To analyze the available literature regarding the safety of atypical antipsychotics in pregnancy and lactation in order to recommend evidence-based strategies for pharmacologic management of psychosis in these conditions. We summarized the results from articles identified via MEDLINE/PubMed/TOXNET (1993-January 31, 2004), using the key terms pregnancy, lactation, breast-feeding, human milk, psychotropic drugs, atypical antipsychotics, olanzapine, quetiapine, risperidone, clozapine, ziprasidone, and aripiprazole. Retrospective studies, clinical observations, and case reports regarding the 6 atypical antipsychotics mentioned above were selected and analyzed. Extensive manual review of pertinent journals and textbooks was also performed. Reviewed studies show that olanzapine and clozapine apparently do not increase the teratogenic risk if administered to pregnant women, while evidence on quetiapine, risperidone, aripiprazole, and ziprasidone is still limited. In contrast, available information is not able to exclude unwanted serious effects associated with the use of all atypical antipsychotics on mother-infant dyads. Furthermore, more than a few studies suggest increased hyperglycemic risk for pregnant women related to atypical antipsychotic therapy during gestation. Finally, published evidence about the effects on long-term infant neurodevelopment of drug exposure through both placenta and breast milk is represented only by sporadic case reports. It is well known that potential consequences of an untreated psychotic episode may be severe and may lead to the mother attempting suicide and/or infanticide. For these reasons, clinicians need to help mothers weigh both fetal and neonatal risks of exposure to drugs against the potential risk they and their infant may incur if the psychiatric illness is not treated. On the other hand, atypical antipsychotics in pregnancy and breast-feeding do not show evident advantages in safety when compared with typical neuroleptic agents. Therefore, we suggest that the most relevant parameters for selecting the best clinical option for pregnant and breast-feeding women with schizophrenia and related disorders remain strongly related to 3 main points: (1). cautious evaluation of the risk/benefit ratio of fetal and neonatal drug exposure, (2). degree of severity of maternal psychiatric illness, and (3). careful preliminary choice of drugs characterized by a balanced safety/efficacy profile.