Management of ovulation induction and intrauterine insemination in infertile patients with hypogonadotropic hypogonadism

Original

Article

Management

of

ovulation

induction

and

intrauterine

insemination

in

infertile

patients

with

hypogonadotropic

hypogonadism

Kiyak

Huseyin

,

Bulut

Berk

,

Karacan

Tolga

*

,

Ozyurek

Eser

,

Gedikbasi

Ali

,

Api

Murat

a_{DepartmentofObstetricsandGynecology,KanuniSultanSuleymanTrainingandResearchHospital,UniversityofHealthSciences,Istanbul,Turkey}

b

DepartmentofObstetricsandGynecology,LivHospital,Istanbul,Turkey

c

DepartmentofObstetricsandGynecology,BagcilarTrainingandResearchHospital,UniversityofHealthSciences,Istanbul,Turkey

d

DepartmentofObstetricsandGynecology,MedipolUniversityHospital,Istanbul,Turkey

ARTICLE INFO

Articlehistory:

Received24December2018

Receivedinrevisedform19March2019

Accepted26March2019

Availableonline29March2019

Keywords: Hypogonadotropichypogonadism Intrauterineinsemination Pregnancyrate Gonadotropinstimulation ABSTRACT

Aim:Toinvestigatetheeffectivenessofovulationinductionandintrauterineinsemination(OI+IUI)in

femalepatientswithhypogonadotropichypogonadism(HH),andtocomparetheoutcomesofdifferent

stimulationprotocolsandcyclecharacteristics.

Materialandmethods:TheoutcomesofOI+IUItreatmentsinpatientswithHHdiagnosedbetween2010

and2018wereretrospectivelyevaluated.Cyclesusingrecombinant(rec)luteinizinghormone(LH)or

human menopausalgonadotropin(hMG) asLHsourceswerecomparedwitheachother.Thecycle

characteristicsandpregnancyratesofthefirstcycleswerecomparedwiththoseofthesecondcyclesin

patientswhounderwent2ormorecycles.

Results:Of104patientsdiagnosedwithWorldHealthOrganizationtype1anovulation,99weretreated

withhMGorrecLH+recfollicle-stimulatinghormone(FSH)inatotalof220cycles.Themeanageofthe

studypatientswas27.8
4.6years(range,19–39years).RecFSH+recLHwasgivenin37cycles,andhMG

wasusedin183cycles.Thehormonevalueswereasfollows:FSH, 1.4
1.6mIU/mL;LH,0.7
1.2mIU/mL;

oestradiol,13(15.8
12.0)pg/mL;andanti-Müllerianhormone,2.1(2.6
1.2)ng/mL.Adominantfollicle

wasobservedin85.7%ofthefirstcyclesandin86.2%ofthesecondcycles.Thetreatmentlasted17.2
5.0

and15.5
3.8daysuntilthehumanchorionicgonadotropin(hCG)administrationdayinthefirstand

second cycles, respectively, and the difference was statistically significant (p<0.05). The cycle

cancellationratewas8.1%(n=3)incyclesdoneusingrecgonadotropinsand29%(n=53)inpatients

stimulatedwithhMG,andthedifferencewasstatisticallysignificant(p<0.05).Thepregnancyrateswere

12.7%and28.3%percycleandperpatient,respectively.ThepregnancyrateinhCG-triggeredpatients

(successfulstimulation)was17.1%percycleinallpatients.

Conclusion:OIwithgonadotropinsandIUIisasafe,efficient,andrelativelycost-effectivetreatment

optioninpatientswithHH,yieldingreasonablepregnancyratespercycleandperpatient.Theuseofrec

FSH+recLHfacilitatescyclemanagementbutdoesnotpositivelycontributetopregnancyratesandis

moreexpensivethansomeotherfeasibleoptions.

©2019ElsevierMassonSAS.Allrightsreserved.

Introduction

Hypogonadotropichypogonadism(HH)isdiagnosedin approx-imately10%ofanovulatorywomen,andthisconditionisclassified asWorldHealthOrganization(WHO)type1anovulation[1].Both hypothalamic amenorrhoea(HA)and idiopathicHH(IHH)have hypothalamiccauses,andtogetherwithhypopituitarism (HP),a

diffusepathologyofthepituitarygland,theseconditionsforma wideclinicalspectrum.

Infertilityduetoanovulationisamajorprobleminwomenof childbearingagewhodesiretoconceive.FSHandLHarerequired for ovulation induction (OI). Gonadotropin-releasing hormone (GnRH)pulsatileinjectionsorpumpsareeffectiveinpatientswith HA andIHH, buttheyarenot usedinthetreatmentof HP.The pulsatile releaseofGnRH providesmonofolliculardevelopment, physiologicoestradiol(E2)levels,andlutealphaseconditionsthat restore normal ovarian function [2,3]. Although gonadotropins provide higher pregnancy rates, their administration is more cumbersomeforpatients,whichdecreasestreatmentcompliance. InpatientswithHP,growthhormone,ACTH,andTSHmightalsobe

* Correspondingauthorat:MerkezMah.MimarSinan Cad.6,Sokak,34100,

Bagcilar,Istanbul,Turkey.

E-mailaddress:tolgakaracan84@gmail.com(K.Tolga).

http://dx.doi.org/10.1016/j.jogoh.2019.03.027

2468-7847/©2019ElsevierMassonSAS.Allrightsreserved.

