Indication of Cranial CT Scan after Post-Traumatic Seizure in Children: A Retrospective Study at CHUJRA, Madagascar

Background: Post-traumatic seizures (PTS) in children are a frequent complication of head trauma. This study aimed to describe the epidemiological and clinical profiles of children admitted with PTS and evaluate the relevance of cranial computed tomography (CT) scan indications based on clinical presentation.

Methods: We conducted a retrospective descriptive study of 78 of 696 children admitted for head trauma in the Department of Neurosurgery at the University Hospital Center Joseph Ravoahangy Andrianavalona between January 2017 and December 2019.

Results: PTS occurred in 20.7% of the cases. A CT scan was performed in 54.2% of the children who had seizures. Most seizures were early (88.5%) or single (80.8%), with a preserved Glasgow score (≥ 13 in 74.4%). CT scans were normal in 35.9% of cases. The main abnormalities were simple skull fractures (25.6%) and depression fractures (23.1%).

Conclusion: CT should be reserved for children with multiple seizures, neurological deficits, low Glasgow scores, or prolonged loss of consciousness.

Introduction

Posttraumatic seizures (PTS) in children are a common complication of head trauma. They most often occur within the first 24 hours after injury. Their incidence ranges from 2% to 20%, depending on the severity of trauma [1]. PTS may indicate an underlying brain injury, such as an epidural hematoma, cerebral contusion, or diffuse brain swelling, and may worsen neurological outcomes owing to secondary metabolic effects.

Brain computed tomography (CT) remains the key imaging tool for detecting intracranial lesions. However, because of radiation exposure, its use must be carefully considered, particularly in pediatric patients [2]. The American College of Radiology (ACR) recommends performing brain CT in children who present with more than one PTS or with a single seizure combined with other warning signs such as prolonged loss of consciousness, neurological deficit, persistent vomiting, or a Glasgow Coma Scale score ≤14 [3].

In our setting, at the neurosurgery department of University Hospital Center Joseph Ravoahangy Andrianavalona (CHUJRA), where access to imaging is limited, it is important to identify clinical profiles that are more likely to be associated with abnormalities seen on CT in children with PTSs. The objective of this study is to describe the epidemiological and clinical profile of children who presented with PTSs and to assess CT scan findings in a resource-limited context.

Methods

We conducted a retrospective descriptive study in the neurosurgery department of CHUJRA in Antananarivo, Madagascar, over a period of three years, from January 1, 2017, to December 31, 2019. We included all patients aged 0–15 years who were admitted for head trauma complicated by at least one PTS and who underwent a brain CT scan, with complete medical records. We excluded patients with non-traumatic seizures, a history of epilepsy or known neurological disorders, and those without a brain CT scan. We used exhaustive sampling that included all eligible patients admitted during the study period.

Data were collected from hospitalization records, CT scan reports, and clinical observation sheets.

The variables analyzed included:

• Frequency of cases;

• Sociodemographic data (age, sex);

• Time to hospital: the time between trauma and arrival at CHUJRA emergency department;

• Time of seizure onset, duration, number, and type of seizures

• Clinical findings on admission (loss of consciousness, associated symptoms such as vomiting or headache, Glasgow Coma Scale score, pupillary status, motor deficit, and local signs)

• Delay to CT scan, CT scan findings

• Management modalities (medical or surgical);

• Length of hospital stay;

• Discharge outcome.

Data were entered and analyzed using IBM SPSS Statistics version 26.0.

Ethical Considerations: This retrospective study was conducted using anonymized data, and ethical approval was not required according to the policy of our institutional review board.

Limitations of the study: This study has several limitations. First, its retrospective design exposes it to potential bias related to the completeness and accuracy of medical records. The sample size was relatively small (78 cases), which may limit the statistical power and generalizability of the findings. Moreover, only patients who underwent a CT scan were included, which could introduce a selection bias and may not represent all children with post-traumatic seizures. Finally, the lack of long-term follow-up did not allow assessment of neurological outcomes or the occurrence of late post-traumatic epilepsy.

Results

We included 78 out of 696 children hospitalized for head trauma in the Neurosurgery Department at CHUJRA, representing 11.2%. Among the 696 patients, 144 (20.7%) had a PTS. Of these, 78 (54.2%) underwent brain CT.

Seizures were most common in children aged 0–3 years (51.3%; n = 40). The mean patient age was 3 years (range, 8–15 years). There was a slight male predominance (51.3%) with a sex ratio of 1.05. Most patients (69.2%) arrived at the hospital within 6 h of trauma. The average time to consultation was 15.8 hours, ranging from 30 min to 11 days.

In terms of seizure timing, 88.5% of seizures occurred within the first week (early seizures), including 10.2% within the first 24 h (immediate seizures). Late seizures (after seven days) were rare (1.3%). In 49.3% of the cases, seizures lasted for less than 5 min. The majority of patients (80.8%) experienced a single seizure, 12.8% had two seizures, 5.1% had three, and only 1.3% had four seizures. Generalized seizures were the most frequent type of seizure (60.3%).

Loss of consciousness was reported in 55.1% of cases, but the duration was often unclear (21.8%). Only 14.1% of the patients had a brief loss of consciousness (< 5 min). Headaches were present in 11.5% of the patients. The Glasgow Coma Scale score was 13–15 in 74.4%, 9–12 in 19.2%, and ≤ 8 in 6.4%. Pupils were equal and reactive in 96.2% of the cases. Anisocoria is rare (3.8%). Motor deficits (hemiparesis or hemiplegia) were observed in 7.7% of the patients.

