Abstract
Background and Objectives
Optimal acute treatment is crucial in pediatric convulsive status epilepticus (CSE). However, data on the benefits of seizure action plans (SAPs) and treatment algorithm implementation in relation to process and outcome measures in CSE are scarce. Our study examines treatment algorithm adherence in benzodiazepine (BZD)-responsive and BZD-resistant groups and specifically focuses on the relationship with personalized SAPs.
Methods
We performed a prospective observational cohort study evaluating patient care processes and outcomes in young patients with CSE lasting ≥5 minutes, requiring admission and an antiseizure medication(s) (ASM) for seizure termination from February 2016 to July 2018. Patients with infantile spasms, invasive EEG monitoring, unclear seizure duration, unclear ASM administration time, or seizure cluster were excluded. We used univariate statistics to analyze the data.
Results
We enrolled 60 patients (median age 3.7 [1.9–7.0] years, 48% female), including 34 BZD-responsive and 26 BZD-resistant patients. Patients who had access to a personalized SAP, even if it was not adhered to, experienced a median (p25-p75) time to first ASM of 6 minutes (5–18; n = 34). By contrast, patients without access to a personalized SAP had a longer median (p25-p75) time to the first ASM of 15 minutes (8–27; n = 24; p < 0.05). The median (p25-p75) time to administer the first ASM was 6 (5–15; n = 33) minutes in BZD-responsive patients vs 15 (6–32; n = 25; p < 0.05) minutes in BZD-resistant patients. Treatment protocol implementation rates were lower in BZD-resistant vs BZD-responsive patients. The median (p25-p75) time to administer the first ASM was 5 minutes (4–5; n = 14) in patients who implemented a personalized SAP compared with 16 minutes (6–31; n = 20; p < 0.001) in patients without SAP implementation. The median (p25-p75) seizure duration in personalized SAP implementation and nonimplementation groups was 9 (6–16; n = 14) and 22 (11–64; n = 20; p < 0.01) minutes, respectively. The intubation rate was 14% for those who implemented the personalized SAP and 53% for those who did not (p < 0.05).
Discussion
Seizure duration was shorter in patients with personalized SAP, and the time to administer ASM was faster. In addition, BZD-resistant patients were less likely to follow treatment protocols, and the time to first-line therapy was slower. SAP and algorithm implementation was associated with a lower intubation rate, indicating potential benefits, including improved process and patient outcome measures.
Introduction
Status epilepticus (SE) is one of the most common pediatric neurologic emergencies. Status epilepticus affects up to 18 children per 100,000 each year,1-3 with mortality rates up to 7%.4 Age, etiology, and SE duration may predict outcome, and among these, SE duration and optimal treatment implementation may often be most easily modified.5-7
Rapid initiation and treatment escalation in pediatric SE have reduced seizure duration and led to more favorable outcomes, including decreased mortality.8 Evidence-based guidelines propose medication algorithms and time lines for each management step.9,10 However, time to rescue medication (RM) administration remains delayed in both prehospital and in-hospital settings.11-14 In addition, protocol adherence and treatment dosing are not always optimal.15,16 Of significance, the likelihood of SE lasting longer than 60 minutes increases by 5% for every minute between the onset of SE and arrival to the emergency department (ED).4 Higher treatment resistance and adverse outcomes are linked to such delays in medication administration.17-22 In this setting, the timely application of appropriately dosed RM is crucial.23-25 Furthermore, the effect of using a personalized seizure action plan (SAP)26,27 on health care utilization, such as reducing emergency visits and hospital admissions, is not consistently guaranteed.25,28 While SAPs based on guidelines are currently legislatively mandated in schools across 23 US states,29 no precise data exist on the relationship between algorithms and appropriately dosed RMs and the relationship between SAPs and outcomes.
The aim of this study was to help close the gap between clinical and legislative practice and the available scientific data. We aim to evaluate prehospital and in-hospital convulsive status epilepticus (CSE) management in a cohort of pediatric patients presenting at a tertiary care center. Specifically, we assess time to treatment, seizure duration, adherence to treatment guidelines, and personalized SAP in benzodiazepine (BZD)-responsive vs BZD-resistant patients to highlight potential improvement areas.
Methods
Study Design and Outcome Measures
We performed a prospective observational cohort study evaluating patient care processes and outcomes of 60 patients at Boston Children's Hospital (BCH). We enrolled participants between February 2016 and July 2018. Patients met the inclusion criteria if they were (1) aged 1 month to 21 years, (2) presented with CSE, defined as a single seizure ≥5 minutes, (3) required hospital admission or had CSE during hospitalization, and (4) needed at least 1 antiseizure medication (ASM) to abort the seizure. In the case of multiple SE episodes, we included a patient's first presentation during this period for analysis. We excluded patients if they had (1) current infantile spasms as the only seizure type, due to differences in treatment approaches and outcomes compared with CSE; (2) invasive EEG monitoring, because treatment protocols for these patients differ significantly and could bias the results; (3) unclear seizure duration; (4) unclear ASM administration time; (5) seizure cluster, because accurately calculating seizure onset and treatment during clusters proved challenging; or (6) nonconvulsive seizures at onset, given their distinct clinical presentation and treatment course compared with CSE.
