There have been significant advances in the clinical psychopharmacology of PTSD. Most, but not all, clinical practice guidelines for PTSD recommend medication as a first-line treatment. Growing understanding of the unique pathophysiology of this disorder has laid the groundwork for rational pharmacotherapy so that newer clinical trials are testing medications that might correct neurobiological alterations shown to be associated with the disorder. The most notable benchmark is approval by U.S. Food and Drug Administration (FDA) for two medications, sertraline and paroxetine, both selective serotonin reuptake inhibitors (SSRIs), as indicated treatment for PTSD. Unfortunately, there have been no new FDA-approved medications for over 10 years.
Although the literature indicates that a number of key neurotransmitter, neuroendocrine, and neuropeptide systems are altered in PTSD (1), the literature on clinical trials of pharmacological agents has focused mostly on antidepressants, especially SSRIs and serotonin/norepinephrine reuptake inhibitors (SNRIs). Medications affecting adrenergic, hypothalamic-pituitary-adrenocortical (HPA), glutamatergic, GABAergic, dopaminergic and other mechanisms of action have received much less attention. Thus, at this point in time, treatment research has focused primarily on clinical trials with medications that were initially developed for treating depression, cardiovascular illnesses (e.g., adrenergic agents), seizure disorders/mood fluctuations, anxiety disorders, and schizophrenia.
Evidence From Clinical Trials
Selective serotonin reuptake inhibitors (SSRIs)
SSRIs are the current pharmacological treatment of choice for PTSD patients as four independent clinical practice guidelines attest (See below). As stated previously, two SSRIs, sertraline and paroxetine, have received FDA approval for PTSD treatment. Multisite randomized clinical trials with both sertraline and paroxetine demonstrated that both agents significantly reduced PTSD symptoms in contrast to placebo and consistently produced remission in 30% of study participants (2–5). Furthermore, when sertraline treatment was extended from 12 to 36 weeks, remission rates increased from 30% to 55% (6). Finally, discontinuation of SSRI treatment is associated with clinical relapse and a return of PTSD symptoms (see Friedman and Davidson  for references).
Based on a total of 12 RCTs on SSRIs, the Cochrane, nine meta-analytic and other recent reviews (8–11) have found paroxetine, sertraline, and fluoxetine superior to placebo. In addition, four of six clinical practice guidelines for PTSD recommend SSRIs as first-line monotherapy for PTSD (12,13–15, 16). One may argue with the criteria utilized by the two guidelines that did not recommend SSRIs: the Institute of Medicine (17) excluded a number of studies included in other guidelines because of more stringent criteria regarding methodology and data analytic strategies, while the UK’s National Institute for Health and Clinical Excellence (18) included unpublished studies and did not consider results with an effect size under 0.5 as a positive trial.
SSRIs have a broad spectrum of action and are effective for PTSD re-experiencing, avoidance/numbing, and hyperarousal symptoms. They also appear to promote rapid improvement in quality of life that is sustained during treatment (19). Given interest in neurogenesis as a potential mechanism for mediating recovery from PTSD, it is noteworthy that following a 9–12 month open-label clinical trial, participants who received paroxetine exhibited a 4.6% increase in mean hippocampal volume as measured by magnetic resonance imaging (MRI) that was associated with improved declarative memory as well as reduction in PTSD symptoms (20).
A common challenge in clinical practice is what to do for the partial responder. With respect to SSRI monotherapy, there have been a few studies testing whether adjunctive pharmacotherapy might benefit partial responders to prolonged exposure and vice versa. Two studies indicate that partial responders to sertraline showed significant improvement when SSRI treatment was augmented with prolonged exposure (21, 22). On the other hand, adjunctive SSRI treatment for prolonged exposure partial responders was ineffective (23). There’s really no theoretical explanation for these findings; perhaps prolonged exposure augmentation of SSRI treatment is more successful than SSRI augmentation of prolonged exposure treatment simply because prolonged exposure is more effective than medication either as a mono- or adjunctive therapeutic agent.
