Intranasal Midazolam vs Rectal Diazepam in
`Acute Childhood Seizures
`Madhumita Bhattacharyya, MD, Veena Kalra, MD, and Sheffali Gulati, MD
`
`One hundred eighty-eight seizure episodes in 46 chil-
`dren were randomly assigned to receive treatment with
`rectal diazepam and intranasal midazolam with doses
`of 0.3 mg/kg body weight and 0.2 mg/kg body weight,
`respectively. Efficacy of the drugs was assessed by drug
`administration time and seizure cessation time. Heart
`rate, blood pressure, respiratory rate, and oxygen
`saturation were measured before and after 5, 10, and
`30 minutes following administration of the drugs in
`both groups. Mean time from arrival of doctor to drug
`administration was 68.3 ⴞ 55.12 seconds in the diaze-
`pam group and 50.6 ⴞ 14.1 seconds in the midazolam
`group (P ⴝ 0.002). Mean time from drug administra-
`tion to cessation of seizure was significantly less in the
`midazolam group than the diazepam group (P ⴝ
`0.005). Mean heart rate and blood pressure did not
`vary significantly between the two drug groups. How-
`ever, mean respiratory rate and oxygen saturation
`differed significantly between the two drug groups at 5,
`10, and 30 minutes after drug administration. Intrana-
`sal midazolam is preferable to rectal diazepam in the
`treatment of acute seizures in children. Its administra-
`tion is easy,
`it has rapid onset of action, has no
`significant effect on respiration and oxygen saturation,
`and is socially acceptable. © 2006 by Elsevier Inc. All
`rights reserved.
`
`Bhattacharyya M, Kalra V, Gulati S. Intranasal midazolam
`vs rectal diazepam in acute childhood seizures. Pediatr
`Neurol 2006;34:355-359.
`
`Introduction
`
`Seizure, a common neurologic medical emergency,
`continues to be associated with significant morbidity and
`mortality in the pediatric age group and affects 4-7% of
`children [1]. Early domiciliary treatment of seizures in the
`
`From the Department of Pediatrics, All India Institute of Medical
`Sciences, Ansari Nagar, New Delhi, India.
`
`community, school, or home with drugs that can be
`administered by parents, teachers, or nonmedical staff may
`be beneficial and can decrease morbidity and mortality [2].
`In planning domiciliary therapy, the safety, ease of admin-
`istration, choice of drug, route of therapy, and the practi-
`cability of familiarization by the user are important issues.
`Various drugs administered through different routes have
`been tried in the management of acute seizures.
`Rectal diazepam has been used successfully for home
`and hospital treatment of acute seizures [3]. Its use may be
`socially embarrassing and undesirable. Moreover, some
`special arrangement is required to administer it, which is
`difficult to arrange in homes, schools, and daycare centers.
`An effective treatment that can be easily administered by
`a more convenient, socially acceptable route is therefore
`needed.
`Midazolam, a benzodiazepine, has been described as an
`alternative rescue medication in the management of acute
`seizures [4,5]. Recent studies have demonstrated intrana-
`sal midazolam to be effective in the management of acute
`childhood seizures [6-9,10]. However, not many compar-
`ative studies have been undertaken, and the search for an
`easily administrable, effective drug to control acute sei-
`zure continues.
`In the light of the above background, the present study
`was undertaken to compare the efficacy and side effects of
`intranasal midazolam and rectal diazepam in the treatment
`of acute childhood seizures.
`
`Materials and Methods
`
`This study was a randomized, controlled, single masked study. All
`types of seizures including febrile seizures and all types of epilepsy in
`children of either sex, ages 3 months to 12 years, who attended the
`Institute’s outpatient department or emergency were included in the
`study. A written consent was obtained from the parents or guardians of
`children regarding their willingness to participate in the study. The study
`was approved by the Institute ethical committee.
`
`Communications should be addressed to:
`Dr. Bhattacharyya; 20/4, N.S.C. Bose Road; Graham’s Land;
`Kolkata – 700040, India.
`E-mail: madhumita57@yahoo.co.in
`Received June 21, 2004; accepted September 14, 2005.
`
`© 2006 by Elsevier Inc. All rights reserved.
