Rudolph's Pediatrics, 22nd Ed.

CHAPTER 560. Evaluation of New Onset Seizures

Alexander Rotenberg

A child’s first seizure is most often a brief, less than 5 minutes in duration, generalized convulsion that is justifiably concerning to physicians and parents. The overall incidence of a single unprovoked seizure approximates 60 per 100,000 persons/year with a peak incidence in the first year of life approximating 130 per 100,000 persons/year.1 In some instances, the first seizure is provoked by an underlying acute illness. In others, it is a manifestation of a chronic syndrome. In about half of the cases, a seizure recurs and the child meets diagnostic criteria for epilepsy.2 Accordingly, the evaluation of a new-onset seizure is centered on identifying its cause and predicting likelihood of recurrence.

Broadly, by the initial history and examination, a first seizure can be categorized into 1 of 3 groups: febrile, acute symptomatic, and remote symptomatic. Management of febrile seizures is discussed in Chapter 559, and of status epilepticus (seizure lasting greater than 5 minutes) in Chapter 561.


An acute symptomatic seizure is triggered by a recent or ongoing neurologic insult such as infection, trauma, stroke, or metabolic disturbance. Examples include seizures in the setting of encephalitis, head injury, or intoxication. Evaluation of an acute symptomatic seizure, even if the seizure is brief, is often emergent and guided by the overall clinical picture. If the history and examination suggest that a seizure is acutely provoked, then tests typically include blood tests to diagnose infection or metabolic disturbance, toxicologic screening, and head computerized tomography (CT) if trauma or hemorrhage is suspected. Lumbar puncture may also be warranted if there is suspicion that an intracranial infection caused the seizure. If after the seizure the patient has not recovered fully, has persistent focal neurologic deficits, or has not regained baseline consciousness, testing may be followed by magnetic resonance imaging (MRI) to evaluate for a structural abnormality, or electroencephalography (EEG) to evaluate for nonconvulsive seizures (see Chapter 561).


If the history and examination indicate that a child’s first seizure occurred without accompanying illness or trauma, and the child has recovered to his or her usual state of health, the seizure can be assumed to be remote symptomatic. In contrast to acute symptomatic seizures, remote symptomatic seizures reflect a chronic underlying vulnerability to seizure, such as a genetic seizure susceptibility or a relatively old brain injury. A common example of a remote symptomatic seizure is one that occurs in a child who had perinatal hypoxic/ischemic brain injury but did not develop seizures until a few years of age. In some reports, remote symptomatic seizures are distinguished from cryptogenic seizure by the presence of neurologic symptoms such as a static encephalopathy or a brain MRI abnormality in the remote symptomatic group, and an otherwise normal neurologic picture (normal neurologic exam, normal development, and normal brain MRI) in the cryptogenic group.3 However, for practical purposes of first seizure management, grouping cryptogenic seizures together with remote symptomatic seizures is useful.

Because remote symptomatic seizures are most often followed by full recovery to the neurologic baseline, and by definition are not associated with an acute central nervous system illness, their workup can be pursued at a slower pace than for acute symptomatic seizures. Evaluation of remote symptomatic seizures nearly always includes MRI and EEG, with other studies determined on a case-by-case basis. Lumbar puncture and head CT are generally not necessary in these cases.


Management of status epilepticus is discussed in Chapter 561. After recovery from a first seizure that is either acute or remote symptomatic, the EEG is among the better tools for estimating the likelihood of seizure recurrence and is indicated in all cases, according to recent practice parameters.4,5 Shinnar et al. identified that the risk of recurrence within 2 years of a first unprovoked seizure was about 30% in patients with a normal EEG, but was 50% to 60% in patients with EEG abnormalities. The risk of recurrence was greatest in this group if the EEG showed epileptiform abnormalities.3

A nonemergent brain MRI is indicated after most cases of first unprovoked seizure in childhood. Recent reports suggest that causal abnormalities for the patient’s seizure are found in as many as 20% of studies, although findings requiring intervention are found in only 1%.6,7 The predictive value of an MRI in estimating the likelihood of recurrence has not been studied as extensively as that of EEG.

In addition to etiology and abnormalities on EEG and MRI, other factors associated with increased probability of seizure recurrence are the seizure’s onset in sleep and prior febrile seizures.3

Whether to initiate antiepileptic treatment after a first seizure remains a controversial issue. A generally accepted practice in pediatric neurology is to defer drug therapy until a recurrence has occurred, unless the imaging and EEG data suggest a very high likelihood of seizure relapse. However, parents and patients can be made aware of the risk of recurrence based on the evaluation and can participate in the decision process.