Abstract Purpose To investigate the association between autograft size and graft failure for hamstrings tendon (HT), patellar tendon (PT) and quadriceps tendon (QT) autografts. Methods Medline, PubMed, Cochrane Library, Embase, Amed and Web of Science were searched at four separate time points, with the most recent search in February 2025. Eligible studies included patients with primary anterior cruciate ligament reconstruction (ACLR) with (graft failures) or without (survivals) graft failure who had specified autograft size for primary ACLR. Graft failure was defined as a re‐rupture of the reconstructed ACL and/or positive pivot shift. Standardised mean differences were calculated for continuous variables, and odds ratios expressed with 95% confidence interval for dichotomous variables of autograft size for survivals versus graft failures. Risk of bias was assessed with RoBANS 2/RoB2 and certainty of evidence with GRADE. Results In total, 46,268 patients, of which 43,660 HT autograft, 2410 PT autograft and 198 QT autograft were covered in 35 studies. An HT autograft size of < 7 mm had 81% greater odds for a graft failure compared to ≥ 7 mm ( p = 0.01), <8 mm HT autograft size had 46% greater odds for a graft failure compared to ≥ 8 mm ( p < 0.0001), <9 mm HT autograft size had 34% greater odds for a graft failure compared to ≥ 9 mm ( p = 0.001). No significant odds for a graft failure were observed for patients with ≥ 10 mm HT autograft size compared to < 10 mm, or for patients with PT autograft. Only two of the included studies provided data on the QT autograft where none presented standard deviation, thus these studies could not be pooled and were presented qualitatively. Conclusion Patients treated with a smaller HT autograft size have greater odds for a graft failure compared to patients with a greater autograft size. There was no association between autograft size for patients treated with PT autograft and graft failure. Surgeons should consider autograft size when performing ACLR as the size of HT autografts influences the risk of a graft failure. Level of Evidence Level IV.
Simonson et al. (Mon,) studied this question.