Prazosin and nightmares: what the large trial found

This article explains research. It is not advice about any medicine. Never start, stop or change a medicine on your own; talk to your doctor or pharmacist.

For years, prazosin was widely used for nightmares linked to post-traumatic stress disorder (PTSD). Then, in 2018, the largest trial yet found that it worked no better than a placebo.^[7] The early results and the large trial are both real findings, and how they fit together says a lot about how medical evidence grows.

What prazosin is

Prazosin is an older medicine first used to lower blood pressure. It blocks alpha-1 receptors, one of the ways noradrenaline (the body’s alarm chemical, also called norepinephrine) acts on blood vessels and the brain. Researchers proposed that in PTSD the brain may respond too strongly to noradrenaline, particularly at night, and that damping this response might quieten trauma nightmares.^[2]

The early trials

Murray Raskind and colleagues, working with the US Department of Veterans Affairs, tested the idea in a series of small, placebo-controlled trials.

• 2003: ten Vietnam veterans with severe trauma nightmares each took prazosin and placebo in turn over 20 weeks. Nightmares and sleep were better on prazosin.^{[2, 5]}
• 2007: 40 veterans were randomly assigned to prazosin or placebo for eight weeks. Among the 34 who could be assessed, prazosin reduced trauma nightmares and improved sleep quality, with large effects, and dreams shifted towards those typical of ordinary dreaming.^[2]
• 2013: 67 active-duty soldiers who had returned from Iraq and Afghanistan took prazosin or placebo for 15 weeks. Prazosin again helped nightmares, sleep quality and day-to-day functioning.^[3]

Smaller trials in civilians with PTSD also reported benefits.^[5] On this evidence, the American Academy of Sleep Medicine’s 2010 best practice guide gave prazosin its strongest rating, Level A, for PTSD-associated nightmares.^[4]

The large trial

In 2018 the New England Journal of Medicine published a much larger test. 304 veterans with chronic PTSD and frequent nightmares were recruited from 13 Veterans Affairs medical centres and randomly assigned to prazosin or placebo for 26 weeks, with the dose raised gradually over the first five weeks. Neither the veterans nor the researchers knew who was taking which.^[1]

The main results were measured at ten weeks: how frequent and distressing the nightmares were (rated by a clinician), sleep quality, and overall change. On all three, there was no meaningful difference between prazosin and placebo, and none at 26 weeks either. Blood pressure did fall slightly more on prazosin, so the medicine was active in the body. The authors concluded that, in these veterans, prazosin did not ease distressing dreams or improve sleep quality.^[1]

Why the results might differ

No one knows for certain. Several explanations have been offered.

• Who took part. Because the trial was long and some people would receive a placebo, it enrolled only veterans who were clinically stable: not drinking heavily, not in family crisis, without suicidal or violent thoughts. Raskind suggested this may have left out the people most likely to respond.^[7]
• Other medicines. More than three in four participants in each group were already taking an antidepressant. The AASM task force wrote that a possible interaction with antidepressants needs to be clarified.^[5]
• Small trials can overestimate. It is common in medicine for small, promising trials to look better than the large trials that follow. That is a general pattern, not proof about prazosin.

How guidelines responded

The AASM’s 2018 position paper moved prazosin from “recommended” to “may be used” for PTSD-associated nightmares, and said plainly that the large trial was the reason. The task force also wrote that many patients clearly respond well in practice, and that prazosin remained the first choice when a medicine is used.^[5] The only treatment the paper recommended outright, for both PTSD-associated nightmares and nightmare disorder, was imagery rehearsal therapy.

The 2023 guideline from the US Department of Veterans Affairs and Department of Defense takes a similarly careful line. It suggests prazosin for nightmares associated with PTSD, as a weak recommendation, while suggesting against it as a treatment for PTSD as a whole.^[6]

What this means for you

• Medication decisions belong with your doctor. If you take prazosin, or are wondering about it, bring your questions to the person who prescribes for you, or to a pharmacist. Do not start, stop or change it on your own.
• A negative trial is about averages. It shows that, across that group, the medicine did no better than placebo. It cannot say what will happen for one person, which is why the guidelines leave room for clinical judgement.
• There are well-supported options without medication. Imagery rehearsal is the treatment sleep medicine recommends; you can read how it works. For nightmares linked to trauma, please do this work with a professional.
• If nightmares come with thoughts of harming yourself, please call your local emergency number or reach out to a crisis line today. Help works, and it is available.

Baku never gives advice about medicines. It is a place to practise the drug-free parts, such as rewriting and rehearsing a dream, alongside whatever you and your doctor decide.