In the 1990s, oncologists first noted anecdotal evidence that SSRIs provided hot flash benefit in men and women with cancer [16, 17]. The first randomized controlled trial (RCT) to investigate an SSRI for hot flashes was published in 2000; venlafaxine showed modest benefit for VMS after 4 weeks in breast cancer survivors or women at high risk for breast cancer [18]. It was not until 2015 that an SSRI was approved by the FDA for the treatment of hot flashes (paroxetine mesylate 7.5 mg daily). Studies show benefit of approximately one hot flash per day with this product [19]. Some oncologists continue to have concern about the theoretical risk of paroxetine use in women taking tamoxifen (potential decrease in effectiveness due to CYP2D6 polymorphism) [20], but clinical evidence to support this risk is lacking [21, 22]. All of the other nonhormonal pharmacotherapy products discussed in this chapter are not FDA approved, and their utilization for menopausal symptom management represents an “off-label” use.

A multitude of reviews are available that describe the efficacy of SSRIs for menopause management [20], but on detailed inspection, only a handful of studies include requisite criteria that allow inter-study comparisons and provide sufficient data to judge efficacy. These criteria include (1) an adequate control group; (2) participants were midlife women with hot flashes (not depressed populations); (3) follow-up was at least 6–8 weeks duration; (4) there was at least 1 week of baseline hot flash data, preferably 2–3 weeks; (5) the population studied had sufficient hot flashes at baseline to evaluate change (at least 6 hot flashes per day); (6) there was data on frequency and severity of hot flashes; and (7) the outcomes were described as change in frequency from baseline compared with placebo.

Using these criteria, both SSRIs and SNRIs decreased hot flashes by approximately 1–3 hot flashes per day above the decreases demonstrated with placebo (Table 3.1). Serotonergic medications with efficacy greater than placebo in RCTs include the SSRIs fluoxetine, escitalopram, and paroxetine and the SNRIs venlafaxine and desvenlafaxine. One study included in Table 3.1 was only 4 weeks in duration, and was an outlier at an improvement of 3.9 hot flashes per day above placebo, but is included in the table because of its size and historical significance [18].

Studies evaluating serotonergic medications for VMS that did not meet the comparison criteria outlined above, and that were not included in Table 3.1, are discussed below. Paroxetine studies not included had either a crossover design [23] or low baseline VMS and limited power [24]. A single study evaluating citalopram lacked baseline measures [25] as did a single study of venlafaxine [26]. Other studies evaluating venlafaxine included a crossover design [27] and an open-label design [28]. Two studies of sertraline incorporated crossover designs [29, 30].

One study of sertraline for VMS met criteria as listed above, but did not find benefit [31]. With approximately 50 women per group, this study had limited power to detect small differences. Notably, there were group differences at baseline. Women taking placebo were older, more likely to be Caucasian, and were of higher socioeconomic status (SES) compared to women in the sertraline group, who were younger, more likely to be African American, and were of lower SES. These differences, in addition to small numbers, may have confounded the null study findings.

There were no trials of fluoxetine that fit inclusion criteria for review. One study lacked baseline measures [25], and another was a crossover design [32]. No RCTs of duloxetine, fluvoxamine, or quetiapine XR exist.

Serotonergic medications are FDA approved for depression and anxiety and are also commonly used to treat post-traumatic stress disorder (PTSD), eating disorders, and obsessive-compulsive disorder [33, 34]. They are known to bind to serotonin receptors in the brain, vascular endothelium, gut, and bone. Adverse effects of serotonergic medications observed in populations for FDA-approved usages (e.g., anxiety and depression) vary, but in general include nausea, insomnia, dizziness, headache, dry mouth, diarrhea, bloating, constipation, insomnia, and sexual dysfunction. Other associations include increased risk of fractures, akathisia, and photosensitivity. In general, these medications are extremely well tolerated and safe in the doses recommended here.

If an SSRI or an SNRI is stopped abruptly, SSRI withdrawal symptoms can occur (dizziness, nausea, diarrhea, sweating, anxiety, irritability), but these symptoms are uncommon in women using SSRIs/SNRIs for hot flashes because of the low doses used. SSRI withdrawal symptoms do not occur with fluoxetine because of its long half-life. When stopping venlafaxine, the recommendation is to reduce the dose by no more than 50 % every 3–4 days until discontinued (personal opinion). For desvenlafaxine, the recommendation is to give a full daily dose (50 mg) less frequently, e.g., alternate days for 1 week. This approach of every-other-day dosing can be used for all SSRIs and SNRIs once you have reached the lowest available dose. In all cases, if withdrawal symptoms emerge during discontinuation, raise the dose to stop the symptoms and then restart withdrawal more slowly. SSRI withdrawal symptoms occur from lowering doses of serotonin and are in contradistinction to the serotonin syndrome (agitation, confusion, dilated pupils, increased heart rate, elevated blood pressure, clonus, diarrhea, sweating, shivering, headache) that is due to an increase in blood serotonin levels [34]. Serotonin syndrome is unlikely to occur with any of the SSRI/SNRI doses recommended for the management of menopausal VMS but can occur with high doses of SSRIs or with combinations of medications (e.g., monoamine oxidase inhibitors [MAOIs] and SSRIs, SSRIs and SNRIs, SSRIs and triptans) [34].

