)	−23.3
1	3 (7.5)	2 (5.0)	+2.5	9 (20.9)	3 (7.0)	+13.9
2	2 (5.0)	2 (5.0)	0.0	4 (9.3)	2 (4.7)	+4.6
3	1 (2.5)	00 (0.0)	+2.5	2 (4.7)	00 (0.0)	+4.7
4	00 (0.0)	00 (0.0)	0.0	00 (0.0)	00 (0.0)	0.0

Data are reported as n (%) unless otherwise indicated. Auto, autograft; Ir-Allo, irradiated allograft.

Discussion

The most important findings of our study were that there were no significant differences in postoperative activity levels and functional outcomes between the 2 groups. However, the differences in physical examination measures and instrumented laxity between the 2 groups were significant.

Both the autograft and allograft have their advantages and disadvantages. In this study, there were 3 patients with nerve damage, with hypoesthesia of the medial saphenous nerve territory, in the Auto group. For allografts, the most important advantage is the lack of donor site morbidity, and the most significant disadvantages are disease transmission, delayed graft incorporation and graft laxity, and failure with prolonged use.² Thus, for allograft use in clinical situations, the irradiated allograft is used most commonly for safety considerations. Gamma irradiation is perhaps the most widely applied method of sterilizing tissue transplants. However, at the dose required for sufficient pathogen inactivation (>30 kGy), it is known to have unacceptable adverse effects.^6,8,12 Both basic science research and biomechanical research showed that allograft irradiation decreases the biomechanical properties of the graft in a dose-dependent fashion.^6,8,12 Yet, a recent study of low-dose (1.0-1.2 Mrad) gamma irradiation of BPTB grafts showed decreased graft stiffness by 20% without any change in biomechanical properties.³⁶ Clinical studies have yielded mixed results regarding whether the irradiation of allografts leads to higher rates of graft failure. Rappe et al²⁶ found a 33% failure rate for irradiated allografts versus 2.4% for nonirradiated allografts. Conversely, Rihn et al²⁷ found no adverse effects of irradiation on clinical outcomes in ACLR with allografts. In former studies, the clinical outcomes of SB reconstruction with autografts versus irradiated allografts were compared, and a significant increase of the failure rate in the irradiated allograft group was found.³⁰ In former studies, the technique chosen for the injured ACL was SB reconstruction; in this current study, although we adopted the anatomic DB reconstruction technique, we still obtained as high as a 30.2% failure rate in the Ir-Allo group according to the KT-2000 arthrometer measurements. When comparing the results of ACLR in the Ir-Allo and Auto groups according to the ADT, Lachman test, and pivot-shift test, we also found an increase in anterior laxity and rotational instability in the Ir-Allo group. The difference was statistically significant. The high laxity rate of irradiated hamstring tendon allografts was in line with the literature comparing irradiated BPTB allografts to nonirradiated allografts, which showed that irradiated grafts had a higher failure rate of 31%.²⁴

Fortunately, as of now, the knee function of patients in the Ir-Allo group is still satisfactory, and activity levels are not adversely affected because of laxity. Further study with a long-term follow-up should be conducted focusing on the functional and activity levels of patients in the Ir-Allo group. With so high a laxity rate, we again do not advocate the irradiated hamstring tendon allograft.

In our study, there were 14 patients (13.1%) excluded from the study at the time of surgery, as they were not indicated for anatomic DB ACLR because the tibial insertion site was <14 mm or the width of the notch was <12 mm. Also, most of the patients were female; this is a relatively high proportion of the study population. The choice of anatomic ACLR techniques is particularly important for Asian patients. DB reconstruction may not always be the optimal choice for Asian patients as they have relatively smaller bones and ligament sizes. Furthermore, DB reconstruction is a highly technically demanding technique and has a learning curve. The success rate of DB reconstruction also correlates well with the number of operations performed per year.²⁸ Even experienced surgeons still have a 5% to 10% failure rate when performing ACLR.²³ In our study, the failure rate of performing anatomic DB reconstruction was 4.7% (5/107). The choice of reconstruction techniques should always be made according to the indications, just as we did in the current study.

Regarding the degenerative changes of the operated knee, evidence suggests that the reconstruction techniques for the ACL influence arthritic progression. Studies have suggested that DB ACLR may reduce the incidence of OA by closely restoring the contact area and pressure.¹⁹ In the study by Sun et al³¹, they found that 29.2% of patients in the DB autograft group and 27.3% of patients in the DB allograft group had a progression in arthritic changes. In our study, 10.0% (4/40) of patients in the Auto group and 32.6% (14/43) of patients in the Ir-Allo group had arthritic progression. Also, there was a significant difference between the 2 groups at the final follow-up. For the Ir-Allo group, a significant difference was also found in the development of OA in the operated knee in comparison to the contralateral knee. This suggests that the progression of OA was related to instability of the operated knee.