`doi:10.1016/j.pediatrneurol.2005.09.006 ● 0887-8994/06/$—see front matter
`
`AQUESTIVE EXHIBIT 1140 Page 0001
`
`Bhattacharyya et al: Acute Childhood Seizures
`
`355
`
`

`

`Table 1. Diagnoses of children under study
`
`Table 3A. Comparative baseline characteristics of the two groups
`of children
`
`Serial
`Number
`
`Diagnosis
`
`No. of Children (%)
`(n ⴝ 46)
`
`Characteristics
`
`Mean
`
`S.D.
`
`P Value
`
`1
`2
`3
`4
`
`5
`
`Epilepsy
`Degenerative brain disease
`Neurocysticercosis
`Other central nervous
`system diseases
`Febrile seizures
`
`18 (39.13%)
`10 (21.73%)
`7 (15.21%)
`6 (13.04%)
`
`5 (10.86%)
`
`Drugs used in this study were intranasal midazolam (0.2 mg/kg) and
`rectal diazepam (0.3 mg/kg). Equal numbers of sealed, unmarked,
`identical envelopes containing the name of the drug to be administered
`were randomized by shuffling. A box containing these envelopes was
`kept in the pediatric ward. When a patient was enrolled into the study,
`randomization to either group was performed by picking an envelope,
`and the indicated medication was administered. Blood sugar and serum
`calcium were assessed before enrollment and after seizure in each
`patient.
`Midazolam was instilled into the anterior nares with the help of a nasal
`dropper, and diazepam was introduced into the rectum with an 8-F size
`infant feeding tube that was inserted 4 cm inside the anal opening.
`Children with hypoglycemic seizures, hypocalcemic seizures, and upper
`respiratory tract infection were excluded from the study.
`The end of the seizure episode (clinically) was defined as the cessation
`of visible epileptic phenomena or return of purposeful response to
`external stimuli. If the seizure did not end within 10 minutes of drug
`administration, the treatment was deemed to be ineffective. Heart rate,
`respiratory rate, blood pressure, and oxygen saturation by pulse oximetry
`were measured before drug administration and monitored at 5 minutes,
`10 minutes, and 30 minutes after drug administration. Recurrence of
`seizures within 60 minutes of drug administration was also evaluated.
`The children were monitored for side effects such as vomiting, excessive
`somnolence, respiratory depression, and apnea after drug administration.
`A stop-watch was used to measure all time accurately by investigators.
`
`Sample Size
`
`A previous clinical study by Scott et al. [11], in which “seizure
`episode” was the unit of randomization, demonstrated the efficacy for
`control of seizure episode in the midazolam group and the diazepam
`group to be approximately 75% and 59%, respectively. Again, “seizure
`cessation time” after administration of the drugs was 6 minutes and 8
`minutes in these two groups, respectively. Thus, considering both factors,
`it was calculated that at least 90 seizure episodes were required to be
`enrolled in each group to produce a statistically significant difference at
`a power of 90% with a P value of ⬍0.05 and odds ratio 0.333.
`
`Statistical Analysis
`
`Data were recorded on a predesigned proforma. Unit of analysis was
`episode of seizure. Covariates between two groups (midazolam and
`
`Table 2. The types of seizure episodes in study children
`
`Serial
`Number
`
`Type of Seizure
`
`Number of Episodes
`(%) (n ⴝ 188)
`
`1
`2
`3
`4
`
`Simple partial seizures
`Generalized tonic clonic seizures
`Myoclonic seizures
`Others, e.g., absence, atonic
`seizures
`
`92 (48.9%)
`70 (37%)
`19 (10.1%)
`7 (3.8%)
`
`Chronologic age (months) (n ⫽ 46)
`Diazepam
`Midazolam
`Age of onset of first seizure (n ⫽
`46)
`Diazepam
`Midazolam
`Developmental age (n ⫽ 46)
`Diazepam
`Midazolam
`
`74.53
`60.47
`
`38.29
`45.35
`
`53.72
`47.56
`
`66.7
`48.06
`
`41.31
`43.76
`
`43.07
`48.42
`
`0.29
`
`0.48
`
`0.22
`
`diazepam) were compared by chi-square test, Fisher’s Exact Test, or
`Student t test. In case of more than one episode of seizure per child,
`repeated-measures analysis of variance using generalized estimation of
`equation was applied. Mann-Whitney U test and Wilcoxon signed rank
`sum test were also applied to determine the pairwise comparison for
`continuous data.