Available

online

at

ScienceDirect

affectedbesidesFSHandLH.Therefore,allotherhormonesthatare deficientshouldbereplaced,includingFSHandLH[2].

Despitecausingmorenon-physiologichyperstimulation, multi-follicular development, and risks of ovarian hyperstimulation syndromeandmultiplepregnancythanGnRHtherapies,OIwith gonadotropinscanbeusedeffectivelyinallclinicalconditions(HA, IHH,andHP)representingHH[4].InOI,treatmentwithFSHalone issufficienttoachievepregnancyinmostwomenwithinfertility. Women with HH have very low LH levels, inadequate oocyte maturation, inadequate follicular growth, low oestrogen levels, lowendometrialscores,andincreasedamountofFSHuse, thus decreasing their pregnancy rates [2,5]. The addition of LH or preparationscontaining

LH improvesthisproblemandincreasesthepregnancyrates [6].WhentheLHsourceisproblematic,thecounterpartofhuman menopausalgonadotropin(hMG)asanLHactiveagentisrecLH.A combinationofrecLHandrecFSHcanbesafelyandeffectively usedforOIinpatientswithHH[7].

Theaimofthisstudywastoinvestigatethecyclecharacteristics andpregnancyoutcomesinsuccessiveintrauterineinsemination (IUI)withOIcyclesinpatientswithHH,aswellaswithrespectto theLHsource.

Materialandmethods

Thismulticentreretrospectivecohortstudywasconductedin the infertility units of Istanbul University of Health Sciences, KanuniSultanSuleymanTrainingandResearchHospital,Bagcilar TrainingandResearchHospital,andMedipolUniversityHospital Gynecology and Obstetrics Clinic. Institutional review board approval(decisionno.21/02/2018-5)wasobtained.Theoutcomes ofOI+IUItreatmentsinpatientswithHHdiagnosedbetween2010 and2018wereretrospectivelyevaluated.CyclesusingrecLHor hMGasLH sourceswerecomparedwitheachother.Inpatients whounderwent 2 or morecycles, thecycle characteristicsand pregnancyratesofthefirstcycleswerecomparedwiththoseofthe secondcycles.

The criteria for the diagnosis of HA include amenorrhoea, anovulation,andnegativeprogesteronechallengetests,aswellas lowoestrogenand gonadotropin levelsand menstrualbleeding withcombineduseoforalcontraceptives.Inthehormoneprofile between2and4daysofmenstruation,TSHandprolactinshouldbe normal, FSH and LH should be <5 mIU/mL, and E2 should be <10pg/mL.

TheinclusioncriteriaforthisstudywereadiagnosisofWHO type1amenorrhoea,adesireforchildbearing,noothercauseof infertility,anda normal genital anatomy.The exclusioncriteria wereanyothercauseofinfertility,abnormalTSHand/orprolactin levels,and a prediagnosed hypothalamic or pituitary lesion on magnetic resonance imaging. Patients with chronic systemic diseases were also excluded from the study. Patients with HP (inwhomotherdeficienthormonesarereplacedbesidesFSHand LH)wereexcludedbecauseadditivehormonesmighthaveadverse effectsonthepregnancyrates.PatientswithHHreceivedE2+P (progesterone)pretreatmentforatleast3months.

Ovulationinduction

Thetreatmentwasstartedonday2or3ofthefirstmenstrual bleedingaftertheendoftheE2+Ptreatment(Cyclo-Progynova1 1mg,ScheringAG,Berlin, Germany).hMG(Menogon1,Ferring, Brazil)orrecFSH(Gonal-F1,MerckSeronoSA,Kiel,Germany)or purified urinary FSH (Puregon1, MSD, Ireland) with rec LH (Luveris1,MerckSeronoSA,Kiel,Germany)at75IUwasstarted at day or 3, and the durations and doses of treatments were adjusted according to the ovarian response. The starting

gonadotropin doses varied according to the patient’s weight, age,andanticipated responsetotreatment.Folliclegrowthwas monitored using transvaginal ultrasound and serum E2 level measurements.Ovulation, defined as thepresence ofa leading follicle>18mminmeandiameter,wastriggeredbyadministering 5000–10,000 U human chorionic gonadotropin (hCG) (Cho-riomon1;IBSA,Switzerland)or250

m

SpermpreparationandIUI

Semen specimens were collected via masturbation after a minimumof2–3daysofsexualabstinence.Semenanalyseswere performedforliquefactiontime,volume,spermcount, concentra-tion,pH,whitebloodcellcount,andKrugermorphologiccriteria. Semen pellets were reconstituted in 0.4mL human tubal fluid mediumforIUI.

TheIUIsamplewasinjectedusingasimplecatheterwiththe patientinthedorsallithotomyposition.IUIwasperformed36– 40hafterovulationtriggering.AfterIUI,thewomenhadabedrest for20min.

Luteal phase support was used routinely in all patients, commencingthedayafterIUI.Itconsistedofprogesteronevaginal geladministration(Crinone8%;MerckSeronoSA,UK)untilweek7 of pregnancy. Oestrogen was alsoadministered tothe patients eitherintheformofatransdermalpatchorasoraltabletsuntil7 weeksofpregnancy.

Clinicalpregnancywasdefinedasasonographicevidenceofan intrauterinegestationalsac.Alivebirthwasdefinedasthebirthof aninfantafter24weeksofgestationwithpostnatalevidenceof life. Multiple pregnancy was defined as the observation of >1 intrauterinegestationalsac[8].

