Amitriptyline, propranolol, and candesartan for chronic headache: evidence and trade-offs when autonomic dysfunction complicates the picture
For someone with both chronic daily headache and autonomic dysfunction, the choice of headache preventive matters more than usual. A review of the evidence and autonomic trade-offs for each option.
Choosing a preventive medication for chronic headache is usually a matter of balancing efficacy against side effects. When the patient also has post-viral autonomic dysfunction — specifically orthostatic tachycardia, fatigue, and exercise intolerance — the calculation changes. Some standard headache preventives are actively beneficial for autonomic symptoms. Others can worsen them significantly. Understanding this intersection is important for making informed choices.
This post covers the three medications I’ve had the most direct experience with and the most reason to investigate: amitriptyline, propranolol, and candesartan. All three are commonly used for headache prevention; all three have meaningful effects on the autonomic nervous system.
Amitriptyline
Headache evidence
Amitriptyline is a tricyclic antidepressant that has been used as a headache preventive for over 50 years. The evidence is among the strongest of any headache preventive: multiple RCTs and meta-analyses confirm it reduces headache frequency and severity in both episodic migraine and chronic daily headache. A 2015 Cochrane review found it more effective than placebo for migraine prevention, with a number needed to treat (NNT) of around 3–4 for a 50% reduction in headache days.
The mechanism isn’t fully understood, but involves serotonin and noradrenaline reuptake inhibition, sodium channel blockade, and inhibition of central sensitisation. It doesn’t need to be taken at antidepressant doses — the headache preventive effect is seen at 10–75mg/day, much lower than psychiatric doses.
Autonomic effects
Amitriptyline has significant effects on the autonomic nervous system, most of which are problematic for someone with POTS or post-viral dysautonomia:
Orthostatic hypotension — amitriptyline has alpha-1 adrenergic blocking effects, meaning it reduces the vasoconstriction response to standing. In a healthy person this is mild and well-tolerated. In someone with already-impaired orthostatic regulation, it can significantly worsen postural lightheadedness and orthostatic tachycardia. This is the most clinically relevant concern.
Anticholinergic effects — dry mouth, constipation, blurred vision, urinary retention, cognitive effects (particularly at higher doses). These are autonomic effects that add to symptom burden.
Heart rate effects — amitriptyline has anticholinergic (heart rate-raising) effects, which counteract the heart rate benefit that some patients with post-viral tachycardia most need to reduce.
Sedation — prominent at initiation, usually attenuating within 2–4 weeks. Can worsen post-viral fatigue significantly in the initial period.
Weight gain — clinically relevant for some patients and associated with worsened metabolic and cardiovascular parameters long-term.
My experience
I trialled amitriptyline at 10mg for four weeks, then 20mg for another four weeks. Headache impact: possible modest reduction in severity, hard to isolate. Autonomic impact: clear worsening of morning lightheadedness and orthostatic symptoms, which was the primary reason I stopped. The trade-off was not acceptable given that the orthostatic component is the more functionally limiting problem.
Amitriptyline can be the right choice for someone whose headache is dominant and whose autonomic symptoms are mild. For someone with significant orthostatic intolerance, the alpha-blocking effect makes it a difficult choice unless there’s a compelling reason to prioritise headache management over autonomic symptoms.
Starting dose and titration
10mg at night. Titrate by 10mg every 2–4 weeks to a maximum of 75mg for headache prevention. Most people find the sweet spot at 25–50mg. Starting low and going slowly reduces the likelihood of intolerable sedation.
Propranolol
Headache evidence
Propranolol is a non-selective beta-blocker with good evidence for migraine prevention. Multiple RCTs support it, and it appears in all major headache prevention guidelines as a first-line option. A 2019 meta-analysis found a 50% reduction in migraine frequency in roughly 50% of patients — comparable to amitriptyline. It’s better studied for episodic migraine than for chronic daily headache specifically, but is used across the spectrum.
The mechanism is thought to involve beta-receptor blockade in the central nervous system reducing cortical excitability, inhibition of vascular serotonin uptake, and membrane-stabilising effects.
Autonomic effects
Propranolol’s autonomic effects are complex and not straightforwardly positive or negative for POTS:
Heart rate reduction — propranolol reduces resting and exertional heart rate by blocking sympathetic cardiac stimulation. In POTS, where elevated heart rate on standing is the defining feature, this could be helpful. And in practice, propranolol at low doses (10–40mg twice daily) is sometimes used specifically to manage heart rate in POTS.
Exercise intolerance — this is the significant problem. Beta-blockers blunt the heart rate response to exercise, which means the heart can’t increase cardiac output appropriately during exertion. In someone already limited by impaired exercise capacity, this can meaningfully reduce the maximum workload achievable and worsen fatigue during and after activity. Studies in post-viral conditions suggest that beta-blockers can reduce exercise capacity significantly in patients with low anaerobic threshold.
Blood pressure effects — propranolol reduces blood pressure. In someone with normal or low blood pressure (common in POTS patients), this can worsen orthostatic hypotension and general fatigue.
