Adding Insult to Injury: Nonconvulsive Seizures in Abusive Head Trauma

Transcription

1 Original Article Adding Insult to Injury: Nonconvulsive Seizures in Abusive Head Trauma Journal of Child Neurology 2015, Vol. 30(13) ª The Author(s) 2015 Reprints and permission: sagepub.com/journalspermissions.nav DOI: / jcn.sagepub.com Mary V. Greiner, MD, MS 1, Hansel M. Greiner, MD 2, Marguerite M. Caré, MD 3, Deanna Owens, MA 2, Robert Shapiro, MD 1, and Katherine Holland, MD, PhD 2 Abstract The primary objectives of this study were to determine the prevalence of nonconvulsive seizures and nonconvulsive status epilepticus in patients with abusive head trauma who underwent electroencephalography (EEG) monitoring and to describe predictive factors for this population. Children with a diagnosis of abusive head trauma were studied retrospectively to determine the rate of EEG monitoring, the rate of nonconvulsive seizures and nonconvulsive status epilepticus, and the associated neuroimaging findings. Over 11 years, 73 of 199 (36.8%) children with abusive head trauma had electroencephalography monitoring performed. Of these, 20 (27.4%) had nonconvulsive seizures and 3 (4.1%) had nonconvulsive status epilepticus. The presence of subarachnoid and cortical T2 / fluid-attenuated inversion recovery signal abnormalities were both significantly associated with the presence of nonconvulsive seizures / nonconvulsive status epilepticus. Nonconvulsive seizures are relatively common in abusive head trauma and may go unrecognized. Specific neuroimaging characteristics increase the likelihood of nonconvulsive seizures on EEG. Keywords EEG, status epilepticus, child abuse, subarachnoid, magnetic resonance imaging (MRI) Received October 17, Received revised January 12, Accepted for publication February 16, Abusive head trauma is the leading cause of traumatic injury death to young children and infants. Abusive head trauma accounts for nearly two-thirds of infant homicides. 1 Infants and children who survive abusive head trauma have brain injury, eye injuries, and fractures in the acute stage. The majority of survivors suffer developmental delays, seizures, motor impairments, feeding difficulties and later behavioral and educational dysfunctions. 2 Victims of abusive head trauma may be at risk for nonconvulsive seizures and nonconvulsive status epilepticus. Nonconvulsive status epilepticus is defined as prolonged electrographic seizure activity resulting in nonconvulsive clinical symptoms, to include alteration of consciousness. 3,4 Nonconvulsive seizures and nonconvulsive status epilepticus are associated with increased mortality and worse shortterm neurologic outcome in critically ill children. 5 Detection of nonconvulsive seizures by electroencephalographic (EEG) monitoring can allow for changes to clinical management to reduce seizures and improve outcome. 6 EEG monitoring may allow an opportunity for intervention to decrease seizures and improve the mortality and outcome of these children. This study is a retrospective chart review to determine the prevalence of nonconvulsive seizures and nonconvulsive status epilepticus in patients with abusive head trauma who underwent EEG monitoring. Secondary aims were to determine magnetic resonance imaging (MRI) findings associated with nonconvulsive seizures and nonconvulsive status epilepticus and to determine developmental outcomes of patients with abusive head trauma and nonconvulsive seizures and nonconvulsive status epilepticus. Patients and Methods This is a retrospective chart review from an 11-year period from January 2001 to December 2011, with approval from the institutional review board at Cincinnati Children s Hospital Medical Center. Subjects were included in this study if they were less than 24 months of age and carried a diagnosis of abusive head trauma. The diagnosis of abusive head trauma was made by a multidisciplinary child abuse team, including board-certified child abuse pediatricians, hospital social workers, investigative community children s services workers, prosecutors, and law enforcement officers. 1 Department of Pediatrics, Cincinnati Children s Hospital Medical Center, Cincinnati, OH, USA 2 Department of Neurology, Cincinnati Children s Hospital Medical Center, Cincinnati, OH, USA 3 Department of Radiology, Cincinnati Children s Hospital Medical Center, Cincinnati, OH, USA Corresponding Author: Mary V. Greiner, MD, MS, Cincinnati Children s Hospital Medical Center, Department of Pediatrics, 3333 Burnet Avenue, ML 3008, Cincinnati, OH 45229, USA. mary.greiner@cchmc.org

