To determine when to use preoperative embolization, we retrospectively reviewed a consecutive series of concurrently treated patients who underwent carotid body tumor resection between 1984 and 1994. Eleven nonembolized tumors (N-EMB group) and 11 embolized tumors (EMB group) were resected. The two groups were similar with respect to demographics and presentation, with the exception that more patients in the EMB group complained of painful neck masses. There was no significant difference in the pretreatment size of the neck mass between the two groups (N-EMB = 4.3 +/- 1.5 cm; N-EMB = 5.1 +/- 2.1 cm). Zero to 6 days after embolization, surgical resection was performed. There was no difference in the distribution of tumors, which were grouped according to Shamblin's classification, between the N-EMB and EMB patients. Two patients in each group required resection of the internal carotid artery, whereas a total of seven cranial nerves were resected. There were no differences in blood loss, number of blood transfusions, operative time, or perioperative morbidity between the N-EMB and EMB groups. Ten patients had new cranial nerve deficits and four of these patients required treatment for tenth nerve paralysis. Overall the total hospital stay was similar in the two groups, but the EMB group had a significantly longer preoperative stay compared to the N-EMB group (1.5 +/- 0.8 vs. 0.8 +/- 0.4 days; p = 0.02). These data show that preoperative embolization does not significantly improve outcome in patients undergoing resection of carotid body tumors measuring 4 to 5 cm. Therefore, in this era of costcontainment, preoperative embolization should not be used in the treatment of midsized carotid body tumors.