Athletes with Bone-Patellar Tendon Bone Autograft for ACL Reconstruction Were Months Slower to Meet Rehabilitation Milestones and Return to Sport Criteria Than Athletes with Hamstring Tendon Autograft or Soft Tissue Allograft – Secondary Analysis from the ACL-SPORTS Trial

Abstract

Objective:

Graft choices for athletes undergoing anterior cruciate ligament reconstruction (ACLR) include bone-patellar tendon-bone (BPTB) and hamstring tendon (HT) autografts, and soft tissue allografts (ALLO). The objective was to assess time to meet clinical milestones by graft type in athletes who completed a RTS program after ACLR.

Design:

Retrospective cohort study

Methods:

79 athletes enrolled after ACLR (ALLO n=18, BPTB n=24, HT n=37). Time from surgery to meet 1) enrollment criteria (≥12 weeks post-op, ≥80% isometric quadriceps strength index (QI), minimal effusion and full knee range of motion (ROM), and 2) RTS criteria (≥90% QI, hop testing limb symmetry, and patient reported outcomes) was calculated. Quadriceps strength, hop performance and patient-reported outcomes were measured before and after training, and at one year post-operative. Descriptive statistics, Chi-square tests, and one-way ANOVAs (α=.05) were used to analyze differences among graft types.

Results:

On average, the BPTB group (28.5±7.6 weeks) took longer to meet enrollment milestones than the HT (22.5±7.6, p=.007) and ALLO (18.9±5.8, p<.001) groups. The BPTB group (44.7±15.8 weeks) took longer from surgery to meet RTS criteria than the HT (32.5±9.9, p=.001) and ALLO (29.3±9.0, p<.001) groups. QI after training for the BPTB group (86.2±11.4) was lower than the HT (96.1±12.9, p=.004) and ALLO (96.9±5.9, p=.009) groups.

Conclusions:

Athletes with a bone-patellar tendon-bone autograft may take longer than athletes with hamstring tendon autograft or a soft tissue allograft to complete post-operative rehabilitation, recover quadriceps strength and meet RTS criteria.

INTRODUCTION

Although ACLR is one of the most common sports medicine orthopedic procedures, graft selection for the surgery is still a highly debated topic, particularly for athletes returning to sport. Many factors can contribute to graft choice, including age, sex, athlete/surgeon preference, return to activity goals, patient outcomes and risk of graft failure. The most common autograft choices are bone-patellar tendon-bone (BPTB) and quadrupled hamstring tendon (HT).^16,27,53

BPTB autografts have been the gold standard for reconstruction²⁶ due to graft stability associated with bone-to-bone fixation.⁸ However, BPTB grafts are linked to anterior knee pain, particularly with kneeling,^20,45 and clinical, radiographic, and histologic abnormalities at the donor site.²⁶ HT autografts have grown in favor due to larger graft diameter and sparing of the knee extensor mechanism.^11,18 Although, HT autografts are associated with complications including tunnel widening, problematic fixation, increased laxity, and poor post-operative functioning of the harvested hamstring.^7,18

The use of allografts rose over the past twenty years due to the lack of donor site morbidity and reduced operative time.^6,47 Allografts have fallen out of favor in the past decade because of a higher risk of graft failure compared to autografts among young, active individuals.^24,25,47 Other risks associated with allograft use include delayed bone integration and infection.¹¹

Medium- and long-term outcomes in strength, range of motion (ROM), knee stability, subjective reports, and patient performance are similar between HT and BPTB autografts.^10,18,20,28 Similar results have also been reported in pain, laxity, ROM, patient-reported outcomes and return to preinjury activity when comparing autografts to allografts.³⁹ The impact of graft choice on outcomes including timelines to meet clinical milestones (e.g. return to sport (RTS) criteria) or response to specific rehabilitation protocols is unclear. Clinicians need this information as they strive to best educate patients regarding post-operative outcomes and timelines, maximize results during rehabilitation, and help athletes prepare for the demands of RTS.

