Hot Flashes: Drugs That Cause or Treat Them (and Why You Get Them)

Hot flashes are caused by a narrowing of the brain's thermoregulatory zone that follows a drop in estrogen, driven largely by overactive neurokinin B signaling in the hypothalamus. Medications that block estrogen or its signaling, such as tamoxifen, aromatase inhibitors, and GnRH agonists, reproduce this same mechanism and are well documented causes of drug-induced hot flashes. Systemic estrogen therapy is the most effective treatment where it is not contraindicated; for people who cannot or choose not to use estrogen, fezolinetant (brand name Veozah), an FDA-approved neurokinin-3 receptor antagonist that reached the US market in May 2023, is the first non-hormonal drug built specifically around this mechanism.

Why do hot flashes happen?

A cluster of neurons in the hypothalamic arcuate nucleus, sometimes called KNDy neurons because they co-express kisspeptin, neurokinin B, and dynorphin, normally has its activity dampened by estrogen. When estrogen falls, this suppression is lost, neurokinin B signaling increases, and the brain's "thermoneutral zone," the narrow range of core temperature the body tolerates without triggering a heat-loss response, narrows. Small, otherwise unnoticed shifts in body temperature then trigger flushing, sweating, and a rise in heart rate. This general mechanism (estrogen withdrawal narrowing the thermoneutral zone through NKB signaling) is well established in the menopause physiology literature, though the exact magnitude of the effect varies between studies and individuals.

A typical episode lasts roughly one to five minutes, with visible skin flushing over the chest and face, a rise in heart rate, and peripheral vasodilation. When this happens during sleep it is called a night sweat, and it can fragment sleep enough to cause daytime fatigue and concentration problems, though the size of that cognitive effect is not precisely quantified in the literature and varies by study population.

Duration varies widely. Large cohort studies of the menopause transition, most notably the Study of Women's Health Across the Nation (SWAN), have followed thousands of women over many years and found that frequent, moderate-to-severe vasomotor symptoms commonly persist for several years, with longer duration in women whose symptoms started earlier, before the final menstrual period, and with disproportionately longer and more frequent symptoms reported among African-American participants. Readers who want the exact median duration figures from SWAN should verify them against the original JAMA Internal Medicine publication rather than assume a specific number, since duration estimates differ somewhat across published SWAN analyses.

Which medications can cause or worsen hot flashes?

Several medications and clinical situations trigger hot flashes by disrupting estrogen signaling, altering central thermoregulation, or provoking vasomotor instability directly. This is an established, mechanism-supported category, though exact frequency percentages below are commonly cited ranges from clinical literature and should be confirmed against a current label or systematic review before being quoted precisely.

Tamoxifen and aromatase inhibitors. Tamoxifen is a selective estrogen receptor modulator (SERM) used in breast cancer treatment and prevention; hot flashes are one of its most common side effects. Aromatase inhibitors (anastrozole, letrozole, exemestane) suppress estrogen synthesis almost completely in postmenopausal women and are generally reported to cause a heavier vasomotor burden than tamoxifen.

GnRH agonists and antagonists. Leuprolide, goserelin, and degarelix suppress ovarian or testicular sex hormone production, a strategy sometimes called medical castration. Men treated with androgen deprivation therapy (ADT) for prostate cancer commonly develop hot flashes through the same hypothalamic mechanism seen in menopause, and symptoms can persist for a period after the drug is stopped.

Raloxifene. This SERM, used for osteoporosis and breast cancer risk reduction, is associated with an increased rate of hot flashes compared with placebo in its trial program.

Opioids and opioid antagonists. Long-term opioid use can suppress hypothalamic GnRH pulsatility, producing low sex hormone states that include hot flashes as a symptom. Naltrexone and naloxone, which block opioid receptors, can precipitate acute withdrawal symptoms, including flushing, in opioid-dependent patients.

Antidepressants, used both ways. Venlafaxine and low-dose paroxetine are used to treat hot flashes (see below), but some patients, especially early in treatment, report new or worsened flushing from SSRIs and SNRIs, plausibly related to acute serotonergic or noradrenergic effects. Bupropion has been reported anecdotally to trigger flushing in some patients.