We screened potential participants for eligibility by conducting a daily review of the neurology service census. Qualifying patients were consented for inclusion in the study. We determined the time to treatment based on information provided by family and emergency medical services (EMS) for prehospital onset and from provider information and hospital records once in hospital. We extracted additional variables from the medical record, including demographics, seizure/epilepsy history, intubation rates, intensive care unit (ICU) admission, and length of stay (LOS). We defined implementation of algorithms as medications administered at the appropriate doses (± 25%), within the appropriate time frame (± 25%), and in the suggested medication administration order. In the outpatient setting, doses are often rounded to standard increments (such as 2.5 mg or 5 mg), leading to slight variations from the recommended mg/kg dosing. Similarly, IV doses may be rounded to facilitate faster administration in emergencies. A 25% deviation from the suggested dose was predetermined as the acceptable cutoff to account for these variations. For standardized treatment algorithms, we used the American Epilepsy Society (AES)30,31 and BCH algorithms available at the time of the seizure. Personalized algorithms, or SAPs, were defined as individualized seizure rescue plans documented in the patient's chart, detailing medication choice, timing, and suggested dosing for at least the initial ASM. These plans varied by patient, with some specifying the initial BZD choice, dosage, and timing after seizure onset. More complex cases included both first-line and second-line ASMs, outlining the name, dose, order, and timing. We evaluated patients based on response to first-line BZD treatment and categorized them as either BZD-responsive (seizure resolved after BZD(s) only) or BZD-resistant (seizure resolution required treatment with additional non-BZD ASMs) patients.
In addition, we conducted a retrospective chart review of the medical records of patients included in the study to assess outpatient care gaps in those with a history of CSE and lack of RM use to understand reasons for not administering RM in the prehospital setting. The pre-BCH hospital setting includes non–health care locations such as homes, schools, or other public places, excluding outside hospitals and outpatient care centers. The outside hospital setting was defined as admission to a non-BCH hospital.
We stored data in REDCap (Research Electronic Data Capture, Vanderbilt University, Nashville, TN).
Statistical Analysis
We used descriptive statistics to summarize demographics and clinical outcome data, including counts and percentages for categorical variables. We summarized medians and interquartile ranges (25th percentile [p25] to 75th percentile [p75]) for continuous variables. When performing a 2-group comparison for BZD-responsive vs BZD-resistant and algorithm implementation vs nonimplementation cases, we used Mann-Whitney U test for continuous variables and applied Fischer exact test to categorical variables. A p value of less than 0.05 was considered significant. We excluded the missing data from the data analysis.
Standard Protocol Approvals, Registrations, and Patient Consents
The BCH IRB approved this study with approval number IRB-P00017871. Informed consent was obtained from all participants, and written consent was provided by pediatric patients' parents or legal guardians. The study was registered on ClinicalTrials.gov with identifier NCT02995759.
Data Availability
The data supporting this study's findings are not publicly available because of confidentiality, ethical restrictions, and privacy restrictions. The data are available upon reasonable request from the corresponding author and with BCH's Institutional Review Board (IRB) permission. Access to the data will be granted in accordance with the IRB guidelines and ethical requirements for data sharing.
Results
Demographics and Seizure Phenotyping
We prospectively enrolled 60 patients (48% female) with a median (p25–p75) age of 3.7 (1.9–7.0) years. The proportion of patients with preexisting epilepsy was 65% (n = 39). At episode onset, patients took a median (p25-p75) of 2 (1–3) ASMs. Seventy-seven percent (n = 30) had a prior episode of SE. RM was prescribed before the episode in 87% (n = 34/39) of patients with epilepsy (Table 1). Approximately half of the seizure onsets occurred outside the BCH hospital setting (pre-BCH hospital; 53.3%, n = 32), and the majority were focal in semiology (60%, n = 36). The most common etiologies were genetic and structural (18.3%, n = 11), structural (20%, n = 12), and febrile (18.3%, n = 11) SE (Figure 1).
Table 1.
