Is Procedural Intravenous Anesthesia Safe for Children? An Analysis of Nationwide Accident Registries in Japan

ABSTRACT

Background

Although pediatric intravenous anesthesia is widely employed outside operating rooms, it poses risks for serious complications such as airway obstruction and apnea. In Japan, the actual numbers and trends of medical accidents associated with pediatric intravenous anesthesia remain unclear, and the safety of procedural intravenous anesthesia in children has not been sufficiently investigated.

Methods

We extracted accident reports on pediatric intravenous anesthesia for medical procedures from a nationwide, anonymized Japanese accident registry database over 13 years. The reports were analyzed to describe the clinical characteristics of the reported severe adverse events.

Results

A total of 79 accident reports related to intravenous anesthesia were registered, accounting for 0.017% of the total number of intravenous anesthesia procedures. Among these, 70 (88.6%) accident cases occurred in children aged 0–9 years and 9 (11.8%) in patients older than 10 years. Four cases resulted in death and eight in cardiopulmonary arrest, all involving serious underlying medical conditions and occurring during prolonged and highly invasive procedures. When categorized by age, six cases (75.0%) involved children aged 0–9 years and 2 (25.0%) involved patients older than 10 years. The most common underlying disease categories were cardiac disease (n = 7) and the most frequent procedures were cardiac catheterization (n = 5). The number of physicians exclusively assigned to anesthesia induction and monitoring using capnometers was not reported.

Conclusions

This nationwide registry analysis characterizes severe pediatric intravenous anesthesia–related adverse events in Japan and highlights the importance of risk stratification and specialist involvement in high‐risk cases.

1. Introduction

In many long‐term medical procedures, such as magnetic resonance imaging (MRI), cardiac catheterization, and gastrointestinal endoscopy, total intravenous anesthesia (TIVA) is required in children to provide sedation and immobility. It is also useful in surgical procedures, particularly in outpatient and intensive care units, offering the significant advantage of a lower risk of malignant hyperthermia compared with inhalation anesthesia [1].

Whereas, intravenous anesthesia poses a risk for complications, such as airway obstruction and apnea, with occasional reports of serious medical accidents [2]. For example, the Japan Pediatric Society reported that approximately 25% of facilities experienced adverse events associated with deep sedation for MRI [3]. Furthermore, the reimbursement for prolonged intravenous anesthesia performed by non‐anesthesiologists in Japan is much lower than that for procedures performed by anesthesiologists, making it difficult to obtain resources for adequate safety equipment and monitoring [4].

Recent analysis of the nationwide open claims database (NDB open data) revealed that long‐term intravenous anesthesia is mostly performed by non‐anesthesiologists, whereas its use by anesthesiologists is declining despite increased government reimbursement support in Japan [5].

In this context, no detailed national survey of facilities or experts has been conducted to determine the extent to which pediatric procedural intravenous anesthesia is safely administered by non‐anesthesiologists in Japan and what the actual details of medical accidents are.

Japan has two nationwide government‐supported medical accident registries, one of which maintains a public, anonymized accident reporting database that is updated annually [6, 7]. We extracted individual reports from the 13‐year database registry and analyzed the details of cases in the accidents related to intravenous anesthesia in Japan. Through these analyses, our study aimed to elucidate and describe the clinical and procedural characteristics of reported severe adverse events.

2. Methods

2.1. Data Source

We utilized medical accident information reports registered with the National Database of Medical Adverse Events Project of the Japan Council for Quality Health Care between January 2010 and September 2023 [8]. This organization is the main entity in charge of evaluation of hospital functionality in Japan. Most hospitals undergoing such an evaluation participate in the registration. In addition, the number of reimbursement claims recorded in the National Database of Health Insurance Claims and Specific Health Checkups of Japan (NDB Open Data), maintained by the Ministry of Health, Labor and Welfare, from April 2014 to March 2023 was used for quantitative and qualitative analyses.

2.2. Extraction Method

Medical accident reports containing the keywords “pediatric,” “intravenous anesthesia,” and/or “sedation” were extracted. Also, from the anesthesia items in the medical procedure section of the NDB Open Data, the total number of intravenous anesthesia claims stratified by sex and age was extracted. The target age group was set to 0–19 years in both databases, according to the age categories defined in each database.

