Antidepressants as a Treatment for Hot Flashes in Women

Denise R. Kockler; Michelle W. McCarthy


Am J Health Syst Pharm. 2004;61(3) 

In This Article

Literature Review

Anecdotal experience suggesting that these antidepressants effectively decreased hot flashes resulted in the initiation of several pilot studies to formally evaluate their efficacy.[6,27,28] To date, seven published clinical trials have evaluated the efficacy of venlafaxine, fluoxetine, and paroxetine for the management of hot flashes.[6,24,25,26,27,28,29] These publications include three pilot studies; three double-blind, placebo-controlled trials; and one open-label continuation study. Change from baseline in hot flash score was the primary endpoint in all but one of the studies. The hot flash score was determined by summing the number of mild hot flashes, two times the number of moderate hot flashes, three times the number of severe hot flashes, and four times the number of very severe hot flashes.[6,24,25,26,27,29] Although the assessment of hot flashes is subjective and performed by study participants, this method is well accepted and has been validated in prior studies evaluating therapy for hot flashes.[34] Tamoxifen has been associated with the development of hot flashes; therefore, concurrent tamoxifen therapy was allowed if doses were stable and continued throughout the study period.[6,24,25,26,27,28,29,35]

Loprinzi et al.[6] conducted an open-label, five-week, pilot study to determine the efficacy of venlafaxine in the management of hot flashes. Eligible patients included women with a history of breast cancer and men who had androgen-deprivation therapy scheduled for six weeks beyond study entry with an Eastern Cooperative Oncology Group (ECOG) scale performance status of 0-1 and a life expectancy of at least six months. Eligible patients had to experience bothersome hot flashes for at least one month before study entry (minimum, 14 per week). Those who received previous therapy with antineoplastics, androgens, progestins, estrogens, and antidepressants were excluded. After a one-week symptom-assessment period (baseline), participants received venlafaxine 12.5 mg twice daily. Primary measures of efficacy included the average number of daily hot flashes per week and average hot flash score per week. Patients collected hot flash data in daily diaries. Safety was assessed by patient reports of selected adverse effects. A total of 31 patients enrolled in this pilot study; 28 were assessable, and 25 completed the five-week study. The majority of study participants were women (82%) and were receiving tamoxifen therapy (68%). The average number of daily hot flashes decreased from 6.6 (range, 3.1-33.9) at baseline to 4.3 (range, 0-19) at study completion. The number of hot flashes decreased by at least 50% in 54% of the study participants. The incidence of severe and very severe hot flashes decreased from 1.4 per day at baseline to 0.1 per day at the end of the study (p < 0.0002). Hot flash scores were reduced by approximately 55% with venlafaxine. Patients reported reductions in fatigue, dry mouth, sweating, difficulty sleeping, and nervousness at study completion. The majority of patients (64%) elected to continue venlafaxine therapy after the study.

Although the authors concluded that low-dose venlafaxine reduced hot flashes in cancer survivors, the numerous limitations and observational design must be considered. Limitations included the lack of p values with the exception of one secondary endpoint, high dropout rate, small sample size, and short study duration. Also, the authors compared the results from this noncontrolled pilot study to those from prior studies in which placebo controls were used.

