Background Hamstring autografts with a diameter of less than 8 mm for ACL reconstruction have an increased risk of failure, but there is no consensus regarding the best method to predict autograft size in ACL reconstruction. Questions/purposes (1) What is the relationship between hamstring cross-section on preoperative MRI and intraoperative autograft size? (2) What is the minimum hamstring tendon cross-sectional area on MRI needed to produce an autograft of at least 8 mm at its thickest point? Methods This was a retrospective cohort study of 68 patients. We collectively reviewed patients who underwent ACL reconstruction by three separate fellowship-trained surgeons at the Carilion Clinic between April 2010 and July 2013. We searched the patient records database of each surgeon using the keyword “ACL”. A total of 293 ACL reconstructions were performed during that time period. Of those, 23% (68 patients) had their preoperative MRI (1.5 T or 3 T magnet) performed at the Carilion Clinic with MRI confirmation of acute total ACL rupture. Exclusion criteria included previous ACL reconstructions, multiligamentous injuries, and history of acute hamstring injuries. After applying the exclusion criteria, there were 29 patients in the 1.5 T magnet group and 39 in the 3 T group. Median age (range) was 29 years (12 to 50) for the 1.5 T group and 19 years (9 to 43) for the 3 T group. The patients were 41% female in the 1.5 T group and 23% female in the 3 T group. Use of 1.5 T or 3 T magnets was based on clinical availability and scheduling. The graft’s preoperative cross-sectional area was compared with the intraoperative graft’s diameter. The MRI measurements were performed by a single musculoskeletal radiologist at the widest point of the medial femoral condyle and at the joint line. Intraoperative measurements were performed by recording the smallest hole the graft could fit through at its widest point. Pearson’s correlation coefficients were calculated to determine the relationship between graft size and tendon cross-sectional area. A simple logistic regression analysis was used to calculate the cutoff cross-sectional areas needed for a graft measuring at least 8 mm at its thickest point. Intrarater reliability was evaluated based on re-measurement of 19 tendons, which produced an overall intraclass correlation coefficient (ICC) of 0.96 95% (CI 0.93 to 0.98). A p value < 0.05 was considered significant. Results In general, the correlation between MRI-measured hamstring thickness and hamstring graft thickness as measured in the operating room were good but not excellent. The three measurements that demonstrated the strongest correlation with graft size in the 1.5 T group were the semitendinosus at the medial femoral condyle (r = 0.69; p < 0.001), the semitendinosus and gracilis at the medial femoral condyle (r = 0.70; p < 0.001), and the mean semitendinosus and gracilis (r = 0.64; p < 0.001). These three measurements had correlation values of 0.53, 0.56, and 0.56, respectively, in the 3 T MRI group (all p values < 0.001). To create an 8-mm hamstring autograft, the mean semitendinosus plus gracilis cutoff values areas were 18.8 mm2 and 17.5 mm2 for the 1.5 T and 3.0 T MRI groups, respectively. Conclusions Imaging performed according to routine knee injury protocol can be used to preoperatively predict the size of hamstring autografts for ACL reconstructions. In clinical practice, this can assist orthopaedic surgeons in graft selection and surgical planning. Level of Evidence Level II, diagnostic study.

Background: In anterior cruciate ligament (ACL) reconstruction, hamstring tendon autografts <8 mm have been associated with increased failure rates. There has been no established modality by which orthopaedic surgeons can preoperatively predict graft sizes.Purpose/Hypothesis: The purposes of this study were to (1) determine whether routine magnetic resonance imaging (MRI) measurement of hamstring tendon cross-sectional area (CSA) can reliably be used by sports medicine fellowship-trained orthopaedic surgeons to predict graft size and (2) determine whether radiologists and sports medicine surgeons are able to discriminate grafts below a predetermined cutoff value. We hypothesized that radiologists will find a correlation between MRI measurement and intraoperative graft size. Similarly, orthopaedic surgeons will be able to correctly estimate the graft size based on MRI measurement.Study Design: Cohort study (diagnosis); Level of evidence, 2.Methods: Included in this study were 30 consecutive patients (15 women and 15 men) (mean age, 23 years [range, 13-43 years]) for whom MRI-determined hamstring tendon CSA and graft size measurements could be compared. Patients were included if they had a preoperative MRI demonstrating acute ACL rupture and were scheduled with 1 of 3 surgeons for a reconstruction performed using the ST and GR tendons. Operative data were collected over 1 year. Sectra imaging software was used to measure the CSA of the semitendinosus (ST) and gracilis (GR) tendons on the preoperative MRIs. Control measurements were performed intraoperatively using a graft sizing block with 0.5-mm increments. Simple linear regression analysis was used to evaluate the ability of MRI measurements to predict autograft size. Logistic regression was used to determine the minimum CSA for a graft of 8 mm. The intraclass correlation coefficient (ICC) was used to evaluate interrater reliability.Results: MRI CSA measurement of the average STGR (ST CSA added to the GR CSA) was a significant predictor of graft size (adjusted R 2 ¼ 0.186; P < .001). The 3 measurements with the strongest correlations with graft size were the ST at the medial femoral condyle (MFC), the STGR at the MFC, and the average STGR. The minimum CSA for the average STGR on MRI to achieve a graft size of 8 mm was 17.168 mm 2 (P < .001). The area under the receiver operating characteristic curve was 0.765. The overall ICC was 0.977. Conclusion:Routine preoperative MRI can be used by both radiologists and orthopaedic surgeons to predict the expected ACL autograft size and identify those below a cutoff of 8 mm. This will help in preoperative planning and graft selection.