paediatric Crohn’s disease Seasonal variability of vitamin D and bone metabolism ininﬂiximab-treated Digestive and Liver Disease

(1)

ContentslistsavailableatScienceDirect

Digestive and Liver Disease

j o u r n a l ho me p a g e :w w w . e l s e v i e r . c o m / l o c a t e / d l d

Alimentary Tract

Seasonal variability of vitamin D and bone metabolism in inﬂiximab-treated paediatric Crohn’s disease

Dolóresz Szabó

^a

, Éva Hosszú

^b

, András Arató

^a

, Katalin Eszter Müller

^a

, Nóra Béres

^a

, Péter László Lakatos

^c

, Mária Papp

^d

, Antal Dezs ˝oﬁ

^a

, Attila J. Szabó

^a,e

, Dániel Sz ˝ucs

^f

, Gabor Veres

^a,∗

a1stDepartmentofPediatrics,SemmelweisUniversity,Budapest,Hungary

b2ndDepartmentofPediatrics,SemmelweisUniversity,Budapest,Hungary

c1stDepartmentofMedicine,SemmelweisUniversity,Budapest,Hungary

dInstituteofInternalMedicine,DepartmentofGastroenterology,UniversityofDebrecen,ClinicalCenter,Debrecen,Hungary

eMTA-SE,PediatricsandNephrologyResearchGroup,Budapest,Hungary

fDepartmentofPediatricsandPaediatricHealthCareCenter,UniversityofSzeged,FacultyofMedicine,Szeged,Hungary

a r t i c l e i n f o

Articlehistory:

Received15October2014 Accepted1May2015 Availableonline19May2015

Keywords:

Bonemineraldensity Bonemarkers Crohn’sdisease Inﬂiximab Paediatric VitaminD

a b s t r a c t

Background:PaediatricCrohn’sdiseasepatientssufferfromseveralcomplications,includinglowbone mineraldensityandinadequateserumlevelsof25-hydroxyvitaminD.

Aims:The aimofthis prospective studywas toaddress theeffectof inﬂiximabtherapy on bone metabolism,bonemineraldensityandvitaminDhomeostasis.Theseasonalvariabilityofserumvitamin Dlevelsinrelationtoinﬂiximabtreatmentwasalsoanalysed.

Methods:Serumosteocalcinandbeta-crosslaps(markersofbonemetabolism),seasonalvariabilityof vitaminD,andbonemineraldensitywereassessedandfollowedthroughouttheyearlongtreatment regimenofinﬂiximabin50consecutivepaediatricpatientswithmoderatetosevereCrohn’sdisease.

Results:Boneformingosteocalcinlevelsweresigniﬁcantly(p<0.001)increasedduringinﬂiximabtherapy.

Incontrast,nosigniﬁcantchangesinbeta-crosslapsandvitaminDlevelswereobserved.VitaminDlevels weresigniﬁcantlydifferentwhenthesummerandwinterperiodswerecomparedatweek0(p=0.039);

however,thisdifferencewasnotdetectedafteroneyearofinfliximabtherapy.Despitethebeneficial clinicaleffectofinfliximab,therewasnosignificantchangeinbonemineraldensityZ-scoresafterone yearoftreatment.

Conclusion:Infliximabmaybeneficiallyaffectbonehomeostasis.Moreover,seasonalvariabilityinvitamin Dlevelsobservedpriortoinitiationofinfliximabtreatmentwasdiminishedafteroneyearoftreatment.

1. Introduction

Crohn’sdisease(CD)isachronicinﬂammatoryboweldisease (IBD) condition affecting the gastrointestinal tract. There is an increasingincidenceofCDinchildhood,whichhasthuscontributed toanincreasednumberofaffectedchildren[1,2].Chronicintestinal inﬂammationleadstolowbonemineraldensity(BMD),afrequent complicationinbothchildrenandadultswithCD[3,4].Ithasbeen hypothesizedthatlowBMDinCDoccurseitherbecauseofthedis- easeitselforduetothetherapeuticsused.AbnormalitiesinBMD,

∗Correspondingauthorat:1stDepartmentofPediatrics,SemmelweisUniversity, 53BókayStreet,1083Budapest,Hungary.Tel.:+3613343743;fax:+3613036077.

E-mailaddress:veres.gabor@med.semmelweis-univ.hu(G.Veres).

whichcanbeevaluatedbydensitometry,canalsobeanalysedby biochemicalmarkersofboneturnoversuchasosteocalcin,amarker ofboneformation,and␤-isomerizedC-terminaltelopeptidefrag- mentsofcollagentypeI(beta-CrossLapsorbCL),anindicatorof boneresorption[5].

