Analysis of bearing wear, whole blood and synovial fluid metal ion concentrations and histopathological findings in patients with failed ASR hip resurfacings

RESEARCH ARTICLE Open Access Analysis o bea ing wea , whole blood and syno ial luid me al ion concen a ions and his opa hological indings in pa ien s wi h ailed ASR hip esu acings La i Leh o i a 1,2* , Aleksi Rei o 2 , Jy ki Pa kkinen 3 , Ha y Ho hi 4 , Johann Henckel 4 , Alis e Ha 4 and An i Eskelinen 2 Abs ac Backg ound: Ad e se Reac ion o Me al Deb is (ARMD) is s ill a majo eason o e ision su ge ies in pa ien s wi h me al-on-me al (MoM) hip eplacemen s. ARMD consis s o a wide ange o al e a ions in pe ip os he ic issues, mos impo an o which a e me allosis, in lamma ion, pseudo umo s and nec osis. S udies in es iga ing his opa hological indings and hei associa ion o implan wea o indi ec measu es o wea ha e yielded inconsis en esul s. The e o e, we aimed o in es iga e bea ing su ace wea olume, whole blood and syno ial luid me al ion concen a ions, his opa hological indings in pe ip os he ic issues and hei associa ions. Me hods: Se en y-eigh pa ien s wi h 85 hips e ised o ARMD we e included in he s udy. P io o e ision su ge y, all pa ien s had whole blood ch omium and cobal ion le els assessed. In e ision su ge y, a syno ial luid sample was aken and analyzed o ch omium and cobal . Pe ip os he ic issue samples we e aken and analyzed o his opa hological indings. Explan ed implan s we e analyzed o bea ing wea olume o bo h ace abula cup and emo al head componen s. Resul s: Volume ic wea o he ailed componen s was highly a iable. The o al wea olume o he head and cup had a s ong co ela ion wi h whole blood ch omium and cobal ion concen a ions (C : ρ=0.80,p<0.001andCo:ρ=0. 84, p< 0.001) and a bi weake co ela ion wi h luid ch omium and cobal ion concen a ions (C : ρ=0.50,p<0.01and Co: ρ=0.41,p= 0.027). Mos issues displayed only low- o-mode a e amoun s o mac ophages and lymphocy es. To al wea olume co ela ed wi h mac ophage shee hickness (ρ=0.25,p= 0.020) and nec osis (ρ=0.35,p<0.01).Whole blood ch omium and cobal ion concen a ions had simila co ela ions. Lymphocy e cu hickness did no co ela e wi h ei he o al wea olume o whole blood me al ion concen a ions, bu co ela ed wi h he g ade o nec osis. Conclusions: Bea ing wea olume co ela ed wi h blood me al ion le els and he deg ee o nec osis and mac ophage in il a ion in pe ip os he ic issues sugges ing a dose- esponse ela ionship. Whole blood me al ion le els a e a use ul ool o clinician o es ima e bea ing wea and subsequen issue esponse. Keywo ds: Me al-on-me al hip eplacemen , Ad e se eac ion o me al deb is, ARMD, ALTR, ALVAL, Wea , His opa hology * Co espondence: [email p o ec ed] 1 Facul y o Medicine, Uni e si y o Tampe e, Tampe e, Finland 2 Coxa Hospi al o Join Replacemen , Tampe e, Finland Full lis o au ho in o ma ion is a ailable a he end o he a icle © The Au ho (s). 2017 Open Access This a icle is dis ibu ed unde he e ms o he C ea i e Commons A ibu ion 4.0 In e na ional License (h p://c ea i ecommons.o g/licenses/by/4.0/), which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided you gi e app op ia e c edi o he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons license, and indica e i changes we e made. The C ea i e Commons Public Domain Dedica ion wai e (h p://c ea i ecommons.o g/publicdomain/ze o/1.0/) applies o he da a made a ailable in his a icle, unless o he wise s a ed. Leh o i a