Sarnoff Mednick - US grants

Winter births have been repeatedly shown to produce an excess of later-diagnosed schizophrenics. Viral infections, in view of their seasonality, have been prominent among the proposed explanations of the winter birth findings. As part of a larger project, we proposed to follow-up a Helsinki cohort of 6,097 consecutive deliveries which occurred during the 1957 type-A2 influenza epidemic. These individuals, 27 years of age at the time of the search, would be followed-up in the Finnish Psychiatric Registers for any evidence of mental illness, including schizophrenia. It was hypothesized that those whose mothers suffered a perinatal influenza infection would evidence an elevated risk of schizophrenia as compared to those whose mothers were free of an infection. In addition, it was hypothesized that among those who experienced an influenza infections, the schizophrenia diagnoses would be concentrated among those who were genetically predisposed among those with a parental schizophrenia diagnosis. It is the specific aim of this proposal to undertake a pilot study to determine the feasibility of pursuing the larger study proposed above. The pilot study will test the hypothesis that systematic examination of the records of Helsinki mental hospitals will reveal an unusual bulge of cases of schizophrenia among those born at or near the time of the 1957 type-A2 influenza epidemic. No study to date has directly examined, in a large population, the possible effects of a perinatal influenza infection on mental illness. If relations are found, the implications for primary prevention will be valuable.

This application for an RSA requests support to complete two related projects. The first is the Danish high risk for schizophenia study; this is a prospective study of children of schizophenic mothers. The are at high risk of becoming schizophrenic. They were examined in 1962, 1967, and 1972. A follow-up is now planned for 1986 when they will average 39 years. Data analyses to date suggest that genetic predisposition for schizophrenia interacts wit environmental insult (perinatal and early childhood trauma) to increase the probability of schizophrenic breakdown. The 1986 follow-up will enable us to test the reliability of these findings. The perinatal data relate to adult CT scan-detected-brain anomalies. Data from other laboratories and from the HR study suggest the possible role of fetal viral infection in the perinatal difficulties. An ongoing population study in Finland is described in which adult psychopathological outcomes are being related to viral infections contracted during pregnancy. Positive preliminary findings in this study have encouraged plans to continue this line of work.

This proposal addresses several problems with the fetal virus-adult schizophrenia research. 1. In almost all of the studies, an increased risk for schizophrenia is related to exposure to the epidemic rather than empirical evidence of an influenza infection. The one exception is our own Helsinki study, but the number of subjects for that study was small. In the context of a population of 5387 Finnish schizophrenics, we propose to determine whether recorded instances of influenza tend to segregate in the second trimester of gestation. 2. We do not know what changes in the fetus, attributable to the viral infection, increase risk for later schizophrenia. We are addressing this problem by examining the neurointegrative and neuromotor functioning of infants whose mothers have a serologically confirmed infection at a known time during gestation. 3. The virus-schizophrenia correlation does not establish whether influenza is the only infectious disease which may increase risk for later schizophrenia. We are examining the effect of a serologically-verified mycoplasma pneumonia infection during gestation on the neurointegrative functioning of the infant. 4. The influenza viral infection is a relatively specific "cause". We attempted to determine whether we could detect a narrowly-defined resultant clinical syndrome. We found that the second-trimester infection produced schizophrenics with marked delusions and reduced levels of thought disorder. We propose to examine the brains of these schizophrenics with MRI to determine whether we can identify a similarly narrowly defined neuroanatomical disturbance underlying the specific clinical syndrome. 5. We propose to examine the possibility that the viral effect is due to antibody reactions to specific fetal brain tissues. We plan to immunostain second trimester fetal neural tissue with IgG and IgM class antibodies and attempt to determine which parts of the brain are recognized by these antibodies. Pilot work has implicated the hippocampus and the basal ganglia.

