| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
ORIGINAL RESEARCH |
From the *Center for Reproductive Sciences, Department of Obstetrics, Gynecology, and Reproductive Sciences, and Department of Epidemiology and Biostatistics, University of California, San Francisco, San Francisco, California.
Address reprint requests to: Robert B. Jaffe, MD, Center for Reproductive Sciences, Department of Obstetrics, Gynecology, and Reproductive Sciences, University of California, San Francisco, 505 Parnassus Avenue, San Francisco, CA 94143–0556; e-mail: jaffer{at}obgyn.ucsf.edu.
 |
ABSTRACT |
|---|
TOP
 ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
METHODS: Eighty postmenopausal women with more than 14 hot flushes per week were randomized to receive treatment with extended-release venlafaxine or placebo. Participants received 37.5 mg daily for 1 week, followed by 75 mg daily for 11 weeks. Daily hot flush severity scores and adverse effects were recorded by subjects. Baseline and monthly follow-up questionnaires assessed patient-perceived hot flush score, quality of life, and sexual function. Participants were treated for 12 weeks.
RESULTS: Of the 80 subjects who enrolled in the study, 40 were in the treatment group and 40 in the control group. Of these, 61 completed the study (treatment, n = 29; control, n = 32). Subjective assessment at monthly visits of the effects of hot flush symptoms on daily living were significantly improved in the treatment group (P < .001). Hot flush severity scores based on daily diaries were somewhat lower in the treatment group, but the between-group difference did not reach statistical significance (P = .25). Three side effects, dry mouth, sleeplessness, and decreased appetite, were significantly more frequent in the venlafaxine group, but others, including dizziness, tremors, anxiety, diarrhea, and rash, were significantly less frequent. Ninety-three percent of participants in the venlafaxine group chose to continue treatment at the conclusion of the study.
CONCLUSION: Extended-release venlafaxine, 75 mg per day, is an effective treatment for postmenopausal hot flushes in otherwise healthy women, based on a significant decrease in patient-perceived hot flush score.
LEVEL OF EVIDENCE: I
Traditionally, estrogen has been the most commonly prescribed medication for the treatment of hot flushes, providing greater than 95% efficacy for relief of vasomotor symptoms. After the publication of the results of the estrogen-progestin arm of the Women's Health Initiative study in 2002, in which an increase in breast cancer risk and thromboembolic events was observed in estrogen-progestin versus placebo users, there was a precipitous decrease in the amount of estrogen prescribed. Interestingly, in the recently released estrogen-only arm of the Women's Health Initiative,5 an increased risk of breast cancer was not observed. Cardiovascular events were neither increased nor decreased in the estrogen-only arm.
Many alternative treatments for hot flushes have been investigated, including megestrol acetate,6 clonidine,7 and selective serotonin-reuptake inhibitors (SSRIs).8–11 Proposed physiologic mechanisms for hot flushes include noradrenergic and serotonergic pathways and their potential interaction with the hypothalamic thermoregulatory center.12–14 Based upon the potential involvement of these neurotransmitter pathways, medications for the treatment of hot flushes have been assessed that inhibit serotonergic and noradrenergic pathways.
In the early 1990s, clonidine, an 
-adrenergic agonist, was found to be effective in decreasing hot flushes, but was associated with significant side effects.12 More recently, SSRIs have been investigated, and paroxetine and fluoxetine have proven to reduce hot flushes significantly when compared with placebos.8,10 Venlafaxine, an antidepressant with both serotonergic and adrenergic inhibitory properties, has been tested at multiple doses for the treatment of hot flushes. In a study by Loprinzi et al,15 venlafaxine 75 mg per day reduced hot flushes by 62%, compared with a 37% reduction with placebo.
