Introduction
Benign convulsions with mild gastroenteritis (CwG) are provoked seizures that occur in infants and young children with acute gastroenteritis, in the absence of fever, hypoglycemia, electrolyte imbalance, or central nervous system infection [1-4]. These seizures frequently present in clusters, posing challenges for both clinicians and parents [2,4-6]. Nevertheless, the prognosis for CwG is generally favorable, with most cases resulting in complete seizure resolution and normal neurodevelopmental outcomes [7-10]. Compared to febrile seizures (FS), patients with CwG have a lower risk of developing epilepsy or experiencing recurrent seizures [8,10,11].
Cluster seizures, defined as two or more seizures within a short period, have been reported in 50% to 75% of CwG cases [2,5,9,12,13]. The median or mean number of seizures in CwG typically ranges from two to three [2,4,5,12]. No clear racial differences in the frequency of cluster seizures in CwG have been documented. Reported rates include 75% in Japan [2], 68.6% and 57.1% in Korea [5,12], 57% in Italy [13], and 63.4% and 36.6% in China [9,14].
A few studies have examined the clinical and laboratory features associated with cluster seizures in CwG [12,14,15]. Comparisons have been made between patients with single versus multiple seizures, two or fewer versus more than two seizures, and less than five versus five or more seizures [12,14,15]. Fan et al. [14] found that serum sodium and calcium levels were lower in patients with cluster seizures compared to those with isolated seizures. You [12] also reported a positive correlation between seizure frequency and patient age. Given the paucity of data, further research is needed to clarify the factors that influence seizure frequency in CwG. Therefore, the present study aims to determine whether clinical presentation and laboratory characteristics vary according to seizure frequency and to identify specific factors that are influenced by seizure frequency.
Materials and Methods
1. Study design and setting
This retrospective cross-sectional study was conducted to compare clinical, laboratory, and neuroimaging data among patients with varying frequencies of seizure attacks, spanning the period from March 2005 to February 2024. Patients were categorized into three groups based on seizure frequency: those with a single seizure (group I), two seizures (group II), and three or more seizures (group III). Additional comparisons were performed between newly defined groupings: group I vs. groups II/III, and groups I/II vs. group III.
2. Study population
The study included patients aged 3 to 36 months who were diagnosed with CwG and admitted to Chonnam National University Hospital (CNUH) between March 2005 and February 2024. CwG was defined as afebrile seizures precipitated by acute gastroenteritis in patients with normal development. Seizures due to fever (body temperature >38.0°C), hypoglycemia, hyponatremia, hypocalcemia, or meningitis were excluded. Cases lacking sufficient information to confirm a diagnosis of CwG were also excluded. Medical records of all enrolled patients were retrospectively reviewed. This study was approved by the ethics committee of CNUH (IRB number: CNUH-2024-151). Written informed consent by the patients was waived due to a retrospective nature of our study.
3. Data collection
Demographic data (age, sex) and clinical history prior to seizure onset, including illness history, developmental status, and family history of seizures, were obtained. Basic clinical information at admission included body temperature (during, before, or after seizures), enteric symptoms (such as vomiting, diarrhea, and abdominal pain), and the date and season of symptom onset. The interval between the onset of enteric symptoms and the onset of seizures was also recorded. Seizure-related data included frequency, the interval from the first to last seizure, changes in seizure type during recurrent episodes, and use of antiseizure medication. For each episode, the seizure type and duration were documented. Information on revisits due to seizure recurrence or developmental concerns after CwG was also reviewed.
Electroencephalography (EEG) and brain imaging studies performed at admission were assessed. Laboratory data collected included initial serum glucose, electrolyte levels, uric acid, blood cell counts, lactate levels, and stool viral reports at admission.
4. Statistical analysis
Data are presented as either the number of cases (%) or as the median with interquartile range (IQR). Categorical variables were analyzed using the Pearson chi-square test, and the Fisher exact test was applied for data categories with fewer than five cases. Continuous variables (e.g., age, days, and laboratory profiles) were compared using the Mann-Whitney U test for two-group comparisons and the Kruskal-Wallis test for comparisons among three groups. Statistical significance was defined as P<0.05. All statistical analyses were conducted using SPSS version 29.0 (IBM SPSS Statistics Corp., Armonk, NY, USA).
