She walked in holding the bottle: propranolol 20 mg, one twice a day, five months of it. Twelve headache days a month, unchanged. “It doesn’t work for me,” she said. Her resting pulse was 78 and her blood pressure was 118/74 — both numbers telling me the drug had barely touched her physiology. She had never been on a dose that had a fair chance. That is the most common reason propranolol “fails” as a migraine preventive: it gets started at 40 mg a day and left there. Propranolol is one of the few migraine preventives with top-tier evidence behind it, but most people who respond do so somewhere between 80 and 160 mg a day — and it takes eight to twelve weeks at that dose before the verdict means anything.
What propranolol is actually doing
Propranolol is a non-selective beta-blocker that has carried an approved indication for migraine prevention for decades, and both the American Academy of Neurology and the American Headache Society place it in the top evidence tier alongside topiramate and divalproex. It has outlasted a lot of fashionable drugs.
Patients are usually told it “relaxes the blood vessels.” That explanation is a leftover from the old vascular theory of migraine and it is almost certainly wrong. The useful effects appear to be central: it dampens noradrenergic traffic out of the brainstem, raises the threshold for the cortical spreading depression that underlies aura, and interacts with serotonin signaling. What matters practically is the consequence — propranolol changes how easily your brain enters an attack. It does nothing at all for an attack already underway. If you take one when the pain starts, you have wasted a tablet.
How well does it work? In controlled trials, roughly four to five people in ten cut their monthly attack count by half or more, against two to three in ten on placebo. That is a real drug with a real effect, and it is also a drug that fails outright for a substantial minority. Nobody should be promised that migraine disappears. The goal is fewer attacks, shorter attacks, and attacks that respond to acute treatment instead of stealing a weekend.
The dose is where most trials go wrong
Starting low is correct — 20 mg twice daily, or 40 mg of a long-acting form, is a sensible opening. The mistake is stopping there. Propranolol should be stepped up every two to four weeks as long as your pulse and blood pressure tolerate it, heading toward 80 to 160 mg a day. Some people need more. The long-acting once-daily preparations are worth asking about, because a drug taken twice a day for months is a drug people quietly stop taking.
The guard rails are simple. I stop climbing if the resting heart rate sits persistently below the mid-50s, if standing up produces real lightheadedness, or if there is any hint of wheeze. Those are the limits — not an arbitrary starting dose that nobody revisited.
There are two clocks running, and people confuse them. The dose clock is how long it takes to reach a therapeutic amount, usually six to eight weeks of careful titration. The response clock only starts once you are at that dose, and it runs another eight to twelve weeks. So a genuine propranolol trial is closer to four months than four weeks. Judge it from a written headache diary rather than memory, because memory is dominated by the worst attack and misses the fact that you had nine bad days instead of fourteen. I have watched patients abandon a drug that was quietly working. If you are not tracking, start a headache diary before you start the tablet.
One thing no preventive can outrun: acute medication overuse. If you are taking triptans or combination painkillers on ten or more days a month, or simple analgesics on fifteen or more, propranolol will look like a failure no matter how well it is dosed. The overuse has to be unwound first, and that is a separate piece of work with its own rules.
Side effects: which to sit out, which to act on
Most of what happens in the first few weeks settles. Mild fatigue, lightheadedness when you stand quickly, cold fingers and toes, unusually vivid dreams — these commonly fade as the dose stabilizes, and pushing through two or three weeks of them is usually rewarded.
Some do not fade, and these are worth acting on rather than tolerating. Exercise intolerance is the one I see underestimated most. Propranolol blunts your peak heart rate, so the run that used to feel effortless now feels like wading. For someone whose training is how they manage their stress — and stress is a migraine trigger — that trade is often a bad one, and I would rather switch drugs than watch a patient give up exercise. Persistent bradycardia, breathlessness or wheeze, and erectile dysfunction all warrant a change rather than patience.
