After we incidentally found on CT extensive esophageal fat accumulations in
a patient with long-term use of steroids, we prospectively evaluated durin
g a 6-month period all CT studies of the chest esophageal lipomatosis and r
elated the findings to the possible use of steroids. The diagnosis of esoph
ageal fat on CT was made by density measurements or if too small for reliab
le density measurements by comparison with mediastinal fat. In 21 of 1320 e
xclusively older male patients the diagnosis of esophageal lipomatosis was
definite in 7 and likely in 14 patients. All fat accumulations were located
in the upper third of the esophagus (mean length 22 +/- 6 mm) and presente
d ring-like (n = 10), irregular (n = 3), or as a horseshoe sparing the post
erior border (n = 8). Tn 20 patients there was an unequivocal history of st
eroid treatment. Associated centripetal fat infiltration was found in 11 pa
tients. None of the patients had swallowing problems. Prolonged use of ster
oids, either orally or inhalationally administered, is associated with esop
hageal lipomatosis. The predisposition for the upper esophagus might be rel
ated to the presence of striated muscle cells in this part of the esophagus
; moreover, inhalational steroid thera-may adversely affect the upper esoph
agus.