Some of the authors of the pilot study then conducted a randomized, placebo-controlled study to further evaluate the efficacy and toxicity of venlafaxine for the treatment of hot flashes in women who had survived breast cancer.[24] Patients included women with a history of breast cancer or those concerned about taking estrogen for fear of breast cancer who were over 18 years of age and had troublesome hot flashes occurring at least 14 times weekly for at least one month before study enrollment. To be eligible, patients should have had a life expectancy of at least six months and ECOG performance status of 0-1. Treatment with raloxifene, tamoxifen, and aromatase inhibitors was permitted if therapy was initiated at least four weeks before study enrollment and continued for the study duration. After a one-week evaluation period (baseline), patients were randomized to receive extended-release venlafaxine 37.5 mg daily for 28 days (n = 56) (group 1); extended-release venlafaxine 37.5 mg daily for 7 days, then 75 mg daily for 21 days (n = 55) (group 2); extended-release venlafaxine 37.5 mg daily for 7 days, 75 mg daily for 7 days, then 150 mg daily for 14 days (n = 54) (group 3); or placebo (n = 56). The primary endpoint was average daily hot flash activity (the number of hot flashes plus hot flash score). Secondary endpoints included changes in mood, assessed by the Beck Depression Inventory, and quality of life, assessed by the Quality of Life instrument, from baseline to end of treatment. The patient characteristics were well matched between the four groups, and approximately 69% of patients were receiving tamoxifen therapy. Complete data were available for 191 of the 221 enrolled patients. After four weeks, the frequency of hot flashes decreased by a median of 19% (95% confidence interval [CI], 14-28%) in the placebo group compared with 30% (95% CI, 22-53%), 46% (95% CI, 36-63%), and 58% (95% CI, 42-67%) in groups 1, 2, and 3, respectively (p < 0.001 for all venlafaxine groups, compared with placebo). The median hot flash score was decreased by 27% (95% CI, 11-34%) for those receiving placebo, 37% (95% CI, 26-54%) for group 1, 61% (95% CI, 60-68%) for group 2, and 61% (95% CI, 48-75%) for group 3 (p < 0.001 for all groups compared with placebo). Hot flash activity was reduced by over 50% in 20% of the placebo patients compared with 45% (95% CI, 33-60%), 63% (95% CI, 47-77%), and 55% (95% CI, 40-69%) of groups 1, 2, and 3, respectively. Depression scores improved by a mean of 1.6 points in the placebo group and 2.4, 4.8, and 3.2 points in groups 1, 2, and 3, respectively. The study reported that overall quality of life decreased by 3 points in the placebo group compared with an average increase of 3 points for the venlafaxine groups (p = 0.02); however, no additional results were provided. Dry mouth, constipation, decreased appetite, and nausea were significantly more common in the venlafaxine-treated patients compared with the placebo group. Venlafaxine 150 mg was associated with more adverse effects than the other doses. The authors concluded that venlafaxine reduced hot flashes in women who were unable or unwilling to receive estrogen and that therapy should be initiated at 37.5 mg daily and increased to 75 mg daily, if necessary. Limitations to this study include the high dropout rate and incomplete presentation of data for several of the comparisons (e.g., changes in blood pressure, differences between patients receiving tamoxifen and those not receiving tamoxifen).

At the completion of the double-blind study, participants were notified of their treatment assignment and had the option of participating in an eight-week, open-label continuation phase.[25] Of the 221 patients in the double-blind study, 157 registered for the continuation phase. At the time of data analysis, complete information was available for 102 patients. Venlafaxine doses ranged from 37.5 to 150 mg per day and were selected by the patient, nurse, and physician. The primary efficacy endpoints were hot flash frequency and score. Tamoxifen was used by 66% of patients. At the completion of this study, 26 patients were receiving 37.5 mg daily, 35 were receiving 75 mg daily, 6 were receiving 112.5 mg daily, and 34 were receiving 150 mg daily. For those initially randomized to placebo, hot flash score decreased by a mean of 62% at the completion of the eight-week period. An additional mean reduction of 26% in hot flash score was reported for the women originally randomized to venlafaxine 37.5 mg. For the 75- and 150-mg groups, the 60% reduction in hot flash score that was achieved in the double-blind study was maintained through the open-label continuation phase. The following adverse events were reported: abnormal sweating (43%), dry mouth (41%), difficulty sleeping (35%), fatigue (33%), constipation (30%), sleepiness (29%), mood changes (13%), nervousness (12%), and loss of appetite (10%). Patients also reported nausea; however, results were presented as a change from baseline instead of overall frequency. The authors concluded that venlafaxine effectively reduced hot flashes in women who were unable or unwilling to take estrogen and that additional benefits were not seen with daily doses in excess of 75 mg. This noncomparative continuation trial did not report any statistical analyses; therefore, any conclusions regarding the extended efficacy of venlafaxine should be made with caution.

