Reappraising prazosin for night-time symptoms in post-traumatic stress disorder

Post-traumatic stress disorder (PTSD) has a lifetime prevalence of approximately 10% and is associated with significant levels of psychological distress, impaired functioning and comorbidity (Kessler et al., 2005). Sleep disturbances are particularly common in PTSD and may have marked negative clinical consequences.

Prazosin, a centrally acting sympatholytic alpha1-adrenergic receptor antagonist, may normalize heightened adrenergic neurotransmission in patients with PTSD. Khachatryan et al. (2016) performed a meta-analysis of six randomized controlled trials (RCT) evaluating the efficacy of prazosin in treating PTSD-related night-time symptoms. The sample sizes were relatively small (20, 26, 29, 50, 67 and 100), and the populations sampled were predominantly male veterans, with only two RCTs including civilians. Outcome measures includ-ed Clinician-Administered PTSD Scale (CAPS) and Clinical Global Impression of Change, for most of the trials. The meta-analysis revealed medium-to-large effect sizes (Khachatryan et al., 2016).

However, the most recent RCT (Raskind et al., 2018), which involved veterans, predominantly men (96.1%), differs from this earlier research in two key ways and is not included in the earlier meta-analysis. First, the sample size (n = 304) is larger than the sum of all the previous studies, and second, there are negative findings with no statistically significant difference between prazosin and placebo in the relief of PTSD-related distressing dreams or improvement in sleep quality. However, it should be noted that an exclusion criterion in this 26-week trial was psychosocial instability. The authors reporting on this RCT argue that the selection bias resulting from recruitment of patients who were mainly clinically stable reduced the likelihood of a response to anti-adrenergic medication (Raskind et al., 2018). Furthermore, Raskind et al. suggest that their exclusion of participants who declined to, or were unable to, cease trazodone, an antidepressant with alpha1-adrenergic receptor antagonistic activity, may have also reduced the potential for clinical improvement in night-time symptoms with prazosin.

Given the substantially larger sample size and negative findings in the Raskind et al.’s RCT, it is likely that it will negatively impact on effect sizes of the combined studies in a meta-analysis, which may be imminent (Prospero 2018 CRD42018110986). Therefore, in the interim, is the evidence for prazosin in the relief of sleep symptoms in PTSD currently equivocal? On this basis, there should be caution in considering prazosin for treatment of PTSD sleep symptoms.