Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Oct 31.
Published in final edited form as: J Autism Dev Disord. 2021 Oct 7;52(10):4301–4310. doi: 10.1007/s10803-021-05305-0

Association Between Exposure of Children to General Anesthesia and Autism Spectrum Disorder

Mariana L Laporta 1, Juraj Sprung 1, Caroline A Fejedelem 1, Dustin T Henning 1, Amy L Weaver 2, Andrew C Hanson 2, Darrell R Schroeder 2, Scott M Myers 3, Robert G Voigt 4, Toby N Weingarten 1, Randall P Flick 1, David O Warner 1
PMCID: PMC9620709  NIHMSID: NIHMS1843078  PMID: 34618293

Abstract

This study tested the hypothesis that exposure of children prior to their third birthday to procedures requiring general anesthesia is associated with an increased incidence of autism spectrum disorder (ASD) in later life. This study employed a nested, 1:2 matched-case control study design using ASD cases identified in a population-based birth cohort of children born in Olmsted County, MN from 1976 to 2000. Matching variables included sex, date of birth, and mother’s age in conditional logistic regression including 499 ASD cases and 998 controls. After adjusting for birth weight and health status, there was no significant association between exposure and ASD (OR 1.27 [95% CI 0.92–1.76]), indicating that general anesthesia is not associated with an increased risk of ASD.

Keywords: Autism, Anesthesia, ASD, Childhood, Neurodevelopmental disorders

Autism spectrum disorder (ASD) is a neurodevelopmental disability characterized by deficits in social communication and restricted/repetitive behaviors (Diagnostic and Statistical Manual of Mental Disorders: DSM-IV-TR, 2000). Although a specific genetic etiology can be identified in a substantial minority of individuals, most cases are currently considered idiopathic and a complex interplay between genomic and environmental factors likely plays an important role (de la Torre-Ubieta et al., 2016; Modabbernia et al., 2017; Muhle et al., 2018). In many countries the estimated prevalence of ASD has increased (Baxter et al., 2015), which may reflect in part improved diagnostics and increased availability of medical care (Hansen et al., 2015; Lundstrom et al., 2015; Myers et al., 2019). However, this increase may also suggest the potential importance of environmental factors (Lyall et al., 2017), such as birth complications and exposure to toxins, infections, and drugs (Christensen et al., 2019; Currenti, 2010; Doja & Roberts, 2006; Modabbernia et al., 2017; Newschaffer et al., 2007).

Exposure of young animals to general anesthetics causes neurodegenerative changes associated with long-term changes in learning, memory, and behavior (Coleman et al., 2016; Jevtovic-Todorovic et al., 2003). In studies of children residing in Olmsted County, MN, multiple (but not single) exposures of young children to anesthesia were associated with an increased incidence of attention-deficit hyperactivity disorder (ADHD) and learning disabilities (Flick et al., 2012; Hu et al., 2017; Sprung et al., 2012; Wilder et al., 2009). Some, but not all, other investigators have found that single exposures to anesthesia at a young age are also associated with ADHD (Ing et al., 2017; Ko et al., 2014; Sedighnejad et al., 2020). Many children with ADHD have difficulties with social interactions, and many children with ASD also meet criteria for ADHD (Antshel & Russo, 2019; Kushki et al., 2019; Llanes et al., 2020; van der Meer et al., 2012). This interaction raises the question of whether the association between undergoing procedures requiring general anesthesia and ADHD may also exist for ASD if there is a similar underlying biology for the social deficits in the two conditions. Two studies find that the use of general anesthesia during caesarean delivery is associated with a higher likelihood of ASD compared with the use of regional anesthesia, with the authors suggesting peripartum anesthesia exposure as a possible cause (Chien et al., 2015; Huberman Samuel et al., 2019). Two other studies have not found a relationship between post-natal exposure to anesthesia and ASD ascertained via clinical diagnosis (Creagh et al., 2016; Ko et al., 2015). Thus, the relationship between exposure to procedures requiring anesthesia and ASD remains unclear.

This study tested the hypothesis that exposure of children prior to their third birthday to procedures requiring general anesthesia is associated with an increased likelihood of ASD. We tested this hypothesis using a population-based birth cohort of children from Olmsted County, MN, taking advantage of prior work in this cohort that ascertained ASD cases using research criteria (Myers et al., 2019).

Methods

The population-based study cohort was assembled as described in prior work (Myers et al., 2019) using birth certificate data for Olmsted County, Minnesota, obtained from the Minnesota Department of Health, and the resources of the Rochester Epidemiology Project (REP), a medical records linkage system that includes records of all outpatient and inpatient medical care in the community, including Mayo Clinic, Olmsted Medical Center, their 3 affiliated hospitals, and several smaller practices. The REP provides centralized longitudinal medical data, including clinical documentation from primary care and specialty clinic visits, emergency department visits, hospitalizations, laboratory and imaging results, and birth and death certificate data. The study was approved by the institutional review boards at Olmsted Medical Center and Mayo Clinic. Access to cumulative school records was made possible through contractual agreement of Mayo Clinic, the Independent School District No. 535 school board, and equivalent authorities governing private schools.

Study Population

All children born between January 1, 1976 and December 31, 2000 to mothers residing in Olmsted County (n = 43,215) at the time of delivery were considered. Among these children, 39,890 had authorized the use of their medical records for research. A total of 31,220 children who still lived in the Olmsted County at or after the age of 3 years were followed from birth until age 21, emigration or death.

