The purpose of our study is to determine if hamstring autograft size can be predicted preoperatively. We will define a relationship between patient body size (BMI, height, and weight) and harvested graft size, as well as define a relationship between the preoperative MRI cross-sectional area (CSA) of hamstring tendons and harvested graft size. This information will be useful as a tool for preoperative planning in graft choice selection.

The pre-operative MRIs of one hundred and four patients (62M, 42F) who underwent ACL reconstruction using hamstring autografts were analyzed. Cross-sectional area (CSA) of the ST and G was measured on a single axial MRI image at the level of the knee joint. Combined CSA of both tendons was then compared to the diameter of the four-strand hamstring autograft measured intra-operatively. Patient BMI, height and weight was also compared to intraoperative hamstring autograft size. Linear regression analysis was then performed to define the relationship and predictive value of body size on graft diameter.

Mean graft size was 7.4mm (range 6 – 9). Average graft size for men and women, 7.6mm and 7.1mm, respectively. Predicting graft size from BMI: r= 0.29, R2= 0.08. Predicting graft size from height: r= 0.52, R2= 0.27. Predicting graft size from weight: r= 0.5, R2= 0.25. On preoperative MRI, the mean CSA of ST and G was 9.8mm2 (range 5.4 – 17.7) and 4.5mm2 (range 1.8 – 9.4) respectively, with a total CSA of 14.3mm2 (range 8.4 – 25). If the total CSA was greater than 12mm2, a graft of 7.0mm or greater could be predicted 93% of the time, with sensitivity and specificity, 78% and 76%, respectively, and a LR of 3.25.

Body size is a poor predictor of hamstring graft size in ACL reconstruction, and therefore a large patient does not always provide a large graft from harvested hamstring tendons. MRI assessment of hamstring tendons can be a useful tool for preoperative planning, providing a strong predictive value of graft size from a simple calculation.