Statisticalanalysis

StatisticalPackagefortheSocialSciencesversion19wasused for statistical analyses (IBM, Armonk, NY, USA). Thedata were givenasmeans
standarddeviations.Invariableswithanormal distribution,group meanswerecompared withindependent t-tests.Variables witha non-normaldistributionwerecompared using theMann-WhitneyU-test,and categoricalvariables were tested using the chi-square test. Cumulative pregnancy rates accordingtothenumberofinterventionsweredeterminedwith the Kaplan-Meier test. A p-value of <0.05 was accepted as statisticallysignificant.

Results

Distributionofpatients

Of104patientswithadiagnosisofWHOtype1anovulation,99 weretreatedwithhMGorrecLH+recFSHinatotalof220cycles. Treatmentwas deniedin5patients,oneofwhomhadSheehan syndrome.Therewere3patientswithpituitarytumourstreated withsurgery,and1patientwithchronicliverdisease.Themean ageofthestudypatientswas27.8
4.6years(range,19–39years). In the group that received treatment, 2 patients had Kallman syndromeand1patientunderwentsurgeryforcerebral haemor-rhage.Therewere25patientswithsecondaryamenorrhoeaand74 patients with primary amenorrhoea. Of all patients, 9 had secondaryinfertilityandtheremaining90hadprimaryinfertility. TwopatientshadafamilyhistoryofHH(Table1).

RecFSH+recLHwasgivenin37cycles,andhMGwasusedin 183cycles.TheadministeredLHdosewas75IU;theFSHand hMGdosesweregivenaccordingtothegonadotropinresponse

ofthepatient(Table2).In6(2.7%)ofthecycles,thepatients discontinued the treatment because ofprolonged unrespon-siveness(Table2).ThepatientflowchartinFig.1 summarizes the chronological order of the successively used treatment options.

Patientdemographics

On physical examination, infantile external genitalia were found in 6 (6.06%) and uterine hypoplasia was detected in 9 (9.09%).Theovarieswerepolycysticin3(3.03%)ofthepatientsand hadamulticysticappearancein4(4.04%).Theovariescouldnotbe visualized using transvaginal ultrasonography in 92.3% of the patients.

Themeanageofthepatientswas27.8
4.6years.Themean bodymassindexwas24.4kg/m2.Themeanageatmenarchewas 17.5
4.9years,andtheaverageinfertilityperiodwas36months. The E2+P pretreatment period was 12 months. The hormone valueswereasfollows:FSH,1.4
1.6mIU/mL;LH,0.7
1.2mIU/ mL; E2,13 (15.8
12.0)pg/mL;and AMH, 2.1 (2.6
1.2) ng/mL (Table1).

Folliculardevelopment

Adominantfolliclewasobservedin80.1%ofthefirstcyclesand in91.5%ofthesecondcycles.Thetreatmentlastedfor17.2
5.0 and15.5
3.8daysuntilthehCGadministrationdayinthefirstand second cycles, respectively, and thedifference was statistically significant (p<0.05). When the first and second cycles were compared,thecyclecancellationratewasfoundtobe33.3%inthe first cycles and 18.8% in the second cycles, with a statistically significantdifference(p<0.05)(Table3).14patientsunderwent thefourthcycle.Of these,onecycleresultedinpregnancy,four cycleswerecancelled(twohadmorethan3growingfolliclesand twohadnogrowingfollicles)andtwosubsequentlyunderwentthe fifthcycle(whichdidnotconceive).These14patientswerenot includedin thecomparison ofcycleoutcomesduetothesmall sample size, and thus Table 3 was only consistentof the data regardingpatientswhounderwentonetothreecycles.

Follicularsize,folliclecount,andendometrialthickness

The mean number of dominant follicles _>18mm was 1.0 (0.8
1.1),andthemeannumberoffolliclesthatwerebetween14 and18mmwas2.0(2.1
1.6).Themeanendometrialthicknesswas 10mm(9.7
2.3mm).

Cyclecancellation

Thecyclecancellationratewas8.1%(n=3)incyclesperformed usingrecgonadotropinsand29%(n=53)inpatientsstimulated withhMG,withastatisticallysignificantdifference(p<0.05).The reasonsforcancellationwereabsentfolliculardevelopmentin22 cycles(10%),hyperstimulationin28cycles(12.7%),andtreatment discontinuationin6cycles(2.7%).Amongthecyclecancellations, 23(33.3%)occurredinthe_firstcycleand13(18.8%)wereduetoa lackofresponse.

Pregnancyratesandoutcomes

Thepregnancyrateswere12.7%and28.3%percycleandper patient, respectively. The rate of pregnancy in hCG-triggered patients (successful cycle) was 17.1% per cycle. No statistically

Table1

ComparisonofcyclecharacteristicsbetweenhMGandrecFSH+recLH.

