How to Improve the Prediction of Quadruple Semitendinosus and Gracilis Autograft Sizes with Magnetic Resonance Imaging and Ultrasonography

OBJECTIVES: To determine if preoperative ultrasonographic measurement of the cross-sectional area (CSA) of the gracilis tendon (GT) and semitendinosus tendon (ST) could predict the diameter of the quadrupled ST-GT autograft as accurately as the magnetic resonance imaging (MRI), and how different magnifications of the MRI slices could influence the accuracy of the measurement. METHODS: 33 patients scheduled for ACL reconstruction with hamstring autograft and with MRI performed at our institution were included. CSA of every GT and ST was calculated with ultrasonography (US) and with 2 different types of measurements on axial 1.5-T MRI at the widest point of the medial femoral epicondyle. In the first type of MRI measurement (MRI-2x method), the images were all analyzed under a 2x-magnified view while in the second type (MRI-4x method), under a 4x-magnified view with the method described by Hamada M et al (Figure 1). The CSA was calculated with the PACS workstation (Centricity Enterprise Web V3.0) with the region of interest tool. Every coupled GT and ST was doubled over around a suture and passed through a closed-hole sizing block. The smallest sizing hole through which the proximal end of the quadrupled ST+GT graft could be pulled with maximal manual force was the final diameter considered for the study PURPOSE: RESULTS: The Pearson correlation coefficient for the US cross- sectional area to the intraoperative graft diameter was 0.506 (p=0.003). When the CSA was major than 16 mm2 resulted in graft diameter of at least 8 mm in 100% of cases. The CSA on MRI of the GT + ST with the MRI-2x and MRI-4x methods were 27.2 ± 4.4 mm2 and 18.8 ± 2.8 mm2 respectively. Intraoperative graft diameter measurement positively correlated with preoperative MRI tendon CSA measurement with both measurement techniques. However, whether the correlation was only moderate with the MRI-2x method (0.536; p=0.001), it was excellent with the MRI-4x method (0.86; p<0.001) (Figure 2). With the MRI-2x measurement method, when the CSA was major than 25 mm2 resulted in graft diameter of at least 8 mm in 95% of the cases. With the MRI-4x measurement method, when the CSA was major than 17 mm2 resulted in graft diameter of at least 8 mm in 100% of cases. CONCLUSION: Preoperative calculation of the cross-sectional area of hamstring tendons with MRI and US can help to reliable estimate quadruple ST+GT graft. In terms of correlation of the CSA with graft diameter, US was comparable to MRI at 2x-magnification, but MRI at 4x-magnification showed superior accuracy. A ST+GT CSA threshold value of 25 mm2, 17mm2, and 16m2 with the MRI-2x, MRI-4x and US methods, respectively, predicted a 4-stranded graft of a minimum 8mm in diameter.