A multicentre cooperative prospective study have been conducted to investigate the factors influencing posttraumatic epilepsy (PTE) and to evaluate the prophylactic effect of anticonvulsants. Since April 1994, patients with head injury have been observed following our protocol as follows; anticonvulsants are administered only to the patients with brain parenchymal injury for one month just after head trauma and no anticonvulsants are administered after one month after trauma to any patients except those with posttraumatic epilepsy (PTE). Brain parenchymal injury included traumatic subarachnoid hemorrhage, acute subdural hematoma, contusion, intracerebral hematoma, and diffuse axonal injury. To April 1996, 635 patients with head injury have been registered and analyzed. During the follow-up period, 14 patients (2.2%) developed PTEs, which had only been observed in patients with brain parenchymal injury. Multiple regression analysis revealed that two factors, early epilepsy and brain parenchymal injury, could contribute to the prediction of PTE. The frequency of PTE in this study was compared with that in our previous retrospective study (Nakamura, 1995), in which anticonvulsants were administered to the patients with head injury. There was no significant difference in the percentage of patients having PTE between the group treated without anticonvulsants in this study and the untreated group in previous retrospective study. Anticonvulsants treatment after head injury was not likely to have a prophylactic effect against the development of PTE.