Two large (12-week and 6-months, respectively) multicenter trials of the SNRI venlafaxine-XR have shown superiority relative to placebo (24, 25). In addition, venlafaxine treatment was associated with improved resilience or the ability to deal with daily stress. The long-term trial also showed (similar to SSRIs) that whereas approximately 30% of patients achieve remission within 12 weeks, a substantial percentage of patients will not exhibit remission in less than 6 months.
Mirtazepine, an antidepressant with both serotonergic and adrenergic activity has proven efficacy for PTSD monotherapy and is recommended as a second-line pharmacotherapeutic agent (8, 15, 16).
Nefazadone and trazodone are two antidepressants that enhance serotonergic activity through both an SSRI action as well as postsynaptic 5-HT2 blockade. Nefazadone has been shown to be as effective as sertraline in a number of RCTs. It is only recommended as a second-line treatment for PTSD, (15, 16) however, because it may cause serious liver toxicity. Indeed, brand name nefazadone (e.g., Serzone) has been withdrawn from the American market because of liver toxicity, although the generic drug is still available. Trazodone has limited efficacy as monotherapy for PTSD. Due to its (antihistaminergic) sedating effects and serotonergic action, it is often used in conjunction with SSRIs to counter medication-induced insomnia (8).
Bupropion potentiates noradrenergic and dopaminergic (but not serotonergic) synaptic activity. There are currently no RCTs indicating that bupropion is an evidence-based treatment for PTSD. Since many clinicians assume that any antidepressant is useful for PTSD, bupropion is sometimes prescribed due to the incorrect assumption that it is evidence-based pharmacotherapy for PTSD.
Tricyclic antidepressants (TCAs) and monoamine oxidase inhibitors (MAOIs)
Tricyclic medications block presynaptic reuptake of both serotonin and norepinephrine, although their relative potency at different reuptake sites varies from one medication to the next. Some TCAs exert their actions primarily on serotonin reuptake (e.g., amitriptyline), others primarily on norepinephrine reuptake (e.g., desipramine) and some on both neurotransmitter systems (e.g., imipramine). TCAs have been tested sparingly, especially in recent years, as interest has shifted to newer antidepressants, especially because new antidepressants have more benign side effect profiles than older agents. This is unfortunate because, as discussed below, TCAs and MAOIs are effective. Another reason why TCAs and MAOIs have not been tested recently is because pharmaceutical companies are not motivated to fund such studies. With regard to clinical trials, RCTs with imipramine (26), amitryptyline (27), and desipramine (28) have demonstrated symptom reduction in PTSD patients (although an RCT with desipramine has reported negative results (29). Unfortunately, TCA blood levels were not measured in any of these studies.
MAOIs block the intraneuronal metabolic break-down of serotonin, norepinephrine, dopamine, and other monoamines, thereby making more available for presynaptic release. It’s been 20 years since MAOIs have been tested. In the only RCT with an MAOI, phenelzine was extremely successful in reducing re-experiencing and arousal symptoms among Vietnam combat veterans with PTSD (26). Otherwise, results have been mixed in open-label trials and a small, methodologically flawed 5-week cross-over study had negative results. Finally, all three PTSD symptom clusters improved during an open trial with the reversible MAO-A inhibitor, moclobemide (for references see Friedman and Davidson ). Although clinicians (and investigators) tend to avoid MAOIs (and to a lesser extent, TCAs) because of potentially serious side effects, they are very effective medications and should be considered as third-line treatments for PTSD if other medications have failed. The “third-line” designation is due to the paucity of clinical trials, potential toxicity, dietary restrictions, and unfamiliarity and discomfort managing MAOIs among practitioners.
Adrenergic dysregulation has been consistently observed among individuals with PTSD (1). Despite this, there have been few clinical trials of adrenergic agents over the years. The medication that has generated the greatest interest in this regard is prazosin, an alpha−1 adrenergic antagonist that has been shown to effectively reduce traumatic nightmares among individuals with PTSD (30, 31). Results have been mixed, however, regarding prazosin’s efficacy for full PTSD. As a result, the VA/DoD clinical practice guideline recommends prazosin as an evidence-based treatment for nightmares but not for other PTSD symptoms (16). It is possible that prazosin’s lack of efficacy for full PTSD is due to its short half-life so that customary bed-time doses are out of the system by morning. A recent RCT with American soldiers who had been deployed to Iraq and/or Afghanistan (32) supports this speculation, since prazosin (when prescribed in divided doses) effectively reduced overall PTSD symptom severity as well as traumatic nightmares.
