a b s t r a c t

Background: Guidelines from the American Academy of Neurology recommend laboratory studies or computed tomography (CT) for children who experience a nonFebrile seizure if anything in their history suggests a clinically significant abnormality.

Objective: To ascertain if any patient or seizure characteristics are associated with a greater likelihood that labo-

ratory studies or CT scan will yield clinically significant results.

Methods: This retrospective case series reviewed 93 children with nonfebrile seizure, who were evaluated in an urban pediatric emergency department (ED) between July 2007 and June 2011.

Results: Laboratory studies were performed in 87% of the study group; 7% of those tests gave clinically significant results. Computed tomographic scans were obtained in 35% of our patients; 9% showed clinically significant find- ings. Presence of an active seizure in the ED or a first nonfebrile seizure had an 8% and 11% difference, respective- ly, for clinically significant laboratory abnormality. Children younger than 2 years showed a 7% difference of clinically significant laboratory abnormality.

Conclusion: This study did not identify statistically significant predictors of laboratory or CT abnormalities for children with nonfebrile seizure presenting to the ED. Age less than 2 years, having an active seizure in the ED, and experiencing a first-time seizure showed a trend toward an increased yield of laboratory testing. In accor- dance with the American Academy of Neurology guidelines, we conclude that the history of a child’s present illness preceding the nonfebrile seizure, not characteristics of the seizure, should be used to determine the need for further testing.

(C) 2015

Introduction

Nonfebrile seizures are a common reason children are brought to emergency departments (EDs) [1,2]. Guidelines issued by the American Academy of Neurology (AAN) recommend laboratory studies or brain computed tomography (CT) for children who have experienced a nonfebrile seizure if their history suggests a clinically significant ab- normality [2]. This still leaves much ambiguity regarding decisions about when to order laboratory studies and CT imaging. Because of the broad nature of the guidelines, many practitioners obtain laboratory studies and emergent CT scans of the brain on every child with a seizure, even those who are alert, are interactive, and have returned to function- al baseline [2]. Laboratory analyses and CT scans are actually of low yield in children who have had a seizure (Table 1) [3-10]. In parallel with this practice, health care providers are being encouraged to reduce the use of brain CT in children to reduce the lifetime risk of malignancy that it

? No disclosures.

* Corresponding author at: Emergency Medicine, University of Maryland, 110 South Paca St, 6th Floor, Suite 200, Baltimore, MD 21201.

E-mail address: [email protected] (A.M. Strobel).

poses [11]. We suspect that physicians continue to request further studies because it is not clear what components of the history of present illness should guide their decisions. In the current study, our primary objective was to determine if characteristics of a nonfebrile seizure would be associated with an increased yield of useful information from laboratory or imaging studies and thus help physicians with the diagnostic evaluation for nonfebrile seizure.

Methods

We performed a retrospective case series of children with nonfebrile seizure, between July 2007 and June 2011, in the Pediatric Emergency Department at the authors’ affiliated medical center. Patients eligible for this study were identified in 3 ways: the chief concern recorded at our ED’s triage station, the primary discharge diagnosis of seizure, or electroencephalographic (EEG) diagnosis of seizure. Children younger than 2 months were excluded because they are a neonatal seizure [12-14]. Children with a fever (temperature >= 38?C or >= 100.4?F) were excluded because our study focused on nonfebrile seizure [15,16]. The resulting eligible study group was 93 children. The institutional review board approved this study.

http://dx.doi.org/10.1016/j.ajem.2015.06.004

0735-6757/(C) 2015

1262 A.M. Strobel et al. / American Journal of Emergency Medicine 33 (2015) 1261-1264

Table 1

Comparison of results from previous investigations of children with seizure and our study

Study	n	Mean age	Nonfebrile (NF) vs febrile (F)	Percent Laboratories Drawn	Clinically significant laboratories	Percent CT obtained	Clinically significant CT	Seizure characteristic: recurrent	Seizure characteristic: focal
Current	93	5.4 y	NF	87%	7%	35%	9%	76%	30%
Landau et al [4]	85	7.5 y	NF	84%	b5%	8%	0%	54%	40%
Sharma et al [3]a	500	5.2 y	NF	NA	NA	91%	8%	NA	28%
Maytal et al [5]	66	4.9 y	NF + F	NA	NA	NA	18%	NA	NA
Akhavan et al [9]	75	NAb	NF + Fc	NA	9.3%	0.1%	0.02%	57%	NA
Scarfone et al [7]	134	5.7 mo	NF + F	50%	7%	NA	NA	48%	NA
Saz et al [6]	55	5.5 y	NF + Fc	92%	0%	NA	6%	NA	NA

Comparing the results of similar previous studies by Landau et al, Sharma et al, Maytal et al, Akhavan et al, Scarfone et al, and Saz et al, only analyzing nonfebrile seizures. Abbreviation: NA, not available.

