If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Sydney Children's Hospital, Department of Pediatric Neurology, High Street, Randwick, NSW, 2031, AustraliaUniversity of New South Wales – Randwick Campus, School of Women's and Children's Health, High Street, Randwick, NSW, 2052, Australia
Sydney Children's Hospital, Department of Pediatric Neurology, High Street, Randwick, NSW, 2031, AustraliaUniversity of New South Wales – Randwick Campus, School of Women's and Children's Health, High Street, Randwick, NSW, 2052, Australia
Sydney Children's Hospital, Department of Pediatric Neurology, High Street, Randwick, NSW, 2031, AustraliaUniversity of New South Wales – Randwick Campus, School of Women's and Children's Health, High Street, Randwick, NSW, 2052, Australia
There is a gap between guidelines on status epilepticus and its management.
•
The proportion of variations ranged from 10.7% to 66%.
•
The most common variation was excessive benzodiazepine use.
•
5.79 times higher odds of respiratory depression with excessive benzodiazepine use.
Abstract
Objective
Due to a gap between published clinical guidelines on status epilepticus SE and clinician management of SE, a systematic review was performed to investigate treatment adherence to SE guidelines and its impact on patient outcomes.
Methods
Medline and Embase searches were conducted for studies appraising adherence to SE guidelines (from 1970 and 1st April 2018). The quality of eligible studies was assessed by QUADAS- 2 criteria. Comparison was made between patients where guidelines were followed and not followed. Various patient outcomes including intubation, ICU admission, morbidity and mortality were compared. A Forest plot was used to investigate the effect of adherence on outcome.
Results
A total of 3424 titles and abstracts were screened from the initial search after removal of duplicates. A total of 441 full text articles were reviewed in detail, and 22 articles were included in this study. The proportion of deviations ranged from 10.7% to 66.1%. Four studies were descriptive. Eighteen studies looked at the adverse effects of non-adherence. Eight studies showed respiratory depression and intubation were associated with excessive benzodiazepine use. A subset analysis showed 5.79 times higher odds of respiratory depression and intubation], if excessive benzodiazepines were given. The next most common variations were delayed management and insufficient treatment. These variations from the guidelines were associated with prolonged seizures.
Conclusions
This review provides preliminary evidence that non-adherence to SE guidelines negatively impacts on patient outcomes. Appropriate and timely treatment is imperative for rapid seizure termination and improving outcomes.
]. Guidelines for the management of SE have been available for many years, yet a number of retrospective studies have confirmed that clinicians often do not follow recommended guidelines on the management of patients with SE [
] and these studies have given conflicting results. However, adherence to guidelines has not been subject to systematic review previously. There has recently been a review published looking at the state of epilepsy practice guidelines [
This study aims to conduct a systematic review investigating adherence to SE clinical practice guidelines and its impact on patient morbidity including respiratory depression and ICU admission.
2. Methods
2.1 Search strategy
This systematic review was performed according to the PRISMA guidelines [
] (Fig. 1). Medline and EMBASE databases were searched for articles published between 1 January 1970 and 1 April 2018, investigating the adherence to SE treatment guidelines.
Fig. 1Flowchart outlining searches strategy for the systematic review (The formatted figure is attached separately as another file).
The following text and MeSH terms were used in our search: (seizures/or status epilepticus and (guideline adherence/or guideline/or clinical protocols or practice guideline) and (deviation from guidelines.mp or benzodiazepines/ad,ae,to or anticonvulsant/ad,ae,to) and (intubation or ICU admission or length of stay or morbidity or mortality or treatment outcome). See detailed search strategy, e-1. The language was limited to English. Only human studies were included.
The review question was broken down to help guide the development of search terms using the PICO structure. Possible synonyms and abbreviations for the terms of interest and subject headings (including MeSH headers) were included.
2.3 Inclusion criteria
1)
All patients: with SE, seizure duration > 5 min
2)
Adult and paediatric studies (age >28 days)
2.4 Exclusion criteria
1)
Personal views and letters to editors
2)
Studies with incomplete data
3)
Conference abstracts
The Medline and Embase searches identified 3773 articles (Fig. 1). Bibliographic references of the studies were also searched. Scopus was hand searched to find relevant and highly cited papers in this area. We hand searched the first authors of included papers to find articles that may not have been appropriately indexed but were relevant to the questions. A total of 67 additional records identified through these searches.
The authors have not included unpublished abstracts and other grey literature as the data is often incomplete and non-peer reviewed.
