Abstract
Background:
The anterior cruciate ligament (ACL) repair technique avoids graft harvest and therefore the risk of donor site morbidity. However, early failure rates after ACL repair with suture ligament augmentation (SLA) remain high.
Purpose:
To compare surgical failure, functional outcomes, return to sport, and joint laxity between adolescents who underwent ACL repair with SLA and those who underwent ACL reconstruction with quadriceps tendon–patellar bone autograft (QPA).
Study Design:
Cohort study; Level of evidence, 3.
Methods:
Adolescent patients (7-18 years old) underwent ACL repair with SLA or ACL reconstruction with QPA. The authors collected data from those who had postoperative information pertaining to repaired ligament or graft failure, range of motion, complications, and return to sport at a minimum of 6 months after surgery. Participants were contacted after surgery to complete study questionnaires.
Results:
The cohort included 22 consecutive patients in the SLA group and 157 in the QPA group. The median duration of follow-up was 2.7 years (interquartile range, 2.0-3.6 years) in the QPA group and 3.2 years (2.2-3.4 years) in the SLA group. After adjustment for sex, age, body mass index, and time from injury to surgery, the hazard of graft failure in the SLA group was 10.66 times (95% CI, 3.41-32.92; P < .0001) that of the QPA group. The cumulative incidence of graft failure in the first 3 years after surgery was 48.8% (95% CI, 28.9%-73.1%) in the SLA group, as opposed to 4.7% (2.1%-10.3%) in the QPA group. There was no difference in return to sport between the groups. Among individuals who did not rerupture their ACL, International Knee Documentation Committee and Lysholm scores were comparable between the groups, as well as range of motion.
Conclusion:
The risk of failure was significantly increased in the SLA group relative to the QPA group. The high risk of failure for the SLA group in this short-term follow-up should be considered when selecting the treatment for adolescent patients with an ACL injury.
Anterior cruciate ligament (ACL) injury is one of the most common knee injuries among adolescent athletes.14,26 Estimates of ACL reconstruction incidence in the United States range from 120,000 to 400,000 cases annually.14,36 ACL injury diagnoses in the pediatric population have risen, perhaps because of earlier sports specialization, increased injury exposures with year-round sports participation, or increased physician awareness. 2 The current gold standard of care for an ACL rupture among pediatric and adolescent patients is ACL reconstruction with autograft.14,29 ACL reconstruction is usually effective at restoring joint stabilization and should be performed soon after the injury, thereby preventing subsequent meniscal or cartilage injury. 24 However, concerns exist regarding return to sport, premature osteoarthritis, neuromuscular deficits, and/or the loss of native biomechanics and proprioception of the knee joint. §
ACL repair is a resurging technique used to treat ACL deficiency. Robson 27 first described successful ACL repair in 1895 by way of suturing the 2 ends of the torn ligament together. ACL repair avoids donor site morbidity associated with autologous graft harvest. 35 Other potential benefits associated with ACL repair over reconstruction include preservation of native biomechanics, proprioception, and gait patterns. ACL repair was popular in the 1970s and 1980s but was discontinued owing to reports of a high rate of failure.8,28,32 These previously reported failure rates of ACL repair surgery included repairs of midsubstance ruptures, which are associated with poor tissue quality and are surrounded by a plasmin-rich synovial fluid environment that is not conducive to tissue repair.13,23 Thus, surgeons have recently focused efforts on repairing proximal avulsions off the femoral footprint. Theoretically, proximal avulsions off the femoral footprint have a superior environment for healing, given the proximity to the bone in the absence of a scaffold. 7 The possibility for conserving the native biomechanics and avoiding graft harvest is taken into consideration by the surgeon when choosing ACL repair over ACL reconstruction. In this study, only proximal avulsions off the femoral footprint with sufficient tissue length and quality of the remaining ACL were considered for ACL repair with suture ligament augmentation (SLA). Promising outcomes have been demonstrated for adult ACL repair with SLA when used to repair proximal ACL avulsions, although existing data are limited. 5
The purpose of this study was to investigate the outcomes of ACL repair with SLA in a cohort of adolescent patients. We compared patients who underwent ACL repair versus ACL reconstruction. We anticipated that ACL repair with SLA and ACL reconstruction with quadriceps tendon–patellar bone autograft (QPA) would demonstrate comparable outcomes. To our knowledge, this study is the first to investigate outcomes of ACL repair with SLA in comparison with ACL reconstruction with autograft among adolescent patients.
