The Leprosarium of Saint-Thomas d’Aizier : the cementochronological proof of the medieval decline of Hansen disease in Europe?

(1)

HAL Id: halshs-01715280

https://halshs.archives-ouvertes.fr/halshs-01715280

Submitted on 22 Feb 2018

HAL is a multi-disciplinary open access archive for the deposit and dissemination of sci- entific research documents, whether they are pub- lished or not. The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.

L’archive ouverte pluridisciplinaire HAL, est destinée au dépôt et à la diffusion de documents scientifiques de niveau recherche, publiés ou non, émanant des établissements d’enseignement et de recherche français ou étrangers, des laboratoires publics ou privés.

Distributed under a Creative Commons Attribution - NonCommercial - NoDerivatives| 4.0

Hansen disease in Europe?

Joël Blondiaux, Stephan Naji, Jean-Pierre Bocquet-Appel, Thomas Colard, Amélie de Broucker, Cécile de Seréville-Niel

To cite this version:

Joël Blondiaux, Stephan Naji, Jean-Pierre Bocquet-Appel, Thomas Colard, Amélie de Broucker, et al..

The Leprosarium of Saint-Thomas d’Aizier : the cementochronological proof of the medieval decline of Hansen disease in Europe?. International Journal of Paleopathology, Elsevier, 2016, 15, pp.140-151.

�10.1016/j.ijpp.2015.02.005�. �halshs-01715280�

(2)

ContentslistsavailableatScienceDirect

International Journal of Paleopathology

jo u r n al h om e p a g e :w w w . e l s e v i e r . c o m / l o c a t e / i j p p

The leprosarium of Saint-Thomas d’Aizier: The cementochronological proof of the medieval decline of Hansen disease in Europe?

Joël Blondiaux

^a,∗

, Stephan Naji

^b

, Jean-Pierre Bocquet-Appel

^c

, Thomas Colard

^d

, Amélie de Broucker

^e

, Cécile de Seréville-Niel

^b

aCentred’EtudesPalèopathologiquesduNord,36rueJulesFerry,59127Walincourt-Selvigny,France

bCentreMicheldeBouärd,CNRS,UMR6273,UniversitédeCaen,Esplanadedelapaix,14032CaenCedex,France

cCNRS,UPR214744,ruedel’AmiralMouchez,75014Paris,France

dLilleUniversity,ForensicTaphonomyUnit—Anthropology,LilleForensicInstitute,59000Lille,France

eLilleUniversity,ForensicTaphonomyUnit—Anthropology,LilleForensicInstitute,59000Lille,France

a r t i c l e i n f o

Articlehistory:

Received3July2013

Receivedinrevisedform17February2015 Accepted18February2015

Keywords:

Leprosy Survivalfunction Middleages Cementum France

a b s t r a c t

ThisstudycomparestheadultsurvivorshipproﬁlesofpeopleinterredintheSaint-Thomasd’Aizier leprosarium,estimatedbycementochronology,toeightarchaeologicalseriesinnorthernFrancedated fromLateAntiquitytotheLateMiddleAges,periodsofsigniﬁcantvisibilityforHansen’sdisease(leprosy).

Thegoalsaretounderstandtheimpactofleprosyonvarioussocialgroupsandtoexplorethecause ofleprosy’sdeclinebyanalyzingmaleandfemalefertility.Survivalratesdifferedbetweenmedieval leprosy-freesitesandtheSaint-Thomasd’Aizierleprosarium,althoughthisdifferencewasstatistically signiﬁcantonlyforthefemaleleprosariumsample.Theselectivefemalefrailty,aconsequenceofsocial exclusionandthecollapseofthequalityoflife,combinedwiththeinfertilityoflepromatouscouples, offeramulti-causalexplanationtotheendoftheexpansionandthendeclineofleprosyinsouthernand westernEuropeancountries.

1. Introduction

Demographicfactorsrelatedtoleprosyinthepasthavebeenrel- ativelyunderexplored,yetthediseaseisknowntoimpactmortality andfertility.Wethereforesuspectthatchangesindemographic parameters played a role in the decline of this disease. Clini- cal analyseshighlightthe majorrole of specific pathologies on mortalityandmorbidity,suchassuicide,tuberculosis(TB),renal insufficiency,andcancer(Tokudomeetal.,1981).Inaddition,the possibilitythatcouplesaffectedbyleprosy wouldhave alower fertility due to male infertility has been raised since the early 20thcentury(McCoy,1913).McCoywasthefirsttodiscussthe frequency oforchiepididymitis (inflammation of thetesticles) in lepromatouspatients. Thesameyear,Barbézieux(1913)argued thatthedeclineofleprosyanditsdisappearanceduringthe17th centuryinFrancewasduetothelegalrequirementthatleproma- touspersonsandtheirdescendants(i.e.,“cagots”)marryaffected

∗Correspondingauthor.Tel.:+33327827439.

E-mailaddresses:jblondiaux@nordnet.fr(J.Blondiaux),stephan.naji@gmail.com (S.Naji),jean-pierre.bocquet-appel@evolhum.cnrs.fr(J.-P.Bocquet-Appel), thomas.colard@free.fr(T.Colard),ameliedebroucker@hotmail.fr(A.deBroucker), cecile.niel@unicaen.fr(C.deSeréville-Niel).

partners,inadditiontomaleinfertility.Morerecently,Beiguelman (1967)collectedalargedemographicdatasetonleprosypatients andconcludedthatsinceorchiepididymitisisfrequentamonglepro- matousmales,manywererenderedsterile.However,inthefertile fractionof lepromatousmales,fecundity wassimilartothat of healthy males.Women’s fertility,however, seemedtobe unaf- fected,asdemonstratedbythecomparablespontaneousabortion rates observedbetweenlepromatousand healthyfemales. Lep- rosyinducedmaleinfertility anda possibleincreasedmortality ofmalesandfemalesduetoleprosyandotherpathogensshould thereforebeexploredtoassesstheirrespectiveeffectsonthedis- ease’s17thcenturydecline.Onesourceofhistoricalinformation isrepresentedbybioarchaeologicaldata.However,thepaucityof demographic analysesfrom archaeological sources, suchas the EnglishmedievalleprosariainChichesterandWinchesterinEng- land(Magiltonetal.,2008;RoffeyandTucker,2012;Tayloretal., 2013)and Næstved in Denmark (Bennike et al., 2005; Møller- Christensen,1953)limitsourunderstandingofthedemography dynamics of Hansen disease (i.e., leprosy) in Medieval Europe.

