Pharmacological prevention of post-traumatic stress disorder and acute stress disorder: a systematic review and meta-analysis

Summary

Background

An increasing number of studies have investigated the pharmacological prevention of post-traumatic stress disorder (PTSD) and acute stress disorder (ASD). This is the first systematic review to examine the effects of pharmacotherapies (eg, β blockers, hydrocortisone, and selective serotonin re-uptake inhibitors) given within the first month after a traumatic or aversive event to prevent PTSD or ASD compared with no pharmacotherapy or placebo control.

Methods

A systematic literature search in PubMed, PsycINFO, Embase, and the Cochrane database of randomised trials was done. Studies included randomised controlled trials, controlled clinical trials, and cohort studies; their overall quality was low to moderate. We computed the pooled incidence risk ratio (IRR): the risk of incidence of PTSD or ASD in the pharmacotherapy groups relative to the incidence of PTSD or ASD in the control groups. Additionally, we computed Hedges' g effect sizes for PTSD or ASD continuous outcomes.

Findings

15 studies met inclusion criteria (1765 individuals). Pharmacotherapy was more effective in preventing PTSD or ASD than placebo or no intervention (14 studies, 1705 individuals, IRR 0·65, 95% CI 0·55–0·78; number needed to treat 11·36), although no effect was found when only randomised controlled trials were included (ten studies, 300 individuals, IRR 0·69, 95% CI 0·40–1·21). Hydrocortisone showed a large effect in reducing the risk of PTSD (five studies, 164 individuals, IRR 0·38, 95% CI 0·16–0·92).

Interpretation

No firm evidence was found for the efficacy of all early pharmacotherapies in the prevention of PTSD or ASD, but hydrocortisone reduced the risk of developing PTSD. The small number of studies and their limited methodological quality cast uncertainty about the effects.

Funding

None.

Introduction

Post-traumatic stress disorder (PTSD) can arise after direct or indirect exposure to trauma involving actual or threatened death, serious injury, or sexual violence in about 9% of survivors, although the conditional risk to develop PTSD varies between the types of trauma studied.¹ DSM-IV PTSD is diagnosed when symptoms are present for at least 1 month after the traumatic event, and is characterised by symptoms of re-experiencing, avoidance, and hyperarousal.² Acute stress disorder (ASD) might already be diagnosed within 2 days to 4 weeks after the trauma, and is characterised by similar symptoms as those included in the PTSD diagnosis, but also includes symptoms of acute dissociation.²

Attempts have been made to prevent the onset of PTSD, but until now most brief early psychological interventions have been unsuccessful³ (but see Rothbaum and colleagues³) and possibly harmful (eg, psychological debriefing⁴). However, brief early treatment of PTSD or ASD once symptoms have developed has been shown to reduce the risk for chronic PTSD.⁵

Based on recent advances in cognitive neuroscience, pharmacological interventions have been proposed as a promising new option in the prevention of PTSD.6, 7 Pharmacological interventions might be most effective if offered immediately after the traumatic event, since cognitive neuroscience studies suggest a 6 h window of time after a traumatic event when fear memories might be disrupted before they are consolidated.⁸

One pharmacological agent suggested to be effective in the prevention of PTSD, is the β adrenergic blocker propranolol. Propranolol has been suggested to interfere with the consolidation of newly acquired emotional memories⁹ by blocking post-synaptic β adrenergic receptors in the basolateral nucleus of the amygdala.⁶ Studies in healthy participants have shown that propranolol given before the presentation of emotionally disturbing materials prevented the heightened recall of these materials associated with placebo ingestion.¹⁰ Moreover, the administration of propranolol in patients attending emergency departments after a traumatic event resulted in decreased psychophysiological responses to personalised trauma scripts⁹ and PTSD incidence¹¹ relative to placebo.

In addition to β blockers, administration of glucocorticoids such as hydrocortisone shortly after trauma has been proposed to prevent PTSD.¹² There is evidence that both stress-induced raised cortisol concentrations and administration of glucocorticoids might result in the impairment of retrieval of declarative memory, including retrieval of emotionally arousing materials.¹³ Recent studies on the administration of hydrocortisone shortly after stressful events support the notion that hydrocortisone might prevent PTSD.¹²

Other pharmacological agents associated with decreased PTSD incidence when given shortly after a traumatic event include opiate analgesics such as morphine,¹⁴ selective serotonin re-uptake inhibitors,¹⁵ and innovative therapies such as α-omega fatty acids.¹⁶

We believe this is the first meta-analysis to investigate the effects of pharmacological prevention of PTSD and ASD in recent trauma survivors. We considered both PTSD and ASD as relevant outcomes. ASD is assumed to be a precursor of PTSD in a proportion of trauma survivors, and most individuals with ASD have been found to develop PTSD.¹⁷ Because this is an emerging research field, we anticipated that the number of randomised controlled trials (RCTs) involving adults only would be restricted. Therefore, we decided to include RCTs, controlled clinical trials (CCTs; trials that included a control group but lacked randomisation), and longitudinal cohort studies, all including either adults or children. Subgroup analyses were done to compare the results for PTSD with those of ASD, RCTs with those of the less rigorous studies, and the results of studies with a placebo control group with studies with a control group that received no intervention or standard pharmacotherapies. Other subgroup comparisons concerned the types of pharmacotherapy administered, not least since their proposed underlying mechanisms are distinct. We also compared studies involving unselected (ie, all trauma survivors) or selected samples (trauma survivors selected on characteristics that are associated with an increased PTSD risk such as an increased heart rate at the emergency room).⁹ We considered the timing of administration of the pharmacotherapies, which could be within the 6 h consolidation window or at a later stage. It has been assumed that administration within the consolidation window would contribute to the efficacy of β blockers and hydrocortisone in preventing PTSD.9, 18 Finally, we compared studies that reported clinician-rated diagnoses with the studies that used self-report instruments.