Other recognized triggers:

• Niacin (nicotinic acid): causes a distinct prostaglandin-mediated cutaneous flush that looks similar to a hot flash but has a different mechanism.
• Calcium channel blockers, particularly nifedipine: peripheral vasodilation can mimic or worsen flushing.
• Calcineurin inhibitors (cyclosporine, tacrolimus): flushing has been reported in transplant recipients.
• Alcohol: accelerates estrogen metabolism and directly dilates cutaneous blood vessels.

If hot flashes start abruptly and closely track the start of a new medication, that timing itself is a useful diagnostic clue and worth mentioning to a prescriber before assuming it is simply menopause.

Which treatments actually work, and how well?

Estrogen-based hormone therapy

Systemic estrogen, with or without a progestogen, is the treatment with the largest and most consistent effect on hot flash frequency and severity across the menopause hormone therapy literature, and it is the reference standard against which non-hormonal drugs are compared. Options include oral estradiol, transdermal patches, gels, and sprays. Women with an intact uterus need a progestogen alongside estrogen to protect the uterine lining; micronized progesterone is generally considered the more physiologic progestogen choice, though comparative cardiovascular and breast safety data between progestogen types come mostly from observational cohorts rather than head-to-head trials, and causal claims about one progestogen being safer than another should be treated cautiously.

The Women's Health Initiative (WHI) context. The WHI is the large randomized trial most often cited when discussing hormone therapy risk. Its combined estrogen-plus-progestin arm found an increased relative risk of invasive breast cancer after several years of use, while its estrogen-only arm, studied in women who had had a hysterectomy, did not show the same increase and in some analyses trended in the opposite direction. Current guideline-level thinking, reflected in Menopause Society (formerly the North American Menopause Society) position statements, generally holds that for women younger than 60 or within about 10 years of menopause onset who do not have contraindications, the benefits of hormone therapy for treating bothersome vasomotor symptoms outweigh the risks for most individuals. Exact hazard ratios and absolute risk numbers from WHI are widely available but vary slightly depending on which follow-up analysis is cited, so a clinician should confirm current figures before quoting them to a patient.

Who should generally avoid systemic estrogen: a personal history of estrogen-receptor-positive breast cancer or other estrogen-sensitive malignancy, a history of venous thromboembolism or stroke, or active liver disease. These are widely recognized contraindications, though the right answer for an individual patient depends on the details of their history and should be worked out with a prescriber.

Fezolinetant (Veozah): the first non-hormonal NK3 antagonist

Fezolinetant blocks the neurokinin-3 receptor on the KNDy neurons described above, interrupting the signal that narrows the thermoneutral zone. The FDA approved it in May 2023, based on the SKYLIGHT phase 3 trial program, for moderate-to-severe vasomotor symptoms associated with menopause. In its pivotal trials, fezolinetant reduced hot flash frequency more than placebo over 12 weeks; exact percentage reductions differ slightly across the published SKYLIGHT 1 and SKYLIGHT 2 papers and should be checked against the current FDA label rather than a single remembered figure. The drug requires baseline liver function testing and periodic monitoring afterward, and it is not recommended in patients with cirrhosis or severe renal impairment. It is not FDA-approved for use in men, and it has not been studied in men on androgen deprivation therapy.

Fezolinetant is generally positioned as the first-choice non-hormonal option for people who cannot or prefer not to use estrogen, including many breast cancer survivors, though the choice between it and an SSRI/SNRI or gabapentin should account for cost, coverage, and individual tolerability.

Low-dose paroxetine (Brisdelle)

Paroxetine 7.5 mg is a lower dose than the antidepressant dose range (20 to 60 mg) and is the only SSRI formulation with FDA approval specifically for vasomotor symptoms. Trials of this low dose have shown a meaningfully greater reduction in hot flash frequency compared with placebo, though, as with the other agents above, exact percentage figures vary by publication and should be verified before being quoted precisely.