e al. BMC Musculoskele al Diso de s (2017) 18:523 DOI 10.1186/s12891-017-1894-5 Backg ound Ad e se eac ion o me al deb is (ARMD) is s ill a majo eason o e ision su ge ies in pa ien s wi h me al-on-me al (MoM) hip implan s. Al hough he use o MoM hip implan s has been widely ceased, mo e han one million pa ien s ha e ecei ed such a de ice [1] and hose ha ha e no been e ised s ill pose an inc eased isk o implan ailu e. ARMD is an umb ella e m desc ibing a wide ange o al e - a ions seen mac o- and mic oscopically in he pe i- p os he ic issue such as me allosis, nec osis, in lamma ion o di e en ypes and so - issue in lam- ma o y lesions e e ed as pseudo umo s [2–4]. Re ie al s udies ha e in es iga ed implan wea and i s associa ion o ARMD. Resul s o hese s udies ha e been inconclusi e as ad e se eac ions ha e been ob- se ed bo h in pa ien s wi h high and low wea ing hip implan s [5–11]. In hei ecen sys ema ic e iew, Campbell e al. concluded ha no clea dose- esponse ela ionship be ween wea and ARMD could be es ab- lished due o he he e ogenei y o he indings in he in- cluded s udies. S udies ha ha e in es iga ed wea o indi ec ma ke s o wea , such as syno ial luid (SF) o whole blood (WB) me al ion concen a ions, and he his opa hological ea u es o ARMD ha e also yielded inconsis en esul s [8, 9, 12–19]. Ex a-a icula issues e ie ed om pa ien s wi h ARMD a y conside ably in hei his ologic p esen a ion. Mos o en issues display p ominen mac ophage in il a ion as a esponse o he cy o oxic me al wea deb is wi h a a iable amoun o lymphocy ic in il a ion, ei he di use o agg ega ed [2,3,8,13].Howe e ,inamino i yo pa ien swi h ARMD, he e is hea y lymphocy ic in il a ion, esem- bling ype IV hype sensi i i y eac ion [17, 20–23]. The p esence o lymphocy es is usually accompanied wi h he p esence o nec osis and his ype o issue esponse was i s e med ALVAL (Asep ic Lympho- cy ic Vasculi is-Associa ed Lesion) by Wille e al. [20]. Te ms ALVAL and ARMD ha e howe e been inapp op ia ely used as synonyms in he ecen li e a u e [24]. Low bea ing wea has been associa ed wi h a suspec ed me al hype sensi i i y esponse in some s udies [8, 9, 16]. Vice e sa, high bea ing wea has been associa ed wi h a mac ophage-domina ed o eign-body esponse. [8, 13]. In addi ion, low WB me al ion le els ha e been associa ed wi h lymphocy e-domina ed issue esponse and high me al ion le els wi h mac ophage- domina ed esponse [17]. Based on hese indings, me al hype sensi i i y o implan -de i ed deb is has been hypo h- esized as a cause o ARMD in pa ien s wi h low-wea ing hip implan s, and cy o oxic, mac ophage domina ed esponse in pa ien s wi h high-wea ing hip implan s [8, 13, 16, 17] Howe e , indings no suppo ing hese hy- po heses ha e been published as well [10, 12, 15, 18, 19]. The his opa hology o ARMD has been well desc ibed bu he li e a u e ega ding i s associa ion o implan wea is inconsis en . I is impo an o unde s and he ue na u e o he associa ion be ween wea and his opa hological indings in ARMD. A e all, i is he his opa hological changes – issue des uc ion and in- lamma ion – ha lead o ailu e o MoM hip implan s. Implan wea canno be measu ed in- i o and hus canno be used in clinical decision making bu he e a e eliable indi ec measu es o wea , such as WB me al ion le els, ha a e commonly used in he ollow-up o pa ien s wi h MoM hip eplacemen s. To gain a be e unde s anding o he ela ionships be ween his opa ho- logical indings, bea ing wea and