- Email: [email protected]
Impact of non-guideline-based treatment of status epilepticus
Accepted Manuscript Impact of non-guideline-based treatment of status epilepticus
James Braun, Elizabeth Gau, Stacy Revelle, Laurie Byrne, Abhay Kumar PII: DOI: Reference:
S0022-510X(17)34321-6 doi:10.1016/j.jns.2017.09.031 JNS 15577
To appear in:
Journal of the Neurological Sciences
Received date: Revised date: Accepted date:
13 April 2017 29 August 2017 21 September 2017
Please cite this article as: James Braun, Elizabeth Gau, Stacy Revelle, Laurie Byrne, Abhay Kumar , Impact of non-guideline-based treatment of status epilepticus. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Jns(2017), doi:10.1016/j.jns.2017.09.031
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Impact of Non-Guideline-Based Treatment of Status Epilepticus James Braun, Pharm.D., BCCCP1; Elizabeth Gau, Pharm.D., BCCCP1; Stacy Revelle, Pharm.D., BCPS1; Laurie Byrne, M.D.2; Abhay Kumar, M.D.3 1
Department of Pharmacy Services
SSM Health Saint Louis University Hospital
IP
T
3635 Vista Ave. St. Louis, MO 63110 2
CR
Division of Emergency Medicine
US
Saint Louis University School of Medicine 3635 Vista Ave.
3
Department of Neurology
M
Saint Louis University School of Medicine
AN
St. Louis, MO 63110
ED
3635 Vista Ave. St. Louis, MO 63110
Abhay Kumar, MD
PT
Corresponding Author:
CE
Saint Louis University School of Medicine
AC
Department of Neurology
3635 Vista Ave. at Grand Blvd. Saint Louis, MO 63110 Fax: (314) 268-5166 Phone: 314-977-4830 Email: [email protected]
Abstract: 1
ACCEPTED MANUSCRIPT Impact of Non-Guideline-Based Treatment of Status Epilepticus
Background: This retrospective study analyzed benzodiazepine usage patterns in relation to
T
guideline recommendations for the treatment of generalized convulsive status
IP
epilepticus (GCSE) as practiced by emergency medical services (EMS) and the
CR
emergency department (ED) of an inner-city hospital. Secondary outcomes of interest
US
were adverse events and admission/discharge outcomes.
AN
Methods:
Records of all patients ≥18 years old diagnosed with GCSE between June 2012 and
ED
M
September 2015 and transported by EMS to our hospital ED were reviewed.
Results:
PT
Of 44 patients analyzed, 43 (98%) had a history of epilepsy. Benzodiazepine utilization
CE
varied; EMS preferred midazolam (69% of cases) while the ED utilized lorazepam (91% of cases). Benzodiazepine dosages used were lower than guideline recommendations.
AC
Seizure activity was aborted with benzodiazepines alone in 22 (50%) patients. Twelve patients (27%) experienced seizure recurrence following SE treatment and achievement of seizure cessation. Twenty-three (52%) patients required intubation after arrival to ED. All 44 patients were admitted; 30 (68%) required admission to the intensive care unit.
Conclusions:
2
ACCEPTED MANUSCRIPT There was consistent underdosing of benzodiazepines in treatment of GCSE in both EMS and ED settings likely resulting in underachievement of seizure cessation, while intubation rates were higher than reported when compared to previous studies. Prospective studies are needed to identify barriers to optimal benzodiazepine usage in
IP
T
GCSE patients.
CR
Keywords:
Status epilepticus; epilepsy; guidelines; benzodiazepines; emergency department;
AN
US
seizures
BZD = benzodiazepine ED = emergency department
ED
AES = American Epilepsy Society
M
Abbreviations:
PT
EMS = emergency medical services
CE
GCS = Glasgow Coma Score
GCSE = generalized convulsive status epilepticus
AC
ICD-9 = International Classification of Diseases, Ninth Revision ICU = intensive care unit LOS = Length of stay NCS = Neurocritical Care Society OSH = outside hospital SE = status epilepticus SZ = seizure Introduction 3
ACCEPTED MANUSCRIPT Patients in GCSE require immediate treatment, as permanent neurological damage can result with continued seizure activity and prolonged seizures are associated with significant morbidity and mortality [1-3]. Treatment of GCSE has evolved over the years with benzodiazepines being the recommended first-line
T
treatment [3-5].
IP
A study of out-of-hospital GCSE management comparing the efficacy of
CR
diazepam and lorazepam found that patients treated with lorazepam were most likely to achieve termination of seizure activity prior to hospital arrival [6]. More recently, a
US
multicenter, double-blind, randomized noninferiority trial compared the efficacy of two
AN
out-of-hospital benzodiazepine treatments provided by emergency medical services (EMS) –10 mg of intramuscular (IM) midazolam or 4 mg of intravenous (IV) lorazepam –
M
for adults presenting in SE. IM midazolam not only met noninferiority requirements, but
ED
also demonstrated superiority to IV lorazepam in terminating seizures before arrival in the ED [7].Based on these studies, recently published guidelines for management of
PT
GCSE patients recommended benzodiazepine dose optimization intended to achieve
CE
seizure cessation while minimizing the risk of significant respiratory depression [4,5]. A retrospective observational study published in 2014 characterized the prehospital and
AC
ED treatment provided to patients presenting with GCSE [8]. Investigators found almost half of the patients (n=170) received more than one dose of benzodiazepine but only 11% of patients received the optimal 0.1 mg/kg lorazepam or equivalent dose of another benzodiazepine. To characterize further the observed discrepancy between benzodiazepine dose recommendations and dosing patterns in practice, we aimed to study benzodiazepine
4
ACCEPTED MANUSCRIPT usage in the treatment of GCSE and its impact on hospitalization patterns and intubation rates at our institution.
Methods
T
Study Population
IP
This retrospective study was conducted at SSM Health Saint Louis University
CR
Hospital (Saint Louis, Missouri), a 356-bed urban, academic teaching hospital that services approximately 45,000 ED visits per year. The electronic medical records
US
(EpicCare EMR, Verona, WI) of patients ≥ 18 years who presented via EMS to the
AN
emergency department (ED) with an initial diagnosis of GCSE between June 2012 and September 2015 were evaluated for inclusion in the study. Patients with history of
M
stroke (ischemic or hemorrhagic) or known brain neoplasm were included. Patients
ED
were excluded if they received GCSE treatment at another hospital prior to transfer, if seizure activity was precipitated by acute neurological injuries (such as hemorrhage,
PT
traumatic brain injury, anoxic brain injury), hypoglycemia, alcohol withdrawal, drug
CE
overdose, cardiac arrest, or bradycardia (defined as a heart rate of < 40 bpm), if the patient had a benzodiazepine allergy, was pregnant, or incarcerated [7]. Patients were
AC
also excluded if they were deemed not to be in GCSE per clinical evaluation by neurology service at the time of their presentation. This study received approval from the Saint Louis University Institutional Review Board and Saint Louis University Hospital Research Compliance Committee; the need for informed consent was waived due to the retrospective nature of the study.
5
ACCEPTED MANUSCRIPT Definitions For the purposes of this analysis, GCSE was defined as 5 minutes or more of continuous clinical seizure activity or acute seizure activity without recovery (return to baseline) between seizures [4,5]. As this study focused on convulsive seizure activity,
T
electroencephalogram data was not utilized. Seizure recurrence was defined as any
IP
return of seizure activity within 12 hours of hospital arrival that required additional
CR
benzodiazepine administration and occurred after cessation of initial seizure activity. A descriptive report of seizure response following medication administration was utilized
US
to determine time to response, since reliable reporting of seizure cessation via timed
AN
flow chart was not always available.
