Introduction

The menopause is almost inevitably associated with the occurrence of hot flashes, but hormone replacement therapy (HRT) using estrogens can usually readily control them. Other hormonal therapy is often used; progestagens (e.g. megesterol) are effective in reducing the frequency of hot flashes, and have the supposed advantage of not exerting an effect on breast cancer morbidity or mortality. More recently, naturally occurring substances containing estrogens have become popular (e.g. soy isoflavones, black cohosh), although many women taking them are unaware that the beneficial effects are related to their estrogen content, with the possible risks that this carries.

Non-hormonal approaches to the treatment of hot flashes include vitamin E, clonidine, belladonna alkaloids, and methyldopa. Success rates with these compounds are very variable, although effectiveness has been shown in randomized clinical studies.

The search for non-hormonal treatment has led to the appraisal of antidepressants drugs in this indication. Both trazadone and paroxetine (serotonin-reuptake inhibitors) have been shown to reduce menopause symptoms. The most recent candidate, venlafaxine, is an antidepressant with activity on both serotonin and noradrenaline uptake; the results of a placebo controlled study are summarized here.

Method

This study, which was coordinated by physicians at Mayo Clinic, USA, enrolled women who had a history of breast cancer, or who were apprehensive of taking estrogen for fear of breast cancer. They had to have hot flashes at least twice a day, present for at least a month. Anti-estrogens (tamoxifen, raloxifene) were allowed, but they should not be taking antineoplastic chemotherapy, other hormones, antidepressants, clonidine, or belladonna alkaloid compounds.

After stratification according to age group (above or below 50), tamoxifen use, duration and frequency of hot flashes, volunteers were randomly assigned to one of 4 groups, and given 37.5 mg, 75 mg, or 150 mg venlafaxine daily, or placebo, in a double-blind fashion. Extended-release venlafaxine was used, and the higher dose levels were achieved by gradual escalation over one or two weeks.

Subjects completed a daily diary hot-flash questionnaire for a baseline week and throughout the 4-week trial period. They also answered quality-of-life questions and completed a Beck depression inventory at the end of each week. Vital signs and side effects were assessed weekly.

Analysis was done of average daily hot-flash activity - the total number, and a score combining the number and severity of hot flashes.

Results

229 women entered the study. After non-completers were eliminated, there were 43, 49, 49 and 50 subjects available for analysis in the 4 treatment groups.

After 4 weeks of treatment, the median decrease in hot-flash scores was significantly greater in all 3 venlafaxine groups, compared to placebo (p<0.0001). While 20% of the placebo group subjects reported a reduction of more than 50% in hot-flash activity, the proportions were 45%, 63% and 55% for the 37.5 mg, 75 mg and 150 mg venlafaxine groups, respectively. (This placebo response is common in such studies.)

Put another way, median hot flash scores were reduced from baseline by 27% in placebo subjects, and in 37%, 61% and 61% for the 3 venlafaxine groups, respectively.

Side effects - principally nausea, dry mouth, anorexia, and constipation - were seen in all the venlafaxine groups in the first week, and persisted in the 75 and 150 mg groups in the subsequent 3 weeks. Libido scores (taken from the Beck inventory) increased in all the groups (venlafaxine and placebo); the scores were higher with venlafaxine, but not significantly so. Depression scores improved in a similar way - again, there were no statistically significant differences between the placebo and venlafaxine groups in this respect.

Overall quality-of-life increased from baseline by an average of 3 points in the 3 venlafaxine groups, and decreased by 3 points in the placebo group (p=0.02).

Comment

The authors conclude from these results that venlafaxine, at a starting dose of 37.5 mg daily, increasing to 75 mg daily if necessary, can relieve the symptoms of menopause within a relatively short time. Side effects are generally tolerable. The magnitude of the response is comparable to that with clonidine, in the same investigators hands. The lack of a marked antidepressant effect of venlafaxine at doses providing relief from hot flashes suggests a different mechanism of action for the two activities.

An accompanying editorial to this article points out that, in selecting breast cancer patients, the investigators have chosen a group especially suited to management by an antidepressant.¹What is now needed, they point out, are efficacy trials in postmenopausal women with no history of breast cancer. If such studies are positive, a new non-hormonal drug for hot flashes will have been added to the physician's armamentarium.