Results
1. Clinical characteristics of patients with CwG
A total of 178 patients diagnosed with CwG were included, of whom 97 were female (54.5%). The cohort was divided into three groups: group I (n=50, 28.1%), group II (n=46, 25.8%), and group III (n=82, 46.1%). Notably, 71.9% of patients experienced cluster seizures. The median age at presentation was 17.5 months (IQR, 14.8 to 22.0). Seizures occurred most frequently during the winter season (48.3%). There were no statistically significant differences in sex, age, or seasonal distribution among the three groups or between the redefined groupings (group I vs. groups II and III; groups I and II vs. group III).
With respect to medical history, five patients (2.8%) had a history of FS, and four (2.2%) had previously experienced CwG. Other organ anomalies were observed in 10 patients (5.6%), with cardiac anomalies—particularly atrial or ventricular septal defects—being the most common (n=6). Family history included FS in 8.4% (n=15), unprovoked seizures in 6.2% (n=11), and CwG in 1.1% (n=2). In group III, a family history of unprovoked seizures was reported in 8.5% of patients, compared to 4% in group I and 4.3% in group II (P=0.227).
Enteric symptoms preceded seizures in 93.8% of cases, with a median interval of 2.0 days from the onset of enteric symptoms to seizure onset (IQR, 1.0 to 3.0 days). Vomiting occurred significantly more often in patients with two or more seizures (86.7%) compared to those with a single seizure (72%, P=0.020). However, the frequencies of other enteric symptoms or non-seizure-related fever did not differ significantly among the groups or between the redefined groupings (Table 1).
2. Characteristics of seizures in patients with CwG
The median seizure count was 2 (range, 1 to 3). In group III, the frequency of seizures per individual ranged from three to 10 episodes (mean±standard deviation: 3.91±1.57). Among the 128 patients who experienced recurrent seizures, 12 (9.4%) exhibited changes in seizure type. The interval between the first and last seizures differed significantly between groups II and III: intervals of <6 hours occurred in 69.6% of group II but only 35.4% of group III (P=0.000), while intervals of 13 to 24 hours were observed in 0% of group II but in 19.5% of group III (P=0.001). Most seizures resolved within 24 hours—97.8% in group II and 93.9% in group III (Table 2).
EEG was performed in 175 patients (98.3%). Diffuse or posterior slowing was observed in 28.3% (36 of 127) of EEGs from groups II and III, compared to 14.6% (seven of 48) in group I; however, this difference did not reach statistical significance (P=0.059). Focal epileptiform discharges were found in only two patients, both from group III. The rate of brain imaging studies was significantly higher in group I (100.0%) compared to groups II and III (91.4%, P=0.032). Mild abnormalities—including transient splenial lesions (n=2), pineal gland or arachnoid cysts, Chiari I malformation, hippocampal sclerosis, and dilated ventricles—were identified in a small subset of patients (n=8, 4.8%), with no significant differences among groups (Table 2).
Antiseizure medications were more commonly administered to patients with cluster seizures, with usage rates of 4%, 43.5%, and 81.7% in groups I, II, and III, respectively (P=0.000) (Table 2 and Fig. 1). Rescue therapy with benzodiazepines, such as lorazepam and diazepam, was also more frequently required in patients with higher seizure counts (P=0.000): 2% in group I, 43.5% in group II, and 74.4% in group III. Benzodiazepines were effective in over 95% of cases in groups I and II, but less so in group III, where 47.5% required second-line therapy. Phenytoin and fosphenytoin were predominantly used in group III, accounting for 93.9% of the 33 total trials, and were mainly administered as second-line treatments (80.1% of 31 trials in group III). In group III, phenytoin was effective in 83.3% of first-line cases (n=6) and in 96% of second-line cases (n=25). Intravenous phenobarbital (n=6) and oral carbamazepine (n=2) were trialed only in group III (Fig. 1). No patient required antiseizure medication after discharge. Post-discharge, 12 patients (6.7%) revisited the hospital due to seizures: seven with FS (3.9%), four with CwG (2.2%), and two with unprovoked seizures (1.1%) (Table 2). Both patients with subsequent unprovoked seizures were from group III.