Fatigue — beta-blocker fatigue is a well-recognised class effect. It tends to be dose-dependent and more prominent with non-selective agents like propranolol than with beta-1-selective agents.
Masking hypoglycaemia — relevant primarily for diabetic patients on insulin, but worth noting.
The POTS-specific evidence
Propranolol has been directly studied in POTS. A crossover trial by Raj et al. (2009) compared low-dose propranolol (20mg), high-dose propranolol (80mg), and placebo. Low-dose propranolol reduced orthostatic heart rate more effectively than placebo, with acceptable tolerability. High-dose propranolol significantly worsened exercise capacity. This suggests that low doses may offer heart rate benefit with manageable exercise trade-offs, while higher doses tip into net harm for exercise-limited patients.
My experience
I tried propranolol at 10mg twice daily. Heart rate on standing was modestly reduced. However, the fatigue effect was prominent — significantly increased fatigue during the working day that exceeded what I’d had without it. I stopped after six weeks. I’m aware that for some people the heart rate benefit outweighs the fatigue cost; the trade-off appears individual.
Candesartan
Headache evidence
Candesartan is an angiotensin II receptor blocker (ARB) primarily used for hypertension and heart failure. Its use in migraine prevention is off-label but reasonably well supported: a Norwegian RCT (Tronvik et al., 2003) found it significantly reduced migraine days compared to placebo, and subsequent observational work has supported this. A 2012 head-to-head trial found candesartan comparable to propranolol for migraine prevention with fewer side effects.
The mechanism is uncertain — ARBs don’t obviously fit the proposed migraine pathophysiology, though hypotheses include effects on cortical spreading depression threshold and vascular effects via the angiotensin system.
Autonomic effects
Candesartan’s autonomic profile is generally more benign for POTS than the other two:
Blood pressure — candesartan lowers blood pressure through vasodilation. This is the primary concern: in someone with POTS who already has inadequate vasoconstriction on standing, an ARB that further reduces vascular tone could worsen orthostatic symptoms.
Minimal heart rate effect — unlike propranolol, candesartan doesn’t blunt the heart rate response to exercise. Exercise capacity is not meaningfully impaired.
No significant fatigue effect — ARBs are generally well tolerated and the fatigue associated with beta-blockers isn’t a class effect for ARBs.
Possible benefit in small fibre neuropathy — there is speculative but mechanistically interesting evidence that angiotensin system modulation has neuroprotective effects and may slow progression of autonomic neuropathy. This hasn’t been tested in POTS specifically.
The practical concern is whether candesartan’s blood pressure-lowering effect worsens orthostatic intolerance. At low doses (4–8mg) the blood pressure effect is modest. Some patients with POTS find it tolerable; others find it worsens lightheadedness. Starting at the lowest available dose and monitoring lying and standing blood pressure in the first few weeks is important.
My experience
I’m currently taking candesartan 4mg. The headache effect is difficult to isolate from natural variation, but I’ve not experienced significant worsening of orthostatic symptoms at this dose. Blood pressure readings, both lying and standing, remain stable. I’ll continue it through a 3-month trial before forming a conclusion.
Making the choice
For someone with both chronic headache and significant autonomic dysfunction:
Amitriptyline — good headache evidence but alpha-blocking effect worsens orthostatic tachycardia. Likely a poor choice if orthostatic symptoms are prominent.
Propranolol — may help with heart rate in POTS at low doses, but impairs exercise capacity and causes fatigue. May be useful for primarily headache-dominated presentations where exercise intolerance is minimal.
Candesartan — weaker but real headache evidence, benign exercise profile, main concern is blood pressure. May be the best fit for someone with POTS where exercise capacity needs to be preserved.
Other options worth noting (outside the scope of this post): topiramate (no autonomic concerns but significant cognitive side effects), sodium valproate (weight and metabolic effects), and for migraine specifically, the newer anti-CGRP medications (erenumab, fremanezumab). Anti-CGRP biologics have a clean side effect profile and no autonomic concerns, but access on the NHS requires failing two prior preventives and can be difficult.
The best approach is an informed discussion with a neurologist who understands the autonomic complication — which is not universal, and may require some advocacy.
References
Silberstein SD, Holland S, Freitag F, et al. Evidence-based guideline update: pharmacological treatment for episodic migraine prevention in adults. Neurology. 2012;78(17):1337–1345.
Tronvik E, Stovner LJ, Helde G, Sand T, Bovim G. Prophylactic treatment of migraine with an angiotensin II receptor blocker: a randomized controlled trial. JAMA. 2003;289(1):65–69.
Stovner LJ, Linde M, Gravdahl GB, et al. A comparative study of candesartan versus propranolol for migraine prophylaxis. Cephalalgia. 2014;34(7):523–532.
Raj SR, Black BK, Biaggioni I, et al. Propranolol decreases tachycardia and improves symptoms in the postural tachycardia syndrome. Circulation. 2009;120(9):725–734.
Amitriptyline versus placebo in the prophylactic treatment of migraine without aura. Cochrane Database Syst Rev. 2015;7:CD010236.