Prazosin for PTSD-Related Nightmares: Rationale and Dosing Strategies for Men and Women

David N. Osser, MD

Associate Professor of Psychiatry
Harvard Medical School

Brockton Division of the VA Boston Healthcare System

Last updated: February 9, 2018

In this video, Dr. David Osser explains the rationale behind the recommendation of using prazosin as first-line treatment for PTSD-related nightmares and disturbed arousals.

He also gives practical advice on different dosing strategies for men and women.

This video is an excerpt from the presentation “Psychopharmacology of PTSD Algorithm”. The full presentation is available for Premium Members of our CME program



In short, the recommendation is that after you have ruled out or managed the various causes of sleep disturbance,


and if the patient has PTSD-related nightmares or disturbed arousals,


we recommend you consider a trial of prazosin as the first line medication treatment.


Rationale for prazosin recommendation


A little bit more background about the rationale for this.

The noradrenergic component was first recognized by Murray Raskind and his colleagues at the University of Washington. They reasoned that since prazosin crosses the blood-brain barrier, it might be effective for the arousal symptoms.


Prazosin a non-sedating agent, unlike a lot of other medications used for sleep.


It actually was first developed as a treatment for hypertension. It was marketed under the brand name Minipress many decades ago.

It didn’t prove to be a great treatment for hypertension because you would develop tolerance.


It is also used to treat symptoms of benign prostatic hyperplasia.

Prazosin is the only commercially available alpha-1 antagonist that crosses the blood-brain barrier, but doxazosin does cross it weakly.

Doxazosin is a much longer acting medication and there has been some research, including one placebo-controlled trial, finding efficacy in PTSD.

We have doxazosin at the end of our algorithm, as another option to consider in the same class as prazosin for overall treatment of PTSD.



There are five placebo-controlled randomized trials with prazosin. The fifth one has not yet been published.

In fact, the only way to see it is to go to and look at the data that they are starting to put up about this study. I will tell you about it in a minute.

The four other studies are all published. They are all from Dr. Raskind’s group, affiliated with the University of Washington. They were published between 2003 and 2013.


The four published studies were all very positive. The impact on PTSD symptoms was mostly twice as great with the drug as on placebo, which can be translated to an effect size of 1.0.


If you look just at the effect on nightmares, it was a 200% increase over placebo for the effect size on nightmares. That would translate to a 2.0 effect size.

Now to put into contrast: What do you think the effect size of SSRIs are on PTSD symptoms?


In a 2015 meta-analysis of all the SSRI studies in PTSD, the effect size was 0.23. That was the mean effect size of all the SSRIs. That’s only 23% better than placebo.

The evidence from these first four prazosin trials is the evidence we have relied on for placing prazosin at the top of our algorithm.


However, the fifth study casts a shadow on the others, a big shadow.

No difference from placebo was found. There was not even much of a small trend in favor of prazosin on the measures that had been reported so far.

They have only started to put up some data on the website as they are required to do. No commentary or explanation has so far been offered by the authors of the study.

What do we think of these results? Why are they negative? We don’t know.

So should we change our algorithm preferring prazosin here for sleep, especially in the light that there have been years now of clinical experience added to it?

There are quite a few, open label and other studies, all supporting the effectiveness of this treatment. Maybe we should change the algorithm.

Until this data has been fully brought to public attention, analyzed and given the considerable clinical experience that has been accumulating, we are keeping prazosin where it is for now. But keep posted.

We recommend you keep checking the website. Check with our website to see if we have decided to change our thinking on it. This is a work in progress.


Dosing prazosin in PTSD

Now the next issue is how do you dose prazosin? This is a very important part of the talk, probably the most important, since we feel prazosin is an important treatment but you need to know how to dose it.


A couple of studies have been published showing that 85% to 90% of patients who are put on prazosin are given a subtherapeutic dose and the clinicians give out prematurely.

The reason seems to be just lack of knowledge. I am going to share that with you right now.


There is also another barrier besides knowing the protocol and that is that this protocol takes some time to administer. Time is one of the biggest barriers to the practice of evidence-based medicine in general.

Anything that takes more time is something that clinicians are likely to reject if it means a change from what they have been doing, because time is very hard to come by.

You have a very busy office or hospital care caseload. And if something is going to take more time, it’s easier to think: “I can’t take this time”, “I’m just gonna keep doing things the old way” and ignore the evidence.


