For a certain kind of trauma, the worst part of the day is the night.
Sections
For a certain kind of trauma, the worst part of the day is the night. The nightmares come back on a loop, the same scene or some warped version of it, you wake up soaked and braced for a fight, your sleep is shot, and the wreckage of that bleeds into everything the next day. Prazosin is the drug that exists in that specific gap, not as a treatment for the whole of post-traumatic stress but as a targeted tool for the part of it that happens after you close your eyes. It is old, it is cheap, it was never designed for any of this, and its story is a genuinely interesting one about a medication that found a second life and then ran headfirst into a trial that complicated everything.
An old blood-pressure pill that found a second job
Prazosin was built decades ago to treat high blood pressure. It is an alpha-1 blocker, which means it sits on a particular kind of adrenaline receptor and keeps norepinephrine from doing its usual job there, and in the body that relaxes blood vessels and drops your pressure. The reason it wandered into psychiatry is that the same adrenaline system runs hot at night in a lot of people with PTSD, and that nighttime surge of norepinephrine is one of the things driving the nightmares and the broken sleep. Prazosin crosses into the brain, dampens that surge, and for a chunk of people the result is that the nightmares fade and the sleep stitches back together. Nobody set out to invent a nightmare drug, it just turned out that quieting the adrenaline that was already keeping these people on guard all night did exactly that.
The evidence got complicated, and that is worth saying plainly
The honest version of the prazosin story is that the research does not point in one clean direction, and anybody who tells you it is a slam dunk or a dud is skipping the interesting part. A run of earlier trials, including a well-known study in active-duty soldiers, found real reductions in trauma nightmares and real improvements in sleep, and a lot of clinicians built genuine trust in the drug on the back of that plus what they saw in their own patients (Raskind 2013, PMID 23846759). Then a large trial run through the VA, the one most people mean when they say the prazosin evidence fell apart, came back negative, with prazosin not beating placebo across the group it was tested in (Raskind 2018, PMID 29414272). That result was real and it deserves to be taken seriously, and it also does not erase the earlier signal, because the population in that trial was relatively stable and may not have carried the high nighttime adrenaline load that the drug specifically targets. The fairest read, and the one more recent work is circling, is that prazosin genuinely helps a subset rather than everyone, and figuring out who that subset is matters more than arguing over whether the drug works in the abstract (Mendes 2025, PMID 39828080).
Who it actually helps
The person who tends to do well on prazosin is the one whose PTSD is wrecking the nights specifically, the recurring trauma nightmares, the waking up in a panic, the sleep that never goes deep, and especially the one who runs physically activated at night, heart going, sweating, on guard. It is a targeted fix for that picture, and when it lands it can be the difference between a person getting actual rest and a person grinding through years of broken sleep. It is much less the drug for someone whose PTSD shows up mainly in the daytime, the avoidance and the hypervigilance and the emotional shutdown, because prazosin is not treating the disorder, it is treating one of its meanest symptoms.
Start tiny, titrate up at bedtime
Usually started at 1mg at bedtime and raised slowly over weeks, often into the single digits and sometimes higher for nightmares. The slow ramp is not caution for its own sake, it is how you avoid the blood-pressure drop that an alpha-1 blocker can cause.
Nights can improve fast
Unlike an antidepressant, the nightmare and sleep effect can show up within days to a couple of weeks of reaching a working dose, so you have a reasonably quick read on whether it is doing anything for you.
Stand up slowly at first
The first doses can drop your blood pressure and leave you dizzy or faint if you stand up fast, especially getting out of bed at night, so the early going calls for moving slowly until your body adjusts.
Starting it without face-planting
The one thing that genuinely trips people up on prazosin is the blood pressure, because the same mechanism that calms the nightmares also relaxes your blood vessels, and the first few doses can leave you lightheaded or even drop you if you stand up too quick. That is exactly why it gets started at a tiny dose at bedtime and raised in small steps, so your body has time to adjust to each level before the next one. The practical advice is dull but it matters, take it at night, get up slowly if you have to use the bathroom in the dark, and give the titration the few weeks it needs rather than jumping the dose to chase a faster result. Handled that way it is generally a well-tolerated drug, the dizziness settles, and the worst of the side-effect list is some nasal stuffiness and the occasional headache.
What it does not do
It is worth being clear that prazosin is not a treatment for PTSD itself, and selling it as one does people a disservice. It does not touch the core of the disorder, the flashbacks during the day, the avoidance, the way trauma reorganizes how a person moves through the world, and the treatments that actually work on that core are the trauma-focused therapies, the structured ones like prolonged exposure and cognitive processing therapy that have the real evidence behind them. Prazosin is an adjunct, a way to give someone their nights back so they have the sleep and the footing to do the harder work in the daylight. Used as the targeted sleep-and-nightmare tool it is, it earns its place, and used as a stand-in for the actual trauma work it falls short, because that was never the job it was for.
The honest read
Prazosin is a cheap, old, generally safe drug that genuinely helps a real subset of people with trauma nightmares get their sleep back, and the fact that the largest single trial did not show a group-wide win is a true and important caveat rather than a verdict that it does not work. If your PTSD is mostly stealing your nights, if you wake up in nightmares and dread sleep because of them, it is a reasonable and low-cost thing to try, with the expectations set honestly, that it might do a lot for you or might do little, that you will know fairly quickly either way, and that it is buying you rest rather than treating the trauma underneath. The real treatment for that underneath is the therapy, and prazosin at its best is what makes a person rested enough to face it.
Sources
- Raskind MA, Peterson K, Williams T, et al. A trial of prazosin for combat trauma PTSD with nightmares in active-duty soldiers returned from Iraq and Afghanistan. Am J Psychiatry. 2013;170(9):1003-1010. PMID 23846759.
- Raskind MA, Peskind ER, Chow B, et al. Trial of Prazosin for Post-Traumatic Stress Disorder in Military Veterans. N Engl J Med. 2018;378(6):507-517. PMID 29414272.
- Mendes TP, et al. Factors impacting prazosin efficacy for nightmares and insomnia in PTSD patients. Prog Neuropsychopharmacol Biol Psychiatry. 2025;136:111228. PMID 39828080.
- FDA prescribing information for prazosin (Minipress) via DailyMed, the source for the approved blood-pressure indication, the dosing and first-dose hypotension caution, and the note that the PTSD-nightmare use described here is off-label.