Patient Characteristics and Time to Treatment
| Patient characteristics | CSE (n = 60) |
| Age at onset (y), median (p25-p75) | 3.7 (1.9–7.0) |
| Female, n (%) | 29 (48) |
| Race, n (%) | |
| White | 30 (50) |
| Black or African American | 5 (8) |
| Asian | 3 (5) |
| Arabic | 9 (15) |
| Other | 9 (15) |
| Not reported | 4 (7) |
| Ethnicity, n (%) | |
| Hispanic | 8 (13) |
| Non-Hispanic | 52 (87) |
| Neurologically healthy other than seizures, n (%) | 20 (33) |
| History of epilepsy, n (%) | 39 (65) |
| Family history of seizures, n (%) | 24 (40) |
| History of developmental delay, n (%) | 36 (60) |
| Number of ASMs at episode onset, median (p25-p75) | 2 (1–3) |
| History of SE in patients with a history of seizures (n = 39), n (%) | 30 (77) |
| RM prescribed before episode in patients with a history of seizures (n = 39), n (%) | 34 (87) |
| Time to treatment and personalized SAP | |||
| Personalized SAP Available (n = 34) |
Personalized SAP Not available (n = 26) |
p Values | |
| Time to first ASM (min), median (p25-p75) | 6 (5–18) n = 34 | 15 (8–27) n = 24 | <0.05 |
| Time to second ASM (minutes), median (p25-p75) | 14 (8–34) n = 20 | 31 (20–64) n = 15 | <0.05 |
| Time to third ASM (min), median (p25-p75) | 32 (21–69) n = 13 | 48 (37–85) n = 12 | 0.082 |
| Time to treatment and seizure duration | |||
| Pearson correlations with total seizure duration (minutes) (n = 60) |
p Values | ||
| Time to first ASM (min), median (p25-p75) | 9 (5–20) n = 58 | 0.878a | <0.001 |
| Time to s ASM (minutes), median (p25-p75) | 24 (10–37) n = 35 | 0.902a | <0.001 |
| Time to third ASM (min), median (p25-p75) | 39 (26–77) n = 25 | 0.931a | <0.001 |
| BZD-responsive vs BZD-resistant groups | |||
| BZD-responsive (n = 34) | BZD-resistant (n = 26) | p Values | |
| Seizure duration (min), median (p25–p75) | 11 (7–20) | 57 (29–91) | <0.001b |
| Number of ASMs needed for seizure abortion, median (p25–p75) | 1 (1–2) | 4 (3–6) | <0.001b |
| Number of non-BZD medications, median (p25–p75) | 3 (3–5) | ||
| Time to first ASM (min), median (p25–p75) | 6 (5–15), n = 33 | 15 (6–32), n = 25 | <0.05b |
| Time to s ASM (min), median (p25–p75) | 10 (8–21), n = 10 | 31 (17–55), n = 25 | <0.01b |
| Time to third ASM (min), median (p25–p75) | 36, n = 1 | 41 (25–78), n = 24 | 0.781b |
| Time to first non-BZD (min), median (p25–p75) | 57 (31–86), n = 26 | ||
Abbreviations: ASM = antiseizure medication(s); BZD = benzodiazepine; CSE = convulsive status epilepticus; SE = status epilepticus.
Correlation is significant at the 0.01 level (2-tailed).
Mann-Whitney U test.
Figure 1. Distribution of Seizure Onset Location, Characteristics, and Etiology in Pediatric Convulsive Status Epilepticus.
(A) Distribution of seizure onset locations. Most of the patients had their onset in the pre-BCH hospital setting (home, school, or other public non–health care location). (B) Seizure semiology. Focal seizures were the most common in this cohort. (C) Seizure etiology. Genetic + structural, structural, and febrile SE were the most common etiologies; those indicated as genetic + structural include structural brain diseases that are known or strongly suspected to have an underlying genetic etiology (e.g., tuberous sclerosis complex, Sturge Weber syndrome) or have an established pathogenic variant with another structural malformation (e.g., SCN1A and hippocampal sclerosis). BCH = Boston Children's Hospital; ED = emergency department; SE = status epilepticus.
Time to Treatment and Seizure Duration
Considering the entire cohort, first, second, and third ASMs were given at a median (p25–p75) of 9 (5–20, n = 58), 24 (10–37; n = 35), and 39 (26–77; n = 25) minutes, respectively (Table 1). Duration of CSE showed a positive correlation with time to administration of the first, second, and third ASMs, with correlation coefficients of 0.878, 0.902, and 0.931, respectively, all with p < 0.001 (Pearson correlation coefficient) (Table 1, Figure 2).
Figure 2. Correlation Between Time to Administer First, Second, and Third Antiseizure Medications and Seizure Duration.
The figure revealed a significant linear correlation between the administered time to the first, second, and third ASMs and duration of seizures. ASM = antiseizure medication.
Time to Treatment and Personalized SAP Availability
Of the total patient population, 34 patients had a personalized SAP, whereas the remaining 26 patients had access to standard protocols such as BCH or AES treatment algorithms. Patients who had access to the personalized SAP, even if it was not adhered to, experienced median (p25–p75) time to first and second ASMs of 6 (5–18; n = 34) and 14 (8–34; n = 20) minutes, respectively. By contrast, patients without access to the personal SAP had a longer median (p25–p75) time to the first and second ASMs of 15 (8–27; n = 24; p < 0.05) and 31 (20–64; n = 15; p < 0.05) minutes , respectively (Table 1).
BZD-Resistant vs BZD-Responsive Patients
The median (p25–p75) time to administration of first, second, and third ASMs was 6 (5–15; n = 33), 10 (8–21; n = 10), and 36 (n = 1) minutes in BZD-responsive patients vs 15 (6–32; n = 25; p < 0.05), 31 (17–55; n = 25; p < 0.01), and 41 (25–78; n = 24; p = 0.781) minutes, respectively, in BZD-resistant patients. The median (p25–p75) time until administering the first non-BZD medication in BZD-resistant patients was 57 minutes (31–86; n = 26). Seizure duration was 11 minutes (7–20; n = 34) in BZD-responsive patients and 57 minutes (29–91; n = 26; p < 0.001) in BZD-resistant patients. The median (p25–p75) number of ASMs administered was 4 (3–6; n = 26) in BZD-resistant patients, which was more than the median (p25–p75) number of BZDs needed (1 [1–2; n = 34; p < 0.001]) for seizure abortion in BZD-responsive patients (Table 1).