In previous venlafaxine studies to treat hot flushes, breast cancer patients, or women with a fear of breast cancer who were unwilling to use estrogen, were recruited. Successful reduction of hot flushes in this population was reported.15 In the present study, we sought to assess whether the beneficial effects of venlafaxine would be observed in the general population, rather than in a population primarily comprising breast cancer patients, using an extended release form of the medication for a longer duration (12 weeks) than has previously been reported.8,10,15
|
MATERIALS AND METHODS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The subjects were recruited initially through newspaper and radio advertising, directed mailings, and flyers posted in physician's offices and clinics. Of these, 80 women were qualified to enter the study. Specific inclusion criteria included natural or surgical menopause, and more than 14 hot flushes per week. Participants were excluded if they had known adverse reactions to antidepressant medications or if they were receiving estrogens, progestins, androgens, antidepressants, or chemotherapy.
Each potential subject was interviewed by the study coordinators (K.M., K.S.M.). After inclusion and exclusion criteria were met, participants were randomly assigned to receive either venlafaxine (treatment) or placebo (control) using a permuted block design (4 per block) and an Excel spreadsheet random number generator. To accomplish blinding of the medication, the pharmacy prepared identical-appearing capsules of 37.5 mg extended-release venlafaxine (Effexor XR, Wyeth Pharmaceuticals, Collegeville, PA) and placebo before randomization. Subjects were told to take 1 tablet daily for 1 week and then 2 tablets daily for 11 weeks. Upon completion of the study, the subjects in the treatment group were given the opportunity to continue treatment, and those in the control group were able to start treatment if desired.
After randomization, but before starting medication, each woman was asked to complete a questionnaire that included questions about quality of life, mood, and sexual function. This questionnaire was based on the Short Form-36 Health Survey (SF-36) mood scale, but was slightly modified. An identical questionnaire was filled out on follow-up visits 4, 8, and 12 weeks after initiating treatment. At each visit including baseline, each woman was asked to assess on a 5-point Likert scale how significantly her hot flushes interfered with daily living. This will be referred to as the patient-perceived hot flush score. In previous studies with venlafaxine and SSRIs8,10,15 for the treatment of hot flushes, a 4–6 week study period was used. By extending the length of treatment to 12 weeks, we hoped to identify any change in effectiveness over this duration of treatment.
All subjects were asked to complete a daily hot flush diary, in which they recorded the number of mild, moderate, severe, and very severe hot flushes per 24 hours. This same hot flush diary has been used in several published studies.8,15–17 Scoring was on a scale from 1 to 4, with 1 being mild, 2 being moderate, 3 being severe, and 4 being very severe. Detailed descriptions of severity scoring were available in each diary. Written descriptions of each of these levels of severity were provided to the subjects and were based on previous, validated studies.15 The women were asked to record potential adverse effects in the daily diary, including loss of appetite, dry mouth, sleeplessness, nausea, dizziness, fatigue, constipation, insomnia, and mood changes.
Statistical Analysis
Chi-squared statistics, t tests, and Wilcoxon tests were used to compare baseline measures between women assigned to active medication and women assigned to placebo (Table 1). The Wilcoxon test was used for comparing alcoholic drinks per week on the basis of inspecting a quantile-quantile plot for evidence of departures from normality. Linear models for repeated measures were used to estimate the treatment effect on repeated monthly assessments of the patient-perceived hot flush scores, with the baseline assessment treated as an outcome and random effects used to account for within-subject correlation. Because we detected both a borderline-significant, between-group difference in favor of placebo at baseline and statistically significant heterogeneity in the difference between groups at months 1, 2, and 3, we made 3 assessments of the treatment effect, using linear combinations of the parameter estimates in a saturated model, allowing for interaction between treatment and time: first, as the average of the between-group differences at months 1, 2, and 3, net of (that is, minus) the difference at baseline; second, as the average of the between-group differences at months 2 and 3, net of the difference at baseline; and third, as a linear trend from baseline to month 3 in the 4 between-group differences. Linear contrasts in the estimated model parameters were also used to assess heterogeneity in the treatment effects. These methods were also used for the other quality-of-life measures assessed monthly (Table 2).
|
|
Using estimates of the standard deviation and correlation between repeated measures based on the data, we estimate that the sample of 61 patients with complete data provided 80% power in 2-sided tests with of 5% to detect a net between-group difference of approximately 9 points in the mean patient-perceived hot flush score in the first analysis and 10 points in the second. In the third analysis, the study was powered to detect a divergence in trends of between 3 and 4 points per month. Details of methods and programs for estimating minimum detectable effects are available from the study statistician (E.V.).