^a Sharma et al only included new-onset nonfebrile seizure in their study.

^b Akhavan et al included children older than 1 month to 12 years.

^c Results were not statistically significant between febrile seizure and nonfebrile seizure.

Historical, clinical, and neuroimaging data were abstracted from the medical record. Historical data included the following: medical record number, age, sex, race, whether this was a first or recurrent seizure, whether it was a generalized or focal seizure, whether the patient was hav- ing an active seizure upon arrival, whether antiepileptics were given in the ED, and medical history. Clinical data included complete blood count, basic metabolic panel, liver function tests, antiepileptic levels, whether neurolo- gy consult was obtained, whether an EEG was obtained, whether the head CT scan was read by an attending radiologist, and the patient’s disposition. The laboratory and CT results were categorized as normal, clinically significant, or not clinically significant. Clinically significant means that an outcome had therapeutic consequences, that is, an intervention could treat the abnormality that precipitated the seizure. Hyponatremia was defined as a sodium concentration less than 130 mmol/L (the point at which hyponatremic seizures become evident) [17,18]. Hypocalcemia was defined as a calcium concentration less than 7.0 mg/dL. Age catego- ries of b 2, 2 to 5, and N 5 years were chosen to allow for direct compar-

ison with the study by Saz and colleagues [6].

Data were analyzed using descriptive statistics. Confidence intervals for differences in proportions were calculated using a normal binomial approximation.

Results

There were 163 Medical records reviewed and 93 children with nonfebrile seizure included in our study. The characteristics of the study group are summarized in Table 2. The median age was 4 years. The ED evaluation components are illustrated in Table 3. Seizure characteristics were evaluated as predictors of statistically significant differences in the results of laboratory tests (Table 4) and Brain CT scans (Table 5).

Active seizure in the ED and first episode of seizure had an 8% and 11% difference, respectively, between clinically significant and normal laboratory results, suggesting a trend toward significance, but not reaching statistical significance (Table 4). No predictor had a significant yield in relation to obtaining clinically significant results on CT scan (Table 5).

Table 2

Characteristics of study population (N = 93)

Predictors No. of patients (%)

Age

2-23 mo 18 (19)

2-5 y 46 (50)

6-19 y 29 (31)

Male	59 (63)	Laboratory tests	81 (87)	5 (7)
Type of seizure		CT scan	33 (35)	3 (9)
First time	22 (24)	Antiepileptic measurement	33 (35)	14 (42)
Focal	28 (30)	EEG	48 (52)	30 (63)
Active in ED	41 (44)	Neurology consults	78 (84)	NA

Discussion

The AAN recommendations base the ordering of laboratory studies or Head CT scans on the patient’s preceding history and physical exam- ination findings rather than seizure characteristics [2]. More focused di- agnostic testing would shorten ED length of stay, increase patient comfort, and decrease radiation exposure [3-11]. Conversely, it must be considered that less testing might cause cases of nonaccidental trau- ma or inborn errors of metabolism to be missed [19-23]. Based on level B evidence, the standard evaluation for first nonfebrile seizure in an adult includes laboratory studies and a head CT scan. A reasonably ben- eficial yield can be expected under circumstances with clear indications including, that is, focal neurologic deficit, persistent altered mental sta- tus, fever, trauma, persistent headache, history of cancer, use of antico- agulant, or suspicion of Acquired immunodeficiency syndrome [24-26]. For pediatric patients, despite the AAN recommendation, emergency physicians have high test rates and low yields for the same diagnostic tests (Table 1) [2-10]. Our results are consistent with previously pub- lished studies that examined the yield of routine laboratory studies and CT imaging in children after a nonfebrile seizure (Table 1) [2-10].