`
`Results
`
`Of 188 seizure episodes in 46 children under study, 96
`episodes were treated with rectal diazepam and 92 with
`intranasal midazolam. The diagnoses of these 46 children
`and the type of seizures are summarized in Table 1 and
`Table 2, respectively. Comparative baseline characteristics
`of the two groups under study are presented in Table 3A
`and 3B.
`After comparing the baseline characteristics between
`the two groups, which did not vary significantly, an
`analysis of the 188 seizure episodes (96 episodes with
`rectal diazepam and 92 episodes with intranasal midazo-
`lam) was undertaken. “Doctor to drug time” (i.e., time
`taken by the doctor to prepare and administer the drug)
`and “seizure cessation time” after administration of the
`drug were significantly shorter in the midazolam group
`
`Table 3B. Comparison of some other baseline characteristics
`
`Diazepam
`
`Midazolam
`
`P Value
`
`Sex (n ⫽ 46)
`Male
`Female
`Category of seizure
`(n ⫽ 46)
`Controlled
`Provoked
`Intractable
`Family history of seizures
`(n ⫽ 46)
`Yes
`No
`History of birth asphyxia
`(n ⫽ 46)
`Yes
`No
`Perinatal history (n ⫽ 46)
`Normal
`Abnormal
`
`67.9%
`32.1%
`
`62.11%
`20.7%
`17.2%
`
`7.1%
`92.9%
`
`89.3%
`10.7%
`
`92.9%
`7.1%
`
`55.6%
`44.4%
`
`68.4%
`10.5%
`21.1%
`
`27.8%
`72.2%
`
`94.4%
`5.6%
`
`94.4%
`5.6%
`
`0.29
`
`0.64
`
`0.071
`
`0.48
`
`0.66
`
`356 PEDIATRIC NEUROLOGY Vol. 34 No. 5
`
`AQUESTIVE EXHIBIT 1140 Page 0002
`
`

`

`Table 4. Comparison of doctor to drug time and drug to seizure
`cessation time in rectal diazepam and intranasal midazolam group
`
`Discussion
`
`Rectal
`Diazepam
`(Seconds)
`Mean
`S.D.
`
`Intranasal
`Midazolam
`(Seconds)
`Mean
`S.D.
`
`68.3
`
`55.1
`
`50.6
`
`14.1
`
`178.6
`
`179.4
`
`116.7
`
`126.9
`
`P Value
`
`0.002
`
`0.005
`
`Doctor to drug
`time
`Drug to seizure
`cessation time
`
`(Table 4). Changes in heart rate, respiratory rate, blood
`pressure, and oxygen saturation, as measured at 5-minute,
`10-minute, and 30-minute intervals after administration of
`drugs in both groups, revealed that mean heart rate and
`blood pressure changes were not statistically different.
`Mean respiratory rate decreased by 1/minute at 5 minutes
`and 4/minute at 10 and 30 minutes after administration of
`rectal diazepam from predrug mean respiratory rate,
`whereas there was no decrease of mean respiratory rate at
`5/minutes and a decline of only 1/minute at 10 minutes
`and 30 minutes after administration of intranasal midazo-
`lam. By repeated-measures of analysis of variance, it was
`found that changes in respiratory rate differed significantly
`between the rectal diazepam group and the intranasal
`midazolam group at 10 minutes and 30 minutes after drug
`administration, with P ⫽ 0.027 and P ⫽ 0.039, respec-
`tively.
`Again, mean oxygen saturation (SaO2) after 5, 10,
`and 30 minutes of intranasal midazolam administration
`did not vary, whereas mean oxygen saturation in the
`rectal diazepam group decreased at 5 minutes and 30
`minutes after administration of the drug from predrug
`mean value. This difference was again statistically
`significant (P ⬍ 0.05). Hypoxia was observed in one
`child treated with rectal diazepam who required oxygen
`inhalation for 7 hours. No significant hypoxia was
`observed in the midazolam group.
`Seizures ceased within 10 minutes of drug adminis-
`tration in 85 of 96 episodes (88.5%) treated with rectal
`diazepam, whereas seizures ended in 89 of 92 episodes
`(96.7%) treated with intranasal midazolam (P ⫽ 0.060).
`Seizures were not controlled in 11 episodes (11.45%) of
`the rectal diazepam group and in 3 episodes (3.26%) of
`the intranasal midazolam group.
