Allograft Augmentation of Hamstring Anterior Cruciate Ligament Grafts is Associated with Increased Graft Failure

OBJECTIVES: ACL reconstruction in adolescents is commonly performed with hamstring autografts. In the adolescent population with very high activity levels, graft rupture is the most feared complication of ACL reconstruction. Young age, higher activity level, allografts, and small graft diameter have been shown to be predictors of graft failure. The applicability of this data to pediatric ACL reconstructions is limited due to heterogeneity of ages, graft constructs, and tunnel techniques. The purpose of this study is to evaluate the association of soft tissue graft constructs and graft rupture following pediatric transphyseal ACL reconstruction. Our hypothesis is that allograft-augmentation of grafts is associated with an increased risk of graft rupture. METHODS: A single-institution retrospective review was performed of consecutive patients. Inclusion criteria were age less than 20 years and transphyseal ACL reconstruction with hamstring autograft, with or without allograft augmentation. Graft constructs included 4-strand doubled semitendinosus and gracilis (4-STG), 5-strand tripled semitendinosus and doubled gracilis (5-STG), 6-strand doubled gracilis and semitendinosus plus allograft (6-STGAllo), and 7-strand tripled semitendinosus and doubled gracilis plus allograft (7-STGAllo). Exclusion criteria included multiligament reconstruction and less than 6 months follow-up. The primary outcome was graft rupture. RESULTS: Three hundred fifty-five patients (157 males, 198 females) with an average age of 15.3 years were identified to meet inclusion criteria. Graft constructs included 4-STG (198), 5-STG (91), 6-STGAllo (65), and 7-STGAllo (1). Average graft diameter was 8.3 mm 4-STG, 8.9 mm 5-STG, and 9.2 mm 6-STGAllo. Age and graft sizes were significantly different across groups with older patients (p <0.001) and larger graft sizes (p <0.001) being found in patients with allograft-augmented grafts. Mean duration of follow-up was 26 months (range 6-56 months). There were 51 graft ruptures (14.3%). The failure rate of each construct was 13.6% 4-STG, 11.9% 5-STG, and 19.7% 6-STGAllo. Time to graft failure was 16 months (range 2-40 months), with 49% of failures occurring before 12 months and 24% after 24 months. Twenty-four patients (6.7%) had a contralateral ACL tear during the follow-up period. Table 1 provides population characteristics by graft rupture status. This data suggests that patients who sustain a graft rupture may be slightly younger (p=0.07) and have lower BMI (p=0.07) than those patients without graft rupture. Odds ratios for graft failure by graft construct, controlling for age and graft size were calculated. Patients with 6-STGAllo grafts had an odds ratio of 2.6 (95% CI: 1.02, 6.50) of graft rupture as compared to 4-STG. CONCLUSION: ACL reconstruction with hamstring autograft combined with soft tissue allograft have a 2.6 times risk of graft rupture as compared to hamstring autograft without augmentation. In situations where the surgeon harvests an inadequately sized 4-strand autograft, we recommend tripling the semitendinosus to produce a larger graft diameter rather than augment with an allograft.