Research with the beta adrenergic antagonist propranolol has been sporadic and yielded mixed results. First, as monotherapy for PTSD, non-RCT trials indicate that it has some beneficial effects (such as reduction of intrusive recollections and reactivity to traumatic stimuli) (33, 34). Second, as a prophylactic agent that can prevent the later development of PTSD in acutely traumatized individuals, after some initially promising findings regarding propranol’s efficacy (35, 36), more recent results have been disappointing (37–39). Finally, a third approach has been research indicating that beta adrenergic blockade in the lateral nucleus of the amygdala disrupts memory reconsolidation, thereby reducing fear memories (40). In such studies patients receive a single dose of propranolol (at weekly intervals) after they actively recall their traumatic experience. Preliminary results are consistent with the hypothesis that PTSD symptoms are significantly reduced due to disruption of reconsolidation of traumatic memories (41). Clearly, more research is needed on this intriguing approach.
Research with alpha−2 adrenergic agonists such as clonidine and guanfacine has been disappointing. Two RCTs with guanfacine in veterans with chronic PTSD, however, have both been negative (42, 43) whereas the sparse, open-label, clinical literature on the efficacy of clonidine in PTSD is generally favorable (34, 44).
Interest in anticonvulsant agents is long-standing because of their antikindling actions (45, 46). More recent research on glutamatergic and GABAergic alterations associated with PTSD has also drawn attention to this class of medications. Finally, interest by the pharmaceutical industry, in recent years, in the development of new anticonvulsant/mood stabilizers has led to clinical trials with PTSD patients.
Except for topiramate, results from RCTs with anticonvulsants have been disappointing. Negative or equivocal findings have resulted from RCTs with valproate (47, 48), lamotrigine (49), and tiagabine (50).
On the other hand, positive results have been found with topiramate monotherapy in three recent RCTs which had a total of 142 participants (51–53). Topiramate had a broad spectrum effect that reduced all PTSD symptoms. According to a recent meta-analyis by the Agency for Healthcare Research and Quality (AHRQ), its efficacy was at least as good as paroxetine and venlafaxine and compared favorably with sertraline, fluoxetine, prazosin, valproate, tiagabine, risperidone and olanzapine (54). This is a very hopeful development, since we now have identified a nonantidepressant medication that appears to be at least as effective as current first-line (SSRI and SNRI) pharmacotherapeutic agents for PTSD.
In summary, topiramate is the only anticonvulsant agent that has proven effective for PTSD. Theoretically, one would predict that a medication like topiramate, which both enhances GABAergic and antagonizes glutamatergic activity, would be effective for PTSD. It is, frankly, unclear why none of the other anticonvulsants have been shown to ameliorate this disorder.
d-Cycloserine (DCS) is a partial NMDA receptor agonist that potentiates learning new behaviors (such as fear extinction) through its NMDA action. It has been shown that d-cycloserine augmentation of exposure therapy has accelerated remission in patients with both acrophobia and social anxiety disorder. As a result, this approach has been suggested for PTSD (11, 55). Although there are a number of trials in progress testing d-cycloserine as an adjunct to exposure therapy, two recently published RCTs have been disappointing. In one, outcomes were no better among mixed trauma survivors receiving exposure therapy + d-cycloserine than among those who received exposure therapy alone (56). In the second, veterans who received exposure therapy + d-cycloserine did significantly worse than those receiving exposure therapy alone (57). At this time it is not apparent why d-cycloserine has proven more effective for acrophobia and social phobia than for PTSD. Hopefully, more positive results will be observed in additional trials. Based on current evidence, however, d-cycloserine cannot be recommended as an adjunct to exposure therapy for PTSD.