`Seizures recurred in 6 of 96 episodes (6.25%) within 60
`minutes of administration of rectal diazepam, and in 3 of
`92 episodes (3%) after administration of intranasal mida-
`zolam. The difference was not statistically significant.
`Side effects such as vomiting and excessive drowsiness
`were observed in 10 of 96 episodes (10.4%) in the rectal
`diazepam group, whereas no such side effects were ob-
`served in the midazolam group. The difference was
`significant statistically (p ⫽ 0.009).
`
`Early termination of seizures is important to prevent
`many adverse consequences and reduce the risk of devel-
`opment of status epilepticus. In a hospital setup, intrave-
`nous diazepam is commonly used for control of acute
`seizures, but it requires prompt establishment of an intra-
`venous line and has the disadvantage of being a respiratory
`depressant [12]. Rectal diazepam is another alternative
`route, but is not always reliable owing to its variable
`bioavailability and wide range of serum concentration
`[13,14]. There is also a risk of child abuse. Episodes of
`acute seizures have also been treated with buccal diaze-
`pam and sublingual lorazepam [15,16]. Administering the
`drugs orally or sublingually is frequently difficult and
`hazardous when children are convulsing. Moreover, ab-
`sorption of diazepam and lorazepam solution is relatively
`slow [16]. Application of drugs to nasal mucosa allows
`rapid absorption of drug into systemic circulation. Mida-
`zolam, a water-soluble benzodiazepine, was found to end
`seizures within 1 to 2 minutes of intranasal administration
`[7-9,17,18]. As such, the present study was undertaken to
`evaluate and compare the efficacy of rectal diazepam with
`intranasal midazolam in terminating acute seizures in
`children.
`Among 188 episodes randomized in the study, 96
`episodes were treated with rectal diazepam with a dose of
`0.3 mg/kg body weight and 92 episodes with intranasal
`midazolam with a dose of 0.2 mg/kg body weight. The
`doses of rectal diazepam used in previous studies are
`variable, ranging from 0.16 to 0.5 mg/kg [11,12,17]. In the
`present study, a midlevel dose of 0.3 mg/kg of rectal
`diazepam was used in order to avoid any cumulative side
`effects of diazepam in children, which was a possibility as
`a child always had a chance to receive diazepam more than
`one time, because the unit of randomization in this study
`was seizure episode. The preparations of diazepam re-
`ported to be used in earlier studies were intravenous
`preparations introduced rectally [12] or commercially
`available prepacked rectal diazepam [11,17]. A rectal tube
`for diazepam and a nasal dropper for midazolam were
`used in the present study as neither prepacked rectal
`diazepam nor midazolam drop or spray were available in
`our country during the period of study.
`
`Doctor to Drug Time
`
`In the current study, the drug administration time was
`observed to be shorter in the midazolam group than in the
`diazepam group (P ⫽ 0.002). Fisgin et al. [7,17] used an
`injector for the introduction of intranasal midazolam,
`through which the drug was introduced within 30 seconds.
`Lahat et al. [18] did not mention the time, but
`they
`dropped the drug immediately in the anterior nares even
`before the establishment of an intravenous line in children.
`Therefore, this easy and shorter administration time for
`
`AQUESTIVE EXHIBIT 1140 Page 0003
`
`Bhattacharyya et al: Acute Childhood Seizures
`
`357
`
`

`

`Table 5. Comparison of earlier studies with present study
`
`Dose of Rectal
`Diazepam
`
`Dose of Intranasal
`Midazolam
`
`Drug to Seizure (Cessation Time mean)
`Rectal
`Intranasal
`Diazepam
`Midazolam
`
`Not used
`Not used
`Not used
`0.3 mg/kg
`0.3 mg/kg
`
`0.2 mg/kg
`0.2 mg/kg
`0.2 mg/kg
`0.2 mg/kg
`0.2 mg/kg
`
`120–300
`178.6 ⫾ 179.5 s
`
`180–500 s
`139.6 ⫾ 129.8 s
`60–120 s
`60–120 s
`116.7 ⫾ 126.9 s
`
`Study Authors
`
`Lahat et al. [18]
`Kutlu et al. [8]
`Fisgin et al. [7]
`Fisgin et al. [17]
`Present study
`
`Abbreviation:
`s ⫽ Seconds
`
`intranasal midazolam plays an important role in the
`management of acute seizures.
