The six primary studies lack the controls that were a key part of the Rochester Epidemiology Project (discussed below), but the rate of seizures in this group was generally higher than the rate in the general population. Furthermore, in most studies it is not possible to determine how many subjects had only a single seizure within 6 months of injury and none later. Thus, the overall proportion of those classified as having post-TBI seizures, ostensibly lasting more than 6 months, might be slightly inflated.
In general, the risk of unprovoked seizures after penetrating TBI is higher than the risk after even the most severe forms of closed TBI. Caveness et al. (1962) compared the number of seizures reported by others in soldiers who sustained both penetrating and nonpenetrating head injuries during World War I (WWI; Credner, 1930; Ascroft, 1941), World War II (WWII; Russell, 1951; Walker and Jablon, 1961), and the Korean War (Caveness and Liss, 1961). They found that seizures were more likely to occur after penetrating head injury (34–43%) than after blunt and blast head injury (12–24%). Of the 317 cases in the WWI cohort, 34.8% reported having seizures compared with 28% of those in the WWII cohort and 24.1% of those in the Korean War cohort. The proportion of patients who had penetrating TBI that later had seizures ranged from 42% in the Korean War cohort (Caveness and Liss, 1961) to 47% in the WWI cohort (Ascroft, 1941).
A followup study was conducted by Caveness (1963) 8–11 years after initial injury. During the followup period, 356 of the original Korean War subjects participated (76.2% of the total and 87.2% of those suitable for followup). Information was collected through mailed questionnaires, physical examinations, interviews with the American Red Cross, and a review of Veterans Administration (VA) records. During the period 1957–1958, additional VA information was available on 84.6% of the participants. Questionnaires were obtained in 1961–1962 from 90.5% of the participants, personal letters from 21.6%, and telephone replies from 9.6% (Caveness, 1963). Caveness (1963) found that of the 356 men, 109 (30.6%) had postinjury seizures; 30 patients had seizures that did not persist beyond 6 months, so 79 (22%) apparently had seizures more than 6 months after injury. Of those with penetrating head wounds, 42% suffered seizures, and 16.4% of those with blunt head wounds had seizures. The authors noted that there was no significant difference in the number of seizures between the total original group and those who were followed for 8–11 years.
Evans (1962) assessed the prevalence of seizures in the same 422 head-injured Korean War veterans at 3–11 years after injury. The overall prevalence was 19.7%. The prevalence was 32% in those with penetrating head injuries, 2% in those with blast wounds, and 8% in those with blunt head injuries.
Phillips (1954) conducted a conditional cohort study of 500 head-injured men admitted within 3 days of injury into the Military Hospital for Head Injuries, Oxford. Information was collected on amnesia, electroencephalographic findings, personal and family history, cerebrospinal fluid (CSF) pressure, epilepsy, intracranial hemorrhage, CSF leak, infection, mental-health changes, condition on discharge, and followup after rehabilitation. The author found that 31 men (6%) developed seizures after injury; 24 had generalized seizures, 5 focal seizures, and 2 mixed seizures. Seven with seizures had focal signs after injury, of whom 5 had focal seizures and 2 had generalized seizures. All the focal seizures occurred within the first 6 days after injury, whereas generalized seizures typically did not typically for several months. It is unclear whether the head injuries were combat-related, and the time between injury and seizure