M
Treatment Protocol
ED
Our institution lacked an established SE treatment protocol during the period of this study, but drug levels of home antiepileptic therapy were obtained and home
PT
therapy was resumed following emergent treatment when applicable. First-line therapy
CE
was defined as benzodiazepine treatment, whether provided by EMS or in the ED; second- and third-line therapies were defined as treatments provided subsequent to
AC
benzodiazepine administration and second-line therapies, respectively. Seizure cessation with benzodiazepines was defined as cessation of convulsive seizure activity at any time after first-line benzodiazepine administration and before treatment advancement to second-line therapies. Prehospital treatment was defined as benzodiazepine therapy provided via EMS personnel prior to ED arrival.
6
ACCEPTED MANUSCRIPT Study Outcomes The primary outcome of interest was benzodiazepine usage patterns of EMS and the ED. This encompassed benzodiazepine selection, the dose given, and route of administration. EMS medication reports were reviewed to collect the details regarding
T
benzodiazepine administrations provided prior to ED arrival. Secondary outcomes were
IP
adverse events (seizure recurrence and the proportion of patients intubated), as well as
CR
admission and discharge outcomes. Demographic characteristics were collected for all
US
patients.
AN
Statistical Analysis
Demographic data were reported via descriptive statistics. Categorical variables
M
were compared via Fischer’s exact test or Χ2 test, as appropriate. Continuous variables
ED
were analyzed using Student’s t test. Statistical analyses were performed using SPSS
PT
(IBM SPSS Statistics 24 for Windows).
CE
Results
A total of 124 patients were identified by their initial ED diagnosis of GCSE, of
AC
which 44 met the inclusion criteria (Figure 1). Of the 80 patients excluded, 29 patients received management for GCSE at other hospital EDs prior to their transfer, while 18 patients were deemed not to be in GCSE per clinical evaluation by neurology service at the time of their presentation. 124 patients were identified via ICD-9 codes, discharge diagnosis, and ED chief complaint
7
ACCEPTED MANUSCRIPT
US
CR
IP
T
80 Were excluded 29 Acute SE was managed at OSH or developed GCSE later in hospital course 18 Did not present in SE (post-ictal or did not meet GCSE criteria per neurology) 14 Duplicate patient chart 5 Missing EMS medication administration report (both dose and route) 5 Prisoners 3 Use of secondary/refractory control concomitantly with or before benzodiazepine administration 3 Did not arrive via EMS 2 Seizure precipitated by major trauma 1 Seizure precipitated by hypoglycemia
AN
44 patients treated for SE by EMS/ED personnel with benzodiazepines and included in final analysis
Figure 1: Study Design
ED
M
ICD-9 = International Classification of Diseases, Ninth Revision; ED = emergency department; SE = status epilepticus; OSH = outside hospital; GCSE = generalized convulsive status epilepticus; EMS = emergency medical services
Table 1 shows patient characteristics; median patient age was 55 years and the
PT
majority of patients had a history of epilepsy (91%). Half of these patients presented
CE
due to noncompliance with prescribed seizure treatment, while a third had a coexisting condition. Fifty percent of patients (n= 22) required initiation of 2 nd-line therapy after
AC
presenting to the ED to abort seizures. Twenty-seven percent of patients (n= 12) experienced seizure recurrence within 12 hours of arrival, mostly occurring in patients (n= 10) whose initial seizures could not be terminated with benzodiazepines alone. All 44 patients were admitted; 68% were admitted to the ICU. Median hospital length of stay (LOS) was 3 days; median ICU LOS was 1 day.
Characteristics
Total (n=44)
Seizure Cessation with Benzodiazepines? Yes No
pvalue 8
ACCEPTED MANUSCRIPT 58 (23-84)
53 (22-88)
0.332
73 (45-107)
77 (51-134)
0.041
15 (54) 20 (50)
13 (46) 20 (50)
0.755 1.000
30 (68) 13 (30) 1 (2)
15 (50) 7 (54) 0
15 (50) 6 (46) 1 (100)
21 (48) 10 (23) 13 (30) 7 (18) 4 (9) 2 (5)
8 (38) 8 (80) 6 (46) 4 (57) 1 (25) 1 (50)
8 (3-15)
6 (3-15)
22 (50)
22 (50)
M
Number of BZD doses administered per patient (EMS and ED combined), n Median (range) 2.0 (1-9) 1.5 (1-5) 3.0 (1-9) Seizure recurrence within 12 hours of arrival, n (%) Hospital admission, n (%) ICU admission, n (%) Endotracheal intubation, n (%)
0.481
T IP
13 (62) 2 (20) 7 (54) 3 (43) 3 (75) 1 (50)
AN
GCS at ED arrival, median (range)* 7 (3-15) Seizure terminated with emergent therapy, n (%)
CR
Male gender, n (%) History of seizures, n (%) Race, n (%) Black Caucasian Asian Precipitating cause of SE, n (%) Noncompliance Idiopathic/breakthrough SZ Coexisting condition Malignancy Previous stroke Other
55 (22-88) 75 (45-134) 28 (64) 40 (91)
US
Age, median (range) Weight (kg), median (range)
0.113
0.617
0.092
2 (17)
10 (83)
0.016
44 (100) 30 (68) 23 (52)
12 (40) 8 (35)
18 (60) 15 (65)
0.104 0.069
Length of hospital admission, days Mean ± SD Median (range) Mode
6.1 ± 8.8 3 (0-50) 2
2.82 ± 3.10 2.0 (0-15) 1
9.41 ± 11.30 5.0 (1-50) 2
0.013
Length of ICU admission, days Mean ± SD Median (range) Mode
4.0 ± 6.4 1 (0-26) 1
1.09 ± 2.36 < 1 (0-11) <1
4.95 ± 7.38 1.0 (0-26) <1
0.026
AC
CE
PT
ED
12 (27)
Table 1: Baseline Characteristics and Outcomes SE = status epilepticus; GCS = Glasgow coma score; EMS = emergency medical services; ED = emergency department; SZ = seizure; ICU = intensive care unit; BZD = benzodiazepine; SD = standard deviation * GCS missing, n=3
One hundred and eight total doses of benzodiazepines were recorded as administered; 39 doses were given to patients via EMS and 69 were given in the ED 9
ACCEPTED MANUSCRIPT (Figure 2). There was little overall variation in dosages of benzodiazepine administered; doses usually administered were 2 mg, 5 mg, and 5 mg for lorazepam, midazolam, and diazepam respectively. EMS utilized midazolam the majority of the times (27 doses; 69%). Of note, midazolam was given IV more often (48%) than IM (30%). Lorazepam
T
was the preferred agent (63 doses; 91%) for treatment of GCSE in the ED, while
IP
midazolam was utilized in the rest. Virtually all of the lorazepam and midazolam doses
CR
in the ED were provided IV (97% and 83%, respectively); commonly administered doses of either agents were 2 mg. Overall, patients received a median of 2 doses of
US
benzodiazepines for the initial treatment of GCSE, but those that required advancement
AN
to 2nd-line therapy to attain seizure control received significantly more doses (3 doses vs 1.5 doses; p=0.008). Four patients (9%) received more than five individual
M
benzodiazepine doses before their therapy was advanced to 2nd-line agents or they
AC
CE
PT
ED
achieved seizure cessation.