Algorithm Implementation
We compared the implementation of BCH, AES, and personalized SAP in BZD-responsive vs BZD-resistant subgroups. Patients with BZD-resistant CSE were less likely to follow treatment protocols when compared with BZD-responsive patients. Specifically, implementation rates were 0% (n = 0/26) vs 41% (n = 14/34; p < 0.001) for the BCH protocol, 15% (n = 4/26) vs 65% (n = 22/34; p < 0.001) for the AES protocol, and 29% (n = 4/14) vs 50% (n = 10/20; p = 0.296) for the personalized SAP, respectively (Table 2).
Table 2.
Algorithm Implementation and Deviations Leading to Non-Adherence
| Algorithm implementation | |||
| BZD-responsive (n = 34) | BZD-resistant (n = 26) | p Values | |
| BCH algorithm implementation, n (%) | 14 (41) n = 34 | 0 (0) n = 26 | <0.001a |
| AES algorithm implementation, n (%) | 22 (65) n = 34 | 4 (15) n = 26 | <0.001a |
| Personalized SAP implementation, n (%) | 10 (50), n = 20 | 4 (29), n = 14 | 0.296a |
| Deviations from algorithms leading to nonadherence (%) | ||||
| Delayed time (>25) | Incorrect dose (> ±25) | Incorrect order | >1 deviation | |
| Lack of implementation of BCH algorithm (n = 46) n (%) | 43 (93) | 25 (54) | 23 (50) | 30 (65) |
| Lack of implementation of AES algorithm (n = 34) n (%) | 18 (53) | 26 (76) | 20 (59) | 21 (62) |
| Lack of implementation of personalized SAP (n = 20) n (%) | 20 (100) | 0 (0) | 9 (45) | 9 (45) |
Abbreviations: AES = American Epilepsy Society; ASM = antiseizure medication(s); BCH = Boston Children's Hospital; BZD = benzodiazepine.
Fisher exact test.
Moreover, the median (p25-p75) seizure duration with and without personalized SAP implementation was 9 (6–16; n = 14) and 22 (11–64; n = 20; p < 0.01) minutes, respectively. Time to the first ASM (minutes) was 5 (4–5; n = 14) minutes in patients following the personalized SAP vs 16 (6–31; n = 20; p < 0.001) minutes in patients without SAP implementation. The median (p25-p75) seizure duration in BCH algorithm implementation and lack of implementation was 7 (5–10; n = 14) and 33 (17–68; n = 46; p < 0.001) minutes and similarly for AES algorithm was 10 (6–16; n = 26) and 45 (22–85; n = 34; p < 0.001) minutes, respectively. Time to the first ASM in BCH algorithm implementation and nonimplementation patients was 5 (5–5; n = 14) and 15 (6–28; n = 44; p < 0.001) minutes, respectively. For the AES algorithm, the corresponding times were 5 (5–9; n = 25) and 17 (7–37; n = 33; p < 0.001) minutes (Table 3).
Table 3.
Clinical Outcomes Based on Algorithm and SAP Implementation
| Variables | Personalized SAP implementation (n = 14) | Personalized SAP nonimplementation (n = 20) |
p Value | BCH algorithm implementation (n = 14) |
BCH algorithm nonimplementation (n = 46) |
p Value | AES algorithm implementation (n = 26) | AES algorithm nonimplementation (n = 34) |
p Value |
| Total seizure duration, median (p25–p75) (min) | 9 (6–16) n = 14 | 22 (11–64) n = 20 | <0.01a | 7 (5–10) n = 14 | 33 (17–68) n = 46 | <0.001a | 10 (6–16) n = 26 | 45 (22–85) n = 34 | <0.001a |
| Time to first ASM, median (p25-p75) (min) | 5 (4–5) n = 14 | 16 (6–31) n = 20 | <0.001a | 5 (5-5) n = 14 | 15 (6–28) n = 44 | <0.001a | 5 (5–9) n = 25 | 17 (7–37) n = 33 | <0.001a |
| Time to s ASM, median (p25–p75) (min) | 10 (8–19) n = 5 | 17 (9–35) n = 15 | 0.203a | 8 (7–9) n = 3 | 25 (11–49) n = 32 | <0.05a | 9 (8–15) n = 8 | 31 (17–55) n = 27 | <0.05a |
| Time to third ASM, median (p25–p75) (min) | 24 (13–31) n = 4 | 39 (23–77) n = 9 | 0.105a | n = 0 | 39 (26–77) n = 25 | — | 29 (22–42) n = 4 | 42 (30–81) n = 21 | 0.159a |
| ICU admission ratec, n (%) | 6 (46) n = 13 | 12 (71) n = 17 | 0.264b | 5 (36) n = 14 | 27 (66) n = 41 | 0.064b | 10 (43), n = 23 | 22 (69), n = 3 | 0.096b |
| ICU LOS median (p25–p75) (d) | 3 (2–8) n = 6 | 3 (1–5) n = 12 | 0.538a | 2 (1–4) n = 5 | 2 (1–5) n = 27 | 0.752a | 1 (1–4) n = 10 | 3 (1–6) n = 22 | 0.208a |
| Intubation rated, n (%) | 2 (14) n = 14 | 10 (53) n = 19 | <0.05b | 1 (7) n = 14 | 20 (45) n = 44 | <0.05b | 4 (17), n = 24 | 17 (50) n = 34 | <0.05b |
Abbreviations: ASM = antiseizure medication; ICU = intensive care unit; LOS = length of stay.
