IUI cycles: a randomized clinical trial Early versus late hCG administration to trigger ovulation in mildstimulated Biology European Journal of Obstetrics & Gynecology andReproductive

(1)

Early versus late hCG administration to trigger ovulation in mild stimulated IUI cycles: a randomized clinical trial

Ana Luisa B. da Silva

^a

, Elisangela Arbo

^a,b

, Renato Fanchin

^a,c,d,

*

aAP-HP,ServicedeGyne´cologie-Obste´triqueetMe´decinedelaReproduction,HoˆpitalAntoineBe´cle`re,Clamart,F-92141,France

bFerringSAS,Gentilly,France

cUnivParis-Sud,Clamart,F-92140,France

dINSERM,U782,Clamart,F-92140,France

1. Introduction

Theadequateappraisaloffolliclepreparednessfortherelease of a mature, reproductively competent oocyte constitutes the foremostobjectiveofthemonitoringoffolliclegrowthandhealth, a standard prerequisite of assisted reproductive technologies (ART).Yet,incontrastwiththemultiplerefinementsmadeinboth thereliability and theflexibility of drugs used to control final follicle maturation and ovulation during recent years, the ultrasonographiccriteriausedtodeterminefollicle/oocytereadi- nessforovulationtriggeringremainimprecise.

Over the last two decades, clinical studies aimed at examining the possible relationship between follicle size, maturation,andoocytereadiness for ovulationled toremark- ablyconﬂictingresults [1–7].Allof themwere focusedonthe

fate of oocytes and/or embryos originated from individual follicles at different sizes in patients undergoing in vitro fertilization-embryotransfer(IVF-ET).Althoughthismethodol- ogyiscontributive,itdoesnotallowdeﬁniteconclusionstobe drawnontheappropriate timingforadministering hCGwhen multiple ovarian follicles are concomitantly developing, espe- cially in non-IVF-ET cycles. Indeed, during controlled ovarian stimulation (COS), numerous follicles at different maturation stages are able to release an oocyte in response to hCG and, therefore, potentially inﬂuence the likelihood of pregnancy.

Moreover,onlyscarceandfragmentarydata[8–11]areavailable onthepractical consequencesofadvancingorpostponinghCG administrationaccording tothe size ofthe leadingfollicle, an issue that may contribute to improvingthe cost-effectiveness andsimplicityofthesetreatments.

Hence, to address this issue, we conducted a prospective randomizedstudy toinvestigatethepossibleinﬂuenceof early versus late hCG administration on pregnancy rates in patients undergoing mild COS for intrauterine insemination (IUI). Our primaryobjectivewastoverifythenon-inferiorityoftheclinical pregnancyrateifhCGisadministeredwhentheleadingfollicle ARTICLE INFO

Articlehistory:

Received14June2011

Receivedinrevisedform4May2012 Accepted27May2012

Keywords:

Intrauterineinsemination Menotropin

HP-hMG IUI

hCGadministration Ovulationtriggering Pregnancyrate

ABSTRACT

Objectives:Toverifynon-inferiorityoftheclinicalpregnancyrateofEarlyhCGadministration(leading folliclesizeswithin16.0–16.9mmindiameter)comparedtoLatehCGadministration(leadingfollicle sizeswithin18.0–18.9mmindiameter).

Studydesign:Prospectiverandomizedtrial.Six hundredandtwelveinfertilewomencandidatesfor intrauterineinsemination(IUI)receivedHP-hMG75IU/daySCfromcycledays4to8andthenasper ovarian response.Ovulation was randomly triggered (hCG 5000IU, IM) when the leading follicle diameterrangedbetweeneither16.0and16.9mm(EarlyhCGgroup,n=227)or18.0and18.9mm(Late hCGgroup,n=207)andIUIwasperformedapproximately36hlater.

Results:Whereaspopulationandspermcharacteristicswerecomparableinbothgroups,thenumberof follicles14mmindiameter(P<0.007)andserumestradiollevels(P<0.001)onthedayofhCGwere lowerintheEarlyversustheLatehCGgroups.Clinical(11.9%versus12.1%)andongoing(11.0%versus 8.6%)pregnancyrates perrandomizedwomenweresimilarinthetwogroups andstatisticalnon- inferiorityofclinicalandongoingpregnancyrateswasdemonstrated.