Methods
Patient Recruitment and Evaluation
After institutional review board approval, we identified all adolescent patients who underwent ACL repair with SLA or ACL reconstruction with QPA performed by a single surgeon between July 2013 and November 2015. Only patients between 7 and 18 years of age (inclusive) who were clinically evaluated at a postoperative minimum of 6 months were included. Patients were excluded on the basis of concomitant grade 3 ligament injuries, documented musculoskeletal disease, or a previous ipsilateral ACL reconstruction (Figure 1). A retrospective chart review was used to collect demographics, skeletal maturity level, operative information, manual Lachman test results, range of motion (ROM), and the date of return to sport, if the participant passed a physical therapist–administered return-to-sport test. The retrospective review was augmented with a phone interview and/or study visit. We contacted patients prospectively and invited them to schedule a long-term follow-up visit and/or complete a phone interview. Body mass index (BMI) measurements were converted to age and sex-specific Z scores with a publicly available program on the website of the Centers for Disease Control and Prevention. 3 Failure was defined as the need for revision surgery or as a magnetic resonance imaging (MRI)–confirmed reruptured ACL. KT-1000 arthrometer measures were obtained prospectively from patients who were seen for a clinic appointment. Return-to-play time was recorded for participants per a clinic visit note or a physical therapy note documenting a return-to-sport date in the medical records or if the patient provided a return-to-sport date during the phone interview.

Enrollment. QPA, quadriceps tendon–patellar bone autograft; SLA, suture ligament augmentation.
Surgical Technique
ACL Repair With SLA
A single surgeon (J.C.A.) who is dually fellowship trained in sports medicine and pediatric orthopaedic surgery and who specializes in pediatric ACL reconstruction performed all cases. The surgeon set 3 required criteria for ACL repair with SLA: proximal avulsion off the femoral footprint (Sherman 1), healthy tissue that accepts suturing, and sufficient tissue length to reach the femoral footprint without gapping at the tear site. The final surgical plan was determined after assessment of the nature of the ACL rupture and the quality of the remaining tissue during arthroscopy. All participants who met the 3 required criteria underwent ACL repair with SLA. No notchplasties were performed in patients who had ACL repair with SLA.
The ACL repair was secured at close to 90° of flexion, and the suture augmentation was secured with the knee in near full extension to avoid excessive stress shielding. FiberTape (Arthrex Inc) was used for the suture augmentation. The Scorpion Suture Passer (Arthrex Inc) was utilized for suturing of the ACL remnant. A single suture (No. 2 FiberWire; Arthrex Inc) was placed in the ACL in either a modified Bunnell suture pattern or a luggage tag formation. 6 The femoral footprint was cleared of soft tissue, and microfracture was performed to stimulate healing. Two drill tunnels were then made in the femoral footprint (all epiphyseal for skeletally immature patients) with a cannulated drill pin. The positions of the tunnels were placed in the anatomic footprint of each bundle. One or 2 tibial tunnels were drilled into the tibial footprint for the braided polyester reinforcement suture. All tunnels were created with a cannulated 4-mm drill. The proximal end of the suture in the ACL was then fixated into the femoral tunnel(s) and tied over a button. The leg was moved into full extension, and the repaired ACL was reinforced with the fiber tape internal brace, bridging the tibia to the femur (Figure 2).

Arthroscopic photo of completed anterior cruciate ligament repair with suture ligament augmentation.
