Released August 17, 2005The following news items present reviews of important, recently published scientific articles selected by TheNorth American Menopause Society (NAMS), the leading nonprofit scientific organization dedicated toimproving women’s health and quality of life through an understanding of menopause. Each has a commentaryfrom a recognized expert that addresses the clinical relevance of the item. Note that opinions expressed in thecommentaries are those of the authors and are not necessarily endorsed by NAMS. Published news items may beviewed on the NAMS Web site (www.menopause.org/news.html).Recurrent vasomotor symptomscommon for EPT users in WHIafter study was discontinuedOckene JK, Barad DH, Cochrane BB, et al. Symptomexperience after discontinuing use of estrogen plusprogestin. JAMA 2005;294:183-193. Evidence level: II-2.Postmenopausal women who discontinue estrogenprogestogentherapy (EPT) are significantly morelikely to experience recurrent vasomotor symptomsthan placebo recipients, especially women who hadsymptoms before starting EPT, according to a crosssectionalsurvey of participants in the Women’sHealth Initiative (WHI). Approximately 8 to 12weeks after the WHI trial was stopped, investigatorsmailed a questionnaire to postmenopausal womenwho were still taking the study pills, either placeboor estrogen plus progestin therapy (0.625 mg/dayconjugated equine estrogens plus 2.5 mg/daymedroxyprogesterone acetate), when the trial wasstopped. A total of 8,405 women completed thesurvey, a response rate of 89.9%. Their mean age atstudy discontinuation was 69.1 years. The primaryend points were vasomotor symptoms and pain orstiffness.After discontinuing the study pills, 21.2% of all EPTusers experienced moderate or severe vasomotorsymptoms compared with 4.8% of placebo users.Conversely, 36.7% of the EPT-withdrawal groupand 59.5% of placebo recipients had no vasomotorsymptoms. Among women who were experiencingthese symptoms at baseline (n = 950), the rates forsymptoms after therapy withdrawal were 55.5% and21.3%, respectively. Among women with novasomotor symptoms at baseline, 6.4% hadsymptoms after stopping EPT versus 1.2% ofplacebo recipients.The adjusted odds ratio (OR) showed that thepresence of symptoms at baseline was the mostimportant factor influencing symptom occurrenceafter treatment withdrawal. Overall, 91.1% of theEPT withdrawal group who reported vasomotorsymptoms had experienced them in the past. Amongwomen who had these symptoms at baseline, theadjusted ORs were 5.36 (95% CI, 4.51-6.38) and3.21 (95% CI, 2.90-3.56) for recurrence of vasomotorand pain or stiffness symptoms, respectively,for the EPT group compared with the placebo group.Women whose symptoms had been relieved by EPTalso were likely to suffer symptoms after EPTwithdrawal: OR, 5.82 (95% CI, 4.92-6.89) for vasomotorsymptoms, and OR, 2.16 (95% CI, 1.90-2.40)for pain or stiffness.Nearly half of the women with symptoms aftertreatment withdrawal reported use of at least onemanagement strategy. Lifestyle management strategiesincluded drinking more fluids, exercising, usingfans, changing diet, using layered or cotton clothing,and drinking less caffeine or alcohol. Medicalmanagement strategies included talking to theirclinician, taking vitamin E, using vaginal lubricants,or taking other medications.Comment. This is a very interesting paper, despitelimitations described by the authors. In this study,women discontinuing EPT or placebo in the WHItrial were followed up 8 to 12 months later todetermine their symptom profile and actions theyhad or had not taken.