`
`Drug to Seizure Cessation Time
`
`The present study demonstrated that the mean time for
`seizure cessation in the intranasal midazolam group was
`significantly shorter than that for rectal diazepam (p ⫽
`0.005) (Table 4). Intranasal midazolam was therefore
`believed to be more effective in controlling acute child-
`hood seizures rapidly, with less seizures cessation time
`than in the rectal diazepam group; this is probably because
`of the water solubility of midazolam and the rapid absorp-
`tion of the drug through the nasal mucosal vasculature.
`Bypassing the portal circulation, it reaches the systemic
`circulation more rapidly than rectal diazepam. These
`results compare favorably with earlier studies (Table 5).
`
`Comparison of Vital Parameters in Both Drug Groups
`
`In this study, the mean change of heart rate and mean
`systolic and diastolic blood pressure at 5, 10, and 30
`minutes did not vary significantly between the rectal
`diazepam group and the intranasal midazolam group.
`Mean respiratory rate decreased in the diazepam group,
`whereas it increased after intranasal midazolam adminis-
`tration from predrug values. This finding indicates that
`intranasal midazolam probably has no significant respira-
`tory depressant effect in children with acute seizures.
`Fisgin et al. [7,17] also detected tachypnea in their study
`children after administration of intranasal midazolam. The
`mean increase in respiratory rate by 1/minute after intra-
`nasal midazolam administration in the present study had
`no clinical significance. A possible explanation for this
`may be nasal mucosal irritation by local application of
`drug.
`This study also revealed a significant difference of
`oxygen saturation as measured by pulse oximeter between
`the diazepam and midazolam groups at 5, 10, and 30
`minutes after drug administration (P ⬍ 0.05). O’Regan et
`al. [14] found a severe decrease in oxygen saturation that
`corrected spontaneously in 1 of 19 children with intracta-
`ble seizures who received intranasal midazolam. No other
`
`studies found any significant fall in oxygen saturation after
`administration of intranasal midazolam. On the contrary,
`Dickmann [12] reported that of 16 children who received
`rectal diazepam, 7 required oxygen alone or oxygen with
`bag valve mask device to combat respiratory depression.
`The present study, which had a sample size larger than the
`previous studies, substantiates earlier reports that intrana-
`sal midazolam appears to have a good safety profile with
`regard to posttherapy oxygen saturation levels.
`
`Antiepileptic Efficacy
`
`Lahat et al. [9] and Kutlu et al. [8] reported that
`intranasal midazolam was effective in ending seizures
`within 10 minutes in 88.4% of study children. The only
`earlier study [17] that compared rectal diazepam with
`intranasal midazolam demonstrated that 20 of 23 (87%)
`children stopped convulsing within 10 minutes of intrana-
`sal midazolam administration and 13 of 22 (60%) children
`receiving rectal diazepam had their seizures controlled
`within 10 minutes (P ⬍ 0.05). The dose of intranasal
`midazolam and rectal diazepam was the same as used in
`the present study, i.e. 0.2 mg/kg and 0.3 mg/kg, respec-
`tively.
`In this study, 85 of 96 episodes (88.5%) in the rectal
`diazepam group and 89 of 92 episodes (96.7%) in the
`intranasal midazolam group were controlled within 10
`minutes of drug administration. Seizures remained uncon-
`trolled in 11 (11.45%) episodes in the diazepam group and
`in 3 (3.26%) in the midazolam group. The difference,
`however, was not statistically significant. Although this
`study had no untreated group for comparison owing to
`ethical constraints, it appears that intranasal midazolam
`may be a good domiciliary strategy for use in epileptic
`subjects. Seizures recurred in six episodes (6.25%) in the
`diazepam group and in three episodes (3.26%) in the
`midazolam group within 60 minutes of drug administra-
`tion. This study thus reveals that intranasal midazolam as
`well as rectal diazepam are equally effective in controlling
`acute seizure within 10 minutes of drug administration and
`that recurrence of seizures may occur in both groups.
`Side effects such as vomiting and excessive drowsiness
`were evident in 10 episodes (10.4%) treated in the rectal
`
`358 PEDIATRIC NEUROLOGY Vol. 34 No. 5
`
`AQUESTIVE EXHIBIT 1140 Page 0004
`
`

`

`diazepam group only. These side effects were observed in
`those children who were treated with the drug multiple
`times for recurrent episodes of seizures. This result is
`believed to be due to the cumulative effect of the drug
`after repeated administration. No such side effects were
`detected in the intranasal midazolam group, even on
`repeated use. This outcome reflects that intranasal mida-
`zolam is a safe drug without any significant side effects
`and can be used in children to control acute seizures. It
`compares favorably with rectal diazepam, with less side
`effects and marginal therapeutic superiority (p ⫽ 0.06).