The overall treatment effect on week-by-week average hot flush severity scores based on the daily diary also was evaluated using a mixed linear model that controlled for the patient-perceived hot flush score reported at baseline and allowed for an arbitrary mean trajectory in the placebo group, using week-specific intercepts. For inference, the weekly hot flush average scores were log-transformed to meet normality assumptions; however, effect estimates are presented on the original scale, for interpretability. Variability in the between-group differences was again apparent across weeks; accordingly, we estimated the treatment difference as the average of the between-group differences at each week, again using linear combinations of the parameter estimates in a model allowing for the interaction between treatment and week. These methods were also used to examine between-group differences in the frequency of hot flushes as well as the severity of reported episodes.
Using the observed standard deviations and within-person correlations, the sample would have provided 80% power in 2-sided tests with of 5% to detect a between-group difference in the average weekly scores of 3 points, if the treatment effect had been constant over time. After accounting for interaction, the minimum detectable summary effect would in most circumstances be larger.
Finally, summary indicators for adverse effects from the daily diaries, defined as any report of adverse effects with a score greater than 2 (scale from 1–7) on any day of follow-up, were analyzed using 
2 statistics. All analyses were performed with SAS 8.02 (SAS Institute, Cary, NC).
|
RESULTS |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The 2 groups were similar in age, race, age at menopause, years since menopause, proportions with surgical versus natural menopause, and smoking history (Table 1). However, the treatment group had higher average patient-perceived hot flush scores and reported more frequent alcohol consumption. In sensitivity analyses, controlling for alcohol use in the linear mixed models did not affect any of our findings.
Mean values of the patient-perceived hot flush score are shown by treatment group and visit in Figure 1. At study entry the treatment group had a higher average score (72.4 versus 61.5; P = .07). At the month 1 follow-up visit, mean scores in both groups had declined substantially and were nearly the same. By completion of the study, average scores in the venlafaxine group had declined further to 35.3, whereas the average in the control group had rebounded. The average difference between the groups at follow-up, net of the baseline difference in favor of placebo, was 21 points (95% confidence interval [CI] 11–32, P < .001). Because there was heterogeneity in the treatment effect across occasions (P = .01), we also assessed the average difference at the months 2 and 3 visits, again net of the baseline difference. This estimate of the treatment effect was 28 points (95% CI 16–39, P < .001). This corresponds to a 51% reduction in the score in the venlafaxine group, compared with a 15% reduction in placebo. The difference between the groups linearly increased from P < .001, with no compelling evidence (P = .25) in an augmented model for departures from linearity.
|
Hot flush severity scores from the daily diary are shown by treatment group and week in Figure 2A. Over the 12-week study period, there was an overall reduction in hot flush severity score in both the treatment and control groups. We found weak evidence for heterogeneity in the treatment effect across weeks (P = .08 on the original scale, P = .16 on the log-transformed scale). The mixed-model estimate of the average reduction in score in the venlafaxine group, taking account of the heterogeneity, was approximately 2.6 points, (95% CI –2.3 to 7.5), but the average difference was not statistically significant on either the original (P = .30) or log-transformed (P = .25) scale. A similar pattern was observed for the frequency of hot flushes (Fig. 2B). Taking account of heterogeneity across weeks (P = .06 on original scale, P = .09 on log scale), the average reduction in the venlafaxine group was 1.4 episodes (95% CI –0.7 to 3.6, P = .20). The average severity of the reported hot flushes was essentially indistinguishable in the venlafaxine and placebo groups (Fig. 2C). Of the broader quality-of-life measures assessed on the monthly questionnaire, mental health and vitality were significantly improved in the treatment group (both P = .005), but there was weak evidence (P = .11) for an adverse effect on sexual function. For these outcomes, heterogeneity in the treatment effects across monthly visits was not apparent. Current health, physical health, and emotional health were similar in the 2 groups (data not shown).
