OBJECTIVE. The purpose of this study was to prospectively compare find
ings from postmortem fetal MR imaging with findings at autopsy. SUBJEC
TS AND METHODS. Twenty-six fetuses were imaged on a 1.5-T MR scanner u
sing two-dimensional and high-resolution three-dimensional fast spin-e
cho techniques immediately before autopsy. The MR images were reviewed
independently by three radiologists who evaluated them for major and
minor malformations. These findings were then compared with those at a
utopsy. RESULTS. The 26 subjects had 47 major and 11 minor malformatio
ns. All three radiologists correctly identified 37 of the major malfor
mations on the MR images (detection rate, 79%), and at least one of th
e three reviewers correctly identified 43 of the abnormalities (detect
ion rate, 91%). Only one of the 11 minor anomalies was identified by a
ny reviewer. Reviewers made six false-positive diagnoses. In two cases
, both with major CNS malformations, MR imaging was superior to autops
y in defining in situ relationships. CONCLUSION. Although autopsy rema
ins the study of choice for evaluating causes of fetal death, MR imagi
ng is an excellent alternative when autopsy is refused. Additionally,
MR imaging may be a valuable adjunct to autopsy for fetuses with CNS a
nomalies.