In the past, the information about the dose-clinical effectiveness of typical antipsychotics was not complete and this led to the risk of extrapyramidal adverse effects. This, together with the intention of improving patients' quality of life and therapeutic compliance, resulted in the development of atypical or second-generation antipsychotics (SGAs). This review will concentrate on the pharmacokinetics and metabolism of clozapine, risperidone, olanzapine, quetiapine, amisulpride, ziprasidone, aripiprazole and sertindole, and will discuss the main aspects of their pharmacodynamics. In psychopharmacology, therapeutic drug monitoring studies have generally concentrated on controlling compliance and avoiding adverse effects by keeping long-term exposure to the minimal effective blood concentration. The rationale for using therapeutic drug monitoring in relation to SGAs is still a matter of debate, but there is growing evidence that it can improve efficacy, especially when patients do not respond to therapeutic doses or when they develop adverse effects. Here, we review the literature concerning the relationships between plasma concentrations of SGAs and clinical responses by dividing the studies on the basis of the length of their observation periods. Studies with clozapine evidenced a positive relationship between plasma concentrations and clinical response, with a threshold of 350-420 ng/mL associated with good clinical response. The usefulness of therapeutic drug monitoring is well established because high plasma concentrations of clozapine can increase the risk of epileptic seizures. Plasma clozapine concentrations seem to be influenced by many factors such as altered cytochrome P450 1A4 activity, age, sex and smoking. The pharmacological effects of risperidone depend on the sum of the plasma concentrations of risperidone and its 9-hydroxyrisperidone metabolite, so monitoring the plasma concentrations of the parent compound alone can lead to erroneous interpretations. Despite a large variability in plasma drug concentrations, the lack of studies using fixed dosages, and discrepancies in the results, it seems that monitoring the plasma concentrations of the active moiety may be useful. However, no therapeutic plasma concentration range for risperidone has yet been clearly established. A plasma threshold concentration for parkinsonian side effects has been found to be 74 ng/mL. Moreover, therapeutic drug monitoring may be particularly useful in the switch between the oral and the long-acting injectable form. The reviewed studies on olanzapine strongly indicate a relationship between clinical outcomes and plasma concentrations. Olanzapine therapeutic drug monitoring can be considered very useful in assessing therapeutic efficacy and controlling adverse events. A therapeutic range of 20-50 ng/mL has been found. There is little evidence in favour of the existence of a relationship between plasma quetiapine concentrations and clinical responses, and an optimal therapeutic range has not been identified. Positron emission tomography studies of receptor blockade indicated a discrepancy between the time course of receptor occupancy and plasma quetiapine concentrations. The value of quetiapine plasma concentration monitoring in clinical practice is still controversial. Preliminary data suggested that a therapeutic plasma amisulpride concentration of 367 ng/mL was associated with clinical improvement. A therapeutic range of 100-400 ng/mL is proposed from non-systematic clinical experience. There is no direct evidence concerning optimal plasma concentration ranges of ziprasidone, aripiprazole or sertindole.