First of all, there are two different protocols for men and women. We don’t know why but for men, you seem to need significantly more to get the same effect.

Where are you going? You are going for a mean average dose of 15.6 mg or about 16 mg. That was the average effective dose at the end of Raskind’s fourth and final published study in 2013.


How did he get there?

He went with 1 mg at bedtime for two nights, then 2 mg for five nights.

Then, 4 for seven nights, 6 for seven nights, 10 for seven nights, 15 for seven nights.


In other words, it took five weeks to get to the average dose and then 25 after seven nights on 15. That is the maximum he used in that study.


Subsequently, he and others have reported using higher doses. Certainly going as high as 25 mg is reasonable if patients are having no side effects and tolerating prazosin well.

That is key with each dosage increase: Are patients tolerating prazosin and having no significant side effects?

Yet, their nightmares are not gone and their disturbed awakenings are not gone.


The goal of treatment is to eliminate these disturbed awakenings and give them a smooth night of sleep.

That is one of the reasons prescribers stop at earlier doses. Maybe they have been having a nightmare every night, maybe two or three times a night for years.

Now, after 4 mg or 6 mg of prazosin, it is down to only two or three nightmares a week. Wow, the patient is very happy.

The doctor may be happy too, but not realizing that if they keep going, the nightmares could be eliminated all together. Two or three nightmares a week and the disturbed nights of sleep are still causing significant impairment and quality of life for many people, compared to having excellent sleep.


In the Raskind protocol for men, he also gave a mid morning dose, usually about 10 or 11 AM. Prazosin is a short-acting drug and it is usually gone by morning.

To give some additional benefit for daytime symptoms, Raskind gave 1 mg, starting at week two for a week.

Then he gave 2 mg for the next two weeks, and then 5 mg for weeks five and six for the daytime dose, which gives an additional boost for daytime symptoms.


The protocol for women is roughly half of that at the end, although it starts the same.

It starts with 1 mg at bedtime for three nights, then goes to 2 mg for the next 11 nights making a total of two weeks on the 2 mg level and then 4 mg for a week, 6 mg for a week.

10 mg was the maximum that they used in that study. The median dose that was effective was 7 mg.



Raskind also gave mid morning doses of 1 mg between weeks two and three and 2 mg between weeks four and five.


You should be aware that there are a published set of case reports from 2014 where individuals received up to 45 mg at bedtime (believe it or not) for treatment-resistant patients who were able to tolerate that much until finally achieving excellent results.

There is a very wide range of dosage, there are some patients who really do improve substantially on 2 mg, occasionally even 1 mg, who are very sensitive to it. But if it doesn’t work, by no means should you quit there.

This is why you get 85% to 90% inadequate trials. Because the doctor may prescribe: ” Take this prazosin 1 mg, if it doesn’t work, go to 2 mg and then come back and see me in a month”

Over the course of the next week or two, the patient gets no benefit. As a matter of fact, they may even think their nightmares are worse, so they stop taking it.

Maybe when they come back to see you, they say: “ This stuff didn’t work”.

If the doctor isn’t really up to date and clear on the dosing protocol, he or she may agree and move on to another, probably much less satisfactory, option. At least if we go by these first four prazosin studies.



  1. Raskind, 2003; Raskind, 2007 ; Taylor, 2008 ; Raskind, 2013
  3. Raskind, M. A., et al (2013). A trial of prazosin for combat trauma PTSD with nightmares in active-duty soldiers returned from Iraq and Afghanistan. American Journal of Psychiatry
  4. Koola MM et al. High-dose prazosin for PTSD. Therapeutic Advances in Psychopharmacology 2014; 4(1):43-4
Published in: ,

Earn CME/SA credits

  • Up to 36 CME credits/year
  • Up to 16 SA credits/year
  • Download more than 70 videos
  • Exclusive updates

Free Course: “SSRIs: The Essentials”

Learn the essentials of SSRIs with our free online course.

One module per week:

  • The Mechanism of Action of SSRIs
  • Fluoxetine
  • Fluvoxamine
  • Paroxetine
  • Sertraline
  • Citalopram and Escitalopram

2019-20 Psychopharmacology Online Update

Earn Up to 36 CME Credits / Year

  • Access exclusive updates: new videos every month
  • Download our entire library: more than 70 videos
  • Earn up to 36 CME credits /year