Identification of Individuals with ASD

ASD case status was determined retrospectively using data from the REP and school documentation and an operational research definition based on Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR) criteria, as describe in detail previously (Myers et al., 2019). First, a group of individuals with neurodevelopmental or psychiatric disorders was identified based on diagnostic codes in their medical records or educational disability classification codes in their school records. The codes included were conditions with signs and symptoms that overlap with the core social, communicative, and behavioral features of ASD or that commonly coexist with ASD. Among the 31,220 eligible individuals in the population-based birth cohort, 4301 were identified as having one or more of these neurodevelopmental/psychiatric disorder or educational disability codes. Of these, 1766 individuals were determined to have potential ASD based on the presence of certain codes or combinations of codes and screening of prioritized relevant parts of the medical records, as described elsewhere (Myers et al., 2019). Records of all 1766 individuals with potential ASD were then manually reviewed to abstract specific data through a systematic, multistage process. The abstractors carefully reviewed all medical record documentation through 21 years of age to identify pertinent descriptive phrases, based on an extensive data dictionary developed by the research team, that mapped to any of the 58 ASD signs or symptoms that contributed to the 12 core criteria (4 social interaction, 4 communication, and 4 restricted, repetitive behavior) in the DSM-IV-TR. Finally, the research team applied the DSM-IV-TR–based criteria to the abstracted information to identify individuals meeting the research definition of ASD. Likely false-positive results were excluded by a manual review process that identified individuals whose signs and symptoms were attributed to other conditions, such as adolescent-onset psychosis, bipolar disorder, or major depression. Cases in the original work (Myers et al., 2019) were ascertained using two research definitions: (1) a broader, more inclusive definition requiring that at least 3 criteria were met (including 2 of the 4 social interaction criteria and at least 1 of the communication criteria or 1 of the restricted, repetitive behavior criteria), based on the DSM-IV-TR criteria for pervasive developmental disorder not otherwise specified and encompassing autistic disorder and Asperger’s disorder), and (2) a more narrow, higher-confidence definition requiring that at least 6 criteria were met (including at least 2 of 4 social interaction criteria, 1 of 4 communication criteria, and 1 of 4 restricted, repetitive behavior criteria), based on DSM-IV-TR criteria for autistic disorder. For the current analysis, the latter definition was used (Myers et al., 2019). Because ADHD is frequently diagnosed in children with ASD and has been associated with early exposure to anesthesia in other studies, we also identified the presence of ADHD in cases and controls by searching the medical records for ICD-9 codes 314.X (Gruschow et al., 2019; Shi et al., 2020).

Identification of Matched Controls

A nested, 1:2 matched case–control design was employed. The date that there was sufficient information in the medical records for a case to meet research criteria for ASD is herein referred to as the “index date”. Case–control matching variables included sex, date of birth (± 90 days), and the mother’s age at the time of the birth (± 2 years). For each ASD case, 2 controls were randomly selected from the birth cohort who (1) met the matching criteria, (2) were still residing in Olmsted County at the time of the index date, and; (3) did not themselves meet ASD criteria prior to the index date.

Identification of Exposure to Procedures Requiring General Anesthesia

All cases and controls who underwent any procedure requiring general anesthesia prior to their 3rd birthday (or prior to the index date if the index date was before their 3rd birthday) at Mayo Clinic Rochester or Olmsted Medical Center, the two facilities that provide medical care in Olmsted County, were identified. Any exposures during cesarean delivery were not included. Anesthesia and surgical information were abstracted and entered manually into the web-based Research Electronic Data Capture (REDCap®) system (Version 9.1.15, 2020 Vanderbilt University, Nashville, Tennessee).

Statistical Analysis

Analyses were performed using conditional logistic regression taking into account the 1:2 matched case–control study design. The explanatory variable of interest was exposure to general anesthesia under the age of 3 years. Exposure was defined using a binary variable (any vs none), a categorical variable defined based on the number of exposures (0, 1, 2 or more), and also using a continuous and categorical variable (< 60 min or ≥ 60 min) representing the cumulative duration of exposure to general anesthesia.

As in our prior work (Flick et al., 2012; Gleich et al., 2014; Hu et al., 2016; Warner et al., 2018) the health status of each child up to age 3 years (or prior to the index date if the index date was before age 3) was quantified using the Johns Hopkins Adjusted Clinical Groups ACG® Case–Mix System (ACG® System Version 12.0, John Hopkins, Baltimore, MD) (Weiner et al., 1991, 1996). This method utilizes International Classification of Diseases, 9th Revision diagnosis codes (ICD-9). Codes for each child were converted to ICD-9 if necessary and assigned to one of 32 unique morbidity clusters (aggregated diagnostic groups [ADGs]).

Since birth weight and health status were not included as matching variables, additional analyses were performed with birth weight (treated as a continuous variable) and 28 ADG indicator variables included as covariates. These indicators represent all ADGs with the exception of those included in the psychosocial category (which maps to the diagnosis of ASD and other behavioral disorders) and pregnancy. Findings are summarized using the odds ratio and corresponding 95% confidence interval (CI). For all continuous variables initial models were fit using restricted cubic splines to assess for evidence of non-linearity. After verifying the lack of significant non-linearity, continuous variables were modeled using linear terms for all subsequent analyses. In all cases, two-tailed P-values < 0.05 are considered statistically significant (Version 9.4, SAS Institute, Inc, Cary, NC).

ASD and ADHD frequently co-exist (Antshel & Russo, 2019; Kushki et al., 2019; Llanes et al., 2020; van der Meer et al., 2012). As a supplemental analysis, cases were categorized into those with ASD only and those with both ASD and ADHD. A supplemental covariate adjusted conditional logistic regression analyses was performed to assess the association between anesthesia exposure and autism for each of these case categories. An additional supplemental analysis was performed to assess whether the association of anesthesia with autism differed based on the age of ASD diagnosis (before vs after 7th birthday). The purpose of this analysis was to approximately the method used in a prior analysis of the association between anesthesia exposure and ASD (Ko et al., 2015) which sought earlier diagnoses (within 3 years of exposure, corresponding approximately to a diagnosis prior to the 7th birthday in the current analysis).