ACL Reconstruction With QPA
The surgical technique used for ACL reconstruction with QPA was previously described in detail by the senior author. 1 Briefly, a No. 15 blade was used to make longitudinal incisions 5 mm medial and lateral from the center of the patella. A sagittal saw was used to cut the trapezoidal bone plug to be approximately 8 to 10 mm deep, 1 cm wide, and 15 mm long. As the bone plug was elevated, the tendon fibers attached to the bone plug were separated from the remaining quadriceps tendon until the graft and bone plug totaled 60 to 70 mm in length. While the graft was being prepared with an adjustable loop button/suspensory fixation system (RT and BTB Tightrope; Arthrex Inc), standard arthroscopic evaluation of the knee joint was performed. At this time, remnants of the native ACL were removed, and any necessary meniscal procedures were performed. The graft was first positioned into the femoral footprint approximately 43% proximal to distal and then into the posterior third of the tibial footprint, in line with the posterior portion of the anterior root of the lateral meniscus.
Rehabilitation
After surgery, patients without meniscal repair were weightbearing as tolerated and immobilized at full extension until able to perform 75 straight-leg raises in 1 day without extensor lag. For patients who underwent a meniscal repair, weightbearing and ROM were limited for 4 to 6 weeks, depending on tear pattern. All participants were encouraged to complete physical therapy at the present institution, where there was a standardized ACL rehabilitation program in place. The rehabilitation regimen was the same between the groups.
Statistical Method
Descriptive statistics were used to summarize the demographics and clinical characteristics in the 2 surgery groups. Chi-square, Student t tests, and Wilcoxon rank sum tests, as appropriate, were used to test for differences in the demographics and clinical characteristics across the 2 surgery groups. A multivariable Cox proportional hazards regression analysis was used to compare both failure-free survival and time to return to play in the 2 groups. Graft failure was modeled as a competing risk in the return-to-play analysis. We also tested the interaction between surgery type and time from injury to surgery. The purpose of this analysis was to test whether the association between surgery failure and time from injury to surgery was different in the 2 surgery groups. A significant P value would provide evidence that the duration of time from injury to surgery has a different effect on the hazard of failure depending on whether the individual underwent the QPA reconstruction or the SLA procedure. Age, sex, BMI Z scores, and time from injury to surgery were included in the multivariable models as potential confounding variables. In the secondary analysis of patients who completed the International Knee Documentation Committee (IKDC) and Lysholm questionnaires and/or underwent an objective measure of joint laxity, Wilcoxon rank sum tests were used to compare IKDC, Lysholm, and KT-1000 arthrometer joint laxity measures in the 2 groups.
Results
The final study population included 22 individuals who underwent ACL repair with SLA. This group was compared with 157 individuals who underwent ACL reconstruction with QPA (Table 1). The median duration of follow-up was 2.7 years (interquartile range [IQR], 2.0-3.6 years) in the QPA group and 3.2 years (IQR, 2.2-3.4 years) in the SLA group. Of the 17 participants who did not have adequate follow-up to be included (Figure 1), 4 had SLA repair, and 13 had QPA reconstruction. No significant chondral work was needed for any of the patients. Age at surgery and skeletal maturity level were significantly different between groups.
Demographics and Clinical Characteristics
Data reported as n(%), unless noted otherwise.
Sex- and age-specific z scores were calculated with population reference values.
After adjustment for BMI Z scores, age, sex, and time from injury to surgery, the hazard of failure in the SLA group was 10.66 times (95% CI, 3.45-32.92; P < .0001) that of the QPA group within the first 36 months after surgery. The cumulative incidence of graft failure in the first 3 years after surgery was 48.8% (95% CI, 28.9%-73.1%) in the SLA group, as compared with 4.7% (95% CI, 2.1%-10.3%) in the QPA group (Figure 3). In the QPA group, 3 patients had graft failure before returning to play (at 10.1, 13.4, and 17.4 months). In the repair group, 4 patients had graft failure before returning to play (at 4.1, 19.4, 18.4, and 10.7 months).