Given the dramatic influence that early return to sport has on second injury risk,^9,17,30 comparing rehabilitation timelines and clinical outcomes at different post-operative stages is important. Therefore, the primary purpose of this analysis was to investigate the time to meet post-operative clinical milestones (enrollment and RTS criteria) according to graft type in athletes who enrolled in a RTS program as part of the ACL Specialized Post-Operative Return to Sports (ACL-SPORTS) clinical trial. Our secondary objective was to compare clinical (functional and patient-reported) outcomes among these athletes by graft type.

METHODS

Participants

Seventy-nine athletes (40 men, 39 women) between the ages of 13 and 55 years, who underwent primary ACLR were included in this secondary analysis of a prospective randomized control trial (parent study registered at Clinicaltrials.gov, NCT01773317). Prior to inclusion, all participants gave written informed consent (or assent if younger than 18 years with parent/guardian written informed consent). Participants were recruited from the University of Delaware Physical Therapy Clinic, and the community via physician and physical therapist (PT) referral, newspaper advertisement and word of mouth. All participants were Level 1 (n=72) or Level 2 (n=7) athletes¹² (minimum 50 hours per year) prior to surgery and intended to return to their prior level of sporting activity following ACLR.

Thirty different orthopedic surgeons performed surgery using the most common graft types, including hamstring tendon (HT) autograft, bone-patella tendon-bone (BPTB) autograft or soft tissue allograft (ALLO). Potential participants were excluded if they: 1) did not meet enrollment criteria by 9 months, 2) had a history of ACLR, 3) had a history of a significant lower extremity injury or surgery or 4) had a grade III concomitant ligamentous injury or large osteochondral defect (>1 cm²). Early post-operative rehabilitation took place at multiple clinics and was not controlled in order to form a more generalizable population. Strict enrollment criteria were used to ensure a homogenous population at the time of training.

Enrollment Criteria

Participants were enrolled in the ACL-SPORTS clinical trial between 3 and 9 months (mean ± standard deviation [SD]: 23.5 ± 8.0 weeks) after surgery, when they met all enrollment criteria. Prior to enrollment, participants had to achieve the following clinical milestones: 1) full knee ROM, 2) minimal to no knee effusion, and 3) quadriceps strength index (QI) ≥ 80%. Participants also had to complete a walk/jog progression prior to enrollment to assess the ability of the athlete’s knee to tolerate increased loading without increased knee soreness or effusion. These stringent enrollment criteria ensured each participant was safe and prepared to begin the higher-level activity required in the training program.

Interventions

All participants completed 10 sessions of a comprehensive RTS program over a 5–7 week period (1–2 sessions per week), either with or without perturbation training¹⁴. Our program was developed based on the primary ACL injury prevention literature^3,49 and included progressive quadriceps strengthening, neuromuscular training activities, plyometrics, and agility exercises. Participants completed the RTS program under one-on-one supervision from a licensed physical therapist in the University of Delaware’s Sports & Orthopedic Physical Therapy Clinic.

Participants progressed through the 10-session ACL-SPORTS protocol;⁵¹ physical therapists used soreness rules¹³ and monitored effusion⁴⁶ to guide clinical decision-making for appropriate progression. Appropriate landing form and lower extremity biomechanics were emphasized throughout the program with feedback from the treating physical therapist. Training variables and progression at each session were adapted based on participant performance. The final stage of the program focused on incorporating distractions and sport-specific skills within the activities of the treatment protocol (e.g., stick handling with direction changes for lacrosse player).

Half of the cohort also received 10 sessions of perturbation training. There were no differences following training in quadriceps strength, hop tests, and self-reported function between the participants in our cohort who received perturbation training and those who did not.^4,5 Thus, the present study did not include the perturbation training variable (i.e. perturbation versus no perturbation) in the analyses.

If participants did not meet the University of Delaware’s RTS criteria at the post-training time point they were educated on a continued home program to address remaining impairments, and seen for additional sessions if deemed necessary by the testing physical therapist. Participants returned for follow-up testing until all RTS criteria were met: ≥ 90% for each of QI, LSI on all 4 hop tests, KOS-ADLS, and GRS. Participants returned for testing again at one year after surgery.