All analyses were performed using the original case–control sets in which two controls for a given case were selected from the pool of all undiagnosed individuals residing in Olmsted County on the date the case was diagnosed. Using this approach, some controls could later be diagnosed with ASD, and therefore included in the study as both a control and as a case. A sensitivity analysis was performed in which the controls that later became cases were replaced with new controls who were never diagnosed with ASD.

At the time of study design, we estimated that there would be approximately 500 cases and 1000 controls. Under the assumption that 15% of the underlying population would be exposed to anesthesia prior to their 3rd birthday (Shi et al., 2018), this sample size provided statistical power of 80% to detect an odds ratio of 1.5.

Results

There were 533 research defined ASD cases identified in the 1976–2000 birth cohort. After excluding 34 cases who denied authorization for the use of their medical records for research, 499 ASD cases and 998 matched controls were analyzed (Table 1). The median age when cases first met research criteria for ASD was 6.9 years [25th, 75th percentile: 4.5, 9.5]; 55 cases (11%) met criteria prior to age 2 years, and 474 cases (95%) prior to age 14 years. Among the 499 individuals with research defined ASD, 194 (39%) had received an ASD diagnosis as part of their clinical care.

Table 1.

Sex, birth weight, maternal age, and health status as measured by Adjusted Clinical Groups (ACG)® Case-Mix System of cases and controls

Variable Cases (N = 499) Controls (N = 998)
Sex, n (%)
 Female 124 (25%) 248 (25%)
 Male 375 (75%) 750 (75%)
Birth weight, grams, n (%)
 < 2500 40 (8%) 45 (5%)
 ≥ 2500 459 (92%) 953 (95%)
Mother’s Age, years 28 (6) 28 (6)
ACG® Case-Mix System-ADG codes
 Acute Minor
  Time limited:minor, n (%) 361 (72%) 635 (64%)
  Time limited:minor-primary infections, n (%) 427 (86%) 814 (82%)
  Injuries/adverse effects:minor, n (%) 215 (43%) 356 (36%)
  Signs/symptoms:minor, n (%) 237 (47%) 382 (38%)
 Acute major
  Time limited:major, n (%) 122 (24%) 153 (15%)
  Time limited:major-primary infections, n (%) 150 (30%) 216 (22%)
  Injuries/adverse effects:major, n (%) 177 (35%) 282 (28%)
  Signs/symptoms:uncertain, n (%) 299 (60%) 535 (54%)
  Signs/symptoms:major, n (%) 284 (57%) 365 (37%)
 Likely to recur
  Allergies, n (%) 52 (10%) 86 (9%)
  Likely to recur:discrete, n (%) 193 (39%) 311 (31%)
  Likely to recur:discrete_infections, n (%) 427 (86%) 806 (81%)
  Dermatologic, n (%) 97 (19%) 181 (18%)
  Discretionary, n (%) 303 (61%) 581 (58%)
 Asthma
  Asthma, n (%) 77 (15%) 92 (9%)
 Chronic medical: unstable
  Likely to recur:progressive, n (%) 4 (0%) 4 (1%)
  Chronic medical:unstable, n (%) 63 (13%) 44 (4%)
  Malignancy, n (%) 2 (0%) 1 (0%)
 Chronic medical: stable
  Chronic medical:stable, n (%) 110 (22%) 98 (10%)
  See and reassure, n (%) 112 (22%) 175 (18%)
 Chronic specialty: stable
  Chronic specialty:stable-orthopedic, n (%) 13 (3%) 15 (2%)
  Chronic specialty:stable-ear, nose, throat, n (%) 52 (10%) 58 (6%)
 Eye/dental
  Chronic specialty:stable-eye, n (%) 205 (41%) 369 (37%)
  Dental, n (%) 13 (3%) 20 (2%)
 Chronic specialty unstable
  Chronic specialty:unstable-orthopedic, n (%) 33 (7%) 17 (2%)
  Chronic specialty:unstable-ear, nose, throat, n (%) 41 (8%) 36 (4%)
  Chronic specialty:unstable-eye, n (%) 40 (8%) 28 (3%)
 Psychosocial
  Psychosocial:time limited:minor, n (%) 106 (21%) 130 (13%)
  Psychosocial:persistent/recurrent, stable, n (%) 163 (33%) 61 (6%)
  Psychosocial:persistent/recurrent,cunstable, n (%) 19 (4%) 1 (0%)
 Preventive/administrative
  Prevention/administrative, n (%) 467 (94%) 978 (98%)
 Pregnancy 0 (0%) 0 (0%)
Open in a new tab

N (%), number and percentage of patients. Analyses were performed using 28 ADG indicator variables included as covariates, a method employed in our prior work where further details may be found (Flick et al., 2012; Gleich et al., 2014; Hu et al., 2016; Warner et al., 2018). These indicators represent all ADGs with the exception of those included in the psychosocial category (which maps to the diagnosis of ASD and other behavioral disorders) and pregnancy

Among cases there were 132 children who underwent 194 procedures requiring general anesthesia under the age of 3 years, and among controls there were 155 children who underwent 230 procedures. Among those who received anesthesia, the median [25th, 75th percentile] age at the first exposure was 14 [7, 24] months for cases and 14 [7, 23] months for controls (rank sum test p = 0.811). The median duration of anesthesia for each procedure was 60 [35, 115] minutes for cases and 61 [30, 105] minutes for controls (rank sum test p = 0.162) and the median total cumulative anesthesia duration was 76 [40, 155] minutes for cases and 70 [30, 119] minutes for controls (rank sum test p = 0.123).The most frequent types of procedures were ear, nose and throat, general and urologic (Table 2). Most procedures included the use of a halogenated inhalational agent.

Table 2.

