Safety of a higher thiamylal induction dose in pediatric sedation for magnetic resonance imaging scans

Abstract

Background

Previous studies have shown that 3.6 (2.8–4.0) mg/kg of intravenous thiamylal is effective and safe for pediatric sedation during magnetic resonance imaging (MRI) scans, although its success rate (96.6%) is slightly lower than that via propofol (99%). This retrospective study aimed to evaluate whether a higher induction dose of thiamylal can reduce MRI scan interruptions without increasing the incidence of adverse events.

Methods

This study included patients aged < 18 years who underwent MRI under thiamylal sedation between January 2021 and December 2022. Intravenous atropine (0.01 mg/kg) and midazolam (0.08 mg/kg) were administered as premedications. An induction intravenous dose of 2–5 mg/kg thiamylal was administered, with additional doses administered as needed. The induction dose was defined as the total amount of thiamylal administered before the scan. Once adequate sedation was achieved, MRI was performed. Any additional doses of thiamylal required because of scan interruption were recorded as rescue doses. Patients who received an induction thiamylal dose of < 3.6 mg/kg were categorized into the lower induction dose group. The primary endpoint was the rate of MRI interruptions. Secondary endpoints included adverse events related to sedation, total thiamylal dose, and total MRI scanning time.

Results

A total of 566 patients, with a median age of 56.4 months (range: 24.0–79.2 months) and median weight of 16.0 kg (range: 10.3–21.0 kg), were included in the analysis. The median induction and total dose of thiamylal per body weight were 4.7 (3.8–5.8) mg/kg and 6.0 (4.6–8.1) mg/kg, respectively. No respiratory or other adverse events were recorded. For 30- to 60-min MRI scans, the higher induction dose group had a lower scan interruption rate, and a lower total thiamylal dose was required than that for the lower induction dose group.

Conclusion

A higher induction dose of thiamylal (5.4 ± 1.6 mg/kg) is safe and is associated with a reduced scan interruption rate for 30- to 60-min MRI scans, as well as a lower total thiamylal dose requirement. However, thiamylal is not an ideal sedative agent for MRI scans longer than 1 h.

Background

Magnetic resonance imaging (MRI) is a noninvasive but time-consuming imaging modality that requires patients to remain still to obtain high-quality images. Sedation is often necessary to maintain immobility in certain populations, including those with claustrophobia, involuntary movement disorders, seizures, and pain, especially in young children. Compared to adult sedation, pediatric sedation presents unique challenges and risks owing to more difficult intravenous access, lower oxygen reserves, and increased oral secretions after crying. Although various strategies exist for conducting MRI in pediatric patients [1], sedation remains unavoidable in certain groups such as children with intellectual disabilities, seizure disorders, or chronic pain.

Understanding the pharmacological properties of sedatives is essential for achieving effective sedation, ensuring high procedural success rates, and supporting the safe and timely discharge of children undergoing outpatient MRI. Prior studies have shown that intravenous thiamylal at a dose of 3.6 (2.8–4.0) mg/kg is effective and safe, though its success rate (96.6%) is slightly lower than that of propofol (99%) MRI sedation protocol [2]. In our pediatric MRI sedation protocol, if patient movement causes scan interruption, the anesthesiologist administers an additional rescue dose of sedation. However, patient movement not only necessitates a rescue dose, but also can disrupt the MRI sequence, prolonging both the MRI and total sedation times.

A 2020 review article [3] highlighted that to ensure complete immobility for up to 1 h in pediatric MRI, deep sedation or general anesthesia is often required. Moreover, in Nishiyama’s study [4], thiamylal (3.1 ± 1.2 mg/kg) with midazolam (0.075 mg/kg) provided stable hemodynamics for intubation in adults—a midazolam dose similar to the premedication used in our MRI sedation protocol.

We conducted this retrospective study to identify the optimal induction dose of thiamylal for pediatric MRI to minimize scan interruptions. We hypothesized that a higher induction dose of thiamylal could reduce the number of MRI scan interruptions without increasing the incidence of adverse events.