Functional Testing and Patient-Reported Outcomes

At the time of enrollment (pre-training) and after finishing the RTS program (post-training), all participants completed a clinical test battery that included: isometric quadriceps strength, single-leg hop testing, Knee Outcome Survey-Activities of Daily Living Scale (KOS-ADLS), and Global Rating Scale (GRS). Additional patient-reported outcome measures collected included the International Knee Documentation Committee 2000 (IKDC 2000) and Knee injury and Osteoarthritis Outcome Score (KOOS).

Quadriceps strength was assessed via maximal voluntary isometric contraction (MVIC) using an electromechanical dynamometer (Kincom; DJO Global, Chula Vista, CA, USA or Biodex, Shirley, NY, USA). MVICs were collected with participants seated with hips and knees flexed to 90 degrees and the axis of rotation of the knee attachment aligned with axis of rotation of the knee.⁴⁴ The participant’s quadriceps index (QI) was calculated by dividing the MVIC of the involved side by the MVIC of the uninvolved side and multiplying by 100.

Hop testing, a reliable and valid performance-based outcome measure following ACL reconstruction,^36,40 was completed using a series of single-limb hop tests as described by Noyes et al.³⁵ This sequence of tests was performed on a line 6 meters long and 15 cm wide.³¹ We measured the single hop (SHP) for distance, the crossover hop (XHP) for distance, the triple hop (TRHP) for distance and the 6-meter timed hop (TIMHP). Each hop was completed on the uninvolved limb first, with two practice trials followed by two measured trials; this sequence was repeated on the involved limb. The mean distance of the two measured trials was calculated for each leg for the single, crossover, and triple hops. We calculated limb symmetry index (LSI) by computing the ratio of the mean distance on the involved limb to the mean distance on the uninvolved limb, multiplied by 100. For the timed hop, the LSI was calculated as a ratio of the uninvolved limb mean time over the involved limb mean time, multiplied by 100.

Participants completed several patient-reported outcomes measures to assess knee symptoms and function. The KOS-ADLS is a valid, reliable, and responsive measurement tool to assess the functional limitations for a varied population with knee injuries and impairments.²³ The GRS is a single question that asks the patient to rate the function of his/her knee on a scale of 0 to 100 with 100 being the function before injury.²² The IKDC is a measure of knee-specific symptoms, function and sports activities, and is valid and reliable for a variety of knee conditions including ACL injury.²¹ The KOOS is a reliable measure often used in the ACL population^2,15,32,52 consisting of 5 subscales assessing patient symptoms, complaints of pain, function in daily life, function during sports and recreational activities, and knee-related quality of life.⁴¹

Statistical Analysis

Statistical analyses were performed using SPSS version 25 (Microsoft, Redmond, Washington, USA). Primary variables of interest were the time from surgery to meet enrollment criteria and the time from surgery to meet RTS criteria. Functional and patient-reported clinical outcomes of secondary interest were: QI, LSI on each hop test, KOS-ADLs, GRS, IKDC and KOOS scores at pre-training, post-training, and one year after ACLR. Demographic characteristics and surgical details were compared across groups using one-way analysis of variance (ANOVA) and Chi-Square tests of proportions for continuous and categorical variables, respectively. We used two one-way ANOVAs with and without the covariates of age and sex to compare across groups our two primary outcomes of interest: 1) time to meet enrollment criteria and 2) time to meet RTS criteria. We used one-way ANOVAs to compare clinical and functional outcomes across groups (secondary analyses). Post-hoc t-tests with Bonferroni corrections were used to test differences between groups if the p-value for the model was statistically significant. A priori statistical significance was set at α = 0.05 for all analyses. No a priori power calculation was performed because this was a secondary analysis of a pre-existing data set; all available data were used in the present study, and the primary analyses of the present study were adequately powered.

RESULTS

Seventy-nine participants completed the RTS program (ALLO=18, BPTB=24, HT=37) (TABLE 1). The ALLO group was older than the BPTB and HT groups (post-hoc p<.001), but HT and BPTB groups did not differ in age (p=1.000). There was an interaction in distribution of sex and graft type, with more men in the ALLO group and more women in the BPTB group. There were no between-group differences for BMI, training group, medial or lateral meniscus treatment, or pre-injury level of sport.

TABLE 1.