Surgeries performed among cases and controls

Procedure type Cases
(N = 499)		 Controls
(N = 998)
Ear, nose and throat 74 (15%) 85 (9%)
General 25 (5%) 32 (3%)
Urologic 18 (4%) 25 (3%)
Ophthalmology 8 (2%) 8 (1%)
Orthopedic 3 (1%) 7 (1%)
Neurologic 3 (1%) 4 (0%)
Oral/Maxillofacial 6 (1%) 6 (1%)
Cardiovascular 6 (1%) 1 (0%)
Other 10 (2%) 8 (1%)
Open in a new tab

Values represent the number (%) of children among cases or controls who had at least 1 of the designated types of surgery. Among cases, 132 children underwent 194 procedures requiring general anesthesia and among controls, 155 children underwent 230 procedures

From the matched analysis that did not include additional covariates, anesthesia exposure was significantly associated with ASD when exposure was quantified as a dichotomous variable (any exposure vs. none), an ordinal variable of the number of exposures (0, 1, or ≥ 2), or the cumulative or categorical duration of exposure (all p < 0.001) (Table 3). However, after adjusting for birth weight and health status, no significant association was detected when exposure was quantified as a dichotomous variable (OR 1.27, [95% CI 0.92–1.76]), the number of exposures (OR 1.39 [0.98–1.98] and 0.93 [0.53–1.64] for 1 and ≥ 2 exposures, respectively, compared to no exposure) or by the cumulative duration of exposure (OR 0.99 [0.95–1.04] per 30 min increase) (Table 3).

Table 3.

Association between exposure to procedures requiring anesthesia before age 3 and incidence of autism spectrum disorder

Anesthetic characteristics Matched analysis without additional
covariates
 Matched analysis, covariate
adjusted
Odds ratio
(95% CI)		 p Odds ratio
(95% CI)		 p
Any anesthetic exposure < .001 0.149
  No 1.00 1.00
  Yes 1.94 (1.49–2.52) 1.27 (0.92–1.76)
Number of anesthetic exposures < .001 0.144
  No exposure 1.00 1.00
  1 exposure 1.88 (1.39–2.54) 1.39 (0.98–1.98)
  ≥ 2 exposures 2.09 (1.33–3.31) 0.93 (0.53–1.64)
Cumulative duration, per 30 mina 1.08 (1.03–1.12) < .001 0.99 (0.95–1.04) 0.772
Categorical duration < .001 0.337
  No exposure 1.00 1.00
  < 60 min 1.62 (1.10–2.39) 1.25 (0.79–1.98)
  ≥ 60 min 2.19 (1.58–3.05) 1.29 (0.87–1.91)
Open in a new tab

Analysis performed using conditional logistic regression taking into account the 1:2 matched set study design (matched on sex, date of birth (± 90 days), and mother’s age at the time of birth (± 2 years)). Additional covariates in the adjusted analysis include birth weight and health status as measured using the Adjusted Clinical Groups (ACG)® Case-Mix System

a

Cumulative duration of exposure to anesthetic before age 3 was analyzed as a continuous variable, with the odds ratio presented for a 30 min increase

In supplemental analyses, among the 499 ASD cases, 305 (61%) children also had a diagnosis of ADHD prior to age 19 years. For covariate adjusted analysis, with anesthesia exposure defined using a dichotomous variable (any exposure vs none), there was no significant association between exposure to anesthesia and ASD without co-existent ADHD (n = 194 cases and 388 controls, OR 1.21 [0.74–1.97]) or ASD with co-existent ADHD (n = 305 cases and 610 controls, OR 1.31 [0.88–1.95]). There was also no significant association between exposure to anesthesia and ASD diagnosed under the age of 7 years (OR 1.32 [0.84–2.08]) or over the age of 7 years (OR 1.23 [0.81–1.87]).

Finally, of the 998 controls included in the primary analyses, 19 subsequently met research criteria for ASD when they were older. From the supplemental analysis that replaced these individuals with alternative controls who never met research criteria for ASD, the pattern of results remained unchanged (covariate adjusted OR 1.30 [0.94–1.80] for any exposure to anesthesia under the age of 3 years versus no exposure; p = 0.114).

Discussion

The main finding of this study is that receiving procedures requiring general anesthesia before the age of 3 years is not associated with an increased incidence of later ASD.

The present report defines ASD cases according to research criteria based on the work of Myers et al. (Myers et al., 2019), who utilized these criteria to determine secular tends in autism incidence. In contrast, most investigators employ the clinical diagnosis of autism according to DSM criteria to analyze prevalence and to examine associations with anesthesia. The latest edition of DSM (DSM-5) combined three diagnostic categories in DSM-IV-TR (ASD, Asperger’s disorder, and pervasive developmental disorder not otherwise specified) into the single category of ASD. We utilize the more narrow, “restricted” definition of ASD described by Myers et al. (which were based on DSM-IV-TR autistic disorder criteria), chosen (1) to be more conservative in our analysis by using the stricter criteria, (2) to reflect diagnostic conventions over the bulk of the time period examined, and; (3) to make the manual review of records for anesthesia exposure more feasible. Prevalence estimates based on this narrower, higher-confidence research definition, were similar to those reported in other studies that used active ascertainment strategies when comparable data (based on ages and calendar years) were examined.

One important finding of Myers et al. relevant to the current analysis is that the clinical diagnosis of ASD was attached to fewer than half of the cases that met research criteria for ASD (Myers et al., 2019). Because other studies of the association between anesthesia and exposure and ASD discussed in the following section use clinical diagnosis to ascertain ASD cases, our results are not strictly comparable with these other studies.

The etiology of ASD is uncertain and likely multifactorial, including genetic factors as well as perinatal,and postnatal environmental factors (Currenti, 2010; Freitag, 2007; Gardener et al., 2011; Modabbernia et al., 2017). The evidence for the importance of genetic and perinatal factors is considerable; examples of risk factors include family history, birth complications, pre-term birth, and advanced paternal age (Gardener et al., 2011). Evidence for the role of postnatal exposure to other environmental factors is not as well-developed, with evidence supporting the association between ASD and exposures to toxins such as inorganic mercury and lead (Modabbernia et al., 2017), and infectious processes such as encephalitis (Atladottir et al, 2010; Marques et al, 2014).

