Da. Rubin et al., NIR DIAGNOSIS OF MENISCAL TEARS OF THE KNEE - VALUE OF FAST SPIN-ECHOVS CONVENTIONAL SPIN-ECHO PULSE SEQUENCES, American journal of roentgenology, 162(5), 1994, pp. 1131-1135
OBJECTIVE. Contrast can be similar on fast spin-echo and conventional
spin-echo MR images, but data acquisition is faster with fast spin-ech
o sequences. This study was designed to evaluate the performance of fa
st spin-echo sequences in the detection of meniscal tears in the knee,
with the established conventional spin-echo technique used as a refer
ence standard. SUBJECTS AND METHODS. We imaged 66 consecutive patients
(129 menisci) who were referred for MR examination with suspected men
iscal tears. We used our routine two-dimensional, multisection, long r
epetition time/double-echo spin-echo sequence and one of two fast spin
-echo sequences. The fast spin-echo parameters were chosen to minimize
the loss of high-resolution detail while otherwise maintaining the se
quence as close as possible to the spin-echo sequence. We then did a r
etrospective evaluation of the fast spin-echo images, using the spin-e
cho images as the gold standard. RESULTS. Fast spin-echo images showed
only 30 (65%) of the 46 meniscal tears seen on the conventional spin-
echo images. In addition, four of the 30 tears seen with both sequence
s were diagnosed with greater confidence on the conventional spin-echo
images. In the cases in which both sequences allowed a diagnosis of d
efinite meniscal tear, the abnormalities tended to be more conspicuous
on the spin-echo images. CONCLUSION. Our results suggest that the fas
t spin-echo sequence should not be used for the diagnosis of meniscal
tears.