Kh. Latief et al., SEARCH FOR A PRIMARY LUNG NEOPLASM IN PATIENTS WITH BRAIN METASTASIS - IS THE CHEST RADIOGRAPH SUFFICIENT, American journal of roentgenology, 168(5), 1997, pp. 1339-1344
OBJECTIVE. We assessed whether chest CT provided an advantage over che
st radiography when diagnosing a primary lung neoplasm in a selected g
roup of patients. MATERIALS AND METHODS. From a retrospective evaluati
on of 925 patients who had a discharge diagnosis of brain metastasis,
we identified 32 patients who presented without a known primary tumor
site and who were investigated subsequently with both chest radiograph
y and CT. Reports of chest radiographs were classified as showing a pr
imary lung neoplasm (positive), as abnormal but nonspecific, or as neg
ative. Patients were categorized as having negative chest radiograph,
negative CT; positive chest radiograph, positive CT; nonspecific chest
radiograph, positive CT; or negative chest radiograph, positive CT. R
adiographic technique and clinical and lesion characteristics were com
pared among these categories. RESULTS. We found negative chest radiogr
aph and negative CT in one patient who ultimately proved to have breas
t cancer. The remaining 31 patients (97%) had primary lung carcinoma.
In 19 (59%) of the 32 patients, chest radiographs and CT were positive
. Twelve patients (38%) had a nonspecific or negative chest radiograph
and positive CT. In the 31 patients with lung carcinoma, the mean dia
meter of lesions in patients with positive chest radiographs was 4.2 c
m, compared with 2.5 cm in patients with normal or nonspecific radiogr
aphs (p <.01). CONCLUSION. Lung cancer is by far the most common cause
of a de novo presentation with brain metastasis. Chest CT is valuable
to supplement chest radiography in patients with metastatic brain dis
ease in whom a primary lesion is sought. Lesion size appears to be the
most important determinant of detectability of a primary tumor on che
st radiographs.