Withoutthisinformation, anyconclusionsaboutthe“declineof leprosy”inEuropeattributedtotheselectiveimpactoftuberculo- sis(Donoghueetal.,2005)ortothesecondaryeffectsofthe14th centuryBlackDeath(Bassukasetal.,2012;Lechat,2002)appear premature.

(3)

Toexplorerelationshipbetweenmortality,fertility,andpale- odemography, a precise adult age-at-death estimation method is required. The only available technique that combines accu- racyandreliabilityiscementochronology,usuallyreferredtoas the“TCA”count.In thepast 30 years,severalresearchers have developedandappliedthistechniqueforhumanage-at-deathesti- mations bycountingthe cementumincrementsof dental roots (Charlesetal.,1986;KagererandGrupe,2001;Solheim,1990;Stott etal.,1982;Wittwer-Backofenetal.,2004).Recentmethodological improvements,bothforproducinghistologyslidesandcounting thecementumincrements,haveresultedinasigniﬁcantincrease intheaccuracyandcontrolsforinter/intra-observercounterror (Blondiauxetal.,2006;Colardetal.,2015;GochaandSchutkowski, 2013;RobbinsSchugetal.,2012;Wedel,2007;Wittwer-Backofen, 2012; Wittwer-Backofen et al., 2008). The cementochronology approach is the only method that allows for the direct count- ingofagrowingtissue(cementum)withoutstatisticalcalibration, Bayesianorotherwise,ortheneedforareferencepopulation.It isonlyrestrictiveinthesensethatitrequiresthepresenceofone toothperindividualwithsuitablepreservation,arequirementsim- ilartothatfor anyotherbiologicalindicator(Najietal., 2014).

Becausewenowhaveatooltoestimatetheage-at-deathofarchae- ologicalskeletalserieswithprecision,paleoepidemiologists and paleodemographershavetheopportunitytoinvestigatethedif- ferentialimpactofvariouspathogensonhistoricandprehistoric serieswithouttheimpedimentofsomeoftheosteologicalpara- dox’sissuesconcerningselectivemortality(Woodetal.,2002).

Theﬁrst objective of this study is to compare demographic proﬁlesofskeletonsfromaFrenchmedievalleprosariumtoeight archaeologicalseriesdatedfromLateAntiquitytotheLateMiddle Agesbysexand byage,estimatedthroughcementochronology.

Theseareallfromageographicallyandpoliticallyrestrictedareaof northernFranceandhavebeenarchaeologicallyandhistorically characterizedin termsof socioeconomiclevels. Thiscontextual backgroundisessentialtointerpretthedifferencesinsurvivorship andallowsustocomparethesamplesbysexandeconomicstatus.

Inaddition,apaleopathologicalcharacterizationoftheskeletons willservefordiscussingtheevolutionanddeclineofleprosyduring theLateMiddleAges.

Therefore, a second objective is to test the hypothesis put forward by Donoghue et al. (2005) that the co-occurrence of tuberculosisandleprosy,associatedwithdepressedsocioeconomic factors,increasedearlymortalityinlocalpopulationsandthuspre- cipitatedthedeclineofleprosy.Sincewehaveidentiﬁedindividuals withpotentiallesionsduetobothleprosyandtuberculosisattwo sites(AizierandArras),wealsohavetheopportunitytodirectly testtheimpactoftheco-occurrenceofM.lepraeandM.tubercu- losisversustheimpactofTBaloneontwodistinctarchaeological samples.

Usingabioarchaeologicalapproach,thatis,usingcombinedhis- torical,archaeological,paleopathological,anddemographicaldata, athirdobjectivewillbetodiscussthequestionofmaleinfertility andelevatedfemalemortalitywithintheleprosariumtoexplore thepathocenosesthatmayhavecontributedtothedeclineoflep- rosyattheendoftheMiddleAgesinEurope.

Ourfourthandﬁnalobjectiveistoconsidertheproposalthatthe conﬁnementofvariouspopulationsinmedievalEuropepreceded andaccompaniedthedeclineofleprosyandmightbelinkedtoits disappearance.

2. Materialsandmethods 2.1. Material

Nineosteologicalseries(Fig.1)housedattheCenterforPale- opathologicalStudies(Walincourt-Selvigny,France)werestudied.

Theyrepresentatotal of1701adultskeletonsdating fromLate Antiquity(450AD)totheLateMiddleAges(1300AD)(Table1).

Theseserieswereselectedbasedonextensiveexcavationsofeach siteandoverallgoodbonepreservation.Becausethearchaeological contextwassystematicallyevaluatedforeachsite,weanticipateno majortaphonomicbiasesandthusconsidertheseseriestoberepre- sentativesamplesoftheregionalpopulation.Fromtheseskeletons, one permanent tooth, mostly monoradicularand anterior, was sampledfromeachof842adultswithobservableteeth.Cemen- tochronology was then used to estimate age-at-death and the correspondingsamplessurvivorshipcurves.Thesampleofadult individualswithidentiﬁedleprosycomesfromthemedievallep- rosariumofSaint-Thomasd’Aizier(STA;Haute-Normandy,France), datedfrom1150to1550AD(Fig.2).Therestoftheseriesrepresents thesampleswithoutobservedleprouslesions.