Parameters recFSH+recLH(n=37){ hMG(n=183){ p-Value

Age(years) 27.8
5.4(26) 27.8
4.4(27) 0.64 & BMI(kg/m2 ) 24.6
2.9(24) 25.3
4.5(24) 0.58 & E2+P(pretreatment)(months) 16.4
21.9(24) 17.2
19.0(12) 0.21 & FSH(mIU/mL) 2.0
2.2(0.8) 1.3
1.4(0.6) 0.06 & LH(mIU/mL) 1.5
2.4(0.3) 0.5
0.6(0.2) 0.07 & E2(pg/mL) 17.0
11.1(14.6) 15.5
12.2(13) 0.35 &_{TSH(mU/L) 2.3
1.8(1.9) 2.1
3.4(1.5) 0.08} Prolactin(ng/mL) 12.6
10.7(10.5) 16.0
34.7(8.9) 0.84 FSHdose 138.5
39.8(150) 157.4
58.6(150) 0.11 LHdose 94.6
32.9(75) 157.4
58.6(150) 0.000 & hCGdays 16.9
4.6(16) 16.1
4.3(15) 0.33 Endometrialthickness 9.7
2.1(10) 9.7
2.4(10) 0.98 Dominantfollicle>18mm 0.7
0.7(1) 0.9
1.1(1) 0.83 Dominantfollicle14–18mm 1.9
1.2(2) 2.1
1.7(2) 0.82 # Cyclecancellation 3(8.1%) 53(29%) 0.008 # ß-hCG(+) 6(16.2%) 22(12.2%) 0.48

{_{Valuesarepresentedasmean
standarddeviation(median).}

#

Valuesarepresentedasnumber(percentage,%).

&

BMI,bodymassindex;FSH,follicle-stimulatinghormone;LH,luteinizinghormone;hMG,humanmenopausalgonadotropin;E2,oestradiol;P,progesterone;TSH,

thyroid-stimulatinghormone;hCG,humanchorionicgonadotropin.

Table2

Comparisonofcycle1andcycle2characteristicsamongthepatients.

Parameters Cycle1(n=99){ _Cycle2(n=71){ _p-Value

FSHdose 133.5(40.2)(150) 165(57.5)(150) 0.000 LHdose 127.4(43.4)(150) 155.2(62.6)(150) 0.000 hCGdays 17.2(5)(17) 15.5(3.8)(15) 0.010 & Endometrialthickness 9.5(2.5)(10) 99.6(2.4)(10) 0.630 & TPMSC 78.1(78.3)(55.6) 78.7(78.4)(55.6) 0.180 Dominantfollicle>18mm 0.8(1)(1) 0.8(1)(1) 0.752 Dominantfollicle14–18mm 1.8(1.4)(2) 2.1(1.8)(2) 0.279 # Gonadotropin hMG 71(81.8%) 57(88.7%) 0.687 FSH+LH 18(18.2%) 13(18.3%) # Cyclecancellation 33(33.3%) 13(18.8%) 0.021 # Bhcg(+) 10(10.1%) 10(14.3%) 0.426

{_{Valuesarepresentedasmean
standarddeviation(median).}

#

Valuesarepresentedasnumber(percentage,%).

&

BMI,bodymassindex;TPMSC,totalprogressivemotilespermcount;FSH,

follicle-stimulatinghormone;LH,luteinizinghormone;Hmg,humanmenopausal

gonadotropin;BMI,bodymassindex;E2,oestradiol;P,progesterone;TSH,

significantdifferencewasfoundbetweenanyof theparameters whenwomenwhoconceivedwerecomparedwiththosewhodid not.Thepregnancyrateinthesecondcyclewas14.1%compared with10.1%inthefirstcycle;however,thedifferencedidnotreach statistical significance. The pregnancy rate was 12.2% in hMG cyclesand16.2%incycleswithrecLH+FSH,withthedifference being not statistically significant. Even more similarly, the cumulativepregnancyrate inpatientsinwhomovulation could betriggeredwas17.6%intherecFSH+recLHgroupand16.9%in thehMGgroup.Thepregnancyresultedinabortionintwopatients. Twopatientshadatwinpregnancy,andtherewereonlytwotriplet pregnanciesoverall.Oneofthetripletpregnanciesspontaneously reduced,whereastheotherwassubjectedtomultifoetal pregnan-cyreduction(MFPR)andcontinuedasatwinpregnancy(Fig.2).

ThemeanFSHdoseusedinthetreatmentwithgonadotropin was150IU,andtheLHdosewas150IU.Themeanperioduntilthe hCGdaywas15.5days,andthemeanendometrialthicknesson hCGdaywas10.0mm.Thedoseofgonadotropinwasincreasedin 66.4%ofthepatientsduringthetreatment.Lutealphasesupport consistedofE2+Pin60.5%ofthepatientsandprogesteronealone in 39.5%. The regimens used in ovarianhyperstimulation were compared.TheLHdosewas75IUintherecombinantgroupand 150 IU in the hMG group, and the difference was significant (p<0.05).

Inpatientswhounderwent2cycles,thecyclecharacteristics andoutcomesofthefirstcycleswerecomparedwiththoseofthe secondcycles.ThetotalFSHandLHdosesusedweresignificantly higherinthesecondcycles(p<0.05).Themeannumberofdays untilhCGadministrationwas15.5daysinthesecondcyclesand 17.2daysinthefirstcycles.Thecancellationratewas33.3%inthe first cycles and 18.8% in the second cycles, with a significant difference(p<0.05).

Whenpatientswithcancellationofcycleswerecomparedwith thosewithoutanycyclecancellation,themeanagewaslowerin the cancellation groupand the E2+P pretreatment period was shorter.ItwasalsoobservedthatthebasalFSH,LH,andoestrogen levelsattheearlyfollicularphasewerelowerinthegroupwith cycle cancellations (p<0.05). In this group, the endometrial thicknessonhCGdaywasalsoless.