`Social acceptability of rectal diazepam is understandably
`less, especially among young females.
`In conclusion, intranasal midazolam was found to be a
`reasonably safe route for terminating acute seizures in
`children. Its antiepileptic effect appeared comparable to
`conventional rectal diazepam. Further, with regard to
`quickness of response, safety, and ease of administration,
`intranasal midazolam was found to be superior.
`Future studies with concurrent electroencephalographic
`documentation are recommended to authenticate the effect
`of intranasal midazolam as an alternative route in the
`management of acute seizures.
`
`References
`[1] Kalra V. Seizure disorders and epilepsy. In: Kalra V, editor.
`Practical paediatric neurology, 1st ed. New Delhi: Arya Publication,
`2002:51-77.
`[2] Alldredge BK, Wall DB, Ferriero DM. Effect of prehospital
`treatment on the outcome of status epilepticus in children. Pediatr Neurol
`1995;12:213-6.
`
`[3] Remy C, Jourdil N, Villemain D, Favel P, Genton P. Intrarectal
`diazepam in epileptic adults. Epilepsia 1992;33:353-8.
`[4] Lahat E, Aladjem M, Eshel G, Bistritzer T, Katz Y. Midazolam
`in treatment of epileptic seizures. Pediatr Neurol 1992;8:215-6.
`[5] Rivera R, Segnini M, Baltodano A, Perez V. Midazolam in the
`treatment of status epilepticus in children. Crit Care Med 1993;21:991-4.
`[6] Koren G. Intranasal midazolam for febrile seizures. BMJ 2000;
`321:64-5.
`[7] Fisgin T, Gurer Y, Senbil N, et al. Nasal midazolam effects on
`childhood acute seizure. J Child Neurol 2000;15:833-5.
`[8] Kutlu NO, Yakinci C, Dogrul M, Durmaz Y. Intranasal mida-
`zolam for prolonged convulsive seizures. Brain Devel 2000;22:300-61.
`[9] Lahat E, Goldman M, Barr J, Bristritzer T, Berkovitch M.
`Comparison of intranasal midazolam with intravenous diazepam for
`treating febrile seizures in children: Prospective randomised study. BMJ
`2000;320:83-6.
`[10] Wallace SJ. Nasal benzodiazepines for management of acute
`childhood seizures? Lancet 1997;349:222.
`[11] Scott RC, Besag FMC, Neville BGR. Buccal midazolam and
`rectal diazepam for treatment of prolonged seizures in childhood and
`adolescence: A randomised trial. Lancet 1999;353:623-6.
`[12] Dickmann RA. Rectal diazepam for prehospital pediatric status
`epilepticus. Ann Emerg Med 1994;23:216-24.
`[13] Scheepers M, Scheepers B, Clarke M, Comish S, Ibitoye M. Is
`intranasal midazolam an effective rescue medication in adolescents and
`adults with severe epilepsy? Seizure 2000;9:417-21.
`[14] O’Regan M, Brown J, Clark M. Nasal rather than rectal
`benzodiazepines in the management of acute childhood seizure? Dev
`Med Child Neurol 1996;38:1037-45.
`[15] Yager JY, Sestria SS. Sublingual lorazepam in childhood serial
`seizures. Am J Dis Child 1988;142:931-2.
`[16] Dhillon S, Osley J, Richens A. Bioavailability of rectally
`administered lorazepam. Br J Clin Pharmacol 1982;13:427-32.
`[17] Fisgin T, Gurer Y, Tezic T, et al. Effects of intranasal
`midazolam and rectal diazepam on acute convulsions in children:
`Prospective randomised study. J Child Neurol 2002;17:123-6.
`[18] Lahat E, Goldman M, Barr J, Eshel G, Berkovitch M. Intrana-
`sal midazolam for childhood seizures. Lancet 1998;352:224.
`
`AQUESTIVE EXHIBIT 1140 Page 0005
`
`Bhattacharyya et al: Acute Childhood Seizures
`
`359
`
`