Of potential relevance to anesthesia exposure, some studies suggest that Cesarean delivery (CD) is associated with an increased likelihood of autism (Curran, et al., 2015a, 2015b; Zhang et al., 2019). Two studies suggest specifically that this association is present for CD conducted with general, but not regional, anesthesia (Chien et al., 2015; Huberman Samuel et al., 2019). Peripartum exposure to general anesthetic drugs was suggested as a possible cause, based on pre-clinical studies showing neurotoxic effects of prenatal and perinatal exposure to general anesthesia, including those in non-human primates (Coleman et al., 2016; Currenti, 2010; Freitag, 2007). However, as noted by others, confounding by indication (i.e., parturients requiring CD differ from those who do not) is difficult to address in these observational studies (Sagi-Dain et al., 2020). Indeed, a recent study using a sibling-matched cohort design found that ASD was not associated with CD, highlighting the importance of confounding (Curran, et al., 2015a, 2015b). In addition, the fetus is exposed only briefly to low concentrations of general anesthetic drugs during CD, as best exemplified by the fact that the neonate is awake at delivery, making it implausible that these drugs would contribute to any such association.

Our results also affirm the potential for confounding to affect conclusions. Despite the use of a matched case–control design, it appears that factors related to differences in health status between children who did and did not receive anesthesia account for the significant association observed in univariate analysis. We noted a similar pattern of results in our prospective matched cohort study (Mayo Anesthesia Safety in Kids) using neuropsychological tests as outcomes. The primary outcome of full-scale intelligence quotient was significantly decreased in children exposed to multiple anesthetics compared to children not exposed to anesthesia, but this association was no longer significant when other cofactors, including health status assessed using the ACG methodology, were included in the analysis (Warner et al., 2018).

Other studies have also examined the association between ASD and exposure to procedures requiring general anesthesia. Two studies utilized Medicare datasets and created a composite outcome of developmental and behavioral disorders that included autism (DiMaggio et al., 2009, 2011). They found associations between exposures and this composite outcome, but did not separately analyze autism as an outcome. A study using a sibling cohort design involving children from Puerto Rico found in univariate analysis no difference in exposures to anesthesia prior to the age of diagnosis, ascertained by parent interview, between children diagnosed with ASD and their non-affected siblings (Creagh et al., 2016). A matched cohort study of children in Taiwan examined the relationship between exposure to procedures requiring anesthesia prior to age 2 years and the clinical diagnosis of ASD within 3 years of exposure, finding no association in adjusted analysis (hazard ratio = 0.93 [95% CI 0.57–1.53]) (Ko et al., 2015). We performed a sensitivity analysis to approximate their criteria of earlier ASD diagnosis (seeking ASD prior to the 7th birthday), which did not change our pattern of results.

Other analyses using a subset of this Olmsted County birth cohort (children born from 1976 to 1982 and from 1996 to 2000) found a significant association between the incidence of ADHD and receiving multiple, but not single, procedures requiring general anesthesia prior to age 3 years (Hu et al., 2017; Sprung et al., 2012). There is considerable overlap between symptoms of ADHD and ASD (Kushki et al., 2019; Llanes et al., 2020); approximately 15% of youth with ADHD are diagnosed with comorbid ASD, and ADHD is the most common comorbidity in children with ASD (diagnosed in up to 70% of children, a rate consistent with our results) (Antshel & Russo, 2019). However, a sensitivity analysis segregating those children with ASD with and without ADHD did not affect the pattern of results, suggesting the specificity of the previously-noted association between multiple anesthesia exposures and ADHD.

In addition to the potential for confounding shared by all observational designs and extensively discussed in this field of investigation (Davidson et al., 2015; Hansen, 2015; Wilder et al., 2009), this analysis has other notable limitations. Although most characteristics of Olmsted County residents are similar to the rest of Minnesota, some differ from the United States population as a whole (St Sauver et al., 2012), such that the results may not apply to other populations. The number of cases may be insufficient to detect a small odds ratio. However, the fact that there was no evidence of any dose–response relationship between exposure intensity and incidence (e.g., the OR for multiply-exposed children was numerically less than the OR for singly-exposed children) makes any significant association unlikely. If the biological processes causing ASD are operative prior to anesthesia exposure, this would bias towards not finding an association between exposure and ASD, even if exposure was a factor in some children. Finally, compared with our prior analyses using subsets of this birth cohort (Flick et al., 2012; Hu et al., 2017; Sprung et al., 2012; Warner et al., 2018; Wilder et al., 2009), some potentially relevant covariates such as parental education and socio-economic status were not available for all cohort members, and thus were not included as adjustors in this analysis.

In conclusion, these findings add to the evidence that exposure of young children to procedures requiring general anesthesia is not itself associated with an increased likelihood of developing ASD in later life, a finding reassuring for both families and physicians.

Acknowledgments

We are grateful to Drs. Slavica Katusic, MD, Emeritus Consultant, Mayo Clinic, Rochester, MN who is now retired and conducted earlier studies to identify incident autism cases in the Olmsted County. We also thank Jenna Steege, APRN, CRNA, Mayo Clinic, Rochester, MN for support in this research project.

Funding

This study used the resources of the Rochester Epidemiology Project (REP) medical records-linkage system, which is supported by the National Institute on Aging (NIA; AG 058738), by the Mayo Clinic Research Committee, and by fees paid annually by REP users. The content of this article is solely the responsibility of the authors and does not represent the official views of the National Institutes of Health (NIH) or the Mayo Clinic. The funders had no role in the design and conduct of the research.

Footnotes

Conflict of interest The authors have no conflict of interest to disclose.

References

  1. Antshel KM, & Russo N (2019). Autism sectrum disorders and ADHD: Overlapping phenomenology, diagnostic issues, and treatment considerations. Current Psychiatry Reports, 21(5), 34. 10.1007/s11920-019-1020-5 [DOI] [PubMed] [Google Scholar]
  2. Atladottir HO, Thorsen P, Schendel DE, Ostergaard L, Lemcke S, & Parner ET (2010). Association of hospitalization for infection in childhood with diagnosis of autism spectrum disorders: A Danish cohort study. Archives of Pediatric and Adolescent Medicine, 164(5), 470–477. 10.1001/archpediatrics.2010.9 [DOI] [PubMed] [Google Scholar]
  3. Baxter AJ, Brugha TS, Erskine HE, Scheurer RW, Vos T, & Scott JG (2015). The epidemiology and global burden of autism spectrum disorders. Psychological Medicine, 45(3), 601–613. 10.1017/S003329171400172X [DOI] [PubMed] [Google Scholar]
  4. Chien LN, Lin HC, Shao YH, Chiou ST, & Chiou HY (2015). Risk of autism associated with general anesthesia during cesarean delivery: A population-based birth-cohort analysis. Journal of Autism and Developmental Disorders, 45(4), 932–942. 10.1007/s10803-014-2247-y [DOI] [PubMed] [Google Scholar]
  5. Christensen DL, Maenner MJ, Bilder D, Constantino JN, Daniels J, Durkin MS, Fitzgerald RT, Kurzius-Spencer M, Pettygrove SD, Robinson C, Shenouda J, White T, Zahorodny W, Pazol K, & Dietz P (2019). Prevalence and characteristics of autism spectrumd disorder among children aged 4 years—early autism and developmental disabilities monitoring network, seven sites, United States, 2010, 2012, and 2014. MMWR Surveilancel Summary, 68(2), 1–19. 10.15585/mmwr.ss6802a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Coleman K, Robertson ND, Dissen GA, Neuringer MD, Martin LD, Cuzon Carlson VC, Kroenke C, Fair D, & Brambrink AM (2016). Isoflurane anesthesia has long-term consequences on motor and behavioral development in infant rhesus macaques. Anesthesiology, 127(1), 74–94. 10.1097/ALN.0000000000001383 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Creagh O, Torres H, Rivera K, Morales-Franqui M, Altieri-Acevedo G, & Warner D (2016). Previous exposure to anesthesia and autism spectrum disorder (ASD): A Puerto Rican population-based sibling cohort study. Boletin De La Asociacion Medica De Puerto Rico, 108(2), 73–80. [PubMed] [Google Scholar]
  8. Curran EA, Dalman C, Kearney PM, Kenny LC, Cryan JF, Dinan TG, & Khashan AS (2015a). Association between obstetric mode of delivery and autism spectrum disorder: A population-based sibling design study. JAMA Psychiatry, 72(9), 935–942. 10.1001/jamapsychiatry.2015.0846 [DOI] [PubMed] [Google Scholar]
  9. Curran EA, O’Neill SM, Cryan JF, Kenny LC, Dinan TG, Khashan AS, & Kearney PM (2015b). Research review: Birth by caesarean section and development of autism spectrum disorder and attention-deficit/hyperactivity disorder: A systematic review and meta-analysis. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 56(5), 500–508. 10.1111/jcpp.12351 [DOI] [PubMed] [Google Scholar]
  10. Currenti SA (2010). Understanding and determining the etiology of autism. Cellular and Molecular Neurobiology, 30(2), 161–171. 10.1007/s10571-009-9453-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Davidson AJ, Becke K, de Graaff J, Giribaldi G, Habre W, Hansen T, Hunt RW, Ing C, Loepke A, McCann ME, Ormond GD, Pini Prato A, Salvo I, Sun L, Vutskits L, Walker S, & Disma N (2015). Anesthesia and the developing brain: A way forward for clinical research. Paediatric Anaesthesia, 25(5), 447–452. 10.1111/pan.12652 [DOI] [PubMed] [Google Scholar]
  12. de la Torre-Ubieta L, Won H, Stein JL, & Geschwind DH (2016). Advancing the understanding of autism disease mechanisms through genetics. Nature Medicine, 22(4), 345–361. 10.1038/nm.4071 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. American Psychiatric Association (2000). Diagnostic and statistical manual of mental disorders (4th edn., Text Revision). Washington, DC: Author. [Google Scholar]
  14. DiMaggio C, Sun LS, Kakavouli A, Byrne MW, & Li G (2009). A retrospective cohort study of the association of anesthesia and hernia repair surgery with behavioral and developmental disorders in young children. Journal of Neurosurgical Anesthesioogyl, 21(4), 286–291. 10.1097/ANA.0b013e3181a71f11 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. DiMaggio C, Sun LS, & Li G (2011). Early childhood exposure to anesthesia and risk of developmental and behavioral disorders in a sibling birth cohort. Anesthesia and Analgesia, 113(5), 1143–1151. 10.1213/ANE.0b013e3182147f42 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Doja A, & Roberts W (2006). Immunizations and autism: A review of the literature. Canadian Journal of Neurological Science, 33(4), 341–346. 10.1017/s031716710000528x [DOI] [PubMed] [Google Scholar]
  17. Flick RP, Katusic SK, Colligan RC, Wilder RT, Voigt RG, Olson MD, Sprung J, Weaver AL, Schroeder DR, & Warner DO (2012). Cognitive and behavioral outcomes after early exposure to anesthesia and surgery. Pediatrics, 128(5), e1053–1061. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Freitag CM (2007). The genetics of autistic disorders and its clinical relevance: A review of the literature. Molecular Psychiatry, 12(1), 2–22. 10.1038/sj.mp.4001896 [DOI] [PubMed] [Google Scholar]