10
ACCEPTED MANUSCRIPT
T P
I R
C S
U N
A
D E
M
T P
E C
C A
Figure 2: EMS and ED Usage Patterns
EMS = emergency medical services; ED = emergency department; SE = status epilepticus; IV = intravenous; IM intramuscular; SD = standard deviation
11
ACCEPTED MANUSCRIPT Table 2 provides additional details regarding patient intubation. All twenty-three patients (52%) underwent endotracheal intubation in the ED upon arrival; the primary rationale were “airway protection” (n= 13, 57%) and “respiratory distress” (n= 5, 22%). The median Glasgow Coma Score (GCS) of these patients upon arrival (prior to
T
intubation) was 5, a significantly lower score compared to patients that were not
IP
intubated (11; p-value = 0.025). The median time to intubation was 52 minutes; 49%
CR
were intubated within 30 minutes of hospital arrival. Only 22% (n= 5) of patients had been advanced to 2nd-line therapy prior to intubation. There was no significant
US
difference in the number of benzodiazepine doses administered prior to advancement to
AN
2nd-line therapy in intubated vs non-intubated patients (2.33 ± 1.96 vs 3.09 ± 2.11,
M
p=0.227). Total (n=23)
AC
CE
PT
ED
Characteristics Reason for intubation, n (%) Airway protection 13 (57) Respiratory distress 5 (22) Seizure control 4 (17) Rationale not provided 1 (4) Intubated patients, n (%) Intubation post-arrival 23 (100) Intubation within 30 11 (48) minutes of hospital arrival Advanced to urgent control therapy (2nd line) prior to 5 (22) intubation GCS prior to intubation, 5 (3-15) median (range) Total number of BZD dose administrations (Mean ± SD) Time from arrival to intubation, minutes Mean ± SD 68.4 ± 64.4 Median (range) 52 (5-235)
Seizure Cessation with Benzodiazepines? Yes No
pvalue
5 (39) 3 (60) 0 0
7 (54) 2 (40) 4 (100) 1 (100)
8 (35)
15 (65)
0.069
5 (46)
6 (55)
0.400
1 (20)
4 (80)
0.059
3 (3-7)
6 (3-15)
0.010
3.09 ± 2.11
2.33 ± 1.96
0.227
42.50 ± 45.5 18.5 (5-121)
82.13 ± 70.0 79.0 (5-235)
0.160
0.091
Table 2: Intubation Characteristics in Detail GCS = Glasgow coma score; BZD = benzodiazepine; SD = standard deviation
12
ACCEPTED MANUSCRIPT Discussion This study highlights a few important points regarding management of GCSE in a real-world setting. We noted consistent underdosing of benzodiazepines during initial treatment of GCSE, resulting in low seizure cessation, higher seizure recurrence and
T
greater admission rates to the hospital. Our findings are complementary to the
IP
observational study of Langer and Fountain in its scope and focus on clinical application
CR
of GCSE therapies in the absence of an institution-specific treatment protocol [8]. We
Table 3.
M
AN
Langer et al. (n=170)* 85 (48) 82 (46)
32 (73) 22 (50)
17 (11)
0
43 (26) 107 (63) 4 (2)
18 (41) 30 (68) 0
CE
PT
Braun et al. (n=44) 26 (59) 29 (66)
146 (86) 98 (56)
ED
Treatment Characteristics and Outcomes, n (%) BZD treatment provided by EMS Multiple BZD doses provided by EMS BZD treatment provided in ED Seizure response to BZD alone Received only a single guidelinerecommended dose of BZD in the ED Intubated following BZD treatment ICU admission Mortality
US
compare our study findings with that of Langer and Fountain through discussion and
* Included analysis of 176 episodes of GCSE Table 3: Treatment Patterns and Outcome Differences in Observational Studies
AC
BZD = benzodiazepine; EMS = emergency medical services; ED = emergency department; ICU = intensive care unit
Treatment recommendations for GCSE in adults provided by both the Neurocritical Care Society (NCS) and the American Epilepsy Society (AES) suggest weight-based benzodiazepine dosing. For an adult patient > 70 kg, a single dose of IM midazolam 10 mg, IV lorazepam 4mg, or IV diazepam 10 mg is recommended for initial treatment [4,5]. The results of our study showed a consistent underdosing of
13
ACCEPTED MANUSCRIPT benzodiazepines for GCSE treatment in both EMS and ED settings. We found midazolam being predominantly used by EMS in pre-hospital treatment of GCSE, although the route used was primarily IV. This is at variance from the current guidelines, which suggest midazolam dose administration via IM route. In contrast, Langer’s EMS
IP
practice could be due to the awareness of recent trial findings.
T
services preferred diazepam, which was utilized 73% of the time [8]. This difference in
CR
In the ED, 32 patients (73%) received benzodiazepine treatment, irrespective of previous benzodiazepine treatment provided by EMS, while 86% of patients in the
US
Langer study received benzodiazepine doses in the ED. Lorazepam was the
AN
benzodiazepine of choice in both EDs, utilized in 86% of the Langer and Fountain patient cohort versus 91% of the doses given in our facility [8]. The ED physician’s
M
practice of benzodiazepine administration was similar to that of EMS (that of consistent
ED
underdosing) regardless of the chosen agent. This could be attributed to provider reluctance in administering weight-appropriate, guideline-recommended dose of
PT
benzodiazepines due to perceived potential for over-sedation and respiratory failure [9].
CE
Administration of lower than recommended initial doses of benzodiazepines likely resulted in seizure cessation in only 50% of patients. This is lower than the seizure
AC
cessation rates of 59-68% reported in trials and the 56% cessation rate to benzodiazepines alone reported in the Langer and Fountain study [6-8,10]. Compared with the findings of Langer and Fountain, a greater proportion of our patients were treated earlier and with multiple benzodiazepine doses via EMS, but this did not translate into a greater proportion achieving seizure cessation. More specific data regarding the average benzodiazepine doses observed in the Langer and Fountain
14
ACCEPTED MANUSCRIPT study was not available, limiting further comparative analysis. Patients receiving suboptimal benzodiazepine doses could potentially explain the increased rate of benzodiazepine failure observed. The inherent danger in utilizing lower initial benzodiazepine dosages relates to seizure continuance and the subsequent
T
neurotransmitter changes that promote more self-sustaining seizures and increased
IP
pharmacoresistance, thereby reducing treatment response to subsequent efforts [11].
CR
While 17 patients (11%) received a single dose of 0.1 mg/kg lorazepam or equivalent dose of another benzodiazepine in the Langer and Fountain study, only one patient in
US
our observation received a single dose of benzodiazepine at a guideline-recommended
AN
dosage [8]. Both studies reflect the need for institution-based GCSE treatment protocols. The AES guidelines furthermore suggest initiation of secondary therapy
M
around 20 minutes following initial emergent treatment, which was not seen in our
recurrence in our cohort [5].
ED
cohort. These delays could further explain comparatively high rates of seizure
PT
Previous clinical trials have reported intubation rates ranging from 10-20%
CE
[6,7,8,12]. We observed overall intubation rates of 52% in our patient cohort, nearly half of them intubated < 30 minutes after patient’s arrival to the ED; 41% of patients
AC
intubated after benzodiazepine administration. In comparison, the overall intubation rate in Langer and Fountain’s study was 44% (75 patients), with only 26% intubated after benzodiazepine administration [8] . Our intubation rates following benzodiazepine administration could be explained by the “low” GCS noted in our cohort. Similar intubation rates were reported in a recent study where 60% patients presenting with SE were intubated in the ED because of their low GCS (median 6) [13]. A low GCS in SE
15
ACCEPTED MANUSCRIPT patients could be secondary to treatment being provided or the disease itself. The discrepancy in intubation rates reported in trials versus those observed in the real-world settings could be potentially explained by provider bias, where patients are intubated predominantly for “airway protection.”