Conclusion:TheseresultssuggestthathCGadministeredwhenthelargestfolliclesizereaches16.0–

16.9mmleadstosimilarclinicalandongoingpregnancyratesaswhenitreaches18.0–18.9mminIUI cycles.

ß2012ElsevierIrelandLtd.Allrightsreserved.

*Corresponding author at: Department of Obstetrics and Gynecology and ReproductiveMedicine,HoˆpitalAntoineBe´cle`re,157,ruedelaPortedeTrivaux, 92141,Clamart,France.Tel.:+33145374053.

E-mailaddress:renato.fanchin@abc.aphp.fr(R.Fanchin).

ContentslistsavailableatSciVerseScienceDirect

European Journal of Obstetrics & Gynecology and Reproductive Biology

j o urn a l hom e pa ge : ww w. e l s e v i e r. c om/ l o ca t e / e j ogr b

0301-2115/$–seefrontmatterß2012ElsevierIrelandLtd.Allrightsreserved.

http://dx.doi.org/10.1016/j.ejogrb.2012.05.034

(2)

sizesare within 16.0–16.9mm or when they are within 18.0–

18.9mmindiameter.

2. Materialsandmethods

FromMarch2004toApril2007weprospectivelyinvestigated infertilewomencandidatesformildCOSandIUI.Thestudywas approved by the pertinent Institutional Review Boards (CPP, HoˆpitaldeBiceˆtre,LeKremlin-Biceˆtre,FranceandAFSSAPS,Saint- Denis,France), and allparticipantssigned an informedconsent beforeinclusion.Allpatientsconformedtothefollowinginclusion criteria:womenaged20–40years;regularmenstrualcycles(28–

32 days); infertility from2 to5 years; 3previous IUIcycles;

normaluterinecavityconﬁrmedbyhysterosalpingographyand/or hysteroscopy;bilateraltubalpatencyconﬁrmedbyhysterosalpin- gographyorlaparoscopy;normalserumFSHandestradiollevels oncycleday31within1yearbeforeinclusion;spermselection testshowingat least10⁶ spermatozoa,grade A/B spermmotility

>30%,>30%ofnormalforms,and/or>30%ofspermsurvivalafter 24horadequatedonorsperm. Exclusioncriteriawereasfollows:

unilateralorbilateraltubalobstruction;endometriosisgradeIIIorIV;

metrorrhagia of unknown origin; present or past malignant, metabolic,orendocrinediseases;cervicalinfection;positiveserology forhepatitisBorC,HIVorsyphilis;anti-spermatozoa antibodies;

positive sperm culture; ejaculation disorders; alcohol or drug addiction; participation in another clinical trial in the previous month.Finally,afterenrolment,exclusioncriteriawere:presenceofa seriousadverseeffect,possiblyduetothetreatment;serumestradiol levels >1500pg/mL; and occurrence of a spontaneous LH surge duringCOSbeforethedayofhCGadministration.

ParticipantswererandomlyallocatedtoreceivehCGinjection whenthemeandiameteroftheleadingfolliclerangedfrom16.0to 16.9mm(EarlyhCGgroup) orfrom18.0to18.9mm(Late hCG group). Groups of randomizationwere automaticallygenerated usingacentralizedtelephonicsystem(InteractiveVoiceResponse System).ThestudywascoordinatedbytheReproductiveMedicine Unit,HoˆpitalAntoineBe´cle`re,Clamart,Franceandconductedby independentcliniciansintheirownmedicalofﬁce(ahundredof clinicians participated). All participants received precise and standardizedinstructionsandthefolliclemeasurementtechnique wasstandardized.Folliclediameterswereassessedusingtrans- vaginal ultrasound scans by determining the two orthogonal diameters(d1and d2)atthelargestfollicleplanewithcalipers placed at the inner follicle borders. Mean follicle diameter corresponded to (d1+d2)/2 and results were expressed in millimeters(mm).

For mild COS, women received highly puriﬁed human menopausalgonadotropin(HP-hMG,Menopur¹,FerringPharma- ceuticals,Gentilly,France),SC,75UI/dayfromcycledays4to8.