Mood deserves an honest word. Older literature claimed beta-blockers cause depression more confidently than current evidence supports, and I do not warn people into a side effect. But if a patient tells me their mood flattened after starting propranolol, I believe them and I switch. There is no prize for winning an argument about the literature while someone feels grey. Weight gain is possible but usually modest — generally less of an issue than with amitriptyline.
Two safety points that get skipped. If you use insulin, propranolol can blunt the early warning signs of a low blood sugar, which needs a proper conversation with whoever manages your diabetes. And never stop propranolol abruptly after months of use — it is tapered. Sudden withdrawal can trigger rebound tachycardia and a blood pressure spike, and in anyone with coronary disease it can provoke angina.
Who it suits, and who should not be on it
Propranolol is my first choice when migraine arrives with company. Coexisting high blood pressure, essential tremor, a fast resting pulse, or prominent physical anxiety all point toward it, because one drug does two jobs. It also appeals to people who want to avoid the weight gain of amitriptyline and the word-finding trouble and tingling that come with topiramate.
It is the wrong drug for asthma or significant bronchospastic COPD — it is non-selective, and that risk is not theoretical. Also avoid it with bradycardia or second- or third-degree heart block, decompensated heart failure, symptomatic low blood pressure, and Raynaud’s phenomenon. Endurance athletes are usually better served by something else, for the reason above.
Two situations call for individual judgment. When someone has frequent or prolonged aura, I am more cautious with beta-blockers; the concern is debated rather than settled, but with good alternatives available I usually reach elsewhere. And in pregnancy, propranolol is among the preventives obstetricians and neurologists are most comfortable with — which is not the same as risk-free. Third-trimester use has been associated with fetal growth restriction and neonatal bradycardia and low blood sugar, so it belongs at the lowest effective dose and in coordination with your obstetrician, never as a solo decision.
One interaction almost nobody is told about: propranolol substantially raises blood levels of rizatriptan. If you take both, the rizatriptan dose should be limited to 5 mg. It is a small detail that routinely falls through the gap between the neurologist and the pharmacy counter.
When it is time to switch
I switch when a patient has spent eight to twelve weeks at 80 to 160 mg a day — or at the highest dose their pulse and pressure allow — with no meaningful drop in headache days. I also switch when we simply cannot get above 40 or 60 mg because the heart rate, the blood pressure, or the breathing will not permit it. That is not a failed drug so much as an unfinished trial, and the right response is to move on rather than push into trouble. And I switch for any side effect that has not faded and costs more than the migraine does.
A partial response is different, and it gets mislabeled as failure constantly. Going from twelve headache days to eight is not nothing — it is a drug that is working and has not been finished. That situation calls for optimizing: raise the dose if there is headroom, add magnesium or riboflavin, protect sleep timing, and treat attacks earlier and harder. The timing of when to switch versus when to wait is worth understanding before you make the call.
There is a larger question hiding behind all of this. If two or three well-run preventives have failed, the problem may not be the drugs at all. Cluster headache, hemicrania continua, unrecognized medication overuse, and raised intracranial pressure are all treated as migraine at some point in their history, and none of them respond to migraine prevention. That pattern — reasonable treatment, correctly dosed, no result — is one of the strongest reasons to revisit the diagnosis itself. I have written about why headaches stop improving and what doctors miss, and that is the piece to read if propranolol is the third thing that has not worked.
Before your next appointment: download the free Headache Red Flags Checklist — the warning signs that mean a headache needs urgent assessment rather than another prescription.
Still stuck after two or more preventives? You can talk to a specialist about your history, your dosing, and whether the diagnosis still fits.
Written by Tae-Ho Guak, MD, board-certified neurologist. This article is for education and does not replace an in-person consultation, diagnosis, or prescription. Do not start, change, or stop any medication — including tapering propranolol — without speaking to your own physician. If you have a headache of sudden thunderclap onset, headache with fever and a stiff neck, weakness, confusion, seizure, or new visual loss, go to an emergency department immediately.
From the textbook. Chapter 15, “Preventive Treatment I: The Oral Preventives,” of my clinical textbook Headache Medicine covers this topic in full for doctors, with the evidence and decision pathways behind it. About the book →