T
All patients included in our study were admitted to the hospital. Sixty-eight
IP
percent of patients were admitted to the ICU, a rate similar to that of the Langer and
CR
Fountain cohort [8]. While Langer and Fountain study characterized the course of seizure management in GCSE patients in detail, it did not report the impact of
US
suboptimal treatment dosing on clinical outcomes like length of ICU and hospital stay. In
AN
our cohort observations, the median ICU LOS was only 1 day as opposed to 3 days in the RAMPART trial [7]. A proportion of patients may not have required intubation if
M
better-defined intubation criteria or a treatment protocol were utilized in the decision-
ED
making process. More definitive root cause analysis of intubation rates in our cohort was limited by the retrospective nature of this study and delineation between intubation
PT
secondary to persistent seizure versus oversedation via benzodiazepines was not
CE
possible. Whether improved guideline compliance would result in reduced intubation rates and ICU admissions is deserving of further study.
AC
Our study has several limitations, primarily stemming from the small sample size of our cohort. This cohort was selected to understand the treatment patterns in patients with initial diagnosis of SE and without any obvious intracranial injuries, which would otherwise have influenced provider approach to care. The small sample size limited further analysis of benzodiazepine dosing, although it provided pertinent observations. All 44 patients in our observations survived, which may be due to small sample size as
16
ACCEPTED MANUSCRIPT well as our extensive exclusion criteria. Quality of data points (particularly the timing of benzodiazepine and other antiepileptic drug administrations in relation to the ongoing seizure activity) collected through the electronic medical records could be affected due to the post-event documentation process frequently utilized in emergent care. We used
T
the NCS and recently updated AES guidelines as a template treatment of SE, even
IP
though they were only recently published or were still in inception at the time of some of
CR
our observations. The guidelines, however, draw from trials that have been wellpublicized and in the public domain for a considerable period prior to their synthesis.
US
Whether treatment decisions and patterns will improve with familiarization of these
AN
guidelines will require further study.
M
Conclusion
ED
Frequent but underdosed benzodiazepine administration is a routine occurrence in the treatment of SE in the EMS and ED settings, while timely advancement to 2 nd-line
PT
urgent control therapy is not. These treatment patterns may explain the increased
CE
incidence of complications observed during SE treatment at our institution. Future efforts will focus on educating providers regarding current guideline dosing
AC
recommendations and more rapid advancement to 2 nd-line therapies; whether ensuring familiarization with guidelines will result in improved treatment of SE and reduce treatment failure will require further study.
Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
17
ACCEPTED MANUSCRIPT References: 1. Lowenstein DH, Alldredge BK. Status epilepticus. N Engl J Med. 1998;338(14):970-976. DOI: 10.1056/NEJM199804023381407 2. Betjemann JP, Lowenstein DH. Status epilepticus in adults. Lancet Neurol.
T
2015;14(6):615-624. DOI: 10.1016/S1474-4422(15)00042-3
IP
3. Feen ES, Bershad EM, Suarez JI. Status epilepticus. South Med J.
CR
2008;101(4):400-406. DOI: 10.1097/SMJ.0b013e31816852b0 4. Brophy GM, Bell R, Claassen J, Alldredge B, Bleck TP, Glauser T, Laroche SM,
US
Riviello JJ Jr, Shutter L, Sperling MR, Treiman DM, Vespa PM; Neurocritical
AN
Care Society Status Epilepticus Guideline Writing Committee. Guidelines for the evaluation and management of status epilepticus. Neurocrit Care. 2012;17(1):3-
M
23. DOI: 10.1007/s12028-012-9695-z
ED
5. Glauser T, Shinnar S, Gloss D, Alldredge B, Arya R, Bainbridge J, Bare M, Bleck T, Dodson WE, Garrity L, Jagoda A, Lowenstein D, Pellock J, Riviello J, Sloan E,
PT
Treiman DM. Evidence-based guideline: treatment of convulsive status
CE
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-
AC
7597-16.1.48
6. Alldredge BK, Gelb AM, Isaacs SM, Corry MD, Allen F, Ulrich S, Gottwald MD, O'Neil N, Neuhaus JM, Segal MR, Lowenstein DH. A comparison of lorazepam, diazepam, and placebo for the treatment of out-of-hospital status epilepticus. N Engl J Med. 2001;345(9):631-637. DOI: 10.1056/NEJMoa002141 7. Silbergleit R, Durkalski V, Lowenstein D, Conwit R, Pancioli A, Palesch Y, Barsan W; NETT Investigators. Intramuscular versus intravenous therapy for 18
ACCEPTED MANUSCRIPT prehospital status epilepticus. N Engl J Med. 2012;366(7):591-600. DOI: 10.1056/NEJMoa1107494 8. Langer JE, Fountain NB. A retrospective observational study of current treatment for generalized convulsive status epilepticus. Epilepsy Behav. 2014;37:95-99.
T
DOI: 10.1016/j.yebeh.2014.06.008
IP
9. Bell DM, Richards G, Dhillon S, Oxley JR, Cromarty J, Sander JW, Patsalos PN.
CR
A comparative pharmacokinetic study of intravenous and intramuscular midazolam in patients with epilepsy. Epilepsy Res. 1991;10(2-3):183-90.
US
http://dx.doi.org/10.1016/0920-1211(91)90011-4
AN
10. Treiman DM, Meyers PD, Walton NY, Collins JF, Colling C, Rowan AJ, Handforth A, Faught E, Calabrese VP, Uthman BM, Ramsay RE, Mamdani MB. A
M
comparison of four treatments for generalized convulsive status epilepticus.
ED
Veterans Affairs Status Epilepticus Cooperative Study Group. N Engl J Med. 1998;339(12):792-798. DOI: 10.1056/NEJM199809173391202
PT
11. Mazarati AM, Baldwin RA, Sankar R, Wasterlain CG. Time-dependent decrease
CE
in the effectiveness of antiepileptic drugs during the course of self-sustaining status epilepticus. Brain Res. 1998;814(1–2):179.
AC
http://dx.doi.org/10.1016/S0006-8993(98)01080-4 12. Vohra TT, Miller JB, Nicholas KS, Varelas PN, Harsh DM, Durkalski V, Silbergleit R, Wang HE; NETT Investigators. Endotracheal intubation in patients treated for prehospital status epilepticus. Neurocrit Care. 2015;23(1):33-43. DOI: 10.1007/s12028-014-0106-5
19
ACCEPTED MANUSCRIPT 13. Tobochnik S, Gutierrez C, Jacobson MP. Characteristics and acute outcomes of ICU patients with initial presentation of seizure. Seizure. 2015;26:94-7. DOI:
AC
CE
PT
ED
M
AN
US
CR
IP
T
10.1016/j.seizure.2015.01.016
20
ACCEPTED MANUSCRIPT Impact of Non-Guideline-Based Treatment of Status Epilepticus
Highlights:
Benzodiazepine underdosing and delayed 2 nd–line AED initiation in GCSE are
These may explain low rates of seizure cessation and high rates of seizure
IP
T
common.
US
These may have led to high intubation rates and subsequent hospitalizations
CE
PT
ED
M
AN
including ICU.