This study had several limitations. The first was that there were some patients excluded from the study at surgery and some patients lost during the follow-up period; the remaining group of patients was relatively small, and the follow-up time was intermediate. This naturally limits the conclusions concerning the insignificant findings in group comparisons. Long-term outcomes with more patients should be further studied. The second was the observer bias. The data were collected by only one surgeon at one institution and were not collected in a blinded fashion. Patients were informed as to the type of surgical procedure that they underwent by the surgeon after surgery, so the data collector may also have been aware at the time of the follow-up. Additionally, the incisions could also have tipped off the observer as to the type of surgery. The third was that the information about patients’ time to return to sports was not collected in the study. Furthermore, this study includes data being limited to ACLR using hamstring tendons as the graft choice, and the results of the study are from surgical procedures performed by one single experienced surgeon. The conclusions may not be generalizable. In addition, in our study, we only showed the changes of the incidence of knee OA after ACLR, but we did not study the exact incidence. The final limitation is that although a standardized therapy protocol was prescribed to all patients postoperatively, the quality and consistency of physical therapy may have varied at outside institutions. This can be a factor that affects clinical outcomes. In addition, although female patients are more likely to suffer injuries of the ACL as reported in the literature, there were few female patients in our study, similar to the reported literature.^7,30 Further study also needs to be conducted.

Conclusion

There were no significant differences in postoperative activity levels and functional outcomes between the Auto and Ir-Allo groups. A significant increase in graft failures as well as in anterior and rotational laxity in patients of the Ir-Allo group was found according to evaluations. We do not advocate the irradiated hamstring tendon allograft for DB ACLR.

Footnotes

The authors declared that they have no conflicts of interest in the authorship and publication of this contribution.

References

Araujo

van Eck

Macalena

. Advances in the three-portal technique for anatomical single- or double-bundle ACL reconstruction. Knee Surg Sports Traumatol Arthrosc. 2011;19:1239-1242.

Barbour

King

. The safe and effective use of allograft tissue: an update. Am J Sports Med. 2003;31:791-797.

Biau

Tournoux

Katsahian

Schranz

Nizard

. Bone-patellar tendon-bone autografts versus hamstring autografts for reconstruction of anterior cruciate ligament: meta-analysis. BMJ. 2006;332:995-1001.

Chechik

Amar

Khashan

Lador

Eyal

Gold

. An international survey on anterior cruciate ligament reconstruction practices. Int Orthop. 2013;37:201-206.

Cohen

. Three-portal technique for anterior cruciate ligament reconstruction: use of a central medial portal. Arthroscopy. 2007;23:325.e1-325.e5.

Curran

Adams

Gill

Steiner

Scheller

. The biomechanical effects of low-dose irradiation on bone-patellar tendon-bone allografts. Am J Sports Med. 2004;32:1131-1135.

Cvetanovich

Mascarenhas

Saccomanno

. Hamstring autograft versus soft-tissue allograft in anterior cruciate ligament reconstruction: a systematic review and meta-analysis of randomized controlled trials. Arthroscopy. 2014;30:1616-1624.

Fideler

Vangsness

Jr Lu

Orlando

Moore

. Gamma irradiation: effects on biomechanical properties of human bone-patellar tendon- bone allografts. Am J Sports Med. 1995;23:643-646.

Freedman

D’Amato

Nedeff

Kaz

Bach

Jr.

Arthroscopic anterior cruciate ligament reconstruction: a metaanalysis comparing patellar tendon and hamstring tendon autografts. Am J Sports Med. 2003;31:2-11.

10.

Gadikota

Seon

Chen

Gill

. In vitro and intraoperative laxities after single-bundle and double-bundle anterior cruciate ligament reconstructions. Arthroscopy. 2011;27:849-860.

11.

Gianotti

Marshall

Hume

Bunt

. Incidence of anterior cruciate ligament injury and other knee ligament injuries: a national population-based study. J Sci Med Sport. 2009;12:622-627.

12.

Grieb

Forng

Bogdansky

. High-dose gamma irradiation for soft tissue allografts: high margin of safety with biomechanical integrity. J Orthop Res. 2006;24:1011-1018.

13.

Hiemstra

Webber

MacDonald

Kriellaars

. Hamstring and quadriceps strength balance in normal and hamstring anterior cruciate ligament-reconstructed subjects. Clin J Sport Med. 2004;14:274-280.

14.

Hussein

van Eck

Cretnik

Dinevski

. Individualized anterior cruciate ligament surgery: a prospective study comparing anatomic single- and double-bundle reconstruction. Am J Sports Med. 2012;40:1781-1788.

15.