clinical ma ke s o wea we aimed o in es iga e bea ing su ace wea ol- ume, WB and SF me al ion concen a ions as clinical ma ke s o wea , and hei associa ions wi h his opa ho- logical indings o he pe ip os he ic issue in pa ien s wi h A icula Su ace Replacemen (ASR) hip esu acing de ice e ised due o ARMD. Based on hep e iousli e a u ewehypo hesized ha 1)low implan wea is associa ed wi h high amoun o lym- phocy es cha ac e is ic o an ALVAL esponse and 2) high implan wea is associa ed wi h high amoun o mac ophages cha ac e is ic o a o eign-body esponse o me al wea deb is. Me hods Be ween he ecall o he ASR MoM hip sys em (Depuy O hopaedics, Wa saw, IN, USA) in Augus 2010 and he end o ou ec ui men pe iod in Janua y 2016, 114 ASR hip esu acing de ices in 107 pa ien s ha e been e ised a ou ins i u ion. All consecu i ely e ised pa ien s who ga e in o med consen and ul illed he ollowing c i e ia we e included in ou s udy: 1) Re ision was due o ARMD, 2) Re ie ed componen s we e a ail- able o bea ing wea analysis and 3) Pe ip os he ic issue sample was a ailable o his opa hologic analysis. A e exclusion, 85 hips in 78 pa ien s we e included in ou s udy. Twen y-one o hese pa ien s we e e e ed o ou ins i u ion om cen al hospi als om o he hospi al dis ic s and 57 pa ien s had had hei index ope a ion (p ima y a h oplas y) and ollow-up a ou ins i u ion. Su ge y was pe o med by o unde he di ec supe i- sion o 14 senio o hopaedic su geons. The s udy was app o ed by he e hical commi ee o Pi kanmaa Hospi al Dis ic (R11006). Re ision su ge y was conside ed i 1) a clea pseudo u- mou (Impe ial class 2A,2B o 3) [25] was obse ed on c oss-sec ional imaging ega dless o symp oms o WB me al ion le els; o 2) he pa ien had ele a ed WB me al ion le els and hip symp oms despi e no mal indings in c oss-sec ional imaging; o 3) he pa ien had a con inu- ously symp oma ic hip o p og essi e symp oms ega dless Leh o i a e al. BMC Musculoskele al Diso de s (2017) 18:523 Page 2 o 11 o imaging indings o me al ion le els. Symp oms included hip pain, discom o , sense o ins abili y, and/o impai ed unc ion o he hip and sounds om he hip (clacking, squeaking). WB me al ion le els we e ega ded as being ele a ed i ei he ch omium o cobal exceeded 5 ppb. Pos - ope a i ely, ailu e was classi ied as being due o ARMD on he basis o he ollowing c i e ia [26, 27]: 1) he e was p es- ence o me allosis o mac oscopic syno i is in he join ; and/o 2) a pseudo umo was ound du ing e ision; and/ o 3) a mode a e o high numbe o pe i ascula lympho- cy es along wi h issue nec osis and/o ib in deposi ion was seen in he his opa hologic sample; and 4) pe iope a- i ely he e was no e idence o componen loosening o pe ip os he ic ac u e. In addi ion, in ec ion was uled ou by ob aining mul iple (a leas i e) bac e ial cul u es du ing e ision su ge y. Bea ing wea analysis The olume o ma e ial loss om he cup and head bea ing su aces was measu ed using a Zeiss P ismo (Ca l Zeiss L d., Rugby, UK) coo dina e measu ing machine (CMM). A o al o 400 pola scan lines on each su ace we e de ined and up o 30,000 da a poin s cap- u ed using a 2 mm uby s ylus; p o ocols o his me hod ha e been p e iously published [28]. An i e a i e leas squa e i ing me hod was used o analyse he aw da a cap u ed by he CMM and he unwo n geome y o he bea ing su ace was used o map egions o ma e ial loss om which he o al olume ic loss was calcula ed. Wea a e (mm3/yea ) was u he calcula ed by di iding o al wea olume in