Received: 4 January 2024 | Revised: 22 June 2024 | Accepted: 17 July 2024

DOI: 10.1111/apa.17364

REVIEW ARTICLE

Effectiveness of pharmacological procedural sedation in

children with autism spectrum disorder: A systematic review
and meta-­analysis

Luisa Zupin1 | Cyril Sahyoun2 | Baruch Krauss3 | Arianna Dagri4 |

Elisabetta Maria Rocco4 | Egidio Barbi1,4 | Fulvio Celsi1

1
Paediatric Department, Institute for
Maternal and Child Health IRCCS Burlo Abstract
Garofolo, Trieste, Italy
Aim: Management of primary healthcare and routine minor procedures for children
2
Division of Paediatric Emergency
Medicine, Children's Hospital of Geneva,
with autism spectrum disorder (ASD) can be challenging; therefore, when behavioural
Geneva University Hospitals, Geneva, strategies fail, sedative medications are often employed. We evaluated the effective-
Switzerland
3
ness of the current pharmacological strategies for managing children with ASD.
Division of Emergency Medicine,
Department of Paediatrics, Boston Methods: We performed a systematic review and meta-­analysis of the current ap-
Children's Hospital, Harvard Medical proaches for procedural sedation in children with ASD.
School, Boston, Massachusetts, USA
4 Results: Twenty studies met inclusion criteria. Dexmedetomidine, midazolam, propo-
Department of Medicine, Surgery and
Health Sciences, University of Trieste, fol and chloral hydrate were the most efficient agents for successful procedures,
Trieste, Italy
while propofol had the highest number of adverse events. The most frequently used
Correspondence agents were dexmedetomidine and midazolam or a combination of the two, and the
Luisa Zupin, Institute for Maternal and
effectiveness of dexmedetomidine plus midazolam was superior to dexmedetomi-
Child Health IRCCS Burlo Garofolo,
Trieste, Italy. dine alone.
Email: [email protected]
Conclusion: Multiple effective drug regimens exist for procedural sedation in children
Funding information with ASD. These results could support the development of specific guidelines for pro-
Ministry of Health, Rome, Italy, in
cedural sedation in children with ASD.
collaboration with the Institute for
Maternal and Child Health IRCCS Burlo
Garofolo, Grant/Award Number: RC 29/23 KEYWORDS
autism spectrum disorders, children, dexmedetomidine, midazolam, procedural sedation

1 | I NTRO D U C TI O N with concurrent conditions, such as epilepsy, anxiety, attention deficit

disorder, self-­harm and depression. The World Health Organization
Autism spectrum disorder (ASD) includes neurodevelopmental condi- estimates a prevalence of ASD of 1 in 100 children worldwide.1,2
tions characterised by impairment of social interaction and communi- However, this figure is likely underestimated, as ASD is not always well
cation, often accompanied by restricted-­repetitive behaviours. ASD monitored or recognised in low and middle-­income countries.1
comprises individuals with intellectual disability, deficient language Managing these patients in the clinical setting raises numerous
skills and children with average or above-­average intellectual function challenges. A new environment, including new sounds, smells, unfa-
with difficulty in social communication. These children often present miliar personnel and medical equipment, can provoke overwhelming

Abbreviations: ASD, autism spectrum disorder; CEM, common effect model; CI, confidence interval; REM, random effect model.

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in
any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.
© 2024 The Author(s). Acta Paediatrica published by John Wiley & Sons Ltd on behalf of Foundation Acta Paediatrica.

Acta Paediatrica. 2024;113:2363–2377.  wileyonlinelibrary.com/journal/apa | 2363

 |

16512227, 2024, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/apa.17364 by Cochrane France, Wiley Online Library on [10/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2364 ZUPIN et al.

anxiety and distress. Therefore, routine primary care, including physi-

cal examinations, venipuncture and immunisation, can be challenging
Key Notes
for both provider and patient. Physical restraint is often required to
successfully complete minor procedures, resulting in traumatic epi- • The study investigated the effectiveness of current
3
sodes for these children. Behavioural interventions, where available, pharmacological strategies employed in medical proce-
often fail to manage the extreme anxiety and distress these children dures in children with autism spectrum disorder (ASD).
exhibit, especially when they have been previously traumatised. 4 • Procedural sedation was efficacious in the clinical man-
As a result, medical care may be postponed indefinitely or, in many agement of children with ASD.
cases, patients undergo general anaesthesia, a limited resource with • These results could support the development of specific
related risks. With the advent of procedural sedation, many of these guidelines for procedural sedation in children with ASD.
procedures can now be done in the outpatient setting (i.e. emergency
department, hospital sedation service) and be a resource for medical
providers.5 performed by importing the references in Zotero citation manager11
Procedural sedation, the use of sedative, analgesic and dissocia- and merging the items in one library.
tive agents to relieve anxiety and pain associated with diagnostic and The articles (title/abstract) were manually screened by three au-
therapeutic procedures outside of the operating theatre, has evolved thors independently, and if discordant, the selection was further dis-
into a widely practiced discipline by a diverse group of specialists.5 cussed with two other authors. Moreover, we employed the ‘revtool’
In a recent survey of current practice at 47 hospitals in the United package for R studio to test if the articles could cluster around topics,12
States caring for children with ASD, less than 30% of existing seda- but they did not, as shown in Figure 1.12 We initially performed the
tion programs included either an ASD-­specific protocol or additional manual screening of the title and abstract, then we retrieved and eval-
time or staffing for sedation of patients with ASD. Most hospitals use uated articles according to the following inclusion/exclusion criteria.
distraction (77%), involvement of parents (94%) or physical restraint Inclusion criteria were:
(45%) for minor procedures. Hospitals affiliated with an autism centre
tend to develop a specific protocol for ASD procedural sedation (71%) • clinical studies on human subjects with autistic spectrum disorder
compared to unaffiliated hospitals (51%). This recent survey found (ASD)
that sedation program directors were not satisfied with their practice • paediatric subjects (ages 0–20 years)
for children with ASD and that more training was needed, and more • use of pharmacological agents
studies were required to define protocols and international guidelines, • use of sedation to perform medical procedures
as well as more information disseminated at academic conferences.6 • English language
There are currently no guidelines published by international med- • Peer-­reviewed paper
ical societies for managing procedural anxiety and pain in children
with ASD. We performed a systematic review and a meta-­analysis to Exclusion criteria:
evaluate the effectiveness of the current pharmacological regimens
for procedural sedation in children with ASD. • ASD patients over 20 years of age
• in vitro and in vivo pre-­clinical studies
• Other no-­English Language
2 | M ATE R I A L S A N D M E TH O DS
Twenty papers fulfilled the inclusion criteria and were selected
We employed Rstudio7,8 to perform a systematic review according for the meta-­analysis (Figure 2).13
to the Preferred Reporting Items for Systematic Review and Meta-­ Three authors independently performed data extraction. The
Analysis (PRISMA) Statement.9 number of successful procedures performed during sedation was
We searched PubMed (https://​pubmed.​ncbi.​nlm.​nih.​gov), em- selected as the primary outcome, the number of adverse effects was
ploying the following optimised key terms: ‘sedation’ or ‘procedural also registered, and other relevant information regarding the results
sedation’ or ‘conscious sedation’ (in ‘Title/Abstract’) and ‘autism’ (in achieved.
‘Title/Abstract’) and paediatric or children (in ‘all fields’). The search
concluded on 30 September 2023, producing 134 results. Keywords
were automatically extracted from 92 articles, while in the remaining 2.1 | Meta-­analysis
42 articles keywords were extracted from the title using the Rapid
Automatic Keyword Extraction (RAKE) method of the Litesearch R We performed the meta-­analysis with the ‘meta’14 packages in
package.10 Rstudio,7 comparing two sedative agents with the ‘metabin’ func-
A literature search was then conducted in PubMed and Web of tion of binary outcome data using odds ratio—OR for the measure-
Science databases, retrieving 134 and 124 items, respectively and ment of treatment. When we did not proceed with a comparison,
then the deduplication resulted in 155 items. The deduplication was as for single-­arm studies or when a single study compared two
16512227, 2024, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/apa.17364 by Cochrane France, Wiley Online Library on [10/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2365

F I G U R E 1 Analysis of clustering performed with Revtool R package. (A) Topic clustering visualisation. (B) Topic frequencies. (C) List of the
|

F I G U R E 2 PRISMA flow diagram for

the selection of the articles.
topic terms.
ZUPIN et al.
 |

16512227, 2024, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/apa.17364 by Cochrane France, Wiley Online Library on [10/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2366 ZUPIN et al.

specific drugs, we utilised the ‘metaprop’ function for proportion were divided into two arms and different drugs were administered. In
data, estimating the overall effect size across the selected inves- 1 (5%) study, the same cohort with ASD underwent procedural seda-
tigations. We used standard and random effect models to syn- tion with two different pharmacological agents at different times.
thesise the results and check heterogenicity. In the case of high The drugs most frequently used were dexmedetomidine (n = 12),
variability, we used the ‘find.outliers’ function and recalculated midazolam (n = 6), chloral hydrate (n = 2), dexmedetomidine + mid-
the models excluding outliers. The ‘InfluenceAnalysis’ function in azolam (n = 3), propofol (n = 2), clonidine (n = 1), Trichlofos sodium
the ‘dmetar’15 package allowed examination of the impact of each (TFS) (n = 1), diazepam (n = 1), fentanyl + pentobarbital (n = 1), ket-
study's heterogeneity. amine (n = 1), midazolam + ketamine (n = 1). The procedures exe-
Since the 20 studies were heterogeneous regarding procedures cuted were electroencephalogram (EEG, n = 6), magnetic resonance
and type of sedation agent, we grouped the studies based on the imaging (MRI, n = 6), auditory brainstem response (ABR, n = 2), dental
drug(s) employed. Moreover, we considered only the groups where procedures (n = 2), emergency department procedures (n = 2), com-
the number of studies was equal or greater than two, while when puterised tomography (CT, n = 1), ophthalmology exams (n = 1), intra-
a single drug was employed in only one study no comparison was venous access (n = 1) and immunisation (n = 1).
possible so we excluded it form the meta-­analysis; therefore, four Considering the sedative agents singularly, we found that dex-
studies were excluded (Kaplan et al.,16 Mehta et al.,17 Pisalchaiyong medetomidine was administered in 12 studies in a total of 815
et al.,18 Ross et al.19). patients. The meta-­analysis showed a medium efficacy of the treat-
The primary outcome was the incidence of successful proce- ment with CEM (0.52, 95% CI = 0.49–0.55), and a medium-­high ef-
dures; the only outcome reported across all the studies. The sec- fectiveness was determined when the REM was applied (0.85, 95%
ondary outcome was the incidence and type of adverse events. The CI = 0.61–0.96) with high heterogeneity in CEM (I2 = 95.30%) and in
common effect model (CEM) and the random effect model (REM) REM (τ 2 = 4.49). There were overall 44 minor adverse events.
were employed to estimate the overall effect size of the analysis. Midazolam was used in six studies of 166 patients. A medium profi-
ciency of midazolam was displayed when CEM was utilised (0.72; 95%
CI = 0.64–0.78), and a medium-­high when REM was applied (0.84; 95%
2.2 | Risk of bias and certainty assessment CI = 0.31–0.98). High heterogeneity was detected with both models
(I2 = 87.50%; τ2 = 7.52). The studies reported five adverse events.
The risk of bias was assessed by using the NHLBI (National Heart, Dexmedetomidine plus midazolam was employed in three stud-
Lung and Blood Institute) Study Quality Assessment Tool for ies in 432 patients. The meta-­
analysis indicated a high efficacy
Observational Cohort and Cross-­
Sectional Studies available at by combining the two drugs with both models (CEM = 0.94; 95%
https://​w ww.​nhlbi.​nih.​gov/​healt​h-­​topics/​study​-­​quali​t y-­​asses​sment​-­​ CI = 0.91–0.96; REM = 0.96; 95% CI = 0.80–0.99). The heterogeneity
20
tools​, composed of 14 items (Table S1). Five authors independently was high in the CEM (I2 = 91.10%) but low in REM (τ2 = 1.75). Eleven
evaluated the selected articles and classified risk of bias of each item adverse events were documented.
as low, moderate, high or critical. We achieved an overall classifi- Chloral hydrate was used in two studies of 145 children. Medium-­
cation, considering the scale with the following numerical values: high efficacy was identified with both models (CEM = 0.86; 95%
low = 0, moderate = 1, high = 2 and critical = 3. The numerical sum of CI = 0.79–0.90; REM = 0.83; 95% CI = 0.61–0.94), showing a high
each item was provided for the ‘overall’ classification in the 4 areas: heterogeneity with CEM (I2 = 91.20%) and low with REM (τ2 = 0.59).
low = 0–10 (0%–25%), moderate = 11–21 (25%–50%), high = 21–31 Eleven adverse events were registered.
(50%–75%) and critical = 32–42 (75%–100%). Propofol was used in two studies on a total of 159 patients. The
We produced a median of all numerical evaluations for each item, meta-­analysis showed high usefulness of propofol (CEM = 0.99; 95%
and the results were visualised with the ‘robvis’ online tool21 (avail- CI = 0.96–1.00; REM = 0.99; 95% CI = 0.96–1.00) with null heteroge-
able at https://​mcgui​nlu.​shiny​apps.​io/​robvis/​). neity (I2 = 0.00%; τ2 = 0.00). Thirty adverse events were reported.
Certainty was assessed in compliance with the Grading of Considering all the drugs together, procedural sedation was an
Recommendations, Assessment, Development and Evaluations efficacious form of intervention in patients with ASD with an overall
(GRADE) framework. 22 REM of 0.90 (95% CI = 0.77–0.96) and CEM of 0.72 (95% CI = 0.69–
0.74) with few side effects (CEM = 0.08; 95% CI = 0.06–0.09;
REM = 0.04; 95% CI = 0.02–0.09). Summary results are presented in
3 | R E S U LT S Figures 3 and 4, the forest plot and Table 2.
Since the heterogeneity of study design was high, we conducted an
3.1 | Meta-­analysis influence analysis (Figures S1 and S2), but no precise results emerged.
Therefore, the analysis of the outliers was performed automatically by
Data from the 20 included studies are presented in Table 1. Six (30%) the ‘meta’ package function. The only group with outliers was the first
were single-­arm studies of children treated with a sedation agent. one, using dexmedetomidine as the only drug. Kaku et al.23 and Lubisch
In 4 (20%), the same sedative was administered to the children with et al.24 were removed according to the analysis and the new forest
ASD and in a control population. In 9 (45%) studies, children with ASD plot was presented in Figure 5. The CEM and REM resulted increased
 TA B L E 1 Data extracted from the 20 articles selected for the meta-­analysis.

Outcome: the number of

Population (age successful procedure conducted
ZUPIN et al.

mean ± standard Route of on the total number of

Article Type of study deviation) Intervention Drug administration Sedation treatment procedure Adverse effect

Abulebda et al. Observational ASD MRI Dexmedetomidine IV Dexmedetomidine: initial Dexmedetomidine 56 on 56 Apnoea:
(2018)28 retrospective study Dexmedetomidine Propofol 2 μg/kg (10′), then infusion Propofol 49 on 49 1 dexmedetomidine
Two treatments n = 56 (7.3 ± 3.3 years) at 1 μg/kg/h 1 propofol
Propofol n = 46 Propofol: 1 mg/kg then
(8.0 ± 3.9 years) infusion 83 μg/kg/min

Ahmed et al. Observational ASD n = 56 MRI Dexmedetomidine IV Dexmedetomidine: initial ASD None
(2014)25 case–control study (6.1 ± 0.3 years) Dexmedetomidine + 2 μg/kg IV (10′) then Dexmedetomidine 46 on 56
Retrospective Control n = 107 midazolam infusion at 1 μg/kg/h Dexmedetomidine + midazolam
study (5.0 ± 0.2 years) Midazolam: IV 10 on 10
0.06 ± 0.07 mg/kg, oral
0.37 ± 0.18 mg/kg

Brown et al. Observational ASD n = 126 (Median Laceration repair Ketamine IV-­ketamine Ketamine: IV mg/kg Midazolam 59 on 64 Not reported for each drug
(2019)36 retrospective study 7 years, IQR 5–11) (24.6%) midazolam (71.9%) IN— 1.7 ± 0.74; IM mg/kg Ketamine 64 on 64
Two treatments Ketamine n = 64 Incision & midazolam (71.9%) 3.23 ± 0.92 midazolam: IV
Midazolam n = 64 drainage (17.5%) mg/kg 0.18 ± 0.15; IN mg/
Diagnostic kg 0.34 ± 0.15
imaging (14.3%)
Physical
examination
(11.9%)
Other in
emergency
department

Capp et al. Case series ASD n = 7 (2–54 years) Dental procedures Midazolam IM—IV IM 0.2–0.3 mg/kg 4 on 7 None
(2010)37 prospective IV 0.1 mg/kg. local
observational anaesthesia lidocaine 2%
study (0.5–2 mL)

Carlone et al. Case series ASD n = 8 Urgent painless Dexmedetomidine IM IM 4 μg/kg 8 on 8 None
(2019)33 observational (10.25 ± 3.20 years; diagnostic
study 5–14) procedures in
the emergency
department

Kaku et al. Observational ASD n = 34 MRI Dexmedetomidine IV Initial: 2 μg/kg, then mcg/ Dexmedetomidine None
(2023)23 case–control (4.67 ± 2.48 years) Dexmedetomidine + kg/h 13 on 34 ASD
prospective study Control n = 31 thiopentone Dexmedetomidine + thiopentone
(5.83 ± 3.09) 21 on 21

(Continues)
| 2367

16512227, 2024, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/apa.17364 by Cochrane France, Wiley Online Library on [10/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
 TA B L E 1 (Continued)
| 2368

Outcome: the number of

Population (age successful procedure conducted
mean ± standard Route of on the total number of
Article Type of study deviation) Intervention Drug administration Sedation treatment procedure Adverse effect

Kamat et al. Retrospective ASD n = 110 MRI Propofol IV 98% propofol ASD (9.6 mg/ 109 on 110 ASD Serious adverse events: Airway
(2018) 41 case–control study (7.5 years; 5.5–10.3) kg) controls (11 mg/ obstruction 15 and 18 (ASD and
Control n = 110 kg/h); in 6 patients NASD), 1 laryngospasm (NASD)
(7.6 years; 5.3–10.3) used methohexital, Adverse events: (ASD and
dexmedetomidine and NASD)
ketamine. Desaturation 5 and 6
Agitation/delirium 3 and 1
Apnoea >15 5 and 3
Unexpected change in HR or
BP 1 and 1

Kaplan et al. Prospective ASD n = 82 EEG Dexmedetomidine Intranasal Dexmedetomidine 3 μg/kg Dexmedetomidine: 34 on 41 Desaturation: 1
(2022)16 (dexmedetomidine) (Dexmedetomidine Trichlofos sodium dexmedetomidine optional 1.5 μg/kg TFS: 18 on 41 dexmedetomidine 1 TFS,
retrospective mean (TFS) oral TFS TSF initial of 50 mg/kg, bradycardia dexmedetomidine 8
(TFS) case series 70.4 ± 32.4 months) and an optional addition of Vomiting 3 TFS
observational TSF 25 mg/kg of TFS 45 min
study (69.5 ± 31.6 months)

Keidan et al. Retrospective ASD n = 183 (CH EEG CH CH oral or rectal Children 1–6 years old CH 99 on 108 CH: 1 oxygen desaturation
(2014)29 observational 4.8 ± 2 years, DEX dexmedetomidine Dexmedetomidine 50 mg/kg Dexmedetomidine 73 on 84 dexmedetomidine: 7
study 8.2 ± 4.2 years) IV Children >6 years old hypotension and/or
Two treatments were sedated with bradycardia, 2 oxygen
dexmedetomidine 1 μg/kg desaturation
over 5 min (a rate of 12 μg/
kg/h). Repeat infusion of
1 μg/kg (max 4) was given
every 5 min

Li et al. (2019)26 Prospective ASD n = 275 CT and/or ABR Dexmedetomidine IN— IN dexmedetomidine 3 μg/ Dexmedetomidine 89 on 136 Dexmedetomidine: 5
randomised (1–12 years) Dexmedetomidine + dexmedetomidine kg oral midazolam 0.2 mg/ dexmedetomidine + midazolam hypotension
controlled trial midazolam Oral—midazolam kg 116 on 139 Dexmedetomidine—midazolam
Two treatments 10 hypotension

Lubisch et al. Retrospective Neurobehavioral MRI (77.8%) Dexmedetomidine IV/oral Dexmedetomidine: IV Dexmedetomidine: 32 on 315 Not reported for each drug
(2009)24 observational disorders n = 315 Dexmedetomidine + dexmedetomidine 2.6 ± 1.7, oral 3.1 ± 1.2 mg/ Dexmedetomidine + midazolam: 1 airway obstruction
study two Autism 83.1% n = 262 midazolam IV/oral/ kg midazolam: IV 279 on 283 dexmedetomidine + midazolam
treatments (6.8 ± 3.9 years; IN—midazolam 0.08 ± 0.09 mg/kg IV, oral
8 months–24 years) 0.45 ± 0.20 mg/kg, IN
0.30 ± 0.16 mg/kg

Luque et al. Retrospective ASD n = 37 ABR CH Oral CH CH: 50 mg/kg <2 years CH: 25 on 37 CH: 7 vomiting
(2021)32 case series (47 months; 24–96) Dexmedetomidine IN and 75 mg/kg > 2 CH + dexmedetomidine: 26 CH + dexmedetomidine: 4
observational dexmedetomidine Dexmedetomidine: IN on 28 vomiting
study 1–3 μg/kg/dose, if rescue Dexmedetomidine: 7 on 9
of CH 2 μg/kg/dose
ZUPIN et al.

16512227, 2024, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/apa.17364 by Cochrane France, Wiley Online Library on [10/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
 TA B L E 1 (Continued)

Outcome: the number of

Population (age successful procedure conducted
ZUPIN et al.

mean ± standard Route of on the total number of

Article Type of study deviation) Intervention Drug administration Sedation treatment procedure Adverse effect

Mcbride et al. Retrospective Autism/intellectual Ophthalmology Midazolam Oral 0.5 mg/kg 30 on 30 None
(2020)38 case series disability n = 30 exams
observational (70.92 months;
study 32–200)

Metha et al. Prospectively ASD n = 27 (median EEG Clonidine Oral Median 5 mg/kg (2–7 mg/ 25 on 27 2 patients heart rate reduction
(2004)17 case series 6 years; 2.2–16.4) kg) 2 patients asymptomatic
observational systolic blood pressure
study reduction

Pisalchaiyong Prospective, Autism n = 13 Dental treatment Diazepam Oral Diazepam 0.3 mg/kg Diazepam: 10 on 13 None
et al. (2004)18 randomised, (8.68 years; 5.8–14.7) Midazolam Midazolam 0.5 mg/kg Midazolam:13 on 13
double-­blind,
cross-­over trial
Two treatments

Penna et al. Parallel, double-­ ASD n = 64 Intravenous Midazolam Oral Midazolam 0.5 mg/kg Midazolam 10 on 32 Midazolam vs.
(2022)39 blind, controlled, (2–59 years) access Midazolam + Midazolam/Ketamine: midazolam + ketamine 18 on 28 Midazolam + ketamine:
randomised clinical ketamine midazolam dose was Agitation 3–2
trial 0.5 mg/kg ketamine dose Vomiting 0–1
Two treatments 3 mg/kg

Ray et al. Retrospective ASD n = 27 EEG Dexmedetomidine IV oral Initial: oral 3.6 ± 0.8 μg/ 27 on 27 Not separately reported for
(2007)30 case series (5.6 ± 2.4 years; 2–11) kg (2.9–4.4 μg/kg) or IV ASD
observational 2.1 ± 0.8 μg/kg (0.5–3.5 μg/
study kg)
Followed by IV 1.5 ± 0.2 μg
(0.5–2.0 μg/kg)

Ross et al. Retrospective ASD n = 41 MRI Fentanyl + IV Fentanyl 1 μg/kg ASD 41 on 41 None
(2005)27 case–control (32.02 ± 3.6 months) pentobarbital Pentobarbital 1–3 mg/kg
observational Controls 43
study (28.16 ± 6.7 months)

Sahyoun et al. Case series ASD n = 29 Blood draw, Dexmedetomidine IN Dexmedetomidine 4 μg/kg 28 on 29 3 Prolonged sedation
(2023)27 observational (7.2 years; 4.5–11.4) immunisation, 1 Vomiting
study EEG 1 Seizure

Shokri et al. Prospective ASD n = 40 EEG Dexmedetomidine IV Dexmedetomidine initial Dexmedetomidine: 11 on 20 Dexmedetomidine: Bradycardia
(2019)31 randomised parallel (dexmedetomidine Midazolam 1 μg/kg (10′) then, 0.7 μg/ Midazolam: 3 on 20 7
group study 7.94 ± 1.77 years, kg/h Midazolam: Bradycardia 2,
Two treatments Midazolam 8.03 ± 1.7) Midazolam initial 0.05 mg/ agitation 4
kg (2′) eventually plus
0.05 mg/kg

Abbreviations: ABR, auditory brainstem response; CH, chloral hydrate; CT, computerised tomography; EEG, electroencephalogram; IM, intramuscular; IN, intranasal; IV, intravenous; MRI, magnetic
resonance imaging.
| 2369

16512227, 2024, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/apa.17364 by Cochrane France, Wiley Online Library on [10/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
 |

16512227, 2024, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/apa.17364 by Cochrane France, Wiley Online Library on [10/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2370 ZUPIN et al.

F I G U R E 3 Forest plot of the meta-­

analysis conducted using as outcome the
number of the successful procedures
completed. Common and random effect
model results are reported as well as
heterogeneity (I2 and τ2). CI, confidence
interval; Events, number of successful
procedures completed; Total, number of
patients.

(CEM = 0.81; 95% CI = 0.77–0.85; REM = 0.90; 95% CI = 0.76–0.97) and When looking at the routes of administration, oral midazolam
heterogeneity lowered (I2 = 62.70%; τ2 = 2.06). (REM = 0.99, CEM = 0.71) and IV propofol (REM = 0.99, CEM = 0.99)
Dexmedetomidine and dexmedetomidine plus midazolam were had the highest level of efficacy, followed by dexmedetomidine IV
24–26
employed in three studies ; the OR was 0.05, with both models, (REM = 0.84, CEM = 0.44) and intranasal (REM = 0.82, CEM = 0.73)
indicating more efficacy of the dexmedetomidine plus midazolam (Figure S3); with a low adverse event rate (Figures S3–S6).
regimen as compared to dexmedetomidine alone. Nevertheless,
the CEM displayed a limited confidence interval range (0.04–0.08),
while the REM contained 1.00 (0.00–1.59). The comparison be- 3.2 | Risk of bias
tween dexmedetomidine and chloral hydrate did not support the
use of one of the two drugs in terms of sedation effectiveness The risk of bias was assessed using the NHLBI's Study Quality
(Figure 6). Assessment Tool for Observational Cohort and Cross-­
Sectional
 |

16512227, 2024, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/apa.17364 by Cochrane France, Wiley Online Library on [10/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ZUPIN et al. 2371

F I G U R E 4 Forest plot of the meta-­

analysis conducted using as outcome the
number of side effects. Common and
random effect model results are reported
as well as heterogeneity (I2 and τ2). CI,
confidence interval; Events, number of
side effects; Total, number of patients.

(Table S1). Ten studies were classified as low and 10 as moderate conducted so far provides a good indication of the likely beneficial
risk of bias (Figure 7). The main critical item was the blinding of the effect. The likelihood that the effect will be substantially different
clinicians involved in evaluating the outcomes. Most studies lacked from what the study shows is moderate’.
sample size justification, power description or variance and effect es-
timates, increasing the risk of bias.
4 | DISCUSSION

3.3 | Certainty assessment We conducted a meta-­analysis of the available literature on proce-

dural sedation in patients with ASD and found that all the studies de-
Certainty was assessed following the GRADE guidelines (Table 3). scribed the significant efficacy of procedural sedation to complete
The overall judgement was ‘moderate’, such as the ‘research medical procedures in children with ASD, independent of the drugs
 |

16512227, 2024, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/apa.17364 by Cochrane France, Wiley Online Library on [10/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2372 ZUPIN et al.

TA B L E 2 Summary of the meta-­analysis findings.

No of No of No of
Drug studies patients events % CEM (95% CI) REM (95% CI) τ2 I2

Number of Successful sedation/procedures conducted

Dexmedetomidine 12 815 424 52% 0.52 [0.49; 0.55] 0.85 [0.61; 0.96] 4.49 95.30%
Midazolam 6 166 119 71% 0.72 [0.64; 0.78] 0.84 [0.31; 0.98] 7.52 87.50%
Dexmedetomidine + midazolam 3 432 405 94% 0.94 [0.91; 0.96] 0.96 [0.80;0.99] 1.71 91.10%
Chloral hydrate 2 145 124 85% 0.86 [0.79; 0.90] 0.83 [0.61; 0.94] 0.59 91.20%
Propofol 2 159 158 99% 0.99 [0.96; 1.00] 0.99 [0.96; 1.00] 0.00 0.00%
Number of side effects
Dexmedetomidine 11 487 44 9% 0.09 [0.07; 0.12] 0.06 [0.02; 0.16] 2.14 65.60%
Midazolam 5 102 5 5% 0.05 [0.02–0.11] 0.04 [0.01–0.15] 0.20 0.00%
Dexmedetomidine + midazolam 3 434 11 2.5% 0.03 [0.01; 0.05] 0.01 [0.00–0.11] 2.11 76.60%
Chloral hydrate 2 145 11 8% 0.08 [0.04; 0.13] 0.01 [0.00; 0.89] 13.42 0.00%
Propofol 2 156 30 19% 0.19 [0.14; 0.26] 0.09 [0.01; 0.45] 1.71 86.2%

Abbreviations: CEM, common effect model; CI, confidence interval; NR, not reported; REM, random effect model.

F I G U R E 5 Forest plot of the meta-­

analysis conducted using as outcome
the number of side effects and removing
outliers. 23,24 Common and random effect
model results are reported as well as
heterogeneity (I2 and τ2). CI, confidence
interval; Events, number of side effects;
Total, number of patients.

employed. The two most investigated single-­drug regimens were As expected, the few adverse events were mainly linked to hemody-
dexmedetomidine and midazolam, primarily for diagnostic imaging. namics, such as bradycardia and hypotension, with minimal impact
These regimens were safe and effective. on respiration. It is well established that in the setting of dexmede-
Considering the studies as single arm based on the pharmaco- tomidine related adverse effects, interventions are rarely required
logical agents used, propofol had the highest efficacy. However, it and may be strongly influenced by underlying and/or concomitant
was employed only in two studies with numerous adverse events abnormalities, such as myocardiopathies, arrythmia, drug induced
and requires intravenous access which can be difficult to obtain in bradycardia and/or hypotension.34 Considering the route of admin-
these patients, often requiring physical restrains or other interven- istration, the use of the IV or IN route appeared to be similar when
tions such as topical anaesthesia or pre-­medication for anxiolysis. using the REM model. The use of the IN route appeared to be supe-
The second most effective drug was dexmedetomidine used in rior in term of successful procedures with respect to the IV when the
12 of the 20 studies with 815 patients. The third was midazolam and CEM model was applied. This may be due to a bias related to both
the last chloral hydrate. However, midazolam, as a single agent, had the analysis (REM is considered more conservative and more appro-
the highest number of sedation failures. priate to heterogeneous studies with respect to CEM35) and to the
Dexmedetomidine appeared to be a safe alternative for sedation type of procedure. Sedation of short and/or non-­painful procedures,
in infants and children. It could be administrated via different routes such as EKG or EEG, will require a lower sedation level compared to a
such as the intranasal (IN), intramuscular (IM) and intravenous (IV). 27 sedation performed for MRI or a painful procedure, showing a major
In our meta-­analysis, seven studies employed intravenous, 23–25,28–31 effectiveness versus a non-­intravenous approach.34
four intranasal,16,26,27,32 two oral (O)24,30 and one intramuscular.33 Midazolam had mild adverse events, mainly concerning agita-
The studies analysed documented 27 cases of adverse events, spe- tion, but the efficacy was as the lowest among the studies anal-
cifically, one apnoea, one desaturation, 15 bradycardia, five hypo- ysed.18,31,36–39 It was mainly administered by orally (n = 3), while IN
tension, three prolonged sedations, one vomiting and one seizure. (n = 1) and IV routes (n = 1) were less represented.
 |

16512227, 2024, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/apa.17364 by Cochrane France, Wiley Online Library on [10/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ZUPIN et al. 2373

F I G U R E 6 Forest plot of the meta-­analysis conducted on the studies where two different drugs are employed. The outcome is the
number of the successful procedures completed. Common and random effect model results are reported as well as heterogeneity (I2 and τ2).
(A) Dexmedetomidine versus dexmedetomidine plus midazolam. (B) Dexmedetomidine versus choral hydrate. CI, confidence interval; Events,
number of successful procedures completed; OR, Odds Ratio; Total, number of patients.

Chloral hydrate has a history of severe adverse events and fatali- 4.1 | Limitations
ties and its production was suspended in the United States in 201240
and banned in some European countries. Nevertheless, its use in The main limitation of this meta-­analysis was the high frequency
procedural sedation in children with ASD showed medium-­high effi- of observational nature of the studies selected. Eleven were retro-
cacy and few adverse events. 29,32 spective observational studies, including three case–control stud-
Due to its adverse effects and the availability of safer and more ies. Three studies were prospective observational studies with one
effective pharmacological choices, the Chloral hydrate use should case–control study, finally two were case series papers. Only four
be limited to settings without any other possible pharmacological works were randomised prospective trials. However, these studies
option. were too few to conduct a sub-­analysis of the current regimens em-
Six studies investigated the combination of two drugs.19,23–26,39 ployed in the procedural sedation of patients with ASD.
In three papers, 24–26 the same combination, dexmedetomidine plus Another limitation is the route of administration, as most of
midazolam, showed high efficacy and low rate of adverse events. the studies used the IV route. Intravenous access is a procedure
Few studies directly compared different drugs (dexmedetomidine that in this population, as a rule, may require interventions, not
versus dexmedetomidine plus midazolam24–26 and dexmedetomi- only in terms of distraction, but often physical restraint or a pre-­
dine versus choral hydrate29,32) but the meta-­analysis did not indi- medication. When looking at the route of administration, IV, IN or
cate the superiority of one drug over the other; however, a trend O did not appear to be different in term of efficacy with REM, al-
suggested that successful sedation was more likely to occur when though IV appeared to be inferior respect to IN or O, when using
children were sedated with the combination of the dexmedetomi- the CEM. Adverse effects did not vary between the routes of ad-
dine plus midazolam, rather than the dexmedetomidine alone. ministration. Finally, the limited number of studies did not allow to
16512227, 2024, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/apa.17364 by Cochrane France, Wiley Online Library on [10/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
|ZUPIN et al.
2374
 |

16512227, 2024, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/apa.17364 by Cochrane France, Wiley Online Library on [10/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ZUPIN et al. 2375

F I G U R E 7 Risk-­of-­bias visualisation by using the 14 items of the NHLBI (National Heart, Lung and Blood Institute)'s Study Quality
Assessment Tool for Observational Cohort and Cross-­Sectional Studies reported in detail in Table S1. (A) ‘Traffic plot’ of risk of bias' results
in each study for each question and (B) weighted bar plots representing the distribution of risk-­of-­bias evaluation within each of the 14
bias items. Two of the authors evaluated independently the selected articles and they classified each item as low, moderate, high or critical
risk of bias. An overall classification was then performed, considering the scale with the following numerical values: Low = 0, moderate = 1,
high = 2 and critical = 3. The numerical sum of each item was employed for the ‘overall’ classification in the four areas, low = 0–10 (0%–25%),
moderate = 11–21 (25%–50%), high = 21–31 (50%–75%) and critical = 32–42 (75%–100%).

TA B L E 3 GRADE summary of findings table.

Certainty assessment

No. of Other
studies Study design Risk of bias Inconsistency Indirectness Imprecision considerations Impact Certainty

Outcome: number of successful procedures

20 Observational Not serious Serious Not serious Not serious None The procedural ⨁⨁⨁◯
studies due to high sedation is Moderate due
heterogeneity an effective to inconsistency
treatment to
allow the success
of the medical
procedures in
children with ASD
Outcome: number of side effects
20 Observational Serious Serious Not serious Not serious None Low number of ⨁⨁◯◯
studies due to no due to high side effects shows Low due to
reporting of heterogeneity the safety of the risk of bias and
mild severe procedures in inconsistency
effects children with ASD

define whether the likelihood of success of each drug, or combina- the efficacy of most of the regimens already in use, moreover, these
tion of drugs, was proportional to the type of procedure or to the data could support the development of evidence-­based guidelines
route of drug administration performed. for procedural sedation in children with ASD.
In patients with ASD non-­pharmacologic strategies can play a
significant role in a successful preparation for sedation, with a sig- AU T H O R C O N T R I B U T I O N S
nificant impact on quality of care. These strategies should include a Luisa Zupin: Conceptualization; data curation; methodol-
previous knowledge of patients' and parents' fears and preferences, ogy; writing – original draft. Cyril Sahyoun: Writing – review
anticipated knowledge of possible distraction strategies, creation and editing; conceptualization; visualization. Baruch Krauss:
of patient-­centred, adequate and quiet environment, with reduced Conceptualization; visualization; writing – review and edit-
noise and light stimuli. Additionally, desensitisation techniques for ing. Arianna Dagri: Writing – review and editing; data curation.
venipuncture or intranasal administration can be employed to im- Elisabetta Maria Rocco: Data curation; writing – review and
prove the overall experience. editing. Egidio Barbi: Conceptualization; visualization; supervi-
sion; writing – review and editing; methodology. Fulvio Celsi:
Conceptualization; methodology; data curation; visualization;
5 | CO N C LU S I O N writing – review and editing.

Multiple effective drug regimens currently exist for procedural se- AC K N OW L E D G E M E N T S

dation in children with ASD. Considering the traumatic experience The authors thank Martina Bradaschia for the English editing of the
that children with ASD often encounter in their medical encoun- paper. Open access funding provided by BIBLIOSAN.
ters, non-­IV regimens would be preferable with respect to the IN
or oral and should be considered to facilitate venous access when F U N D I N G I N FO R M AT I O N
deeper levels of sedation are required. These alternatives are not This work was supported by the Ministry of Health, Rome, Italy, in
only effective but tend to be more acceptable from the parents' and collaboration with the Institute for Maternal and Child Health IRCCS
patients' perspective. The results of this meta-­analysis highlighted Burlo Garofolo, Trieste, Italy (RC 29/23).
 |

16512227, 2024, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/apa.17364 by Cochrane France, Wiley Online Library on [10/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2376 ZUPIN et al.

C O N FL I C T O F I N T E R E S T S TAT E M E N T 17. Mehta UC, Patel I, Castello FV. EEG sedation for chil-
The authors have no relevant financial or non-­financial interests to dren with autism. J Dev Behav Pediatr. 2004;25(2):102-­ 4.
doi:10.1097/00004703-­200404000-­0 0005
disclose.
18. Pisalchaiyong T, Trairatvorakul C, Jirakijja J, Yuktarnonda W.
Comparison of the effectiveness of oral diazepam and midaz-
ORCID olam for the sedation of autistic patients during dental treatment.
Luisa Zupin https://orcid.org/0000-0001-5886-9129 Pediatr Dent. 2005;27(3):198-­206.
19. Ross AK, Hazlett HC, Garrett NT, Wilkerson C, Piven J. Moderate
Egidio Barbi https://orcid.org/0000-0002-6343-846X
sedation for MRI in young children with autism. Pediatr Radiol.
2005;35(9):867-­71. doi:10.1007/s00247-­0 05-­1499-­2
REFERENCES 20. NHLBI (National Heart, Lung and Blood Institute NH Lung and
1. WHO. WHO autism. 2023. Accessed December 5, 2023. https://​ Blood Institute). Study Quality Assessment Tools – Observational
www.​w ho.​i nt/​n ews-­​r oom/​f act-­​s heets/​d etail/​a utis​m -­​s pect​r um-­​ Cohort and Cross-­ Sectional Studies. 2021. Accessed December
disor​ders 5, 2023. https://​w ww.​nhlbi.​nih.​gov/​healt​h-­​topics/​study​-­​quali​t y-­​
2. Zeidan J, Fombonne E, Scorah J, et al. Global prevalence of au- asses​sment​-­​tools​
tism: a systematic review update. Autism Res. 2022;15(5):778-­90. 21. McGuinness LA, Higgins JPT. Risk-­of-­bias VISualization (robvis):
doi:10.1002/aur.2696 an R package and shiny web app for visualizing risk-­of-­bias as-
3. Povey C. Helping children on the autism spectrum deal with hos- sessments. Res Synth Methods. 2021;12(1):55-­ 61. doi:10.1002/
pital admissions. Arch Dis Child. 2016;101(12):1081. doi:10.1136/ jrsm.1411
archdischild-­2016-­311299 22. The GRADE Working Group. GRADE handbook for grading quality
4. Slifer KJ, Hankinson JC, Zettler MA, et al. Distraction, exposure of evidence and strength of recommendations. 2013. http://​guide​
therapy, counterconditioning, and topical anesthetic for acute pain lined​evelo​pment.​org/​handbook
management during needle sticks in children with intellectual and 23. Kaku SM, Bansal S, Gs Rao U, Bharath RD, Srinath S, Girimaji SC.
developmental disabilities. Clin Pediatr (Phila). 2011;50(8):688-­97. Higher anesthetic dose requirement for sedation in children with
doi:10.1177/0009922811398959 autism spectrum disorder compared to neuro-­ atypical controls
5. Krauss B, Green SM. Procedural sedation and analgesia – a prospective observational study. Res Autism Spectr Disord.
in children. Lancet. 2006;367(9512):766-­ 8 0. doi:10.1016/ 2023;101:102086. doi:10.1016/j.rasd.2022.102086
S0140-­6736(06)68230-­5 24. Lubisch N, Roskos R, Berkenbosch JW. Dexmedetomidine for
6. Kamat PP, Bryan LN, McCracken CE, Simon HK, Berkenbosch JW, procedural sedation in children with autism and other behav-
Grunwell JR. Procedural sedation in children with autism spectrum ior disorders. Pediatr Neurol. 2009;41(2):88-­ 94. doi:10.1016/j.
disorders: a survey of current practice patterns of the society for pediatrneurol.2009.02.006
pediatric sedation members. Paediatr Anaesth. 2018;28(6):552-­7. 25. Ahmed SS, Unland T, Slaven JE, Nitu ME, Rigby MR. Successful use
doi:10.1111/pan.13387 of intravenous Dexmedetomidine for magnetic resonance imaging
7. Rstudio. Rstudio: IDE Integrated Development Environment for R. sedation in autistic children. South Med J. 2014;107(9):559-­ 6 4.
2023. http://​w ww.​rstud​io.​com/​ doi:10.14423/SMJ.0000000000000160
8. R Core Team. R: A Language and Environment for Statistical 26. Li BL, Yuen VM, Zhang N, et al. A comparison of intranasal
Computing. R Foundation for Statistical Computing; 2023. http://​ Dexmedetomidine and Dexmedetomidine plus buccal midaz-
www.​R-­​proje​c t.​org olam for non-­ painful procedural sedation in children with au-
9. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 state- tism. J Autism Dev Disord. 2019;49(9):3798-­ 8 06. doi:10.1007/
ment: an updated guideline for reporting systematic reviews. BMJ. s10803-­019-­0 4095-­w
2021;372:n71. doi:10.1136/bmj.n71 27. Sahyoun C, Krauss B, Bevacqua M, Antonsen A, Jardinier L, Barbi
10. Grames EM, Stillman AN, Tingley MW, Elphick CS. An automated E. Safety and efficacy associated with a family-­centered proce-
approach to identifying search terms for systematic reviews dural sedation protocol for children with autism Spectrum disorder
using keyword co-­ occurrence networks. Methods Ecol Evol. or developmental delay. JAMA Netw Open. 2023;6(5):e2315974.
2019;10(10):1645-­54. doi:10.1111/2041-­210X.13268 doi:10.1001/jamanetworkopen.2023.15974
11. Roy Rosenzweig Center for History and New Media. Zotero 28. Abulebda K, Louer R, Lutfi R, Ahmed SS. A comparison of safety and
[Computer software]. 2016 www.​zotero.​org/​download efficacy of Dexmedetomidine and Propofol in children with autism
12. Westgate MJ. Revtools: an R package to support article screening and autism Spectrum disorders undergoing magnetic resonance
for evidence synthesis. Res Synth Methods. 2019;10(4):606-­ 14. imaging. J Autism Dev Disord. 2018;48(9):3127-­32. doi:10.1007/
doi:10.1002/jrsm.1374 s10803- ­018-­3582-­1
13. Haddaway NR, Page MJ, Pritchard CC, McGuinness LA. 29. Keidan I, Ben-­Menachem E, Tzadok M, Ben-­Zeev B, Berkenstadt
PRISMA2020: An R package and Shiny app for producing PRISMA H. Electroencephalography for children with autistic spectrum
2020-­ compliant flow diagrams, with interactivity for optimised disorder: a sedation protocol. Pediatr Anesth. 2015;25(2):200-­5.
digital transparency and Open Synthesis. Campbell Syst Rev. doi:10.1111/pan.12510
2022;18(2):e1230. doi:10.1002/cl2.1230 30. Ray T, Tobias JD. Dexmedetomidine for sedation during elec-
14. Balduzzi S, Rücker G, Schwarzer G. How to perform a meta-­analysis troencephalographic analysis in children with autism, pervasive
with R: a practical tutorial. Evid Based Ment Health. 2019;22(4):153-­ developmental disorders, and seizure disorders. J Clin Anesth.
60. doi:10.1136/ebmental-­2019-­3 00117 2008;20(5):364-­8. doi:10.1016/j.jclinane.2008.03.004
15. Harrer M, Cuijpers P, Furukawa T, Ebert DD. dmetar: Companion R 31. Shokri H, Kasem AA. Dexmedetomidine versus midazolam seda-
Package For The Guide “Doing Meta-­Analysis in R”. 2019 R package tion for autistic children undergoing electroencephalogram: a pro-
version 0.1.0. http://​dmetar.​prote​c tlab.​org/​ spective randomized trial. Ain-­Shams J Anesthesiol. 2019;11(1):33.
16. Kaplan E, Shifeldrim A, Kraus D, et al. Intranasal dexmedetomidine doi:10.1186/s42077-­019-­0 047-­5
vs oral triclofos sodium for sedation of children with autism under- 32. Luque CG, Atkins-­ L abelle C, Pauwels J, Costello R, Kozak FK,
going electroencephalograms. Eur J Paediatr Neurol. 2022;37:19-­ Chadha NK. Intranasal Dexmedetomidine increases the suc-
24. doi:10.1016/j.ejpn.2022.01.005 cessful sedation of children with autism for out-­patient auditory
 |

16512227, 2024, 11, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/apa.17364 by Cochrane France, Wiley Online Library on [10/02/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
ZUPIN et al. 2377

brainstem response hearing tests. Int J Pediatr Otorhinolaryngol. ketamine plus midazolam as preanesthetic medication in autism
2021;151:110945. doi:10.1016/j.ijporl.2021.110945 spectrum disorder: double-­blind randomized clinical trial. Braz J
33. Carlone G, Trombetta A, Amoroso S, Poropat F, Barbi E, Cozzi G. Anesth. 2023;73(3):283-­90. doi:10.1016/j.bjane.2022.09.003
Intramuscular Dexmedetomidine, a feasible option for children 40. Farr KM, Moffett BS, Jones JL, Rogers AP, Chumpitazi CE.
with autism Spectrum disorders needing urgent procedural se- Chloral hydrate sedation in a Dexmedetomidine era. Hosp Pharm.
dation. Pediatr Emerg Care. 2019;35(6):e116-­ e117. doi:10.1097/ 2020;55(4):236-­9. doi:10.1177/0018578719836639
PEC.0000000000001776 41. Kamat PP, Karaga MK, Wisniewski BL, et al. Outpatient procedural
34. Mahmoud M, Barbi E, Mason KP. Dexmedetomidine: What's new sedation of patients with autism Spectrum disorders for magnetic
for pediatrics? A narrative review. J Clin Med. 2020;9(9):2724. resonance imaging of the brain using Propofol. J Child Neurol.
doi:10.3390/jcm9092724 2018;33(5):313-­9. doi:10.1177/0883073817753908
35. Nikolakopoulou A, Mavridis D, Salanti G. How to interpret meta-­
analysis models: fixed effect and random effects meta-­analyses. Evid
Based Ment Health. 2014;17(2):64. doi:10.1136/eb-­2014-­101794 S U P P O R T I N G I N FO R M AT I O N
36. Brown JJ, Gray JM, Roback MG, Sethuraman U, Farooqi A, Additional supporting information can be found online in the
Kannikeswaran N. Procedural sedation in children with autism Supporting Information section at the end of this article.
spectrum disorders in the emergency department. Am J Emerg
Med. 2019;37(8):1404-­8. doi:10.1016/j.ajem.2018.10.025
37. Capp PL, de Faria ME, Siqueira SR, Cillo MT, Prado EG, de Siqueira JT.
Special care dentistry: midazolam conscious sedation for patients
with neurological diseases. Eur J Paediatr Dent. 2010;11(4):162-­4. How to cite this article: Zupin L, Sahyoun C, Krauss B, Dagri A,
38. McBride GR, Stephenson KAJ, Comer G, Flanagan O. The use of Rocco EM, Barbi E, et al. Effectiveness of pharmacological
Oral midazolam to facilitate the ophthalmic examination of children procedural sedation in children with autism spectrum disorder:
with autism and developmental disorders. J Autism Dev Disord.
A systematic review and meta-­analysis. Acta Paediatr.
2021;51(5):1678-­82. doi:10.1007/s10803-­020-­0 4658-­2
39. Penna HDM, Paiva APM, Romano AJM, Alves RL, Nascimento Junior 2024;113:2363–2377. https://doi.org/10.1111/apa.17364
PD, Módolo NSP. Comparison between oral midazolam versus oral