Zhao

. A meta-analysis of hamstring autografts versus bone-patellar tendon-bone autografts for reconstruction of the anterior cruciate ligament. Knee. 2011;18:287-293.

16.

Luites

Wymenga

Blankevoort

Kooloos

Verdonschot

. Computer-assisted anatomically placed double-bundle ACL reconstruction: an in vitro experiment with different tension angles for the AM and the PL graft. Med Eng Phys. 2012;34:1031-1036.

17.

Mae

Shino

Miyama

. Single- versus two-femoral socket anterior cruciate ligament reconstruction technique: biomechanical analysis using a robotic simulator. Arthroscopy. 2001;17:708-716.

18.

Meredick

Vance

Appleby

Lubowitz

. Outcome of single-bundle versus double-bundle reconstruction of the anterior cruciate ligament: a meta-analysis. Am J Sports Med. 2008;36:1414-1421.

19.

Morimoto

Ferretti

Ekdahl

Smolinski

. Tibiofemoral joint contact area and pressure after single- and double-bundle anterior cruciate ligament reconstruction. Arthroscopy. 2009;25:62-69.

20.

Mott

. Semitendinosus anatomic reconstruction for cruciate ligament insufficiency. Clin Orthop Relat Res. 1983;172:90-92.

21.

Muller

Hofbauer

Wongcharoenwatana

. Indications and contraindications for double-bundle ACL reconstruction. Int Orthop. 2013;37:239-246.

22.

Nakamura

Horibe

Sasaki

. Evaluation of active knee flexion and hamstring strength after anterior cruciate ligament reconstruction using hamstring tendons. Arthroscopy. 2002;18:598-602.

23.

Prodromos

Howell

Johnson

Lawhorn

. Controversies in soft-tissue anterior cruciate ligament reconstruction: grafts, bundles, tunnels, fixation, and harvest. J Am Acad Orthop Surg. 2008;16:376-384.

24.

Prodromos

Joyce

Shi

. A meta-analysis of stability of autografts compared to allografts after anterior cruciate ligament reconstruction. Knee Surg Sports Traumatol Arthrosc. 2007;15:851-856.

25.

Radford

Amis

. Biomechanics of a double prosthetic ligament in the anterior cruciate deficient knee. J Bone Joint Surg Br. 1990;72:1038-1043.

26.

Rappe

Horodyski

Meister

Indelicato

. Nonirradiated versus irradiated Achilles allograft: in vivo failure comparison. Am J Sports Med. 2007;35:1653-1658.

27.

Rihn

Irrgang

Chhabra

Harner

. Does irradiation affect the clinical outcome of patellar tendon allograft ACL reconstruction? Knee Surg Sports Traumatol Arthrosc. 2006;14:885-896.

28.

Robertsson

Knutson

Lewold

Lidqren

. The routine of surgical management reduces failure after unicompartmental knee arthroplasty. J Bone Joint Surg Br. 2001;83:45-49.

29.

Segawa

Omori

Koga

Kameo

Iida

Tanaka

. Rotational muscle strength of the limb after anterior cruciate ligament reconstruction using semitendinosus and gracilis tendon. Arthroscopy. 2002;18:177-182.

30.

Sun

Zhang

Wang

. Arthroscopic anterior cruciate ligament reconstruction with at least 2.5 years’ follow-up comparing hamstring tendon autograft and irradiated allograft. Arthroscopy. 2011;27:1195-1202.

31.

Sun

Chen

Wang

Chen

. Prospective randomized comparison of knee stability and joint degeneration for double- and single-bundle ACL reconstruction. Knee Surg Sports Traumatol Arthrosc. 2015;23:1171-1178.

32.

van Eck

Schreiber

Liu

. The anatomic approach to primary, revision and augmentation anterior cruciate ligament reconstruction. Knee Surg Sports Traumatol Arthrosc. 2010;18:1154-1163.

33.

van Eck

Widhalm

Murawski

. Individualized anatomic anterior cruciate ligament reconstruction. Phys Sportsmed. 2015;43:87-92.

34.

West

Harner

. Graft selection in anterior cruciate ligament reconstruction. J Am Acad Orthop Surg. 2005;13:197-207.

35.

Yagi

Wong

Kanamori

Debski

Woo

. Biomechanical analysis of an anatomic anterior cruciate ligament reconstruction. Am J Sports Med. 2002;30:660-666.

36.

Yanke

Bell

Lee

. The biomechanical effects of 1.0 to 1.2 Mrad of gamma irradiation on human bone-patellar tendon-bone allografts. Am J Sports Med. 2013;41:835-840.

37.

Yao

Wen

Cheung

. Deterioration of stress distribution due to tunnel creation in single-bundle and double-bundle anterior cruciate ligament reconstructions. Ann Biomed Eng. 2012;40:1554-1567.