|
Multiple potential adverse effects were evaluated. Statistically significant increases in dry mouth (81% versus 44%) and sleeplessness (88% versus 47%), as well as decreased appetite (81% versus 53%), were observed in the venlafaxine group, but dizziness (6% versus 56%), tremors (0% versus 33%), anxiety (13% versus 57%), diarrhea (9% versus 44%), and rash (13% versus 47%) were all significantly less frequent. Reports of constipation, nausea, forgetfulness, vision problems, chills, fever, fatigue, headache, decreased libido, and increased appetite were similar in the 2 groups (data not shown).
|
DISCUSSION |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Very importantly, the study indicates that extended-release venlafaxine, at 75 mg per day, can significantly reduce the patient-perceived effects of hot flushes on activities of daily life, compared with placebo. Most patients seek medical attention for their hot flushes because they interfere with daily living. Common presenting complaints related to hot flushes include interference with work, family life, sleep, and sexual function. Based on the data obtained, venlafaxine can be considered an effective treatment option, because a patient's perception of her hot flushes is paramount. The effectiveness of treatment is further supported by the observation that 93% of subjects in the treatment group chose to continue treatment with venlafaxine after completion of the study. However, the estimated difference between the treatment and control groups in average diary-based weekly hot flush severity scores was only 2 points (P = .25). This resulted from a reduction in frequency; there was no apparent reduction in severity of reported episodes. Although the upper confidence limit of 8.3 points for the between-group difference in the severity scores was consistent with a clinically meaningful effect, it may be that the patients' perceptions of the impact of their hot flushes on daily life was significantly improved by the treatment even though the reduction in the more direct measure of hot flush severity is relatively small. This difference may be due to the antidepressant effect of venlafaxine. It is possible that the improvement in the sense of well-being brought about by the venlafaxine altered the women's perceptions of their hot flushes. There are results in the current study that suggest that the antidepressant effect in the treatment group is significant. When comparing the 2 groups, there is a significant difference in the patients' perceptions of mental health (P < .001) and vitality (P 
.001). It is likely that hot flushes lead to interference with daily activities, annoyance, lack of sleep, fatigue, and ultimately, alteration in mood, even resulting in dysphoria or depression. The positive effects of venlafaxine (eg, patient-perceived hot flush score) may be due to mood elevation and/or an increased ability to cope with hot flushes.
Several hypotheses have been suggested to explain the cause of hot flushes. It has been hypothesized that hot flushes are triggered within the hypothalamus through the action of 2-adrenergic receptors on noradrenergic neurons. Freedman et al12,14,18 tested this hypothesis by administering clonidine (an 2-adrenergic agonist) and yohimbine (an 2-adrenergic antagonist) to postmenopausal women with and without hot flushes. Clonidine significantly decreased the number of hot flushes in the symptomatic group and significantly increased the amount of peripheral heating required to evoke a hot flush. Yohimbine had the opposite effect. Clonidine has proven to be a successful treatment for hot flushes but has not been used extensively because of its adverse-effect profile. The above findings, as well as several other published reports,7,12,18 support the role of a central 2-adrenergic mechanism in initiating hot flushes.