2.1.1. Saint-Thomasd’Aizierleprosarium

ThetotalSTAsampleiscomprisedof186(88.3%)adultsandonly 24(11.6%)subadultsidentifiedinsitu,eitherasindividualburials orascommingledremains.Eightwereburiedinthesameaisleof thechapel,fiveareinthenorthcemetery,twoareinthesouthern sector,andoneteenagerwasinasingletomblocatedatthechevet ofthebuilding.Itisnoteworthythatthegravesoftheyoungest weregroupedincloseproximityalongthenorthwallofthenave orundertheeaves(substillicidio).Theyprobablybenefitedfroma privilegedlocationinthecemetery,whiletheburialsofolderchil- dren,inthemostnortheasternarea,appearmoredispersedamong theadultburials.Overall,theexcavationandtaphonomicstudyof thecemeterydidnotrevealanyburialswithmorethanoneindivid- ual.Sexestimationbasedonthemorphologyandsizeofthepelvis (Bruzek,2002;BuikstraandUbelaker,1994)waspossiblefor125 individuals,includingtwoolderteenagers.AtAizier,70malesare presentalongwith55females(sexratio=1.27).Amalepredom- inancewasobservedforindividualsburiedinsidethechapel(12 malesfor3females).

2.1.2. Comparativesites

Eightsites,totaling1519adults,makeupthecomparativesam- ple.Archaeologicalcontextualinformationwasusedtoestimate thesettlementtype(rural,urban,andreligious)andtoassessthe socialstatus(wealthyorpoor)(Table1).

1.Theurbansiteofthe“ÎlotdelaBoucherie”inAmiens(France)was excavatedbetween2006and2007.This5200m²siterevealed thepresenceofa vastcemeteryaroundanamphitheaterand contained277burials. Thesystematic studyof theskeletons andassociatedartifactsofferedadetailedpictureofasigniﬁcant sampleoftheAmienspopulationduringLateAntiquity(290–395 AD).Availableinformationincludesoverallhealthassessments aswellasthesocialandpossibleethniccompositionofthepre- dominantlyyoungmaleworkersandmilleteatersofGermanic origin(Binet,2012a,b;Blondiaux,2012).

2.The Arras-Prefecture site is strictly contemporary with STA (1200–1300 AD) and is composedof lower-class individuals buried around thechevet of the medievalcathedral of Arras (Jacqueset al.,1988).Thisurbancemeteryyielded 81 skele- tons,including65completeandwell-preservedadults.Acluster ofﬁveburialscontainedﬁveskeletonswithboneandjointTB (Blondiauxetal.,1992).

3.“LesRuesdesVignes”isaMerovingiancemetery(480–680AD), with269 burials having been excavatedfrom 1980 to1982 (Blondiaux,1986).Sexdiagnosishasbeenconﬁrmedon64.5%

oftheadultsample(62malesand69females).Atleastthree casesofboneandjointTBhavebeenrecorded.

(4)

Fig.1.LocationofthetownofAizier(Eure)inHaute-Normandy,France.

4.ThesiteofLisieux-Michelet(866burials),datedfrom280to750 AD,waspartofawealthyGallo-Romancitywithanimportant militarygarrison(Paillardetal.,2006).

5.Twenty burials were excavatedfrom the choir of the abbey church of Saint-Pierre-sur-Dives (Normandy) in 1997. These upper-classburialsdatefromthe12thand13thcenturiesADand consistedof19maleskeletonsandonlyonefemale(Alduc-Le Bagousseetal.,2004).

6.The site of the “Marmousets” at Vendeuil-Caply (France) has beenarchaeologicallyinterpretedasaruralsite,eventhough itscemeteryislocatednearvastculturalandreligiousbuildings (vicus)(Piton,2009).Twocontiguousburialareasrepresenttwo chronologicalphases.Theearliestarea(178 burials,250–380 AD)only yielded a few gravegoods.Within this area, there is a distinct cluster of individuals with lesions of probable

tubercularorigin.Themorerecentarea(45burials,500–600AD) wasrepletewithgravegoods.Withinthissecondarea,therewas aclusterof7individualswithbonelesionssuggestiveofleprosy (acro-osteolysis,corticalresorption,tarsaldestruction) 7.Vronwasexcavatedfrom1971to1975andisdatedfrom308

to700AD(Seillier,2006).Theinitialcomponentofthesiteisa GermanicsettlementontheLitusSaxonicum.Thesiterevealed 219individuals,amongwhichhalfwerepositivelydiagnosedas males(41),females(36),andsubadults(38).NosignsofTBhave beenobserved.

8.Wandignies-Hamage is a female monastic community dated from the 7th to 11th centuries AD (Louis and Blondiaux, 2009).Aclearsegregationofsexandsocialorigins ismarked withinthecemeteries,bothinsideandoutsidethecommunity buildings.

Table1

Summarystatisticsforthesitesusedinthestudywiththenumberofteeth(842)usedfortheageestimationbycementochronologyrepresenting842adults(49.5%).

N=number,LA=lateantiquity,LMA=latemiddleages,EMA=earlymiddleages.

Sites Skeleton(N) Adult(N) Sub-adult(N) Tooth(TCA)(N) Periods Context Socialstatus Sources

AmiensBoucherie 277 148 129 147 LA3rd–4thc. Urban,workers Poor Binet(2012a,b)

ArrasPréfecture 81 65 16 30 LMA10–12thc. Urban Poor Jacquesetal.

(1988)

LesRuesdesVignes 269 203 66 127 EMA6–7thc. Rural Rich Blondiaux

(1986)

LisieuxMichelet1 648 458 190 94 LA4thc. Urban Rich Paillardetal.

(2006)

LisieuxMichelet2 218 155 63 35 EMA6–7thc. Urban Rich Paillardetal.

(2006)

SaintPierresurDives 20 20 0 17 LMA12–13thc. Rural Rich Alduc-Le

Bagousseetal.

(2004)

SaintThomasd’Aizier 206 182 24 63 LMA12–15th Leprosarium Poor Nieletal.

(2007)

Vendeuil-Caply1 178 115 63 90 LA3rd–4thc. Rural Poor Piton(2009)

Vendeuil-Caply2 45 35 10 30 EMA5–6thc. Rural Poor Piton(2009)

Vron1 116 92 24 46 LA4thc. Rural,migrants Poor Seillier(2006)

Vron2 47 41 6 39 EMA5c. Rural Rich Seillier(2006)

Vron3 56 48 8 37 EMA6–7thc. Rural Poor Seillier(2006)

Wandignies-Hamage 167 139 28 87 EMA6–9thc. Nunnery Rich Louisand

Blondiaux (2009)

Total 2328 1701 627 842

(5)

Fig.2. ArchaeologicalsiteofthemedievalleprosariumofSaint-Thomasd’Aizier (photoby.C.deSeréville-Niel).

