Paroxetine for Hot Flashes

Low-dose paroxetine is its own FDA-approved product, not an antidepressant prescription written small. Each capsule holds 9.69 mg of paroxetine mesylate, equal to 7.5 mg of paroxetine base, and carries one approved indication: the treatment of moderate to severe vasomotor symptoms associated with menopause. FDA approved the original application, NDA 204516, on June 28, 2013, and a generic 7.5 mg capsule now carries that same vasomotor indication. The molecule did not change. The dose and the approved job did.

The label does more than stay quiet about psychiatry. Under Limitations of Use it states the capsule is not indicated for any psychiatric condition, that its dosage is lower than the dosage used for major depressive disorder, obsessive compulsive disorder, panic disorder, generalized anxiety disorder, social anxiety disorder and post-traumatic stress disorder, and that safety and effectiveness at this lower dosage have not been established for any of them. If you need paroxetine for a psychiatric diagnosis, the label directs stopping this capsule and starting a product indicated for that use. Two prescriptions, one active ingredient, two separate decisions.

How it quiets a hot flash is not settled science. The label states directly that the mechanism of action for vasomotor symptoms is unknown, and that this product is not an estrogen. What is documented is narrower: paroxetine blocks serotonin uptake into human platelets at clinically relevant doses and has weak effects on norepinephrine and dopamine reuptake. Any explanation you read that ties serotonin neatly to your body's temperature set point is a hypothesis the label itself declines to endorse.

Two Phase 3 trials in 1,174 postmenopausal women carried the approval, and entry required a minimum of 7 to 8 moderate to severe episodes a day for the 30 days before dosing. In the 12-week trial of 606 women, median daily frequency fell by 5.9 from a baseline median of 10.4 on the capsule and by 5.0 on placebo, a treatment difference of 0.9 episodes a day. In the 24-week trial of 568 women, frequency fell by 5.6 from a baseline of 9.9 against 3.9 on placebo, a difference of 1.7.

Read those two lines together and the placebo column is the part most summaries drop. A dummy capsule removed 3.9 to 5.0 hot flashes a day. The medication's own contribution above that was roughly one to under two episodes daily. At week 24 in the longer trial, 48% reached a 50% or greater cut in frequency against 36% on placebo. Severity barely moved: the week 12 severity difference was 0.04 in the first trial and did not reach significance, and 0.05 in the second on a scale where baseline severity sat at 2.5.

A percentage would make all of this sound bigger than it is. Going from ten hot flashes a day to four is a genuine change in your day, and so is a placebo arm going from ten to five. Whether roughly one extra avoided episode a day justifies a daily SSRI with a CYP2D6 interaction is a judgment a prescriber makes with you, weighing your symptom load against everything else you take. None of this is medical advice.

Paroxetine is a strong CYP2D6 inhibitor, and the label's interaction table lists reduced plasma concentrations of tamoxifen's active metabolite with the recommendation to consider avoiding concomitant use. The warnings section is blunter: some studies show tamoxifen efficacy, measured by breast cancer relapse and mortality, may be reduced alongside paroxetine because paroxetine irreversibly inhibits CYP2D6 and lowers tamoxifen blood levels, while other studies failed to demonstrate that risk. Tamoxifen has to be converted before it works.

Block the converting enzyme and you are no longer interfering with a hot flash drug. You are interfering with a cancer drug.

The two anchor studies disagree, and both deserve to be on the table. An Ontario cohort of 2,430 women taking tamoxifen plus a single SSRI found that absolute increases of 25%, 50% and 75% in the share of tamoxifen time overlapping paroxetine were associated with 24%, 54% and 91% increases in the risk of breast cancer death, with no comparable signal for other antidepressants. A five-database US study of 6,067 and 8,465 tamoxifen starters found a pooled hazard ratio for death of 0.96, 95% confidence interval 0.88 to 1.06, for potent CYP2D6-inhibiting SSRIs versus other SSRIs.

A systematic review of 15 studies covering close to 100,000 women concluded there was no consistent negative effect from concurrent antidepressant and tamoxifen use and argued the guidelines should be revisited. A scientific argument is not the same as clearance. The label recommendation still stands, other approved hot flash options do not touch CYP2D6 at all, and if you take tamoxifen this is the first thing to raise, ahead of price, ahead of convenience. Bring the full list of what you swallow, including anything you buy without a prescription.

The 2023 nonhormone therapy position statement of The North American Menopause Society recommends SSRIs and SNRIs at Level I evidence for vasomotor symptoms, alongside cognitive behavioral therapy, clinical hypnosis, gabapentin and fezolinetant, while concluding that hormone therapy remains the most effective treatment and should be considered within 10 years of the final menstrual period. The same statement frames nonhormone options as the route when hormone therapy is contraindicated, naming estrogen-dependent cancers and cardiovascular disease, or when it is simply not wanted.