Methods

Patient selection

This study was approved by the Institutional Review Board of the Taipei Veterans General Hospital, Taipei, Taiwan (IRB-TPEVGH no. 2024-08-018AC). We carefully reviewed the electronic medical records of patients who underwent MRI under sedation between January 2021 and December 2022. Patients aged > 18 years, those treated with sedation or rescue agents other than thiamylal, and those with missing data were excluded (Fig. 1).

Fig. 1.

Patient selection flow chart

Anesthesia management

Oral intake of clear liquid, breast milk, or other nutrients was restricted to 2, 4, or 6 h prior to sedation. There were no restrictions on the sleep duration. Intravenous atropine (0.01 mg/kg) and midazolam (0.08 mg/kg) were administered as premedications in the waiting room after peripheral vein access was established. After the patients were transported to the MRI scanning bed, they were sedated with intravenous thiamylal (2–5 mg/kg), and an additional dose of thiamylal (1–2 mg/kg) was administered by the anesthesiologist in charge. MRI was performed after achieving adequate sedation. When MRI scanning could not be continued owing to inadequate sedation, it was temporarily discontinued, and a rescue dose was administered. One radiologist, who was not involved in the sedation process, evaluated the MRI quality. If any movement affected the image quality, the MR series was repeated after adequate sedation. However, the maximum total thiamylal dose was not set.

Safety management was implemented, including observation during and after sedation and emergency care. Each patient received routine oxygen via a nasal cannula at 3–5 L/min. The anesthesiologist performed the chin-lift position according to the patient’s clinical situation. Body temperature, heart rate, respiratory rate, blood pressure, and transcutaneous oxygen saturation were measured before and after MRI. Heart rate, respiratory rate, transcutaneous oxygen saturation, and partial pressure of end-tidal carbon dioxide were observed and recorded every 5 min during MRI. The induction and total doses of thiamylal, time of administration, MRI start time, number of scan interruptions, MRI end time, time of confirmation of the awake state, and adverse events during the periprocedural period were also recorded. After MRI, the patients were transferred to the observation room in the pediatric outpatient department or ward. Observation and monitoring continued until the patients were awake. Awake state was defined as meeting the discharge criteria of the American Academy of Pediatrics guidelines [5].

Data collection and endpoints

Patient records, including demographic data such as age, height, weight, past medical history, medication dose, image type, and procedure time, were collected from the electronic anesthesia records. After data collection, random samples of the collected data were examined to ensure data quality. Patients were divided into two groups according to the induction dose of thiamylal. Patients who received a thiamylal dose of < 3.6 mg/kg were assigned to the lower induction dose group; the remaining patients were assigned to the higher induction dose group. The primary endpoint was the rate of MRI scan interruption, and secondary endpoints included adverse events related to sedation, total thiamylal dose, and total MRI scan time.

Statistical analysis

The study data are presented as the means ± standard deviations, medians (interquartile ranges [IQRs], 25th–75th percentiles), and frequencies (%). We compared patient characteristics between the two groups using the Student’s t-test, Mann–Whitney U test, or chi-squared test and analysis of variance, as appropriate. Factors associated with an additional rescue dose of thiamylal were analyzed using a logistic regression model. Statistical significance was set at P < 0.05. All analyses were performed using IBM SPSS Statistics for Windows (version 25.0; IBM Corp., Armonk, NY, USA).

Results

Patient characteristics

During the study period, 1034 patients received sedated MRI scans. After excluding 480 patients who met the exclusion criteria, 564 patients were included in the analysis (Fig. 1). The median age and weight of the patients were 55.8 months (24.1–79.1 months) and 16.0 kg (10.3–21.0 kg), respectively. The majority of the scans were brain MRIs (n = 374, 66.3%), followed by brain with whole spine (n = 74, 13.1%) and spine scans (n = 64, 11.3%).

The overall median induction dose of thiamylal per body weight was 4.7 mg/kg (IQR: 3.8–5.8 mg/kg), and the total thiamylal dose per body weight was 6.0 mg/kg (IQR: 4.6–8.1 mg/kg). No respiratory or other adverse events, such as circulatory collapse, abdominal symptoms, allergic reactions, asthma attacks, or drug extravasation, were observed.