Inaddition toinﬂammation,severalothermechanismshave beenimplicatedinCD-relatedreductionofBMD,suchasmalab- sorptionand malnutrition,genetics,drugtherapy (e.g.corticos- teroids),lowdietaryintakeofessentialvitamins(e.g.25-hydroxy vitamin D), and a lack of physical activity. Release of chronic pro-inﬂammatorycytokinesplaysapivotalroleintheregulation and maintenance of bone health. Tumournecrosis factor alpha (TNF-␣),interleukin-1␤(IL-1␤)andIL-6␤stimulateboneresorp- tionthroughabalancedmodulationofosteoclastandosteoblast activationanddifferentiation[6,7].While previousstudieshave http://dx.doi.org/10.1016/j.dld.2015.05.006

(2)

implicatedcorticosteroidsinimpairedBMD[8],othertreatments havebeenshowntohaveabeneficialeffectonbonethroughreg- ulationofpro-inflammatorycytokines.Infliximab(IFX),achimeric monoclonal antibody therapeutic targeted against TNF-␣, may improve bone mineralization [9].Several differentmechanisms contributetothepositiveeffectsofIFX.WhileIFX-mediatedreduc- tionofpro-inflammatorycytokines,suchasTNF-␣,IL-6andIL-1␤, representanimportantmodeofaction[10,11],theIFX-induced reductioningutinflammationalsopromotesimprovedabsorption ofcriticalbonenutrientssuchasvitaminDandcalcium.More- over,animprovementinpatienthealthleadstoincreasedphysical activity,whichmayalsoleadtoanincreaseboneformation[6].

VitaminD is an essential nutrient requiredfor proper bone mineralization.Vitamin Dhasalsobeen shown toregulatethe immunesystemduetoitsindispensableroleinthedevelopment ofself-tolerance.Studiesinanimalandinvitromodelshaveshown thatvitamin Dfunctionsin theformationofa normal immune responseandenhancementofcytokineproduction[12,13].Recent studieshaverevealedahigherprevalenceofvitaminDdeficiency inadultsandchildrenwithIBD[3,14].ThebenefitsofvitaminD indicatethatitmaybeanadditionaltreatmentoptioninCD.An earlierstudy,however,demonstratedconflictingresultswhenCD patientswithlowBMDweretreatedwithcalciumandvitaminD [15].

VitaminDlevelsﬂuctuatewiththedifferentseasonsoftheyear.

ThelatewintermonthsareassociatedwithlowervitaminDlevels bothinhealthyindividualsandIBDpatients[13,16,17].Theeffect ofseasonsonvitaminDlevelsinpatientsreceivingIFXtherapy, however,hasnotyetbeenstudied.

TheprimaryaimofthisstudywastoevaluatetheeffectofIFX treatmentonclinicalparameters,bonemetabolism,bonemineral density,andvitaminDhomeostasisinmoderatetoseverepaedi- atricCDpatients.Wealsoprovideacharacterizationoftheseasonal variability ofvitamin D levelsatbaseline and follow-up in our patientpopulation.

2. Patientsandmethods

2.1. Patients

Ourprospectivestudywasperformedinthe1stDepartment of Paediatrics of Semmelweis University from January 2009to December 31, 2013.Fiftychildren withmoderate tosevere CD who wereresistant toconventional therapy (azathioprine,sys- temicsteroids)andhadaPaediatricCrohn’sDiseaseActivityIndex (PCDAI)>30wereincludedintheanalysisofvitaminDlevel,bone markersand BMDchanges duringtheyearlong IFXtherapy. To monitortheseasonalchangesofserumvitaminDlevels,anaddi- tional25childrenwithCDwhoweretreatedwithIFXwereincluded (onlythevitaminDserumlevelwasavailable).

Inclusioncriteriawereresistancetoconventionaltherapyforat leastthreemonthsfollowingtheirdiagnosis,aPCDAIscorehigher than30whenIFXtherapywasinitiated,andvitaminDsupplemen- tationforatleastthreemonthspriortoIFXtherapy.Administration ofazothiprineorsteroidswasnotgroundsforexclusion.

Controlgroup.Acontrolgroupcontaining34childrenwithCD inclinicalandIFX-free(PCDAI≤12.5)remissionwasalsoincluded inthisstudy.VitaminDserumlevelwasmeasuredineverypatient fromthecontrolgroup.Additionally,14patientswerealsoeval- uatedforosteocalcinandbCLlevels.Inthepatientsanalysedfor osteocalcinandbCLlevels,steroidtreatmentwasnotallowedasa formofremissioninduction.

Written,informedconsentwasobtainedfromparentspriorto studyenrollment.ThestudywasapprovedbytheSemmelweisUni- versityRegionalandInstitutionalCommitteeandResearchEthics.

2.2. Methods

Vitamin D and calcium supplementation. During treatment, patients(IFXandcontrolgroups)received1000U/dayvitaminD and500mg/daycalciumsupplements.

PatientsintheIFXtreatmentgroupreceiveda5mg/kgintra- venousinfusionofIFXasaninductiontherapyatweek0,2and6, andasamaintenancetherapyateveryfollowing8thweek,based ontheguidelinesforIFXtreatmentinpaediatricCD[18].