Graft/repair survival.
Age at surgery (P = .1600), BMI Z scores (P = .8227), and sex (P = .2800) were not associated with the hazard of surgery failure. Time from injury to surgery was not associated with hazard of failure (hazard ratio per 1-month increase, 0.91; 95% CI, 0.71-1.17; P = .4598) (Table 2). The relationship between time to surgery and hazard of graft failure did not differ by group (P = .5855).
Time From Injury to Surgery by Group and Graft Failure Status
Among the individuals who did not require revision surgery or complete the postoperative evaluation before graft failure, the median IKDC score was 90.8 (IQR, 83.9-97.7) in the SLA group (n = 8), as compared with 95.4 (IQR, 89.7-98.9) in the QPA group (n = 110). The median Lysholm score was 100 (IQR, 89.5-100.0) in the SLA group (n = 8) and 100 (IQR, 90.0-100.0) in the QPA group (n = 106). Based on the Wilcoxon rank sum test, there was no difference in IKDC scores (P = .3681) or Lysholm (P = .7464) scores between groups. There was also no difference (P = .3932) in the proportion of individuals with flexion/extension ROM limitations at final follow-up in the QPA group (4.6%) versus the SLA group (1.9%). ROM was recorded from physical therapy notes and clinic visit notes. Due to differences in provider documentation preferences, the absolute ROM measurement was not consistently recorded. However, each provider noted whether the ROM was full or not. Therefore, we elected to analyze ROM as a binary full/incomplete variable in the statistical analysis. The increase in side-to-side joint laxity comparisons between the surgical joint and the uninvolved joint was comparable between the groups (Table 3). KT-1000 measurements were available for 33 patients who underwent ACL reconstruction with QPA and 2 who underwent ACL repair with SLA.
Increased Joint Laxity of the Surgical Joint Compared With the Uninvolved Joint a
IQR, interquartile range.
Individual did not suffer graft failure or postoperative evaluation obtained before graft failure.
Median time to return to play was similar for the SLA group (11.9 months; 95% CI, 7.5-14.0 months) and the QPA group (11.2 months; 95% CI, 10.0-12.0 months). After adjustment for sex, age at surgery, BMI, and time from injury to surgery, there was no difference in the hazard of return-to-play in the SLA group versus the QPA group (hazard ratio, 0.69; 95% CI, 0.38-1.24; P = .2165).
Discussion
ACL repair augmentation technology, including improved anchors, instruments, sutures, and soft tissue fixation, has contributed to a renewed interest in ACL repair techniques. 7 However, outcomes of ACL repair relative to ACL reconstruction among adolescents have not been studied. The results from our investigation question the suitability of this technique of ACL repair for treatment of ACL rupture among adolescent patients. The cumulative incidence of graft failure in the first 3 years after surgery was significantly increased in the SLA group versus the QPA group, at 48.8% and 4.7%, respectively (SLA: 95% CI, 28.9%-73.1% QPA: 95% CI, 2.1%-10.3%) (Figure 3). We speculate that patients may have returned to more aggressive activity from the first season to the second season. Between 12 and 18 months after surgery, athletes may not have been playing as aggressively as they normally would have, despite returning to play. In the repair group, it is also possible that the ligament augmentation suture finally underwent fatigue failure at 18 to 24 months. Before that, the ligament may not have healed, but the augmentation suture was stabilizing the knee.