A total of 277 patients (49.1%) underwent MRI without interruption after thiamylal induction. However, 176 patients (31.2%) experienced one interruption and 69 patients (12.2%) experienced two interruptions because of patient movement. Furthermore, one scan had nine interruptions because of patient movement. The median duration of the MRI procedure was 60 min (IQR: 45–60 min) (Table 1).

Table 1.

Patient characteristics

	n = 564
Male (%)	311 (55.1)
Age (month)	55.8 (24.1–79.1)
Age < 1 year	71 (12.6)
Weight (kg)	16.0 (10.3–21.0)
ASA-PS
I	110 (19.5)
II	419 (74.3)
III	35 (6.2)
Image type
Brain	375 (66.5)
Brain + Spine	74 (13.1)
Spine	64 (11.3)
Extremity	19 (3.4)
Thoracic	2 (0.4)
Abdomen	12 (2.2)
Orbits	13 (2.3)
Others	5 (1.0)
Induction thiamylal dose (mg/kg)	4.7 (3.8–5.8)
Total thiamylal dose (mg/kg)	6.0 (4.6–8.1)
Scan interruption	287 (50.9)
Times of scan interruption
1	176 (31.2)
2	69 (12.2)
3	25 (4.4)
> 3 times	17 (3.0)
Total scan time (min)	60 (45–60)

Data are presented as the n (%) or median (interquartile range)

ASA-PS: American Society of Anesthesiologists Physical Status

Adverse events

No adverse events were recorded during the study period.

Grouping as per thiamylal induction dose

In total, 111 patients were included in the lower induction thiamylal dose group (< 3.6 mg/kg). The patients in the lower induction thiamylal dose group were older (73.2 ± 40 vs. 51.4 ± 33.6 months, P < 0.001), heavier (20.5 ± 8.3 vs. 15.3 ± 6.6 kg, P < 0.001), received a lower mean induction thiamylal dose (2.8 ± 0.6 vs. 5.5 ± 1.6 mg/kg, P < 0.001), received a lower total thiamylal dose (4.6 ± 2.2 vs. 7.2 ± 3.3 mg/kg, P < 0.001), had a similar total scanning time (54.7 ± 15.9 vs. 55.9 ± 20.8 min, P = 0.50), and had a similar scan interruption rate (58.6% vs. 49%, P = 0.073) compared with those in the higher induction thiamylal dose group (Table 2).

Table 2.

Patients grouped according to the thiamylal induction dose

Group	Lower induction dose group	Higher induction dose group	P value
Age (month)	73.2 ± 40	51.4 ± 33.6	< 0.001
Weight (kg)	20.5 ± 8.3	15.3 ± 6.6	< 0.001
Induction thiamylal dose (mg/kg)	2.8 ± 0.6	5.5 ± 1.6	< 0.001
Total thiamylal dose (mg/kg)	4.6 ± 2.2	7.2 ± 3.3	< 0.001
Scan interruption	65 (58.6)	222 (49.0)	0.07
Total scan time (min)	54.7 ± 15.9	55.9 ± 20.8	0.50

Data are presented as means ± standard deviations or n (%)

Patients who received < 3.6 mg/kg induction dose of thiamylal were included in the lower induction dose group; remaining patients were included in the higher induction dose group

Grouping by total MRI scan time

The number of patients who underwent MRI within 30 min, 30–60 min, and more than 1 h was 71 (12.6%), 409 (72.5%), and 84 (14.9%), respectively. The mean induction thiamylal dose was similar across the three groups (5.2 ± 1.9, 4.9 ± 1.8, and 5.1 ± 1.8 mg/kg, respectively; P = 0.22). The mean total thiamylal dose (5.3 ± 1.9, 6.3 ± 2.6, and 9.9 ± 4.8 mg/kg) was significantly greater in the more than 1 h group (P < 0.001), but there was no difference between the other two groups (P = 0.05). The scan interruption rates (7.0%, 49.6%, and 94%, respectively; P < 0.001) were significantly different among the three groups (Table 3).