IMPACT-IIIandPCDAI.Tomeasurehealth-relatedqualityoflife (QoL)theIMPACT-IIIquestionnaire–developedbyOtleyetal.and validatedinHungary–wasadministeredtopatients.IMPACT-IIIisa self-reportedquestionnaireconsistingof35queries.Onaﬁvepoint Likertscale,childrenindicatedtheextenttowhichtheybelongto eachquerybyspeciﬁcaspectsoftheirhealthcondition.IMPACT- IIIscoresrangefrom35to175,withahigherscoreindicativeofa betterQoL[19,20].Inaddition,PCDAIwasevaluatedandfollowed overtime[21].

BonemarkersandvitaminD.LevelsofserumosteocalcinandbCL, markersofbonemetabolismandboneturnoverrespectively,and 25-hydroxicholecalciferol(vitaminD)quantityweremeasuredat week0,week6,week30andweek53.Bloodsampleswerecollected byvenepunctureintoavacutainertubewithnoadditive,andthe serumswereprocessedbeforetheIFXtreatmentstarted.Samples forthevitaminDanalysiswereprotectedfromsunlightandwere post-processedwithinafewhoursofcollectionfromthepatient.A standardconsensusregardingthenormalrangeofvitaminDlevels hasnotyetbeenestablished.Using therecentguidelinesestab- lishedbytheEndocrineSocietyin2011,vitaminDdeficiencywas definedas≤20ng/mLwhilevitaminDinsufficiencywasdefinedas 21–29ng/mL[16,22].

BasedonthedateoftheﬁrstIFXtherapy,seasonalvariabilityof vitaminDlevelwasmeasured.Thesummerperiodwasdesignated asthe21stofMarchthrough22ndofSeptemberwhiletherestof theyearwasdesignatedthewinterperiod.

Procedures.Immunoassayanalyseswerecarriedoutwithelec- trochemiluminescence immunoassay (ECLIA) methods (Elecsys N-MIDOsteocalcinandElecsysbeta-CrossLaps,Roche^®)andwith achemiluminescenceimmunoassay(25-OHVitaminD,LIAISON^®, DiaSorin),whichwereperformedaccordingtothemanufacturer’s instructions.

Whole-bodyandlumbarspineDual-energyX-rayabsorptiome- try(DEXA)(type:QDRDiscovery,model:DiscoveryA(S/N83638), HOLOGIC,Inc.,USA)scanswereusedprimarilytoevaluateBMD status before IFX treatmentand at the conclusionof one year ofIFX treatment.Allscanswereperformedbythesameopera- tor.Thescanprovidedmeasurementsoflumbarandwhole-body areaBMD (g/cm²). Z-scores werecalculated by subtracting the standard deviation of measured BMD from the expected BMD of individuals of the same age and sex. A Z-score <−2.0 was reportedasreducedbonemineralmass[23].Whole-bodyDEXA measurements served as a readout for cortical bone health, whereaslumbarspineDEXAservedasanassessmentoftrabecular bone.

2.3. Statistics

AllanalyseswereperformedusingSPSS22(IBM^®,Somers,NY) statisticalsoftware.BasedontheresultsofKolmogorow–Smirnov and Shapiro–Wilk normalitytests, thePCDAI, IMPACT-III,bone markersandvitaminDdatasetsfollowedanon-normaldistribu- tion.TheBMDdatashowedanormaldistribution.Todetermine signiﬁcance,theFriedmantestwasusedasanon-parametrictest andapairedt-testwasusedasaparametricprobe.Ap-valueof

<0.05 was setas thethreshold for statistical signiﬁcance. Pair- wisecomparisonswereperformedwithaBonferronicorrectionfor

(3)

Table1

Characteristicsofthestudypopulation.

Totalnumberofpatients 50

Malegender(%) 19(38)

Meanage;years(±SD) 14.8(±2.4)

Meandiseaseduration:years(±SD) 2.1(±2.0) MedicaltreatmentwhenIFXtherapywasintroduced;n(%) Azathioprine+5-ASA+steroid+antibiotic 3(6) Azathioprine+5-ASA+steroid 9(18)

(4methilprednisolone, 5budesonide) Azathioprine+5-ASA+antibiotic 3(6)

Azathioprine+5-ASA 24(48)

Azathioprine+steroid 1(2)

Azathioprine+antibiotic 1(2)

5-ASA+antibiotic 2(4)

5-ASA+steroid 2(4)

Azathioprine 2(4)

5-ASA 3(6)

5-ASA:5-aminosalicylicacid,IFX:inﬂiximab.

multiplecomparisons.Ap-valueof0.01wasconsideredstatistically signiﬁcant.

To determine the correlation between the laboratory and clinicalparameters,theSpearmancorrelationcoefﬁcientwascal- culated.

3. Results 3.1. Patientdata

The IFX group included 50 children (38% males, mean age 14.8±2.4years,range8–18.6).Themeanelapsedtimesincediag- nosiswas2.1±2.1years(Table1).

Thecontrol group included34 children (55.8%males, mean age14.5±3.6years).Themeanelapsedtimesincediagnosiswas 3.15±2.88years.Inthecontrolgroup,noneofthepatientsreceived IFX therapy, and wereall treated withazathioprine and 5-ASA thetimeof measurement;threepatientsalsoreceivedsystemic steroids.

