Background Microvascular Decompression (MVD) operation is the most reliable treatment for hemifacial spasm (HFS), but it causes many complications. The aim of this retrospective study was to investigate the factors relavent to the effects and postoperative complications of microvascular decompression on hemifacial spasm. Methods A total of 1200 HFS patients treated with MVD were studied retrospectively. The root exit zone (REZ) of the facial nerve was exposed through the infraflocculus approach, the offending vessels were identified and separated from the REZ, and a Teflon graft was interposed between the offending vessels and the brain stem. Brain stem auditory evoked potential (AEP) was monitored intraoperatively. Results The offending vessels can be identified in all patients. The anteroinferior cerebellar artery was the main offending vessel (42.6%). Patients with vertebral artery compression had a multiple vascular compression fashion. Follow-up for 2-10 years (mean 4.2 years) showed that 88.7% patients were cured and 5.6% relieved, with an effective rate of 94.3%. Recurrence rate was 3.2%, and the ineffective rate was 2.6%. The most frequent complication was hearing dysfunction (2.8%). Conclusions MVD is the most definitive treatment method of HFS. The key procedures of this operation include adequate exposure of the REZ, identification of the offending vessels, and proper positioning of Teflon grafts. Complications can be reduced effectively by utilizing a real-time AEP monitoring during the operation. Background Microvascular Decompression (MVD) operation is the most reliable treatment for hemifacial spasm (HFS), but it causes found many complications. The aim of this retrospective study was to investigate the factors relavent to the effects and postoperative complications of microvascular decompression on hemifacial spasm. Methods A total of 1200 HFS patients treated with MVD were studied retrospectively. The root exit zone (REZ) of the facial nerve was exposed through the infraflocculus approach, the offending vessels were identified and separated from the REZ, and a Teflon graft was interposed between The Thefending vessels and the brain stem. Brain stem auditory evoked potential (AEP) was monitored intraoperatively. Results The offending vessels could be identified in all patients. The anteroinferior cerebellar artery was the main offending vessel (42.6%). Patients with vertebral artery compression had a multiple vascular compression fashion. Follow-up for 2-10 years (mean 4.2 years) sho wed that 88.7% patients were cured and 5.6% relieved, with an effective rate of 94.3%. Recurrence rate was 3.2%, and the ineffective rate was 2.6%. The most frequent complication was hearing dysfunction (2.8%). Conclusions MVD is the The key procedures of this operation include adequate exposure of the REZ, identification of the offending vessels, and proper positioning of Teflon grafts. Complications can be reduced effectively by utilizing a real-time AEP monitoring during the operation.