Maturitas 47 (2004) 151155Menometrorrhagia in the perimenopause is associatedwith increased serum estradiolM.H. Moena,, H. Kahnb, K.S. Bjervec, T.B. HalvorsendaDepartment of Obstetrics and Gynaecology, St. Olavs Hospital, Trondheim University Hospital, N-7006 Trondheim, NorwaybIla Medical Centre, Trondheim, NorwaycDepartment of Clinical Chemistry, St. Olavs Hospital, Trondheim University Hospital, N-7006 Trondheim, NorwaydDepartment of Pathology, St. Olavs Hospital, Trondheim University Hospital, N-7006 Trondheim, NorwayReceived 16 May 2002; received in revised form 3 February 2003; accepted 11 June 2003AbstractObjective: The purpose of this study was to evaluate the possible association between menometrorrhagia and the level ofendogenous estrogen in perimenopausal women. Methods: A prospective controlled study in which 28 perimenopausal women>40 years presenting with menometrorrhagia were compared with 28 age-matched (2 years) women with normal cyclicalmenstrual periods concerning levels of estradiol and follicle-stimulating hormone (FSH). Neither of the two groups had receivedsexual hormone treatment at least in 2 weeks preceding the hormonal assessment. Results: The serum level estradiol in thepatientswassignicantlyhigherthaninthecontrols(0.55nmol/lversus0.24nmol/l),whereasFSHwasnotsignicantlydifferent between the two groups (21.2 IU/l versus 11.8 IU/l). Twenty of the 28 patients had performed at histologic examinationoftheendometrium,and10ofthese(50%)hadsignsofendometrialhyperplasia.Norelationshipwasfoundbetweentheendometrialhistologyandtheestradiollevel. Conclusions:Anassociationbetweenahighendogenousestradiollevelandmenometrorrhagia in the perimenopause was demonstrated. This may have implications for the choice of treatment in this groupof women. It is proposed that this type of bleeding disturbances should be controlled by progestins only, and not with combinedestrogenprogestin treatment. Suppression of the associated hyperestrogenism could be achieved by use of oral contraceptivesor GnRH agonists. 2003 Elsevier Ireland Ltd. All rights reserved.Keywords: Menometrorrhagia; Serum estradiol; Perimenopause; Endometrium1. IntroductionMenometrorrhagia with irregular as well as heavyuterine bleeding is a common phenomenon in the yearspreceding the menopause. This symptom has a greatCorresponding author. Tel.: +47-73-868746;fax: +47-73-867602.E-mailaddress: mette.moen@medisin.ntnu.no (M.H. Moen).impact upon the quality of a womans life. For manywomenit might bescaring, inconvenient andoftendemands further investigation and treatment that ulti-mately might be hysterectomy. A previous study [1]has showed that 18% of perimenopausal women hadbleeding disturbances indicating histologic evaluationof theendometrium. Themost commonreasonforthesemenstrual irregularitiesisirregularormissingovulation with progesterone deciency. Many physi-0378-5122/$ see front matter 2003 Elsevier Ireland Ltd. All rights reserved.doi:10.1016/S0378-5122(03)00250-0152 M.H.Moenetal./Maturitas 47 (2004) 151155cians, however, usecombinedhormonereplacementtherapy (HRT) for treatment of bleeding disturbancesin women approaching the menopause. This wasclearlydemonstratedinasurveyamongNorwegiangynecologists answering a questionnaire about HRT,disclosing that 73%of the colleagues consideredmetrorrhagiatobeanabsoluteorrelativeindicationfor HRT [2]. An addition of estrogens could aggravatethebleedingdisturbance, whichcouldbeexplainedby exposing the endometrium to excessive estrogen.The aim of this study was to investigate the estrogenlevel in women presenting with menometrorrhagia inthe years before the menopause.2. Material and methods2.1. PatientsTwenty-eight womenmorethan40yearsof agereferred to the general gynecological outpatient clinicwithmenometrorrhagiadenedas heavyandun-predictable bleeding episodeswere consecutivelyincluded. If the referral letter raised any suspicion ofmalignancy, these patients were directed to the sectionof gynecological oncologyandwerethusexcludedfrom the study. Duration of the bleeding disturbancewas from a short period to several months and evenyears. Commonlydifferent treatmentshadbeenof-feredsuchas antibrinolyticagents or sexual hor-mones.Inseveralcaseshaddilatationandcurettagebeenperformedpreviously. Endometrial tissuewasobtainedfrom20patients, mostlybyuseofPipelleaspirationdeviceTM[3] for histological evaluation.The reason for not performing an investigation of theendometrium in all cases was that it had recently beenperformed, or that it was decided initially to try hor-monal treatment, a routine that has been proposed inour country [4]. Use of vaginal ultrasound was often,but not routinely carried out. A blood test for measur-ing estradiol and follicle-stimulating hormone (FSH)was