OBJECTIVE To assess the role of CT brain scans as a routine restaging procedure after primary, aggressive, drug or radiation therapy of unresectable lung cancer. If early, asymptomatic brain metastases are detected and treated, survival could be improved relative to the patients showing brain involvement in a later CT scan performed during the follow-up, at the onset of neurological symptoms. METHODS One hundred patients affected with lung cancer, unresectable on account of histology (small-cell carcinoma) or advanced stage (III, IV) were submitted to chemo- and/or radiotherapy, after a clinical staging including brain CT, which was negative in all patients. Brain CT was also repeated at the end of therapy (restaging), in the absence of any neurological symptom. Further scans were obtained during the subsequent follow-up only when clinical symptoms occurred, suggesting metastases to the brain. Survival values were analyzed in the patients whose brain involvement was detected during restaging, vs those showing symptomatic brain metastases during the follow-up. RESULTS Only 4 patients had asymptomatic metastases, diagnosed with the restaging brain CT scan. Their survival rate was significantly lower than that of the 20 patients whose brain involvement was shown by a follow-up CT scan, performed after the onset of neurological symptoms. However, death was rarely a consequence of brain metastases: primary or other metastatic sites were involved in the terminal events, in the greatest majority of these cases. CONCLUSIONS The sudden, asymptomatic brain involvement, detected at restaging CT scan after primary therapy for unresectable lung cancer, does not correlate with a better prognosis than symptomatic metastases, diagnosed later with a follow-up CT obtained performed for clinical suspicion. Therefore the use of restaging CT scan is not warranted, as a routine procedure, except for the clinical trials intended to define optimal treatment schedules.