Dj. Waters et al., THE DETECTION OF PULMONARY METASTASES BY HELICAL CT - A CLINICOPATHOLOGICAL STUDY IN DOGS, Journal of computer assisted tomography, 22(2), 1998, pp. 235-240
Purpose: We sought to determine the accuracy of helical CT in the dete
ction of pulmonary metastases. Method: Four anesthetized dogs with met
astatic osteosarcoma underwent helical CT with a collimation of 5 mm,
a pitch of 2, and a reconstruction interval of 5 mm. All macroscopical
ly evident metastases were recorded at autopsy, CT images were indepen
dently reviewed by 10 radiologists and compared with pathologic result
s. Alternate slices in the dog with the most metastases were microscop
ically examined in their entirety. Results: Pathologic examination of
the lungs revealed 132 macroscopically evident pulmonary metastases, o
f which 74 (56%) were detected by at least one reader, Forty-four of t
he 99 (44%) metastases of less than or equal to 5 mm in diameter were
detected by at least one reader compared with 30 of 33 (91%) metastase
s of >5 mm in diameter (p < 0.0001). The 10 readers reported a total o
f 107 false positives. Complete microscopy of alternate slices in the
dog with the most metastases (n = 68) revealed an additional 38 microm
etastases of less than or equal to 3 mm in diameter, None of the 32 mi
crometastases of less than or equal to 1 mm were detected by CT. Concl
usion: Helical CT has some limitations in the detection of pulmonary m
etastases.