  19. Gardener H, Spiegelman D, & Buka SL (2011). Perinatal and neonatal risk factors for autism: A comprehensive meta-analysis. Pediatrics, 128(2), 344–355. 10.1542/peds.2010-1036 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Gleich SJ, Flick R, Hu D, Zaccariello MJ, Colligan RC, Katusic SK, Schroeder DR, Hanson A, Buenvenida S, Wilder RT, Sprung J, Voigt RG, Paule MG, Chelonis JJ, & Warner DO (2014). Neurodevelopment of children exposed to anesthesia: Design of the mayo anesthesia safety in kids (MASK) study. Contemporary Clinical Trials, 41, 45–54. 10.1016/j.cct.2014.12.020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Gruschow SM, Yerys BE, Power TJ, Durbin DR, & Curry AE (2019). Validation of the use of electronic health records for classification of ADHD Status. Journal of Attention Disorders, 23(13), 1647–1655. 10.1177/1087054716672337 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Hansen SN, Schendel DE, & Parner ET (2015). Explaining the increase in the prevalence of autism spectrum disorders: The proportion attributable to changes in reporting practices. JAMA Pediatrics, 169(1), 56–62. 10.1001/jamapediatrics.2014.1893 [DOI] [PubMed] [Google Scholar]
  23. Hansen TG (2015). Anesthesia-related neurotoxicity and the developing animal brain is not a significant problem in children. Paediatric Anaesthesia, 25(1), 65–72. 10.1111/pan.12548 [DOI] [PubMed] [Google Scholar]
  24. Hu D, Flick RP, Gleich SJ, Scanlon MM, Zaccariello MJ, Colligan RC, Katusic SK, Schroeder DR, Hanson AC, Buenvenida SL, Wilder RT, Sprung J, & Warner DO (2016). Construction and characterization of a population-based cohort to study the association of anesthesia exposure with neurodevelopmental outcomes. PLoS ONE, 11(5), e0155288. 10.1371/journal.pone.0155288 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Hu D, Flick RP, Zaccariello MJ, Colligan RC, Katusic SK, Schroeder DR, Hanson AC, Buenvenida SL, Gleich SJ, Wilder RT, Sprung J, & Warner DO (2017). Association between exposure of young children to procedures requiring general anesthesia and learning and behavioral outcomes in a population-based birth cohort. Anesthesiology, 127(2), 227–240. 10.1097/ALN.0000000000001735 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Huberman Samuel M, Meiri G, Dinstein I, Flusser H, Michaelovski A, Bashiri A, & Menashe I (2019). Exposure to general anesthesia may contribute to the association between Cesarean delivery and autism spectrum disorder. Journal of Autism and Developmental Disorders, 49(8), 3127–3135. 10.1007/s10803-019-04034-9 [DOI] [PubMed] [Google Scholar]
  27. Ing C, Sun M, Olfson M, DiMaggio CJ, Sun LS, Wall MM, & Li G (2017). Age at exposure to surgery and anesthesia in children and association with mental disorder diagnosis. Anesthesia and Analgesia. 10.1213/ANE.0000000000002423 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Jevtovic-Todorovic V, Hartman RE, Izumi Y, Benshoff ND, Dikranian K, Zorumski CF, Olney JW, & Wozniak DF (2003). Early exposure to common anesthetic agents causes widespread neurodegeneration in the developing rat brain and persistent learning deficits. Journal of Neuroscience, 23(3), 876–882. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Ko WR, Huang JY, Chiang YC, Nfor ON, Ko PC, Jan SR, Lung CC, Chang HC, Lin LY, & Liaw YP (2015). Risk of autistic disorder after exposure to general anaesthesia and surgery: A nationwide, retrospective matched cohort study. European Journal of Anaesthesiology, 32(5), 303–310. 10.1097/EJA.0000000000000130 [DOI] [PubMed] [Google Scholar]
  30. Ko WR, Liaw YP, Huang JY, Zhao DH, Chang HC, Ko PC, Jan SR, Nfor ON, Chiang YC, & Lin LY (2014). Exposure to general anesthesia in early life and the risk of attention deficit/hyperactivity disorder development: A nationwide, retrospective matched-cohort study. Paediatric Anaesthesia, 24(7), 741–748. 10.1111/pan.12371 [DOI] [PubMed] [Google Scholar]
  31. Kushki A, Anagnostou E, Hammill C, Duez P, Brian J, Iaboni A, Schachar R, Crosbie J, Arnold P, & Lerch JP (2019). Examining overlap and homogeneity in ASD, ADHD, and OCD: A data-driven, diagnosis-agnostic approach. Translational Psychiatry, 9(1), 318. 10.1038/s41398-019-0631-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Llanes E, Blacher J, Stavropoulos K, & Eisenhower A (2020). Parent and teacher reports of comorbid anxiety and ADHD symptoms in children with ASD. Journal of Autism and Developmental Disorders, 50(5), 1520–1531. 10.1007/s10803-018-3701-z [DOI] [PubMed] [Google Scholar]
  33. Lundström S, Reichenberg A, Melke J, Råstam M, Kerekes N, Lichtenstein P, Gillberg C, & Anckarsäter H (2015). Autism spectrum disorders and coexisting disorders in a nationwide Swedish twin study. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 56(6), 702–710. 10.1111/jcpp.12329 [DOI] [PubMed] [Google Scholar]
  34. Lyall K, Croen L, Daniels J, Fallin MD, Ladd-Acosta C, Lee BK, Park BY, Snyder NW, Schendel D, Volk H, Windham GC, & Newschaffer C (2017). The changing epidemiology of autism spectrum disorders. Annual Review of Public Health, 38, 81–102. 10.1146/annurev-publhealth-031816-044318 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Marques F, Brito MJ, Conde M, Pinto M, & Moreira A (2014). Autism spectrum disorder secondary to enterovirsu encephalitis. Journal of Child Neurology, 29(5), 708–714. 10.1177/0883073813508314 [DOI] [PubMed] [Google Scholar]
