1. Standards. There are insufficient data to support a treatment standard for this topic.

2. Guidelines. Prophylactic use of anti-seizure therapy is not recommended for children with severe traumatic brain injury (TBI) for preventing late posttraumatic seizures (PTS).

3. Options. Prophylactic anti-seizure therapy may be considered as a treatment option to prevent early PTS in young pediatric patients and infants at high risk for seizures following head injury.

4. Indications from Adult Guidelines. Prophylactic use of phenytoin, carbamazepine, or phenobarbital is not recommended for preventing late PTS in adults (1).

   Use of phenytoin and carbamazepine has been shown to decrease the risk of early PTS in adults. There is no evidence that outcome is improved. It is a treatment option that anticonvulsants may be used to prevent early PTS in adult patients at high risk for seizures after TBI.

Infants and small children are known to have lower seizure thresholds2. Infants can have subtle seizures, making clinical detection difficult3. An electroencephalogram may be needed to determine whether an infant is experiencing seizures. Treatment of neonatal seizures can be difficult, and single-drug therapy is not effective for the majority of infants4.

Posttraumatic seizures are classified as early (occurring within 7 days) or late (occurring after 7 days) following injury. As outlined in the adult guidelines, "It is desirable to prevent both early and late PTS; however it is also desirable to avoid neurobehavioral and other side effects of the medications that are ineffective in preventing seizures. Prophylaxis for PTS refers to the practice of administering anticonvulsants to patients following head injury, to prevent seizures. The rationale for routine seizure prophylaxis is as follows. There is a relatively high incidence of PTS in head-injured patients, and there are potential benefits to preventing seizures following head injury"1.

In the acute period after severe TBI, seizures increase brain metabolic demands, increase intracranial pressure, and may lead to secondary brain injury. The occurrence of late PTS may be associated with accidental injury and psychological consequences. Prevention of early seizures has been suggested to prevent the development of chronic epilepsy(5, 6). On the other hand, anticonvulsants have been associated with adverse side effects such as impaired learning, rashes, Stevens Johnson syndrome, hematic abnormalities, and ataxia, and also with neurobehavioral side effects(5-7). Finally compliance with long-term seizure prophylaxis in children after severe TBI has been poor7. Critical evaluation of prophylactic anticonvulsant use in children with severe TBI is therefore needed.

We searched Medline and Healthstar from 1966 to 2001 by using the search strategy for this question (see Appendix A) and supplemented the results with literature recommended by peers or identified from reference lists. Of 31 potentially relevant studies, three were used as evidence for this question (Table 1).

Infants and children are reported to have greater risk of early PTS compared with adults after severe TBI. The reported incidence of early PTS after severe TBI in children varies from approximately 20 to 39%8 -11. Both low Glasgow Coma Scale (GCS) scores8 -11 and young age are associated with increased risk of early PTS10 -13. After adjustment for GCS and duration of coma, children <2 yrs of age have almost a three-fold greater risk of early PTS compared with older children up to 12 yrs of age (odds ratio, 2.96; 95% confidence interval, 1.42- 6.21)9. The incidence of early PTS after abusive head injury has not been reported. Studies report that the majority of early pediatric PTS occurs within the first 24 hrs after injury10, 11.

The rate of late PTS among children with severe traumatic brain injury ranges from 7 to 12%7, 8, which is slightly lower than among adults after severe nonpenetrating TBI (9-13%)5, 8. The relationship of patient age and severity of head injury to the risk of late PTS among children after severe TBI injury has not been reported. Jennet14 reported that among patients with depressed skull fractures, the incidence of late PTS was 12% in children <5 yrs and 20% in children 5-16 yrs of age. The rate of late PTS among patients >16 yrs of age was 9%. The relation between early PTS in children with severe TBI and late PTS has not been well studied. Among children <3 yrs old who had brain trauma and early PTS, 12% of infants and toddlers experienced late PTS, and risk of late PTS was significantly greater among children <1 yr of age compared with 2- and 3-yr-old children15.

Studies Regarding Late PTS

Young et al.7 conducted a randomized, double-blind, placebo-controlled study of late PTS prophylaxis using phenytoin or placebo in 41 children after head trauma. Children were changed to phenobarbital if they developed delayed hypersensitivity to phenytoin. Follow-up was for 18 months. The authors did not report the incidence of early PTS in the study population. The incidence of late PTS in the treated group was 12% vs. 6% in the placebo group. This difference was not statistically significant. Compliance was poor. By the third month of study, only half of the patients had a level of ≥10 µg/mL, and at 1 yr only 10% had the targeted level of 10-20 µg/mL. No patient with a plasma concentration >10 µg/mL had a seizure, and therefore the possibility remains that higher compliance levels might be more effective in preventing late PTS. Because of small sample size and poor compliance with the prescribed anticonvulsant regimen, this study is graded as class II evidence.