2 Greiner et al 1779 For eligible subjects, it was determined by chart review whether EEG monitoring was obtained during the admission. All EEGs were performed using a continuous video EEG technique and using the standard 10/20 system of electrode placement, except when neurosurgical considerations related to wounds and drains required placement of fewer electrodes. If EEG monitoring was obtained, it was categorized as routine (<60 minutes in duration) or prolonged. Prolonged EEG monitoring was quantified in hours. EEG Analysis All EEG monitoring reports were reviewed for findings, if applicable. All records were reviewed by a pediatric epileptologist (HG) to determine presence of epileptiform discharges, nonconvulsive seizures, and/or nonconvulsive status epilepticus as defined by Abend et al. 7 EEG reports were assessed for (a) nonconvulsive seizures defined as rhythmic epileptic activity with evolution in frequency, amplitude, and location with no clear clinical correlate; (b) nonconvulsive status epilepticus, defined as 30 minutes of continuous epileptic activity or at least 30 minutes of activity within a 1-hour period with no clear clinical correlate; and/or (c) convulsive seizure, defined as an electrographic seizure with clinical correlate. The number, duration, and locations of seizure onset and seizure spread were recorded. For the data analysis, EEG results were categorized into normal, nonconvulsive seizures or nonconvulsive status epilepticus, electroclinical seizures only, electrocerebral silence, and other abnormal. For statistical analysis, because nonconvulsive status epilepticus is a prolonged form of nonconvulsive seizures, both nonconvulsive seizures and nonconvulsive status epilepticus were combined to form the variable nonconvulsive seizures. Brain MRI Analysis For subjects identified with nonconvulsive seizures or nonconvulsive status epilepticus, all magnetic resonance imaging (MRI) neuroimaging reports were reviewed. In addition, a pediatric neuroradiologist (MC) blinded to EEG results reviewed all MRI neuroimaging completed within 14 days of diagnosis to identify intracranial findings, to include subdural, epidural, and subarachnoid, parenchymal, T2 / fluid-attenuated inversion recovery signal abnormality, diffusion restriction and edema. When potentially equivocal results existed for a particular variable, the variable was treated as positive, to best reflect clinical practices. For example, if an MRI was read as possible subarachnoid, that was interpreted as positive for subarachnoid. Developmental Outcome Further chart review was completed on each patient to look for evidence of developmental outcome, including diagnoses of developmental delays, therapy appointments, or follow-up in a postinjury clinic. The postinjury clinic is a consultation clinic for children with a history of abusive head trauma that provides developmental testing every 3 months to 1 year by a developmental pediatrician using the Mullen Scales of Early Learning, a comprehensive measure of cognitive functioning designed for use in children age birth through 68 months. In this study, testing results were used to assess the impact of nonconvulsive seizures on developmental outcome. Statistical Analysis Statistical analysis was performed using the Fisher exact test. Table 1. Demographics. All subjects with AHT Subjects with monitored EEG Subjects with NCS/NCSE N Age, mo (range) 6.55 ( ) 5.25 ( ) 3.46 ( ) Gender (%) Males 122 (61.3) 44 (60.3) 9 (45) MRI 111 (56) 50 (68) 16 (80) neuroimaging Mortality 26 (13) 3 (4) 1 (5) Abbreviations: AHT, abusive head trauma; EEG, electroencephalography; MRI, magnetic resonance imaging; NCS, nonconvulsive seizures; NCSE, nonconvulsive status epilepticus. Results Demographics Over an 11-year period, 199 subjects were identified with the diagnosis of abusive head trauma. These children ranged in age from 0.03 to 23.8 months, with an average age of 6.6 months. There were 122 males and 77 females (Table 1). Of the 199 subjects with abusive head trauma, 73 (36.8%) had EEG monitoring during their admission hospitalization. For 59 of the 73 (81%), this consisted of a 1-hour routine EEG. Only 14 patients (19%) had a prolonged EEG. These ranged from 11 hours to 120 hours. Prevalence of Nonconvulsive Seizures and Nonconvulsive Status Epilepticus Of the 73 subjects with EEG monitoring, 20 (27.4%) had nonconvulsive seizures. Eleven of these subjects (55%)wereidentified with prolonged EEG, and 9 (45%) were identified with routine EEG. Three subjects (4.1%) had nonconvulsive status epilepticus. Two of the 3 subjects identified with nonconvulsive status epilepticus also had clinically apparent seizures. One was identified on routine EEG, 2 had prolonged EEGs (Figure 1). MRI Findings Of the 73 subjects with EEG monitoring, 50 (68.5%) had MRIs performed within 2 weeks of admission. Looking at this subset of children with EEG monitoring and MRI imaging, all 50 (100%) had subdural, 39 (78%) had subarachnoid, 20 (40%) had parenchymal and none had epidural. Thirty-four (68%) had diffusion restriction, 35 (70%) had acute T2 / fluid-attenuated inversion recovery signal abnormality, 15 (30%) had hippocampal signal abnormality, and 1 (2%) had contrast enhancement. Two imaging findings were found to be significantly associated with the finding of nonconvulsive seizures on EEG: the presence of subarachnoid (Fisher exact test, P ¼.010) and the presence of abnormal T2 / fluid-attenuated inversion recovery signal (Fisher exact test, P ¼.019). The presence of parenchymal, diffusion restriction, MRI enhancement, and hippocampal signal abnormality was not

3 1780 Journal of Child Neurology 30(13) who followed up in the postinjury clinic, and 13 (81%) of them had formal developmental testing. Of this group, 10 (77%) had evidence of development delay and 3 (23%) did not. The difference between groups was not statistically significant (Fisher exact test, P ¼.47). All 3 subjects with nonconvulsive status epilepticus were followed in the postinjury clinic with formal developmental testing. All 3 demonstrated developmental delays. Figure 1. Study flow based on type of EEG performed and result. Abbreviations: AHT, abusive head trauma; EEG, electroencephalography; NCS, nonconvulsive seizures; NCSE, nonconvulsive status epilepticus; Sz, seizure. Replacing Figure 1 with higher quality image, attached. associated with the presence of nonconvulsive seizures (Fisher exact test, P >.05) (Table 2). There were 2 subjects in this subset who received both EEG monitoring and MRI who were identified with nonconvulsive status epilepticus, a 1.5-month-old female and an 8.5-oldmonth female. Both subjects were noted to have subdural, subarachnoid, T2 / fluid-attenuated inversion recovery signal abnormality, diffusion restriction, and hippocampal signal abnormality. Neither of these subjects had parenchymal or MRI enhancement (Figure 2). Developmental Outcome Sixty-one subjects out of the 73 (83.6%) with EEG monitoring had follow-up at the Children s Hospital, allowing for chart review of developmental outcome. Follow-up was defined as at least 1 subsequent appointment in any division within our hospital. Forty-five of the 61 subjects (73.8%) had evidence of developmental delay described in clinical or therapy notes. Fifteen of the 19 (78.9%) patients with nonconvulsive seizures on EEG during admission demonstrated developmental delay, whereas 30 of 42 (71%) subjects without nonconvulsive seizures on EEG during admission demonstrated developmental delay. There was no statistical difference between these 2 groups (Fisher exact test, P ¼.75). Forty-two subjects of 72 (58%) with EEG monitoring were followed up in the postinjury clinic. There were 26 subjects without nonconvulsive seizures on EEG monitoring who followed up in the postinjury clinic, and 23 (88%) of those had formal developmental testing. Of this group, 14 (61%) had evidence of developmental delay and 9 (39%) did not. There were 16 subjects with nonconvulsive seizures on EEG monitoring Discussion Key Findings There has been increased focus on nonconvulsive seizures in high-risk neonates and critically ill children, yet the specific subpopulation of children with abusive head trauma has not been studied in a large sample. Previous studies have demonstrated a 20% to 23% prevalence of nonconvulsive seizures in critically ill adults 8-10 and a 30% to 47% prevalence in critically ill infants using prolonged EEG or continuous EEG monitoring. 