Mann-Whitney U test.
Fisher exact test.
ICU admission because of CSE.
Intubation because of CSE.
We evaluated selected outcome measures in patients related to BCH, AES, and personalized SAP implementation vs nonimplementation. The intubation rate was 7% (n = 1/14) in the BCH algorithm implementation group and 45% (n = 20/44; p < 0.05) in the nonimplementation group, and for AES, it was 17% (n = 4/24) and 50% (n = 17/34; p < 0.05), respectively. In the personalized SAP implementation group, intubation rates were lower than in the nonimplementation group (14%; n = 2/14 vs 53%; n = 10/19; p < 0.05) (Table 3).
Deviations From Algorithm Implementation
In assessing the nonimplementation of the BCH algorithm, we found that the recommended timing for administering ASM was delayed in 43 of 46 nonimplementation patients, accounting for 93% of the cases. In addition, 54% (25/46) of patients received an inappropriate dose. ASMs were not administered in the suggested order for 50% (23/46) of patients. Based on the AES and personalized SAP, the proposed timing of ASM administration was not followed in 53% (18/34) and 100% (20/20) patients, respectively. Based on the AES algorithm, ASMs were not administered in the suggested dose and order in 76% (26/34) and 59% (20/34) patients, respectively. According to the personalized SAP, all patients stayed within the recommended dose, and 45% (9/20) did not follow the suggested ASM order (Table 2).
Moreover, a retrospective evaluation of outpatient care gaps revealed that, of 60 patients with CSE, 32 (53%) had a prehospital CSE onset, of whom 18 (56%) had a history of seizures. Among those with prehospital seizure onset and a history of seizures, 10 (55%) received RM from family members or school personnel before arrival, 3 (17%) received RM from EMS, and 5 (28%) received RM upon arrival at the hospital or another care facility. Our analysis through the chart review identified 3 primary barriers to RM administration by caregivers: (1) caregivers failed to recognize the seizure onset (12.5%; 1/8); (2) RM was unavailable (37.5%; 3/8); and (3) RM was not administered when patients either called 911 or were driven to the hospital (50%; 4/8).
Discussion
While previous studies have established the challenges and risks associated with CSE, our study uniquely highlights the effectiveness of SAPs and standardized treatment algorithms in addressing these challenges, specifically in reducing treatment delays, shortening seizure duration, and improving seizure outcomes. In our study, deviation from standardized treatment algorithms was associated with BZD resistance; compared with young BZD-responsive patients, these patients experienced longer seizures that required more aggressive treatment of seizure cessation. The implementation of personalized SAPs and standardized treatment algorithms was associated with lower intubation rates. Future studies should aim to assess whether personalized SAPs and standardized treatment algorithms improve outcome metrics in a larger population and compare the effectiveness of standardized vs personalized treatment algorithms.
SAPs are legal mandates in 23 states, and we aimed to provide more definitive insights into the effectiveness of these interventions.29 Previous studies suggested that SAPs do not clearly decrease health care utilization.25 While an initial analysis suggests multiple barriers to implementing improved SE care, an individualized analysis of patient-specific challenges and an adjustable care plan that considers individual aspects may improve care, particularly in patients with a high risk of seizure clusters or SE.32 Potential options for improved care in selected patients include individualized SAPs and care process maps,32 among other tools that may facilitate care improvements. Although SAPs are primarily designed for prehospital management, they may also provide essential guidance to medical teams during the initial treatment phases in the ED and inpatient settings. This study investigated the impact of personalized SAP on the time to treatment in young patients with CSE. Patients with available personalized SAPs tend to receive more timely treatment of seizures, regardless of whether they adhere to the personalized SAP. This suggests that the caregivers have been effectively trained in accordance with the personalized algorithm. Furthermore, patients who had implemented personalized SAPs experienced a shorter time to first and second ASM administration and a lower rate of intubation than those without the implementation of personalized SAPs. In our study, SAP implementation did not significantly affect ICU admission or ICU LOS, potentially confounded by the small sample size.
Based on our findings, SAPs may need to incorporate clear guidance on the timing and dosing of seizure interventions, particularly for administering first-line ASM. Furthermore, SAPs may need to be tailored to the individual needs and seizure patterns of each patient, including those with BZD-resistant or refractory seizures.