taken. Two of the patients had been treated withhormones (17-estradiol combined with norethis-terone acetate) and this was discontinued at least 14days prior to the endocrine investigation. The half-lifeof estradiol is 1014 h and it is not accumulated in thebody. At latest in a week after stopping treatment itis not expected to nd any exogenous estradiol in thebody(informationfromNovoNordicA/S,producerof the type of HRT used for the two women).2.2. ControlsFor each patient an age-matched (2 years) con-trol was selected. The control group consisted ofwomen consulting a general practitioner (HK) fornon-gynecological complaints. A prerequisite for en-teringthestudywasahistoryof normal menstrualperiods and knowledge of the date when the last men-strual period started. The controls were not allowed touse any kind of sexual hormones. Blood for estradiolandFSHanalyseswastakenfromeachwomaninthe control group at random during the cycles in themorning or early afternoon.2.3. EthicsThe study was approved by the Regional Committeefor Medical Research Ethics. Written informed con-sent was obtained from the controls.2.4. Methods for hormone analyses17-estradiol was measured on an Immulite Anal-yser (Diagnostic Products Corporation, Los Ange-les, CA)usingreagentsandcalibratorssuppliedbythemanufacturer. FSHwasmeasuredonanES600Analyser (Boehringer Mannheim GmbH, Mannheim,Germany) using reagents fromthe manufacturer.Themethodwascalibratedagainst theinternationalstandard IRP 78/549.Thenormal valuesfor estradiol andFSHareasfollowsinthefollicular phase: 0.05Estradiol(nmol/l)0.55 (0.57) 0.24 (0.13) P=0.031FSH (IU/l) 21.2 (23.1) 11.8 (3.7) P>0.05Table 2Age (years), estradiol concentration (nmol/l) and endometrial his-tology of 20 perimenopausal patients with menometrorrhagia (dataare presented according to increasing levels of estradiol)Age Estradiol Endometrial histology49 0.07 Inactive52 0.12 Irregular complex hyperplasia49 0.18 Irregular complex hyperplasia52 0.19 Hormone stimulated46 0.24 Simple hyperplasia48 0.30 Proliferative51 0.30 Simple hyperplasia with secretory signs52 0.32 Inactive43 0.33 Proliferative50 0.33 Proliferative54 0.43 Complex hyperplasia55 0.43 Proliferative46 0.57 Complex hyperplasia45 0.66 Focal hyperplasia48 0.71 Proliferative50 0.78 Simple hyperplasia45 0.90 Proliferative54 1.40 Simple hyperplasia52 1.80 Proliferative45 2.50 Simple hyperplasiathan 1.0 nmol/l. These high values were controlled ina later test, which disclosed estradiol level of the samemagnitude.FSHwas higher among patients than controls(21.2 23.1 IU/l versus 11.8 3.7 IU/l), but this dif-ference did not reach statistical signicance (Table 1).Elevated FSH level (above 26 IU/l) was found in nineof the 28 patients and in four of 28 controls despiteregular menstrual periods (not statistically different).Theendometrium(Table2)from20patientswasevaluatedbythesamepathologist (TBH) andwasfound to be histologically inactive in two cases. Inone case, secretory signs were demonstrated, indicat-ing that this woman had an ovulatory cycle. Prolifer-ation and signs of hormone stimulation was found ineightpatients, whilehyperplasiawasdetectedin10cases (50%). No case of malignancy was found in thestudygroup. InTable2it isdemonstratedthat hy-perplasia was found independently of estradiol levels.The mean estradiol level in the patients with endome-trial hyperplasia was not signicantly different fromthe mean estradiol level in patients without hyperpla-sia (0.72 nmol/l versus 0.54 nmol/l).154 M.H.Moenetal./Maturitas 47 (2004) 1511554. DiscussionThis study disclosed perimenopausal womenwith menometrorrhagia having more than doublethe amount of endogenous estradiol compared toage-matched women with regular menstrual periods.Toour knowledgethis is therst controlledstudydemonstrating a signicantly higher estradiol level inperimenopausal women with menometrorrhagia com-pared to age-matched women with normal menstrualcycles. An association between menstrual irregu-larities andrelative hyperestrogenismhas howeverpreviously been anecdotally reported [5,6].Intheyearsprecedingthemenopause, agradualdecreaseinestrogenlevelhasbeendemonstratedincross-sectional andlongitudinal studies[79]. Lon-gitudinal studies with close monitoring of hormonesinagingwomenhave however discloseda higherlevel of estradiol inolder thaninyounger womenof reproductive age [10]. This phenomenonis ex-plained by decreasing levels of inhibin followed by ariseinFSH. WefoundhigherFSHinpatientswithmenometrorrhagiacomparedwithcontrols, but thisdifference was not statistically signicant. We did nothavepossibilities for measuringinhibins. Probably,there is a great uctuation in estrogen production aswellasalternatingovulatoryandanovulatorycycleswithunpredictableproductionofprogesterone. Thisis