Table 3.

Patients grouped as per the total magnetic resonance image scanning time Note that 94% of magnetic resonance imaging scan for more than 1 h experienced scan interruption

Scanning time (min)	< 30 (n = 71)	30 to 60 (n = 409)	> 60 (n = 84)	P value
Induction thiamylal dose (mg/kg)	5.2 ± 1.9	4.9 ± 1.8	5.1 ± 1.8	0.22
Total thiamylal dose (mg/kg)	5.3 ± 1.9	6.3 ± 2.6	9.9 ± 4.8	< 0.001
Scan interruption	5 (7.0)	203 (49.6)	79 (94)	< 0.001

Data are presented as means ± standard deviations or n (%)

Analysis after prolonged MRI scan

There were 84 patients who received MRI for more than 1 h. These patients were excluded from the study. The patients in the lower induction thiamylal dose group were also older (71.5 ± 37.4 vs. 52.6 ± 33.8 months, P < 0.001), heavier (20.1 ± 7.7 vs. 15.5 ± 6.7 kg, P < 0.001), received a lower mean induction thiamylal dose (2.8 ± 0.6 vs. 5.4 ± 1.6 mg/kg, P < 0.001), and received a lower total thiamylal dose (4.2 ± 1.9 vs. 6.6 ± 2.5 mg/kg, P < 0.001), with a similar total scanning time (50.6 ± 10.2 vs. 49.2 ± 10.8 mg/kg, P = 0.23); however, a higher frequency of scan interruption (55% vs. 40%, P < 0.05) was noted than for those in the higher induction thiamylal dose group.

Factors related to scan interruption

Patients younger than 1 year (P < 0.05), those with a lower induction thiamylal dose (P < 0.01), and those with a longer total MRI scanning time (P < 0.001) experienced scan interruption (Table 4).

Table 4.

Factors related to scan interruption

	Β	se	std β	OR (95% CI)	P value
Age < 1 year	− 0.86	0.36	− 0.07	0.43 (0.21–0.85)	< 0.05
Lower induction thiamylal dose	− 0.19	0.07	− 0.08	0.83 (0.72–0.94)	< 0.01
MRI scanning time	0.12	0.01	0.31	1.13 (1.10–1.16)	< 0.001
AUC = 0.80

A lower induction thiamylal dose was defined as dose < 3.6 mg/kg

β: Regression coefficients; se: standard error; std β: standardized regression coefficients; OR, odds ratio; CI, confidence interval; AUC: area under curve; MRI: magnetic resonance imaging

Subgroup analysis with scan interruption

As shown in Table 5, patients who received scans within 30 min with or without scan interruption had a similar mean induction thiamylal dose (4.1 ± 1.9 and 5.3 ± 1.9 mg/kg, P = 0.18) and total thiamylal dose (6.6 ± 1.6 and 5.3 ± 1.9 mg/kg, P = 0.13).

Table 5.

Patients grouped as per the total magnetic resonance scanning time and subgrouped according to scan interruption

Scanning time (min)		With scan interruption	Without scan interruption	P value
Within 30 min	Induction thiamylal dose (mg/kg)	4.1 ± 1.9	5.3 ± 1.9	0.18
	Total thiamylal dose (mg/kg)	6.6 ± 1.6	5.3 ± 1.9	0.13
30 to 60 min	Induction thiamylal dose (mg/kg)	4.6 ± 1.6	5.1 ± 1.9	< 0.001
	Total thiamylal dose (mg/kg)	7.4 ± 2.8	5.1 ± 1.9	< 0.001

Note that patients in 30- to 60-min group received a lower induction thiamylal dose, higher total thiamylal dose, and experienced scan interruption

However, when MRI scanning continued for 30 to 60 min, patients in the with scan interruption group received a lower induction thiamylal dose (4.6 ± 1.6 and 5.1 ± 1.9 mg/kg, P < 0.01), but a higher total thiamylal dose (7.4 ± 2.8 and 5.1 ± 1.9 mg/kg, P < 0.001).

More than 94% of patients experienced scan interruption when MRI scanning was performed for more than 1 h.