AC
CR
recurrence.
21
James Braun, Elizabeth Gau, Stacy Revelle, Laurie Byrne, Abhay Kumar PII: DOI: Reference:
S0022-510X(17)34321-6 doi:10.1016/j.jns.2017.09.031 JNS 15577
To appear in:
Journal of the Neurological Sciences
Received date: Revised date: Accepted date:
13 April 2017 29 August 2017 21 September 2017
Please cite this article as: James Braun, Elizabeth Gau, Stacy Revelle, Laurie Byrne, Abhay Kumar , Impact of non-guideline-based treatment of status epilepticus. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Jns(2017), doi:10.1016/j.jns.2017.09.031
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Impact of Non-Guideline-Based Treatment of Status Epilepticus James Braun, Pharm.D., BCCCP1; Elizabeth Gau, Pharm.D., BCCCP1; Stacy Revelle, Pharm.D., BCPS1; Laurie Byrne, M.D.2; Abhay Kumar, M.D.3 1
Department of Pharmacy Services
SSM Health Saint Louis University Hospital
IP
T
3635 Vista Ave. St. Louis, MO 63110 2
CR
Division of Emergency Medicine
US
Saint Louis University School of Medicine 3635 Vista Ave.
3
Department of Neurology
M
Saint Louis University School of Medicine
AN
St. Louis, MO 63110
ED
3635 Vista Ave. St. Louis, MO 63110
Abhay Kumar, MD
PT
Corresponding Author:
CE
Saint Louis University School of Medicine
AC
Department of Neurology
3635 Vista Ave. at Grand Blvd. Saint Louis, MO 63110 Fax: (314) 268-5166 Phone: 314-977-4830 Email: [email protected]
Abstract: 1
ACCEPTED MANUSCRIPT Impact of Non-Guideline-Based Treatment of Status Epilepticus
Background: This retrospective study analyzed benzodiazepine usage patterns in relation to
T
guideline recommendations for the treatment of generalized convulsive status
IP
epilepticus (GCSE) as practiced by emergency medical services (EMS) and the
CR
emergency department (ED) of an inner-city hospital. Secondary outcomes of interest
US
were adverse events and admission/discharge outcomes.
AN
Methods:
Records of all patients ≥18 years old diagnosed with GCSE between June 2012 and
ED
M
September 2015 and transported by EMS to our hospital ED were reviewed.
Results:
PT
Of 44 patients analyzed, 43 (98%) had a history of epilepsy. Benzodiazepine utilization
CE
varied; EMS preferred midazolam (69% of cases) while the ED utilized lorazepam (91% of cases). Benzodiazepine dosages used were lower than guideline recommendations.
AC
Seizure activity was aborted with benzodiazepines alone in 22 (50%) patients. Twelve patients (27%) experienced seizure recurrence following SE treatment and achievement of seizure cessation. Twenty-three (52%) patients required intubation after arrival to ED. All 44 patients were admitted; 30 (68%) required admission to the intensive care unit.
Conclusions:
2
ACCEPTED MANUSCRIPT There was consistent underdosing of benzodiazepines in treatment of GCSE in both EMS and ED settings likely resulting in underachievement of seizure cessation, while intubation rates were higher than reported when compared to previous studies. Prospective studies are needed to identify barriers to optimal benzodiazepine usage in
IP
T
GCSE patients.
CR
Keywords:
Status epilepticus; epilepsy; guidelines; benzodiazepines; emergency department;
AN
US
seizures
BZD = benzodiazepine ED = emergency department
ED
AES = American Epilepsy Society
M
Abbreviations:
PT
EMS = emergency medical services
CE
GCS = Glasgow Coma Score
GCSE = generalized convulsive status epilepticus
AC
ICD-9 = International Classification of Diseases, Ninth Revision ICU = intensive care unit LOS = Length of stay NCS = Neurocritical Care Society OSH = outside hospital SE = status epilepticus SZ = seizure Introduction 3
ACCEPTED MANUSCRIPT Patients in GCSE require immediate treatment, as permanent neurological damage can result with continued seizure activity and prolonged seizures are associated with significant morbidity and mortality [1-3]. Treatment of GCSE has evolved over the years with benzodiazepines being the recommended first-line
T
treatment [3-5].
IP
A study of out-of-hospital GCSE management comparing the efficacy of
CR
diazepam and lorazepam found that patients treated with lorazepam were most likely to achieve termination of seizure activity prior to hospital arrival [6]. More recently, a
US
multicenter, double-blind, randomized noninferiority trial compared the efficacy of two
AN
out-of-hospital benzodiazepine treatments provided by emergency medical services (EMS) –10 mg of intramuscular (IM) midazolam or 4 mg of intravenous (IV) lorazepam –
M
for adults presenting in SE. IM midazolam not only met noninferiority requirements, but
ED
also demonstrated superiority to IV lorazepam in terminating seizures before arrival in the ED [7].Based on these studies, recently published guidelines for management of
PT
GCSE patients recommended benzodiazepine dose optimization intended to achieve
CE
seizure cessation while minimizing the risk of significant respiratory depression [4,5]. A retrospective observational study published in 2014 characterized the prehospital and
AC
ED treatment provided to patients presenting with GCSE [8]. Investigators found almost half of the patients (n=170) received more than one dose of benzodiazepine but only 11% of patients received the optimal 0.1 mg/kg lorazepam or equivalent dose of another benzodiazepine. To characterize further the observed discrepancy between benzodiazepine dose recommendations and dosing patterns in practice, we aimed to study benzodiazepine
4
ACCEPTED MANUSCRIPT usage in the treatment of GCSE and its impact on hospitalization patterns and intubation rates at our institution.
Methods
T
Study Population
IP
This retrospective study was conducted at SSM Health Saint Louis University
CR
Hospital (Saint Louis, Missouri), a 356-bed urban, academic teaching hospital that services approximately 45,000 ED visits per year. The electronic medical records
US
(EpicCare EMR, Verona, WI) of patients ≥ 18 years who presented via EMS to the
AN
emergency department (ED) with an initial diagnosis of GCSE between June 2012 and September 2015 were evaluated for inclusion in the study. Patients with history of
M
stroke (ischemic or hemorrhagic) or known brain neoplasm were included. Patients
ED
were excluded if they received GCSE treatment at another hospital prior to transfer, if seizure activity was precipitated by acute neurological injuries (such as hemorrhage,
PT
traumatic brain injury, anoxic brain injury), hypoglycemia, alcohol withdrawal, drug
CE
overdose, cardiac arrest, or bradycardia (defined as a heart rate of < 40 bpm), if the patient had a benzodiazepine allergy, was pregnant, or incarcerated [7]. Patients were
AC
also excluded if they were deemed not to be in GCSE per clinical evaluation by neurology service at the time of their presentation. This study received approval from the Saint Louis University Institutional Review Board and Saint Louis University Hospital Research Compliance Committee; the need for informed consent was waived due to the retrospective nature of the study.
5
ACCEPTED MANUSCRIPT Definitions For the purposes of this analysis, GCSE was defined as 5 minutes or more of continuous clinical seizure activity or acute seizure activity without recovery (return to baseline) between seizures [4,5]. As this study focused on convulsive seizure activity,
T
electroencephalogram data was not utilized. Seizure recurrence was defined as any
IP
return of seizure activity within 12 hours of hospital arrival that required additional
CR
benzodiazepine administration and occurred after cessation of initial seizure activity. A descriptive report of seizure response following medication administration was utilized
US
to determine time to response, since reliable reporting of seizure cessation via timed
AN
flow chart was not always available.