Ovarian hyperstimulation developed in 1 patient. Adverse effectssuchasconstipation,nausea,abdominalpain,andheadache developed in 25 patients(25.2%). None of thepatients discon-tinuedtreatmentbecauseofadverseeffects.

Discussion

HHisarare,heterogeneousspectrumofdisorders character-ized by gonadotropin deficiency, hypo-oestrogenism, amenor-rhoea,andinfertilityassociatedwithanovulation,mostofwhich developasaresultofdefectsinGnRHrelease.Theconditionmay becongenital,asinIHH,orcandeveloplaterinlife,asinHPandHA

Fig.1.Treatmentflowchartforthepatientcohort.

Table3

CumulativefolliculardevelopmentratesandpregnancyratesinpatientstreatedwithconcomitantfollitropinalphaandluteotropinalphaandhMG.

Cycle1(N=99) Cycle2(N=71) Cycle3(N=34)

hMG(n) recLH+rec

FSH(n)

Total%(n/N) hMG(n) recLH+rec

FSH(n)

Total%(n/N) hMG(n) recLH+rec

FSH(n)

Total%(n/N)

Cumulativefolliculardevelopmentrate*,α,β 65 15 80.8%(80/99) 52 13 91.5%(65/71) 27 6 97%(33/34)

Cumulativefolliculardevelopmentrate,

patientswhoreceivedhCG*,α,β

53 13 66.6%(66/99) 45 13 80.2%(57/71) 22 6 82.3%(28/34)

Cumulativepregnancyrate*,α,β 9 1 10.1%(10/99) 7 3 14%(10/71) 5 2 20.5%(7/34)

Cumulativepregnancyrate,patientswho

receivedhCG*,α,β

9 1 14.7%(10/68) 7 3 17.5%(10/57) 5 2 25%(7/28)

Cumulativeon-goingpregnancyrate*,α,β 8 1 9.1%(9/99) 6 3 12.6%(9/71) 5 2 20.5%(7/34)

Cyclecancellationm,α,β 30 3 33.3%(33/99) 13 0 18.3%(13/71) 6 0 17.6%(6/34)

Rec,recombinant;hCG,humanchorionicgonadotropin;FSH,follicle-stimulatinghormone;LH,luteinizinghormone;Hmg,humanmenopausalgonadotropin.

Datagivenas%(n/N).

*

Cycle1vs.2:Nosignificantdifferenceamongthegroupsforthenotedvariable.

m_{Cycle1vs.2:Asignificantdifferenceamongthegroupsforthenotedvariable(p<0.05).}

α_{Cycle1vs.3:Nosignificantdifferenceamongthegroupsforthenotedvariable.}

β_{Cycle2vs.3:Nosignificantdifferenceamongthegroupsforthenotedvariable.}

[2].Regardlessoftheunderlyingaetiology,bothFSHandLHare neededtorestorenormalovarianfunctionandpreventfollicular growtharrest[9,10].

Pulsatile GnRH treatment is an effective and appropriate methodtoachievenormal ovarianfunction, withadvantagesof inducing mono-follicular development, physiologic oestrogen levels, and normal luteal phase function; however, it is not effectiveinHPandmostpatientsfinditdifficulttocarryapump continuously[2,7]. Daily injectionof gonadotropins is a better tolerated and more suitable treatment option for OI [4]. This therapeuticmodalityisassociatedwithincreasedrisksofmultiple folliculargrowthandovarianhyperstimulationsyndrome,owing tothesupra-physiologicfollicularstimulus[11].

AlthoughFSH is theprimary hormone for optimalfollicular development,LHis alsorequiredfor thefullmaturationof the follicleandthefertilizationcapacityoftheoocyte.In embryogen-esis,meioticdivision,whichstopsatmeiosisI,resumesafterLH activity.Anotherprocessthatis inducedbyLH isfollicularwall proteolysis,whichleadstofollicleruptureandoocyterelease[12]. WhenonlyFSHisusedforOI,itresultsinasmallerincreasein oestrogenconcentration,poorendometrialscores,poorfollicular luteinization,anddecreasedoocytefertilizationrates.Asaresultof the reduced concentration of LH, ovarian androgen production decreases, leading to diminished ovarian oestrogen synthesis, decreasedimplantationrates,andincreasedmiscarriagerates[13]. In a recLH dose study, in theOI of patientswith HH, the recommendedoptimalLHdosewas75IU.WhentheLHdosewas 75IU,therewas88%folliculargrowth,whichincreasedto100%at 225IU;however,thefertilizationratedecreasedwhen225IULH wasused(LHceilingtheory)[14].

InwomenwithHH,theovarianvolume,antralfolliclecount, andAMHlevelsdonotpredicttreatmentresponse[15,16].AMH levelsarelowerinthesewomenthaninnormalpatients.Thelevel of AMH increases after gonadotropin administration [16]. The absenceof testsandfindingstodetectovarianreservemakesit difficulttomanagetheOIofpatientswithHH. Comparedwith patientswho haveothercauses of infertility, patientswithHH experiencemorecyclecancellationsbecauseofinsufficientovarian responseandhyperstimulationsyndrome[17].

