Peut-on prédire, dès l’enfance, les risques de développer la schizophrénie à l’âge adulte ? Une étude rétrospective centrée sur l’hypersensibilité prémorbide ; premiers résultats

Abstract

It has been repeatedly found that 25 to 40% of the schizophrenic (SZ) patients had presented, in childhood, premorbid signs such as social withdrawal and minor neuro-psychological deficits. These signs were mainly found in patients with poor prognosis. Clinical indications suggest, however, that early premorbid sensitivity signs might exist in the whole population of SZ patients. The implications of such potential findings for treatment and prevention are evident. The aim of the study was to explore the nature and the importance of such signs and their potential value for prediction. MethodAs the premorbid schizophrenia signs that have been occasionally mentioned are often idiosyncratic and non specific, an exploratory inventory had to be done beforehand by exploring literature, clinical files, biographies, etc. This allowed the preparation of a 344 items questionnaire (about 3/4 of the questions relating to affectivity). The questions were organized according to a multi-choice design (generally five levels of intensity, from null to extreme). This exploratory questionnaire was retrospective and self-administered. PopulationThe questionnaire was answered by 29 families of adult SZ patients and of 27 Control (CT) subjects, paired by sex, age and socio-cultural level. ResultsThis paper presents the results for the 160 questions covering the ages from birth to the end of elementary school (∼ age 12). The most discriminative differences appeared from the visual comparison of the histograms of the responses intensity for SZ and for CT subjects. In about half of the cases, the histograms were similar, or quite similar, in the two populations (Fig. 1). But for 78 questions, the histograms (Fig. 2 et 3) revealed that the extreme levels of intensity were much more frequent in the SZ than in the CT population (table 1). This observation induced to credit each extreme response with a score, weighted according to the magnitude of this difference. Summing up the various scores of each subject made it possible to attribute individual scores. A score of 12 or more was found exclusively in the SZ sample (11 SZ, 0 CT); a score of 3 to<12 was found in 14 SZ and 3 CT; a negative (“normality”) score in 9 CT and 0 SZ. Conclusion and discussionThis exploratory sample of 29 reports on non symptomatic children who later became SZ, suggests that, beside defective premorbid signs in a minority of subjects, affective premorbid signs are present in most cases. Their tentative qualitative analysis suggests that pre-SZ children do not differ substantially from CT Ss with respect to common nervous features such as: Intrensic sleep characteristics, language mechanisms, emotivity and ordinary affectivity, including depression and culpability, etc. But, they are hypersensitive to some interpersonal situations: They might react excessively to the fear of losing attachments, of being unable to develop and defend a territory and to make social relationships. Combined eventually with developmental signs, affective premorbid signs could predict schizophrenia vulnerability. They could, also, provide clues to improved treatment of adult patients and, possibly, to preventive attempts. What make these non-specific signs potentially pre-SZ are: 1) their high intensity; and 2) their poorly organized accumulation in individual subjects (reflecting a disturbance of neural organization?). Alternatively, one third of the CT are correctly discriminated as “normal prone”. This, too, could inspire treatment and prevention. The coherence of the findings supports the validity of this retrospective study. However, the necessary duplication study is under way with a shorter list of questions and a larger number of subjects. From an ethical point of view, a public dissemination of these preliminary results would not be appropriate.