2.2. Methods

Adultmedianage-at-deathandsurvivorshipcurvesestimated bycementochronologyarecomparedtoassessthemortalitydif- ferencesamongthearchaeologicalsamplesandsubsamples.The sexestimationsfortheskeletonsusedtheBruzek(2002)method ortheBuikstraandUbelaker(1994)standards,dependingonthe skeletalpartsavailableforobservation.

2.2.1. Individualadult-ageestimation:Cementochronology

Age estimationswere made bycementochronology. Cemen- tumisthe interfacebetweentheperiodontalligamentand the rootofthetooth,and itservesfor theattachmentofthetooth in the alveolar bone (Hillson, 1996). This histological method, developed in animal biology (Linhart, 1973) and later applied tohumanremains(Charles etal., 1986;Stottet al.,1982), and requiresobservations of toothroots cross-sectionsfor alternat- ingdarkand clearbandingincrements,which arevisible when thinsectionsaremicroscopicallyviewedunderdirecttransmitted light(Wittwer-Backofenetal.,2004).AsdescribedbyBlondiaux and colleagues(Blondiaux et al.,2006), sectionswereobtained frommonoradicularteeth(mostlypremolars)andembeddedin anepoxyresin(AralditeBrot).Theresultingblockwastrimmedto reducecuttingtime.Eighttransversesectionsof80–100␮mwere thenobtainedforeachtoothusingamicrometriclow-speeddia- mondsaw(BuehlerIsomet).Unpolishedsectionsweremounted usingCanadianbalsam.Eachfieldwasobservedinnormal light at200×magnificationonaNikonLabophotmicroscopeequipped withadigitalcamera.Twentymicroscopicfieldspertoothwere selectedand photographed.Theimageswerethendisplayedon acomputer,andthecementumincrementswerecountedutiliz- ingAdobePhotoshopsoftware.Themodalaveragecountsadded to the averagedate of tooth eruption (Liversidge et al., 1998) wererecordedonanautomatedspreadsheet,whichincludedthe standard errorspecific tothereference sample (softwareby B.

Bertrand,ArchaeologicalServiceCAD,Douai,France).

To control for biases due to alveolar bone destruction in Hansen’s disease, Magniez (2012) compared the frequency of dentalconditionsof theSTAadults withthose ofthecompara- tiverural,urban,richandpoorpopulations.TheSTAoralhealth didnot seem to differ from the rural and poor samples,with the exception of teeth exposed by alveolar resorption due to

rhinomaxillary syndrome (Andersen and Manchester, 1992).

Therefore,theanteriormaxillaryteeth(whenpreserved)werenot usedforcementochronologyatSTA.Mostofthepremolarsdidnot exhibitadditionalmicroscopiccementolysisrelativetotheother archaeologicalteethusedinthisstudy.Thisassessmentisofimpor- tancewhendealingwiththeantemortemerosionandpostmortem diagenesisoftoothcementum.

2.2.2. Survivalfunctions

Thestatisticalsurvivorshipfunctionappliedtotheageatdeath datawastheKaplan–Meiertechnique(KaplanandMeier,1958;

SPSS software). Kaplan–Meier-estimated survivorship functions area seriesofhorizontalstepsofdecreasingsizethatallowan approachtotheactualsurvival inthepopulationwhen asufﬁ- cientlylargesampleisused.Thevalueofthesurvivorshipfunction betweensuccessivesamplesisconsideredconstant.Ifnotruncation orcensoringisreached,theKaplan–Meierestimateisequivalentto theempiricaldistributionfunction.

Thelogrank-test(Mantel–Cox)statisticcomparestheestimates ofthehazardfunctionsfortwogroupsateachobservedevent.It isconstructedbyﬁrstcomputingtheobservedandexpectednum- berofeventsinoneofthegroupsforeachobservedevent,and thenaddingthesetoobtainanoverallsummaryacrossalltime pointswherethereisanevent.Thesurvivorshipcurveshavebeen traced according tothe standardized survivorship (cumulative) commonlyusedforthestudyofsurvivalrates,andtheconventional p-valueof0.05wasusedtorejectthenullhypothesisoftherebeing nodifferencebetweentwogroups.

2.2.3. Pathology

The diagnosisof leprous lesions uses Andersen, Manchester and colleagues’ criteria for rhinomaxillary syndrome, tibiofibu- lar periosteal lesions, infective lesions of the metatarsals and metacarpals,concentricdiaphyseallesionsofthemetatarsalsand phalanges,manualphalangealgrooves,andcupandpeginterpha- langealjointlesions(AndersenandManchester,1992;Andersen etal.,1994).Ofthe43skeletonswithfacialbonespreservedinSTA, 34(76.4%)presentatleasttwocriteriaofrhinomaxillarysyndrome (Fig.3),thushavingoneofthehighestproportionsoflepromatous casesstudiedinFrancethus far(Colardet al.,2011).Regarding theperiostealreactionofthelowerlimbs,a radiographicstudy associatedwithmicroscopicverificationwasappliedto35pairs oftibiaeandfibulae.Ofthese,85%(31pairs)hadradiographically visibleappositionsofnewboneofvariablethicknessesandexten- sion,whileonly55%weremacroscopicallyvisible.Ofthefourpairs

“free”ofmacroscopicorradiographiclesions,twoshowedremod- eledappositiononmicroscopicsections(Rufﬁnetal.,2011).The thicknessandextentofperiostealappositionwaspositivelycorre- latedwithage,butrhinomaxillarysyndromewasnot(Rufﬁnetal., 2011).

LesionscharacteristicofdisseminatedTBwereincludedinour samples to examine leprosy and TB covariance. These lesions include lumbar and thoracic vertebral cavitations, remodeled periostealappositionsofthepleural surfacesof theribs, hyper- trophicpulmonaryosteopathyandboneandjointlesions.Evenif someoftheselesionsarenon-speciﬁc,webelievethatitisvalidto usethemasexpressionsofTBwithnecessaryrestrictions(Nicklisch etal.,2012).Sevenskeletonsexhibitedremodeledperiostealappo- sitionsofthepleuralsurfacesoftheribs, possiblyrelatedtoTB.

