2.1 Inclusion/exclusion criteria

This prospective cohort study was approved by the Thomas Jefferson University Institutional Review Board prior to the initiation of the study (IRB #11D.303). Skeletally mature patients aged 14–25 years old undergoing primary ACL reconstruction at a single multi-center institution were prospectively identified before surgery between 2010 and 2018. Of note, it is currently the practice of our surgeons to treat patients of this age cohort with BTB autograft, however, during the period of study it was routine practice to give patients the option to make an informed decision for BTB allograft or autograft. Patients (1) less than 14 years old or over 25 years old at the time of surgery, (2) undergoing revision procedures, (3) who preferred reconstruction with hamstring autograft or allograft, and/or (4) undergoing simultaneous or staged posterior cruciate ligament (PCL), medial collateral ligament (MCL), posterolateral corner (PLC), or lateral collateral ligament (LCL) reconstruction were excluded (Fig. 1). Patients with concomitant meniscal pathologies and/or who underwent concomitant meniscal procedures were considered for inclusion. Those identified who met inclusion and exclusion criteria and wished to participate were given the opportunity to choose bone-patellar tendon-bone (BTB) allograft or autograft based on their preferences and after a thorough discussion of the risk and benefit profile by the attending physician through an informed consent process.

Close

Fig. 1 Consort Diagram demonstrating inclusion of patients in final analysis.

Export to PPT

2.2 Surgical technique

Primary ACL reconstruction was conducted under general anesthesia, supplemented with regional anesthesia for all patients. An initial examination under anesthesia documented ligamentous laxity and positive pivot-shift findings. BTB allografts were obtained from the Musculoskeletal Tissue Foundation (MTF, Edison, NJ), using deep-frozen allografts without chemical treatment or gamma irradiation for sterilization. BTB autografts were harvested via a single midline incision at the central one-third of the patellar tendon. Grafts were prepared and sized uniformly. Excess bone from autograft preparation filled patellar and tibial defects, with tendon edges approximated using interrupted #1 vicryl suture and paratenon closure using running 2-0 vicryl suture. Following complete arthroscopic knee examination, ACL reconstruction proceeded. Tibial tunnel drilling occurred via a tibial guide inserted through the anteromedial portal at the anatomic footprint. Femoral tunnel establishment was trans-tibial or through an accessory anteromedial portal, based on surgeon preference. Graft fixation was achieved using bioabsorbable interference screws for both femoral and tibial ends.

2.3 Postoperative rehabilitation

Postoperatively, patients were fitted with a hinged knee brace locked in extension and were given instructions to immediately be weight-bearing as tolerated. All patients followed identical rehabilitation protocols, commencing physical therapy at 10–14 days post-surgery with the aim of achieving 0–100⁰ knee range of motion by postoperative day 14, regardless of concomitant procedures. Criteria for RTS were consistent across all cohorts, requiring a minimum of 6 months of rehabilitation, quadriceps and hamstring strength exceeding 90 % of the opposite leg, single-leg vertical leap reaching at least 50 % of the opposite leg, ability to jog and run at full speed without pain or limp, capability to perform uninhibited acceleration and deceleration drills, and absence of significant quadriceps atrophy, consistent for all included patients.

2.4 Data collection

Chart review was performed on all participating patients to collect demographic and injury information including age, body mass index (BMI), mechanism of injury, and preoperative sport participation. Patients were asked to complete the Marx Activity Rating Scale survey (Fig. 2a) at 6 weeks, 3-, 4.5-, 6-, 9- and 12 months postoperatively which assessed their perceived ability to run, decelerate, cut, and pivot on a scale from zero (ability to perform less than one time per month) to four (ability to perform four or more times per week).¹⁵ Scores from this survey were summative and reported as “Part 1” scores (minimum score of 0, maximum score of 16). A second survey with 12 ‘yes’ or ‘no’ questions (Fig. 2b) evaluated whether patients felt they could perform specific activities; ‘yes’ responses were given a score of one, while ‘no’ responses were given a score of zero and were reported as “Part 2” scores (minimum of 0, maximum of 12). Individual and total scores within each survey were summarized to compare subjective ability to RTS.¹⁵ Electronic surveys were administered and collected using RedCap (Vanderbilt University, Nashville, Tennessee). Operative notes were reviewed to collect details including concomitant pathologies and procedures. Postoperative clinical notes were also reviewed at 6- and 12 months for clinical assessment including range of motion, presence of effusion or joint line tenderness, and evaluation using pivoting, anterior drawer, McMurray, and Lachman tests. Preoperative sport participation was also collected.

Close

Fig. 2a Part 1 Survey given to patients postoperatively at 6 weeks, 3 months, 4.5 months, 6 months, 9 months, and 12 months.