Athletes Demographics and Graft Type Groups

	ALLO (n=18)	BPTB (n=24)	HT (n=37)	Model p-value
Age at surgery (yrs)	30.5±10.5	18.3±3.2	18.5±3.3	<.001
BMI (kg/m2)	27.1±3.3	25.9±3.4	25.6±3.2	.298
Sex (F/M)	5/13	16/8	18/19	.044
Training Group (SAPP/SAPP+Pert)	10/8	13/11	17/20	.734
Medial Meniscus Treatment*	10 none, 5 partial meniscectomy, 2 repair	13 none, 1 partial meniscectomy, 3 repair	22 none, 8 partial meniscectomy, 7 repair	.479
Lateral Meniscus Treatment*	10 none, 6 partial meniscectomy, 1 repair	8 none, 7 partial meniscectomy, 2 repair	19 none, 13 partial meniscectomy, 5 repair	.915
Pre-Injury Level of Sport	16 level I, 2 level II	22 level I, 2 level II	34 level I, 3 level II	.929

Abbreviations: ALLO, allograft; BPTB, bone-patella tendon-bone autograft; HT, hamstring tendon autograft; BMI, Body Mass Index; F, female; M, male; SAPP, Strength, Agility, Plyometric and Secondary Prevention; SAPP+Pert, Strength, Agility, Plyometric and Primary Prevention plus Perturbation Training

^*Meniscus treatment data was not available for 8 athletes

Time to Meet Post-Operative Clinical Milestones

There was a significant effect of graft type for the time from surgery to meet enrollment criteria in the unadjusted model (p < 0.001); graft type remained the only significant predictor (p = 0.005) in the full model including the covariates of age (p = .789) and sex (p = .237). There was a significant effect of graft type for the time from surgery to meet RTS criteria in the unadjusted model (p < 0.001); graft type was the only significant predictor (p = 0.001) in the full model including the co-variates of age (p = .804) and sex (p = .973).

The BPTB group took longer to meet post-operative clinical milestones for enrollment and RTS (FIGURE 1). The BPTB group took 6 weeks longer than the HT group (p=.007) and 9.5 weeks longer than the ALLO group (p<.001) to meet the clinical milestones for enrollment. There was no difference (p=.283) in time to enrollment between HT and ALLO. Seventy-five participants met all RTS criteria at some point after finishing the RTS prevention program. Of the 4 who did not meet the criteria, 3 had BPTB grafts and 1 had a HT graft. For time from surgery to meet RTS criteria, the BPTB group took 12 weeks longer than HT (p=.001) and 15.5 weeks longer than ALLO (p<.001), but there was no difference (p=1.000) between HT and ALLO.

FIGURE 1.

Time to Meet Enrollment and RTS Criteria by Group. Abbreviations: RTS, Return to Sport; ALLO, allograft; BPTB, bone-patella tendon-bone autograft, HT; hamstring tendon autograft

Clinical Outcomes: Functional Testing and Patient Reported Outcome Measures

There were group differences in clinical outcome measures at enrollment (APPENDIX A) and post-training (APPENDIX B), but not at one year after ACLR (APPENDIX C).

There were differences between groups for all hop tests and the KOS. There were no differences between groups for the GRS, IKDC, or any of the KOOS sub-scales.

After completing the RTS program, 79 participants completed strength testing and patient-reported outcome measures. Sixty-four participants completed post-training hop testing; testing was deferred for 15 participants due to knee effusion, QI < 80%, and/or pain (TABLE 2a). The QI for BPTB group was lower than the HT (p=.004) and ALLO (p=.009) groups, but there was no difference (p=1.000) between HT and ALLO (FIGURE 2). The BPTB group had a lower KOS score than the HT group (p=.007); there were no differences between ALLO and BPTB (p=.479) or ALLO and HT (p=.548) groups.

TABLE 2a.

Reasons by Group for Not Completing Hop Testing at Post Training

ALLO (n=0)	BPTB (n=8)	HT (n=7)
N/A	Pain: 1 Quad strength/activation: 4 Combination of pain, quad strength and/or effusion: 3	Effusion: 3 Quad strength/activation: 4

Abbreviations: N/A: Not applicable; ALLO, allograft; BPTB, bone-patella tendon-bone autograft, HT; hamstring tendon autograft

Figure 2.