M
Treatment Protocol
ED
Our institution lacked an established SE treatment protocol during the period of this study, but drug levels of home antiepileptic therapy were obtained and home
PT
therapy was resumed following emergent treatment when applicable. First-line therapy
CE
was defined as benzodiazepine treatment, whether provided by EMS or in the ED; second- and third-line therapies were defined as treatments provided subsequent to
AC
benzodiazepine administration and second-line therapies, respectively. Seizure cessation with benzodiazepines was defined as cessation of convulsive seizure activity at any time after first-line benzodiazepine administration and before treatment advancement to second-line therapies. Prehospital treatment was defined as benzodiazepine therapy provided via EMS personnel prior to ED arrival.
6
ACCEPTED MANUSCRIPT Study Outcomes The primary outcome of interest was benzodiazepine usage patterns of EMS and the ED. This encompassed benzodiazepine selection, the dose given, and route of administration. EMS medication reports were reviewed to collect the details regarding
T
benzodiazepine administrations provided prior to ED arrival. Secondary outcomes were
IP
adverse events (seizure recurrence and the proportion of patients intubated), as well as
CR
admission and discharge outcomes. Demographic characteristics were collected for all
US
patients.
AN
Statistical Analysis
Demographic data were reported via descriptive statistics. Categorical variables
M
were compared via Fischer’s exact test or Χ2 test, as appropriate. Continuous variables
ED
were analyzed using Student’s t test. Statistical analyses were performed using SPSS
PT
(IBM SPSS Statistics 24 for Windows).
CE
Results
A total of 124 patients were identified by their initial ED diagnosis of GCSE, of
AC
which 44 met the inclusion criteria (Figure 1). Of the 80 patients excluded, 29 patients received management for GCSE at other hospital EDs prior to their transfer, while 18 patients were deemed not to be in GCSE per clinical evaluation by neurology service at the time of their presentation. 124 patients were identified via ICD-9 codes, discharge diagnosis, and ED chief complaint
7
ACCEPTED MANUSCRIPT
US
CR
IP
T
80 Were excluded 29 Acute SE was managed at OSH or developed GCSE later in hospital course 18 Did not present in SE (post-ictal or did not meet GCSE criteria per neurology) 14 Duplicate patient chart 5 Missing EMS medication administration report (both dose and route) 5 Prisoners 3 Use of secondary/refractory control concomitantly with or before benzodiazepine administration 3 Did not arrive via EMS 2 Seizure precipitated by major trauma 1 Seizure precipitated by hypoglycemia
AN
44 patients treated for SE by EMS/ED personnel with benzodiazepines and included in final analysis
Figure 1: Study Design
ED
M
ICD-9 = International Classification of Diseases, Ninth Revision; ED = emergency department; SE = status epilepticus; OSH = outside hospital; GCSE = generalized convulsive status epilepticus; EMS = emergency medical services
Table 1 shows patient characteristics; median patient age was 55 years and the
PT
majority of patients had a history of epilepsy (91%). Half of these patients presented
CE
due to noncompliance with prescribed seizure treatment, while a third had a coexisting condition. Fifty percent of patients (n= 22) required initiation of 2 nd-line therapy after
AC
presenting to the ED to abort seizures. Twenty-seven percent of patients (n= 12) experienced seizure recurrence within 12 hours of arrival, mostly occurring in patients (n= 10) whose initial seizures could not be terminated with benzodiazepines alone. All 44 patients were admitted; 68% were admitted to the ICU. Median hospital length of stay (LOS) was 3 days; median ICU LOS was 1 day.
Characteristics
Total (n=44)
Seizure Cessation with Benzodiazepines? Yes No
pvalue 8
ACCEPTED MANUSCRIPT 58 (23-84)
53 (22-88)
0.332
73 (45-107)
77 (51-134)
0.041
15 (54) 20 (50)
13 (46) 20 (50)
0.755 1.000
30 (68) 13 (30) 1 (2)
15 (50) 7 (54) 0
15 (50) 6 (46) 1 (100)
21 (48) 10 (23) 13 (30) 7 (18) 4 (9) 2 (5)
8 (38) 8 (80) 6 (46) 4 (57) 1 (25) 1 (50)
8 (3-15)
6 (3-15)
22 (50)
22 (50)
M
Number of BZD doses administered per patient (EMS and ED combined), n Median (range) 2.0 (1-9) 1.5 (1-5) 3.0 (1-9) Seizure recurrence within 12 hours of arrival, n (%) Hospital admission, n (%) ICU admission, n (%) Endotracheal intubation, n (%)
0.481
T IP
13 (62) 2 (20) 7 (54) 3 (43) 3 (75) 1 (50)
AN
GCS at ED arrival, median (range)* 7 (3-15) Seizure terminated with emergent therapy, n (%)
CR
Male gender, n (%) History of seizures, n (%) Race, n (%) Black Caucasian Asian Precipitating cause of SE, n (%) Noncompliance Idiopathic/breakthrough SZ Coexisting condition Malignancy Previous stroke Other
55 (22-88) 75 (45-134) 28 (64) 40 (91)
US
Age, median (range) Weight (kg), median (range)
0.113
0.617
0.092
2 (17)
10 (83)
0.016
44 (100) 30 (68) 23 (52)
12 (40) 8 (35)
18 (60) 15 (65)
0.104 0.069
Length of hospital admission, days Mean ± SD Median (range) Mode
6.1 ± 8.8 3 (0-50) 2
2.82 ± 3.10 2.0 (0-15) 1
9.41 ± 11.30 5.0 (1-50) 2
0.013
Length of ICU admission, days Mean ± SD Median (range) Mode
4.0 ± 6.4 1 (0-26) 1
1.09 ± 2.36 < 1 (0-11) <1
4.95 ± 7.38 1.0 (0-26) <1
0.026
AC
CE
PT
ED
12 (27)
Table 1: Baseline Characteristics and Outcomes SE = status epilepticus; GCS = Glasgow coma score; EMS = emergency medical services; ED = emergency department; SZ = seizure; ICU = intensive care unit; BZD = benzodiazepine; SD = standard deviation * GCS missing, n=3
One hundred and eight total doses of benzodiazepines were recorded as administered; 39 doses were given to patients via EMS and 69 were given in the ED 9
ACCEPTED MANUSCRIPT (Figure 2). There was little overall variation in dosages of benzodiazepine administered; doses usually administered were 2 mg, 5 mg, and 5 mg for lorazepam, midazolam, and diazepam respectively. EMS utilized midazolam the majority of the times (27 doses; 69%). Of note, midazolam was given IV more often (48%) than IM (30%). Lorazepam
T
was the preferred agent (63 doses; 91%) for treatment of GCSE in the ED, while
IP
midazolam was utilized in the rest. Virtually all of the lorazepam and midazolam doses
CR
in the ED were provided IV (97% and 83%, respectively); commonly administered doses of either agents were 2 mg. Overall, patients received a median of 2 doses of
US
benzodiazepines for the initial treatment of GCSE, but those that required advancement
AN
to 2nd-line therapy to attain seizure control received significantly more doses (3 doses vs 1.5 doses; p=0.008). Four patients (9%) received more than five individual
M
benzodiazepine doses before their therapy was advanced to 2nd-line agents or they
AC
CE
PT
ED
achieved seizure cessation.
