Uterine Artery Embolization Treatment ofUterine Fibroids: Effect on Ovarian Function inYounger WomenAdel Ahmad, MD, Laila Qadan, MD, Naheda Hassan, MD, and Kenneth Najarian, MDPURPOSE: Toevaluatehowuterinearteryembolization(UAE)treatmentforuterinefibroids(UF)affectsovarianfunction in young Middle Eastern women.MATERIALS AND METHODS: In this prospective study, 32 patients (mean age, 34 y; range, 2645 y) underwent UAEtreatment of symptomatic fibroids. Serum follicle-stimulating hormone (FSH) levels were measured before and aftertheembolizationtreatment.Preprocedurallevelsweredeterminedontheseconddayofthemenstrualcycle.Post-procedural levels were measured 3 months and 6 months after embolization. A detailed history of menstrual cycleswas obtained before and after UAE.RESULTS:ThirtypremenopausalpatientshadnormalmensesbeforeUAE.MeanFSHlevelsbeforeand3monthsafter UAE were 6.83 IU/L 1.8 and 6.99 IU/L 1.67, respectively (P .66). Normal menstruation resumed 23 monthsafter the procedure. In two perimenopausal women, who had irregular menses and decreased ovarian reserve, meanFSHlevelsincreasedtransientlyfrom22and30IU/Lto40and48IU/L, respectively, 3monthsafterUAE; theydeveloped transient amenorrhea.CONCLUSION: In this study, UAE had no clinically relevant adverse effects on normally functioning ovaries andcould be used safely in the treatment of symptomatic fibroids in premenopausal women. Larger studies are requiredfor further support of this observation.Index terms: Fibroid Ovary, function Uterine arteries, embolization Uterus, neoplasmsJ Vasc Interv Radiol 2002; 13:10171020Abbreviations: FSH follicle-stimulating hormone, UAE uterine artery embolization, UF uterine fibroidsUTERINEfibroids(UF) arethemostcommon type of pelvic tumor and20%40% are present in women aged35 years and older. The traditionaltreatmentforsymptomaticfibroidsishysterectomy (13). Myomectomy hasbeen used in women who wish to pre-serve future childbearing, but hasbeen associated with greater risks (eg,greater loss of blood, longer proce-dure, longer hospital stay, and higherpostoperative morbidity) than hyster-ectomy. Moreover, after myomec-tomy, recurrence of fibroids can occurinas manyas 20%25%of patients(13). During the last 5 years, uterineartery embolization (UAE) hasemerged as a successful alternative tosurgical treatment (46). However,controversy still surrounds this proce-dure because of a number of concerns,includingthe postprocedural riskofovarian failure. The primary objectiveofthisstudywastoaddressthiscon-cern, namely, ovarian dysfunction as aresult of UAE, inresponsetoseveralrecentcasereportsthatindicatedtran-sient or permanent ovarian failure afterthisprocedure(610). Becauseagrad-ual rise in plasma gonadotropin levels istheearliestevidenceofovarianfailure(11,12), follicle stimulating hormone(FSH) was used as the key indicator ofdeclining ovarian function.MATERIAL AND METHODSPatientsBetweenOctober 1997andMarch2001, 32 women (mean age, 34 y;range, 2645 y) underwent UAE treat-ment of symptomatic fibroids. Patientswereincludediftheyhadsymptom-aticfibroids, werepre- orperimeno-pausal, andhadnodesireforfuturepregnancy.Patients were referred after a gyne-From the Departments of Radiology (A.A.) and In-ternal Medicine and Endocrinology (L.Q.), Jabriyah;Maternity Hospital (N.H.), Ministry of Health, Shu-waikh, Kuwait; and Department of Radiology(K.N.), University of Vermont, Burlington, Vermont.Received February 19, 2002; revision requestedApril 8; revisionreceivedandacceptedJune 14.Address correspondence toA.A., Department ofRadiology, Kuwait University, Faculty of Medicine,P.O. Box 24923, Safat 13110, Kuwait; E-mail:adelaaa@hotmail.comNone of the authors has identifieda conflict ofinterest. SIR, 20021017cologist determined that symptomswere mainly relatedto UF andre-quired invasive intervention. In-formed consent was obtained fromeach patient. The study was approvedby the Institutional Review Board forevaluating human research in Kuwait.Baseline workupincludedbimanualpelvic examination (performed by thesame gynecologist), measurement ofthelevelsofFSH, andacombinationof endovaginal and transabdominalultrasonography (US) performed byan experiencedradiologist to deter-mine the number and size of fibroids.Options