ResearchlinkingsuchribsandatypicalspinallesionstoTBisincon- clusive,however,possiblybecausetherearefewcasesconﬁrmed withancientDNA(aDNA)andmycolipidsanalyses,butalsobecause thereisalackofprecisestudiesofradiographs,anatomicpathol- ogy,orpre-antibioticclinicallyconﬁrmedcases(Baker,1999;Haas etal.,2000;Pfeiffer,1984;SantosandRoberts,2006;Spekkeretal., 2012).

(6)

Fig.3. Saint-Thomasd’Aizier(#904),male,aged47.3±1.8yrs,frontalviewofthe rhino-maxillarysyndrome.

Aspinesample(STA642)withlumbarandthoracicvertebral cavitations(Fig.4)wasassociatedwithcribrafemorisandcribra orbitalia.Vertebralcavitationsarepresentassmooth-walledcir- cumferentialresorptiveslesionsofthoracicandlumbarvertebral bodies(Baker,1999).Theextensiveandirregularpittingisasso- ciatedtosuperﬁcialremodelingandporosities.Cribrafemorisor Allen’sfossa,isa lesionformedbyatrophyofthecorticalbone locatedsymmetricallyontheanteriorsurfaceofthefemoralneck

reminiscentofaninflammatoryresponse(Finnegan,1978).Cribra orbitaliawereidentifiedasclustersoffiveormoremicro-sieves locatedontheceilingoftheorbits. Additionaldescriptionsand scoringforthelesion’sdegreeandactivityfollowedBuikstraand Ubelaker(1994).

Aradiographicstudyofa19-year-oldsubadultvertebralbodies (STA642)clearlyrevealedcenteredtranslucentzonessurrounded bydenseandmostly irregularareas.Theselesionsaredifferent fromtheresidualsinusvenousgroovescommonlyseeninjuve- nilesinthecentralanteriorportionofthevertebralbody(Keats andAnderson,2012)anddescribedasHahncleftandtransverse growthrecoverystresslines(Freyschmidtetal.,2002).Thecom- parisonofthe“cavitations”andtopographyofthevertebralwalls ofthisindividualtodescriptionspublishedfromautopsysamples (Ménard,1888)andtoapictureoftubercularcaries(Fig.5)suggests thatthisadolescentcouldcombinebothHansen’sdiseaseandan earlystageof“tubercular”caries.Inhispublication,Ménard,1888 classifiedthislesionasasuperficialandearlystageoftuberculosis, verydifferentfromPott’sdisease,whichwouldbeamorelocalized, destructive,andchronicformofTB.Thisdescriptionislaterusedby SorrelandDejerine-Sorrel(1932)as“diffuseandsuperficial”,and oftenlethalinchildrenandadolescents.Freyschmidtetal.(2002) describethismultiplevertebraltuberculosisassuperficialanterior tuberculosisspondylitis.Thiscasecouldillustratethecoexistence ofthetwomycobacteriainfections,M.lepraeandM.tuberculosis, andconfirmthedeleteriousimpactofTBinanindividualcarrying thestigmaoflepromatousleprosy.AbonesampleofSTA642taken byMarkSpigelmanhasbeensubmittedtoaDNAanalysistoHelen Donoghue’slaboratoryinLondon.

3. Results

3.1. Adultsurvivalbysiteandsex

Thereisadeﬁnitehierarchyofsurvivorshiplevelsbasedonthe socialbackgroundofeachsampleasestimatedfromitsarchaeo- logicalcontext.Amongthefemalesamples(Fig.6andTable2),the

Fig.4. SaintThomasd’Aizier;(1a)–(1d)#642,female,aged19yrs,rhinomaxillarysyndromewithperforationofthepalate,vertebralcavitations(anteriorandlateralright, withradiographylateralright).

(7)

Fig.5. CopyofPlate11fromVictorMénard’sbook(1888),drawnbyKarmanski.The pictureshowsvertebralcavitationsinafiveyear-oldchildwhodiedfromtubercu- losis.Theassociatedcommentstates:“wecanseeonthesurfaceofthevertebrae erosions,smallulcerations,superficialcavities,someempty,somefulloffongosity andpus[...].Undertherightparietalpleurathereisaprominenttubercularabscess symptomaticofanalterationofthethirdrib.”Thissuperficialformdescribedby MénardisvisiblymoreadvancedthanforSTA-482.

highestsurvivorshipcurveisinferredforthearistocraticnunsfrom themonasteryofHamage.Themiddlesurvivorshipcurvesarefrom Vron(rural)andtwourbanfemaleseriesinAmiens.Thearistocratic religioussitehasthehighestmeanage-at-death(MAD)andthelep- rosariumthelowest,whilethethreeruralsecularcemeteriesare intermediate.However,allthesitesexceptHamagepresentsimilar survivalpatternsduringtheﬁrstdecadeofadultlife.

Thesurvivorshipcurvesfortheeightmaleadultseries(Fig.7 andTable3)followasimilarpattern.Thelowestsurvivalatallages

Table2

Summarystatisticsforadultfemalesurvivalcurvesbysite.MAD=meanage-at- death;CI=conﬁdenceinterval±95%.

Females Teeth(N) MAD CI−95% CI+95%

St-Thomasd’Aizier 25 31.9 26.9 36.9

Amiens 55 34.7 31.7 37.7

Vron 43 40.1 33.8 46.4

Lisieux 31 46.8 33.7 49.5

Hamage 58 57.1 53.9 60.3

Fig.6. Survivorshipcurvesof5adultfemalegroups.

Fig.7.Survivorshipcurvesof8adultmalegroups.

before50isfromSTAleprosarium,whilethehighestsurvivorship involvesthetwoaristocraticchurchburialsites(StPierre-sur-Dive andHamage),withMADsof65and59.4years,respectively.The othersixruralandurbanmaleseriesfollowvarioussurvivorship patterns,dependingontheirsocioeconomiclevels,similartothe

Table3

Summarystatisticsforadultmalesurvivalcurvesbysite.MAD=meanage-at-death;

CI=conﬁdenceinterval±95%.