Discussion

The efficacy and safety of intravenous thiamylal in pediatric sedated MRI have been proven in previous studies [2]. However, while 10.2% of the patients in previous studies reported respiratory complications, no such events were observed in our study. A key difference was the use of routine oxygen administration and airway management in our protocol, including nasal cannula oxygenation and chin-lift positioning, which effectively prevented desaturation. In our study, respiratory depression was judged based on the respiratory rate, ensuring timely interventions. This suggests that using a higher induction dose of thiamylal (4.7 [3.8–5.8] mg/kg) with proper airway management may be safer than a lower dose (3.6 [2.8–4.0] mg/kg) without oxygen support.

Moreover, Lin et al. [6] reported a 1.8% desaturation rate in pediatric patients receiving a 5 mg/kg thiamylal loading dose for procedural sedation in emergency settings, with higher supplemental doses increasing the risk of hypoxia. In contrast, our use of lower rescue doses (1–2 mg/kg) resulted in no cases of desaturation, further underscoring the importance of dose adjustments and proper monitoring.

In this study, no alternative sedatives were used as rescue agents, resulting in a 100% success rate for thiamylal sedation. If patients who required other agents (e.g., ketamine or propofol) were included, the success rate decreased to 98.3%.

According to Irie et al. [2], there was no significant difference in the induction dose of thiamylal per weight between age groups. Interestingly, our findings indicate that older pediatric patients tend to receive a lower induction dose of thiamylal per kilogram, which likely reflects their lower sedation requirements. However, older age was not a risk factor for the need for rescue doses, suggesting that sedation protocols can be individualized according to the patient’s age and weight.

The relationship between scan duration and thiamylal dosing also highlights its limitations. For MRI scans lasting 30–60 min, patients who received lower induction doses experienced higher interruption rates and required higher total thiamylal doses. Even with a higher induction dose (5.1 ± 1.8 mg/kg), 94% of scans exceeding 1 h experienced interruptions, suggesting that thiamylal may not be the optimal sedative for procedures lasting longer than 60 min.

This study had some limitations. First, confounding variables, such as patient medical history, surgical history, indications for MRI scans, and the protocol design of MRI studies, were not fully controlled in the analysis. A prospective study with a larger sample size is required to validate these findings. However, as 79.7% of the MRI scans in this study focused on the brain, which is the most common indication for MRI, we believe that the results provide meaningful clinical insights for this prevalent MRI setting.

Second, the retrospective design of this study has inherent limitations. Most sedated pediatric MRI scans at our institution are performed in an outpatient setting, which may lead to an underestimation of late-onset adverse events that occur after discharge. Lastly, selection bias was a concern, as patients who received medications other than thiamylal were excluded from the analysis. This may have resulted in overestimation of the MRI scan success rate.

Conclusion

For MRI scans performed under 30 min, the interruption rates were similar between the lower- and higher-dose induction thiamylal groups. However, for scans lasting 30–60 min, a higher induction dose reduced the interruption rates and total thiamylal dose requirements without increasing the adverse events. Thiamylal is not an ideal agent for procedures exceeding 1 h, as interruption rates remain high despite higher induction doses. Future studies comparing thiamylal and propofol could further clarify their roles in pediatric sedation during MRI.

Abbreviations

MRI

Magnetic resonance imaging

IQR

Interquartile range

Author contributions

YSL contributed to the data acquisition, conceptualization, statistical analysis, and manuscript drafting. SPL contributed to data acquisition and statistical analysis. WKC reviewed and revised the manuscript. CKT contributed to the study coordination and manuscript revision. All the authors have read and agreed to the published version of this manuscript.

Funding

This study was supported by a grant from the Anesthesiology Research and Development Foundation (ARDF11401) and the National Science and Technology Council (NSTC112-2314-B-A49 -068 -MY3).

Availability of data and materials

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

This study was approved by the Institutional Review Board of the Taipei Veterans General Hospital, Taipei, Taiwan (IRB-TPEVGH no. 2024-08-018AC). The requirement for written informed consent was waived, and all included patients were de-identified before the analysis.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.