Anterior cruciate ligament (ACL) reconstruction is widely used to restore knee stability after injury, but the risk of revision surgery increases when the autograft size is inadequate. Ultrasound (US) measurements of preoperative target tendons have been applied to predict the intraoperative autograft size, with various outcomes across different studies. This systematic review and meta-analysis aimed to summarize the evidence and investigate the usefulness of US in predicting autograft size. Electronic databases were searched for relevant studies from inception to 19 January 2022. The primary outcome was the correlation between the preoperative US measurements of donor tendons and intraoperative autograft size. The secondary outcomes encompassed the predictive performance of US for autograft size and the comparison between US and magnetic resonance imaging (MRI) for preoperative tendon measurements. Nine studies, comprising 249 patients, were enrolled. The preoperative US measurements of the donor tendons demonstrated a significant positive correlation with their intraoperative autograft diameter, with a pooled correlation coefficient of 0.443 (95% confidence interval [CI], 0.266–0.591, p < 0.001) for the gracilis and semitendinosus autograft, 0.525 (95% CI, 0.114–0.783, p = 0.015) for the semitendinosus autograft, and 0.475 (95% CI, 0.187–0.687, p = 0.002) for the gracilis autograft. The pooled sensitivity and specificity of US imaging in predicting the autograft diameter were 0.83 (95% CI 0.57–0.95) and 0.70 (95% CI, 0.36–0.91), respectively. Moreover, no significant differences were observed between US and MRI measurements in predicting the sizes of the gracilis and semitendinosus autografts. Preoperative US measurements of the target tendons were moderately correlated with the intraoperative autograft size. US imaging has a discriminative performance similar to that of MRI in predicting the autograft size. A standardized US scanning protocol is needed for future studies to minimize the variations in tendon measurements across different investigators and increase the comparability of US imaging with intraoperative findings.

No abstract