2.5 Screening
A total of 3424 records were screened after removal of 416 duplicates (Fig. 1). We screened titles and abstracts and excluded all irrelevant publications. If the paper was deemed relevant by any of the authors or the relevance was not clear from the title and abstract, then the full text was retrieved.
2.6 Eligibility
A total of 441 full-text articles were reviewed and 419 articles excluded; of these 233 were comparisons between medications and/or compared various routes of administration; 83 were opinions, review articles and letters to the editor; 77 were guidelines or comments on guidelines; 14 articles had duplicate or incomplete data; 10 were expert commentaries and 2 were physician perspectives.
2.7 Study quality
For quality assessment, a template based on Quality Assessment of Diagnostic Accuracy Studies tool (QUADAS-2) [
The search strategy identified 22 articles, which were included in the systematic review. Included studies were appraised for their risk of bias and applicability (Fig. 2a & b). The quality assessment was used for evaluative purposes. No studies were excluded based on quality assessment.
Fig. 2a Proportion of studies with risk of bias (%) as assessed by QUADAS-2. b Proportion of studies with low, high or unclear concerns regarding applicability (%) as assessed by QUADAS-2.
]. Eighteen studies also looked at the effect of non-adherence on various patient outcomes and morbidity including respiratory depression, intubation and ICU admission [
]. This review did not investigate individual drug comparisons and nor did we look at impact of the route of administration.
3.2 Characteristics of the target population and studies
The definition of SE varied between studies. Eleven of the 22 included studies have used the classical definition of seizure duration >30 min (Table 1). Five out of the 22 studies have used the operational definition of SE to include >5 min of seizure activity [
The SE clinical practice guidelines followed by all the various studies used similar first and second line anticonvulsant recommendations. The first line agent was always a benzodiazepine (BZD) e.g. lorazepam or midazolam and the second line was a long acting anticonvulsant e.g. phenytoin. However, there were slight variations in the guidelines. In two studies, levetiracetam [
Fig. 3Forest plot – the odds ratios (95%CI), of adverse outcomes due to non-adherence to treatment guidelines. With non-adherence, these results suggest there are higher odds of adverse outcome for most deviations from guidelines. For excessive benzodiazepines in particular, the pooled OR shows much higher odds for adverse outcomes (OR = 5.79).
]. In a recent multicentre, prospective study of 218 patients in the United States 66.1% patients received the first line benzodiazepines 10 min or more after the seizure onset [
] reported a non-adherence rate of 53.2% and the odds of ICU admission were 3.1 times higher if there was non-adherence to the protocol (Fig. 3). A Swiss study of 263 adults reported that adherence to treatment guidelines had no effect on morbidity, mortality or likelihood of seizure control [
The data from studies was summarized in the form of a forest box-plot (see Fig. 3) where odds ratios could be calculated. These results suggest there are higher odds of adverse outcome for most deviations from guidelines. Worse outcomes were seen in both paediatric [
] reported excessive BZD use. For excessive benzodiazepines, the pooled OR (Fig. 3) shows much higher odds (OR = 5.79) for adverse outcomes including respiratory depression, intubation and ICU admission [
], showed 39.7% (31/78) received at least three doses of benzodiazepines prior to escalation to non-benzodiazepine medications. An American study done prospectively in paediatric ICU patients, found respiratory compromise was more common in patients who received a BZD dose greater than recommended guidelines [
]. A paediatric emergency department study in New York looked at type of benzodiazepine and intubation rates. This study reported that diazepam use resulted in a higher number of intubations as compared to lorazepam [
]. The New York study demonstrated 12 out of 26 patients required endotracheal intubation that the authors considered was related to excessive BZD administration [
]. It compared 173 adult patients who received standard BZD doses to those receiving excessive doses (>30% above recommended dose, n = 29) and quantitated the morbidity and mortality. The study found a statistically significant increase in the intubation rate with 45% (13/29) of the “excessive” BZD patients intubated for airway protection, versus only 8% (13/173) where the BZD dose was appropriate (p < .001) (Box plot1). A Georgian paediatric study found that 13/17 patients who received excessive BZDs required to be intubated [
A paediatric study of 47 children, found three times higher odds of excessive benzodiazepines in the children with convulsive SE admitted to intensive care, when compared to the ward group [
] when excessive BZDs were given. For excessive benzodiazepine use, the pooled OR shows much higher odds (OR = 5.79) for adverse outcomes including respiratory depression, intubation and ICU admission.