Previous case series reported positive outcomes among patients undergoing ACL repair with SLA.31,36 Smith et al 31 reported that 3 pediatric patients all demonstrated complete ACL healing at postoperative 3 months, returned to activities at 4 months, and showed good objective outcomes at >2 years. A case report of one of the first patients to undergo ACL repair with internal brace ligament augmentation demonstrated full ACL healing and good functional outcomes at postoperative 2 years. 37 For a minimum of 1 year, MacKay et al 20 followed 68 patients (age range, 16-60 years; median age, 28 years) who underwent ACL repair with internal brace ligament augmentation. Given a low incidence of reintervention (6%) and meaningful improvements in functional knee measurements, the authors concluded that the repair outcomes were comparable with those in the current literature after primary ACL reconstruction. However, these previous reports are limited by the absence of an appropriate comparison group as well as by heterogeneity in age and activity level. 12 In our adolescent population, the risk of failure was significantly higher among patients who underwent ACL repair versus QPA reconstruction. To our knowledge, this study is the first to follow a cohort of adolescent patients who underwent ACL repair and to compare the outcomes with those who underwent ACL reconstruction.
One factor known to affect ACL healing is the time from injury to surgery. Previous reports suggested that the native ACL is often amenable to repair within 3 months of injury when the remnant is high-quality tissue. Surgical delay may lead to contraction of the torn ends of the ligament and thereby more gapping at the repair site. 20 In the current study, the time from injury to surgery was not associated with surgical failure in either group (Table 1). This is likely due to strict adherence to the requirements for attempting ACL repair. Future research should aim to further investigate the time from injury to surgery as an indication or contraindication for ACL repair.
Among patients who did not suffer a second ACL injury, IKDC and Lysholm scores were comparable between the groups. Given that patient-reported knee function may be comparable between the techniques when successful, it remains important for clinicians to weigh potential benefits of ACL repair against the likelihood of surgical failure. Murray and Fleming 22 demonstrated prevention of arthritic changes after ACL repair in a porcine model. The potential for these results to be translatable to humans is reason to continue research in this area.
The present study is limited by a relatively small sample size in the SLA ACL repair group and the nonrandomized nature of treatment selection. The present study does not address long-term outcomes. However, the high short-term failure rate of ACL repair remains important to consider. We chose to include consecutive patients with a minimum of 6 months after surgery because of the high early failure rate (Figure 3). To maintain a comparable follow-up period between the groups, we did not consider participants past the first 3 years after ACL surgery. The strengths of this study include the prospective questionnaire component and the inclusion of a single orthopaedic surgeon’s patient population.
MRI information was not available to include in this study. Because of cost, no patients had postoperative MRI at follow-up clinic visits unless they sustained a new injury. Future prospective studies should include MRI data to determine the course of ACL graft/repair healing. This information would be particularly useful in determining if joint stability among those who underwent ACL repair was due to the SLA or a healed ACL.
Overall, the risk of surgical failure was significantly higher in the ACL repair group as compared with the QPA reconstruction group. The high risk of failure is an important surgical consideration, as research showed that revision ACL surgery may lead to worse outcomes than primary ACL reconstruction.10,17,38,39 ACL repair remains an alternative to ACL reconstruction and has the potential to preserve native kinematics and prevent the progression of osteoarthritis. However, ACL repair is an evolving technique. On the basis of current evidence, we encourage surgeons to consider the high rate of early failure associated with SLA ACL repair when choosing the optimal treatment for adolescent patients with a ruptured ACL.
Supplemental Material
Supp_ICMJE_Disclosure_Form – Supplemental material for ACL Repair With Suture Ligament Augmentation Is Associated With a High Failure Rate Among Adolescent Patients
Supplemental material, Supp_ICMJE_Disclosure_Form for ACL Repair With Suture Ligament Augmentation Is Associated With a High Failure Rate Among Adolescent Patients by Alexia G. Gagliardi, Patrick M. Carry, Harin B. Parikh, Jessica L. Traver, David R. Howell and Jay C. Albright in The American Journal of Sports Medicine
Footnotes
Presented at the 44th annual meeting of the AOSSM, San Diego, California, July 2018.
One or more of the authors has declared the following potential conflict of interest or source of funding: J.C.A. has received speaker’s fees, travel, education, and hospitality payments from Arthrex; education payments from Wardlow Enterprises and Gemini Medical; and travel and hospitality payments from DJO, LLC. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