Males Teeth(N) MAD CI−95% CI+95%

St-Thomasd’Aizier 26 39 35 43

Amiens 92 40.9 37.5 50.3

Vron 42 47.3 43.7 50.9

Lisieux 72 49 46 51.9

Hamage 21 59.4 43.1 75.7

Vendeuil 64 52.6 49.1 56.1

RuedesVignes 42 57.3 51.7 62.9

StPierresurDive 16 65 51.3 78.7

(8)

Table4

Summarystatisticsforadultmaleandfemalesurvivalcurves.MAD=meanage-at- death;CI=conﬁdenceinterval±95%.

Teeth(N) MAD CI−95% CI+95%

Amiensmale 92 39 36.6 44

Amiensfemale 55 33.8 30.7 36.8

Vronmale 42 46.8 42.6 51

Vronfemale 43 40.1 34.8 45.4

STAmale 26 39.3 30.55 48.1

STAfemale 25 29.4 24.48 34.32

STAindeterminate 12 38.9 32.3 45.5

STAtotal 63 35.3 32.6 38

Arrasmale 16 38 34.21 41.38

Arrasfemale 13 38 32.13 43.07

Arrastotal 29 38 34.9 41.3

trendobserved forthefemalecurves. It isnoteworthythatthe AmiensproﬁleisclosetothatofSaint-Thomasd’Aizier,eventhough itsMADishigher.Amiensishypotheticallythe“poorest”ofthesites forsurvivorshipandsocialbackgroundafterSTA.

For both sexes, the lowest survival function is for the STA medievalleprosarium,withMADsof31.9(Table2)and39(Table3) forthefemalesandmales,respectively,whicharebothstatistically signiﬁcant(p<0.001).

3.2. Adultsurvivalbysex(pooled)

Inourlargerseries(Amiens,Vron,andSTA;Table4),weobserve thatfemalesurvivalfrom20to50yearsfollowsthesamelower curveasthemales,eventhoughonlySTAreachedstatisticalsig- niﬁcance(Figs.8–10).Followingthisgapinthesurvivalcurves,all threesamplesshowconsiderableoverlap.Theundeterminedsex componentofSTA(Fig.10)doesnotaffectthedifferencebetween maleandfemalesurvivalanditssigniﬁcance.

3.3. Adultsurvivalandtheco-occurrenceoftuberculosisand leprosy

BoththeSTA(withleprosy)andArras-Prefecture(withoutlep- rosy)sitesincludeindividualsconsistentwithapossiblediagnosis

Fig.8.Amiens4thcenturyAD(N=155).

Fig.9.Vron4–7thcenturiesAD(N=134).

ofTB:boneandjointlesions(STA,n=1;Arras,n=3)andperiosteal appositionsonthevisceralribsurfaces(STA,n=7;Arras, n=2).

Thecomparisonbetweentheirsurvivalcurvesofferednostatis- ticallysigniﬁcantdifferences(Fig.11andTable4).Consistentwith theresultsfromtheleprosarium,thesurvivalrateofthefemales fromSTAwassigniﬁcantlyreducedincomparisontotheirmale counterpartsandtheArrassample(Fig.12andTable4).

4. Discussion

4.1. Themortalityofadultfemales

Applying cementochronology to a largearchaeological sam- pletoestimatesurvivorshipcurvesprovidesuniqueopportunities to explore the impact of pathology on human groups. The

Fig.10.Saint-Thomasd’Aizier,Survivorshipcurvebysex(N=63).

(9)

Fig.11.Saint-ThomasadultsversusArras-Préfectureadults.

Fig.12.SurvivorshipofSaint-Thomasd’AizierfemalesandmalesversusArras femalesandmales.

survivorshipcurvesobtainedintheSTAleprosariumﬁrstsuggesta highermortalityforbothsexescomparedtothesamplesburiedin othercemeteries.STAspeciﬁcallysuggestsanincreasedmortality ofthefemalesaffectedbyleprosy.Howcanweexplainthispattern?

Recent clinical studies on cohorts of patients with leprosy (Table5)haverevealedspeciﬁccausesofmortality,tuberculosis

Table5

Summaryofmodernstudiesonleprosaria.

Source Location Patient(N)

PowellandSwan(1955) Carville(LA,USA) 50 KolonelandHirohata(1977) Hawai,(HI,USA) 1123

Tokudomeetal.(1981) Japan 2383

Glaziouetal.(1993) FrenchPolynesia 673

and cancer, in particular, that could lead to a decrease in life expectancyandpartiallyexplaintheincreasedmortalitypatternfor thefemalesinSTA.PowellandSwan(1955)reporteda20%mortal- ityratefromTBand18%fromneoplasms,whileGlaziouetal.(1993) observeda13%mortalityratedirectlyfromTB.Otherresearchers havespecificallyreportedsignificantlyhighermortalityamonglep- romatousfemalesfromcancersofthecervixandesophagusaswell asfromsuicides(Koloneland Hirohata,1977; Tokudomeetal., 1981).Finally,athirdmajorcauseofdeathreportedamongpatients withleprosyintwomodernsamplesisrenalinsufficiency,possibly asaconsequenceofchronicurinaryinfections(PowellandSwan, 1955;Tokudomeetal.,1981).

Donoghue et al. (2005) more recently suggested that the immunologicalchangesfoundinmulti-bacillaryleprosy,inasso- ciationwiththesocioeconomicimpactonthosesufferingfromthe disease,ledtoanincreasedmortalityfromtuberculosisandthere- fore,tothehistoricaldeclineinleprosy.Thus,it islegitimateto lookforevidenceofTBorcancer,amongotherdiseases,inlepro- matousarchaeologicalsamplestoassesstheirpotentialimpacton mortality,especiallyonthatoffemales.

However,thereappearstobenosigniﬁcantimpactofM.tuber- culosistoexplainthefemaleincreasedmortalityintheSTAsample.