Key Elements from the Adult Guidelines Relevant to Pediatric TBI

The "Guidelines for the Management of [Adult] Severe Traumatic Brain Injury"1 reported on six well-controlled studies of prophylaxis for late PTS and concluded that anticonvulsants administered prophylactically do not significantly reduce the incidence of late PTS. The American Academy of Physical Medicine and Rehabilitation published similar conclusions that "antiepileptic drugs are not recommended after 1 wk for preventing late posttraumatic seizures in patients without a history of seizures following nonpenetrating traumatic brain injury."16

The study of pediatric patients7, combined with the evidence from adult studies1, do not support use of prophylactic anti-seizure therapy for prevention of late PTS in children; however, the evidence in children is at the level of a guideline rather than the standard level recommended for adult patients.

Studies Regarding Early PTS

Lewis et al.10 reported a retrospective cohort study of children treated at a single urban trauma center with head injuries, comparing risk of PTS among children with severe TBI who received prophylactic phenytoin with children with severe TBI who did not receive anticonvulsants. Use of prophylactic phenytoin was determined by the caregivers. Thirty-one children had severe TBI. Patients who received prophylactic phenytoin (n = 17) were compared with children (n = 13) who did not receive prophylactic anti-seizure medications. (One child received multiple anti-seizure medications and was excluded.) Nine (53%) of the children who did not receive prophylactic phenytoin had a PTS, whereas two (15%) of the children who received prophylactic phenytoin had a PTS. Outcome was not reported.

Tilford et al.17 reported on a retrospective cohort of 477 children treated for head trauma at three pediatric intensive care units. One hundred and twentyeight had severe TBI. The study evaluated differences in therapies used for head trauma patients at the three intensive care units. Use of anti-seizure medication varied significantly by center, ranging from 10% to 35%. Type of anticonvulsant medications used was not specified. The overall incidence of PTS was 9%, and the authors did not report the univariate analysis for difference in seizure rate by use of an anticonvulsant or whether all of the PTS occurred in the first week after injury. In a multivariate analysis that included the entire cohort of patients, use of an anti-seizure medication was significantly associated with improved survival (odds ratio, 0.17; 95% confidence interval, 0.04-0.7). The model adjusted for severe TBI (GCS ≤8 on admission), dilated and unresponsive pupils, serum bicarbonate level, use of vasoactive medications, hyperglycemia, use of a neuromuscular blocking agent, use of an intracranial pressure monitor, and hospital site of care.

Key Elements from the Adult Guidelines Relevant to Pediatric TBI

The "Guidelines for the Management of [Adult] Severe Traumatic Brain Injury"1 reviewed two well-controlled studies of prophylaxis for early PTS in adults and concluded that anticonvulsants administered prophylactically reduce the incidence of early PTS but do notimproveultimateoutcome.Consideration of prophylactic use of anticonvulsants is recommended as a treatment option to prevent early PTS in the adult guidelines1.

Late PTS

Prophylactic anticonvulsant therapy is not recommended to prevent late PTS in children. If a late PTS occurs, the patient should be managed in accordance with standard approaches to patients with new-onset seizures.

Early PTS

Phenytoin has been shown to reduce the incidence of early PTS in a single study of children with severe TBI10. Tilford et al.17 reported an association between use of anticonvulsant therapy and improved survival; however, children with severe TBI were not separately evaluated and the medication used was not specified. Anticonvulsant therapy to prevent the occurrence of early PTS in high-risk children during the first week following severe TBI is recommended as a treatment option.

The scientific evidence suggests that children, especially infants and toddlers, with severe TBI have a greater risk of early PTS than do adults with severe TBI.

• Further studies regarding anti-seizure prophylaxis of early PTS after severe TBI in infants and children are needed. A large observational study17 reported an association between use of anticonvulsant medication and lower mortality rate, raising the possibility that children may benefit more than adults from prophylactic anticonvulsant medications; however, this observation needs to be confirmed.
• Potential differences in effects of anticonvulsant use with age have not been evaluated.
• Further studies regarding risk of both early and late PTS in infants with abusive head trauma as well as mortality rate information are needed. Prophylactic anticonvulsant therapy is not recommended to prevent late posttraumatic seizures in children.