Quad Index by Group at Enrollment, Post Training and One Year. Abbreviations: QI, Quad Index; Abbreviations: ALLO, allograft; BPTB, bone-patella tendon-bone autograft, HT; hamstring tendon autograft

Sixty-six participants completed all testing (strength, hop testing, and patient reported outcomes) at one year post-ACLR. Among participants with incomplete data, 10 completed all testing with the exception of hop testing, which was deferred due to pain, decreased QI, increased effusion or second injury (Table 2b). One participant only completed patient-reported outcome measures as she lived out of the area. Two participants were lost to follow-up and did not complete any testing at one year. There were no between group differences for any clinical variable at one year.

TABLE 2b.

Reasons by Group for Not Completing Hop Testing at One Year

ALLO (n=1)	BPTB (n=3)	HT (n=6)
2nd injury: 1	Effusion: 1 Quad strength/activation: 1 Pain: 1	Effusion: 1 Quad strength/activation: 2 2nd Injury: 3

Abbreviations: ALLO, allograft; BPTB, bone-patella tendon-bone autograft, HT; hamstring tendon autograft

DISCUSSION

The objective of this secondary analysis was to compare timelines to meet important post-operative clinical milestones between athletes with different ACLR graft types who enrolled in a RTS program. Athletes with BPTB grafts took longer to meet enrollment and RTS criteria than those who received a HT autograft or a soft tissue allograft. Accounting for age and sex, graft type was the discriminating factor, suggesting that graft type has an impact on rehabilitation and return to play timelines after ACLR.

Our enrollment criteria encompassed important late stage clinical markers that identify when a patient is ready to move on to higher level activity such as running, agility and plyometrics. The BPTB group took 1.5 months longer from ACLR than the HT group and 2.5 months longer than the ALLO group to achieve these milestones, indicating that those with BPTB graft may take longer to recover full knee ROM, reduce effusion, and recover quadriceps strength after ACLR. The BPTB group did not meet RTS criteria until over 10 months after surgery; the ALLO and HT groups met these criteria at 6.5 and 7.5 months post-operative, respectively. The time difference could not simply be attributed to the BPTB group starting the training later than the other groups, but was also due to the additional time it took after completing the RTS program to meet these RTS criteria.

Our secondary aim was to assess differences between graft type with regard to functional and patient-reported outcomes at different stages of rehabilitation (enrollment, post-training and one-year post-operative). The BPTB group had a lower QI after completing the RTS program, suggesting that athletes with a BPTB graft may require additional time or interventions to restore quadriceps strength after ACLR. Clinicians should consider continuing unilateral quadriceps strengthening in the RTS phase of rehabilitation, focusing on providing appropriate (heavy resistance) loading to facilitate strength gains in athletes with a BPTB graft. By one year, outcomes did not differ between groups, indicating that the BPTB group ‘caught up’ to their peers in the time period between post-training and one year.

Our results have important clinical implications for graft selection as post-operative outcomes and timelines for rehabilitation may inform an athlete’s graft selection for ACLR. If an athlete is considering an accelerated RTS following ACLR, opting for an allograft or HT autograft may facilitate faster rehabilitation and return to play. However, in light of growing evidence, RTS timelines and risk of second injury need to be considered. At least 1 in 4 young athletes experience a second ACL injury following RTS,^37,38,48 and these injuries usually occur early after return with minimal athletic exposure.³⁸ For every 1 month delay in RTS up to 9 months, the rate of knee re-injury can be halved.¹⁷ In light of the timeframes for each graft type meeting RTS criteria in this study, the ALLO and HT groups may be at a higher risk of sustaining another knee injury when they return to sport as they meet RTS criteria earlier than the BPTB group.