Laboratory studies were performed in 87% of our study group; their results were clinically significant in 7%. In the study by Landau et al [4], laboratory studies were performed in 84% of their patients, with signif- icant results in less than 5%. Their study population and exclusion criteria were similar to ours. We also obtained similar results, increasing the validity of our study despite the small sample size and different geo- graphic location. The median age in the 2 groups was 7.5 years in the study by Landau et al and 4 years in ours. Landau et al calculated prev- alence ratios of 5.0 for active seizure in the ED and 4.8 for first time sei- zure, making these predictors more likely to have a clinically significant yield (Table 4).

Head CT imaging was performed in 35% of our patients with a 9% clinically significant yield. Previous studies did not identify predictors that increase the yield of clinically significant findings with brain CT im- aging [3-5]. Only Sharma and colleagues [3] specifically looked at nonfebrile, new-onset seizure and tried to identify clinical variables that would differentiate high- and low-risk groups for clinically signifi- cant Neuroimaging studies. Using recursive partition analysis, they found that a focal seizure occurring before 33 months of age and a pre- disposing history increased the yield of neuroimaging to 26%. Our use of

Table 3

Tests obtained during ed evaluation

No. of obtained (%) No. of positive (%)

Discharged 24 (26)

Abbreviation: NA, not available.

A.M. Strobel et al. / American Journal of Emergency Medicine 33 (2015) 1261-1264 1263

Table 4

Results of analysis of possible predictors of abnormal laboratory values in children with nonfebrile seizure

Predictor	Fraction with predictor with abnormal laboratory result	Fraction without predictor with abnormal laboratory result	Percent difference (95% CI)	Prevalence ratio (95% CI)
Seizure
characteristic:
Active in the ED	4/41 (10%)	1/52 (2%)	8% (-2 to +18)	5.0 (0.6-44)
First nonfebrilea	3/22 (14%)	2/71 (3%)	11% (-4 to +26)	4.8 (0.9-27)
Focalb	2/28 (7%)	3/65 (5%)	3% (-8 to +13)	1.5 (0.3-8.8)
Age (y)
b2	2/18 (11%)	3/75 (4%)	7% (-8 to 22)	3 (0.5-15)
2-5	1/46 (2%)	4/47 (9%)
N 5	2/29 (7%)	3/64 (5%)

^a All others with history of seizures.

^b All others with Generalized seizures.

Table 5

Results of analysis of possible predictors of abnormal scan in children with nonfebrile seizure

studies were obtained from 11% of children younger than 2 years, 2% for children between the ages of 2 and 5 years, and 7% for those older than 5 years. Age less than 2 years were 3 times more likely to have a clinically significant laboratory value (see Table 4). Age seems to be as-

Predictor Fraction with predictor with abnormal CT

Fraction without predictor with abnormal CT

Percent difference (95% CI)

Prevalence ratio (95% CI)

sociated with abnormal laboratory values for patients younger than 2 years. None of the children younger than 2 years had abnormal head CT results. Therefore, no cases of missed abusive head trauma were found with CT imaging in this vulnerable age group, so seizure alone

Seizure characteristic

should not necessitate CT imaging to find nonaccidental trauma

[19-21]. Our findings are in contrast to an update by the AAN, suggest-

Active in the ED First nonfebrile^a

1/41 (2%) 2/52 (4%) -1% (-8 to +6) 0.6 (0.06-6.8)

0/22 (0%) 3/71 (4%) -4% (-9 to 0) 0 (nc)

ing increased yield of emergent CT neuroimaging in children younger than 6 months with seizure [27].

Among the patients with laboratory abnormalities, the history of

Focal^b 0/28 (0%) 3/65 (5%) -5% (-12 to +3) 0 (nc)

Age (y)
b2	0/18 (0%)	3/75 (4%)
2-5	2/46 (4%)	1/47 (2%)
N 5	1/29 (3%)	3/64 (5%)

Abbreviation: nc, not calculated due to zero denominator.

^a All others with history of seizures.

