Are Attributes of Pregnancy and the Delivery Room Experience Related to Development of Autism? A Review of the Perinatal and Labor Risk Factors and Autism

Table 1

Summary of autism studies.

Study	Summary/major findings
Bilder et al. [3]	Population study based on birth records in Utah . Study found that an association between ASD and breech presentation may have a common etiology. Authors found that advanced maternal age and parity occurred more frequently in ASD children. Primary cesarean delivery was a significant perianal factor. No significant associations were established between neonatal factors and ASD.
Roberts et al. [4]	Population-based longitudinal cohort study with 116,430 women. Maternal abuse exposure in childhood was found to be directly correlated with increased risk of autism in offspring. Amount of abuse exposure also directly correlated with level of autism.
Christensen et al. [6]	Population study of all Denmark children born alive from the years 1996 to 2006. The study, which utilized national records, found a significant absolute risk between valproate exposure during pregnancy and autism in offspring.
Williams et al. [7]	Population-based data were used to examine risk factors for autism. Risk factors found to be significantly associated with autism were male gender, premature birth, maternal age greater than or equal to 35, mother not being born in Australia, and multiple birth.
Glasson et al. [8]	Association of autism spectrum disorders with perinatal obstetric factor was analyzed for a cohort of children. It was determined that a single obstetric factor is not likely to cause autism. Increased obstetric factors may be associated with autism because of underlying genetic and environmental factors.
Haglund and Källén [9]	Obstetrical information was reviewed for 250 children diagnosed with autism or Asperger syndrome using the Swedish Medical Birth Registry. Authors found that obstetric factors like prematurity, macrosomia, and low Apgar scores were associated with autism but not with Asperger syndrome. Authors also found positive association between autism and mothers born outside Nordic countries.
Lampi et al. [10]	Validity of the Finnish registry for diagnoses of autism was examined. A sample of 95 people from the Finnish Hospital Discharge Register who had been diagnosed with childhood autism or pervasive developmental disorder/pervasive developmental disorder not otherwise specified or Asperger’s syndrome was examined. It was determined that the Finnish registry was valid for childhood autism diagnoses.
Gunnes et al. [11]	This study examined if there was an increased risk of autistic disorder in children born within a year of a sibling. Norway registry data were utilized to analyze 223,476 sibling pairs. An association between interpregnancy intervals less than one year and increased risk of autistic disorder was found.
Langridge et al. [12]	Study of 383,153 singletons born alive in Western Australia was conducted. Increased risk of autism spectrum disorder was found with the following factors: hypertension in pregnancy, asthma, antepartum hemorrhage (some types), breech presentation, need for resuscitation at birth, elective cesarean sections, preterm birth, and urinary tract infections.
Mamidala et al. [13]	A retrospective cohort of 942 children from India was examined. Twenty-five pre-, peri-, and neonatal risk factors of autism spectrum disorder were examined. Advanced maternal age, birth asphyxia, fetal distress, delayed birth cry, gestational respiratory infections, preterm birth, labor complications, and neonatal jaundice were found be significantly associated with increased risk of autism spectrum disorder.
Polo-Kantola et al. [14]	Seventy-one postmenopausal women were examined to evaluate estrogen replacement therapy and its effects on nocturnal periodic limb movements. It was found that estrogen replacement therapy does not alter nocturnal periodic limb movement incidence or intensity when it is given in doses used for climacteric symptoms.
May-Benson et al. [15]	Retrospective chart review was utilized to investigate 1000 children diagnosed with sensory processing disorder and 467 with autism spectrum disorder as well as sensory processing disorder. Jaundice was found to be three to four times more prevalent in sensory processing disorder and autism spectrum disorder children compared to typical children. Other factors found to be of higher incidence in these children were breech position, cord wrap and prolapse, assisted delivery methods, and high birth weight.
Kuzniewicz et al. [16]	A retrospective cohort of subjects born at greater than or equal to 24 weeks in Kaiser Permanente Northern California hospitals ( = 195,021) was investigated. Autism spectrum disorder was found to be about 3 times more prevalent in infants <27 weeks than in infants born term. Shorter gestation time was found to be directly associated with increased risk of autism spectrum disorder. Intracranial hemorrhage and high frequency ventilation were also associated with autism spectrum disorder for infants born <34 weeks.