6,11-14 Our retrospective study was limited by mostly brief (<60-minute) recordings and inconsistent practice patterns and indications for monitoring over time. Despite these limitations, we identified nonconvulsive seizures in more than one-quarter of children with abusive head trauma who had at least a routine EEG. Given that previous studies have found nonconvulsive seizures only after more than 1 hour of monitoring in over half of patients ultimately identified, the prevalence in our study is startling and likely represents a significant underestimate. Further, we identified key imaging predictors associated with an increased risk of nonconvulsive seizures. Because children with abusive head trauma are typically diagnosed using head computed tomography or brain MRI, this result is immediately clinically relevant to help guide selection of candidates for prolonged EEG monitoring. Mechanisms of Epileptogenesis in Abusive Head Trauma Abusive head trauma substantially increases risk of seizures compared to accidental head trauma. 15,16 A small study performed by Hasbani and colleagues 17 suggests this very idea, finding nonconvulsive electrographic seizures in 8 of 12 (57%) children in the pediatric intensive care unit. Our study looked at 79 children with EEG monitoring and found a rate of 26.6%. The lower percentage found in our study may reflect a different/larger population, or may partially be because our study was not restricted to the intensive care unit. Children with abusive head trauma also have worse outcomes than children with accidental traumatic brain injury. 18,19 The exact reason for this is unclear, but it may be related to the different mechanisms of injury. A case series of 19 confessions of abusive head trauma from perpetrators found 13 confessions to involve shaking and impact, rather than impact alone. 20 Another series of confessions found 100% reporting violent shaking and only 24% reporting impact. 21 This mechanism of shaking alone, or shaking with impact, may result in increased

4 Greiner et al 1781 Table 2. Head Magnetic Resonance Imaging (MRI) Findings Associated With NCS/NCSE. Subdural Subarachnoid Parenchymal Epidural Diffusion restriction Acute T2/ FLAIR signal abnormality Hippocampal signal abnormality MRI enhancement No seizures (n ¼ 34) 34 (100%) 23 (59%) 13 (38%) 0 (0%) 22 (65%) 20 (59%) 9 (26%) 0 (0%) NCS or NCSE (n ¼ 16) 16 (100%) 16* (100%) 7 (44%) 0 (0%) 12 (75%) 15** (94%) 6 (38%) 1 (6%) All EEG monitoring (n ¼ 50) 50 (100%) 39 (78%) 20 (40%) 0 (0%) 34 (68%) 35 (70%) 15 (30%) 1 (2%) Abbreviations: EEG, electroencephalography; FLAIR, fluid-attenuated inversion recovery; NCS, nonconvulsive seizures; NCSE, nonconvulsive status epilepticus. *Fisher exact test, P ¼.010. **Fisher exact test, P ¼.019. Figure 2. An 8 month old girl presented with encephalopathy and multiple injuries including liver and splenic lacerations and healing rib fractures. Two days later while in the intensive care unit, brief jerking events were noted by a nurse. EEG was performed. (A) Coronal re-formatted Head CT showing a right frontoparietal subdural hematoma (yellow arrow) and subtle hypoattenuation in the left parieto-occipital region consistent with acute parenchymal injury (double yellow arrow). (B) 3D reconstruction of the acute head CT showing multiple comminuted left parieto-occipital fractures. (C) Admission day 5 MRI Brain, Diffusion -weighted imaging showing bilateral (L > R) parietooccipital diffusion restriction consistent with acute infarction. (D) There is corresponding hyperintense T2 signal in the area of major infarction. (E) 8 month post-admission Head CT demonstrating chronic volume loss (CVL) with encephalomalacia and compensatory third and lateral ventricular dilatation. (F) Section of prolonged video EEG (10-20 standard, bipolar montage) showing waxing/waning rhythmic 1 Hz spike/sharp and wave complexes in the left temporal region. Electrographic seizures started within the first hour of recording, improving with multiple anticonvulsants. The patient was seen in followup at 18 months from time of injury. She has global developmental delay, receiving services for the blind.