It is certainly not surprising that more than half ofthe women with vasomotor symptoms atrandomization to active EPT also reported thesesymptoms after discontinuing use of them. This hasalways been a dilemma of clinical practice — weadvise women to take EPT for “the shortest duration,etc.” But if symptoms recur, they may need to goback on treatment, and despite the best intentions,short-term therapy inevitably becomes long-termtherapy. The problem remains as to how to deal withthe situation.To me, one of the most interesting findings is thewithdrawal placebo effect. Women on placeboreported a 21.3% increase in symptoms after placebowas discontinued. We all know that studies ofpostmenopausal therapies for vasomotor symptomsshow a significant placebo response, with 30% to50% of participants reporting reduced symptoms.What has not been reported before is the recurrenceof symptoms with discontinuation of placebo, anarea ripe for further research.The 30% to 50% response with placebo is about thesame level of response as reported in studies of soy,isoflavones, and other herbal products. It is,therefore, not surprising that the authors state, “Theuse of herbal or natural hormones by respondents inthe current study was reported as one of the leasteffective strategies.” This supports recommendingagainst expensive but largely ineffective over-thecounterremedies.The overall lesson from this article appears to be asuggestion that part of the counseling of womenconsidering ET/EPT is to note that symptoms mayonly be delayed, and they have at least a 50-50chance of recurring when such therapy is discontinued.Wulf H. Utian, MD, PhDArthur H. Bill Professor Emeritusof Reproductive BiologyCase Western Reserve University School of MedicineConsultant in Women’s HealthThe Cleveland Clinic FoundationExecutive DirectorThe North American Menopause SocietyCleveland, OHSuboptimal vitamin D levelscommon among women usingosteoporosis therapyHolick MF, Siris ES, Binkley N, et al. Prevalence ofvitamin D inadequacy among postmenopausal NorthAmerican women receiving osteoporosis therapy. J ClinEndocrinol Metab 2005;90:3215-3224. Evidence level: II-2.More than half of North American women receivingosteoporosis therapy have low serum levels ofvitamin D, according to this cohort study. Serumlevels were obtained from 1,536 postmenopausalwomen older than age 55 years (mean age, 71) whohad been receiving osteoporosis therapy for at least 3months. Participants were evenly distributed acrossgeographic regions based on latitude.The mean serum level of 25-hydroxyvitamin D was30.4 ng/mL. Approximately 52% of the women hadsuboptimal vitamin D levels, defined as below30 ng/mL; 36% were below 25 ng/mL, and 18%were below 20 ng/mL. A multivariate analysis foundeight variables associated with low vitamin D levels:older than age 80, nonwhite race, body mass indexgreater than 30 kg/m 2 , use of therapeutic agentsknown to decrease vitamin D levels, lack ofexercise, physician did not discuss importance ofvitamin D supplements with patient, low educationlevel (did not finish high school), and taking a dailyvitamin D supplement with less than 400 IU.Comment. Holick and colleagues provide importantevidence that the public health message on thebenefits of vitamin D has not been deliveredsuccessfully to physicians or patients. More than50% of women already receiving therapy for thetreatment or prevention of osteoporosis did not reachadequate serum 25-hydroxyvitamin D levels of 30ng/mL (75 nmol/L). This is not surprising when 40%of women reported that they consumed less than400 IU of vitamin D per day. In fact, despite beingtreated for osteoporosis, about one-third neverdiscussed the importance of vitamin D in bonehealth with their physicians.Based on recent evidence from two meta-analysesof well-designed randomised controlled trials,consumption of at least 700 to 800 IU/day vitamin Dis needed for optimal prevention of both fracture[Bischoff-Ferrari JAMA 2005] and falls. [Bischoff-Ferrari JAMA 2004] A recent pragmatic trialsuggests that even more vitamin D may be needed2
3for individuals starting at 25-hydroxyvitamin Dlevels below 16 ng/mL (40 nmol/L) [Grant Lancet2005] to reach the target serum 25-hydroxyvitaminD level of 30 ng/mL (75 nmol/L). [Dawson-HughesOsteoporos Int 2005] Every strategy for theprevention or treatment of osteoporosis shouldinclude vitamin D in a dose of at least 700 to 800IU/day. This is especially attractive as vitamin D iswell-tolerated and inexpensive.Heike A. Bischoff-Ferrari, MD, MPHAssistant ProfessorDepartment of Rheumatology andInstitute for Physical MedicineUniversity HospitalZurich, SwitzerlandStudies find no substantial linkbetween low androgen levels andsexual function or androgen levelsand natural menopauseDavis SR, Davison SL, Donath S, Bell RJ. Circulatingandrogen levels and self-reported sexual function inwomen. JAMA 2005;294:91-96. Evidence level: II-2.Low serum androgen levels are not associated