3.5 Delay in initiation of treatment of SE
Five out of 22 studies examined if there was a delay in initiation of treatment. These include delay in pre-hospital treatment, delay in arrival to emergency department, or delay in administration of first and second line antiepileptic agent. In a recent multicentre, prospective study 66.1% patients received the first line benzodiazepines 10 min or more after the seizure onset [
]. The multivariate analysis showed these patients higher odds of death after adjusting for confounders (adjusted odds ratio, 11.0), In another study, treatment was initiated after an hour following the onset of SE in 62/225 patients [
]. A delay in presentation to the emergency department was associated with a longer duration of seizure. For a delay in treatment of every minute from onset of SE, there was a 5% increase in the risk of the episode being longer lasting, in excess of 60 min [
]. A recent In a retrospective study of 45 adults, delayed management was associated with significantly longer seizure duration (median: 95 versus 38 min); compared to where management was instituted in the recommended timeframes (p ≤ 0.02) [
]. The criteria for insufficient treatment included the doses administered were too low, wrong route of administration was used, or a delay to move on to the next line of treatment. Of the 346 patients (≥15years) of SE looked at retrospectively “insufficient” treatment was found in 22.5% in patients who had negative sequelae. In patients with good outcome, insufficient treatment was seen in only 8% [
]. It was a prospective study comparing management and prognosis of SE according to the academic tertiary versus peripheral hospital setting, done in Switzerland [
This systematic review is the first to investigate the effect of clinician adherence to SE clinical practice guidelines and its impact on patients. Although there are clinical practice guidelines for SE, it is clear from this systematic review that SE clinical practice guidelines are not being followed. All of the studies in this systematic review confirmed the management of status epilepticus varied from the clinical practice guidelines. Non-adherence to SE management guidelines were associated with an increased chance of worse outcome including ICU admission [
]. The systematic review found that most studies violated the guidelines and the main protocol deviations were excessive benzodiazepine use and delay to treatment [
]. Comparisons between medications and/or various routes of administration were not examined in this systematic review. For example important randomized controlled trials such as RAMPART [
] found that IM midazolam is no different to IV lorazepam for the management of SE. Intramuscular administration can be given more rapidly than IV administration but this was not felt to be an important contributor to guideline violations and was not examined. This systematic review included both adult and paediatric age groups. Our study shows that there was deviation from guidelines seen in both paediatric studies [
]. The included studies were mostly emergency department and ICU based studies and did not look at if the diagnosis or initiation of treatment was made in consultation with a specialist neurologist. The guidelines are designed to enable early initiation of treatment and seeking speciality consultation may delay this.
This systematic review is not without limitations. This systematic review did not include unpublished data, abstracts from conferences and clinician personal communications. Low-level evidence studies have not been included in this review.
Twelve of the 22 included studies are retrospective studies [
]. Various studies have chosen different inclusion criteria for status epilepticus (Table 1) and different outcome measures (Table 2). We were unable to conduct a meta-analysis due to the heterogeneity of the included studies.
The need to understand why clinicians do not follow SE clinical practice guidelines is paramount. One of the outstanding characteristics of our current health care systems is the gap between what should be done and what is delivered [
]. This review would form the basis for quality improvement in the management of status epilepticus, to understand and improve the factors responsible for non-adherence to guidelines.
5. Conclusions
Implementation of SE guidelines may improve management. To provide maximum benefit to the patients, it is imperative that appropriate care is instituted for patients presenting with a seizure. This review highlights that deviations from guidelines exist and have been studied to some extent. The adverse effect of non-compliance and its impact on patient care has not been vigorously and systematically studied. This review provides evidence that non-adherence to SE guidelines leads to adverse outcomes. Most of the studies consistently showed adverse outcomes were related to the use of excess benzodiazepines.
Conflicts of interest
None.
Disclosure
The authors report no disclosures relevant to the manuscript.
Author contributions
Preena Uppal contributed to the study design, literature review, data analysis, data interpretation, and writing. Michael Cardamone contributed to the study design, literature review, data interpretation, and writing. John Lawson, contributed to the study design, data interpretation, and writing.
Study funding
No funding was received for this study.
Acknowledgments
The authors thank Dr. Nancy Briggs, Senior Statistical Consultant, UNSW for statistical analysis.
The authors thank Dr. Tejaswi Kandula, Paediatric neurologist for suggestions and proofreading.
The authors thank Robyn Howard for proofreading.
The authors thank Toni Le Roux & Cheng Siu from the UNSW Library for assistance with the search strategy.
References
Chin R.F.
Neville B.G.
Peckham C.
Wade A.
Bedford H.
Scott R.C.
Treatment of community-onset, childhood convulsive status epilepticus: a prospective, population-based study.
30486-7&id=gr1.jpg){kind=link}
30486-7&id=gr2.jpg){kind=link}
30486-7&id=gr3.jpg){kind=link}