of intervention (UAE,myomectomy, hysterectomy), goals,andpossiblecomplications wereex-plainedindetail tothepatients. Pa-tients were informedthat long-termeffects of UAEonfertilityhave notbeenestablishedbecauseit isarela-tively new procedure that not enoughpatients who desire future fertilityhaveundergone. Theywereadvisednot toattempt toconceive after theprocedure.Study EndpointsThe primary endpoint of the studywas to determine the effect of UAE onovarian function. The secondary end-pointwastoestablishtheefficacyofUAE for the treatment of UF.ProcedureTheright commonfemoral arterywas accessed with use of the Seldingertechnique. Pelvic arteriography wasperformedonadigital angiographicunit (Advantax LCA; GE Medical Sys-tems, Milwaukee, WI). Selective cath-eterization of the contralateral uterinearterywas performedwithuseof a4-F, C-1 glide catheter (Terumo, Som-erset, NJ). Innine patients, a turboTracker-18microcatheter(BostonSci-entific/Medi-tech, Galway, Ireland)was employed in a coaxial fashion tosecure a more distal position in the leftuterine artery to start embolization.A Waltman loop was formed with a4-F, C-1catheter andusedtoselec-tivelycatheterizetheipsilateral uter-ineartery. In29patients, akinkwasencountered and the catheter waschangedtoa5-F, C-1glidecatheter(Cordis Europa, Roden, TheNether-lands). AWaltmanloopwasformedandusedtocatheterize the anteriordivision of the ipsilateral internal iliacartery. A Tracker-18 microcatheter-seeker combination was used for sub-selective catheterizationandemboli-zation of the ipsilateral uterine artery.Polyvinyl alcohol particles(500700 m; Cook, Bloomington, IN)wereusedforembolizationinallpa-tients. An average of 600 mg (400800mg)wasusedforbilateral emboliza-tion. Theembolizationendpoint wasthe cessation of antegrade blood flowin the uterine artery.Immediately before starting theprocedure, patients were administered1gof intravenousRocephin(Roche,Basel, Switzerland) and a Foley cathe-ter was inserted and not removed un-til 6hoursaftertheprocedure. Con-scious sedation was achieved withintravenous midazolam (25 mg;Roche) and fentanyl (50125 cg; Jans-senPharmaceutica, Beerse, Belgium).Theprocedurewas completedinanaverageof90minutes(range, 45120min)andanaverageof120mL(60150 mL) of Omnipaque (240 mgI/mL;Nycomed, Cork, Ireland), dilutedinsalinesolution, wasused. All proce-dures were performed by the same in-terventional radiologist (A.A.).The procedures were uneventfulexcept for one, inwhichthepatientdeveloped a groin hematoma. She didnot require blood transfusion and thehematoma resolved spontaneously.Follow-upPatients underwent gynecologicand a combination of transabdominaland endovaginal US on follow-up ex-aminations performed 1 and 6 weeks,and3and6monthsafter emboliza-tion. US was performed by an experi-encedbody-imagingtrainedradiolo-gist, whowas blindedtothestudy.Number, size, and volume of fibroidswere recorded and compared withpreprocedural measurements. LevelsofFSHwereobtainedonthesecondday of menstrual cycles 3 and 6months after embolization. Patientswere monitoredfor development ofsymptoms of vasomotor instability(hot flashes), and resumption ofmenstruation.Statistical AnalysisThe formula for a prolate ellipsoid(volume 0.5233 D1 D2 D3),whereD1, D2, andD3arethethreemaximumlongitudinal, anteroposte-rior, and transverse diameters, wasused to calculate the UF volumes (13).Thedatawereanalyzedwithuseofthe SPSS software package (SPSS, Chi-cago, IL). Continuousvariableswereexpressed as means SD. The pairedStudent t test was usedtocomparechanges in serum FSH levels at 3- and6-month follow-up. The Wilcoxonsigned-rank test was used to comparepercentreductioninfibroidvolumes.P values .05 were consideredsignificant.RESULTSThirty-two patients (mean age,34 y; range, 2645 y) were included inthisstudy. Symptomswereexcessivemenstrual bleeding (n 8), pelvicpain or pressure (n 10), or both (n 14). Nine patients were anemic be-cause of menorrhagia. Twopatientshadirregularmenstrualperiodswithvague symptoms of insomnia and de-creased energy. Four patients had un-dergone previous myomectomies withrecurrenceoffibroids. Sevenpatientshad undergone previous cesarean sec-tionsandonehadleftovarianresec-tionbecauseoftorsionofanovariancyst.Patients