Withregardtoseasonalvariability,36initiatedtreatmentinthe winter(42.8%,meanage14.6±3years)while39patientsbegan theirtreatmentduringthesummer(46.4%,meanage14.7±2.4 years).

3.2. EffectofIFXonPCDAIandIMPACT-III

PCDAI scores were assessed to monitor the efﬁcacy of IFX inductionandmaintenancetherapy.PCDAIdecreasedsigniﬁcantly (p<0.001) during the treatmentperiod when compared tothe initialscores.ThemedianPCDAIscorewas35.0(25thpercentile [pc-25]30;75thpercentile[pc-75]42.5)atweek0anddecreased to10.0(pc-25,-75:1.3,20)byweek6(p<0.0001).Duringmain- tenancetherapy, thePCDAIwas10.0 (pc-25,-75:5,20)and5.0 (pc-25,-75:0,25)atweek30andweek53,respectively(p<0.0001).

Whencomparedwiththebaseline,allchangesintheIFXgroup weresigniﬁcant.

AfteroneyearofIFXtherapy,33/50oftheIFX-treatedchildren wereinsteroid-freeremission(66%,PCDAI≤12.5).

IMPACT-III scores were similarly improved after IFX treat- mentwithsigniﬁcantdifferencesateach monitoringtime point (p<0.0001).ThemedianIMPACT-IIIscorewas116.5(pc-25,-75:

106.5,131)atbaselineandincreasedto142(pc-25,-75:130,149) byweek6(p<0.0001).Duringthemaintenancetherapy period, theIMPACT-IIIscorewas140(pc-25,-75:129.5,156.75)and146 (pc-25,-75:127,153.75) atweek30and week53, respectively (p<0.0001).

20 30 40 50 60 70

0 200 400 600 800 1000

Week 0 Week 6 Week 30 1 year

Osteocalcin [ng/mL]

beta-crosslaps [pg/mL]

bCL median OC median

Fig.1.Developmentofbeta-crosslapsandosteocalcinmediansduringinﬂiximab therapycourse.bCL:beta-crosslaps;OC:osteocalcin.

3.3. EffectofIFXonbonemarkers

AfterIFX therapy, serum osteocalcin levelsincreased signif- icantly (²=18.61, p<0.001). A post hoc analysis revealed a significantchangeintheosteocalcinlevelfromweek0toweek6 (median45.43;p<0.004)andtoweek30(median52.26;p=0.001), butnotfromweek0toweek53(median37.67).Whilesignificant changesinosteocalcinlevelswereobserved,wedidnotobserve significantchangesintheserumbCLlevels(p=0.105)atanypoint duringtheyearlongIFXtreatment(Table2andFig.1).

WhentheosteocalcinandbCLlevelsofpatientsinthecontrol groupwerecomparedwiththoseofIFX-treatedpatients,wedid notobservesigniﬁcantdifferencesatthebeginning(osteocalcin:

p=0.6;bCL:p=0.08)orendoftheyearlongIFXtreatmentregimen (osteocalcin:p=0.66;bCL:p=0.32),includingonlythosepatients whowereinremission(Table2).

3.4. VitaminDlevelsandseasonalvariabilityofvitaminD

AfteroneyearofIFXtreatment,therewerenosigniﬁcantdif- ferencesinthevitaminDlevelwhentheinitialserumlevelprior totreatmentwascomparedtotheendpointserumlevel(p=0.099;

Table2).

TheratioofvitaminDdeficiencytovitaminDinsufficiencywas improvedafterthecourseofIFXtherapy.Theproportionofpatients withavitaminDdeficiencydecreasedfrom57.4%attheonsetof thestudyto40.0%afteroneyearofIFX.Whenthestudybegan,18%

ofpatientshadaserumvitaminDlevel>29ng/mL,andthisper- centageincreasedto22%byweek53(Fig.2).Inthecontrolgroup, 25%ofpatientshadavitaminDdeﬁciencyand21%hadvitaminD level>29ng/mL(Table2).WhenvitaminDlevelswerecompared

0%

20%

40%

60%

80%

100%

week 0 1 year

Adatsor4

>29 ng/ml 20-29 ng/ml

20 ng/ml

≤

Fig.2.StratiﬁcationsofvitaminDlevelsatbaselineandafteroneyearofinﬂiximab therapyinpaediatricpatientswithmoderatetosevereCrohn’sdisease.

(4)

Table2 Medianscoresofserumosteocalcin,beta-crosslapsandvitaminD,andthesigniﬁcanceresultsofaFriedmananalysis. NumberofpatientsWeek0 Median(pc-25,-75)Week6 Median(pc-25,-75)Week30 Median(pc-25,-75)Week53 Median(pc-25,-75)Signiﬁcance(Friedman’s)Controlgroup Median(pc-25,-75) Osteocalcin(ng/mL)n=3331.82(19.8–52.2)45.43(31.2–63.9)52.26(34.7–78.9)37.67(22.7–59.7)p<0.000156.9(38.2–96.7) Beta-crosslaps(pg/mL)n=39674(514–1046)723(442–1163)821(497–1205)554(407–863)p=0.105258(235–640) Vitamin-D(ng/mL)n=3918.30(12.4–25)22.50(18.3–26.7)20.10(14.9–24.2)21.95(15.8–27.4)p=0.09924.2(20.2–27.6) pc-25,-75=percentile25andpercentile75.