Our data suggest a linear correlation between the time to treatment and seizure duration. Furthermore, BZD resistance may be associated with delays in treatment and longer seizure duration, resulting in a need for more ASMs to abort SE compared with BZD-responsive patients. These results match other investigations showing that the time to treatment and duration of SE are associated.7,10,17 Treating seizures sooner is supported by changes in excitatory and inhibitory mechanisms related to SE leading to BZD resistance and prolonged seizures33,34 and brain damage occurring in animals following prolonged seizures.35 In addition, data from the pSERG (pediatric SE Research Group) cohort demonstrated that among patients treated within 10 minutes, no deaths occurred, and in patients treated after 10 minutes, up to 5% of patients died.8 Therefore, earlier treatment and appropriately dosed treatment15 may reduce SE duration, improve outcomes, and save lives.
Other studies also show that compliance with standardized treatment guidelines leads to faster seizure cessation and better seizure control.36 However, there is a lack of implementation of standard treatment, with notable deviations, including delays in initiating treatment and inappropriate dosing of BZDs.11,15,37 In our study, we found that while there is a significant delay in administering the first BZD in BZD-resistant patients, subsequent delays in administering additional ASMs and the first non-BZD medication likely contribute to further nonadherence to treatment protocols. The extended time required to administer multiple ASMs, combined with the significant differences in seizure duration, suggests that BZD-resistant patients not only face delayed treatment but also require more aggressive interventions, which may further contribute to deviations from the established protocol. Several factors may contribute to the lower implementation rates in the BZD-resistant subgroup. One possible explanation is the complex and refractory nature of BZD-resistant CSE, which may require more specialized and individualized treatment approaches.38 While previous studies highlighted the relationship between suboptimal treatment and outcomes,7,10 our study adds to the literature that implementing algorithms is related to shorter times to treatment. These findings emphasize the importance of algorithm implementation in optimizing seizure control and improving patient outcomes, particularly in the context of the BZD-resistant group. Furthermore, our study highlights particular areas for improvement in algorithm adherence, including improving time to appropriately dosed ASMs administered in a suggested order, the availability of a SAP and RM, and the timely identification of seizure onset in the prehospital setting.
The sample size in each group is relatively small, which may limit the ability to detect significant differences. Using a specific medical setting can cause selection bias, limiting the generalizability of the findings. Missing data were excluded from the analysis and were presumed to be missing randomly; hence, the results were less likely to be skewed. Information bias may have occurred due to supplementing clinical information from electronic health records, and recall bias may have affected family data acquisition. To mitigate these biases, we obtained data from various sources for verification, including families, EMS records, and health care professionals' documentation during prospective enrollment. Future studies could consider wearables and passive data acquisition to document the time from seizure onset to treatment. In addition, we acknowledge the importance of conducting follow-up studies to evaluate potential barriers to timely implementation of ASMs in inpatient settings.
Further research, including larger sample sizes and prospective designs, may provide more robust evidence regarding the effect of standardized vs personalized algorithm adherence on ICU admission rates and LOS. Other outcome measures and long-term effects of algorithm adherence or follow-up data may add to current knowledge.
Evidence related to treatment algorithm provision and SAPs is limited. However, SAPs are now legislatively mandated in many states. Our study provides the postguideline implementation evaluation, suggesting that personalized SAPs and standardized medication algorithms may contribute to shorter time to treatment and seizure duration and lower intubation rates.
Author Contributions
C.M. Stredny: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; study concept or design; analysis or interpretation of data. S. Rostamian: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; study concept or design; analysis or interpretation of data. T.A. Sheehan: major role in the acquisition of data; study concept or design. M. Gaínza-Lein: major role in the acquisition of data; study concept or design. S. Carey: major role in the acquisition of data; study concept or design. S. Lewis: major role in the acquisition of data. J. Clark: major role in the acquisition of data; study concept or design. B. Bucciarelli: major role in the acquisition of data; study concept or design. M. Chiujdea: major role in the acquisition of data; study concept or design. A. Antonetty: major role in the acquisition of data; study concept or design. B. Zhang: drafting/revision of the manuscript for content, including medical writing for content; analysis or interpretation of data. L.F. Atunes Ortega: major role in the acquisition of data. L. Voke: major role in the acquisition of data. T. Loddenkemper: drafting/revision of the manuscript for content, including medical writing for content; major role in the acquisition of data; study concept or design; analysis or interpretation of data.
Study Funding
This study was supported by the Epilepsy Research Fund and Fred Lovejoy Research and Education Fund.
Disclosure
T. Loddenkemper receives research support from NIH, Epitel, and MIKU. He received past research support from Upsher Smith, Proximagen, and UCB. He is part of patent applications to detect and predict clinical outcomes, and to detect, manage, diagnose, and treat neurological conditions, epilepsy, and seizures. Some of T. Loddenkemper's trainees received salary support from international foundations/societies and academic centers while working in his laboratory; C. M. Stredny receives grant support from the Pediatric Epilepsy Research Foundation. She is an unpaid member of the medical and scientific advisory boards of the Autoimmune Encephalitis Alliance and NORSE Institute; S. Rostamian, T. Sheehan, Gaínza-Lein, S. Carey, S. Lewis, J. Clark, B. Bucciarelli, M. Chiujdea, A. Antonetty, B. Zhang, L. F. Atunes Ortega and L. Voke report no disclosures relevant to the manuscript. Full disclosure form information provided by the authors is available with the full text of this article at Neurology.org/cp.