AmongthecomorbiditiesintheindividualsexhumedfromSTA, theonecaseofpossibleboneandjointtuberculosis,along with thesevencasesofperiostealappositionsofthepleuralsurfacesof theribs,suggestthatTBwasendemic.Theyinvolvefourwomen aged 19 to 34 years and four men aged 29 to 73 years. It is clearlynotpossibletoprojectdeﬁnitiveconclusionsfromsucha smallsampleregardingtheimpactofTB ondifferentialmortal- ityofbothsexes.Takingintoaccountaproportionatenumberof possibletuberculosiscasesintheArrassample(n=5),wecannot concludethatthemortalitypatternsarecomparableinbothloca- tions.Similarly,theabsenceofpathologicallyvisibleneoplasmsin theosteologicalseriesisinconclusiveastotheirpotentialeffect onlepromatousmortalities.Finally, regardingthepresence and impactofrenalinsufﬁciencyintheSTAsample,wearecurrently usingamicroscopicexaminationofcorticalbonetoevaluateits potentialprevalence.Illustrativeoftheseissues,aroundconcre- tionwasdiscoveredinthepelviccavityofa46year-oldmale(#STA 659).X-raydiffractionandinfraredspectrometryrevealedacon- centricmineralstructurewithacentralcavitycharacteristicofa vesicalcalculus(Aufderheideetal.(1998).

AsDonoghueetal.(2005)havesuggested,analyzingleprosaria foraDNA,andparticularlythelipidmarkersofM.tuberculosis,may make itpossible toestimatingtheprevalenceof thediseasein peopleburiedinleprosariaandthereforetheroleofTBinthese institutions’declineandclosure.Finally,femalemortalityshould beexploredbyconsideringmaternalmortalityandtheprecarious statusofwomeninthishighlyhierarchicalsociety,wherefemale marginalization,fooddeprivation,andstricterreclusionwerecom- mon(Bériac, 1988,1984).Someofthefactorsimplicatedinthe modernstudiescitedabove,suchasthesuiciderate,mayberel- evant.Presently,thecomparativeanalysisofSTAtothenorthern Frenchsampleshasrevealedthattuberculosiswaspresent,even thoughitappearstohaveplayedasecondaryroleasacofactorfor thelepromatousmortalityrate.

4.2. Infertilityandmaternalmortality

BecauseSTAisconsideredtoincludemanyindividualswithlep- rosy,thelow proportionofimmatureskeletonsposesquestions (juvenilityindex:5–14yrs/20+yrs=0.103;15P5ratio=5–19yrs/5+

yrs=0.119)(Bocquet-Appel,2008;Bocquet-Appeletal.,2008).The absenceofchildburialscannotbeattributedtotaphonomicpro- cesses,especiallywhencomparedtoothersecularsites(Table1).

Theoverall preservationof subadultskeletonswascomparable,

(10)

Fig.13. Saint-Thomasd’Aizier(#636a,636b):femaleandfetusinutero.(A)Burialwithfetalbonesunderlinedinred;(B)burial,insituview;(C)fetalpositionoutlinedin redaccordingtothebonedispersalinthemother’spelvis.(Forinterpretationofthereferencestocolorinthisﬁgurelegend,thereaderisreferredtothewebversionofthis article.)

and no empty graves were identiﬁed. Another hypothesis that couldexplaintheabsenceofchildrencouldbethatthechildrenof male,femalesandcoupleshostedintheleprosariumwereremoved fromtheirparents earlyand ultimately pursuedtheirlife else- where (Bériac, 1988).There were no speciﬁcor only sketchily

mentioned regulationsonchild exclusionand expulsion,family status,betrothal,marriage,divorceandsexsegregation.Although such explanationshave beensuggested in few medievaltexts, theirstrictapplicationisuncertain(Brenner,2007).Onefetusand three neonates’remains suggestof activesexuallife withinthe

(11)

residentcommunityatSTAandelsewhere(Lewis,2002).There- fore,thescarcityofsubadultburialsisthepossiblerealpictureof theoverallinfertilitywithintheleprosarium’sinmates.

Increasedfemalemortalitymaybeimplicatedin thegeneral infertilityofreclusivecouples,oritcouldbeadirectconsequence ofmaternalmortality.Thismaybeillustratedbyoneburialofa younglepromatouswomanwithherat-termfetusinutero,which wasfreeofvisiblepathology(Fig.13).

Earlierstudiesconvincinglyhighlightedtheimpactofleprosy onmaleinfertilityasopposedtonormalfertilityinfemalepatients withleprosy(Beiguelman,1967).Thiswasprobablytrueinfavor- ablesocioeconomicconditions, which wasnot thecase atSTA, whereisolationandalowincomewouldhavebeenparticularly harmful(Brenner,2007).Maternalmortalitywouldhavecertainly exacerbatedthedeclineinfertilityofthissmallcommunity.Allof thesereasonscouldexplaintherelativelylownumberofimmature samplesburiedatSTA.

4.3. Thedeclineofleprosy

Afterthetheoryof cross-immunization betweenM.tubercu- losisand M.leprae raised,longago by Chaussinand(1944) and Manchester(1984),questionedsincethenbyWilburetal.(2002), the hypothesis of Donoghue et al. (2005) implicating tubercu- losisin the decline of endemic leprosy is attractive. However, manyauthorshaveclaimed thatthesimultaneousdevelopment ofenhancedimmunityagainstleprosyduetotheselectiveeffect thattheBlackDeathcouldhaveplayedaroleintherapiddisap- pearanceofHansen’sdiseaseduringthe14thand15thcenturies AD(Bassukasetal.,2012;Lechat,2002).

Thesearesomeofthemanyalternativehypothesizedcauses forthedeclineofleprosyinEurope(e.g.,low-fertility,socialexclu- sion,andlowincomeassociatedwithleprosy).However,theresults ofourresearchsuggestanassociationbetweenhighermortality ratesandreducedfertilitythatmayhavecontributedaswell,since weknowthat,despitetheabsenceofcongenitalforms,themain routeoftransmissionoccursbydropletinfectionwithinhouseholds (Moetetal.,2006;Raoetal.,1975).