Thereafter, dose adjustments were made according to ovarian responseuntilthecriteriaforhCGadministrationweremet.The daily HP-hMG dose never exceeded 300IU. Transvaginal ultra- soundswereperformedatleastoncycledays4and9forfollicle measurementandalsoserumlevelsofestradiol,progesterone,and LHwereveriﬁed.Forovulationinduction,5000IUofhCG,IM,was administeredinbothgroups(GonadotrophineChorioniqueEndo¹, OrganonSA,Puteaux,France).Approximately36hlatter,IUIwas performed. Sperm wascollected on site by masturbation, then evaluatedandpreparedwithin1hbydoublecentrifugationusing standardizedprotocols.Lutealphasesupportwasobtainedwith natural micronized progesterone (Estima Ge´¹, Efﬁk, Bie`vres, France), 600mg/day, by thevaginal route until thepregnancy test(serumß-hCGmeasurement),14dayslater.Compliancewas assessedbycomparingtheamountofmedicationdeliveredbyand returnedtoinvestigators.

Theprimaryoutcomewastheclinicalpregnancyrate,deﬁned asthepresenceofanintrauterinegestationalsacwithfetalcardiac activity 4 weeks after IUI. Secondary outcomes were ongoing intrauterinepregnancyrateat10weeksafterIUIandincidenceofa premature LH surge before hCG administration (serum LH

>10mIU/mL).

Allstatisticswereperformedbyanindependentgroup(ICTA- PM,Fontaine-les-Dijon,France),usingSASsoftware(version8.2, SAS Institute Inc., NC, USA). Toassess theequivalence of both methods, a non-inferiorityanalysiswas conducted.Considering estimatedclinicalpregnancyrateat12%,atypeIerrorof0.05,a statisticalpowerof80%,andamaximuminter-groupdifferencein clinical pregnancy rates of 2.7% witha lowerlimit of the 95%

conﬁdenceintervalof 8%,260patientsineachgroupwouldbe necessary accordingtothebilateralDunnett andGent method.

Assumingalikelihoodof30%non-analyzablepatients,itwouldbe necessarytoinclude338patientsineachgroup.Dataarepresented asmeansandstandarddeviationswheneverthenormaldistribu- tionisconﬁrmed.Qualitativedataaredescribedaspercentages, anddifferencesbetweenbothgroupswereassessedbyPearson’s ChiSquaretest.

3. Results

Apatientﬂow-chartisdepictedinFig.1.Asshown,635patients wererandomizedtooneofthetwostudygroups(EarlyhCGgroup, n=317;LatehCGgroup,n=318).Fromthese,309patientsfrom theEarlyhCGgroupand303fromtheLatehCGeffectivelystarted theovarianstimulationprotocolandweredeﬁnedastheintention to treat (ITT) population. Finally, 434 patients received hCG accordingtoprotocolrequirements(EarlyhCGgroup,n=227;Late hCG group, n=207), thereby constituting theper-protocol (PP) population.Fourpatientshadcyclecancellationduetoexcessive folliculargrowth.

PopulationcharacteristicsintheEarlyandLatehCGgroupsin thePPpopulationaredetailedinTable1.Patientsfrombothgroups were similarwithregard tobaseline characteristics and sperm characteristics. Homogeneity analysis showed that ITT and PP populationswerestrictlycomparable(datanotshown).

Dataonovarianresponsetostimulationinbothgroupsinthe PPpopulationaresummarizedinTable2.Despitethefactthat,in theEarlyhCGgroup,folliclesizecriteriatoadministerhCGshould havebeenmetearlierthanintheLatehCGgroup,boththetotal HP-hMGdoseadministeredandthecycledayinwhichhCGwas administered were comparable in both groups. However, the number of follicles 14mm in diameter on the day of hCG administrationwaslowerintheEarlyhCGgroupaswellasthe prevalenceof3follicles14mm. Serumestradiol(P<0.001) butnotprogesterone(P=0.513)levelswereloweronthedayof hCGadministrationintheEarlyhCGgroup.Moreover,serumLH levelsonthedayofhCGadministrationwerelower(P<0.03)in thisgroup.