believedtobetheexplanationfor bleedingdis-turbances and episodes of ushing experienced by anumber of women before the permanent cessation ofestradiol production takes place.Prolonged estrogen stimulation without proges-tional inuence causes proliferation and hyperpla-siaoftheendometriumwithunpredictablebleedingepisodes whentheendometrial liningbreaks downand is expelled. In our study, no association betweenestradiol level and hyperplasia could be demonstrated.Thiscouldbeexplainedbyaforegoingstimulationof the endometrium by excessive, but now decliningestrogen.Treatment regimes with progestins alone, and estro-gensandprogestinsincombinationarewidelyusedwith little consensus about the optimum treatment isthe statement in a Cochrane Review [11].The optimal treatment should probably be se-quential progestin to counteract the estrogen-inducedendometrial proliferation and hyperplasia. The effec-tiveness of norethisterone or micronized progesteroneused in treating premenopausal bleeding disorders hasbeen shown in a randomized study where all cases ofhyperplasia were reversed after 36 months treatment[12]. In a randomized controlled study, using an oralcontraceptivefortreatingdysfunctional bleedingef-fect was obtained in 80% of cases [13]. Their patientswere however around 30 years of age where it is morenatural to use contraceptive pills than in women closetothemenopause. It shouldbeborninmindthatoral contraceptivesincontrast toHRTsuppresstheendogenous production of estradiol and progesterone.Another optionfor controllingdysfunctional bleed-ing is to combine a gonadotropin-releasing hormoneagonistwithHRT[14]. Inthiswaytheendogenoushormone productionis blockedandreplacedbyabalanced estrogenprogestin combination.The present studywas not designedtoevaluatetreatment strategies for perimenopausal bleeding dis-turbances, but the demonstration of elevated estradiollevels in women with menometrorrhagia indicates thatadding estrogen to progestin treatment is unnecessary.Suppression of the endogenous estrogen production byoral contraceptives or GnRH agonists is treatment op-tion, but side effects and cost make them less suitable.Further research is recommended to establish the roleofdifferenttreatmentmodalitiesinthemanagementof perimenopausal menometrorrhagia.References[1] Seltzer VL, Benjamin F, Deutsch S. Perimenopausal bleedingpatternsandpathologicndings.JAmMedWomensAssoc1990;45:1324.[2] Nilsen ST, Fredriksen T, Iversen OE, Moen MH. Attitude toand practice of Norwegian gynecologists concerning hormonereplacement therapyinclimacteric. TidsskrNorLaegeforen1998;118:29403.[3] CornierE. ThePipelle: adisposabledeviceforendometrialbiopsy. Am J Obstet Gynecol 1984;148:10910.[4] Kirschner R, NesheimBI, Urnes A, LangebrekkeA, SkarOJ. Examination and treatment of gynecological bleedingdisorders. Tidsskr Nor Laegeforen 1992;112:27624.[5] Shideler SE, De Vane GW, Kalra PS, Benirschke K,Lasley BL. Ovarian pituitary hormone interactions during theperimenopause. Maturitas 1989;11:3319.[6] Santoro N, Brown JR, Adel T, Skurnick JH. Characterizationofreproductivehormonaldynamicsintheperimenopause. JClin Endocrinol Metab 1996;81:1495501.M.H.Moenetal./Maturitas 47 (2004) 151155 155[7] Burger HG, Dudley EC, Hopper JL, et al. The endocrinologyof the menopausal transition: a cross-sectional studyof apopulation-basedsample. JClinEndocrinolMetab1995;80:353745.[8] Rannevik G, Jeppsson S, Johnell O, Bjerre B, Laurell-BorulfY, Svanberg. Alongitudinal study of the perimenopausaltransition: altered proles of steroid and pituitary hormones,SHBG and bone mineral density, Maturitas 1995;21:10313.[9] OverlieI, MoenMH, MorkridL, SkjaeraasenJS, HolteA.The endocrine transition around menopausea ve yearsprospective study with proles of gonadotropines, estrogens,androgens andSHBGamonghealthywomen. ActaObstetGynecol Scand 1999;78:6427.[10] Welt CK, McNicholl DJ, Taylor AE. Female reproductiveaging is marked by decreased secretion of dimeric inhibin. JClin Endocrinol Metab 1999;84:10511.[11] Hickey M, Higham, Fraser IS. Progestins versus estrogenand progesterons for irregular uterine bleeding associatedwith anovulation, In: The Cochrane Library, Issue 1. Oxford:Update Software, 2001 (Cochrane Review).[12] Saarikoski S, Yliskoski M, Penttila I. Sequential use ofnorethisterone and natural progesterone in pre-menopausalbleeding disorders. Maturitas 1990;12:8997.[13] Davis A, Godwin A, Lippman J, Olson W, KafrissenM. Triphasic norgestimate-ethinyl estradiol for treatingdysfunctional uterine bleeding. Obstet Gynecol 2000;96:91320.[14] Thomas EJ, Okuda KJ, Thomas NM. The combinationof a depot gonadotrophin releasing hormone agonist andcyclical hormone replacement therapy for dysfunctionaluterine bleeding. Br J Obstet Gynaecol 1991;98:11559.