Cases of inhalation anesthesia with a closed circuit and TIVA performed in the operating room for surgery as well as cases in which the administration of intravenous anesthetic was not confirmed in the report were excluded from the analysis.

2.3. Quantitative and Qualitative Analyses

Details of the accidents and trends over time were analyzed based on the report descriptions. The relative proportion of extracted medical accident reports to the total number of intravenous anesthesia claims obtained from the NDB Open Data and its temporal trends were evaluated. The analysis was conducted for the 10‐year period from 2014 to 2023, corresponding to the overlapping period available in both databases. Age was categorized into 0–9 and 10–19 years, following the age categories in the Japanese reimbursement system for intravenous anesthesia [4].

The data for items other than age were tabulated by the number of accidents each year, gender, department in charge, number of physicians and years of experience at the time of the accident, disease category, examination and treatment, number and type of sedative drugs used, complications, treatment for complications, as well as prognosis and outcome. Among the complications, cases involving death and cardiopulmonary arrest were selected, and background risk factors were evaluated from the reports.

2.4. Statistical Analysis

The Cochran–Armitage test was employed to test for trends, with p‐value < 0.05 considered to indicate statistical significance. GraphPad Prism (version 8.1.0; GraphPad Software, San Diego, CA, USA) was used for statistical analysis.

2.5. Ethical Considerations

We asked permission to use the information in the database from the secretariat of the Japan Institute for Health Care Excellence. This study is a secondary analysis of publicly available anonymous case information. Therefore, ethical review was waived by the Juntendo University School of Medicine Ethics Review Committee.

3. Results

3.1. Number of Reported Accidents and Relative Occupancy

During the observation period, 79 accident reports related to deep sedation with intravenous anesthesia were registered, accounting for 0.017% of the total number of intravenous anesthesia procedures (460,734). The annual trends in the number and the relative incidence of reports showed no significant increasing or decreasing trends over the study period (Figure 1) (p = 0.76, Cochrane–Armitage test).

FIGURE 1.

(A) Actual number of accidents related to intravenous anesthesia reported in the database from 2010 to 2023. (B) Number of accidents involving children per 100,000 anesthesia procedures in the database from 2014 to 2023. (p = 0.76, Cochrane–Armitage test) There was a 3‐year discrepancy in the observation periods between the total number of intravenous anesthesia procedures and the number of intravenous anesthesia–related accident reports.

3.2. Background of the Accidents

A summary of accident case information is presented in Table 1.

TABLE 1.

Demographics of patients involved in anesthesia‐related accidents.

	n (n = 79)	%
A. Patient characteristics and provider categories
Age of patient a
0–9 years	70	88.6
10–19 years	9	11.4
Sex of patient
Male	45	56.9
Female	34	43.0
Provider categories
Doctor (1 person)	58	73.4
Doctor (2 persons)	12	15.2
Doctor (≧ 3 persons)	9	11.4
Years of occupational experience
2	6	5.0
2–5 years	18	15.0
5–10 years	35	29.2
≧ 10 years	49	40.8
No data	12	10.0
B. Clinical department, primary diagnosis, and medical practice
Clinical department
Pediatrics	67	84.8
Pediatric surgery	8	10.1
Anesthesiology	4	5.1
Neurosurgery	4	5.1
Cardiovascular surgery	3	3.8
Radiology	3	3.8
Otolaryngology	2	2.5
Ophthalmology	2	2.5
Cardiology	1	1.3
Emergency	1	1.3
Orthopedic surgery	1	1.3
Internal medicine	1	1.3
Surgery	1	1.3
Neurology	1	1.3
Primary diagnosis
Cardiac	23	29.1
Hematology‐oncology	16	20.5
Neurological	15	19.0
Congenital and genetic disorders (chromosomal or genetic abnormality)	7	8.9
Prematurity‐related	6	7.6
Liver‐related	5	6.3
Infectious	3	3.8
Trauma	3	3.8
Dermatologic	2	2.5
Ophthalmological	2	2.5
Orthopedical	2	2.5
Respiratory	2	2.5
Rheumatological	2	2.5
Surgical	2	2.5
Gastroenterological	1	1.3
Metabolic	1	1.3
Otolaryngological	1	1.3
Renal	1	1.3
Others	2	2.5
Medical practice
MRI	21	26.6
Cardiac catheterization	11	13.9
Motion suppression	11	13.9
Central venous cannulation	5	6.3
CT scan	5	6.3
Bone marrow aspiration/biopsy	4	5.1
Lumbar puncture	4	5.1
Nuclear medicine study	4	5.1
Anticonvulsant therapy	3	3.8
Defibrillation	2	2.5
Angiography	1	1.3
Drain tube insertion	1	1.3
Liver biopsy	1	1.3
Lower GI series	1	1.3
Proton therapy	1	1.3
Renal biopsy	1	1.3
Respiratory examination	1	1.3
Upper GI/gastrostomy	1	1.3
Ophthalmology treatment	1	1.3
C. Responsible agent(s)
Number of agents
1	40	50.6
2	18	22.7
≧ 3	6	7.6
No data	15	19.0
Agents for intravenous anesthesia
Opioid analgesics
Ketamine hydrochloride	7	8.9
Fentanyl citrate	4	5.1
Pentazocine	3	3.8
Pethidine hydrochloride	1	1.3
Nonopioid analgesics
Midazolam	28	35.4
Thiopental sodium	15	19.0
Thiamylal sodium	12	15.2
Dexmedetomidine hydrochloride	4	5.1
Propofol	2	2.5
Diazepam	1	1.3
Phenobarbital	1	1.3
Secobarbital sodium	1	1.3
Combination‐use
Triclofos sodium or chloral hydrate	9	11.1
No data	15	19.0

