Of all benign salivary gland tumors of the parotid gland, pleomorphic adenoma (mixed tumor) is the most common. It accounts for 60% to 70% of all benign tumors of the parotid gland. This neoplasm arises in patients in the fourth to sixth decade of life, with a female predominance. The surgical excision of this lesion continues to be the subject of major debate. The goal is to avoid facial disability yet attain complete resection without perforation of the capsule/pseudocapsule. The purpose of our study is to compare 2 surgical techniques performed at the Ear, Nose, and Throat and Maxillofacial Departments, Rambam Medical Center, Haifa, Israel, and determine which is preferable in treating this lesion. We reviewed 48 patients who underwent excision of pleomorphic adenoma of the parotid gland between 1996 and 2005 at Rambam Medical Center: 18 were treated surgically with the classical superficial parotidectomy (SP) technique, using an anterograde approach, and 30 were treated with retrograde partial superficial parotidectomy (PSP). We compared the 2 surgical techniques in terms of surgical time, histopathologic size of the lesion, amount of excised healthy parotid tissue, histologic margin, and the preservation of the capsule/pseudocapsule. We also made clinical records of temporary or definitive injury to the facial nerve, which branches of the facial nerve were temporarily or definitively injured, the occurrence of Frey syndrome, esthetic satisfaction, and the amount of recurrence or infection after surgery. Of the 48 patients, 19 (39.6%) were male and 29 (60.4%) were female, with a mean age (+/- SD) of 43.8 +/- 16.97 years (median, 50 years; range, 12-79 years). We found a significant difference (P = .029) in mean surgical time (+/- SD): 171 +/- 49.7 minutes (median, 165 minutes) when performing the classical SP and 145 +/- 42.7 minutes (median, 130 minutes) when performing the retrograde PSP. Much more healthy parotid tissue was taken out with the classical procedure (mean, 51.4 +/- 13.6 mm; median, 50 mm) than with the retrograde PSP technique (mean, 39.2 +/- 11.8; median, 35 mm) (P = .01). There was a significant difference (P = .0003) in facial nerve injuries: 39% of patients did not report any facial deficit in the SP group compared with 90% in the PSP group. In the SP group, only 3 patients reported a permanent deficit, and in the PSP group, only 3 patients had a temporary deficit (compared with 8 in the SP group). The main injuries occurred in the mandibular branch with both techniques: 6 SP and 2 PSP. There was no difference in esthetic satisfaction: 72.2% of patients in the SP group and 80% in the PSP group had no esthetic complaints. In the SP group, patients mainly complained about swelling (3 patients), and in the retrograde PSP group, the main complaint was depression (4 patients). Frey syndrome was found in 9 patients in the retrograde PSP group and 4 in the classical SP group (with an overall rate of 27.7%). The lesion recurred in only 2 patients--1 in each group. With both of the techniques, we found satisfactory results. In the majority of cases, retrograde PSP is a superior technique to the classical SP, although Frey syndrome is more often observed with the former.