A clinically important interaction: paroxetine is a strong CYP2D6 inhibitor and can substantially reduce the conversion of tamoxifen to its active metabolite, endoxifen. Patients taking tamoxifen for breast cancer should generally avoid paroxetine and consider venlafaxine, gabapentin, or fezolinetant instead, a point worth raising explicitly with an oncologist or prescriber rather than assuming any SSRI is interchangeable.

Venlafaxine and other SNRIs

Venlafaxine, typically used off-label at doses below the standard antidepressant range, is one of the most widely used non-hormonal options for hot flashes, including in breast cancer survivors who cannot take estrogen or paroxetine. Desvenlafaxine and duloxetine appear to have similar, though not necessarily identical, effects in comparative literature. All SNRIs require a gradual taper rather than abrupt discontinuation to avoid discontinuation symptoms.

Gabapentin

Gabapentin is used off-label for hot flashes and appears as a recognized second-line non-hormonal option in menopause society guidance, despite lacking a specific FDA indication for this use. Sedation is the main limiting side effect, which is why some clinicians favor giving the full daily dose at bedtime for patients whose primary complaint is night sweats rather than daytime flashes.

Clonidine

Clonidine, an alpha-2 adrenergic agonist, reduces hot flash frequency to a smaller degree than estrogen, fezolinetant, or the antidepressant options, and its use is limited by orthostatic hypotension and dry mouth. It remains a reasonable fallback when other options are contraindicated or not tolerated, particularly outside of patients already on other blood-pressure-lowering drugs.

Oxybutynin

Oxybutynin, an anticholinergic drug primarily used for overactive bladder, has shown efficacy for hot flashes in trial data and is included as an option in some obstetric and gynecologic society guidance, particularly useful in patients who also have overactive bladder symptoms. Dry mouth is the main dose-limiting side effect.

Progestins (megestrol acetate, medroxyprogesterone acetate)

Progestin-only regimens can meaningfully reduce hot flash frequency, but they are not typically used as first-line therapy because megestrol acetate is a synthetic progestogen with theoretical potential to stimulate hormone-sensitive tissue, and its long-term safety in breast cancer survivors specifically is debated rather than settled.

Non-drug approaches

Cognitive behavioral therapy (CBT) has been studied for hot flashes and appears to help people cope with symptoms and reduce how bothersome they feel, though its effect on the actual frequency of hot flashes is less consistent across trials. Keeping the bedroom cool, wearing moisture-wicking fabrics, and modest weight loss are recommended as reasonable adjuncts in guideline documents, though the size of their effect is less precisely quantified than for the drug therapies above and mostly comes from observational or small trial data. Isoflavone (phytoestrogen) supplements show modest and inconsistent effects across meta-analyses and are not considered a substitute for pharmacologic therapy in people with severe symptoms.

One small triple-blind randomized trial examined chamomile (Matricaria chamomilla) for its effect on quality of life in women with menopausal symptoms; this is a single trial in a herbal supplement rather than an established treatment, and its findings should be treated as preliminary rather than practice-changing until replicated (Effect of Matricaria chamomilla on the Quality of Life Among Women with Menopausal Symptoms).

Is hormone therapy actually used the way guidelines suggest?

Guideline statements about who should be offered hormone therapy do not automatically describe who receives it or who stays on it in practice. Observational research on hormone therapy initiation and interruption in real-world US cohorts has found that use patterns are shaped by factors well beyond the classic contraindication checklist, including co-occurring conditions and substance use, and that treatment interruption is common for reasons that have nothing to do with efficacy (Substance use and menopausal hormone therapy: treatment initiation and interruption among US women with and without HIV, 2008-2019). The practical takeaway is that a patient's actual hormone therapy history, including gaps and restarts, is worth reviewing explicitly rather than assumed to match a guideline pathway.

A decision framework for choosing a hot flash treatment

This is an original decision aid built for this article. It is meant to organize a conversation with a prescriber, not to replace one, and it does not set an individual dose.

Step 1: Rule out a drug or disease cause before assuming "menopause."

Step 2: If it is menopause-related, check for contraindications to estrogen.