5 1782 Journal of Child Neurology 30(13) global brain injury compared to impact alone. This may increase the likelihood of convulsive seizures, which are associated with worse outcomes in patients with abusive head trauma. 22 Neurodevelopmental Outcome Our study did not demonstrate worse developmental outcomes for children with nonconvulsive seizures compared to children without identified nonconvulsive seizures. However, the measures used for many patients were imprecise, amounting to obvious need for therapy. Rigorous developmental testing was not available in a majority of the study population; further studies using standardized developmental testing are necessary to determine if developmental outcome is worse in children surviving abusive head trauma complicated by nonconvulsive seizures. Previous studies in other populations have shown a correlation between nonconvulsive seizures and poor outcome. Morbidity of nonconvulsive seizuresishigh, 5 and identification of nonconvulsive seizures prompts management changes. 6 The question remains whether effective treatment or prevention of nonconvulsive seizures can improve morbidity or prognosis. Neuroimaging Predictors of Nonconvulsive Seizures We found an association between neuroradiologic findings of both subarachnoid s and T2 / fluid-attenuated inversion recovery signal abnormality and nonconvulsive seizures. The literature reflects a rate of nonconvulsive seizures of around 7% to 20% in adult patients with subarachnoid ; however, this includes nontraumatic cases In our study, nonconvulsive seizures were found in 41% of children with subarachnoid who had EEG monitoring. Although the true prevalence of nonconvulsive seizures in the entire population of children with abusive head trauma and subarachnoid s is unknown, subarachnoid represents an obvious cortical irritant that can trigger seizures. Abnormal T2 / fluid-attenuated inversion recovery signal in the cerebral cortex is a sensitive biomarker for localized neuronal cytotoxic edema. It can be seen as both a cause and effect of seizures. The presence of subarachnoid and/or abnormal cortical T2 / fluid-attenuated inversion recovery signal should trigger a high suspicion for nonconvulsive seizures and therefore an indication for EEG monitoring. Future Directions Identification of nonconvulsive seizures and nonconvulsive status epilepticus by EEG changes management in most cases, requiring initiation and/or escalation of anticonvulsant therapy. 5 Further, patients with trauma as an etiology may be at increased risk for permanent structural changes after nonconvulsive status epilepticus. 6 Seizure prophylaxis in children and older adults with accidental head trauma is common, although Figure 3. EEG ordering patterns in children with AHT over time. Prolonged EEG 60 minutes. Routine EEG < 60 minutes. still with controversy on timing and antiepileptic choice. 26,27 This report suggests that children with abusive head trauma should be considered in further studies on efficacy of posttraumatic seizure prophylaxis. Limitations This study has limitations. Only 34.5% of children had EEG monitoring during admission for abusive head trauma. EEG ordering patterns were looked at over the time period of this study but a consistent trend was not identified (Figure 3). Unfortunately, it was not possible to determine the reason for EEG monitoring in each case. It is likely that children with a more severe presentation received EEG monitoring and that the rate of nonconvulsive seizures for the nonmonitored may be lower. However, admission to the pediatric intensive care unit was not significantly associated with EEG monitoring or the finding of nonconvulsive seizures in this study, suggesting