Pneumoencephalography, brain imaging studies and neuropathology research have demonstrated anomalies in the brains of schizophrenics. Empirical studies have suggested that the source of these brain anomalies may be complications of pregnancy and delivery. The research on obstetrical complications (OCs) and schizophrenia has utilized two basic methods to obtain information on the OCs of the adult schizophrenic subjects: recall by the mother of the adult schizophrenic and clinical records produced by mid-wives or obstetrical nurses. The clinical records, have the great advantage of being unbiased and prospective. Studies have, however, shown clinical records to be incomplete; almost no information is included on pregnancy complications. In view of this, in the 60's we prepared two perinatal research samples for utilization at this time. Both contain a data bank of excellent information on pregnancy and delivery collected during fetal development and at birth. The first is a Finnish sample of 171 children with schizophrenic mm others (HR offspring) and matched controls; the second is a birth cohort of 9,006 consecutive deliveries which formed the basis of a very carefully conducted Danish perinatal research project. Both samples were born in the early 60's; the offspring are now adults. We intend to examine them with diagnostic interviews and CT-scans to test the following chief hypotheses: 1) adult schizophrenics will have suffered a higher rate of severe OCs than controls. 2) delivery complications will interact with genetic risk to increase the risk of schizophrenia and the risk of ventriculomegaly. 3) genetic risk will relate positively to increase in cortical and cerebellar dysgenesis. 4) type and severity, of OCs in HR offspring will influence the type of schizophrenic symptoms displayed in adulthood.

DESCRIPTION (Adapted from applicant's abstract): We propose a "two-hit" working model to guide our research on the etiology of schizophrenia. The first proposed hit consists of a disruption of fetal neural development induced by genetic or teratogenic factors. The second hit may take the form of 1) brain damage caused by delivery complications or 2) stressful, non-optimal rearing circumstances. The nature of the second hit may determine the course of illness. We plan to test this working model in three projects: Project 1. 1962 High Risk Project. This project has followed a sample of 311 children of schizophrenic mothers and controls for the past 33 years. We have descriptions of the lives and the biological and social characteristics for each of the individuals. Of the 311, 33 have become schizophrenic. We propose to examine this data bank to attempt to understand what combination of characteristics and events is associated with schizophrenic outcome. Project 2. Fetal Virus Project. We have recently found that genetic predisposition and teratogenic factors, the two first hits, seem to be associated with different clinical outcomes within the schizophrenia diagnostic umbrella. We propose to replicate this finding in a larger population to differentiate effects of genetic vs. teratogenic factors. Project 3. Perinatal Disturbance and Adult Schizophrenia. We have been following a Danish Perinatal Cohort of 9,125 consecutive deliveries. We have ascertained their adult psychiatric status. In two studies we have noted that: 1) early institutional rearing and 2) delivery complications predict to schizophrenia among those at genetic risk (with schizophrenic parents).

There is a considerable body of research which finds that schizophrenics engage in elevated levels of violent behavior. Rates of violence among schizophrenics are frequently compared to rates in the general population. Schizophrenics, however, tend to have relatively low socioeconomic status (SES) and, depending on the sampling frame, may tend to be relatively young and male. Schizophrenics also tend to abuse alcohol and drugs. Since low SES, youth, maleness and substance abuse are characteristic attributes of violent offenders, a part of the excess of violence noted for schizophrenics may be related to the methods of sampling in these studies. There are three positions which may be differentiated in understanding the empirical findings of excessive levels of violence among schizophrenics: 1. The findings are due to sampling errors. When proper demographic controls and controls for substance abuse are utilized, the differences will disappear. 2. Generally, schizophrenics are not violent. Only schizophrenics in the midst of an acute episode might commit violent acts. 3. Antisocial behavior, including violence, is a part of the schizophrenia genetic spectrum. We propose to examine these three positions in the context of two cohorts. The first is a total birth cohort of all individuals born between January 1, 1944 and December 31, 1947 in Denmark (N=358,180). Due to the existence of highly accurate national registers, data are available on every arrest for violence and every hospitalization for mental illness that occurred for individuals in this cohort through the age of 44-47 years. This cohort will be utilized to test hypotheses related to the first two positions listed above. Hypotheses related to the third position will be examined in the context of a total cohort of adoptions (The Copenhagen Adoption Cohort) which includes all nonfamilial adoptions (N= 14,427) in Denmark, 1924-1947. Violent offending in the biological parents will be correlated with schizophrenia in the adopted-away offspring.