A history of estrogen-receptor-positive breast cancer, venous thromboembolism, stroke, or active liver disease shifts the conversation toward non-hormonal options first. Absence of these does not automatically mean estrogen is the right choice, but it keeps it on the table for shared decision-making with a prescriber, generally most favorably in women under 60 or within about 10 years of menopause onset.

Step 3: Match a non-hormonal option to what else is going on, if estrogen is not an option or not wanted.

Step 4: Set a reassessment point rather than staying on an ineffective drug indefinitely.

Most non-hormonal options show their main effect within about 8 to 12 weeks. If a treatment has not produced a clear improvement by then, the more useful next step is usually switching rather than adding a second agent, since combining drug classes for hot flashes has not been well tested in randomized trials.

When should you see a clinician rather than manage this on your own?

Most hot flashes reflect ordinary hormonal change and do not require urgent evaluation. Some presentations should prompt a visit sooner rather than later:

• Hot flashes beginning before age 40
• Hot flashes in a man who has never taken androgen-suppressive therapy
• Hot flashes accompanied by unintentional weight loss, drenching night sweats, or lymph node swelling
• Hot flashes that started abruptly after a new prescription

Diagnosis of ordinary menopausal hot flashes is clinical, based on the symptom pattern, and does not require a specific test. When the picture is atypical, clinicians typically use FSH testing to help confirm ovarian insufficiency, TSH to rule out thyroid disease (which can mimic vasomotor symptoms), and, when carcinoid syndrome or pheochromocytoma are genuinely suspected, urine or plasma catecholamine or metanephrine testing. These are used to exclude alternative causes, not to diagnose typical menopausal hot flashes themselves.

Special populations

Men on androgen deprivation therapy. Hot flashes are a common side effect of ADT for prostate cancer, driven by the same hypothalamic mechanism seen in menopause. Low-dose progestins, venlafaxine, and, outside the United States, cyproterone acetate have supporting evidence in this population. Fezolinetant has not been studied in men and is not an appropriate substitute without further evidence.

Breast cancer survivors. This is the population where avoiding estrogen and avoiding paroxetine-tamoxifen interactions matters most. Systematic review evidence in breast cancer survivors supports SSRIs and SNRIs, as well as gabapentin and clonidine, as non-hormonal options with meaningful, if generally smaller-than-estrogen, effects; exact effect sizes vary by review and should be checked against a current Cochrane or comparable systematic review rather than a single remembered number.

Transgender women. Adequate-dose exogenous estrogen effectively eliminates hot flashes in transgender women, and stopping or reducing estrogen, including around some gender-affirming surgeries, can reproduce the same vasomotor symptoms seen in menopause.

People with fibromyalgia or chronic pain conditions during the menopause transition. Vasomotor symptoms and chronic widespread pain conditions can overlap and interact during midlife hormonal change, an area of ongoing research rather than settled practice; a clinician managing hot flashes in someone with significant comorbid pain should account for that overlap when choosing a drug, since some non-hormonal hot flash treatments (for example, gabapentin) are also used for pain.

What is established, what is plausible, and what is not established

Established: estrogen withdrawal narrows the hypothalamic thermoneutral zone through neurokinin B signaling; drugs that block estrogen or its production (tamoxifen, aromatase inhibitors, GnRH agonists) commonly cause hot flashes; systemic estrogen therapy is the most effective treatment where it is not contraindicated; fezolinetant is FDA-approved specifically for this indication as of May 2023; paroxetine meaningfully reduces tamoxifen's active metabolite and should generally be avoided in patients on tamoxifen.

Plausible but not firmly quantified: the precise percentage reductions in hot flash frequency for several of the drugs discussed above vary meaningfully across published trials, and single-number claims (for example, an exact percentage reduction for gabapentin, clonidine, or oxybutynin) should be checked against the specific trial or a current systematic review rather than treated as fixed facts. The effect of core-temperature cooling strategies and modest weight loss is directionally supported but not tightly quantified.

Not established: combining an SSRI/SNRI with gabapentin for additive effect, benefit of fezolinetant in men, and the comparative long-term breast safety of different progestogens used alongside estrogen are all areas where the evidence is incomplete or the studies available are observational rather than randomized.