Thisretrospective multicentre cohort study includeda high amountofdataforthisuncommonpatientgroup.Inthestudy,220 OI+IUIcycleswereperformedin99patientswithHH.Patientshad anaverageof2.26OI+IUIcycles,rangingfromonetofivecycles.In this study, the response to therapy, cycle characteristics, and pregnancyrateswithdifferentLHsources(hMGandrecFSH+rec LH) were investigated. In addition, the characteristics of the patientswithatleasttwoOI+IUIcycleswerecompared.Ofthe patients, 4 had thyroid surgery and two received medical treatmentforadrenalinsufficiency.

Inourstudygroup,theoverallresponseratetogonadotropin treatmentwas66.6%(66/99)inthe_firstcycles.Inthestudyby Dubourdieuet al.,the ovarianresponserate toGnRHpulsatile treatmentwas73%andthattogonadotropintreatmentwas60%. Ontheotherhand,theclinicalpregnancyrateswere45%and15% with pulsatile treatment and gonadotropin treatment, respec-tively[18].

In manystudies, a 74–87% follicular growthrate and a 22% pregnancyratewerereportedwhenFSHandLHwereused[11]. Thepregnancyratepercyclewas12.7%(28patients),andthemean numberofOI+IUIcyclesperpatientwas2.2;28.3%pregnancyrate wasachieved.Inourdata,theFSHorhMGdosewaschangedin 66.4%of thecycles duringtheOIofthepatients.Shohametal. reporteda dosechange of 33.3% [10]. This is indicative of the difficultmanagementofOI.

In60.5%ofthecycles,lutealphasesupportwasprovidedwith E2+P and with progesterone in in 39.5%. In their in vitro

fertilization study conductedin patients withHH, Mumusoglu et al. reported that progesterone was sufficient for the luteal phase[19].

OneofthemostimportantfactorsaffectingOImanagementin patientswithHHiscyclecancellation.Fromouroveralldata,hCG couldnotbegivenin25.4%(n=56)ofthecycles,andthesecycle werecancelled.Thereasonsforcyclecancellationwerefailureto developanyfolliclesin10.0%(n=22)ofthecycles,presenceof>3 dominantfolliclesduetoexcessiveresponsetothestimulationin 12.7%(n=28),andfailuretoadheretothetreatmentbecauseof prolongedfollicularphaseandunresponsivenesstothe stimula-tionin2.7%(n=6).Cycliccancellationswereseenin33.3%inthe firstcycleofOIandin18.8%inthesecondcycle,andthedifference wasstatisticallysignificant(p<0.005).

WhenevaluatedaccordingtotheLHsourceused,thefrequency ofcancellationwas29.0%inthehMGgroupand8.1%intherec FSH+recLHgroup,whichwasstatisticallysignificantlydifferent (p<0.005).RecFSH+recLHfacilitatesmanagementandincreases thehCGdeliveryrate.

Toourknowledge,thereareonly2literaturereportscomparing highlypurifiedhMGandrecombinanttreatment(FSHandLH)in HHcases.Caroneetal.reportedthattheovulationoutcomewas betterwithhighlypurifiedhMG thanwithrecFSHand recLH, whereas the pregnancy rates were better with the combined recombinant treatment. Papaleo et al. only reported a more favourable pregnancy rate with the combined recombinant treatmentoption.Inourstudy,asuccessfulfolliculardevelopment was moreprobablewiththecombinedrecombinanttreatment; however, the pregnancy rate could not reach a significantly improvedlevelwiththerecombinanttreatmentmodality[20,21]. No significant difference was found in the basic and cycle characteristicswithrespecttoconception or thecycleorder in successivecycles.Theper-cyclepregnancyratewas12.7%(n=28), andtheper-patientpregnancyratewas28.3%.Thepregnancyrates were10.1%inthefirstcycleand14.1%inthesecondcycle,andthe differencewasnotstatisticallysignificant.Accordingtoourdata, thecyclesthat usedrecFSH+recLHas theLH sourcecouldbe managed more easily and yielded a higher rate of cycles appropriate for hCG triggering. However, these advantages did notinfluencethepregnancyrates.

Among our patients, 3 women had twinpregnancies and 2 women had triplet pregnancies. One of thetriplet pregnancies spontaneouslyreducedtoatwinpregnancy,andtheotherhada selectiveMFPR.Therate ofmultiplepregnancywas 14.2%.Two (7.1%) pregnancies resulted in abortion. The relatively high miscarriageratesreportedinthestudybyBalenetal.werenot observed inourstudy [4].The rate ofon-going pregnancywas 11.8%percycleand26.2%perpatient.

In this study, the rate of cycles that were appropriate for triggeringwithhCGwas66.7%and81.2%inthefirstandsecond cycles,respectively.Themeancycleduration(
_SD)inthesecond cycleswasalsosignificantlyshorterthaninthefirstcycles(15.5
3.8daysvs.17.2
5days,p<0.05).Furthermore,thepregnancy rateswerehigherinthesecondcycle.Therelativelybettercourse ofthesecondcyclescanbeattributedtothestimulanteffectofthe gonadotropins administered in the first cycles, acting as a pretreatment. In 2.7% of the cycles, the patients discontinued thetreatmentbecausetherewerenodominantfollicles,and18.2% ofthepatients(n=18)electednottocontinuewithasecondcycle ofOI+IUI.Althoughearlierstudiessuggestedthattreatmentwith OI+IUIwasa feasiblealternative,ourstudyrevealedthatabout 45%(45of99)ofthepatientsbeingtreatedconventionallydropped outofourprogramfollowingthesecondcycle.