Loprinzi et al.[26] conducted an eight-week, double-blind, placebo-controlled crossover trial, evaluating the efficacy of fluoxetine on the frequency and severity of hot flashes. Participants included 87 women with a history of breast cancer (or apprehension regarding the use of estrogen) and a minimum of 14 hot flashes per week for at least one month. Those who received antineoplastics, androgens, estrogens, progestins, or warfarin and previous fluoxetine or antidepressant therapy (within two years of the study) were excluded. Tamoxifen and raloxifene were permitted if doses were stable and continued during the study. Following a one-week symptom assessment period (baseline), patients were stratified and then randomized to receive fluoxetine 20 mg or placebo daily for four weeks. After the initial four weeks of treatment, patients were crossed over to the other treatment group. The primary efficacy variables were the change from baseline for hot flash scores and frequency of hot flashes. Patients recorded hot flash data daily and toxicity data weekly in diaries. Of the 87 women who were enrolled in the trial, 6 canceled their enrollment before receiving any study medication. Of the 81 remaining participants, 72 provided baseline data, 68 provided data before crossover, and 66 provided complete data. Fifty-four percent of the patients used concomitant tamoxifen therapy. During the first four weeks of the crossover phase, the median number of daily hot flashes decreased by 3.4 and 2.5 for patients taking fluoxetine and placebo, respectively (p = 0.54). Reductions in hot flash scores were reported in both treatment groups (50% [fluoxetine] versus 36% [placebo]); however, the between-group difference was not statistically significant (p = 0.35). Subsequent analysis of the crossover data indicated an improvement in hot flash scores and a reduction in the number of daily hot flashes when fluoxetine (p = 0.02) was compared with placebo (p = 0.01). Similar adverse events were reported for the two treatment groups, except for an increased rate of dry mouth with fluoxetine-treated patients. The authors concluded that fluoxetine 20 mg modestly improved hot flashes. However, no differences in primary endpoints between fluoxetine and placebo were reported during the initial phase of the study. Other study limitations included high dropout rate, small sample size, short study duration, and lack of washout period. Additional studies evaluating fluoxetine for hot flashes are warranted.

Stearns et al.[28] conducted a single-institution, open-label, pilot study to assess the efficacy of paroxetine in reducing the frequency and severity of hot flashes in breast cancer survivors. Eligible patients included women over 18 years of age with a history of breast cancer who were without detectable disease with a history of at least 14 hot flashes per week for at least one month, ECOG performance status of 0-2, and a life expectancy of more than six months. Treatment with vitamin E was allowed if it was used regularly for at least six weeks before study enrollment and continued throughout the study. Women taking other agents for hot flashes, hormonal therapy other than tamoxifen, or antidepressants were excluded. The primary endpoints, hot flash frequency and severity, were documented in daily diaries. Patients completed symptom-assessment questionnaires to document secondary endpoints, menopausal symptoms, selected adverse effects of paroxetine, and quality of life as assessed by the EuroQOL survey. The 30 women who enrolled in the trial received paroxetine 10 mg daily for one week. Participants then received paroxetine 20 mg daily for an additional four weeks.

Twenty-seven women completed the entire study. Three patients discontinued paroxetine therapy because of adverse effects: two because of somnolence and one because of anxiety. Tamoxifen and vitamin E were used by 80% and 73% of the study participants, respectively. The mean reduction in the frequency of hot flashes was 67% (95% CI, 56-79%), and 20 women (67%) reported a more than 50% reduction in the average daily number of hot flashes. Of the 26 women who completed the entire study and severity score questionnaires, the mean reduction in hot flash severity score at week 6 was 75% (95% CI, 66-85%). Further, 22 women (73%) reported a more than 50% reduction in hot flash severity scores, and 25 women (83%) elected to continue paroxetine therapy at the end of the study. Reported adverse effects included somnolence, dry mouth, nausea, dizziness, tremor, anxiety, and sexual dysfunction.

Significant improvements from baseline were reported in depression (p = 0.02) (assessed by the Center for Epidemiologic Studies Depression Scale), sleep (p = 0.0002) (assessed by Medical Outcomes Study 9 item Sleep Scale), anxiety (p = 0.0005) (assessed by Hospital Anxiety and Depression Scale), and quality of life (p = 0.004). Although the authors suggested that paroxetine is an effective therapy for hot flashes in breast cancer survivors, a randomized, placebo-controlled trial to determine the effectiveness of paroxetine in hot flashes is underway. This study was limited by the small sample size, open-label and uncontrolled design, and short duration. Also, the changes in hot flash frequency and hot flash severity scores were depicted graphically, and the exact values for these endpoints were not provided. Thirty patients were needed to provide 80% power to detect a 50% reduction in hot flashes; however, the primary endpoint was analyzed with results for only 27 patients. In addition, the statistical analysis of the primary endpoints was made using historical, placebo-controlled data for comparison.