withlow levels of female sexual function, according tothis cross-sectional, community-based study fromAustralia. Serum levels of total and free testosterone,androstenedione, and dehydroepiandrosterone sulfate(DHEAS) were obtained from 1,423 randomlyselected women aged 18 to 75 years, 85% of whomwere older than 35. Women were excluded if theywere pregnant, taking psychiatric medications ororal contraceptives, had abnormal thyroid function,or had diagnosed polycystic ovarian syndrome. Selfreportson the Profile of Female Sexual Functionwere used to assess sexual function.A comparative analysis found no clinicallysignificant relationship between low levels of totalor free testosterone or androstenedione and sexualfunction scores. A low score for sexualresponsiveness in women aged 45 years and olderwas significantly associated with the lowest DHEASlevels (odds ratio [OR], 3.90; 95% CI, 1.54-9.81).For women aged 18 to 44 years, the lowest DHEASlevels were significantly associated with low sexualdesire (OR, 3.89; 95% CI, 1.27-11.67), sexualarousal (OR, 6.39; 95% CI, 2.30-17.73), and sexualresponsiveness (OR, 6.59: 95% CI, 2.37-18.34).Davison SL, Bell R, Donath S, Montalto JG, Davis SR.Androgen levels in adult females: changes with age,menopause, and oophorectomy. J Clin Endocrinol Metab2005;90:3847-3853. Evidence level: II-2.An analysis of data from the same study indicatesthat although serum androgen levels decline sharplyin the early postmenopausal year, the decline is not aresult of natural menopause, and the postmenopausalovary appears to continue to produce testosterone.For this analysis, serum levels of total testosterone(T), calculated free T, DHEAS, and androstenedionewere compared from both the study population and areference population of 595 women free of factorsknown to affect androgen levels.Serum levels of all four androgen measures declinedwith age (P
changes and sexual function, except in specificsubgroups of women. Women aged 18 to 44 yearswith the lowest DHEAS levels had lower sexualdesire, sexual arousal, and sexual responsiveness;women aged 45 years and older with the lowestDHEAS levels had lower sexual function.While the “holy grail” of the psychoneuroendocrinologistwould be to prove that a hormone (ie, anandrogen) could direct a behavior, the likelihood ofmaking such a finding in this study seems extremelyremote. Given the provisos listed above (eg,biological, psychological, social, situational) and theextremely complex nature of sexual functioncombined with the primitive nature of ourinstruments for assessing sexual function (ie, TheProfile of Female Sexual Function [PFSF]), it ismore surprising that these investigators found anycorrelations between the hormones and behaviorsthey assessed.The absence of strong associations of hormones withaging or menopause should not be confused with theresults of treatment studies of women with lowsexual desire or other sexual dysfunctions. Severalpublished randomized controlled trials indicate thatexogenous testosterone in both oral and nonoralformulations have a positive effect on sexualfunction, primarily desire, arousal, and orgasmicresponse, in women after spontaneous or surgicallyinduced menopause.The studies under review here should also give theclinician great pause when considering whether tomeasure androgens in a clinical setting as an indexor a marker of sexual dysfunction. Laboratorytesting of testosterone levels, for example, should beused only to monitor for supraphysiologictestosterone concentrations before and/or duringtherapy, not to diagnose testosterone insufficiency orsexual dysfunction.James A. Simon, MDClinical ProfessorGeorge Washington UniversityWashington, DCMedical DirectorThe Women’s Health Research CenterLaurel, MDRates of less severe lupusflares increased by CEE/MPABuyon JP, Petri MA, Kim MY, et al. The effect ofcombined estrogen and progesterone hormone replacementtherapy on disease activity in systemic lupuserythematosus: a randomized trial. Ann Intern Med 2005;142:953-962. Evidence level: I.Combined estrogen and progestin therapy (EPT) isassociated with a modest, but statistically significant,increase in mild to moderate flares in womenwith systemic lupus erythematosus (SLE), accordingto this randomized, double-blind, placebo-controlledtrial. In all, 351 postmenopausal women (mean age,50 years) with inactive or stable-active SLE wereassigned to 12 months of treatment with eitherplacebo or oral EPT (0.625 mg/day conjugatedequine estrogens [CEE] with 5 mg/day medroxyprogesteroneacetate [MPA] added 12 days permonth). The end point was flare rate during the trial.At 12 months, mild to moderate flares weresignificantly increased in the EPT group comparedwith placebo. The rate per person-year was 1.14 forEPT and 0.86 for placebo, yielding a 34% increasedrelative risk for EPT (P = 0.01). However, the rate ofsevere flare was low in both groups: 0.081 for EPTand 0.049 for placebo, a nonsignificant