Berendsen has proposed a hypothesis for the role of serotonin (5-hydroxytryptamine [5-HT]) in the initiation of hot flushes.13 In this model, estrogen withdrawal leads to a decreased serotonin blood level and an increased 5-HT2A receptor sensitivity in the hypothalamus (involved in thermoregulation). In response to stimuli (eg, stress, caffeine), there is an increased release of 5-HT moduline, a protein. This leads to the blockade of 5-HT1B receptors and a subsequent increased release of 5-HT. The increased serotonin stimulates the 5-HT2A receptors in the hypothalamus (discussed above), changing the set-point temperature. Autonomic reactions are triggered to cool the body, resulting in a hot flush. The use of SSRIs for the treatment of hot flushes is reasonable, based on this hypothesis. Several SSRIs have proven to be effective in clinical trials.8–11
Venlafaxine affects both serotonin and norepinephrine reuptake and therefore addresses both of the potential physiologic mechanisms outlined above. At lower doses, it is primarily involved in serotonin reuptake inhibition (similar to SSRIs). At higher doses, venlafaxine has serotonergic and noradrenergic effects.
Adverse effects, were noted in the treatment group when compared with the control group. These adverse effects, including dry mouth, sleeplessness, and decreased appetite, were significantly more frequent in the venlafaxine group, but others, including dizziness, tremors, anxiety, diarrhea, and rash, were significantly less frequent. We assume that the subjects felt that the benefits of treatment outweighed the discomfort of any adverse effects, because 93% of treated women chose to continue treatment at the conclusion of the study.
There were several limitations of this study. There was no run-in period, and baseline data were not obtained before treatment for hot flush severity scores and daily adverse effects. Further, a larger sample size might have revealed a significant difference in hot flush severity scores.
Thus, this study indicates that extended-release venlafaxine is an effective treatment option for postmenopausal hot flushes. Women perceived that the severity of their hot flushes, particularly their interference with daily activities, is significantly decreased with treatment. Although several adverse effects were noted in the venlafaxine group, most women (93%) continued treatment beyond the conclusion of the study, suggesting that the benefits of treatment outweighed the discomfort of the adverse effects.
|
Footnotes |
|---|
Received June 25, 2004. Received in revised form September 1, 2004. Accepted September 9, 2004.
doi:10.1097/01.AOG.0000147840.06947.46
|
REFERENCES |
|---|
|
TOP
ABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Hannan J. The flushings of the menopause. London (UK): Bailliere, Tindall and Cox; 1927.
3. Thompson B, Hart SA, Durno D. Menopausal age and symptomatology in general practice. J Biosoc Sci 1973;5:71–82.[Medline]
4. Rodstrom K, Bengtsson C, Lissner L, Milsom I, Sundh V, Bjorkelund C. A longitudinal study of the treatment of hot flushes: the population study of women in Gothenburg during a quarter of a century. Menopause 2002;9:156–61.[Medline]
5. Manson JE, Hsia J, Johnson KC, Rossouw JE, Assaf AR, Lasser NL, et al. Estrogen plus progestin and the risk of coronary heart disease. N Engl J Med 2003;349:523–34.
6. Loprinzi CL, Michalak JC, Quella SK, O'Fallon JR, Hatfield AK, Nelimark RA, et al. Megestrol acetate for the prevention of hot flashes. N Engl J Med 1994;331:347–52.
7. Pandya KJ, Raubertas RF, Flynn PJ, Hynes HE, Rosenbluth RJ, Kirshner JJ, et al. Oral clonidine in postmenopausal patients with breast cancer experiencing tamoxifen-induced hot flashes: a University of Rochester Cancer Center Community Clinical Oncology Program study. Ann Intern Med 2000;132:788–93.
8. Loprinzi CL, Sloan JA, Perez EA, Quella SK, Stella PJ, Mailliard JA, et al. Phase III evaluation of fluoxetine for treatment of hot flashes. J Clin Oncol 2002;20:1578–83.
9. Roth AJ, Scher HI. Sertraline relieves hot flashes secondary to medical castration as treatment of advanced prostate cancer. Psychooncology 1998;7:129–32.[Medline]
10. Stearns V, Beebe KL, Iyengar M, Dube E. Paroxetine controlled release in the treatment of menopausal hot flashes: a randomized controlled trial. JAMA 2003;289:2827–34.