10
ACCEPTED MANUSCRIPT
T P
I R
C S
U N
A
D E
M
T P
E C
C A
Figure 2: EMS and ED Usage Patterns
EMS = emergency medical services; ED = emergency department; SE = status epilepticus; IV = intravenous; IM intramuscular; SD = standard deviation
11
ACCEPTED MANUSCRIPT Table 2 provides additional details regarding patient intubation. All twenty-three patients (52%) underwent endotracheal intubation in the ED upon arrival; the primary rationale were “airway protection” (n= 13, 57%) and “respiratory distress” (n= 5, 22%). The median Glasgow Coma Score (GCS) of these patients upon arrival (prior to
T
intubation) was 5, a significantly lower score compared to patients that were not
IP
intubated (11; p-value = 0.025). The median time to intubation was 52 minutes; 49%
CR
were intubated within 30 minutes of hospital arrival. Only 22% (n= 5) of patients had been advanced to 2nd-line therapy prior to intubation. There was no significant
US
difference in the number of benzodiazepine doses administered prior to advancement to
AN
2nd-line therapy in intubated vs non-intubated patients (2.33 ± 1.96 vs 3.09 ± 2.11,
M
p=0.227). Total (n=23)
AC
CE
PT
ED
Characteristics Reason for intubation, n (%) Airway protection 13 (57) Respiratory distress 5 (22) Seizure control 4 (17) Rationale not provided 1 (4) Intubated patients, n (%) Intubation post-arrival 23 (100) Intubation within 30 11 (48) minutes of hospital arrival Advanced to urgent control therapy (2nd line) prior to 5 (22) intubation GCS prior to intubation, 5 (3-15) median (range) Total number of BZD dose administrations (Mean ± SD) Time from arrival to intubation, minutes Mean ± SD 68.4 ± 64.4 Median (range) 52 (5-235)
Seizure Cessation with Benzodiazepines? Yes No
pvalue
5 (39) 3 (60) 0 0
7 (54) 2 (40) 4 (100) 1 (100)
8 (35)
15 (65)
0.069
5 (46)
6 (55)
0.400
1 (20)
4 (80)
0.059
3 (3-7)
6 (3-15)
0.010
3.09 ± 2.11
2.33 ± 1.96
0.227
42.50 ± 45.5 18.5 (5-121)
82.13 ± 70.0 79.0 (5-235)
0.160
0.091
Table 2: Intubation Characteristics in Detail GCS = Glasgow coma score; BZD = benzodiazepine; SD = standard deviation
12
ACCEPTED MANUSCRIPT Discussion This study highlights a few important points regarding management of GCSE in a real-world setting. We noted consistent underdosing of benzodiazepines during initial treatment of GCSE, resulting in low seizure cessation, higher seizure recurrence and
T
greater admission rates to the hospital. Our findings are complementary to the
IP
observational study of Langer and Fountain in its scope and focus on clinical application
CR
of GCSE therapies in the absence of an institution-specific treatment protocol [8]. We
Table 3.
M
AN
Langer et al. (n=170)* 85 (48) 82 (46)
32 (73) 22 (50)
17 (11)
0
43 (26) 107 (63) 4 (2)
18 (41) 30 (68) 0
CE
PT
Braun et al. (n=44) 26 (59) 29 (66)
146 (86) 98 (56)
ED
Treatment Characteristics and Outcomes, n (%) BZD treatment provided by EMS Multiple BZD doses provided by EMS BZD treatment provided in ED Seizure response to BZD alone Received only a single guidelinerecommended dose of BZD in the ED Intubated following BZD treatment ICU admission Mortality
US
compare our study findings with that of Langer and Fountain through discussion and
* Included analysis of 176 episodes of GCSE Table 3: Treatment Patterns and Outcome Differences in Observational Studies
AC
BZD = benzodiazepine; EMS = emergency medical services; ED = emergency department; ICU = intensive care unit
Treatment recommendations for GCSE in adults provided by both the Neurocritical Care Society (NCS) and the American Epilepsy Society (AES) suggest weight-based benzodiazepine dosing. For an adult patient > 70 kg, a single dose of IM midazolam 10 mg, IV lorazepam 4mg, or IV diazepam 10 mg is recommended for initial treatment [4,5]. The results of our study showed a consistent underdosing of
13
ACCEPTED MANUSCRIPT benzodiazepines for GCSE treatment in both EMS and ED settings. We found midazolam being predominantly used by EMS in pre-hospital treatment of GCSE, although the route used was primarily IV. This is at variance from the current guidelines, which suggest midazolam dose administration via IM route. In contrast, Langer’s EMS
IP
practice could be due to the awareness of recent trial findings.
T
services preferred diazepam, which was utilized 73% of the time [8]. This difference in
CR
In the ED, 32 patients (73%) received benzodiazepine treatment, irrespective of previous benzodiazepine treatment provided by EMS, while 86% of patients in the
US
Langer study received benzodiazepine doses in the ED. Lorazepam was the
AN
benzodiazepine of choice in both EDs, utilized in 86% of the Langer and Fountain patient cohort versus 91% of the doses given in our facility [8]. The ED physician’s
M
practice of benzodiazepine administration was similar to that of EMS (that of consistent
ED
underdosing) regardless of the chosen agent. This could be attributed to provider reluctance in administering weight-appropriate, guideline-recommended dose of
PT
benzodiazepines due to perceived potential for over-sedation and respiratory failure [9].
CE
Administration of lower than recommended initial doses of benzodiazepines likely resulted in seizure cessation in only 50% of patients. This is lower than the seizure
AC
cessation rates of 59-68% reported in trials and the 56% cessation rate to benzodiazepines alone reported in the Langer and Fountain study [6-8,10]. Compared with the findings of Langer and Fountain, a greater proportion of our patients were treated earlier and with multiple benzodiazepine doses via EMS, but this did not translate into a greater proportion achieving seizure cessation. More specific data regarding the average benzodiazepine doses observed in the Langer and Fountain
14
ACCEPTED MANUSCRIPT study was not available, limiting further comparative analysis. Patients receiving suboptimal benzodiazepine doses could potentially explain the increased rate of benzodiazepine failure observed. The inherent danger in utilizing lower initial benzodiazepine dosages relates to seizure continuance and the subsequent
T
neurotransmitter changes that promote more self-sustaining seizures and increased
IP
pharmacoresistance, thereby reducing treatment response to subsequent efforts [11].
CR
While 17 patients (11%) received a single dose of 0.1 mg/kg lorazepam or equivalent dose of another benzodiazepine in the Langer and Fountain study, only one patient in
US
our observation received a single dose of benzodiazepine at a guideline-recommended
AN
dosage [8]. Both studies reflect the need for institution-based GCSE treatment protocols. The AES guidelines furthermore suggest initiation of secondary therapy
M
around 20 minutes following initial emergent treatment, which was not seen in our
recurrence in our cohort [5].
ED
cohort. These delays could further explain comparatively high rates of seizure
PT
Previous clinical trials have reported intubation rates ranging from 10-20%
CE
[6,7,8,12]. We observed overall intubation rates of 52% in our patient cohort, nearly half of them intubated < 30 minutes after patient’s arrival to the ED; 41% of patients
AC
intubated after benzodiazepine administration. In comparison, the overall intubation rate in Langer and Fountain’s study was 44% (75 patients), with only 26% intubated after benzodiazepine administration [8] . Our intubation rates following benzodiazepine administration could be explained by the “low” GCS noted in our cohort. Similar intubation rates were reported in a recent study where 60% patients presenting with SE were intubated in the ED because of their low GCS (median 6) [13]. A low GCS in SE
15
ACCEPTED MANUSCRIPT patients could be secondary to treatment being provided or the disease itself. The discrepancy in intubation rates reported in trials versus those observed in the real-world settings could be potentially explained by provider bias, where patients are intubated predominantly for “airway protection.”