betweengroupdifference. Overall, the probability ofexperiencing any type of flare was significantlyincreased for EPT recipients: 0.64 vs 0.51 forplacebo (P = 0.01).Comment. General dogma suggests that in womenwith SLE, hormone therapy increases the risk offlare. The results of this study can be characterizedas “good news-bad news” — although so-calledsevere flares were not increased, mild to moderateflares were increased.I am concerned that the study presentation tends tominimize the risk to patients. What was described asmoderate flares were clinically significant (pleuritis,pericarditis, need for significant steroid increase,etc). Furthermore, 82% of the women randomized tothis study had inactive disease at baseline, whichleads one to wonder what the results would havebeen if EPT had been given to women with moreactive disease at onset.4Nevertheless, as the authors note, the story remainsessentially the same — estrogen-containing hormone
5therapy can be considered in women with SLE if thegains outweigh the risks.Roland W. Moskowitz, MDProfessor of MedicineCase Western Reserve University School of MedicineUniversity HospitalsCleveland, OHComment. Use of estrogen therapy by women withSLE has been controversial owing to data from bothanimal and human studies on the ability of estrogento either induce or exacerbate lupus. Retrospectiveand prospective studies have found that estrogen useeither has no effect on the rate of lupus flare or itproduces a modest increase in disease frequency.This study by Buyon and colleagues is the largestrandomized, placebo-controlled trial of EPT in SLEpatients to date. It uses rigorous standardization ofpatient assessments to increase accuracy betweencenters when determining the presence of mild,moderate, and severe disease flares. Results indicatethat while EPT increased the frequency of all flares,there was no difference between groups in thenumber of severe flares, the most significantoutcome. If EPT is deemed necessary in postmenopausalwomen with SLE, the data on induction ofsevere flares should not discourage its use. However,clinicians need to be aware that EPT may increasethe risk of mild to moderate flares.Thomas J. Lang MD, PhDAssistant Professor of MedicineUniversity of Maryland School of MedicineBaltimore, MDGenetic traits have substantialeffects on age at menopauseMurabito JM, Yang Q, Fox C, Wilson PWF, Cuples LA.Heritability of age at natural menopause in theFramingham Heart Study. J Clin Endocrinol Metab 2005;90:3427-3430. Evidence level: II-3.More than half of the individual difference in age atmenopause appears to be attributable to inheritedfactors, according to these data from theFramingham Heart Study, a community-based,epidemiologic study. Family correlations regardingage at natural menopause were evaluated in 1,500women from the original cohort and 932 women inan offspring cohort. The study sample included thefollowing pairs: mother-daughter (n = 622), sistersister(n = 474), grandmother-granddaughter(n = 29), aunt-niece (n = 258), and first cousins(n = 165) from 1,296 extended families.The mean age at menopause was 49.1 years for theoriginal cohort and 49.4 years for the offspringcohort. According to results from a multivariableadjustedanalysis (adjusted for body mass index,cigarette smoking, and number of children), geneticfactors accounted for 52% of the variation in age atmenopause for the overall study population(95% CI, 0.35-0.69; P
6Another consideration not noted by the authors isthat genetic studies cannot differentiate geneticsfrom family environmental factors. Lifestyle factorssuch as diet and activity and other householdexposures shared among family members could haveinfluenced the age at menopause.Despite these limitations, the data in this reportsuggest that the mother’s age at menopause is animportant factor determining that for daughters, evenwhen controlling for the effects of body mass index,cigarette smoking, parity, alcohol intake, oralcontraceptive use, and age at menarche. Studies ofother U.S. ethnic groups of women are needed toascertain whether the heritability of age atmenopause is predicted equally well by mother’s ageat menopause. In addition, the influence ofcontemporary environmental factors on age atmenopause should be explored, including trends inobesity, smoking, activity patterns, parity, and oralcontraceptive use.The “Baby Boomers” are likely to differ from theirmothers in each of these factors, and in turn, theirdaughters are likely to reflect secular trends in theirbehavior. Until work with a more diverse populationof mother-daughter and sibling pairs is available,these data provide important evidence that age atmenopause is heritable. Further studies are neededthat address heritability of symptoms during themenopause transition.Nancy Fugate Woods, PhD, RN, FAANDean, School of NursingUniversity of WashingtonSeattle, WALow-dose aspirin provideswomen minimal cancer protectionCook NR, Lee I-M, Gaziano