Acta Neurochir (Wien) (1997) 139:421-425 Acta Neuroehirurgiea 9 Springer-Verlag 1997 Printed in Austria A Prospective Study of Microvascular Decompression for Trigeminal Neuralgia H. Sletteb01 and P. K. Eide 2 1Department of Neurosurgery, The National Hospital, University of Oslo and 2 Department of Neurosurgery, Ullev~l Hospital, University of Oslo, Oslo, Norway Summary In a prospective study of 25 patients with trigeminal neuralgia (TN), we examined the results of microvascular decompression (MVD). Initial pain relief was complete in 22 patients and partial in one. There were two primary failures. After a median observation time of 38 months, 20 of the 22 patients still were completely free of pain, and one patient reported then 50% pain relief. A vascular compression of the trigeminal root was found intra-operatively in 23 patients. No serious complications occurred. Minor but bother- some dyaesthesias were reported by two patients (8%). The results were satisfactory when compared to other MVD studies. Keywords: Trigeminat neuralgia; microvascular decompres- sion; long-term follow-up. Introduction In 1927 the Norwegian neurosurgeon Vilhelm Magnus described a patient with severe trigeminal neuralgia (TN) secondary to "pressure from an inter- nal carotid artery aneurysm on the surface of the Gas- serian ganglion" [11]. To our knowledge, this is the first report of a vascular aetiology of TN. Later, in 1934, Dandy [5] suggested a vascular aetiology in at least 45% of patients with TN. Gardner [8] was the first to relieve the vascular compression surgically, and later Jannetta developed and refined the method called microvascular decompression (MVD). Jannetta found vascular compressions of the trigeminal root in more than 90% of the patients with TN [9]. Pain relief 10 years after MVD may be expected in 70-80% of the patients [4, 20]. Nevertheless, objections against MVD have been raised due to higher morbidity and mortality when compared to percutaneous procedures and controversy with regard to the role of vascular compression in the aetiology of TN [1, 18]. These objections have limited the use of MVD in many neu- rosurgical departments. Percutaneous procedures have been preferred in spite of a higher recurrence rate. With this background, the present prospective study was undertaken to further evaluate the efficacy of MVD for TN when taking into account the compli- cations. Patients and Methods This prospective study was performed during the 5-year period from 1989 to 1994, and includes our 25 first patients undergoing MVD. The following data were recorded in the research protocol: demographic data, medication before and after surgery, previous surgical treatment, pain characteristics, pre-operative radiological findings, postoperative complications, postoperative pain relief. Exclusion criteria were: multiple sclerosis, poor general health or age above 68 years. All the patients had typical TN, not sufficient- ly relieved by carbamazepine or phenytoin. In 23 patients the pain syndrome was characterized by unilateral intermittent pain, often triggered by non-painful stimulation, and with "pain-free periods. Two of these patients also had continuous background pain. In all these patients TN was to some degree reduced by carbamazepine. The two patients who bad failed to respond to MVD bad continu- ous pain with a dysaesthetic quality in addition to intermittent pain. In both patients the pain was resistant to carbamazepine. Patholog- ical pain including allodynia and wind-up like pain [6], could be evoked by non-painful stimulation in the affected trigeminal skin area of both these two patients. Demographic data are presented in Table 1. In 17 of 25 patients TN could be provoked by movement (e. g., flexion or rotation of the head, or by stamping tbe foot on the floor), whereas the pain disappeared when the patient rested in the supine position. Prior to MVD, 13 patients had been treated by var- ious surgical methods (Table 1). Exploration of the posterior fossa was performed through a small suboccipital retromastoid craniectomy with the patient in the lateral decubitus position, as described by Jannetta [9]. A Iateral supracerebellar route was chosen [14] and the petrosal vein was sacrificed. Compressing vessels were retracted from the nerve root