Graft rupture rates may not be equivalent between graft types, particularly among young athletes. The evidence is clear that the use of an autograft is preferred over an allograft for primary ACLR in young active individuals due to the increased risk of revision with allograft.^6,24,33,47 The trend toward the use of allografts in older patients was reflected in our study by the age difference in the ALLO group versus the HT and BPTB groups. However, our ALLO group was highly active. With regard to differences in autograft choice, similar mid- to long-term clinical, patient-reported and performance outcomes have been reported between HT and BPTB autografts^19,27,53. However, when comparing HT and BPTB grafts, failure rates are slightly higher in HT autografts and there is a lower risk of revision with BPTB autografts.^16,42,43 In the case of our findings, the delay in meeting RTS criteria gives the graft a longer time to incorporate and may contribute to the lower risk of graft rupture with BPTB as these athletes return to play later, at a point when the risk of second injury has declined. Future research should explore relationships between second ACL injury and graft type and timing of RTS.

There are limitations to this study. Post-operative rehabilitation prior to enrollment was not standardized and many surgeons were involved, but this also improves the generalizability. All athletes had to meet common enrollment criteria prior to participating in the RTS program, ensuring a standardized starting point. Although many PTs from the University of Delaware treated the participants in the RTS program, all PTs used the ACL-SPORTS protocol⁵⁰ to progress the athletes throughout the program. Again, this increases generalizability, as any clinician could use the protocol to complete the RTS program. There was an unequal distribution between groups in sex and age, with more men in the ALLO group and more women in the BPTB group; the ALLO group was older than the HS and BPTB groups. However, graft type alone was the discriminating factor when adjusting for sex and age. Although all participants completed the comprehensive RTS program, half also received additional perturbation training. Participants who received perturbation training were equally distributed among the graft types. Previously published data from our clinical trial indicate that there are no differences in clinical or functional outcomes in those completing the RTS program with or without perturbation training at 1 or 2 years after ACLR.^4,5

CONCLUSION

Athletes who received a BPTB autograft took between 1.5 and 2.5 months longer to meet key rehabilitation milestones than athletes who received an allograft or a hamstring tendon autograft, and up to 4 months longer to meet return to sport criteria. The BPTB group also had quadriceps weakness (86% QI) after RTS training whereas the ALLO (97% QI) and HT (96% QI) groups did not. In our cohort, athletes with BPTB graft needed additional time to complete post-operative rehabilitation, recover quadriceps strength and meet RTS criteria, delaying their return to the field.

KEY POINTS.

Findings:

Athletes who received a BPTB autograft took longer than those who received a soft tissue allograft or a hamstring tendon autograft to reach important post-operative clinical milestones, including recovering quadriceps strength and meeting RTS criteria.

Implications:

The increased time to achieve post-operative clinical and RTS milestones means athletes with BPTB grafts returned to sport later after ACLR than athletes with other graft types; this delay may help protect these athletes from second ACL injury.

Caution:

ACLR was completed by many (30) orthopedic surgeons, and early rehabilitation was not controlled prior to enrollment. All participants were athletes planning to return to Level I-II sporting activity, which may limit the generalizability of these findings to all athletic populations.

APPENDIX A.

Enrollment (Pre-Training) Clinical Outcomes

	ALLO	BPTB	HT	Model p- value
SHP (LSI)	75.8±18.6	73.6±14.5	84.5±11.5	.010
XHP (LSI)	82.8±14.3	81.0±16.4	90.5±13.6	.034
TRHP (LSI)	83.7±13.8	82.4±11.4	90.4±10.4	.020
TIMHP (LSI)	89.7±10.7	88.5±8.7	94.7±7.8	.018
KOS (%)	92.1±4.8	90.4±7.2	94.9±5.6	.017
GRS (%)	77.8±7.5	78.4±7.1	80.4±10.8	.529
IKDC (%)	76.9±7.4	77.4±10.7	81.0±6.6	.124
KOOS_Pain (%)	89.4±6.6	90.7±7.3	93.0±6.8	.156
KOOS_Symptom (%)	84.9±9.1	86.5±8.8	83.0±10.3	.388
KOOS_ADL (%)	96.7±3.3	97.1±3.7	98.3±3.3	.217
KOOS_Sport/Rec (%)	77.8±13.7	74.8±16.4	83.0±12.9	.086
KOOS_QoL (%)	55.6±10.0	54.7±16.4	62.3±17.6	.130

Abbreviations: Abbreviations: ALLO, allograft; BPTB, bone-patella tendon-bone autograft; HT, hamstring tendon autograft; LSI, Limb Symmetry Index; SHP, Single hop for distance; XHP, Crossover hop for distance; TRHP, Triple hop for distance; TIMHP, Timed hop; KOS, Knee Outcome Survey Activities of Daily Living; GRS, Global Rating of Perceived Function; IKDC, International Knee Documentation Subjective Knee Form 2000; KOOS, Knee Injury and Osteoarthritis Outcome Score; ADL, Activities of daily living; QoL, Quality of life

APPENDIX B.

