OBJECTIVE. The objective of this study was to assess the ability of sonogra
phy to reveal Baker's cysts using MR imaging as a gold standard.
MATERIALS AND METHODS. The study group consisted of 36 consecutive knees in
36 patients evaluated with both MR imaging and sonography. Inclusion crite
ria included axial proton density-weighted or T2-weighted MR images. a sono
graphy report that documented the evaluation of the popliteal region of the
knee, and sonographic and MR images that were available for review. The MR
images were retrospectively reviewed for the presence of Bakers cyst (flui
d signal between the semimembranosus and medial gastrocnemius tendons). Son
ography reports were compared with the MR imaging results. The sonographic
images were also retrospectively reviewed to determine whether any characte
ristic findings on sonography were significantly associated with the presen
ce of Baker's cyst on MR imaging.
RESULTS. Retrospective review of MR images revealed 21 Baker's cysts, one m
yxoid liposarcoma, one meniscal cyst. and 13 examinations with normal findi
ngs. The sonography reports revealed that the 21 Baker's cysts were correct
ly diagnosed, whereas the meniscal cyst and myxoid liposarcoma were misdiag
nosed as Baker's cysts, Retrospective review of sonographic images showed a
100% sensitivity, specificity, positive predictive value, negative predict
ive value, and accuracy in the diagnosis of Baker's cyst when hypoechoic or
anechoic fluid was present between the semimembranosus and medial gastrocn
emius tendons. No other sonographic characteristics were significant.
CONCLUSION. identification of fluid between the semimembranosus and medial
gastrocnemius tendons in communication with a posterior knee cyst indicates
Baker's cyst with 100% accuracy.