References
- 1.Wu YW, Shek DW, Garcia PA, Zhao S, Johnston SC. Incidence and mortality of generalized convulsive status epilepticus in California. Neurology. 2002;58(7):1070-1076. doi: 10.1212/wnl.58.7.1070 [DOI] [PubMed] [Google Scholar]
- 2.Coeytaux A, Jallon P, Galobardes B, Morabia A. Incidence of status epilepticus in French-speaking Switzerland: (EPISTAR). Neurology. 2000;55(5):693-697. doi: 10.1212/wnl.55.5.693 [DOI] [PubMed] [Google Scholar]
- 3.Chin RFM, Neville BGR, Peckham C, et al. Incidence, cause, and short-term outcome of convulsive status epilepticus in childhood: prospective population-based study. Lancet. 2006;368(9531):222-229. doi: 10.1016/S0140-6736(06)69043-0 [DOI] [PubMed] [Google Scholar]
- 4.Chin RFM, Neville BGR, Scott RC. A systematic review of the epidemiology of status epilepticus. Eur J Neurol. 2004;11(12):800-810. doi: 10.1111/j.1468-1331.2004.00943.x [DOI] [PubMed] [Google Scholar]
- 5.Raspall-Chaure M, Chin RFM, Neville BG, Scott RC. Outcome of paediatric convulsive status epilepticus: a systematic review. Lancet Neurol. 2006;5(9):769-779. doi: 10.1016/S1474-4422(06)70546-4 [DOI] [PubMed] [Google Scholar]
- 6.Maytal J, Shinnar S, Moshé SL, Alvarez LA. Low morbidity and mortality of status epilepticus in children. Pediatrics. 1989;83(3):323-331. doi: 10.1542/peds.83.3.323 [DOI] [PubMed] [Google Scholar]
- 7.Logroscino G, Hesdorffer DC, Cascino GD, Annegers JF, Bagiella E, Hauser WA. Long-term mortality after a first episode of status epilepticus. Neurology. 2002;58(4):537-541. doi: 10.1212/wnl.58.4.537 [DOI] [PubMed] [Google Scholar]
- 8.Gaínza-Lein M, Sánchez Fernández I, Jackson M, et al. Association of time to treatment with short-term outcomes for pediatric patients with refractory convulsive status epilepticus. JAMA Neurol. 2018;75(4):410-418. doi: 10.1001/jamaneurol.2017.4382 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Abend NS, Loddenkemper T. Pediatric status epilepticus management. Curr Opin Pediatr. 2014;26(6):668-674. doi: 10.1097/MOP.0000000000000154 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Brophy GM, Bell R, Claassen J, et al. Guidelines for the evaluation and management of status epilepticus. Neurocrit Care. 2012;17(1):3-23. doi: 10.1007/s12028-012-9695-z [DOI] [PubMed] [Google Scholar]
- 11.Sánchez Fernández I, Abend NS, Agadi S, et al. Time from convulsive status epilepticus onset to anticonvulsant administration in children. Neurology. 2015;84(23):2304-2311. doi: 10.1212/WNL.0000000000001673 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Pellock JM, Marmarou A, DeLorenzo R. Time to treatment in prolonged seizure episodes. Epilepsy Behav. 2004;5(2):192-196. doi: 10.1016/j.yebeh.2003.12.012 [DOI] [PubMed] [Google Scholar]
- 13.Lewena S, Pennington V, Acworth J, et al. Emergency management of pediatric convulsive status epilepticus: a multicenter study of 542 patients. Pediatr Emerg Care. 2009;25(2):83-87. doi: 10.1097/PEC.0b013e318196ea6e [DOI] [PubMed] [Google Scholar]
- 14.Seinfeld S, Shinnar S, Sun S, et al. Emergency management of febrile status epilepticus: results of the FEBSTAT study. Epilepsia. 2014;55(3):388-395. doi: 10.1111/epi.12526 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Vasquez A, Gaínza-Lein M, Abend NS, et al. First-line medication dosing in pediatric refractory status epilepticus. Neurology. 2020;95(19):e2683-e2696. doi: 10.1212/WNL.0000000000010828 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Sathe AG, Underwood E, Coles LD, et al. Patterns of benzodiazepine underdosing in the established status epilepticus treatment trial. Epilepsia. 2021;62(3):795-806. doi: 10.1111/epi.16825 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Chin RFM, Neville BGR, Peckham C, Wade A, Bedford H, Scott RC. Treatment of community-onset, childhood convulsive status epilepticus: a prospective, population-based study. Lancet Neurol. 2008;7(8):696-703. doi: 10.1016/S1474-4422(08)70141-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Alldredge BK, Wall DB, Ferriero DM. Effect of prehospital treatment on the outcome of status epilepticus in children. Pediatr Neurol. 1995;12(3):213-216. doi: 10.1016/0887-8994(95)00044-g [DOI] [PubMed] [Google Scholar]
- 19.Eriksson K, Metsäranta P, Huhtala H, Auvinen A, Kuusela AL, Koivikko M. Treatment delay and the risk of prolonged status epilepticus. Neurology. 2005;65(8):1316-1318. doi: 10.1212/01.wnl.0000180959.31355.92 [DOI] [PubMed] [Google Scholar]