underwent technically suc-cessful embolizationof bothuterinearteries. All patients had at least 30%reduction in fibroid size. The most sig-nificant differencewasadecreaseinfibroid volume from a median of 283.5cm3(range, 140500cm3)to151cm3(range, 0300 cm3; P .001) at6-month follow-up US.Thirty patients had normal baselineovarian function, including normalmenstrual cycles and mean FSH levels(SD) of 6.8IU/L1.8. At 3- and6-month follow-up, FSH levels inthesepatientswere6.99IU/L1.67(P .66) and 6.7 IU/L 1.18 (P .62),respectively. Normal menstrual cyclesresumed23monthsaftertheproce-dure. Two perimenopausal women,who had irregular menses beforeUAE, had high baseline FSH levels (30IU/Land22IU/L, respectively). At3-month follow-up, FSH levels had in-creasedto48IU/Land40IU/L, re-spectively, andpatientsreportedam-enorrhea and hot flashes. At 6-monthfollow-up, FSH had returned to base-linelevels(26IU/Land27IU/L, re-1018 UAE Treatment of Uterine Fibroids in Younger WomenOctober 2002 JVIRspectively). This indicated transientovarian dysfunction. In these two pa-tients, normal menstrual cycles re-sumed 8 and 10 months after the pro-cedure, respectively.DISCUSSIONUAE is not a new procedure. It hasbeenusedfor treatment of obstetricemergenciesfor morethan20years.UAE has been considered to be a safe,well-tolerated, and effective proce-dure. During this time, it was not re-portedtoadverselyaffect mensesorsubsequent pregnancies (1420).However, with the most recent use ofUAE in treating fibroids, cases of tran-sient and permanent amenorrhea havebeen reported (610). There is concernthat UAE might adversely affect ovar-ian function, an issue that we addressin this study.Thepurposeof thisstudywastoevaluateovarianfunctionbeforeandafter UAE treatment of UF. We had nospecific age criterion for choosing pa-tients,butpatientswereaskednottoattempt toconceive after the proce-dure. Middle-Easternwomenusuallycomplete child-bearing at a young age.Therefore, the usual unintentional biasagainst choosing younger patientswith fibroids did not exist in this studyand most of our patients were be-tween 30 and 35 years of age. All ex-cepttwo, whowereperimenopausal,were considered premenopausal.Perimenopause is an ill-defined in-terval that begins with sporadic abnor-malities or failure of ovulation (mani-festingasintermittentFSHelevation)and progresses to end stage persistentanovulationat a highly variable rate(11). Menstrual irregularities arenotuncommon during this period (11,12).AlthoughabsoluteFSHvaluescouldbemisleadingindistinguishingpre-fromperimenopausal women, it ap-pears that a cutoff point of 24 IU/L isreasonablefromaclinical standpoint(11,12,21). Persistentanovulationandcessation of menstruation for at least 1year defines menopause and is associ-ated with substantially increased FSHlevels, usuallygreater than40IU/L(11,12).Inthisstudy, wefoundthat onlythe two perimenopausal women, whohad proven decreased ovarian reservebefore the procedure (higher FSH andirregularmenses), wereadverselyaf-fected. They experienced transientovarian dysfunction manifested bytransientamenorrheaandatransientmild elevation of FSH level. However,premenopausal patients, who had nor-mal menstrual and hormonal patterns,experienced no change after theprocedure.To explain the postprocedural tran-sient ovarian dysfunction in someperimenopausal women, wehadthefollowing hypotheses: (i) patients whohad higher baseline FSHlevels arelikelytohaveworsebaselineovarianfunctionand, therefore, ahighersen-sitivity to vascular insults; (ii) patientswithfailingovariescouldhavecom-pensatory enlargement of the ovarianbranch of uterine artery, which predis-poses them to a higher number of em-bolization particles deposited at theseovarian branches and a higher risk ofpostprocedural ischemia(thisisonlytheoretical because we have no angio-graphic proof); (iii) the perimeno-pausal phaseisanill-definedperiodclinically and hormonally, with erraticsecretion of estradiol and intermittentworsening of ovarian function andFSH elevation. Could the minimalchange of hormonal levels in some oftheperimenopausalwomenbejustanatural phenomenon, regardless ofUAE?We believe that our study, in agree-ment withtwoother reports(22,23),confirms that younger patients(younger