Export to PPT

Close

Fig. 2b Part 2 Survey given to patients postoperatively at 6 weeks, 3 months, 4.5 months, 6 months, 9 months, and 12 months.

Export to PPT

2.5 Statistical analysis

Comparison of demographics and outcomes was done between autograft or allograft BTB ACL reconstruction cohorts. Comparison for continuous data was done using T-tests or Mann-Whitney U tests and comparison for categorical data was analyzed using Fisher's exact test. A p-value of < 0.05 was deemed significant. All statistical analysis was done using R Studio (Version 3.6.3, Vienna, Austria).

3.1 Part 1 (Marx Activity Rating Scale) results

Members of the allograft cohort have a significantly higher mean Part 1 score at the 6 month time point (13.7 ± 3.22 vs. 11.0 ± 4.66 in the autograft group, P = .032) (Fig. 3, Table 2a). However, at no other survey time before or after was there a significant difference in subjective ability between groups.

Close

Fig. 3 Plotted results of Part 1 survey, Allo = Allograft group, Auto = Autograft group.

Export to PPT

Higher scores for ability to cut were recorded in patients in the autograft cohort at 3 months (0.68 ± 1.12 of a maximum of 4 in the autograft group, vs 0.07 ± 0.27 in the allograft group, P = .003). In contrast to this, at 6 months, higher scores for ability in running, cutting, and decelerating were measured in the allograft cohort (3.77 ± 0.44 vs 3.31 ± 1.04, P = .033), (3.15 ± 1.21 vs 2.28 ± 1.54, P = .048), and (3.62 ± 0.65 vs 2.81 ± 1.37, P = .008) respectively. However, at no other follow-up time were significant differences observed (Table 2b).

3.2 Part 2 results

Part 2 survey scores were summarized and individual responses were analyzed. There were no significant differences between groups at any time point with regard to either total scores (Fig. 4, Table 3a) or responses to individual questions (Table 3b).

Close

Fig. 4 Plotted results of Part 2 survey, Allo = Allograft group, Auto = Autograft group.

Export to PPT

3.3 Clinical outcomes

Three patients (6.98 %) who received autograft and 1 patient (6.25 %) who received allograft went on to graft failure requiring revision ACL reconstruction. The three autograft revisions patients had returned to regular recreational or elite level of sport (basketball or lacrosse) prior to re-injury. None had any reports of laxity or prior graft injuries. These patients experienced retears while playing their respective sports between 2.5 and 4 years following initial surgery. All underwent revision ACL reconstruction with BTB allografts. The member of the allograft cohort who underwent reconstruction had also returned to regular level of basketball participation and sustained a retear again while participating approximately 3 years following initial ACL reconstruction. Notably, this patient had experienced a partial sprain of their ACL while playing basketball 12 months after initial surgery (and 2 years prior to complete graft retear) for which they underwent successful nonoperative treatment with rehabilitation.

Postoperative clinical notes at 6 months were available for 50 patients. Follow-up evaluation at this time did not reveal any significant differences among members of either cohort. Clinical evaluation at 12 months was available for 31 patients. At this time, one patient within the allograft cohort who sustained a partial ACL sprain two weeks prior to evaluation while playing basketball (confirmed with MRI) had resultant 1A laxity on Lachman exam. This patient went on to full recovery with nonoperative treatment including rehabilitation. Following total recovery this patient later sustained a graft tear while playing basketball 3 years following initial surgery.

4.1 Limitations

This study has several limitations that must be considered. Firstly, there were incomplete patient survey responses at each follow-up time point, (6 weeks–89.8 %, 3 months–89.9 %, 4.5 months–79.7 %, 6 months–83.1 %, 9 months–74.6 %. 12 months–66.1 %) which could introduce selection bias given the relative differences in cohort sizes. Moreover, the study's design allowed participants to choose their graft type (BTB allograft vs. autograft), potentially introducing selection bias as certain participants may have preferred one graft type over another. While demographic characteristics were similar between groups, a larger randomized cohort would help eliminate any selection preference. Additionally, the inclusion of both anteromedial and trans-tibial drilling surgical techniques may introduce variability in postoperative recovery and perceived ability to return to sport, which could be addressed in future studies with larger sample sizes. Furthermore, the study may have been underpowered to detect statistically significant differences due to its small sample size, although similar trends have been observed in other patient-reported outcome scores (Lysholm,^16,24 Tegner activity scale,²⁵ and International Knee Documentation Committee (IKDC) scores ¹¹). Moreover, attempting to draw correlations between a patient's early subjective readiness to return to sport and subsequent ACL graft failure rates may be limited by the lack of ACL re-ruptures within the first postoperative year and the small number of recorded re-ruptures overall. Finally, changes in practice during the investigation period, particularly the elimination of the option for BTB allograft in this demographic, may have contributed to the small sample size of patients in this study.