The trial group tells you who the evidence describes. Mean age was 55 with a range of 40 to 73, roughly a fifth had surgical menopause, and enrollment was 65% and 76% White with 33% and 22% Black or African American across the two trials. That second figure matters, because menopause trials are frequently criticized for enrolling almost entirely white cohorts, and these two did not. Anyone under 65 fits the studied group; the trials did not enroll enough women aged 65 and over to say whether the response differs.

The exclusions are firm rather than cautionary. It is contraindicated in pregnancy, is not indicated before menopause, and is contraindicated with monoamine oxidase inhibitors taken within the previous 14 days, with thioridazine, with pimozide, and in anyone with known hypersensitivity to paroxetine. The label also advises avoiding SSRIs entirely if you have untreated anatomically narrow angles in the eye without a patent iridectomy, which is the kind of history a symptom questionnaire will not ask about unless it is built to.

Two neurokinin antagonists now sit beside the SSRI route. Fezolinetant, an NK3 receptor antagonist first approved in 2023, carries a boxed warning for hepatotoxicity and requires liver testing before starting, monthly for the first three months, then at six and nine months. Elinzanetant, first approved in 2025, blocks both NK1 and NK3 receptors and is dosed at 120 mg, two 60 mg capsules, once daily at bedtime, with baseline liver testing before the first dose. Neither is an estrogen either, so the same reasoning that rules out hormone therapy does not rule these out.

Their trial numbers look larger, and the comparison is not as clean as it appears. Fezolinetant 45 mg reduced daily frequency by 2.6 and 2.5 episodes more than placebo at week 12 across its two trials. Elinzanetant 120 mg reduced it by 3.22 and 3.24 more than placebo at week 12 in OASIS 1 and OASIS 2. Those are least squares means from mixed models. The paroxetine figures are median changes analyzed with a rank-based method, in separate trials, with different entry thresholds and different baseline burdens. Setting the three side by side in a league table would flatter the newer drugs by construction rather than by evidence.

So the honest position is that no head-to-head trial exists, and the differences that are documented are practical ones. The SSRI route brings a CYP2D6 interaction and a psychiatric class warning. The neurokinin route brings liver monitoring and, in the case of fezolinetant, a boxed warning of its own. A prescriber weighs your medication list, your liver history and your symptom burden. Nothing in the published record supports naming a winner for everyone.

The recommended dosage is 7.5 mg once daily at bedtime, with or without food. There is no titration schedule and no approved step up, which is a real difference from the psychiatric product: paroxetine tablets start at 20 mg daily for major depressive disorder and rise in 10 mg increments at intervals of at least a week to a maximum of 50 mg, or 60 mg for obsessive compulsive disorder and panic disorder. The hot flash dose is a fraction of the lowest routine antidepressant dose and it does not move.

Timing is worth knowing before you judge whether it works. In a 14-day dosing study in 24 healthy postmenopausal women, steady-state concentrations were reached at about 12 days for most, with peak concentrations at a median of 6 hours, although the label notes it can take substantially longer in an occasional case. Both approval trials measured their co-primary endpoints at week 4 and week 12, and both showed a significant frequency separation from placebo by week 4. Four weeks is a fair first checkpoint. Twelve is the one the label was built on.

The interaction list runs well past tamoxifen and is the reason a medication review is not a formality here. Thioridazine and pimozide are contraindicated outright for QT prolongation risk, and the label flags dose consequences for tricyclic antidepressants, risperidone, atomoxetine, warfarin, digoxin and theophylline, plus caution with other CYP2D6 substrates such as nortriptyline, fluoxetine, phenothiazines and Type 1C antiarrhythmics. Bleeding risk rises alongside aspirin, NSAIDs, antiplatelet drugs and anticoagulants, and the everyday ibuprofen habit counts.

The boxed warning states that SSRIs increased the risk of suicidal thoughts and behavior in pediatric and young adult groups in short-term trials for major depressive disorder and other psychiatric disorders, and that because this capsule is an SSRI it warrants close monitoring for the emergence of those thoughts. The same label adds the caveat that makes the box hard to size at this dose: there is limited information about suicidal thoughts and behaviors in postmenopausal women using it for vasomotor symptoms, because the trials excluded anyone with a present or past psychiatric disorder.