In patients with HH, several studies have reported more successful cycle management and results with gonadotropin pretreatment.PretreatmentwithrecLHhasbeenshowntoincrease

thefollicleretrievalandpregnancyrates[13].Similarly,theresultsof cyclesweremoresuccessfulinpatientswhounderwent pretreat-mentwith hMG[16]. On theother hand, short-termoestrogen pretreatmenthasalimitedeffectontheuterinesize[22].Women withabsoluteGnRHdeficienciestendtohavelonger follicularphases becausethepituitary glandneeds tobeprimedfor severaldays beforeactivesecretionofFSHandLH[2].

Otherresearchershavesoughttoincreasethesuccessrateof stimulation.Awwadet al. showed thatovariansteriodogenesis, folliculogenesis,endometrialthickening,andfollicular luteiniza-tionafterhCGwereimprovedwhendailydosageofrecLH was administeredin4equalpartsduringthestimulation[9].

Thelimitationsofthisstudyincludetherecruitmentofpatients frommultiplecentres,theapplicationof treatmentbymultiple physicians,and theretrospectivestudydesign. Experience with patients with HH is gained over many years. Accordingly, physiciansare oftenunable tocomplete thelearning curve for thispatientgroup.

Inconclusion,OIwithgonadotropinsandIUIisasafe,efficient, andrelativelycost-effectivetreatmentoptionin theHHpatient group,andreasonablepregnancyratespercycleandperpatient canbeobtained.RecFSH+recLHfacilitatescyclemanagementbut doesnot positivelycontribute to pregnancy rates and is more expensivethansomeotherfeasibleoptions.Thefindingthatthe secondcycleshad improvedratesof successfulstimulationand pregnancyratescomparedwiththefirstcycles isin agreement with the concept of pretreatment with gonadotropins before initiatingthedefinitetreatment.However,approximatelyhalfof all patients discontinue the treatment and seek alternative treatments.Thishighcyclecancellationrateseemstobethemost obviousreason for discontinuation. Finally, we believe that, to improvepatientcompliancewhilemaintainingpregnancysuccess, thereisaneedforstudiesinvestigatingtreatmentsthatcanreduce cyclecancellations.

References

[1]DeSouzaMJ,MillerBE,LoucksAB,LucianoAA,PescatelloLS,CampbellCG, etal.Highfrequencyoflutealphasedeficiencyandanovulationinrecreational womenrunners:bluntedelevationinfollicle-stimulatinghormoneobserved duringluteal-folliculartransition.JClinEndocrinolMetab1998;83:4220–32, doi:http://dx.doi.org/10.1210/jcem.83.12.5334.

[2]YasminE,DaviesM,ConwayG,BalenAH.BritishFertilitySociety.’Ovulation inductioninWHOType1anovulation:Guidelinesforpractice’.Producedon behalfoftheBFSPolicyandPracticeCommittee.HumFertil(Camb) 2013;16:228–34,doi:http://dx.doi.org/10.3109/14647273.2013.829673. [3]LetterieGS,CoddingtonCC,CollinsRL,MerriamGR.Ovulationinductionusing

s.c.Pulsatilegonadotrophin-releasinghormone:effectivenessofdifferent pulsefrequencies.HumReprod1996;11:19–22,doi:http://dx.doi.org/10.1093/ oxfordjournals.humrep.a019015.

[4]BalenAH,BraatDD,WestC,PatelA,JacobsHS.Cumulativeconceptionandlive birthratesafterthetreatmentofanovulatoryinfertility:safetyandefficacyof ovulationinductionin200patients.HumReprod1994;9:1563–70,doi:http:// dx.doi.org/10.1093/oxfordjournals.humrep.a138750.

[5]CouzinetB,LestratN,BraillyS,ForestM,SchaisonG.Stimulationofovarian follicularmaturationwithpurefollicle-stimulatinghormoneinwomenwith gonadotropindeficiency.JClinEndocrinolMetab1988;66:552–6,doi:http:// dx.doi.org/10.1210/jcem-66-3-552.

[6]ShohamZ,SmithH,YekoT,O’BrienF,HemseyG,O’DeaL.RecombinantLH (lutropinalfa)forthetreatmentofhypogonadotrophicwomenwithprofound LHdeficiency:arandomized,double-blind,placebo-controlled, proof-of-efficacystudy.ClinEndocrinol(Oxf)2008;69:471–8,doi:http://dx.doi.org/ 10.1111/j.1365-2265.2008.03299.x.

[7]Kaufmann R, DunnR, VaughnT, Hughes G, O’BrienF, HemseyG, etal. Recombinanthumanluteinizinghormone,lutropinalfa,fortheinductionof folliculardevelopmentandpregnancyinprofoundlygonadotrophin-deficient women.ClinEndocrinol(Oxf)2007;67:563–9.

[8]GandhiAR,CarvalhoLF,NutterB,FalconeT.Determiningthefertilitybenefitof controlledovarianhyperstimulationwithintrauterineinseminationafter operativelaparoscopyinpatientswithendometriosis.JMinimInvasive Gynecol2014;21:101–8,doi:http://dx.doi.org/10.1016/j.jmig.2013.07.009. [9]AwwadJT,FarraC,MitriF,AbdallahMA,JaoudehMA,GhazeeriG.Splitdaily

recombinanthumanLHdoseinhypogonadotrophichypogonadism:a nonrandomizedcontrolledpilotstudy.ReprodBiomedOnline2013;26:88–92, doi:http://dx.doi.org/10.1016/j.rbmo.2012.09.016.

