Summary

MRI and CT scans show similar intra- and inter- rater reliability for tunnel measurement following failed ACL reconstruction and both demonstrate higher reliability than plain radiographs.

Abstract

Background

Multiple imaging modalities are used to assess tunnel diameter following ACL reconstruction in preparation for revision surgery. The aim of this study is to assess the intra- and inter- rater reliability of each imaging modality for the measurement of tunnel morphology.

Methods

26 patients were identified who had pre-revision plain films, MRI, and CT imaging of the knee. Two reviewers independently measured tibial and femoral tunnel diameter for each patient using all three imaging modalities. For radiographs, the tibial and femoral tunnels were measured as the widest segment on both AP and lateral views. For both MRI and CT, the long and short axes of the tunnels were measured. Intra- and inter- class correlation (ICC) values were computed to determine intra-rater and inter-rater reliability for each modality. Values less than 0.5, between 0.5 and 0.75, 0.75 and 0.9, and greater than 0.9 were considered poor, moderate, good, and excellent reliability, respectively. In addition, one reviewer reviewed all images in duplicate and the mean tunnel measurement for each location (tibia short axis, tibia long axis, femur short axis, femur long axis) was calculated. The mean tunnel measurements for each location were compared between MRI and CT and between XR and CT using Student’s t-tests.

Results

Inter-rater (Table 1) reliability was moderate for tibial plain radiographs and poor for femoral plain radiographs. Inter-rater reliability was moderate for tibial MRI and good for femoral MRI. Inter-rater reliability was moderate - good for tibial CT, and good - excellent for femoral CT. There was no difference in mean tunnel measurements when comparing MRI and CT scan (tibia short axis p= 0.76, tibia long axis p= 0.15, femur short axis p= 0.12, and femur long axis p= 0.84). However, mean XR measurements did differ from mean MRI measurements at all locations (tibia short axis p= 0.036, tibia long axis p= 0.01, femur short axis p= 0.02, and femur long axis p< 0.001). XR estimates of tunnel diameter differed from CT by >2mm 50%, 79%, 75%, and 92% of the time at the short tibia, long tibia, short femur, and long femur sites respectively. XR estimates of tunnel diameter differed from CT by >5mm 4%, 38%, 21%, and 75% of the time at the short tibia, long tibia, short femur, and long femur sites respectively. MRI estimates of tunnel diameter differed from CT by >2mm 38%, 50%, 63%, and 79% of the time at the short tibia, long tibia, short femur, and long femur sites respectively. XR estimates of tunnel diameter differed from CT by >5mm 13%, 21%, 0% and 29% of the time at the short tibia, long tibia, short femur, and long femur sites respectively.

Conclusions

MRI and CT scans show similar intra- and inter- rater reliability for tunnel measurement following failed ACL reconstruction and both demonstrate higher reliability than plain radiographs. In addition, at all locations, mean tunnel measurements did not differ when comparing measurements on MRI to CT. Mean tunnel measurements did differ at all locations when comparting XR to CT. Taken together, these findings suggest that MRI may be equivalent to CT for tibial tunnel measurement.