The pooled safety data are small but they are not empty. Across the Phase 2 and Phase 3 trials, 635 women received the capsule and 641 received placebo; three on the capsule reported serious suicidal ideation and one reported a suicide attempt, with no such events reported on placebo. Discontinuation for an adverse reaction ran 4.7% on the capsule against 3.7% on placebo. Four events out of 635 cannot establish a rate, and zero out of 641 cannot rule one out. That ambiguity is precisely why the class warning stayed on a product with a non-psychiatric indication.

The everyday side effects are milder and better quantified. Headache occurred in 6.3% against 4.8% on placebo, fatigue, malaise or lethargy in 4.9% against 2.8%, and nausea or vomiting in 4.3% against 2.3%, with nausea concentrated in the first four weeks and fatigue mostly in the first week, both easing with continued dosing. The label also lists screening for bipolar disorder before starting, plus warnings on serotonin syndrome, hyponatremia, bone fracture association, seizures and sexual dysfunction including reduced libido and delayed or absent orgasm.

Certain symptoms appeared more often when the capsule was stopped than when placebo was stopped, and the label names them: increased dreaming or nightmares, muscle cramps, spasms and twitching, headache, nervousness or anxiety, fatigue and tiredness, a restless feeling in the legs, and trouble sleeping. The label adds that while these were generally self-limiting, serious discontinuation symptoms have been reported with other paroxetine products, particularly after stopping abruptly.

Here is the labeling asymmetry that almost nobody writing about this drug mentions. The paroxetine tablet label carries an explicit instruction in its dosage section to reduce the dosage gradually rather than stop abruptly, while the 7.5 mg capsule label lists discontinuation reactions in its adverse reactions section and prints no dose-reduction instruction at all. The difference is in what the two labels tell a prescriber to do at very different doses. It is not evidence that coming off the low dose is uneventful, and there is no smaller approved strength to step down through.

Which makes the exit plan a question to ask at the start rather than at the end. Ask how a prescriber would have you stop, what to expect in the first week off it, and whether hot flashes returning is being counted as withdrawal or as the original symptom coming back.

One firm rule survives stopping: at least 14 days must pass after the last dose before a monoamine oxidase inhibitor is started.

Every figure here comes from the CMS National Average Drug Acquisition Cost file, which is what pharmacies pay to buy a drug. It is not a retail price, a cash price or a subscription. At the August 19, 2026 effective date, generic paroxetine mesylate 7.5 mg capsules were $2.63265 each and generic paroxetine hydrochloride 20 mg tablets were $0.06308 each. Thirty of the 7.5 mg capsules cost a pharmacy about $79 to acquire. Thirty of the 20 mg tablets cost about $1.89. Forty-two times the price per unit for a smaller dose of the same drug substance is a real number, and the explanation is regulatory rather than pharmaceutical. They are different salts in different dosage forms approved for different indications, and only the capsule carries a vasomotor indication. A 20 mg tablet cannot be substituted at the pharmacy counter for a 7.5 mg capsule, and no fraction of a tablet reproduces 7.5 mg of paroxetine base. Any page implying you can get the same product cheaper by splitting tablets is describing something a prescription cannot authorize.

For scale against the newer route, fezolinetant 45 mg tablets were $18.58505 each and elinzanetant 60 mg capsules were $9.98057 each at the same effective date, both still classified as brand. At label dosing that is roughly $558 a month of acquisition cost for one and roughly $599 for the other, against about $79 for the generic capsule. What any of this costs you depends on your coverage, your pharmacy and any manufacturer program, none of which acquisition cost predicts.

The label builds several steps into prescribing this, and they are the steps a thin intake form skips. Screening for bipolar disorder, including a detailed psychiatric and family history, is specified before starting; pregnancy is a contraindication; and three drug categories are contraindicated outright. A service that collects a hot flash count and ships a capsule without ever seeing what else you take has not done the work the label describes. Whether an online route or an in-person one suits you, that review is the part to insist on.

Come prepared and the visit gets shorter and better. Have your full medication list, including aspirin, ibuprofen and anything bought off a shelf. Say outright whether you take tamoxifen or any breast cancer endocrine therapy. Mention narrow-angle glaucoma, a seizure history, diuretics or previous low sodium, and any history of mania. Ask what counts as moderate to severe in your case, given trial entry required 7 to 8 episodes a day, and ask what happens if four weeks brings nothing.

ManyTreatments compares treatment options and the services that offer them. We do not prescribe. Whether any prescription medication is appropriate for you is a decision for a licensed clinician, and nothing on this page replaces that consultation or counts as medical advice.

Two questions separate a considered prescription from a transaction. What is the plan at week 12 if the response is small, and how would you come off it? Ask both before the first capsule.