JM, et al. Low-dose aspirinin the primary prevention of cancer: the Women’s HealthStudy: a randomized controlled trial. JAMA 2005;294:47-55. Evidence level: I.Low-dose aspirin does not lower the risk of totalcancer or of breast, colorectal, or other site-specificcancers, according to data from the Women’s HealthStudy. In this randomized, placebo-controlled study,39,876 women aged 45 years and older wereassigned to receive either aspirin (100 mg) orplacebo every other day. Participants did not have ahistory of cancer, cardiovascular disease, or othermajor chronic illness. Follow-up lasted an average of10.1 years.Compared with placebo, aspirin had no effect on therelative risks (RR) for total cancer (RR, 1.01; 95%CI, 0.94-1.08), breast cancer (RR, 0.98; 95% CI,0.87-1.09), colorectal cancer (RR, 0.97; 95% CI,0.77-1.24), or any other site-specific cancer exceptlung cancer (RR, 0.70; 95% CI, 0.50-0.99). Analysesbased on follow-up time or interaction with vitaminE found no differential effects for aspirin.Comment. Animal research and epidemiologicstudies have suggested that aspirin may preventcancer development at various primary sites by 20%to 50%. Two small-scale randomized trials demonstratedno chemopreventive effect from low-doseand high-dose aspirin on colorectal cancer.However, large-scale, randomized controlled trialswere needed to assess the effect of aspirin oncancers of all sites.The study by Cook and colleagues is a wellconducted,large-scale, long-term, randomized,controlled, primary prevention trial. It providesstrong evidence that in healthy women, low-doseaspirin does not reduce the incidence and mortalityof total cancer, breast cancer, colorectal cancer, andother site-specific cancers, except for a trend inreducing lung cancer. These results do notnecessarily refute previous research demonstratingthat a higher dose of aspirin (≥325 mg/day) mayprevent certain cancers. In addition, it cannot beextrapolated that other nonsteroidal antiinflammatoryagents will not be as useful aschemopreventive agents. Because it is possible thatnonsteroidal agents may interfere in early events inthe carcinogenesis process, it is important to notethat participants were not women at high risk.Finally, this study did demonstrate a protectiveeffect of low-dose aspirin on lung cancer, whichmay warrant further research.Ting Bao, MDSenior Medicine ResidentJohns Hopkins Bayview Medical CenterBaltimore, MDVered Stearns, MDAssistant Professor of OncologySidney Kimmel Comprehensive Cancer CenterJohns Hopkins School of MedicineBaltimore, MD
Presentation of POF diagnosis keyto woman’s emotional reactionGroff AA, Covington SN, Halverson LR, et al. Assessingthe emotional needs of women with spontaneouspremature ovarian failure. Fertil Steril 2005;83:1734-1741. Evidence level: II-3.How a clinician presents the diagnosis of prematureovarian failure (POF) can have a significant impacton the woman’s emotional reaction, according to thisobservational study by the National Institutes ofHealth. For this study, structured telephoneinterviews were conducted with 100 womendiagnosed with spontaneous POF (median age atdiagnosis, 28 years; range, 13.5 to 39.0 years). Thediagnosis was based on the woman having at least 4months of amenorrhea before age 40, which wasassociated with two serum FSH levels in thepostmenopausal range sampled at least 1 monthapart. The structured interview used 40 questions,which were designed to determine the manner inwhich they were informed of the POF diagnosis,their emotional response, and the type of emotionalsupport.Overall, 71% were not satisfied with the manner inwhich their physician informed them of thediagnosis, and 89% reported experiencing moderateto severe emotional distress at the time, includinganger, depression, and feeling older, less healthy, orless feminine. A direct correlation was foundbetween their degree of emotional distress and themanner in which they were informed (P = 0.01).Women were significantly more satisfied with theclinician’s presentation if they were emotionallyprepared (P
8Counseling about Early Menopause?There Is a Resource to HelpToday more women are reaching menopause earlier than the typical age– either spontaneously or through medical means, including bilateraloophorectomy or ovarian damage through chemotherapy or pelvicradiation therapy. These women have special needs, regarding not onlymenopause symptom relief and reducing long-term health risks, butalso emotional support due to their loss of fertility earlier than planned.The Early Menopause Guidebook is a comprehensive, affordable,64-page booklet designed especially for this population.Place your order today through the NAMS Web site(www.menopause.org).First to Know ® is a registered trademark ofThe North American Menopause SocietyCopyright © 2005The North American Menopause SocietyAll rights reserved.Post Office Box 94527Cleveland, OH 44101, USATel 440/442-7550Fax firstname.lastname@example.org