Table3

EffectofinﬂiximabtherapyonbonemineraldensityinchildrenwithCrohn’sdisease.

Week0 Mean(±SD)

1year Mean(±SD)

Signiﬁcance n=29 Lumbar2-4Z-score −0.779(±1.233) −0.728(±1.566) p=0.985 BodyBMDZ-score −0.827(±0.137) −0.872(±0.138) p=0.155 BMD:bonemineraldensity.

betweenthecontrolandIFXpatientsinremission,nosigniﬁcant differenceswereobservedatthestudyonset(p=0.093)orstudy conclusion(p=0.042).

Seasonal variability wasassessed to determineif vitamin D levelswereinfluencedbythewinterorsummerperiod.Indeed,sig- nificantdifferencesintheserumlevelofvitaminDwereobserved betweenthesummerandwinterperiodatweek0(p=0.039).After oneyearofIFXtherapy,however,thissignificancewasnolonger apparent(p=0.426).ThemedianvitaminDvalueswere16.5ng/mL (pc-25,-75:10.3,20.77)and20.6ng/mL(pc-25,-75:14.15,27)at week0,and18.1ng/mL(pc-25,-75:14.85,24.7)and23.6ng/mL (pc-25,-75:17,28.1)atweek53inthewinterandsummergroups, respectively.Atthebeginningofthestudy,69%ofpatientswere severelydeficientand19%ofpatientswereinsufficientinthewin- terperiodgroup,whilethe48.6%ofpatientsweredeficientand 28.6%wereinsufficientinthesummerperiodgroup.Weobserved animprovedratioofthevitaminDdeficiencyandinsufficiencyin bothgroupsafteroneyearofIFXtreatment(Fig.3).

3.5. EffectsofIFXonbonemineraldensity

PriortoinitiationofIFXtreatment,18.3%ofthechildreninthe IFXgrouphadareducedlumbarBMDZ-score(BMDZ-score<−2.0).

In29/50cases,repeatedDEXAresultswereavailable.Whenthis subsetofpatientswasanalysed,therewerenosigniﬁcantchanges inBMDZ-scoresovertime(lumbarandtotalbodyBMDZ-scores werep=0.985,andp=0.155,respectively)(Table3).

3.6. Resultsofthecorrelationanalysis

ASpearmancorrelationanalysisindicatedaweaknegativecor- relationbetweenthevitamin Dlevel and thePCDAI atweek0 (=−0.303) andat week53 (=−0.26),and between thevita- minDandosteocalcinlevelsatweek0(=−0.241)andweek6 (=−0.315).

4. Discussion

Inthisstudy,weconfirmedthebeneficialeffectofayearlongIFX therapycourseontheclinicalstatus(decreasedPCDAI,improved IMPACT-III),andthusQoL,ofpaediatricpatientswithmoderateto severeCD.Despitefavourableclinicalchanges,wedidnotobserve significantimprovementintheBMDZ-scoreafteroneyearoftreat- ment.Similarly,therewerenosignificantdifferencesinserumbCL levelsorvitaminDlevels.TreatmentwithIFX,however,wasasso- ciatedwithasignificantincreaseinserumlevelsofosteocalcin.

Thegoalofthisstudywastodeterminewhetherbonebiomarker levels and BMD were affected by one year of IFX therapy in children with moderate tosevere CD. Osteocalcin, a markerof bone osteoblast activity,improvedsignificantly over thecourse ofIFXtherapy,withtheexceptionofweek53.SerumbCLlevels didnotchangesignificantly.Whiledecreasesindiseaseactivity (lowerPCDAIscore)couldgiverisetofavourablechangesinbone metabolism,ourcorrelationanalysisdidnotsupportthisrelation- shipinthepresentstudy.Whencomparedwiththecontrolgroup, therewerenosignificantdifferencesinbonemarkerlevelsbetween thetwo therapytypes (IFXvs.standard). Futurestudiesshould

(5)

0%

20%

40%

60%

80%

100%

Winter, week 0 Winter, 1 year Summer, week 0 Summer, 1 year

>29 ng/ml 20-29 ng/ml

20 ng/ml

≤

Fig.3.StratiﬁcationsofvitaminDlevelatbaselineandafteroneyearofinﬂiximabtreatmentwithregardtoseasonoftreatmentinitiation.

investigatehowBMD,osteocalcinandbCLareaffectedby“deep remission”tounderstandtherelationshipofbonehealthtoCD.In adultCDpatients,Mihellerandcolleaguesalsoobservedasigniﬁ- cantincreaseinosteocalcinlevelsafterIFXtherapy[24].Similarto ourstudy,nosigniﬁcantchangesinbCLlevelswereobserved[24].