- 20.Goodkin HP, Yeh JL, Kapur J. Status epilepticus increases the intracellular accumulation of GABAA receptors. J Neurosci. 2005;25(23):5511-5520. doi: 10.1523/JNEUROSCI.0900-05.2005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Maegaki Y, Kurozawa Y, Tamasaki A, et al. Early predictors of status epilepticus-associated mortality and morbidity in children. Brain Dev. 2015;37(5):478-486. doi: 10.1016/j.braindev.2014.08.004 [DOI] [PubMed] [Google Scholar]
- 22.Naylor DE, Liu H, Wasterlain CG. Trafficking of GABA(A) receptors, loss of inhibition, and a mechanism for pharmacoresistance in status epilepticus. J Neurosci. 2005;25(34):7724-7733. doi: 10.1523/JNEUROSCI.4944-04.2005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Aliasgharpour M, Dehgahn Nayeri N, Yadegary MA, Haghani H. Effects of an educational program on self-management in patients with epilepsy. Seizure. 2013;22(1):48-52. doi: 10.1016/j.seizure.2012.10.005 [DOI] [PubMed] [Google Scholar]
- 24.Besag FMC. The department of health action plan “improving services for people with epilepsy”: a significant advance or only a first step? Seizure. 2004;13(8):553-564. doi: 10.1016/j.seizure.2004.01.005 [DOI] [PubMed] [Google Scholar]
- 25.Roundy LM, Filloux FM, Kerr L, Rimer A, Bonkowsky JL. Seizure action plans do not reduce health care utilization in pediatric epilepsy patients. J Child Neurol. 2016;31(4):433-438. doi: 10.1177/0883073815597755 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Seizure Action Plans. Epilepsy foundation. Accessed May 9, 2024. epilepsy.com/preparedness-safety/action-plans
- 27.What is a SAP? Seizure action plans. Accessed May 9, 2024. seizureactionplans.org/seizure-action-plans/
- 28.Albert DVF, Moreland JJ, Salvator A, et al. Seizure action plans for pediatric patients with epilepsy: a randomized controlled trial. J Child Neurol. 2019;34(11):666-673. doi: 10.1177/0883073819846810 [DOI] [PubMed] [Google Scholar]
- 29.Advocacy: Seizure Safe Schools. Epilepsy foundation. Accessed April 9, 2024. epilepsy.com/advocacy/priorities/seizure-safe-schools
- 30.Glauser T, Shinnar S, Gloss D, et al. Evidence-based guideline: treatment of convulsive status epilepticus in children and adults: report of the guideline committee of the American epilepsy society. Epilepsy Curr. 2016;16(1):48-61. doi: 10.5698/1535-7597-16.1.48 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Guideline for Treatment of Prolonged Seizures in Children and Adults. 2016. Accessed October 4, 2024. aesnet.org/clinical-care/clinical-guidance/guideline-prolonged-seizures [Google Scholar]
- 32.Jackson MC, Vasquez A, Ojo O, et al. Identifying barriers to care in the pediatric acute seizure care pathway. Int J Integr Care. 2022;22(1):28. doi: 10.5334/ijic.5598 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Kapur J. Role of NMDA receptors in the pathophysiology and treatment of status epilepticus. Epilepsia Open. 2018;3(suppl suppl 2):165-168. doi: 10.1002/epi4.12270 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Burman RJ, Rosch RE, Wilmshurst JM, et al. Why won't it stop? The dynamics of benzodiazepine resistance in status epilepticus. Nat Rev Neurol. 2022;18(7):428-441. doi: 10.1038/s41582-022-00664-3 [DOI] [PubMed] [Google Scholar]
- 35.Lothman E. The biochemical basis and pathophysiology of status epilepticus. Neurology. 1990;40(5 suppl 2):13-23. [PubMed] [Google Scholar]
- 36.Aranda A, Foucart G, Ducassé JL, Grolleau S, McGonigal A, Valton L. Generalized convulsive status epilepticus management in adults: a cohort study with evaluation of professional practice. Epilepsia. 2010;51(10):2159-2167. doi: 10.1111/j.1528-1167.2010.02688.x [DOI] [PubMed] [Google Scholar]
- 37.Uppal P, Cardamone M, Lawson JA. Outcomes of deviation from treatment guidelines in status epilepticus: a systematic review. Seizure. 2018;58:147-153. doi: 10.1016/j.seizure.2018.04.005 [DOI] [PubMed] [Google Scholar]
- 38.Rai S, Drislane FW. Treatment of refractory and super-refractory status epilepticus. Neurotherapeutics. 2018;15(3):697-712. doi: 10.1007/s13311-018-0640-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data supporting this study's findings are not publicly available because of confidentiality, ethical restrictions, and privacy restrictions. The data are available upon reasonable request from the corresponding author and with BCH's Institutional Review Board (IRB) permission. Access to the data will be granted in accordance with the IRB guidelines and ethical requirements for data sharing.