In total, 461 seizure episodes were documented: 50 in group I, 92 in group II, and 319 in group III. Seizure duration and type differed significantly among groups (P=0.004 and P=0.025, respectively). In group III, the majority of seizures (98.4%) resolved within 5 minutes, while 10% of seizures in group I (vs. 1.6% in group III) lasted longer than 5 minutes. Focal tonic-clonic or clonic seizures occurred only in group II (n=2) and group I (n=1) (Table 3).
3. Laboratory findings in patients with CwG
Norovirus genogroup II (GII) was the most frequently detected pathogen in stool samples, accounting for 61.2% of 116 cases tested, followed by group A rotavirus (16.7% of 162 cases) and enteric adenovirus (10.3% of 116 cases). In group I, norovirus GII was detected in 50.0% of cases (vs. 65.5% in groups II and III, P=0.126), and rotavirus in 21.7% (vs. 14.7% in groups II and III, P=0.275). No significant differences were observed in the distribution of viral pathogens among the three groups or between the redefined groupings (Table 4).
A statistically significant difference was found in median platelet count and serum calcium levels between patients with two or fewer seizures and those with three or more: 312×10³/mm³ (IQR, 257 to 368) vs. 277×10³/mm³ (IQR, 221 to 333), P=0.024; and 9.7 mg/dL (IQR, 9.3 to 10.1) vs. 9.5 mg/dL (IQR, 9.1 to 9.9), P=0.038, respectively. Uric acid levels were elevated, with a median of 8.1 mg/dL (IQR, 6.3 to 9.9), but there were no significant differences among the original groups or the redefined groupings (Table 5).
Discussion
The clinical presentation and laboratory characteristics of CwG vary with seizure frequency, influencing factors such as vomiting, seizure duration and type, platelet counts, and serum calcium levels, as demonstrated in this study.
A previous study by You [12], which included 105 patients, found a positive correlation between age and seizure frequency. The mean age of patients with two or fewer seizures was 18.16±7.08 months, while it was 22.33±6.34 months in those with three or more seizures [12]. In our study, the median age of all CwG patients was 17.5 months (IQR, 14.8 to 22.0), with no significant differences in age, sex, or season across groups stratified by seizure frequency. This aligns with findings by Fan et al. [14], who also reported no significant differences in age, sex, or season between isolated and cluster seizure groups.
You [12] reported that the frequency or severity of enteric symptoms was not associated with seizure frequency. In contrast, our study found that vomiting was significantly more frequent in patients with two or more seizures compared to those with a single seizure. The frequencies of other enteric symptoms and fever unrelated to seizures did not significantly differ between groups. The viral composition identified in our study also did not significantly vary among groups. This finding is consistent with previous studies that have reported no significant differences in causative viral pathogens based on seizure frequency [12,14].
Seizure characteristics, including the interval from enteric symptom onset to seizure onset, seizure duration, and seizure type, were previously compared by Fan et al. [14], who found no significant differences between isolated and cluster seizure groups. In our study, the median interval from enteric symptom onset to seizure onset was 2.0 days (IQR, 1.0 to 3.0), and this did not differ significantly between groups, mirroring the results of Fan et al. [14]. However, differences were observed in seizure duration and type: the majority of seizure episodes in patients with three or more seizures resolved within 5 minutes (98.4%), whereas 10% of episodes in patients with isolated seizures persisted beyond 5 minutes. Focal tonic-clonic or clonic seizures were only observed in patients with two or fewer seizures. The interval between the first and last seizure in patients with cluster seizures also varied significantly between groups II and III in our study, particularly for intervals of less than 6 hours (69.6% vs. 35.4%) and 13-24 hours (0% vs. 19.5%).
Diffuse or posterior slowing on EEG was observed in 28.3% of patients with cluster seizures, compared to 14.6% of those with a single seizure, although this did not reach statistical significance. Focal epileptiform discharges were noted in only two patients in group III. These findings are consistent with those of Fan et al. [14]. In our study, two patients who experienced one or two brief seizures (<5 minutes) presented with reversible splenial lesions, which have been previously documented in CwG patients [16-18].
Benzodiazepines were effective in over 95% of patients with one or two seizures, but were less effective in group III, where 47.5% of patients required second-line therapies. Among patients with three or more seizures, phenytoin was effective as a first-line treatment in 83.3% of cases (six patients) and as a second-line treatment in 96% of cases (25 patients). The efficacy of phenytoin as a second-line therapy was reported as 80% in the study by Okumura et al. [19].