that illness severity may not have been a direct correlate to EEG monitoring. Nonconvulsive seizures may be underidentified in our population because it is reported that more than half of children found to have nonconvulsive seizures on EEG did so only after the first hour of monitoring and, therefore, would not be picked up by routine EEG. 12 Therefore, it is likely that seizures in some of the children in our study were missed because they received only a routine EEG rather than a prolonged EEG monitoring. We were also unable to identify the children in this study who may have been treated with anticonvulsants. If some children were treated with prophylactic anticonvulsants, the prevalence may be even higher than we detected. Finally, during the period of this study, MRI studies were often performed later in the clinical course, making them less useful for prediction of nonconvulsive seizures. Reviewing initial head CT, however, which was completed within 24 hours of admission for all children, 82% of subjects with subarachnoid noted on MRI had the finding noted on the computed tomography study as well. Furthermore, with advances in medical technology, CT identification of subarachnoid will

6 Greiner et al 1783 likely continue to improve and MRI will likely be completed earlier in the clinical course to assist with clinical decision making. Conclusion This study suggests that nonconvulsive seizures are relatively common in abusive head trauma patients and may go unrecognized. Subarachnoid and abnormal T2 / fluidattenuated inversion recovery signal abnormality on MRI increases the risk for nonconvulsive seizures. Identification of nonconvulsive seizures would change patient management. Further study is needed to determine if treatment will improve outcomes. EEG monitoring should be strongly considered in all cases of suspected abusive head trauma. Acknowledgments Katrina Peariso, MD, for assisting with significant revisions to the manuscript before submission. Author Note An abstract of this manuscript was presented at the 2012 Ray E. Helfer Society Annual Meeting (platform) and the 2013 American Epilepsy Society Annual Meeting (poster). Author Contributions MVG designed the study, acquired the data, and drafted the manuscript. MVG, HMG, MMC, DO, and KH analyzed and interpreted the data. HMG, MMC, RS, and KH contributed to revising the article critically. All authors made substantial intellectual contributions and gave final approval for the version submitted for publication. Declaration of Conflicting Interests The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Funding The authors received no financial support for the research, authorship, and/or publication of this article. Ethical Approval This study was reviewed and approved by the Cincinnati Children s Hospital Medical Center Institutional Review Board. Study # References 1. Overpeck MD, Brenner RA, Trumble AC, Trifiletti LB, Berendes HW. Risk factors for infant homicide in the United States. N Engl JMed. 1998;339: Makaroff KL, Putnam FW. Outcomes of infants and children with inflicted traumatic brain injury. Dev Med Child Neurol. 2003;45: Walker M, Cross H, Smith S, et al. Nonconvulsive status epilepticus: Epilepsy Research Foundation workshop reports. Epileptic Disord. 2005;7: Brophy GM, Bell R, Claassen J, et al. Guidelines for the evaluation and management of status epilepticus. Neurocrit Care. 2012; 17: Abend NS, Sanchez SM, Berg RA, Dlugos DJ, Topjian AA. Treatment of electrographic seizures and status epilepticus in critically ill children: a single center experience. Seizure. 