Al-Farsi et al. [17]	A case control study was performed with 102 autism spectrum disorder subjects and 102 healthy control subjects. Association between late initiation of breastfeeding, bottle-feeding, prelacteal feeding, and nonintake of colostrum was found to be associated with autism spectrum disorder. Decreased risk of autism spectrum disorder was found with exclusive breastfeeding and continued breastfeeding.
Wilkerson et al. [18]	The authors conducted a study involving mothers of 183 mothers of autistic children and 209 mothers of normal children. The mothers in the study completed a Maternal Perinatal Scale which surveyed complications of pregnancies and medical conditions. It was found that 65% of autistic cases could be grouped using differences in gestational age, maternal morphology, and intrauterine stress. In addition, urinary tract infection, high temperatures, and depression also contributed to the separation of groups.
Zhang et al. [19]	A case control study was conducted of 190 children in China with and without autism. Nine risk factors were found to be associated with autism: maternal second hand smoke exposure, advanced paternal age at delivery (>30 years of age), maternal medical conditions not related to the pregnancy, gravidity >1, mother being unhappy, nuchal cord, gestational complications, and edema.
Volk et al. [20]	A study conducted with 252 cases of autism spectrum disorder and 156 control subjects was examined. MET rs1858830 CC genotype as well as air pollution exposure may increase risk of autism spectrum disorder.
Roberts et al. [21]	A study was conducted that examined the association of air pollution exposure and autism spectrum disorder. This has 325 cases and 22,101 controls. Significant association between air pollutants and autism spectrum disorder was found.
Bergman et al. [22]	This study examined 22 women who were given terbutaline intravenously before delivery and were delivered by elective cesarean section. The authors found that venous umbilical blood slowly but continuously uptook terbutaline.
Zerrate et al. [23]	This study examined the effects of terbutaline on newborn rats to determine if terbutaline is a risk factor for autism spectrum disorder. The authors found that overstimulation of the beta-2 adrenergic receptor, which terbutaline is an agonist for, is associated with behavioral abnormalities which are similar to that of autism.
Connors et al. [24]	Dizygotic twin data were analyzed for association of autism spectrum disorders and terbutaline exposure. It was found that continuous exposure for 2 weeks or more of terbutaline has an increased risk for autism spectrum disorders in twins that are dizygotic. In addition, a significant association was found between 16G and 27E polymorphisms in the beta-2 adrenergic receptor and autism.
Cheslack-Postava et al. [25]	It was examined if there was association with the ADRB2 allele for the gene coding for the beta-2 adrenergic receptor and autism. This study found that the Glu27 allele of the ADRB2 gene may be involved with increased risk of autism.
Croen et al. [26]	A case control study with children born from the years 1995–1999 at hospitals in California was conducted to examine the association between autism spectrum disorder and prenatal exposure to beta-2 adrenergic receptor agonists like terbutaline. It was found that terbutaline was the only beta-2 adrenergic receptor agonist that was associated with increased risk of autism spectrum disorder, but only when exposure was greater than or equal to 2 days during the third trimester.
Witter et al. [27]	The authors of this paper found that overstimulation of the beta-2 adrenergic receptor during certain periods of time in prenatal development may have an association with autism spectrum disorder due to induced behavioral and functional teratogenesis.
Abisror et al. [28]	This retrospective study compared outcomes of babies born from mothers diagnosed with primary antiphospholipid syndrome and mothers diagnosed with systemic lupus erythematosus . Autism spectrum disorder was found in the offspring of mothers diagnosed with antiphospholipid syndrome.
Brown et al. [29]	A cohort study was conducted that examined inflammation during pregnancy and correlation with autism spectrum disorder of offspring. Significant association was found between offspring autism and increased C-reactive protein levels in the mother.
Banach et al. [30]	This study investigated the relationship between autism and gender as well as nonverbal IQ. Single child daughters with autism showed lower nonverbal IQ than single child sons with autism. There appears to be variability in nonverbal IQ and gender in autism.
Rasalam et al. [31]	The authors of this study explored in long term in utero exposure to anticonvulsant drugs and risk of autism using 260 children. An association between in utero exposure to anticonvulsant drugs and autism spectrum disorder development was found.
Surén et al. [32]	A population-based study with 85,176 children was conducted to examine relationship between prenatal supplements of folic acid and risk of development of autism spectrum disorder. The authors found an association between folic acid supplementation during pregnancy and decreased risk of autism spectrum disorder in offspring.
Davis et al. [33]	A study utilizing records of 70 subjects exposed to cocaine in utero was conducted to examine developmental abnormalities. Autism was found in high frequency in offspring of mothers using cocaine during pregnancy (11.4%).
Tran et al. [34]	This was a population-based study that utilized the Finnish Medical Birth Register. The study examined risk of autism spectrum disorder in offspring of mothers who smoke tobacco during the entire length of pregnancy. The authors found no association between autism and maternal smoking.
Volk et al. [35]	The authors of this paper conducted a population-based study that was case controlled and looked for relationships between air pollution created by traffic, quality of air, and autism. The study had 279 subjects with autism and 245 control subjects. An association was found between autism and exposure to pollution caused by traffic.
Croen et al. [36]	This population-based case controlled study attempted to determine association between maternal antidepressant utilization during pregnancy and autism spectrum disorder in offspring. In utero exposure to antidepressants may be associated with autism development, but the authors suggest that more studies should be conducted.
Rai et al. [37]	The authors conducted a population-based case controlled study and examined risk of autism spectrum disorders in offspring of mothers using serotonin selective reuptake inhibitors and tricyclic antidepressants in pregnancy. An increased risk of autism was found in children exposed in utero to serotonin selective reuptake inhibitors and tricyclic antidepressants.
Simpson et al. [38]	This study utilized rats and examined serotonin levels and how alterations in serotonin may affect neurodevelopment and development of autism. Cortical organization of rats was examined after serotonin selective reuptake inhibitor exposure pre- and postnatally. Disruptions in brain features were found as well as changes in behavior. The authors call for further research of serotonin selective reuptake inhibitor exposure and autism development in humans.
Hviid et al. [39]	A study utilizing 3892 subjects diagnosed with autism spectrum disorder was examined for association between serotonin selective reuptake inhibitor exposure in utero and autism spectrum disorder development. Significant association was not found but further studies are warranted.
Lyall et al. [40]	This was a nested case control study that utilized the Nurses’ Health Study II. The analysis found that therapy for reproductive assistance as well as infertility history did not show an increased risk of autism spectrum disorder in offspring unless the mother was 35 years of age or older.
Bay et al. [41]	This was a cohort study that used the Danish National Health Registers to investigate association between fertility treatments and autism spectrum disorder, as well as other mental disorders. A significant increase in risk of autism spectrum disorder was seen in offspring of mothers who utilized fertility treatments, but the increase in risk was low.
Mamidala et al. [42]	A study was conducted in India to explore hormonal interventions and the association with autism spectrum disorder. The study utilized 942 children, 471 of whom had been diagnosed with autism spectrum disorders. Analysis of data collected revealed that hormonal intervention used by the mother significantly increased the risk of offspring having autism spectrum disorder.
Matsuishi et al. [43]	The authors investigated the prevalence of autistic disorder diagnosed in children at St. Mary’s Hospital. The number of children identified having autistic disorder was higher than previously reported values in the nation of Japan.
Schendel and Bhasin [44]	A retrospective cohort study was conducted that compared weight at birth and gestational age of born children to rates of autism. Autism was found less than other developmental disabilities in preterm children and in low birth weight term children. However, autism did appear more when the child was also intellectually disabled.
Schrieken et al. [45]	This study used 96 children with autism spectrum disorder and a control of 163 children to examine if abnormalities in growth may be linked to autism spectrum disorder. Autism spectrum disorder was seen more in children who were first born or premature or had low birth weight. Results suggested that autism spectrum disorder may have a different reason for abnormal growth than premature/low birth weight children without autism spectrum disorder due to a difference in head circumference compared to height in the two groups of children.
Grandgeorge et al. [46]	The study utilized 422 children with autism spectrum disorder and 153 control children and measured attributes of these children. A significant finding of increased incidence of macrocephaly and microcephaly in autism spectrum disorder children was found.
Abel et al. [47]	A Scandinavian population study was performed that examined the relationship between growth of the fetus and autism spectrum disorder development. It was found that difference in fetal growth either too high or too low may be significant for development of autism spectrum disorder.
Bilder et al. [48]	This was a population-based autism spectrum disorder cohort study . Weight gain during pregnancy was found to be significantly associated with increased risk of autism spectrum disorder in offspring. However, prepregnancy weight and body mass index was not significantly associated with increased risk of autism spectrum disorder in offspring.
Moss and Chugani [49]	This study investigated prepregnancy weight and its relationship to autism in offspring. The authors found increased incidence of autism in children with very low birth weight, which correlated with mothers that were underweight or overweight.
Movsas et al. [50]	A secondary analysis of a study was performed to explore the link between abnormalities seen in the craniums of neonates via ultrasound and subsequent risk of autism spectrum disorder. It was found that particular types of white matter injury were associated with increased risk of autism spectrum disorder diagnoses, as well as ventricular enlargement.
Gregory et al. [51]	This study investigated the relationship between risk of autism diagnosis in induced and/or augmented births and births that were not. A total of 625,042 subjects were studied, of which greater than 5500 were documented as being autistic. The authors found increased incidence of autism diagnosis in induced births as well as augmented births.
Juul-Dam et al. [52]	This study investigated pre-, peri-, and neonatal factors in children that were diagnosed as autistic and compared these factors’ incidence with a control population. Most factors had about the same incidence between the two populations. However, increased incidence of uterine bleeding and decreased incidence of contraceptive use as well as decreased incidence of infection of the vagina of the mother were associated with autistic children.
Maramara et al. [53]	A study was conducted with 268 subjects in New Jersey diagnosed with autism spectrum disorder. A questionnaire was given to the subjects and some factors were compared to the New Jersey population statistics. It was found that the subjects of the study had significantly higher incidence of the following factors: advanced maternal age of 35 years or older, decreased weight at birth, hypoxia, long labor time, bleeding from the vagina, multiple gestation, and prematurity.
Eaton et al. [54]	Records from Denmark were used for this study, which attempted to find the relationship between obstetric complications and psychopathology risk with a focus on autism. The authors found an association and determined that birth weight and growth rate as well as Apgar score taken at 5 minutes were the best predictors of increased risk.
Hultman et al. [55]	This nested case control study utilized the Swedish Birth Register to utilize information of 408 children diagnosed with autism and examined perinatal risk factors. Associations were found between autism diagnosis and smoking during pregnancy, birth of the mother outside of the regions of North America and Europe, delivery via cesarean section, a low Apgar score taken at 5 minutes, congenital malformations, and being small for gestational age. Associations were not found between autism and circumference of the head, being one of twins, diabetes of the mother, or season during which birth took place.
Larsson et al. [56]	This case control study looked at the relationship between autism and perinatal factors, socioeconomic status, and the psychiatric history of parents. Data were used from the Denmark registries. No significant relationship was found between autism and socioeconomic status, weight for age of gestation, age of the parents, parity, and visits before birth. However, association of autism and environmental factors prenatally and the psychiatric history of parents was found.
Gardener et al. [57]	A meta-analysis was conducted using research available on three databases (PsycInfo, PubMed, and Embase) and 40 studies were utilized. More than 60 neonatal and perinatal risk factors were examined for their relationship to autism, and many were found to be associated. These included nonnormal presentation, hyperbilirubinemia, complications with the umbilical cord, Rh antigen or blood type incompatibility, distress with the fetus, anemia of the neonate, injury or trauma during birth, aspiration of meconium, multiple birth, difficulties with feeding, hemorrhage of the mother, nonoptimal 5-minute Apgar score, and congenital malformations.
Russell et al. [58]	This was a retrospective study that analyzed a UK cohort to investigate if social factors and demographic factors may be part of the diagnosis of autism spectrum disorder in children who are diagnosed. It was found that boys were more likely to be diagnosed with autism spectrum disorder than girls with similar symptoms.
Casanova et al. [59]	The purpose of this study was to examine brains of subjects previously diagnosed with autism spectrum disorder for abnormal thinning of the cerebral cortex as well as neuronal morphometric abnormalities. Foci of decreased width of cortex were found in brains of autism spectrum disorder individuals, as well as a decrease in size of the neurons of affected locations. The pathology seen correlates with focal cortical dysplasia diagnosis, which may provide an explanation for the high incidence of seizures in individuals with autism spectrum disorder.