Fan et al. [14] demonstrated that serum sodium and calcium levels were lower in patients with cluster seizures compared to those with a single seizure. In our cohort, there was no significant difference in serum sodium or calcium levels between the two groups. However, median serum calcium levels were notably lower in patients with three or more seizures compared to those with two or fewer seizures. Similarly, median platelet counts were significantly lower in patients with three or more seizures. The overall median serum sodium level in our cohort was slightly low at 135.0 mEq/L (IQR, 133.0 to 137.0), consistent with previously reported values [20,21]. The median serum uric acid level was elevated at 8.1 mg/dL (IQR, 6.3 to 9.9), which aligns with findings from prior studies [5,15,22,23].
A decreased platelet count and lower serum calcium level have also been reported in patients with recurrent or complex FSs [24,25]. Platelets are recognized as mechanical scavengers that secrete chemokines in response to infection or inflammation [26]. Infection leads to platelet activation and aggregation, initiating a cascade that results in platelet consumption [24,26]. Therefore, patients with CwG who experience recurrent seizures (≥3 episodes) and have a reduced platelet count may have more severe inflammation than those with only one or two seizures.
Chen et al. [25] suggested that patients with FSs and lower serum calcium levels may be more susceptible to seizures due to a lowered seizure threshold, resulting in more frequent episodes than those with higher calcium levels. In cases of CwG, Yeom et al. [27] reported that serum calcium levels were decreased in patients with frequent seizures following the introduction of the rotavirus vaccine. They hypothesized that seizure susceptibility might be increased due to reduced serum calcium levels, potentially caused by rotavirus enterotoxin (nonstructural protein 4 [NSP4]), which enhances cell membrane permeability to Ca²+ [27]. However, the pathophysiological mechanisms associated with other viruses remain unclear.
In our study, the recurrence rate of afebrile seizures requiring hospital revisits was 3.4% (n=6), including 2.2% (n=4) with CwG and 1.1% (n=2) with unprovoked seizures, with no significant differences among groups. Previous studies with follow-up periods exceeding two years have identified cluster seizures within 24 hours as a risk factor for CwG relapse [6,8]. Reported relapse rates of afebrile seizures range from 5.6% to 7.7% [5,7-9], with the majority being CwG cases, accounting for approximately 4.8% to 6.3% of all CwG patients [6,8-10].
This study has several strengths. First, unlike previous studies that compared only two groups, we classified patients into three distinct groups—single seizure (group I), two seizures (group II), and three or more seizures (group III)—and conducted comparisons between newly defined group pairs for a more nuanced analysis. Second, our data set was comprehensive, including demographic information, detailed clinical history, seizure characteristics, laboratory, electrophysiological, and neuroimaging data, as well as treatment outcomes. Third, this study included the largest cohort to date among studies comparing clinical and laboratory characteristics of isolated versus clustered seizures in CwG. However, the retrospective single-center design and limited sample size may restrict the generalizability of our findings. Therefore, future large-scale, multicenter, and prospective studies are needed to validate these results. It would also be beneficial to examine these findings in diverse racial groups to enhance their broader applicability.
In conclusion, certain clinical and laboratory characteristics of CwG vary according to seizure frequency. While age, sex, season, and virus type in stool did not differ significantly between groups, vomiting occurred more frequently in patients with two or more seizures. Brief seizure episodes (<5 minutes) were more common in patients with three or more seizures. Focal tonic-clonic or clonic seizures were observed exclusively in patients with two or fewer seizures. Patients with three or more seizures had lower initial platelet counts and serum calcium levels than those with two or fewer seizures. In these patients, benzodiazepines were less effective, with a failure rate of about 50%, whereas phenytoin was effective in over 80% of cases as either first- or second-line treatment. Therefore, in patients with cluster seizures (≥3 episodes) and CwG, early use of phenytoin or fosphenytoin may be considered, while repeated use of benzodiazepines should be avoided. Certain clinical and laboratory characteristics may serve as biomarkers for predicting cluster seizures (≥3 episodes) in CwG, providing valuable guidance for selecting antiseizure medications.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Cited