2013;22: Abend NS, Topjian AA, Gutierrez-Colina AM, Donnelly M, Clancy RR, Dlugos DJ. Impact of continuous EEG monitoring on clinical management in critically ill children. Neurocrit Care. 2011;15: Abend NS, Gutierrez-Colina AM, Topjian AA, et al. Nonconvulsive seizures are common in critically ill children. Neurology. 2011;76: Vespa PM, McArthur DL, Xu Y, et al. Nonconvulsive seizures after traumatic brain injury are associated with hippocampal atrophy. Neurology. 2010;75: Dunne JW, Summers QA, Stewart-Wynne EG. Non-convulsive status epilepticus: a prospective study in an adult general hospital. QJMed. 1987;62: Laccheo I, Sonmezturk H, Bhatt AB, et al. Non-convulsive status epilepticus and non-convulsive seizures in neurological ICU patients. Neurocrit Care. 2015;22: Wusthoff CJ, Dlugos DJ, Gutierrez-Colina A, et al. Electrographic seizures during therapeutic hypothermia for neonatal hypoxic-ischemic encephalopathy. J Child Neurol. 2011;26: Schreiber JM, Zelleke T, Gaillard WD, Kaulas H, Dean N, Carpenter JL. Continuous video EEG for patients with acute encephalopathy in a pediatric intensive care unit. Neurocritical Care. 2012;17: Greiner HM, Holland K, Leach JL, Horn PS, Hershey AD, Rose DF. Nonconvulsive status epilepticus: the encephalopathic pediatric patient. Pediatrics. 2012;129:E748-E Shah DK, Zempel J, Barton T, Lukas K, Inder TE. Electrographic seizures in preterm infants during the first week of life are associated with cerebral injury. Pediatr Res. 2010;67: Bechtel K, Stoessel K, Leventhal JM, et al. Characteristics that distinguish accidental from abusive injury in hospitalized young children with head trauma. Pediatrics. 2004;114: Liesemer K, Bratton SL, Zebrack CM, Brockmeyer D, Statler KD. Early post-traumatic seizures in moderate to severe pediatric traumatic brain injury: rates, risk factors, and clinical features. J Neurotrauma. 2011;28: Hasbani DM, Topjian AA, Friess SH, et al. Nonconvulsive electrographic seizures are common in children with abusive head trauma. Pediatr Crit Care Med. 2013;14: Adamo MA, Drazin D, Smith C, Waldman JB. Comparison of accidental and nonaccidental traumatic brain injuries in infants and toddlers: demographics, neurosurgical interventions, and outcomes. J Neurosurg Pediatr. 2009;4: Zakhary MM, Wesolowski JR, Sewick AE, et al. Prevalence and etiology of intracranial in term children under the age of two years: a retrospective study of computerized tomographic

7 1784 Journal of Child Neurology 30(13) imaging and clinical outcome in 798 children. Acad Radiol. 2009; 16: De Leeuw M, Beuls E, Parizel P, Jorens P, Jacobs W. Confessed abusive blunt head trauma. Am J Forensic Med Pathol. 2013;34: Adamsbaum C, Grabar S, Mejean N, Rey-Salmon C. Abusive head trauma: judicial admissions highlight violent and repetitive shaking. Pediatrics. 2010;126: Greiner MV, Lawrence AP, Horn P, Newmeyer AJ, Makoroff KL. Early clinical indicators of developmental outcome in abusive head trauma. Childs Nerv Syst. 2012;28: Claassen J, Mayer SA, Hirsch LJ. Continuous EEG monitoring in patients with subarachnoid. J Clin Neurophysiol. 2005;22: Dennis LJ, Claassen J, Hirsch LJ, Emerson RG, Connolly ES, Mayer SA. Nonconvulsive status epilepticus after subarachnoid. Neurosurgery. 2002;51: ; discussion O Connor KL, Westover MB, Phillips MT, et al. High risk for seizures following subarachnoid regardless of referral bias. Neurocrit Care. 2014;21: Chang BS, Lowenstein DH; Quality Standards Subcommittee of the American Academy of Neurology. Practice parameter: antiepileptic drug prophylaxis in severe traumatic brain injury: report of the Quality Standards Subcommittee of the American Academy of Neurology. Neurology. 2003;60: Torbic H, Forni AA, Anger KE, Degrado JR, Greenwood BC. Use of antiepileptics for seizure prophylaxis after traumatic brain injury. Am J Health Syst Pharm. 2013;70: