Introduction

Autism spectrum disorder (ASD), referred to as autism in this study, is a complex neurodevelopmental condition that typically manifests in early childhood. Globally, the prevalence of autism is approximately 0.76%, with reported rates in China ranging between 0.3% and 1%1,2. Autism is characterized by deficits in social communication, restricted interests, repetitive behaviors, and various challenges affecting learning, communication, and daily life3. Despite extensive research, the exact etiology of autism remains unclear, although it is widely believed to involve a combination of genetic, environmental, and neurobiological factors4.

Functional magnetic resonance imaging (fMRI) has become an invaluable tool for exploring the neural mechanisms underlying autism-related functional abnormalities5. However, performing fMRI in children with autism presents unique challenges. These children often face heightened anxiety, sensory sensitivities, and difficulty following behavioral instructions, such as remaining still during scanning6. Movement during imaging can compromise data quality, making sedation or anesthesia necessary to ensure successful examinations. While sedation can help mitigate these challenges, existing research on sedation in children with autism is limited, primarily consisting of case reports and small observational studies7,8,9. This lack of standardized guidelines underscores the need for evidence-based protocols tailored to this population.

Propofol, a widely used anesthetic agent, is favored for its rapid onset, predictable recovery profile, and suitability for target-controlled infusion (TCI)10. However, the optimal propofol concentration for achieving effective sedation with minimal risks in children with autism undergoing fMRI remains unclear. To address this gap, our study evaluated the safety, efficacy, and hemodynamic stability of propofol-induced intravenous deep sedation in 120 preschool children with autism. By comparing different TCI concentration levels, we aimed to provide insights into the best sedation practices for this vulnerable population, contributing to improved management strategies in pediatric neuroimaging.

Materials and methods

Ethical considerations

The study protocol received approval from the Ethics Committee of the Children’s Hospital of Chongqing Medical University (IRB #139/2021, approved on April 27, 2021) and was prospectively registered in the Chinese Clinical Trial Registry (Registration No.: ChiCTR2100050071, registered on August 16, 2021). Enrollment of the first participant began on September 6, 2021. Conducted at the Children’s Hospital of Chongqing Medical University, the study adhered to the ethical standards outlined in the Helsinki Declaration. Written informed consent was obtained from the parents or legal guardians of all patients before enrollment.

Subjects

From September 2021 to July 2023, we enrolled children with ASD from our hospital’s neurology and pediatrics clinics who required fMRI due to language or behavioral issues. Inclusion criteria were a diagnosis of autism according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5), age between 2 and 6 years, right-handedness, no significant findings on routine MRI, and American Society of Anesthesiologists (ASA) physical status I–II. Exclusion criteria included ASD with fragile X syndrome, childhood disintegrative disorder, or Rett syndrome; significant head trauma or other neurological conditions; and lack of parental consent.

Anesthesia management and data collection

All children were required to fast for 6–8 h and refrain from drinking water for 2 h before the examination. Upon arrival at the observation room, two attending anesthesiologists established a peripheral venous access and collected baseline information (gender, age, weight, body mass index (BMI), ASA classification). Prior to the examination, one anesthesiologist administered an intravenous induction of propofol at a dose of 3–4 mg/kg. The examination began when the bispectral index (BIS) reached 50–70 and the Ramsay sedation score (Table S1) was ≥ 5, with vital signs continuously monitored11,12. A target-controlled infusion pump (TCI, TCI-V, VERYARK, Guangxi) was placed outside the MRI room in the observation area, and the infusion was connected to the child via an extended infusion line. The child’s age, height, and weight were input into the system, and the Paedfusor model was used to set the target-controlled infusion of propofol to achieve an effect-site concentration. Children were divided into two groups based on the target effect-site concentration of propofol: the 2.0 µg/mL group (60 children) and the 2.5 µg/mL group (60 children).

During the examination, MR-compatible monitoring equipment was used to monitor oxygen saturation (SpO2), heart rate (HR), respiratory rate (RR), end-tidal carbon dioxide (ETCO2), mean systolic blood pressure (mSBP), and mean diastolic blood pressure (mDBP). ETCO2 was continuously monitored via a nasal cannula-type CO2 monitoring tube.2An ETCO2 level > 50 mmHg indicated hypercapnia, while RR falling 20% below baseline signaled respiratory depression. Upon observing these conditions, drug administration was immediately halted, and jaw-thrust maneuver was initiated to provide assisted ventilation. Vital signs, including SpO2, HR, RR, ETCO2, mSBP, mDBP, and Ramsay sedation scores, were recorded every 5 min. If any movement or a Ramsay sedation score below 5 occurred, an additional dose of propofol (0.5–1 mg/kg) was administered13. Another anesthesiologist was responsible for monitoring and recording the child’s vital signs, anesthesia time, examination duration, recovery time, maintenance dose of propofol, adverse reactions, and movement status. Following the examination, the child was transferred to the recovery area for emergence from anesthesia and was only discharged when the Sedation Agitation Scale (SAS, Table S2) score reached 3–4 points under parental supervision14.

Outcomes

The primary outcome measure of the study was the proportion of first successful sedation, defined as the ratio of cases in which no additional propofol was required during the examination to the total number of cases. Successful sedation was defined as obtaining high-quality MRI images after sedation. The secondary outcome measures included vital signs during the examination (HR, RR, mSBP, mDBP, SpO2, ETCO2), incidence of adverse events (tachycardia, bradycardia, hypotension, hypertension, hypoxemia, hypercapnia, respiratory depression), and sedation parameters (BIS, propofol dose, anesthesia duration, examination duration, recovery time, SAS Score, Ramsay Sedation Score).

Statistical analysis

Randomization was performed using computer-generated random number sequences with allocation concealment achieved through opaque, sealed envelopes. Participants were allocated in a 1:1 ratio to either the 2.0 µg/mL group or the 2.5 µg/mL group. Continuous variables are presented as mean ± standard deviation (SD) or median [IQR], while categorical variables are presented as numbers (%). The normality of the data was assessed using the Shapiro-Wilk test, and based on the results, either independent t-tests or Mann-Whitney U tests were used for analysis. Proportions were compared using chi-square tests or Fisher’s exact test. All p-values were two-sided, and p < 0.05 was considered statistically significant. All analyses were performed using R.

Results

Demographic characteristics

A total of 125 children with ASD were initially recruited for this study. Due to intracranial structural abnormalities in 4 children and refusal to participate by 1 family, 120 children were ultimately included, with the enrollment process shown in Fig. 1. Based on the TCI effect-site concentration settings, they were divided into two groups (2.0 group and 2.5 group), with 60 children in each group. There were no statistically significant differences between the two groups in terms of gender, age, weight, height, BMI, or ASA classification (Table 1).

Fig. 1
Fig. 1
Full size image

Flow diagram of the inclusion process for children with autism spectrum disorder (ASD).

Table 1 Patients’ demographics and characteristics.
Full size table

Primary outcome measures

Regarding the primary outcome, the first-time sedation success rate was 91.67% (n = 55) in the 2.0 group and 96.67% (n = 58) in the 2.5 group, with no statistically significant difference between the two groups (p = 0.439). The detailed results are presented in Table 2.

Table 2 Comparison of vital signs and adverse events between two groups.
Full size table

Secondary outcome measures

The mean systolic blood pressure in the 2.5 group was lower than that in the 2.0 group (2.0 group vs. 2.5 group; 96.11 [94.25,102] vs. 93.54 [87.81,96.92]; median [IQR]; p = 0.001). The trend charts of vital signs for both groups are shown in Fig. 2. There was no statistically significant difference in the incidence of adverse events between the two groups, although the frequency of respiratory and circulatory depression was lower in the 2.0 group. The 2.0 group demonstrated a higher mean BIS during sedation (59.92 ± 3.52 vs. 56.21 ± 3.58; mean ± SD; p < 0.001), required a lower maintenance dose of propofol (10.6 [9.39, 12.28] vs. 13.26 [12.07, 14.74]; median [IQR]; p < 0.001), had a shorter recovery time (22 [15.75, 25] vs. 24 [18, 31.25]; median [IQR]; p = 0.042), and experienced a smoother emergence profile. Detailed data are presented in Tables 2 and 3.

Fig. 2
Fig. 2
Full size image

Trend chart of vital sign changes in children with autism spectrum disorder during functional magnetic resonance imaging (fMRI). (a) Heart Rate (HR); (b) Mean Arterial Pressure (MAP); (c) Respiratory Rate (RR); (d) Oxygen Saturation (SpO2); (e) End-Tidal Carbon Dioxide (ETCO2); (f) Bispectral Index (BIS).

Table 3 Comparison of sedation parameters between two groups.
Full size table

Discussion

This study systematically compared the safety and efficacy of two different TCI concentrations of propofol for sedation during cranial fMRI in children with ASD. To the best of our knowledge, this is the first prospective study to directly compare effect-site concentrations of 2.0 µg/mL versus 2.5 µg/mL propofol for fMRI sedation in children with ASD.