^a Age categories were based on the original data.

A total of 70 (88.6%) accident cases involved the 0–9‐year‐old age group and 9 cases (11.4%) involved the 10–19‐year‐old age group. In 58 cases (73.4%), a single physician was responsible for anesthesia management, and in 49 cases (40.8%), the accident occurred under the care of a physician with an experience of over 10 years (Table 1A).

In terms of underlying diseases, cardiovascular diseases accounted for the highest number (n = 23, 29.1%), followed by hematological and oncological diseases (n = 16, 20.5%) and then neurological diseases (n = 15, 19.0%). Among procedures requiring sedation, MRI was the most common (n = 21, 26.6%), followed by cardiac catheterization and motion suppression (both n = 11, 13.9%) (Table 1B).

In 40 cases (50.6%), a single intravenous sedative or anesthetic agent was used. Among these, barbiturates were the most commonly reported, with phenobarbital, secobarbital, thiopental, and thiamylal accounting for 28 cases (35.4%) when combined. Meanwhile, triclofos and choral hydrate, which are used for relatively mild sedation, were reported in 9 cases (11.1%) (Table 1C).

3.3. Complications and Prognosis

A summary of complications is presented in Table 3.

TABLE 3.

Summary of fatal cases and cases requiring cardiopulmonary resuscitation.

	Date	Age	Sex	Clinical department	Categories of primary diagnosis	Provider categories	Medical practice	Medicine of anesthesia	Prognosis and outcomes
Case 1	2010	10–19 years	M	Pediatrics	Cardiac Congenital and genetic disorders	1 doctor (≧ 10 years)	Cardiac catheterization	Thiopental sodium	Death
Case 2	2012	0–9 years	F	Pediatrics Pediatric surgery	Metabolic	2 doctors (2–5 years) 1 doctor (5–10 years) 3 doctors (≧ 10 years)	Liver biopsy	Midazolam Thiamylal sodium	Death
Case 3	2013	0–9 years	F	Pediatrics	Cardiac	1 doctor (5–10 years) 1 doctor (≧ 10 years)	Cardiac catheterization	No data	Death
Case 4	2013	0–9 years	F	Pediatrics Pediatric surgery Anesthesiology Cardiovascular surgery	Cardiac	1 doctor (2–5 years) 1 doctor (5–10 years) 4 doctors (≧ 10 years)	Cardiac Catheterization	Diazepam Thiopental sodium	Death
Case 5	2016	0–9 years	M	Pediatrics	Cardiac	1 doctor (≧ 10 years)	Cardiac Catheterization	No data	Possibility of permanent disease
Case 6	2018	0–9 years	M	Cardiovascular Surgery	Cardiac	2 doctors (2–5 years) 2 doctors (≧ 10 years)	Motion suppression	Midazolam Dexmedetomidine Hydrochloride Combination‐use; Triclofos sodium	Possibility of permanent disease
Case 7	2019	10–19 years	F	Pediatrics Anesthesiology	Cardiac	1 doctor (≧ 10 years)	Motion suppression	No data	No permanent disease
Case 8	2022	0–9 years	F	Pediatrics	Cardiac Congenital and genetic disorders	1 doctor (2–5 years) 1 doctor (≧ 10 years)	Cardiac catheterization	Thiopental sodium	Possibility of permanent disease

The most common complication was hypoxemia (n = 60, 75.9%). Persistent decreases in oxygen saturation were observed in 36 cases (45.6%), apnea in 13 (16.5%), bradycardia in 7 (8.9%), and hypotension in 4 (5.1%). Cardiopulmonary arrest occurred in five cases (6.3%) (Table 2A).