The most common local sign was scalp hematoma (34.6%), followed by facial bruising and depressed skull fractures (11.5% of each). Brain CT scans were performed after 6 h in 60% of cases.

CT findings were normal in 35.9% of the cases. The most frequent lesions were simple skull fractures (25.6%), followed by depressed fractures (23.1%), cerebral contusions (19.2%), extradural hematomas (12.8%), acute subdural hemorrhage (9%), subarachnoid hemorrhage (11.5%), and pneumocephalus (2.6%).

We classified the cases into 9 clinical groups based on seizure characteristics, GCS, and focal signs (Table I).

Table I. Clinical Groups and Corresponding CT Scan Findings in Children with Post-Traumatic Seizures (n = 78)

Group	Definition	Normal CT n (%)	Abnormal CT n (%)
Group 1a	Single early seizure, GCS 13-15, no focal signs	20 (25.6)	20 (25.6)
Group 1b	Single immediate seizure, GCS 13-15, no focal signs	1 (1.3)	0
Group 2	Single early seizure, GCS 13-15, with focal signs	2 (2.6)	7 (9)
Group 3a	Single early seizure, GCS 9-12, no focal signs	2 (2.6)	0
Group 3b	Single immediate seizure, GCS 9-12, no focal signs	1 (1.3)	1 (1.3)
Group 4	Single early seizure, GCS 9-12, with focal signs	0	1 (1.3)
Group 5a	Single early seizure, GCS 3-8, no focal signs	0	3 (4)
Group 5b	Single late seizure, GCS 3-8, no focal signs	0	5 (6.4)
Group 6a	Two early seizures, GCS 13-15, no focal signs	2 (2.6)	2 (2.6)
Group 6b	Two immediate seizures, GCS 13-15, no focal signs	0	1 (1.3)
Group 7	Two immediate seizures, GCS 9-12, with focal signs	0	4 (5.2)
Group 8	Three early seizures, GCS 13-15, no focal signs	0	4 (5.2)
Group 9	Four early seizures, GCS 9-12, with focal signs	0	5 (6.4)

All the patients received analgesics. Anticonvulsants were administered in 60.3% of the patients and anti-edema treatment in 35.9%. Emergency surgery was needed in 10.3% (n = 8) of cases: five extradural hematoma evacuations, one depressed fracture elevation, one cranio-cerebral wound debridement, and one case not operated due to lack of resources. Hospital stays lasted 3–7 days in 51.3% of the cases, with a mean of 4.01 days (range: 2–15). Most of the patients (93.5%) were discharged. One death was reported.

Discussion

In our study, PTS was observed in 20.7% of children hospitalized for traumatic brain injury (TBI), a proportion higher than that reported in some international studies. This may be explained by the local epidemiological context and low rate of early medical consultations. Most patients were very young (mean age, 3 years), which aligns with the existing literature showing that PTS is more common in children aged < 5 years [4].

Most seizures were early onset (88.5%) and isolated (80.8%), indicating a low-risk profile according to current guidelines. The Glasgow Coma Scale (GCS) score was 13 or higher in > 74% of cases. According to the American College of Emergency Physicians, a CT scan is not required when the neurological exam is normal and no other risk factors are present [3].

In our cohort, 35.9% of the brain CT scans were normal. The abnormalities were mainly simple or depressed skull fractures. This raises the question of whether routine CT imaging is necessary for all children presenting with PTS, especially in cases with a single, brief seizure, preserved GCS, and no focal neurological signs.

Several authors have shown that abnormal CT findings are rare in children with isolated seizures, especially in the absence of prolonged loss of consciousness or neurological deficits [4], [5]. Badawy et al. [6] found significant associations between the likelihood of CT abnormalities and longer delays before the onset of post-traumatic symptoms (p = 0.006) as well as longer seizure duration (p < 0.001).

In resource-limited settings, adapting these algorithms may help optimize imaging decisions, reduce unnecessary radiation exposure, and better identify high-risk cases.

Our findings support this approach: normal CT scans were common in Group 1 patients (single seizure, GCS 13–15, and no focal signs). Conversely, abnormal findings were more frequent in groups with multiple seizures, lower GCS scores, and neurological deficits. These observations are consistent with the literature-based decision rules aimed at reducing unnecessary CT scans.4 In clinical practice, implementing decision algorithms based on these risk factors may help optimize imaging, lower healthcare costs, and reduce radiation exposure.

Based on our study findings and published literature, we recommend the following:

• CT scan should be indicated in PTSs in children if:

∘ There are multiple seizures (≥ 2), regardless of GCS score;

∘ GCS score is < 13;

∘ There are focal neurological signs;

∘ There is prolonged loss of consciousness (> 5 minutes);

∘ The seizure occurs more than 7 days after trauma.

• CT scan is not systematically indicated in cases of:

∘ Single brief seizure (< 5 minutes);

∘ GCS score ≥ 13;

∘ Normal neurological examination;

∘ No or brief loss of consciousness;

∘ No focal neurological signs.

These recommendations are consistent with the validated PECARN [7], NICE [8], and CATCH [9] algorithms adapted to the context of low-resource settings.

Conclusion

Post-traumatic seizures in children are a common complication of head injury, but most occur early, are isolated, and of short duration, with a preserved Glasgow Coma Scale score. In this context, CT scans were often normal or showed only minor lesions. Therefore, systematic CT scanning is not always warranted. Indications should be restricted to children with multiple seizures, low Glasgow scores, neurological deficits, or prolonged loss of consciousness. Adapting these recommendations to resource-limited settings such as Madagascar could reduce unnecessary radiation exposure and optimize the use of imaging resources.

Conflict of Interest

No potential conflict of interest relevant to this article was reported.