11. Weitzner MA, Moncello J, Jacobsen PB, Minton S. A pilot trial of paroxetine for the treatment of hot flashes and associated symptoms in women with breast cancer. J Pain Symptom Manage 2002;23:337–45.[Medline]
12. Freedman RR, Woodward S, Sabharwal SC. Alpha 2-adrenergic mechanism in menopausal hot flushes. Obstet Gynecol 1990;76:573–8.
13. Berendsen HH. The role of serotonin in hot flushes. Maturitas 2000;36:155–64.[Medline]
14. Freedman RR, Krell W. Reduced thermoregulatory null zone in postmenopausal women with hot flashes. Am J Obstet Gynecol 1999;181:66–70.[Medline]
15. Loprinzi CL, Kugler JW, Sloan JA, Mailliard JA, LaVasseur BI, Barton DL, et al. Venlafaxine in management of hot flashes in survivors of breast cancer: a randomised controlled trial. Lancet 2000;356:2059–63.[Medline]
16. Barton D, La VB, Loprinzi C, Novotny P, Wilwerding MB, Sloan J. Venlafaxine for the control of hot flashes: results of a longitudinal continuation study. Oncol Nurs Forum 2002;29:33–40.[Medline]
17. Loprinzi L, Barton DL, Sloan JA, Zahasky KM, Smith de AR, Pruthi S, et al. Pilot evaluation of gabapentin for treating hot flashes. Mayo Clin Proc 2002;77:1159–63.[Medline]
18. Freedman RR, Dinsay R. Clonidine raises the sweating threshold in symptomatic but not in asymptomatic postmenopausal women. Fertil Steril 2000;74:20–3.[Medline]
This article has been cited by other articles:
|
|
 |
|
  D. Grady, B. Cohen, J. Tice, M. Kristof, A. Olyaie, and G. F. Sawaya Ineffectiveness of Sertraline for Treatment of Menopausal Hot Flushes: A Randomized Controlled Trial Obstet. Gynecol., April 1, 2007; 109(4): 823 - 830. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. L. Loprinzi, J. W. Kugler, D. L. Barton, A. C. Dueck, L. K. Tschetter, R. A. Nelimark, E. P. Balcueva, K. N. Burger, P. J. Novotny, M. D. Carlson, et al. Phase III Trial of Gabapentin Alone or in Conjunction With an Antidepressant in the Management of Hot Flashes in Women Who Have Inadequate Control With an Antidepressant Alone: NCCTG N03C5 J. Clin. Oncol., January 20, 2007; 25(3): 308 - 312. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. S. Carpenter, A. M. Storniolo, S. Johns, P. O. Monahan, F. Azzouz, J. L. Elam, C. S. Johnson, and R. C. Shelton Randomized, Double-Blind, Placebo-Controlled Crossover Trials of Venlafaxine for Hot Flashes After Breast Cancer Oncologist, January 1, 2007; 12(1): 124 - 135. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Grady Management of Menopausal Symptoms N. Engl. J. Med., November 30, 2006; 355(22): 2338 - 2347. [Full Text] [PDF]
|
|
|
|
|
|
A. H. Boekhout, J. H. Beijnen, and J. H.M. Schellens Symptoms and treatment in cancer therapy-induced early menopause. Oncologist, June 1, 2006; 11(6): 641 - 654. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. D. Nelson, K. K. Vesco, E. Haney, R. Fu, A. Nedrow, J. Miller, C. Nicolaidis, M. Walker, and L. Humphrey Nonhormonal therapies for menopausal hot flashes: systematic review and meta-analysis. JAMA, May 3, 2006; 295(17): 2057 - 2071. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Tice and D. Grady Alternatives to estrogen for treatment of hot flashes: are they effective and safe? JAMA, May 3, 2006; 295(17): 2076 - 2078. [Full Text] [PDF]
|
|
|
|
 |
|
 W. L. Wolfman HRT and antidepressants Can. Med. Assoc. J., August 2, 2005; 173(3): 237 - 237. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