Although fMRI offers a unique perspective for elucidating the neural mechanisms of ASD15,16, its application in the pediatric population is severely limited by challenges such as loud noise, prolonged scanning times, and the confined environment of the procedure6. Sedation is crucial for the successful completion of the examination; however, no standardized sedation protocol currently exists for this specific scenario. Previous research has indicated that sedative agents may interfere with the blood oxygen level-dependent (BOLD) signal through effects on neurovascular coupling, thereby constraining the image quality and interpretive validity of fMRI17. Consequently, exploring sedation strategies that simultaneously ensure high-quality imaging and patient safety is of paramount importance. Several prior studies provide a rationale for the use of propofol in this field: Bernal et al. demonstrated its position as the sedative of choice for pediatric fMRI17; Zhou et al. confirmed the efficacy and recovery advantages of its use, both alone and in combination with dexmedetomidine18; particularly importantly, Tang et al. found that while propofol suppresses responses to auditory stimuli in certain cortical areas, its impact on the primary auditory cortex is limited19, preserving the potential for detecting key neural activity signals in fMRI studies. Based on this evidence, this study focused on the refined adjustment of propofol TCI concentrations, aiming to provide an optimized sedation option for clinical practice.

In this study, we used propofol for induction and maintained sedation with TCI. TCI ensures stable drug concentrations at the effect site or in plasma, minimizing fluctuations and maintaining consistent sedation depth20. The effective TCI concentration required for deep sedation typically ranges from 2.0 to 3.0 µg/mL21. Based on this concentration range, the study further compared the clinical efficacy of two target concentrations (2.0 µg/mL vs. 2.5 µg/mL). Our findings indicate that both TCI concentrations of propofol provided effective and reliable sedation for fMRI. Objective monitoring via BIS confirmed a dose-dependent sedative effect, with the lower concentration group maintaining a significantly lighter yet adequate plane of sedation. This electrophysiological profile aligns with the superior clinical outcomes observed in the 2.0 µg/mL group, which included faster recovery times and a lower total propofol requirement. Furthermore, a trend toward a higher incidence of self-limiting respiratory events was noted at the higher concentration, reinforcing the dose-dependent nature of propofol’s side effects. This is consistent with the findings of Dueck et al.22, who reported dose-dependent attenuation of auditory cortex activation, highlighting propofol’s broader influence on central processing. Therefore, by integrating the BIS-correlated sedation depth with the enhanced recovery and safety profile, we recommend a TCI concentration of 2.0 µg/mL as the optimal balance for preserving BOLD signal integrity while minimizing patient risk.

In the 2.5 µg/mL group, one child with ASD exhibited emergence agitation during recovery (SAS score: 7), accompanied by aggressive behavior. Intranasal dexmedetomidine (3 µg/kg) was administered, leading to resolution without further complications. Martini noted that preschool children are at higher risk for emergence agitation, particularly in unfamiliar environments23. In our study, ASD children experienced heightened difficulty being separated from parents in such settings, increasing their susceptibility to agitation during recovery. Special attention should be given to this risk when sedating ASD children. Moreover, the sedation needs of ASD children may vary significantly due to the impact of ASD on the central nervous system. Elmorsy found that ASD may alter sedative drug requirements24. In our study, one child with a Childhood Autism Rating Scale (CARS) score of 36 (severe ASD) required a significantly higher propofol dose. CARS scores can be used to assess ASD severity and guide individualized sedation strategies25,26. Tailoring protocols based on ASD severity may improve sedation outcomes, enhance MRI success rates, and reduce the risk of adverse effects, such as respiratory depression (e.g., hypoxemia and hypercapnia).

Despite offering valuable insights, our study has limitations. The small sample size and single-center design may limit the generalizability of the findings. Future multicenter, randomized controlled trials are needed to validate our results. Additionally, while this study focused on propofol monotherapy, the sedation needs of ASD children may vary. Future research should explore the efficacy of combining propofol with other sedatives, particularly for mitigating emergence agitation and other potential adverse effects.

Conclusion

This is the first study to systematically compare two propofol target-controlled infusion concentrations (2.0 µg/mL vs. 2.5 µg/mL) for efficacy and safety during fMRI in children with autism spectrum disorder. By jointly assessing scan success, emergence time, drug consumption and adverse events, we demonstrate that the lower concentration achieves highly effective sedation while allowing faster recovery, lower doses and a better safety profile, providing direct evidence for individualized sedation in this special population.