T
All patients included in our study were admitted to the hospital. Sixty-eight
IP
percent of patients were admitted to the ICU, a rate similar to that of the Langer and
CR
Fountain cohort [8]. While Langer and Fountain study characterized the course of seizure management in GCSE patients in detail, it did not report the impact of
US
suboptimal treatment dosing on clinical outcomes like length of ICU and hospital stay. In
AN
our cohort observations, the median ICU LOS was only 1 day as opposed to 3 days in the RAMPART trial [7]. A proportion of patients may not have required intubation if
M
better-defined intubation criteria or a treatment protocol were utilized in the decision-
ED
making process. More definitive root cause analysis of intubation rates in our cohort was limited by the retrospective nature of this study and delineation between intubation
PT
secondary to persistent seizure versus oversedation via benzodiazepines was not
CE
possible. Whether improved guideline compliance would result in reduced intubation rates and ICU admissions is deserving of further study.
AC
Our study has several limitations, primarily stemming from the small sample size of our cohort. This cohort was selected to understand the treatment patterns in patients with initial diagnosis of SE and without any obvious intracranial injuries, which would otherwise have influenced provider approach to care. The small sample size limited further analysis of benzodiazepine dosing, although it provided pertinent observations. All 44 patients in our observations survived, which may be due to small sample size as
16
ACCEPTED MANUSCRIPT well as our extensive exclusion criteria. Quality of data points (particularly the timing of benzodiazepine and other antiepileptic drug administrations in relation to the ongoing seizure activity) collected through the electronic medical records could be affected due to the post-event documentation process frequently utilized in emergent care. We used
T
the NCS and recently updated AES guidelines as a template treatment of SE, even
IP
though they were only recently published or were still in inception at the time of some of
CR
our observations. The guidelines, however, draw from trials that have been wellpublicized and in the public domain for a considerable period prior to their synthesis.
US
Whether treatment decisions and patterns will improve with familiarization of these
AN
guidelines will require further study.
M
Conclusion
ED
Frequent but underdosed benzodiazepine administration is a routine occurrence in the treatment of SE in the EMS and ED settings, while timely advancement to 2 nd-line
PT
urgent control therapy is not. These treatment patterns may explain the increased
CE
incidence of complications observed during SE treatment at our institution. Future efforts will focus on educating providers regarding current guideline dosing
AC
recommendations and more rapid advancement to 2 nd-line therapies; whether ensuring familiarization with guidelines will result in improved treatment of SE and reduce treatment failure will require further study.
Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
17
ACCEPTED MANUSCRIPT References: 1. Lowenstein DH, Alldredge BK. Status epilepticus. N Engl J Med. 1998;338(14):970-976. DOI: 10.1056/NEJM199804023381407 2. Betjemann JP, Lowenstein DH. Status epilepticus in adults. Lancet Neurol.
T
2015;14(6):615-624. DOI: 10.1016/S1474-4422(15)00042-3
IP
3. Feen ES, Bershad EM, Suarez JI. Status epilepticus. South Med J.
CR
2008;101(4):400-406. DOI: 10.1097/SMJ.0b013e31816852b0 4. Brophy GM, Bell R, Claassen J, Alldredge B, Bleck TP, Glauser T, Laroche SM,
US
Riviello JJ Jr, Shutter L, Sperling MR, Treiman DM, Vespa PM; Neurocritical
AN
Care Society Status Epilepticus Guideline Writing Committee. Guidelines for the evaluation and management of status epilepticus. Neurocrit Care. 2012;17(1):3-
M
23. DOI: 10.1007/s12028-012-9695-z
ED
5. Glauser T, Shinnar S, Gloss D, Alldredge B, Arya R, Bainbridge J, Bare M, Bleck T, Dodson WE, Garrity L, Jagoda A, Lowenstein D, Pellock J, Riviello J, Sloan E,
PT
Treiman DM. Evidence-based guideline: treatment of convulsive status
CE
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-
AC
7597-16.1.48
6. Alldredge BK, Gelb AM, Isaacs SM, Corry MD, Allen F, Ulrich S, Gottwald MD, O'Neil N, Neuhaus JM, Segal MR, Lowenstein DH. A comparison of lorazepam, diazepam, and placebo for the treatment of out-of-hospital status epilepticus. N Engl J Med. 2001;345(9):631-637. DOI: 10.1056/NEJMoa002141 7. Silbergleit R, Durkalski V, Lowenstein D, Conwit R, Pancioli A, Palesch Y, Barsan W; NETT Investigators. Intramuscular versus intravenous therapy for 18
ACCEPTED MANUSCRIPT prehospital status epilepticus. N Engl J Med. 2012;366(7):591-600. DOI: 10.1056/NEJMoa1107494 8. Langer JE, Fountain NB. A retrospective observational study of current treatment for generalized convulsive status epilepticus. Epilepsy Behav. 2014;37:95-99.
T
DOI: 10.1016/j.yebeh.2014.06.008
IP
9. Bell DM, Richards G, Dhillon S, Oxley JR, Cromarty J, Sander JW, Patsalos PN.
CR
A comparative pharmacokinetic study of intravenous and intramuscular midazolam in patients with epilepsy. Epilepsy Res. 1991;10(2-3):183-90.
US
http://dx.doi.org/10.1016/0920-1211(91)90011-4
AN
10. Treiman DM, Meyers PD, Walton NY, Collins JF, Colling C, Rowan AJ, Handforth A, Faught E, Calabrese VP, Uthman BM, Ramsay RE, Mamdani MB. A
M
comparison of four treatments for generalized convulsive status epilepticus.
ED
Veterans Affairs Status Epilepticus Cooperative Study Group. N Engl J Med. 1998;339(12):792-798. DOI: 10.1056/NEJM199809173391202
PT
11. Mazarati AM, Baldwin RA, Sankar R, Wasterlain CG. Time-dependent decrease
CE
in the effectiveness of antiepileptic drugs during the course of self-sustaining status epilepticus. Brain Res. 1998;814(1–2):179.
AC
http://dx.doi.org/10.1016/S0006-8993(98)01080-4 12. Vohra TT, Miller JB, Nicholas KS, Varelas PN, Harsh DM, Durkalski V, Silbergleit R, Wang HE; NETT Investigators. Endotracheal intubation in patients treated for prehospital status epilepticus. Neurocrit Care. 2015;23(1):33-43. DOI: 10.1007/s12028-014-0106-5
19
ACCEPTED MANUSCRIPT 13. Tobochnik S, Gutierrez C, Jacobson MP. Characteristics and acute outcomes of ICU patients with initial presentation of seizure. Seizure. 2015;26:94-7. DOI:
AC
CE
PT
ED
M
AN
US
CR
IP
T
10.1016/j.seizure.2015.01.016
20
ACCEPTED MANUSCRIPT Impact of Non-Guideline-Based Treatment of Status Epilepticus
Highlights:
Benzodiazepine underdosing and delayed 2 nd–line AED initiation in GCSE are
These may explain low rates of seizure cessation and high rates of seizure
IP
T
common.
US
These may have led to high intubation rates and subsequent hospitalizations
CE
PT
ED
M
AN
including ICU.
AC
CR
recurrence.
21