Clinical outcomes in both ITT and PP populations are presentedinTable3.Clinicalpregnancyrateswerecomparable in theEarlyandLatehCGgroupsin bothITT(11.7%and9.6%, respectively)andPP(11.9%and12.1%,respectively)populations.

Non-inferiorityof theclinicalpregnancy ratein theEarlyhCG groupascomparedtotheLatehCGgroupwasdemonstratedin both populations (P<0.001 and P<0.007 for ITT and PP populations). Also, in both ITT and PP populations, ongoing pregnancyratesintheEarly(11.0%and11.5%,respectively)and LatehCGgroups(8.6%and10.7%,respectively)werecomparable andnon-inferiorityof theEarlyhCGgroup wasdemonstrated.

Finally, the multiple pregnancy rate was not statistically differentbetweengroups.

(3)

4. Comment

ThepresenttrialcomparedtheefﬁcacyoftwostrategiesofhCG administrationforovulationtriggeringinHP-hMGstimulatedIUI cycles according to leading follicle sizes. It examined the hypothesis that, in this modality of treatment, preovulatory follicles can release a reproductively competent oocyte at a diameter as small as 16mm and that prolonging ovarian stimulation further to achieve larger follicle sizes would not improvetreatmenteffectiveness.Toreducetheriskofoverlapping betweenfolliclesizegroupsandmisinterpretationofresults,we

considered onlyfollicles includedintotwo distinctand speciﬁc diameter frames. Non-inferiority analysis demonstrated similar clinicalandongoingpregnancyratesirrespectiveofwhetherhCG wasadministeredwhentheleadingfolliclewas16.0–16.9mmor 18.0–18.9mmindiameter.Wealsoobservedthat,althoughthe numberoffollicles>13mmwassigniﬁcantlylargerintheLateas compared to the Early hCG group, multiple pregnancy rate remainedsimilarin bothgroups. Asthepresentstudy wasnot poweredtodetectdifferencesinmultiplepregnanciesrate,itdoes notallowaconclusiononthepossibleincreaseintheprevalenceof thiscomplicationwhenhCGinjectionispostponed.

Another potential inconvenience often associated with the prolongingofCOSistheriskofaprematureLHsurge.Despitethe factthatmedianserumLHlevelsonthedayofhCGadministration wereslightlyincreasedintheLatehCGgroup,wefailedtoobserve asigniﬁcantdifferenceintheincidenceofprematureLHsurges betweenthetwostrategies.Besidesthecleardifferenceinleading folliclesizes,dissimilaritiesintheintensityofovarianresponseto stimulation, afterthe5-day 75IU/day course,may constitutea possibleadditionalexplanationforthisphenomenon.

The present studydidnot aim at speciﬁcally assessing the reproductivepotentialof16.0–16.9mmfolliclesascomparedto 18.0–18.9mm follicles. Given that the coexistence of 18.0–

18.9mm follicles with smaller follicles in the Late hCGgroup could not be excluded, it is not possible to assertthat 18.0–

18.9mm follicles are as healthy as 16.0–16.9mm follicles, despite the lack of difference observed in pregnancy rates betweenbothgroups.Ourdatasimplyindicatethatcycleswith leading follicles measuring18.0–18.9mm in diameterare not associatedwithanincreasedchanceofobtainingaclinicaloran ongoingpregnancyascompared tothosein whichtheleading follicles measured 16.0–16.9mm. In contrast with this, Rich- mond et al. [11] observed thatwhereas 17mm follicles are Randomization

n=635

Early hCG group^* (Intention to treat population)

n=309

Late hCG group^¶ (Intention to treat population)

n=303 Treatment not initiated

n=23

Exclusions^§

-Major protocol deviation (n=52)^£ -hCG not administered^#(n=30)

Exclusions^§

-Major protocol deviation (n=65)^£ -hCG not administered^#(n=31)

Late hCG group (Per-protocol population)

n=207 Early hCG group

(Per-protocol population) n=227

Fig.1.Patients’ﬂow-chart.*hCGadministrationwhenleadingfolliclereached16.0–16.9mmindiameter;^ôhCGadministrationwhenleadingfolliclereached18.0–18.9mm indiameter;^§eachpatientcouldbeexcludedformorethanonereason;£majorprotocoldeviationatinclusionweretubalpermeabilitynotassessed(1patientintheEarly hCGgroupand2intheLatehCGgroup),tubalobstruction(1patientintheEarlyhCGgroupand2intheLatehCGgroup),endometriosisstageIII/IV(1patientintheLatehCG group).MajorprotocoldeviationduringCOSwereanotherhormonaltreatmentduringCOS(2patientsintheEarlyhCGgroupand4intheLatehCGgroup),leadingfolliclesize onthehCGdaynotcorrespondingtorandomizationgroups(51patientsintheEarlyhCGgroupand60intheLatehCGgroup).^#Duetoinsufﬁcientfolliculargrowth,serum estradiollevels>1500pg/mL,orprematureLHpeak(serumLHlevels10mUI/mL)beforethecriteriaforhCGadministrationweremet.

Table1

PopulationcharacteristicsintheEarly(16.0–16.9mm)andLate(18.0–18.9mm) hCGgroupsinthePPpopulation.

EarlyhCGgroupLatehCGgroupP (n=227) (n=207)

Age(years) 30.93.8 31.03.8 0.823

Bodymassindex(Kg/m²) 22.74.2 22.33.5 0.846 Durationofinfertility(months) 38.322.2 38.320.0 0.484 Typeofinfertility(%)

Primary 68.2 68.5 0.945

Secondary 31.8 31.5

Etiologyofinfertility(%)

Female 26.4 22.2 0.545

Male 21.2 25.2

Mixte 24.2 27.0

Unexplained 28.2 25.6

Endometriosis(stagesIorII)(%) 12.3 8.6 0.334 No.ofpreviousIUI(%)

0 77.0 69.1 0.198

1 10.2 15.9

2 12.4 13.0

3 0.4 2.0

PasthistoryofpregnancyafterART(%) 7.5 9.2 0.348

(4)

associatedwith74%of attributable fetal hearts obtainedafter IUI, 16mm or 15mm follicles are associated only with respectively 31% (P<0.0001) and 8.1% (P<0.0009) of them, suggesting an improved reproductive potential of 17mm follicles. The inherent difﬁculties for establishingan accurate linkbetweenthefolliclediameterandtherespectivegestational sacin IUIcycles mayexplain,at leastinpart,the discrepancy between theirresults andours.Some otherfew,retrospective trials,studyingdifferentfolliclesizesthanthoseelectedinthe present investigation and including several thousands of stimulatedIUIcycles,alternatelyshowed thatpregnancy rates are reduced when follicle diameter is either 20mm or<15mm [8],similarwhen folliclediametersare15–19mm or20mm[9],orinﬂuencedbythenumberof12mmfollicles onthedayof hCGadministration[10].

Thepossiblerelationshipbetweenfolliclesizeandinvitrofate of the ensuing oocyte in IVF-ET cycles has been addressed in differentstudies[1–7].Mostofthemconsideredthesizeoffollicles on the day of oocyte retrieval instead of the day of hCG administration. Haines and Emes failed to associate either fertilization rates or embryo morphology withthe diameterof theoriginatingfollicle[1].Incontrast,Dubeyetal.showedhigher fertilizationrateswhentheoocytecamefromfollicleswithamean diameter16mm thanwhen it came fromthose14mm[2].

Similarly, Arnot et al. identiﬁed a signiﬁcant improvement of fertilization and ensuing embryo grading with the progressive

increase inthefolliclevolumes[3].Theseresultswerepartially conﬁrmedbyEctorsetal., whoobserveddecreasedfertilization ratesbutnotembryomorphologywhenoocyteswereoriginated from<2mLfollicles[4].Inarecentstudy,bothfertilizationand embryomorphologywereobservedtobeaffectedbysmallfollicle sizeonthedayofoocyteretrieval[7].Othersobservedareduction intheabilityofanoocytetogenerateapregnancywhenitcame from<2mLfolliclesafterconventionalIVF-ET,butnotIVF-ETwith ICSI,asifICSIcouldovercomepossibleinsufﬁcienciesofoocytes originatedfromsmallfollicles[5].Anotherextensiveanalysisofa totalof9 933oocytes concludedthatfolliclesizeand develop- mentalcapacityoftheoocytearenotcloselyrelatedandIVF-ET outcome of oocytes/embryos from 1 or>1mL follicles [6].

Together, these data suggest that very small follicle sizes are possibly associatedwithadecreasein oocyte/embryofunctions buttheabilityoftheseoocytestogenerateapregnancyremains controversial.

Anotherinterestingissueobservedwasthat,althoughhCGwas deliberatelyadministeredearlierintheEarlyhCGgroup,boththe totalHP-hMGdoseadministeredandthedurationofstimulation remained similar in both groups. These unexpected results possiblyreﬂectpatient-to-patientdifferencesintermsofgonado- tropinrequirementsforCOS.Indeed,speciﬁcmodulationofHP- hMGdosesfromcycleday9onwardscouldhaveeitherslowedor hastenedthepaceoffolliclegrowthirrespectiveoftherandomi- zationgroup.Thismethodologicalcontingencypreventedusfrom Table2

Clinicalparameters,ovarianresponsetostimulationandspermcharacteristicsintheEarly(16.0–16.9mm)andLate(18.0–18.9mm)hCGgroupsinthePPpopulation.

EarlyhCGgroup LatehCGgroup P

(n=227) (n=207)

TotalHP-hMGdose(IU) 639282 648250 0.183

DailyHP-hMGdose(IU) 80.714.2 80.411.9 0.607

DayofhCGadministration(cycleday) 11.72.2 11.61.9 0.416

No.offollicles14mm^a,b 1.560.14 1.690.07 0.025

Prevalenceoffollicles14mm^a

1or2(%) 94.7 86.0 0.007

3(%) 5.3 14.0

Endometriumthickness(mm)^a 8.71.6 9.01.6 0.056

Serumestradiollevel(pg/mL)^a,b 338.415.22 403.914.67 0.001

Serumprogesteronelevel(ng/mL)^a,b 0.850.1 0.680.038 0.513

SerumLHlevel(IU/L)^a,b 6.100.44 6.930.48 0.022

IncidenceofprematureLHsurge(%)^c 13.4 18.1 0.177

SpermcharacteristicsatIUI

Volumeofinseminate(mL) 0.250.11 0.270.11 0.346

Inseminatedmotilespermcount(10⁶)^d 8.1[5.7–10.5] 10.8[5.0–16.7] 0.830

aOnthedayofhCGadministration.

b Meanstandarddeviation.

cDeﬁnedasserumLHlevels10mIU/mL.

d Mean[95%CI].

Table3

IUIclinicaloutcomeinbothITTandPPpopulations.

EarlyhCGgroup LatehCGgroup D(95%CIs)^a P^y P^z

ITTpopulation

n 309 303 – – –

Clinicalpregnancies,n(%)^b 36(11.7) 29(9.6) 2.1( 3.1–7.3) 0.001 0.404

Ongoingpregnancies,n(%)^c 34(11.0) 26(8.6) 2.4( 2.6–7.4) 0.001 0.314

PPpopulation

n 227 207 – – –

Clinicalpregnancies,n(%)^b 27(11.9) 25(12.1) 0.2( 6.8–6.4) 0.007 0.953

Ongoingpregnancies,n(%)^c 26(11.4) 22(10.6) 0.8( 5.5–7.2) 0.002 0.784

No.ofmultiplepregnancies^d 2 0 – – 0.223

aDifferenceinclinicalandongoingpregnancyratesbetweenbothgroupswith95%continuity-correctedWaldconﬁdenceintervals.

b Deﬁnedasthepresenceofanintrauterinegestationalsacwithfetalcardiacactivityat4weeksafterIUI.

cDeﬁnedasthepresenceofanongoingintrauterinepregnancyat10weeksafterIUI.

d Onlytwinpregnancieswereobserved.

y NoninferioritybetweenbothgroupsdemonstratedwhenP<0.025;unilateralx²ofDunnettandGent.

z DifferencebetweenbothgroupsdemonstratedwhenP<0.05;bilateralx²ofPearson.

(5)

drawingdeﬁniteconclusionsonthepresumablecost-effectiveness advantagesofearlyhCGadministration.

Inconclusion,ourresultsfailedtodemonstrateaclearbenefitin targetinglarger leading folliclesizesfor hCG administration in mild-stimulatedcyclesforIUI.Byextrapolation,theyalsosuggest thatprolongingCOStoobtainleadingfollicles18mm iscost- ineffective and potentially more complex. From a practical standpoint,our presentdata are reassuring withregard to the possibilityofschedulingIUItoconvenientdaysforphysiciansand patients. Also, they are in line with data from a previous investigation conducted in IVF-ET cycles showing the lack of outcomeimprovementwhenhCGadministrationwaspostponed for1or 2days[12].Yet,ourstudyincludedaselectednormo- responder population submitted to precise criteria for hCG administration. The results may be different in other clinical pictures or when different criteria for hCG administration are considered. Further studies focusing on the consequences of administering hCG at different follicle size ranges than those chosenbythepresentstudyandindifferentCOSprotocols,maybe including new technological refinements for assessing follicle volumesat ultrasound scans, are needed to define the proper ultrasonographic criteria for triggering final follicle maturation andovulationwithhCGinmildstimulatedIUIcycles.

Acknowledgements

This clinical trail was supported by the French division of FerringPharmaceuticals,Gentilly,France.ALBdaSilvahasnothing to disclose. E Arbo is Medical Advisor at Ferring France since December2010.RFanchinhasnothingtodisclose.

References

[1]HainesCJ,EmesAL.Therelationshipbetweenfolliclediameter,fertilization rate,andmicroscopicembryoquality.FertilityandSterility1991;55:205–7.

[2]DubeyAK,WangHA,DuffyP,PenziasAS.Thecorrelationbetweenfollicular measurements, oocyte morphology,and fertilization ratesinan in vitro fertilizationprogram.FertilityandSterility1995;64:787–90.

[3]ArnotAM,VandekerckhoveP,DeBonoMA,RutherfordAJ.Follicularvolume andnumberduringinvitrofertilization:associationwithoocytedevelopmen- talcapacityandpregnancyrate.HumanReproduction1995;10:256–61.

[4]EctorsFJ,VanderzwalmenP,VanHoeckJ,NijsM,VerhaegenG,DelvigneA, etal.Relationshipofhumanfolliculardiameterwithoocytefertilizationand developmentafterinvitrofertilizationorintracytoplasmicsperminjection.

HumanReproduction1997;12:2002–5.

[5]BerghC, Broden H,Lundin K,Hamberger L. Comparison of fertilization, cleavageandpregnancyratesofoocytesfromlargeandsmallfollicles.Human Reproduction1998;13:1912–5.

[6]SalhaO,NugentD,DadaT,KaufmannS,LevettS,JennerL,etal.Therelation- shipbetweenfollicularﬂuidaspiratevolumeandoocytematurityininvitro fertilizationcycles.HumanReproduction1998;13:1901–6.

[7]Rosen MP,Shen S, DobsonAT, RinaudoPF, McCullochCE, CedarsMI.A quantitativeassessmentoffolliclesizeonoocytedevelopmentalcompetence.

FertilityandSterility2008;90:684–90.

[8]Ghosh C, Buck G, Priore R, Wacktawski-WendeJ, Severino M.Follicular responseandpregnancyamonginfertilewomenundergoingovulationinduc- tionandintrauterineinsemination.FertilityandSterility2003;80:328–35.

[9]Ibe´ricoG,VioqueJ,ArizaN,LozanoJM,RocaM,Lla´cerJ,etal.Analysisoffactors inﬂuencingpregnancyratesinhomologousintrauterineinsemination.Fertili- tyandSterility2004;81:1308–13.

[10]DickeyRP,TaylorSN,LuPY,SartorBM,RyePH,PyrzakR.Relationshipoffollicle numbersandestradiollevelstomultipleimplantationin3608intrauterine inseminationcycles.FertilityandSterility2001;75:69–78.

[11]RichmondJR,DeshpandeN,LyallH,YatesRW,FlemingR.Folliculardiameters inconceptioncycleswithandwithoutmultiplepregnancyafterstimulated ovulationinduction.HumanReproduction2005;20:756–60.

[12]TanSL,BalenA,elHusseinE,etal.Aprospectiverandomizedstudyofthe optimumtimingofhumanchorionicgonadotropinadministrationafterpitui- tary desensitization in in vitro fertilization. Fertility and Sterility 1992;57:1259–64.