The efficacy and safety of paroxetine in the treatment of hot flashes in 13 women undergoing treatment for breast cancer were evaluated in an open-label, pilot study.[27] Eligible patients were over 18 years of age, had hot flashes at least moderate in severity (i.e., at least a 3 on a 1-5 scale), had a ECOG performance status rating of <2.5, and were currently undergoing or had received in the previous six months adjuvant chemotherapy for breast cancer. Of the participants, 8 (62%) were receiving chemotherapy, 9 (69%) were taking tamoxifen or anastrozole, and 5 (38%) were premenopausal. Paroxetine therapy was initiated at 10 mg daily for three days and then increased to 20 mg daily. Hot flash severity, the primary endpoint, was assessed one month after therapy initiation using a 5-point Likert scale, where 0 = not at all severe and 4 = extremely severe. At follow-up, mean hot flash severity decreased from 3.62 at baseline to 2.08 (p = 0.002). Although the majority of patients (85%) were still having hot flashes, the percentage of patients rating their hot flash severity as "quite a bit" or "extremely severe" had decreased from 100% at baseline to 38% (p = 0.008). Patients also reported an improvement in sleep quality from a mean of 1.85 at baseline to 0.77 after paroxetine therapy. Significant reductions in general, emotional, mental, and total fatigue and an improvement in vigor were noted. The percentage of women exhibiting clinically significant depressive symptomatology decreased from 75% at baseline to 25% after paroxetine therapy (p = 0.03). There were no reports of adverse reactions requiring dosage adjustment or study discontinuation. The authors concluded that paroxetine was beneficial for women of any menopausal status receiving chemotherapy resulting in hot flashes; however, this conclusion is limited by the small sample, lack of controls, open-label design, and lack of information about the statistical tests used.

A multicenter, randomized, double-blind, placebo-controlled trial evaluated the efficacy and safety of extended-release paroxetine for the treatment of vasomotor symptoms in 165 menopausal women.[29] Eligible patients were at least 18 years of age who had a minimum of two hot flashes per day or 14 hot flashes per week and had stopped hormone therapy for at least six weeks before hot flash symptom assessment (baseline). Before study enrollment, psychotropic agents (i.e., tricyclic antidepressants, selective noradrenaline-reuptake inhibitors, selective serotonin-reuptake inhibitors, lithium and oral neuroleptics, all sedatives and hypnotics, monoamine oxidase inhibitors, and depot neuroleptics) must have been stopped for a specified time period. Concomitant treatment with tamoxifen, raloxifene, and vitamin E was allowed if used regularly for at least three months before screening and continued during the study period. After a one-week placebo run-in phase, patients were randomized to receive extended-release paroxetine 12.5 mg/day (n = 51), extended-release paroxetine 25 mg/day (n = 58), or placebo (n = 56) for six weeks. All patients recorded the frequency and severity of their hot flashes in daily diaries. Adverse events were reported at scheduled study visits. The primary endpoint was the mean change in the daily hot flash composite score from baseline to week 6 for patients receiving extended-release paroxetine 25 mg/day or placebo. Of the 165 participants, 139 completed the study. Tamoxifen, raloxifene, or vitamin E was used by 36 (21.8%) of the patients. A mean reduction in hot flash composite score compared with placebo was reported at week 6 for extended-release paroxetine 12.5 mg/day (-4.7 [95% CI, -8.1 to -1.3]) (p = 0.007) and 25 mg/day (-3.6 [95% CI, -6.8 to -0.4]) (p = 0.03). After six weeks of treatment, mean daily hot flash frequency was reduced in all groups (mean reductions of 3.3, 3.2, and 1.8 for those receiving 12.5 mg/day, 25 mg/day, and placebo, respectively). Adjusted mean differences for hot flash frequency were -1.55 (95% CI, -2.75 to -0.34) (p = 0.01) and -1.50 (95% CI, -2.66 to -0.34) (p = 0.01) when 12.5 and 25 mg/day were compared with placebo, respectively. Adverse events most frequently reported for patients randomized to extended-release paroxetine were mild to moderate headache, nausea, and insomnia. Of the 165 randomized patients, 14 did not complete the study because of adverse events (12.5 mg/day [4 patients], 25 mg/day [8 patients], and placebo [2 patients]). The authors concluded that extended-release paroxetine might be an effective and acceptable alternative in treating menopause-related hot flashes. Although paroxetine reduced hot flash scores, patients still had a mean of three or more hot flashes daily. Study limitations included a high dropout rate, short study duration, lack of power, and decreased enrollment of African-American and Asian women.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.