Post-Training Clinical Outcomes

	ALLO	BPTB	HT	Model p- value
SHP (LSI)	89.5±13.3	92.0±11.0	96.6±9.8	.100
XHP (LSI)	96.4±6.6	95.5±8.0	97.7±7.6	.601
TRHP (LSI)	95.1±5.1	94.0±7.1	98.4±6.2	.047
TIMHP (LSI)	97.5±5.4	98.9±8.3	101.2±7.2	.206
KOS (%)	94.8±3.8	92.6±6.9	96.9±4.4	.009
GRS (%)	87.3±8.9	87.3±5.5	86.4±9.6	.892
IKDC (%)	84.4±8.5	84.3±10.3	88.9±7.6	.072
KOOS_Pain (%)	92.3±5.8	93.2±5.8	95.3±6.3	.161
KOOS_Symptom (%)	85.1±8.2	86.0±12.5	84.6±13.0	.898
KOOS_ADL (%)	98.9±1.7	98.8±2.1	98.3±3.2	.610
KOOS_Sport/Rec (%)	87.5±10.6	87.5±13.5	90.3±12.0	.604
KOOS_QoL (%)	67.7±12.9	60.9±21.8	71.1±18.4	.117

Abbreviations: Abbreviations: ALLO, allograft; BPTB, bone-patella tendon-bone autograft; HT, hamstring tendon autograft; LSI, Limb Symmetry Index; SHP, Single hop for distance; XHP, Crossover hop for distance; TRHP, Triple hop for distance; TIMHP, Timed hop; KOS, Knee Outcome Survey Activities of Daily Living; GRS, Global Rating of Perceived Function; IKDC, International Knee Documentation Subjective Knee Form 2000; KOOS, Knee Injury and Osteoarthritis Outcome Score; ADL, Activities of daily living; QoL, Quality of life

APPENDIX C.

One-Year Clinical Outcomes

	ALLO	BPTB	HT	Model p-value
SHP (LSI)	94.8±9.7	97.6±7.0	99.5±7.3	.153
XHP (LSI)	99.7±6.6	98.1±6.2	100.1±9.0	.656
TRHP (LSI)	98.9±7.5	97.4±5.5	99.4±6.3	.548
TIMHP (LSI)	102.6±5.6	102.3±5.3	102.0±7.0	.960
KOS (%)	95.6±5.9	95.5±5.5	98.0±3.0	.075
GRS (%)	93.1±8.8	91.9±13.6	95.4±6.7	.370
IKDC (%)	91.6±9.9	90.4±10.2	94.6±7.4	.181
KOOS_Pain (%)	95.1±5.0	95.5±4.7	97.0±4.2	.270
KOOS_Symptom (%)	86.5±12.3	90.1±11.3	90.4±9.7	.441
KOOS_ADL (%)	98.4±4.0	99.0±1.9	99.6±1.0	.185
KOOS_Sport/Rec (%)	93.1±9.6	91.3±10.7	94.7±8.9	.412
KOOS_QoL (%)	82.6±16.4	76.4±13.5	82.0±17.3	.347

Abbreviations: ALLO, allograft; BPTB, bone-patella tendon-bone autograft; HT, hamstring tendon autograft; LSI, Limb Symmetry Index; SHP, Single hop for distance; XHP, Crossover hop for distance; TRHP, Triple hop for distance; TIMHP, Timed hop; KOS, Knee Outcome Survey Activities of Daily Living; GRS, Global Rating of Perceived Function; IKDC, International Knee Documentation Subjective Knee Form 2000; KOOS, Knee Injury and Osteoarthritis Outcome Score; ADL, Activities of daily living; QoL, Quality of life