InasubsetoftheREACHstudy,boneformationmarkers(serum bonespecificalkalinephosphataseandN-terminalpropeptideof type1collagen)and resorption(urineC-telopeptideofcollagen cross-linksanddeoxypyrodinoline)changedsignificantly,though thisresultincludedonlytheinductiontherapyperiodandnolong- termdata[25].Whilearecentstudyreportedsignificantchanges intheanthropometryqualityofchildren throughoutaoneyear IFXcourse,theydidnotobservechangesinbiomarkersofbone metabolismofBMD[26].

Inourstudy,BMDZ-scoreswerenotsigniﬁcantlydifferentafter oneyearofIFXtherapy.Areductioninbonemineralmassperage groupwasdetectedin18%ofpatientsattheonsetofthestudy.

InanItalian study,Paganelliandcolleaguesreportedthat31.4%

of theCD patientsin the study had a low BMD (Z-score≤−2) [7].Anotherpopulation-basedstudyinchildrenandadolescents withIBDshowedthatalmost50%ofpatientshaddecreasedBMD (Z-score≤−1),whileapproximately25%ofthestudypopulation hadalowBMD(Z-score≤−2)[27].

Ourresults,aswellasdatafromtheliterature,highlightthe importance of routine patient follow up and improvement of bonemetabolism,sinceimprovementofBMDcannotbeassumed.

Becausepeakbonemassisachievedduringchildhood,ﬁndingan optimaltreatmenttomaintain/improvebonehealthisessentialin paediatricCDpatients.

BecausevitaminDplaysrolesinimmunesystemregulationin additiontoitsrolesinbonehealthandhomeostasis,monitoring vitaminDlevelsshouldbeanimportantaspectinthetreatment ofCD[28,29].Thus,ensuringpatientshaveadequateserumvita- minDlevelsmayfurtherincreasetheefficacyoftreatmentinIBD, asa previous studyrevealed [30]. A recentlypublished review reportedthatlow vitaminDlevelswereaglobalprobleminall agegroups[16].InasurveyfromtheUnitedStates,35.8%paedi- atricIBDpatientshadaserumvitaminDconcentration≤15ng/mL [13].Similarly,Levinetal.reportedthat19%ofchildrenwithIBD hadavitaminDdeficiencywhile38%hadaninsufficiency[31].In ourstudy,57%and25%ofthepatientshadvitaminDdeficiency andinsufficiency,respectively,priortoIFXtherapyinitiation.Our resultshaveahigherincidenceofdeficiencyandinsufficiencydue tothethresholdselectedforourstudyversusthethresholdapplied tothepreviousstudies[22].

Little datais currentlyavailable onthe correlationbetween IFXandvitaminDlevels.ThestudyfromPichlerandcolleagues reported a signiﬁcant increase in vitamin D level as result of IFX therapy [26].Yet a recentstudy foundthat 1,25-dihydroxy

vitaminDserumconcentrationswereincreasedwithoutconcomi- tantchangesin25-hydroxyvitaminDafterIFXinductiontherapy.

Thestable25-hydroxyvitaminDlevelsindicatethattheincreases in1,25-hydroxyvitaminDlevelswerenotduetoimprovedgut absorptionofvitaminDbutrathermaybeduetothesuppressive effectofinﬂammationonrenalactivationofvitaminD[32].

WhenIFX-treatedCDpatientsinremissionwerecomparedwith thecontrolgroup,clinicalremissiondidnotsigniﬁcantlyimprove serumvitaminDlevels.Futurestudiesshouldinvestigatewhether patientsin‘deepremission’showdifferentcorrelationswithvita- minDlevels.

ThesuggestedvitaminDintakeinhealthychildrenis600U/day.

TheEndocrineSocietyrecommendsa doseof2000U/dayforat leastsixweeksinpatientswithadeficiency.Inthepresentstudy, patientsreceiveda1000U/dayvitaminDsupplementationfor≥15 months.Infuturestudies,ahigherdose(2000U/day)mightberec- ommended,thoughin inflammatoryconditions suchasIBDthe vitaminDlevelisinfluencedbyseveralmechanisms(e.g.intestinal absorption).Thus,theeffectofanincreaseddoseonserumvitamin Dlevelsisdifficulttopredict.

WhenpatientvitaminDlevelswereanalysedwithrespectto therapy initiationseason,patientswhoinitiated IFX duringthe summermonthshadasigniﬁcantlyhighervitaminDlevel.Atbase- line,22%ofpatientsthatinitiatedtreatmentduringthesummer seasonhadvitaminDserumlevels>29ng/mL,ascomparedto11.5%

of patientswhobegan treatmentin thewinter months.Severe vitaminDdeficiencywasobservedinmorethan69%ofpatients in thewinter groupand nearly50% of patientsinthesummer groupattheonsetoftreatment.WhilevitaminDlevelsdidnot changesignificantlyafteroneyearofIFX,theratioofvitaminD deficiency(<20ng/mL)decreasedregardlessoftheseason.These findingssuggestthat IFXmayreducetheseasonalvariabilityof vitaminDlevelsduringtreatment.Inthewintergroup,vitamin Dlevelsdidnotdecrease,and weobservedanincreasingtrend duringIFXtreatment.Althoughsignificantlongtermimprovement in vitamin D levels wasnot observed after IFX administration, decreasesinvitamin Dlevelsduring thewinterseasondidnot occur,thoughthenatureoftherelationshipbetweenvitaminD andIFXisnotyetunderstood.Previousstudieshavealsoreported aneffectofseasonalityofvitaminDlevelsinIBD.McCarthyand colleaguesfoundlowervitaminDconcentrationsinCD patients than in control subjectsduring both seasons. Moreover, lower vitaminDconcentrationswereobservedduringlatewintercom- paredtolate summerbothin CDpatientsand healthycontrols.

TheauthorsalsofoundthatpatientsreceivingvitaminDsupple- mentationdidnothaveseasonalvariation[17].Incontrast,other studieshavenotidentiﬁedasigniﬁcantassociationbetweenvita- minDlevelandseasonalityinpatientswithulcerativecolitisorIBD [33].

(6)

Insummary,IFXtherapydemonstratedaclearclinicalbenefit inpaediatricpatientswithmoderatetosevereCD.Moreover,the additionofIFXtherapyimprovedbonehealthinitially,thoughthis beneficialeffectwasdiminishedafteroneyearoftreatment.More- over,despitesupplementaladministrationofthevitaminDand calcium,lowvitamin DlevelsinCDpatientspersisted.Seasonal variabilityofvitaminDlevelswasobservedpriortoIFXinitiation butthesedifferenceswerenolongerapparentafteroneyearofIFX treatment.WedidnotobserveanexclusiveinfluenceofIFXtherapy onbonemetabolismandvitaminDlevels.

Conﬂictofinterest Nonedeclared.

Acknowledgements

ThisstudywassupportedbyOTKAK108688,OTKAK100909, OTKAK105530,OTKAK108655,OTKAPD105361,KMR12-1-2012- 0074,andLP008/2014.

References

[1]Muller KE, LakatosPL,Arato A,et al. Incidence, Paris classiﬁcation,and follow-upinanationwideincidentcohortofpaediatricpatientswithinﬂam- matoryboweldisease.JournalofPediatricGastroenterologyandNutrition 2013;57:576–82.

[2]LovaszBD,LakatosL,HorvathA,etal.Incidenceratesanddiseasecourseof paediatricinﬂammatoryboweldiseasesinWesternHungarybetween1977 and2011.DigestiveandLiverDisease2014;46:405–11.

[3]El-MataryW,SikoraS,SpadyD.Bonemineraldensity,vitaminD,anddis- easeactivityinchildrennewlydiagnosedwithinﬂammatoryboweldisease.

DigestiveDiseasesandSciences2011;56:825–9.

[4]SylvesterFA,WyzgaN,HyamsJS,etal.Naturalhistoryofbonemetabolismand bonemineraldensityinchildrenwithinﬂammatoryboweldisease.Inﬂamma- toryBowelDiseases2007;13:42–50.

[5]RisteliL,RisteliJ.Biochemicalmarkersofbonemetabolism.AnnalsofMedicine 1993;25:385–93.

[6]VeerappanSG,O’MorainCA,DalyJS,etal.Reviewarticle:theeffectsofanti- tumournecrosisfactor-alphaonbonemetabolismininﬂammatorybowel disease.AlimentaryPharmacologyandTherapeutics2011;33:1261–72.

[7]PaganelliM,AlbaneseC,BorrelliO,etal.Inﬂammationisthemaindetermi- nantoflowbonemineraldensityinpaediatricinﬂammatoryboweldisease.

InﬂammatoryBowelDiseases2007;13:416–23.

[8]TsampalierosA,LamCK,SpencerJC,etal.Long-terminﬂammationandgluco- corticoidtherapyimpairskeletalmodelingduringgrowthinchildhoodCrohn disease.JournalofClinicalEndocrinologyandMetabolism2013;98:3438–45.

[9]HyamsJ,WaltersTD,CrandallW,etal.Safetyandefﬁcacyofmaintenance inﬂiximabtherapyformoderate-to-severeCrohn’sdiseaseinchildren:REACH open-labelextension.CurrentMedicalResearchandOpinion2011;27:651–62.

[10]DaneseS.Mechanismsofactionofinﬂiximabininﬂammatoryboweldisease:

ananti-inﬂammatorymultitasker.DigestiveandLiverDisease2008;40(Suppl.

2):S225–8.

[11]ArmuzziA,DePascalisB,FedeliP,etal.InﬂiximabinCrohn’sdisease:earlyand long-termtreatment.DigestiveandLiverDisease2008;40(Suppl.2):S271–9.

[12]IijimaH,ShinzakiS,TakeharaT.TheimportanceofvitaminsDandKforthebone healthandimmunefunctionininﬂammatoryboweldisease.CurrentOpinion inClinicalNutritionandMetabolicCare2012;15:635–40.