[10]ShohamZ,BalenA,PatelA,JacobsHS.Resultsofovulationinductionusing humanmenopausalgonadotropinorpurifiedfollicle-stimulatinghormonein hypogonadotropichypogonadismpatients.FertilSteril1991;56:1048–53,doi: http://dx.doi.org/10.1016/S0015-0282(16)54715-3.

[11]KrauseB,OhlingerR,HaaseA.Lutropinalpha,recombinanthumanluteinizing hormone,forthestimulationoffolliculardevelopmentinprofoundly LH-deficienthypogonadotropichypogonadalwomen:areview.Biologics 2009;3:337–47,doi:http://dx.doi.org/10.2147/BTT.S3307.

[12]CaroneD,CaropresoC,VittiA,ChiappettaR.Efficacyofdifferentgonadotropin combinationstosupportovulationinductioninWHOtypeIanovulation infertility:clinicalevidencesofhumanrecombinantFSH/humanrecombinant LHina2:1ratioandhighlypurifiedhumanmenopausalgonadotropin stimulationprotocols.JEndocrinolInvest2012;35:996–1002,doi:http://dx. doi.org/10.3275/8657.

[13]BalaschJ,FábreguesF,CarmonaF,CasamitjanaR,Tena-SempereM.Ovarian luteinizinghormoneprimingprecedingfollicle-stimulatinghormone stimulation:clinicalandendocrineeffectsinwomenwithlong-term hypogonadotropichypogonadism.JClinEndocrinolMetab2009;94:2367–73, doi:http://dx.doi.org/10.1210/jc.2009-0262.

[14]TheEuropeanRecombinantHumanLHStudyGroup.Recombinanthuman luteinizinghormone(LH)tosupportrecombinanthumanfollicle-stimulating hormone(FSH)-InducedfolliculardevelopmentinLH-andFSH-Deficient anovulatorywomen:adose-findingstudy.JClinEndocrinolMetab 1998;83:1507–14,doi:http://dx.doi.org/10.1210/jcem.83.5.4770.

[15]Bry-GauillardH,Larrat-LedouxF,LevaillantJM,MassinN,MaioneL,BeauI, etal.Anti-mullerianhormoneandovarianmorphologyinwomenwith isolatedhypogonadotropicHypogonadism/Kallmannsyndrome:effectsof recombinanthumanFSH.JClinEndocrinolMetab2017;102:1102–11,doi: http://dx.doi.org/10.1210/jc.2016-3799.

[16]ChanC,LiuK.Clinicalpregnancyinawomanwithidiopathichypogonadotropic hypogonadismand low AMH:utilityofovarianreserve markersinIHH.J Assist Reprod Genet2014;31:1317–21,doi:http://dx.doi.org/10.1007/s10815-014-0312-2. [17]MilsomS,DugganK,O’sullivanS,OgilvieM,GunnAJ.Treatmentofinfertility

withhypogonadotropichypogonadism:10-yearexperienceinAuckland,New Zealand.AustNZJObstetGynaecol2012;52:293–8,doi:http://dx.doi.org/ 10.1111/j.1479-828X.2012.01450.x.

[18]DubourdieuS, Fréour T, Dessolle L, BarrièreP. Prospective, randomized comparisonbetweenpulsatileGnRHtherapyandcombinedgonadotropin (FSH+LH)treatmentforovulationinductioninwomenwithhypothalamic amenorrheaandunderlyingpolycysticovarysyndrome.EurJObstetGynecol ReprodBiol2013;168:45–8,doi:http://dx.doi.org/10.1016/j.

ejogrb.2012.12.016.

[19]MumusogluS,AtaB,TuranV,DemirB,KahyaogluI,AslanK,etal.Does pituitarysuppressionaffectlivebirthrateinwomenwithcongenital hypogonadotrophichypogonadismundergoingintra-cytoplasmicsperm injection?Amulticentercohortstudy.GynecolEndocrinol2017;33:728–32, doi:http://dx.doi.org/10.1080/09513590.2017.1318278.

[20]CaroneD,CaropresoC,VittiA,ChiappettaR.Efficacyofdifferentgonadotropin combinationstosupportovulationinductioninWHOtypeIanovulation infertility:clinicalevidencesofhumanrecombinantFSH/humanrecombinant LHina2:1ratioandhighlypurifiedhumanmenopausalgonadotropin stimulationprotocols.JEndocrinolInvest2012;35:996–1002,doi:http://dx. doi.org/10.3275/8657.

[21]PapaleoE,AlviggiC,ColomboGL,PisanelliC,RipellinoC,LongobardiS,etal. Cost-effectivenessanalysisontheuseofrFSH+rLHforthetreatmentof anovulationinhypogonadotropichypogonadalwomen.TherClinRiskMang 2014;10:479–84,doi:http://dx.doi.org/10.2147/TCRM.S62351.

[22]Kim HJ, Lee DY, Yoon BK, Choi D. Uterine developmentafter estrogen replacementtherapyinwomenwithdifferentetiologiesofprimary hypogonadism.JPediatrAdolescGynecol2016;29:344–7,doi:http://dx.doi. org/10.1016/j.jpag.2015.11.011.