Although antipsychotic medications are extremely valuable in both the acute and long-term management of schizophrenia, a substantial subgroup of patients derive little or only partial benefit. A variety of strategies to treat such patients are reviewed.

As with other drugs it is necessary to look for changes induced by anxiolytics on vital signs, laboratory parameters and adverse events. In return, in a more specific way for anxiolytics, we will look at side effects at the central nervous system level with psychological and physiological battery tests. We will also assess the safety of use of anxiolytics in certain specific conditions, such as overdose or withdrawal and in certain populations such as the elderly, neonates and children. The assessment of safety and side effects, whatever the drug type studied, must come early in the developing process of a drug (phases I, II and III).

In 55 chronic DSM III-R schizophrenics the occurrence of obstetric complications (OCs) was investigated using the familial/sporadic strategy and Leonhard's unsystematic/systematic distinction. The overall frequency and severity of OCs did not differ between patients and controls. A sub-sample of patients, whose genetic risk was supposed to be high in both classification systems (diagnosis of unsystematic and familial schizophrenia), had significantly fewer OCs than controls on the Lewis and Murray scale (P < 0.05). With reference to previous reports of increased mortality rates in the offspring of schizophrenics, high genetic risk and additional perinatal stressors may increase perinatal mortality. In contrast, patients whose genetic risk was supposed to be low in both systems (diagnosis of systematic and sporadic schizophrenia) showed a trend to an increased frequency of OCs in the Fuchs scale. In the context of the recently reported highly significantly increased rate of maternal infections during midgestation in these patients, it was supposed that perinatal complications may be of some aetiological importance in schizophrenics with low genetic risk.

The hypothesis that paranoid and non-paranoid schizophrenics are differentially associated with unilateral hemisphere dysfunction was tested on 12 paranoid and 12 non-paranoid schizophrenics, as well as on 24 affective patients and 24 normal controls. The subjects were presented for 150 ms with series of digit-pairs and open rings to the left or right visual fields. Overall recognition of digits and localization of gaps in the rings were better for the right than for the left visual field. However, performance of the paranoid and non-paranoid schizophrenics was relatively poorer in response to the right and left visual field stimuli, respectively. Since these data do not correspond to the findings obtained in the auditory modality, they were interpreted as indicating modality-specific associations of paranoid schizophrenia with left hemisphere dysfunction, and of non-paranoid schizophrenia with right hemisphere dysfunction.

The P3 components (P3a and P3b) of the event-related potentials have been analysed in sixteen in-patients with a major depressive episode (DSM III-R) according to their subtype - anxious-agitated and impulsive versus retarded and blunted-affect - assuming that these two opposite dimensions of the depressive mood would rely on different physiopathological mechanisms. The P3a component, which indexes automatic cognitive processes, was dramatically decreased in the retarded and blunted-affect group. This suggests that this special type of cognitive dysfunction in depression is rather related to negative symptoms than to depression itself.

In a double-blind study carried out on elderly patients (older than 65 years) the efficacy and tolerability of the new antidepressant moclobemide was compared. Moclobemide belongs to a new class of substances called RIMA (Reversible inhibitor of the monoamine oxidase type A). Fluvoxamine, a selective reuptake-inhibitor of 5-HT, belongs to a class of antidepressants known for their better tolerability compared to tricyclic especially with elderly patients. Forty elderly patients (mean age 75 years) with major depression (according to DSM III) were randomized to receive either moclobemide (300 mg) or fluvoxamine (100 mg) twice daily. Dosages were increased when necessary on day 8, to a maximum of 450 mg moclobemide or 200 mg fluvoxamine and in most cases were maintained at these levels for the remainder of the study period (four weeks). Moclobemide was more effective than fluvoxamine showing a marked antidepressant effect and an earlier effect on psychomotor retardation. The two drugs were well tolerated showing a low incidence of side effects.

We report on the case of a patient who developed an acute meningitis and, after a period of about two weeks, without any neuropsychiatric problems, an acute paranoid-hallucinatory and catatonic syndrome. The symptomatology is discussed, in relation with the diagnostic difficulties of differentiating between a biphasic meningo-encephalitis with an organic psychosis or a first manifestation of an endogenous psychosis.