[13]PappaHM,GrandRJ,GordonCM.ReportonthevitaminDstatusofadult andpaediatric patientswith inflammatoryboweldisease and itssignifi- canceforbonehealthanddisease.InflammatoryBowelDiseases2006;12:

1162–74.

[14]JorgensenSP,HvasCL,AgnholtJ,etal.ActiveCrohn’sdiseaseisassociatedwith lowvitaminDlevels.JournalofCrohn’sandColitis2013;7:e407–13.

[15]BakkerSF,DikVK,WitteBI,etal.Increaseinbonemineraldensityinstrictly treatedCrohn’sdiseasepatientswithconcomitantcalciumandvitaminD supplementation.JournalofCrohn’sandColitis2013;7:377–84.

[16]PalaciosC,GonzalezL.IsvitaminDdeﬁciencyamajorglobalpublichealth problem? Journal of Steroid Biochemistry and Molecular Biology 2013.

pii:S0960-0760(13)00233-1.

[17]McCarthyD,DugganP,O’BrienM,etal.SeasonalityofvitaminDstatusand boneturnoverinpatientswithCrohn’sdisease.AlimentaryPharmacologyand Therapeutics2005;21:1073–83.

[18]HyamsJ,CrandallW,KugathasanS,etal.Inductionandmaintenanceinﬂiximab therapyforthetreatmentofmoderate-to-severeCrohn’sdiseaseinchildren.

Gastroenterology2007;132:863–73,quiz1165–6.

[19]OtleyA,SmithC,NicholasD,etal.TheIMPACTquestionnaire:avalidmeasureof health-relatedqualityoflifeinpaediatricinﬂammatoryboweldisease.Journal ofPediatricGastroenterologyandNutrition2002;35:557–63.

[20]SzaboD,KokonyeiG,AratoA,etal.Autoregressivecross-laggedmodelsof IMPACT-IIIandPaediatricCrohn’sDiseaseActivityindexesduringoneyear inﬂiximabtherapyin paediatricpatientswith Crohn’sdisease.Journalof Crohn’sandColitis2014;8:747–55.

[21]HyamsJS,FerryGD,MandelFS,etal.Developmentandvalidationofapaedi- atricCrohn’sdiseaseactivityindex.JournalofPediatricGastroenterologyand Nutrition1991;12:439–47.

[22]Holick MF,Binkley NC, Bischoff-Ferrari HA, et al. Evaluation, treatment, and prevention of vitamin D deﬁciency: an Endocrine Society clinical practiceguideline.JournalofClinicalEndocrinologyandMetabolism2011;96:

1911–30.

[23]SylvesterFA.IBDandskeletalhealth:childrenarenotsmalladults!Inﬂamma- toryBowelDiseases2005;11:1020–3.

[24]MihellerP,MuzesG,RaczK,etal.ChangesofOPGandRANKLconcentra- tionsinCrohn’sdiseaseafterinﬂiximabtherapy.InﬂammatoryBowelDiseases 2007;13:1379–84.

[25]Thayu M,Leonard MB, Hyams JS,et al. Improvementin biomarkers of boneformation during inﬂiximab therapyin paediatric Crohn’s disease:

resultsoftheREACHstudy.ClinicalGastroenterologyandHepatology2008;6:

1378–84.

[26]PichlerJ,HanslikA,DietrichHuberW,etal.Paediatricpatientswithinﬂam- matoryboweldiseasewhoreceivedinﬂiximabexperiencedimprovedgrowth andbonehealth.ActaPaediatrica2014;103:e69–75.

[27]SchmidtS,MellstromD,NorjavaaraE,etal.Lowbonemineraldensityinchil- drenandadolescentswithinﬂammatoryboweldisease:apopulation-based studyfromWesternSweden.InﬂammatoryBowelDiseases2009;15:1844–50.

[28]GargM,LubelJS,SparrowMP,etal.Reviewarticle:vitaminDandinﬂamma- toryboweldisease–establishedconceptsandfuturedirections.Alimentary PharmacologyandTherapeutics2012;36:324–44.

[29]OoiJH,ChenJ,CantornaMT.VitaminDregulationofimmunefunctioninthe gut:whydoTcellshavevitaminDreceptors?MolecularAspectsofMedicine 2012;33:77–82.

[30]NicholsonI,DalzellAM,El-MataryW.VitaminDasatherapyforcolitis:a systematicreview.JournalofCrohn’sandColitis2012;6:405–11.

[31]LevinAD,WadheraV,LeachST,etal.VitaminDdeﬁciencyinchildrenwith inﬂammatoryboweldisease.DigestiveDiseasesandSciences2011;56:830–6.

[32]AugustineMV,LeonardMB,ThayuM,etal.ChangesinvitaminD-relatedmin- eralmetabolismfollowinginductionwithanti-tumornecrosisfactor-alpha therapyinCrohn’sdisease.JournalofClinicalEndocrinologyandMetabolism 2014:jc20133846.

[33]BlanckS,AberraF.Vitaminddeﬁciencyisassociatedwithulcerativecolitis diseaseactivity.DigestiveDiseasesandSciences2013;58:1698–702.