  36. Modabbernia A, Velthorst E, & Reichenberg A (2017). Environmental risk factors for autism: An evidence-based review of systematic reviews and meta-analyses. Molecular Autism, 8, 13. 10.1186/s13229-017-0121-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Muhle RA, Reed HE, Stratigos KA, & Veenstra-Vander-Weele J (2018). The emerging clinical neuroscience of autism spectrum disorder: A review. JAMA Psychiatry, 75(5), 514–523. 10.1001/jamapsychiatry.2017.4685 [DOI] [PubMed] [Google Scholar]
  38. Myers SM, Voigt RG, Colligan RC, Weaver AL, Storlie CB, Stoeckel RE, Port JD, & Katusic SK (2019). Autism spectrum disorder: Incidence and time trends over two decades in a population-based birth cohort. Journal of Autism and Developmental Disorders, 49(4), 1455–1474. 10.1007/s10803-018-3834-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Newschaffer CJ, Croen LA, Daniels J, Giarelli E, Grether JK, Levy SE, Mandell DS, Miller LA, Pinto-Martin J, Reaven J, Reynolds AM, Rice CE, Schendel D, & Windham GC (2007). The epidemiology of autism spectrum disorders. Annual Review of Public Health, 28, 235–258. 10.1146/annurev.publhealth.28.021406.144007 [DOI] [PubMed] [Google Scholar]
  40. Sagi-Dain L, Kedar R, Bardicef M, & Riskin S (2020). Numerous confounders affecting the alleged association between cesarean deliveries under general anesthesia and autism spectrum disorder. Journal of Autism and Developmental Disorders, 50(2), 688–690. 10.1007/s10803-019-04247-y [DOI] [PubMed] [Google Scholar]
  41. Sedighnejad A, Soltanipour S, Saberi A, Kousha M, Bidabadi E, Biazar G, & Naderi N (2020). Risk of attention deficit hyperactivity disorder dfter early exposure to general anesthesia: A case control study. Iranian Journal of Pediatrics, 30(3), e99976. [Google Scholar]
  42. Shi Y, Hu D, Rodgers EL, Katusic SK, Gleich SJ, Hanson AC, Schroeder DR, Flick RP, & Warner DO (2018). Epidemiology of general anesthesia prior to age 3 in a population-based birth cohort. Paediatric Anaesthesia, 28(6), 513–519. 10.1111/pan.13359 [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Shi Y, Schulte PJ, Hanson AC, Zaccariello MJ, Hu D, Crow S, Flick RP, & Warner DO (2020). Utility of medical record diagnostic codes to ascertain attention-deficit/hyperactivity disorder and learning disabilities in populations of children. BMC Pediatrics, 20(1), 510. 10.1186/s12887-020-02411-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Sprung J, Flick RP, Katusic SK, Colligan RC, Barbaresi WJ, Bojanić K, Welch TL, Olson MD, Hanson AC, Schroeder DR, Wilder RT, & Warner DO (2012). Attention-deficit/hyperactivity disorder after early exposure to procedures requiring general anesthesia. Mayo Clinic Proceedings, 87(2), 120–129. 10.1016/j.mayocp.2011.11.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. St Sauver JL, Grossardt BR, Yawn BP, Melton LJ 3rd., Pankratz JJ, Brue SM, & Rocca WA (2012). Data resource profile: The Rochester Epidemiology Project (REP) medical records-linkage system. International Journal of Epidemiology, 41(6), 1614–1624. 10.1093/ije/dys195 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. van der Meer JM, Oerlemans AM, van Steijn DJ, Lappenschaar MG, de Sonneville LM, Buitelaar JK, & Rommelse NN (2012). Are autism spectrum disorder and attention-deficit/hyperactivity disorder different manifestations of one overarching disorder? Cognitive and symptom evidence from a clinical and population-based sample. Journal of the American Academy of Child and Adolescent Psychiatry, 51 (11), 1160–1172.e1163. 10.1016/j.jaac.2012.08.024 [DOI] [PubMed] [Google Scholar]
  47. Warner DO, Zaccariello MJ, Katusic SK, Schroeder DR, Hanson AC, Schulte PJ, Buenvenida SL, Gleich SJ, Wilder RT, Sprung J, Hu D, Voigt RG, Paule MG, Chelonis JJ, & Flick RP (2018). Neuropsychological and behavioral outcomes after exposure of young children to procedures requiring general anesthesia: The Mayo Anesthesia Safety in Kids (MASK) study. Anesthesiology, 129(1), 89–105. 10.1097/ALN.0000000000002232 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Weiner JP, Dobson A, Maxwell SL, Coleman K, Starfield B, & Anderson GF (1996). Risk-adjusted Medicare capitation rates using ambulatory and inpatient diagnoses. Health Care Financing Review, 17(3), 77–99. [PMC free article] [PubMed] [Google Scholar]
  49. Weiner JP, Starfield BH, Steinwachs DM, & Mumford LM (1991). Development and application of a population-oriented measure of ambulatory care case-mix. Medical Care, 29(5), 452–472. [DOI] [PubMed] [Google Scholar]
  50. Wilder RT, Flick RP, Sprung J, Katusic SK, Barbaresi WJ, Mickelson C, Gleich SJ, Schroeder DR, Weaver AL, & Warner DO (2009). Early exposure to anesthesia and learning disabilities in a population-based birth cohort. Anesthesiology, 110(4), 796–804. 10.1097/01.anes.0000344728.34332.5d [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Zhang T, Sidorchuk A, Sevilla-Cermeño L, Vilaplana-Pérez A, Chang Z, Larsson H, Mataix-Cols D, & Fernández de la Cruz L (2019). Association of Cesarean delivery with risk of neurodevelopmental and psychiatric disorders in the offspring: A systematic review and meta-analysis. JAMA Network Open, 2(8), e1910236. 10.1001/jamanetworkopen.2019.10236 [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES