Early life risk factors for schizophrenia - studies of foetal and perinatal complications

Abstract

Background: Despite intensive research for decades the causes of schizophrenia are largely unknown. A major lead in scientific work is the neurodevelopmental hypothesis i.e. a vulnerability for the illness is believed to be caused by developmental defects of the CNS during foetal life or childhood. The causes may be of genetic or environmental origin. Studies of the brain have been one source of the hypothesis.Additional support for the hypothesis was gathered from research on obstetric complications (OCs) and the later development of schizophrenia. Some thirty studies have been performed on the subject but have been criticised on methodological grounds. There has been a need for further research to investigate the association. The aims of the thesis were to examine in large samples:· Whether there is an association between OCs and psychiatric illness, especially schizophrenia. · Whether this possible risk is associated with any specific kind of complications, according to a proposed classification with four subgroups.· Whether the effect of OCs could be confounded by maternal psychotic illness. · Whether the association between OCs and schizophrenia is modified by gender, age at onset or maternal psychotic illness. · Whether the ICD-9 diagnoses of schizophrenia in the National Inpatient Register used in the studies are in accordance with the criteria-based DSM-IV diagnosis.Methods: OCs were classified into four groups in chronological order because alterations in the brain may differ in topography due to the maturation of the foetus/infant. In a nation-wide cohort study 507,516 persons were followed-up with regard to early age at onset schizophrenia (238 cases). In a case-control study the birth records were scrutinised for 524 cases of schizophrenia and 1,043 controls, born 1960 and onwards. In addition, 509 jaundiced infants and 509 controls were followed-up regarding psychiatric inpatient care. Data on maternal psychotic illness were assessed by record linkage. The diagnosis of schizophrenia was validated from 98 psychiatric case records. Results and conclusions: There was some support for an increased risk of schizophrenia associated with all four proposed groups of OCs: 1. foetal malnutrition/ growth impairment (pre-eclampsia RR 2.5; 95% CI 1.3-4.5, ponderal index<20 RR 3.4; 95% CI 1.1-10.5), 2. prematurity (RR 3.4; 95% CI 1.4-8.2), 3. signs of asphyxia at birth (Apgar score<7 OR 2.7; 95% CI 1.5-4.8), and 4. jaundice (OR 2.1; 95%CI 1.1-3.7), though the latter association was weaker. Further, an increased risk for psychiatric inpatient care was found among persons who had suffered from jaundice compared to non-jaundiced infants (OR 2.5; 95% CI 1.1-5.5). Thus, it seems as if complications at different points of time during foetal life, delivery, and may be neonatal life are associated with schizophrenia. This is supported by other studies of early life risk factors.There was an increased risk for psychotic mothers to have OCs (RR 1.2; 95%CI 1.2-1.3) but there was no support for maternal psychotic illness being a confounder of the association between OCs and schizophrenia. The effect from unknown confounders could not of course be ruled out. Other types of studies are needed to understand the causal relationships. The risk of schizophrenia associated with OCs was not modified by gender, age at onset of illness or the presence of maternal psychotic illness. Even larger studies are needed to further investigate interaction effects.The ICD-9 diagnoses of schizophrenia in the National Inpatient Register were true positives according to DSM-IV criteria in 86% of cases. This is satisfactory for epidemiological studies.