Diagnostic efficacy in knee MRI comparing conventional technique and multiplanar reconstruction with one-millimeter FSE PDW images

Background: Magnetic resonance ( MR) imaging has proved to be an excellent tool in diagnosing injuries of the cruciate ligaments and menisci. However, multiple planes and sometimes optimal oblique or double-oblique scan planes are needed due to the variability in the positioning of important structures, which means there is a lower throughput and longer scanning time. Purpose: To compare the performance of a 1-mm-thickness fast spin-echo ( FSE) proton-density-weighted ( PDW) MR imaging technique with multiplanar reconstruction ( MPR) in diagnosing tears of the menisci and cruciate ligaments with that of conventional MR imaging. Material and Methods: Twenty-five consecutive patients underwent preoperative conventional and 1-mm-thickness FSE PDW MR imaging with subsequent knee arthroscopic surgery. Two musculoskeletal radiologists evaluated the status of the cruciate ligaments and menisci using two sets of MR images ( method A: conventional images including seven sequences, taking 26 min; method B: 1-mm-thickness FSE PDW images with MPR, taking 7 min 20 s). The diagnostic efficacies of both methods for tears of the cruciate ligament and menisci were calculated and compared. Results: Arthroscopic surgery revealed 10 anterior cruciate ligament ( ACL) tears, one posterior cruciate ligament ( PCL) tear, and 26 meniscal tears. The diagnostic values of both methods were 100% for a cruciate ligament tear. The diagnostic values ( sensitivity, specificity, accuracy, positive predictive value, and negative predictive value) for meniscal tears were 90%, 100%, 96%, 100%, and 94% for method A, and 95%, 100%, 98%, 100%, and 97% for method B, respectively. There were no significant differences in the diagnostic values between methods A and B. Conclusion: 1-mm-slice-thickness FSE PDW imaging with MPR showed comparable performance in diagnosing tears of the cruciate ligaments and menisci to conventional sequences but the scan time was much shorter. Therefore, this technique ( method B) might improve the throughput of a 3T MR imaging system.