TABLE 2.

Complications, treatments, and outcomes of patients.

	n (n = 79)	%
A. Complications
Adverse events
Desaturation below 90%	36	45.6
Prolonged sedation	23	29.1
Overdose of intravenous anesthesia	18	22.8
Misadministration of medicine	14	17.7
Apnea	13	16.5
Bradycardia	7	8.9
Operation failure/surgical complications	6	7.6
Cardiac arrest	8	10.1
Blood pressure reductions	4	5.1
Secretions/vomiting	3	3.8
Allergic reaction (rash)	2	2.5
Nerve paralysis	2	2.5
Seizure (unanticipated) during sedation	2	2.5
Others	7	8.9
B. Treatments against complications
Treatments
Oxygen inhalation	29	36.7
Bag‐mask ventilation	25	31.6
CPR	8	10.1
Tracheal intubation and need ventilator management	6	7.6
Reoperation/surgical treatments	4	5.1
Antagonist administration	3	3.8
Airway (oral or nasal)	2	2.5
Anaphylaxis treatment	2	2.5
Blood transfusion	2	2.5
Extracellular fluid infusion	2	2.5
Suction	2	2.5
No treatment	19	24.1
No data	12	15.2
C. Prognosis and outcomes of patients
Prognosis and outcomes
Death	4	5.1
Possibility of permanent disease	17	21.0
No permanent disease	58	71.6

Abbreviation: CPR, cardiopulmonary resuscitation.

The interventions for these complications included cardiopulmonary resuscitation (n = 8, 10.1%), tracheal intubation and ventilation (n = 6, 7.6%), bag‐mask ventilation (n = 25, 31.6%), oxygen administration (n = 29, 36.7%), and airway insertion (n = 2, 2.5%) (Table 2B). Four (5.1%) patients died, 17 (21.0%) had possible sequelae, and 58 (71.6%) had no sequelae (Table 2C).

Details of fatal cases and cases requiring cardiopulmonary resuscitation are presented in Table 3. There were four deaths and eight cases of cardiopulmonary arrest; among the latter, three were left with possible residual disabilities after resuscitation, whereas one showed no lasting impairment. When categorized by age, six cases (75.0%) involved children aged 0–9 years, whereas 2 cases (25.0%) involved individuals aged 10–19 years. The patient disease categories included cardiac disease (n = 7), congenital and genetic disorders (n = 2), and metabolic disease (n = 1). The examinations and procedures included cardiac catheterization (n = 5), liver biopsy (n = 1), and motion control (n = 2). Thiopental was used in two cases of death and one case requiring cardiopulmonary resuscitation, whereas midazolam was used in one case of death and one case requiring cardiopulmonary resuscitation.

The number of physicians exclusively assigned to anesthesia induction and monitoring using capnometers was not reported.

4. Discussion

This study analyzed data from a 13‐year nationwide Japanese database of accident reports, focusing on pediatric medical accidents related to intravenous anesthesia during tests and procedures. It also examined the contributing factors. In Japan, the vast majority of such anesthesia is administered by non‐anesthesiologists, and no comprehensive nationwide safety reports that do not target specific areas have been published [5].

Technical fees for deep sedation with intravenous anesthesia in children continue to increase, alongside fees for intravenous anesthesia (category 3) administered by professional anesthesiologists in 2020 [4]. However, the number of reported accidents and relative occupancy rates in the database have not shown a decline. It may be challenging to make procedure‐related pediatric intravenous anesthesia a safe medical practice simply by securing medical resources in economic terms.