Similar criteria were applied to examine existing data on GABAergic pharmacotherapies for VMS:

Using these inclusion criteria, studies evaluating gabapentin immediate release (IR), gabapentin gradual release (GR), and pregabalin have shown benefit for reducing VMS, similar to SSRIs and SNRIs – a reduction of approximately 1–3 hot flashes per day above placebo (Table 3.2). A single study of gabapentin IR of 4-week duration is included in Table 3.1 because of the number of women in the study and its otherwise strong design, but we acknowledge it did not strictly meet the criteria outlined above [35]. Studies evaluating gabapentin IR and not included in Table 3.2 either did not have a control group and were not blinded [36] or did not display outcomes as hot flash frequency, so they could not be compared with other studies [37].

Gabapentin is FDA approved as an adjuvant for seizures and for the treatment of postherpetic neuralgia. It is commonly used to relieve other types of neuropathic pain (diabetic neuropathy and central neuropathic pain) and restless legs syndrome. It is also commonly prescribed for off-label use to treat anxiety, insomnia, and mood disorders. Pregabalin, like gabapentin, is commonly used for seizure disorders and neuropathic pain. It is used off label for anxiety and is FDA approved for use in fibromyalgia. Gabapentin and pregabalin are structural analogues of the inhibitory neurotransmitter γ-aminobutyric acid (GABA).

The most common side effects of gabapentin and pregabalin, observed in over 10 % of individuals when used for FDA-approved indications, are dizziness and drowsiness. Both drugs may also produce sexual dysfunction in a smaller percentage of patients (loss of libido and inability to orgasm) [38]. Additional potential side effects are ataxia, blurred vision, headache, diplopia, euphoria, confusion, vivid dreams, irritability, memory impairment, tremors, dysarthria, paresthesias, vertigo, dry mouth, GI side effects, weight gain, and peripheral edema. Notably, pregabalin can increase creatine kinase and decrease platelets (package insert [http://​labeling.​pfizer.​com/​ShowLabeling.​aspx?​id=​561]).

It is important to titrate doses to balance benefit with side effects. This is most effectively done by increasing doses slowly and to dose GABAergic medications predominantly before bed to take advantage of potential sleep benefit of the drowsiness side effect.

Similar criteria that were applied to serotonerigic and gabaergic medications were used to examine clonidine for the treatment of VMS.

SSRIs for VMS have not been compared head-to-head with “standard-dose” hormonal therapies. A single trial included venlafaxine and low-dose oral estrogen and showed similar benefit [44] (Fig. 3.2). A single study of gabapentin and standard-dose estrogen showed similar benefit, but there were only 20 women in each arm, and the gabapentin dosing was quite high, often a dose not tolerated by many women (placebo, gabapentin 2400 mg, and conjugated equine estrogen 0.625 mg) [37] (Fig. 3.2).

Estrogen benefits for VMS follow dose-response curves (Fig. 3.3). This does not appear to be so for serotonergic or GABAergic medications. Rather, there seems to be a threshold dose at which serotonergic and GABAergic medications become effective and higher doses do not have added benefit for VMS (unlike mood and pain disorders). Whether choosing an SSRI, SNRI, GABAergic medications, or clonidine, the “rule of thumb” is to use low-dose therapies. Almost across the board, low-dose formulations appear just as effective as the higher-dose formulations used for mood, pain, or hypertensive disorders, and lower doses afford a lower-risk side effect profile (Fig. 3.4). Benefit from serotonergic and GABAergic medications may plateau by 4 weeks, whereas improvement with estrogen may plateau by 8 weeks and is dose dependent [45]. Optimal doses of serotonergic and GABAergic medications are listed in bold type in Tables 3.1 and 3.2.

Mood, sleep, sexual function, weight gain, and quality of life are important considerations when choosing a nonhormonal pharmacotherapy, just as when considering hormonal options. Very few RCTs have evaluated serotonergic or GABAergic medication effects beyond those described for VMS benefit, and head-to-head comparisons with hormone therapy are limited. RCT findings are summarized in Table 3.3. Additional details are provided below.