than 40 years) are morelikely to have normal ovarian functionthat wouldnot be compromisedbyUAE. AlthoughChrismanet al (22)attempted to exclude perimenopausalwomen (with use of menstrual historyand FSH criteria), age remained an im-portant factor in determining the endresult after UAE, withwomenolderthan 45 years at higher risk of ovarianfailure after UAE. Therefore, age, men-strualhistory, andFSHlevelsshouldbe consideredinpredicting ovariandysfunction after UAE.WeconcludethatUAEappearstobe a safe procedure and women withnormal ovarian function are not likelyto develop postprocedural ovarianfailure. However, because the numberofourpatients, especiallythosewhowereperimenopausal, wassmall, webelieve that larger multicenter studiesare required to confirmour obser-vations.References1. Reidy JF, Bradley EA. Uterine arteryembolizationforfibroiddisease. Car-diovasc Intervent Radiol 1998; 21:357360.2. SuttonCJG. Treatmentoflargeuter-ine fibroids. Br J Obstet Gynaecol 1996;103:494496.3. Hutchins FZ. Abdominal myomec-tomyasatreatment forsymptomaticuterinefibroids. Obstet Gynecol ClinNorth Am 1995; 22:781789.4. RavinaJH, HerbreteauD, Ciraru-Vig-neronN, etal. Arterialembolizationto treat uterine myomata. Lancet 1995;346:671672.5. Goodwin SC, Vendantham S, McLucasB, etal. Preliminaryexperiencewithuterine artery embolization for uterinefibroids. J Vasc Interv Radiol 1997;8:517526.6. BradleyEA, ReidyJF, FormanRG, etal. Transcatheter uterine artery em-bolization to treat large uterine fi-broids. Br J Obstet Gynaecol 1998; 105:235240.7. Vashisht A, Studd JW, Carey AH, et al.Fibroidembolization: atechniquenotwithout significant complications. Br JObstet Gynaecol 2000; 107:11661170.8. Amato P, Roberts AC. Transientovarian failure: a complication of uter-ine artery embolization. Fertil Steril2001; 75:4384399. GoodwinS, McLucasB, LeeM, etal.Uterine artery embolization for thetreatment of uterine leiomyomata:midtermresults. J VascIntervRadiol1999; 10:11591165.10. SpiesJ, Scialli A, JhaR, etal. Initialresultsfromuterinefibroidemboliza-tionfor symptomatic leiomyomata. JVasc Interv Radiol 1999; 10:11491157.11. Cobin RH, Bledose MB, Futterweit W,et al. AACEmedical guidelines forclinical practice for management ofmenopause. Endocr Pract 1999; 5:354366.12. Goodman NF. Defining the meno-pause: what are the clinically relevantissues? (editorial) Endocr Pract 1998;4:177178.13. Orsini LF, Salardi S, Pilu G, Bovicelli L,CacciariE. Pelvicorgansinpremen-archeal girls: real-time ultrasonogra-phy. Radiology 1984; 153:11311614. Stancato-PasikA, MittyHA, RichardHM, Eshkar N. Obstetric embolo-therapy: effect onmenses andpreg-nancy. Radiology 1997; 204:791793.15. VedanthamS, GoodwinSC, McLucasB, MohrG. Uterinearteryemboliza-tion: an underused method of control-lingpelvichemorrhage. AmJ ObstetGynecol 1997; 176:938948.Ahmad et al 1019Volume 13 Number 1016. PattinsonHA, DunphyBC, WoodS,Saliken J. Cervical pregnancy follow-ing in vitro fertilization: evacuation af-ter uterine artery embolization withsubsequent successful intrauterinepregnancy. Aust N Z J Obstet Gynaecol1994; 34:492493.17. McIvor J, Cameron EW. Pregnancyafter uterine artery embolization tocontrol hemorrhage fromgestationaltrophoblastictumor. BrJRadiol1996;69:624629.18. Kung FT, Chang SY. Efficacy ofmethotroxatetreatment inviableandnonviablecervical pregnancies. AmJObstet Gynecol 1999; 181:14381444.19. Frates MC, Benson CB, Doubilet PM, etal. Cervical ectopic pregnancy: re-sults of conservative treatment. Radiol-ogy 1994; 191:773775.20. PelageJP, SoyerP, RepiquetD, etal.Secondary postpartum hemorrhage:treatment with selective arterial embo-lization. Radiology 1999; 212:385389.21. Stellato RK, Crawford SL, McKinlay SM,Longcope C. Can follicle-stimulatinghormone be used to define menopausalstatus? Endocr Pract 1998; 4:137141.22. Chrisman HB, Saker MB, Ryu RK, et al.Theimpact ofuterinefibroidemboli-zationonresumptionof menses andovarianfunction. JVascIntervRadiol2000; 11:699703.23. Spies J, RothA, Gonsalves S, et al.Ovarianfunctionafter uterine arteryembolizationforleiomyomata: assess-ment with use of serum follicle stimu-lating hormone assay. J Vasc Interv Ra-diol 2001; 12:437442.1020 UAE Treatment of Uterine Fibroids in Younger WomenOctober 2002 JVIR