Nonhormonal Management of Menopause-associated Vasomotor Symptoms: 2015 Position Statement of The North American Menopause Society

The North American Menopause Society (NAMS). 2015;22(11):1155-1174. 

In This Article

Prescription Therapies

A low-dose paroxetine salt (7.5 mg/d) is the only nonhormonal pharmaceutical approved by the US Food and Drug Administration (FDA) for the treatment of moderate to severe menopausal VMS, with improvements found in VMS frequency and severity up to 24 months and improvements in sleep disruption without negative effects on libido or weight gain.[120] Level I evidence

Many selective serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine uptake inhibitors (SNRIs), along with gabapentin and clonidine, have been tested and have shown some degree of efficacy in symptomatic menopausal women. Onset of action is rapid, usually within 2 weeks. Clinical trials performed for FDA approval are evaluated for significance over placebo at 4 weeks and 12 weeks, and persistence is evaluated at 12 to 24 weeks. Level I to II evidence

Limited evidence suggests that menopausal HT is more effective than nonhormonal agents in reducing the frequency and severity of hot flashes. Head-to-head RCTs are limited, with comparisons including varying types, doses, and routes of administration with nonhormonal agents.[10,121,122,123,124] One trial[121] reported that 75 mg per day of venlafaxine was as effective as a low-dose oral estradiol 0.5 mg per day. In this RCT, oral estradiol reduced the frequency of hot flashes by 2.3 more per day than placebo (P < 0.001), whereas venlafaxine reduced the frequency of hot flashes by 1.8 more per day than placebo (P=0.005) However, this trial did not allow dose escalation, in which case estradiol would be expected to provide 77% improvement in hot flashes on average.[125] Level II evidence

Selective Serotonin Reuptake Inhibitors and Serotoninnorepinephrine Reuptake Inhibitors

Meta-analyses,[10,126,127] a pooled analysis,[128] a Cochrane review,[124] and a review focused on evidence in cancer survivors[129] provide evidence that SSRIs and SNRIs are associated with mild to moderate improvements in symptomatic postmenopausal women, regardless of whether menopause is natural or surgical. The reviews are limited by variability in inclusion criteria, population tested, dosing, length of treatment, and outcomes tested. Those with statistically significant reductions in hot flashes in large, randomized, double-blind, placebo-controlled trials of symptomatic women include paroxetine, escitalopram, citalopram, venlafaxine, and desvenlafaxine. Reduction in hot flashes varies from 25% to 69%, with improvements in composite hot flash frequency and severity from 27% to 61%. Less consistent results have been seen with sertraline and fluoxetine (statistically insignificant trend toward improvement in hot flashes).[10,130,131,132,133,134] Level I to II evidence

Contraindications to SSRIs and SNRIs include prior neuroleptic syndrome, serotonin syndrome (beware of possible synergy with other medications), and concurrent use of monoamine oxidase inhibitors. Exercise caution for patients with bipolar disease, uncontrolled seizures, liver or kidney insufficiency, uncontrolled hyponatremia or poorly controlled hypertension, concurrent use of other SSRIs or SNRIs, or relevant polymorphisms in cytochrome P450 enzyme pathways. For women using tamoxifen, coadministration of SSRIs may lead to inhibition of CYP2D6 (the enzyme that converts tamoxifen to its most active metabolite, endoxifen). The most potent inhibition of CYP2D6 occurs with paroxetine and fluoxetine, so these should be avoided in tamoxifen users. Safer choices include venlafaxine or desvenlafaxine (SNRIs) or escitalopram or citalopram (SSRIs). Black box warnings include uncommon suicidal thoughts within first few months. Possible reported risks include increased risk of bone fracture (mixed reports of bone loss and fracture)[135,136]; SNRIs may produce significant nausea or dizziness, which improves after 1 to 2 weeks.

Suggested dosing for paroxetine salt is 7.5 mg per day; for paroxetine, 10 to 25 mg per day; for escitalopram, 10 to 20 mg per day; for citalopram, 10 to 20 mg perday; for desvenlafaxine, 100 to 150 mg per day; and for venlafaxine, 37.5 to 150 mg per day ( Table 3 ).


Gabapentin is an FDA-approved antiepileptic drug used for diabetic neuropathy and postherpetic neuralgia. In multiple trials at 900mg (300mg 3x/d), it improved the frequency and severity of VMS.[137,138] Adverse events include dizziness, unsteadiness, and drowsiness at week 1 (compared with placebo), which improves by week 2 and is back to baseline by week 4. Gabapentin extended release at asymmetric dosing of 1,800mg per day was shown to be effective but was not approved by FDA for VMS. Higher doses of gabapentin (titrated to 2,400mg/d) were as effective as estrogen (CEE 0.625/d) at reducing hot flash severity scores in a placebo-controlled trial. However, adverse events of gabapentin at this dose included dizziness, headache, and disorientation, so the effectiveness was limited by these.[123] Gabapentin may be a good choice for women with disruptive sleep from VMS because drowsiness is an adverse event. Pregabalin is effective in relieving hot flashes but is less well studied.[139] Level I evidence

Black box warnings for gabapentin and pregabalin include uncommon suicidal thoughts or behaviors. Adverse events include drowsiness, dizziness, and impaired balance or coordination. Pregabalin may impair memory or concentration.

Suggested dosing for gabapentin is 900 mg per day to 2,400 mg per day, and for pregabalin, 150 to 300 mg per day ( Table 3 ).


Clonidine is a centrally active α-2 adrenergic agonist that has been shown to be modestly more effective than placebo[10] but less effective than SSRIs, SSNRIs, and gabapentin.[10,124]However, it is used infrequently because of adverse events, including hypotension, lightheadedness, headache, dry mouth, dizziness, sedation, and constipation. Sudden cessation can lead to significant elevations in blood pressure. Level II evidence