Benzodiazepines, which act at GABA-A receptors, are not recommended for PTSD treatment in most clinical practice guidelines (unless their use can be justified for comorbid disorders) (15, 16). In an RCT, alprazolam monotherapy did not reduce core re-experiencing or avoidant/numbing symptoms, although it did lead to improvement in insomnia and generalized anxiety (58). Furthermore, prophylactic treatment of recently traumatized emergency room patients with clonazepam (59) or the hypnotic benzodiazepine temazepam (60) did not prevent the later development of PTSD. In addition to their lack of efficacy for PTSD (61), benzodiazepine treatment is associated with significant risks. These include their abuse potential, concerns about sedation (and loss of balance) in the elderly, and the possibility that they might interfere with the psychological processes needed to benefit from CBT. Despite all this evidence that risks appear to far outweigh benefits, these medications are widely used to treat PTSD (62).
Atypical Antipsychotic Medications
The major interest in atypical antipsychotics in PTSD has been as adjunctive treatment for nonpsychotic partial responders to antidepressants. A number of small, single-site studies (involving risperidone and olanzapine), all done before 2011, indicated that adjunctive atypical antipsychotic treatment significantly improved outcomes for patients who failed to achieve improvement following antidepressant treatment (See Friedman et al. (8) for references). Because these results were limited by small sample sizes, a large multisite RCT involving 247 veterans was undertaken to test the efficacy of adjunctive risperidone for antidepressant non/partial responders (63). The findings were completely negative. Risperidone was no better than placebo augmentation. As a result, the 2010 VA/DoD PTSD Clinical Practice Guideline, which had previously recommended adjunctive atypical antipsychotic treatment, changed its recommendations for treatment of PTSD without psychotic symptoms to: 1) a specific recommendation against risperidone augmentation; and 2) a conclusion that evidence was inconclusive regarding adjunctive use of any other atypical antipsychotic. In addition to lack of benefit, atypical antipsychotics have serious metabolic side effects (e.g., hyperglycemia, diabetes, heart disease).
With respect to atypical antipsychotic monotherapy for PTSD, results have been mixed in a few small studies (64, 65) that have been judged to lack sufficient rigor to be included in the recent AHRQ meta-analysis (54). At this time, atypical antipsychotics are not recommended for monotherapy of PTSD (15, 16).
Finally, it has been proposed that atypical antipsychotics should be used when PTSD is associated with co-occurring psychotic symptoms (such as auditory and visual hallucinations or paranoid delusions, often related to hypervigilance). Indeed, one published RCT showed that atypical antipsychotics (e.g., risperidone) did ameliorate psychotic symptoms although they were not effective in reducing PTSD symptoms (66). This use of atypical antipsychotics would be consistent with clinical practice guidelines that recommend atypical antipsychotics for co-occurring psychotic symptoms but not for PTSD per se (15, 16).
Finally, given that there are no RCTs with first-generation antipsychotics indicating efficacy in PTSD, and given abundant evidence regarding potential neuromuscular side effects, they are not recommended for PTSD treatment.
The major challenge for the future is to develop and test pharmacological agents that are designed to address specific pathophysiological abnormalities associated with PTSD. It might be expected that such a theory-driven approach would result in more effective treatments than has been the case with agents originally utilized for other disorders such as antidepressants, antiadrenergics, anticonvulsants and (atypical) antipsychotics.
Medications designed primarily for PTSD might include CRF antagonists, NPY enhancers, or more specific adrenergic, serotonergic, glutamatergic, GABAergic agents or medications affecting HPA mechanisms (67). Agents that promote neurogenesis should be a focus for future research. Emerging knowledge on the psychobiology of dissociation points to possible roles for medications acting on NMDA, AMPA, alpha−2 adrenergic and 5HT−2 receptors. Other classes of medications of great interest might include endocannabinoids, oxytocin, neurokinin/SubstanceP, dopamine-beta-hydroxylase, pregnenalone, hydrocortisone, and other pharmacological agents (61). Such clinical trials are investigating preventive agents, monotherapeutic interventions, adjunctive treatments and medications that enhance psychotherapy. In short, we are on the threshold of exciting new developments in research on pharmacotherapy for PTSD with the hope that effective novel treatments will soon be available for clinical use.