^b All others with generalized seizures.

present illness or the medical history suggested the test result. The histories of children with clinically significant results included the fol- lowing: history of end-stage renal disease, decreased mental status, improper formula mixing, trileptal toxicity, and being 4 days postpar- tum (Table 6). Associated findings were hypocalcemia, urosepsis, hyponatremia, trileptal toxicity, and eclampsia, respectively. Two of the children with clinically significant head CT scans had a

Table 6

Patients with clinically significant abnormal laboratory and CT imaging

Patient Abnormal laboratory finding Seizure characteristics Pertinent history

Hypocalcemia (6.7 mg/dL) First, focal, active 5-mo-old female, ESRD on Peritoneal dialysis, FTT

Hyponatremia (117 mEq/L) First, generalized, active 3-mo-old male, dilute formula mixing

Hyponatremia (123 mEq/L) Recurrent, focal, resolved prior to arrival

2-y-old male, trileptal toxicity for seizure disorder due to history of left Middle cerebral artery aneurysm

Leukocytosis (WBC 18000)

and thrombocytopenia (83000)

Transaminitis, metabolic acidosis

(bicarbonate 9 mEq/L), leukocytosis (WBC 19300)

Recurrent, generalized, active 6-y-old female, ex-25-wk twin VP shunt, altered

mental status due to urosepsis

Recurrent, generalized, active 16-y-old female, eclampsia, 4 d postpartum

Patient Abnormal CT finding Seizure characteristics Pertinent history

Large bilateral choroid fissure cyst with mass effect on hippocampus

Recurrent, generalized, active 4-y-old male, Phelan-McDermid syndrome, chromosome 22q13 deletion, developmental delay

Enlarged left ventricle Recurrent, generalized, active 6-y-old female, ex-25-wk twin, VP shunt,

altered mental status due to urosepsis

Increased ventriculomegaly Recurrent, generalized, resolved 4-y-old male, VP shunt secondary to

nonaccidental traumatic brain injury

A description of patient history of present illness and medical history for patient with clinically significant laboratory or brain CT Imaging results. Of note, a sex or age difference was not found.

ESRD, end-stage renal disease; FTT, failure to thrive, X; VP, Ventriculoperitoneal, X; WBC, white blood cell count.

predictors such as focal vs generalized seizure, persistent active seizure in the ED, and first or recurrent seizure and did not reveal clinically sig- nificant predictors that could increase the yield of CT imaging (Table 5). Saz and colleagues [6] recommended further workup in children younger than 2 years with focal or active (defined as N 10 minutes) sei- zure. We examined different age categories to determine if age might serve as a predictor of increased yield of laboratory testing or emergent CT neuroimaging (Tables 4 and 5). Clinically significant laboratory

ventriculoperitoneal shunt, and the other had a chromosomal abnor- mality with developmental delay.

Limitations

The limitations of our study include its retrospective nature and the possibility of inaccuracies during data collection. Our small sample size prevented the study from reaching statistical significance. However, as

1264 A.M. Strobel et al. / American Journal of Emergency Medicine 33 (2015) 1261-1264

mentioned above, our results are similar to those reported by other inves- tigators (Table 1). Our study was not blinded, presenting the possibility of experimenter bias. Unfortunately, numerous medical records had incom- plete documentation of the neurologic examination, trauma preceding seizure, last dose of antipyretics, or inconsistencies in reporting the results, so we could not reliably use this aspect of the assessment as a predictor. However, the neurologic examination should be considered for future studies aiming to risk stratify the diagnostic evaluation of pediatric patients with nonfebrile seizures. We did not have complete documentation of whether or not trauma preceded the seizure or timing of last dose of antipyretic prior to seizure. Future prospective studies may want to include these when determining inclusion criteria.

Conclusion

We were unable to identify statistically significant predictors of laboratory or CT abnormalities for children with nonfebrile seizure pre- senting to the ED. The predictors that showed a trend toward signifi- cance in regard to increased yield of clinically significant laboratory findings were as follows: age less than 2 years, having an active seizure in the ED, and experiencing a first seizure. In accordance with AAN guidelines, we conclude that the history of present illness preceding the seizure should be used to determine the need for further testing, rather than the seizure itself in otherwise healthy children back to their neurologic baseline.