Most of the registered cases were below 9 years old. However, a non‐negligible number of accidents, including fatal ones, occurred in schoolchildren and adolescents aged 10 years and older, who are ineligible for the additional pediatric technical fee. Furthermore, a large case–control study conducted at a university hospital in Japan found no significant difference in accident rates by age among pediatric sedation cases [9]. Many school‐age children requiring sedation for procedures have developmental or neurological disorders, and their sedation is never easy. Multicenter, large‐scale data from a North American consortium revealed a trend toward older age for pediatric procedural sedation [10]. Based on this data, the current maximum age limit of 3 years for additional technical fees in pediatric sedation may be insufficient. However, this study does not demonstrate that age itself is a risk factor; it merely indicates that severe events occur beyond early childhood. If clinical severity can be clearly stratified by age groups in future studies, policy interventions targeting older children may also need to be considered.

The popular use of barbiturates in children who have developed complications is a serious concern. Owing to their strong suppression on cardiopulmonary function, these drugs are not commonly used in North America [10]. In addition, the relative risk of complications with barbiturates for sedation during MRI scans was high in Japan, based on claims data analysis [11]. However, the data in this study only describe the drugs involved in accident cases and do not allow estimation of relative risk for each drug; therefore, further studies incorporating claims data on sedative drug use in Japan are warranted to evaluate drug‐specific risks.

The finding that most accidents resulting in death or cardiopulmonary arrest occurred during cardiac catheterization in patients with serious underlying conditions, such as complex cardiac or congenital diseases, represents highly important descriptive data and suggests that medical safety measures and considerations may be insufficient. The recommendation on sedation during MRI examinations from the Japanese Pediatric Society also points out the importance of risk assessment prior to sedation [12]. In Japan, unlike inhalation anesthesia, intravenous anesthesia does not incur additional technical fees based on the patient's underlying conditions or the complexity of the procedure. Risk stratification by underlying disease and additional technical fees may lead to safer induction of intravenous anesthesia.

Recently, the government established a generous technical fee for prolonged sedation (additional fees for pediatric MRI sedation) for MRI, for which clinical data has been accumulated. However, the requirements for calculation are overly stringent, and this fee is rarely calculated [13]. There is a need to design a technical fee for a more comprehensive subject that is more readily available to non‐anesthesiologists.

This study has the following limitations: Selection bias exists in enrollment as this is a retrospective study of a case registry. Also, reporting to the accident registry of the Japan Council for Quality Health Care may be biased toward relatively large secondary and tertiary hospitals, which may lead to discrepancies between the reported cases and the true underlying denominator. Details such as accurate age information, underlying diseases, and facility names are missing owing to the anonymization rule. For example, because of limitations in the available denominator data, the denominator for the 16–18‐year age group could not be obtained. Detailed data on monitoring equipment and the involvement of anesthesiology departments are limited, as the reports mainly rely on free‐text descriptions. The latter part of the inclusion period may have been influenced by changes in disease structure owing to the novel coronavirus pandemic. Some of these limitations may be alleviated by more detailed national surveys led by academic societies or consortia, such as those already conducted in association with MRI and cardiac catheterization [3, 14].

In conclusion, this nationwide registry analysis characterizes severe pediatric intravenous anesthesia–related adverse events in Japan. Serious accident cases included those with serious underlying medical conditions, and stratification of the technical fee based on underlying medical conditions, as is done with inhalation anesthesia, may help ensure the appropriate allocation of medical resources. The findings emphasize the importance of risk stratification and specialist involvement in high‐risk cases. A more detailed national survey is warranted to examine measures to ensure the safety of pediatric procedural sedation.

Author Contributions

M.M. and E.I. developed the concept and initial design of the study and were responsible for data collection, statistical analysis, and figure preparation and participated in discussions. S.Y., A.E., Y.N., S.Y., I.I., Y.S., S.T., T.K., Y.O., T.S., and H.S. supervised the study design and participated in the preparation and discussion of the final draft of the paper. All researchers participated in the writing of the paper and agreed to its submission for publication. All authors confirmed and approved the final version of the manuscript.

Funding

The authors have nothing to report.

Disclosure

A portion of the data from this study was sent to the Health Insurance Bureau of the Ministry of Health, Labor and Welfare as reference material for the 2024 revision of reimbursement for medical services. The authors of this study encompassed social insurance directors and committee members from several academic organizations, including the chair of the social insurance committee (EI) of the Japan Pediatric Society. The views expressed in this paper are those of the authors as individuals and do not represent the views of their respective organizations.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.