Theimpactofcomplexsocialandpoliticalfactorsonindividual health,asassessedthroughnon-specificindicatorsofthequalityof life,hasbeendemonstratedinmodernpopulations(Audureauetal., 2013).Significantdeleteriouseffectsonoutcomesincludingmor- tality,morbidityandhealth-relatedbehaviorswasobservedwhen individualsweresuddenlydeprivedofwork,socialinteractionand autonomy(Audureauet al.,2013).Examples ofsuchconditions includeforcingpatientstousetheirownchestsascoffins,sharing singleunhealthytimbercabinsordormitory,andlivingfromalms (NielandTruc,2009;Nieletal.,2007).Localconditionsleadingtoan overalldecreasedqualityoflifeandlowsurvivalofpatients,added tolowfertility,surelyheavilyimplicateeitherthedirectimpactof diseasesonlifeexpectancyoronfertility,andundoubtedlyaccel- eratedtheinmatesdemiseandtheclosureoftheinstitution.

5. Conclusion

Thefirstobjectiveofthisstudywastocomparedemographic profilesofaFrenchmedievalleprosariumtoeightarchaeological seriesdatedfromLateAntiquitytotheLateMiddleAges,estimated throughcementochronology.Theresultsclearlydefineda social hierarchyofsurvivorshiplevelswithskeletonsshowingsignsof leprosysystematicallyassociatedwiththehighestmortalityand female’smortalitysystematicallyhigherthanmales.

The second objective tested the hypothesis that the co- occurrenceoftuberculosisandleprosyincreasedearlymortality inlocalpopulationsandthusprecipitatedthedeclineofleprosy.

ConclusionsindicatethattheimpactofHansen’sdiseaseisindeed perceptibleatSTA,butotherunassessedimportantfactorsmay have affected the morbidity of this population. Among other possiblecauses,wefoundthattuberculosisandchronicurinary infection,eventhoughdiagnosedatSTA,probablyonlyplayeda secondaryroleasacofactorforthelepromatousmortalityrate.

Thethirdobjectivewastodiscussthequestionofmaleinfertil- ityandelevatedfemalemortalitywithintheleprosariumtoexplore thepathocenosesthatmayhavecontributedtothedeclineoflep- rosyattheendoftheMiddleAgesinEurope.EvidencefromtheSTA leprosariumseemtofavorincreasedmaternalandfemalemortal- itycombinedtolepromatousmaleinfertilityasthemaincausefor thedeclineofleprosyin17thcenturyEurope.

Ourfinalobjectiveconsideredtheproposalthattheconfinement ofvariouspopulationsinmedievalEuropeprecededandaccompa- niedthedeclineanddisappearanceofleprosy.Thedemographic impactofthelowqualityoflifeforfamilieswithHansen’sdisease intheseinstitutionsafterthe14thclearlyplayedacatalyticcon- textinwhichtheroleofpathocenosessuchastheGreatPlagueand endemicTBleadtothedeclineanddisappearanceofleprosy.This complexdynamicbetweensocialenvironmentandpathocenoses shouldthereforebemeticulouslyexploretorefineourcomprehen- sionofleprosyevolutionattheendoftheMiddleAge.

Acknowledgements

TheauthorsthankJaneBuikstraandthreeanonymousreviewers whoprovidedcommentsthatgreatlyimprovedthismanuscript,as wellasthehelpoftheAmericanManuscriptEditors.Thanksalso gotoMarie-CécileTruc,RaphaëlleLefebvre,DamienJeanneandthe archaeologistsoftheGroupeArchéologiqueduValdeSeineandthe InstitutNationaldeRecherchesArchéologiquesPréventives.Weare alsogratefultoHelenDonoghueandMarkSpigelmanfortheirassis- tanceandresponsibilityofthepaleomicrobiologicalstudyofSTA material.Theﬁrstauthoralsopaystributetothoseaffectedbylep- rosymet,from1973to1977,duringweeklyclinicsonthebanksof theHooghlyinHowrah,duringmonthlyclinicsonroadsidesofSan- talPargana(Bihar),topeopleoftheleprosariumofChandraghona (ChittagongHillTracts)andtotwopatientsdiagnosedwithleprosy attheThanaHealthCenterofShariakandi(BograDistrict).

AppendixA. Supplementarydata

Supplementarydataassociatedwiththisarticlecanbefound,in theonlineversion,atdoi:10.1016/j.ijpp.2015.02.005.

References

Alduc-LeBagousse,A.,Blondiaux,J.,Desloges,J.,Maneuvrier,C.,2004.Lessépultures d’unsanctuairebénédictinnormand:lechoeurdel’abbatialedeSaint-Pierre- sur-Dives.In:InhumationsetÉdiﬁcesReligieuxAuMoyenÂgeEntreLoireet Seine,TablesRondesDuCRAHAM.PublicationsduCRAHM,Caen,pp.181–206.

Andersen,J.G.,Manchester,K.,1992.Therhinomaxillarysyndromeinleprosy:aclinical,radiologicalandpalaeopathologicalstudy.Int.J.Osteoarchaeol.2,121–129.

Andersen,J.G.,Manchester,K.,Roberts,C.,1994.Septicbonechangesinleprosy:

aclinical,radiologicalandpalaeopathologicalreview.Int.J.Osteoarchaeol.4, 21–30.

Audureau,E.,Rican,S.,Coste,J.,2013.Fromdeindustrializationtoindividualhealth- relatedqualityoflife:multilevelevidenceofcontextualpredictors,mediators andmodulatorsacrossFrenchregions,2003.HealthPlace22,140–152.

Aufderheide,A.C.,Rodriguez-Martin,C.,Langjoen,O.M.,1998.Urinarydiseases.In:

TheCambridgeEncyclopediaofHumanPaleopathology.CambridgeUniversity Press,Cambridge,pp.284–286.

Baker,B.J.,1999.In:Palﬁ,G.,Dutour,O.,Deak,J.,Hutas,I.(Eds.),TuberculosisPast andPresent.GoldenBookandTuberculosisFoundation,Budapest/Szeged,pp.

301–307.

Bassukas,I.D.,Gaitanis,G.,Hundeiker,M.,2012.Leprosyandthenaturalselection forpsoriasis.Med.Hypotheses78,183–190.

Beiguelman,B.,1967.Leprosyandgenetics.Areviewofpastresearchwithremarks concerningfutureinvestigations.Bull.WorldHealthOrgan.37,461–476.