References

1. Martindale JL, Goldstein JN, Pallin DJ. Emergency department seizure epidemiology. Emerg Med Clin North Am 2011;29:15-27.
2. Hirtz D, Ashwal S, Berg A, Bettis D, Camfield P, Crumrine P, et al. Practice parameter: evaluating a first nonfebrile seizure in children: report of the Quality Standards Sub- committee of the American Academy of Neurology, the Child Neurology Society, and the American Epilepsy Society. Neurology 2000;55:616-27.
3. Sharma S, Riviello JJ, Harper MB, Baskin MN. The role of emergent neuroimaging in children with new-onset afebrile seizures. Pediatrics 2003;111:1-5.
4. Landau YE, Waisman Y, Shuper A. Management of children with non-febrile seizures in the emergency department. Paediatr Neurol 2010;14:439-44.
5. Maytal J, Krauss JM, Novak G, Nagelberg J, Patel M. The role of brain computed tomography in evaluating children with new onset of seizures in the emergency department. Epilepsia 2000;41:950-4.
6. Saz EU, Polat TB, Gokben S. Are Laboratory workups really needed on management of pediatric seizures at emergency rooms? J Med Med Sci 2010;1:241-4.
7. Scarfone RJ, Pond K, Thompson K, Fall I. Utility of laboratory testing for infants with seizures. Pediatr Emerg Care 2000;16:309-12.
8. Valencia I, Sklar E, Blanco F, Lipsky C, Pradell L, Joffe M, et al. The role of routine serum laboratory tests in children presenting to the emergency department with unprovoked seizures. Clin Pediatr 2003;6:511-7.
9. Akhavan KS, Modares MM, Fallah R. Utility of laboratory studies in seizures of children older than one month. Singapore Med J 2009;50:814-6.
10. Nypaver MM, Reynolds SL, Tanz RR, Davis AT. Emergency department laboratory eval- uation of children with seizures: dogma or dilemma? Pediatr Emerg Care 1992;8:13-6.
11. Miglioretti DL, Johnson E, Williams A, Greenlee RT, Weinmann S, Solberg LI, et al. The use of computed tomography in pediatrics and the associated radiation exposure and estimated cancer risk. JAMA Pediatr 2013:1-8.
12. Hill Alan. Neonatal seizures. Pediatr Rev 2000;21:117-21.
13. Gillam-Krakauer Maria, Carter Brian. Neonatal hypoxia and seizures. Pediatr Rev 2012;33:387-97.
14. Olson DM. Neonatal seizures. Neoreviews 2012;13:213-23.
15. Bachur RG, Harper MB. predictive model for serious bacterial infections among infants less than 3 months of age. Pediatrics 2001;108:311-6.
16. Subcommitte on Febrile Seizures. Febrile seizures: guidelines for the neurodiagnostic evaluation of the child with a simple febrile seizure. Pediatrics 2011;127:389-94.
17. Garvey MA, Gaillard WD, Rusin JA, Ochsenschlager D, Weinstein S, Conry JA, et al. Emergent brain computed tomography in children with seizures: who is most likely to benefit. J Pediatr 1998;133:664-9.
18. Farrar HC, Chande VT, Fitzpatrick DF, Shema SJ. Hyponatremia as a cause of seizures in infants: a retrospective analysis of incidence, severity, and Clinical predictors. Ann Emerg Med 1995;26:42-8.
19. Berry PL, Belsha CW. Hyponatremia. Pediatr Clin North Am 1990;37:351-63.
20. Jenny C, Hymel KP, Ritzen A, Reinert SE, Hay TC. Analysis of missed cases of abusive head trauma. JAMA 1999;281:621-6.
21. Sugar NF, Taylor JA, Feldman KW. Bruises in infants and toddlers: those who don’t cruise rarely bruise. Puget Sound Pediatric Network. Arch Pediatr Adolesc Med 1999;153:399-403.
22. Bechtel K, Stoessel K, Leventhal JM, Ogle E, Teague B, Lavietes S, et al. Characteristics that distinguish accidental from abusive injury in hospitalized young children with head trauma. Pediatrics 2004;114:165-8.
23. Rahman S, Footitt EJ, Varadkar S, Clayton PT. Inborn errors of metabolism causing epilepsy. Dev Med Child Neurol 2013;55:23-36.
24. Burton BK. Inborn errors of metabolism in infancy: a guide to diagnosis. Pediatrics 1998;102:1-9.
25. Clinical policy: critical issues in the evaluation and management of adult patients pre- senting to the emergency department with seizures. Ann Emerg Med 2004;43:605-25.
26. Turnbull TL, Vanden Hoek TL, Howes DS, Eisner RF. Utility of laboratory testing in the emer- gency department patient with a new-onset seizure. Ann Emerg Med 1990;19:373-7.
27. Henneman PL, DeRoos F, Lewis RJ. Determining the need for admission in patients with new-onset seizures. Ann Emerg Med 1994;24:1108-14.