This article reviews the literature on the long-term pharmacological treatment of post-traumatic stress disorder (PTSD). A PUBMED search was conducted; only studies on the effects of long-term (>14-weeks) pharmacological treatment for PTSD in adults or children were considered. Our search identified three randomised, double-blind, placebo-controlled studies (one each for sertraline, fluoxetine and risperidone), four open-label studies (one each for sertraline, paroxetine, nefazodone and valproate), one retrospective case series (clozapine) and one pooled analysis (sertraline). All studies involved adult populations, with the exception of the study of clozapine. The studies demonstrate that long-term treatment of PTSD with SSRIs effectively maintains the previous treatment response and improvement in quality of life, converts more patients to responder status and accounts for one-third of overall treatment gains. Greater PTSD severity predicts a longer time to response to these drugs. Discontinuation of SSRI treatment after 12 weeks results in a greater risk of relapse and symptom exacerbation compared with extended treatment. In addition to improved PTSD symptoms, extended treatment with paroxetine improves verbal declarative memory and increases hippocampal volume. Long-term treatment of PTSD with atypical antipsychotics (risperidone and clozapine), non-SSRI antidepressants (nefazodone) and antiepileptic drugs (AEDs; valproate) also appears to result in significant improvements in PTSD symptoms. In conclusion, long-term treatment of PTSD with SSRIs improves the psychiatric and clinical outcome of patients with the disorder and prevents relapse and symptom exacerbation. The effect of other agents (atypical antipsychotics, AEDs and other psychotropic medications) requires further controlled study.