cubic millime e s by implan a ion ime in yea s. His opa hological analysis o he pe ip os he ic issue Du ing e e y hip e ision, a sample o he in lamed syno ia o pseudo umo was ob ained. Fo his opa ho- logical analysis, each issue sample was o malin ixed. Se e al 10 μm mic o ome sec ions we e made and em- bedded in pa a in. S anda d hema oxylin and eosin s aining was used. The sec ions we e examined his olog- ically unde no mal ligh wi h a Nikon Eclipse 50i (Nikon Co po a ion, Shinagawa, Tokyo, Japan). The samples we e g aded by a senio musculoskele al pa h- ologis (JP) using sco ing p inciples adop ed om he s udy by Na u e al. [2] ( e med Na u g ading in ou s udy) and he ALVAL sco e p e iously desc ibed by Campbell e al. [8]. The Na u g ading consis ed o ollowing pa ame e s: 1) lymphocy e cu hickness, 2) whe he lymphocy ic in- il a e was di use o agg ega ed, 3) p esence o ge mi- nal cen e s, 4) his iocy e shee hickness, 5) me al pa icle load wi hin his iocy es, 6) ex en o issue nec o- sis, 7) p esence o plasma cells and 8) p esence o g anu- lomas. Lymphocy e cu hickness was calcula ed using a g a icule. An a e age o i e measu emen s was aken and g aded as 0–3 (absen , 0.25 mm, 0.25–0.75 mm, >0.75 mm). Thickness o his iocy e shee s was also cal- cula ed using a g a icule and g aded 0–3 (absen , <1 mm, 1–2 mm, >2 mm). Me al pa icle load wi hin his iocy es was g aded as 0–4 as done in he assessmen o i on decomposi ion in li e cells [29, 30]. The ex en o o e all issue nec osis in a sample was g aded based on he su ace nec osis yping acco ding o Da ies e al. [22]. Type 1 su ace con ains in ac syno ial epi helium. Type 2 su ace shows loss o syno ial epi helial cells wi hou ib in deposi ion. In ype 3 su ace he e is ib in deposi ion and in ype 4 su ace he e is ex ensi e nec o- sis and loss o a chi ec u e. The ex en o ype 4 su ace nec osis was used o g ade he o e all issue nec osis in a gi en sample, as desc ibed by Na u e al. [2]. In g ade 4 nec osis, mo e han 75% o he issue sample showed ype 4 su ace nec osis. In g ade 3 nec osis, be ween 25 and 75% showed ype 4 su ace nec osis. In g ade 2 ne- c osis ei he less han 25% o he issue showed ype 4 su ace nec osis o he issue showed ype 3 su ace. In g ade 1 nec osis, he sample consis ed o ype 2 su ace. ALVAL sco ing consis s o h ee subsco es: syno ial lining (0-3p), issue o ganiza ion (0-3p) and in lamma- o y in il a e (0-4p). Bo h syno ial lining and issue o ganiza ion e lec he deg ee o nec osis and highe sco es mean highe deg ee o nec osis. In lamma o y in- il a e sco e e lec s he p edominan in lamma o y cell ype on a spec um: 0 poin s means minimal in il a es, 1p means p edominan ly mac ophages, 2p means bo h mac ophages and di use/pe i ascula lymphocy es, 3p means mos ly lymphocy es in agg ega es and some mac- ophages and 4p means la ge lymphocy e agg ega es and li le o no mac ophages. Whole blood and syno ial luid me al analysis Since Janua y 2012, WB me al ion (Co and C ) con- cen a ions ha e been ou inely measu ed as a pa o he sys ema ic ollow-up p og am o pa ien s wi h MoM hip eplacemen s a ou ins i u ion. All pa ien s unde wen WB analysis o Co/C ollowing sampling om he an ecubi al ein using a 21-gauge needle connec ed o a Vacu aine sys em (Bec on, Dickinson and Company, F anklin Lakes, NJ, USA) and ace- elemen blood ubes con aining sodium e hylenedi- amine e aace ic acid (EDTA). S anda d ope a ing p ocedu es we e es ablished a he Finnish Ins i u e o Occupa ional Heal h o Co and C measu emen using dynamic eac ion cell induc i ely coupled plasma (quad ipole) mass spec ome y (Agilen 7500 cx, Agilen Technologies, San a Cla a, CA, USA). The labo a o y echnicians we e blinded o all clinical ou comes. The samples we e p ese ed in +6 °C o +8 °C p io o analysis. Leh o i a e al. BMC Musculoskele al Diso de s (2017) 18:523 Page 3 o 11 Since Oc obe 2011, ou MoM hip e ision p o ocol has in ol ed pe iope a i e SF aspi a ion, which is always aken be o e opening he deep ascia using a s anda d 18- o 20-gauge needle connec ed o a Vacu aine sys em (Bec on, Dickinson and Company, F anklin Lakes, New Je sey) and ace elemen ubes con aining sodium EDTA. Simila p ocedu es we e used o SF me al ion concen a ion measu emen as desc ibed abo e o WB. S a is ical me hods Spea man ank co ela ion was used o s udy he associ- a ions be ween wea olume, WB and SF me al ion concen a ions, and his opa hological indings due o non-no mal dis ibu ion o hese a iables. Medians we e calcula ed o wea olume, WB and SF me al ion concen a ions. To compa e hese median alues be ween di e en subg oups, nonpa ame ic Mann- Whi ney U- es was used. When analyzing he co el- a ion be ween WB me al ion concen a ions and o he ac o s, we only included pa ien s wi h unila e al hip a h oplas ies (57 hips) o a oid he con ounding e ec o me al ions being eleased o he blood om he o he implan . The h eshold o s a is ical signi icance was se o 0.05. The analyses we e conduc ed using IBM SPSS so wa e (IBM Co p. Released 2012. IBM SPSS S a is ics o Windows, Ve sion 21.0. A monk, NY: IBM Co p.). Resul s O he 85 hips included in he s udy, 56 we e explan ed om emale pa ien s and 29 om male pa ien s. Mean age a he ime o he e ision su ge y was 57.3 yea s (SD 10.3 yea s). Mean ollow-up ime be ween index op- e a ion and e ision su ge y was 5.4 yea s (SD 1.8 yea s). Volume ic wea analysis o he explan ed componen s demons a ed a wide ange o wea in bo h he ace abu- la cup and emo al head (Table 1). Wea a es we e also highly a iable wi h a median o 9.0 mm3/yea ( ange 1.1…99.7 mm3/yea ). In a as majo i y o he compo- nen s (85.1%), he emo al head was mo e wo n han he ace abula cup. Median a io o head wea o cup wea was 1.7 ( ange, 0.5…10). In addi ion o ac ual olume ic componen wea , also WB and SF me al ion le els, se - ing as indi ec ma ke s o wea , we e highly a iable (Table 2). The o al wea olume o he head and cup s ongly co ela ed wi h WB me al ion concen a ions (C : ρ= 0.80, p< 0.001 and Co: 0.84, p< 0.001) and mode a ely wi h SF me al ion concen a ions (C : ρ= 0.50, p< 0.01 and Co: ρ= 0.41, p= 0.027). Wea a e had sligh ly s onge co ela ion wi h WB me al ion concen- a ions (C : ρ= 0.87, p< 0.001 and Co: 0.89, p< 0.001) and SF me al ion concen a ions (C : ρ= 0.71, p< 0.001 and Co: 0.66, p< 0.01) han o al wea olume. His ologically, a iable amoun s o mac ophages, lym- phocy es and nec osis we e seen in he issue samples. One o mo e ge minal cen e s we e p esen in 5 samples (5.7% o all samples). One o mo e g anulomas we e p esen in 14 samples (16.5% o all samples). All issue samples e inced a leas some deg ee o mac ophage in- il a ion (mac ophage shee hickness sco e o a leas 1) and in mos cases i was low o mode a e (Fig. 1). In ega d o lymphocy e in il a ion, mos issues e inced li le o no lymphocy es and in only a mino i y o he samples he in il a e was p ominen (Fig. 2). All cases wi h hea y lymphocy e in il a ion (sco es 2 o 3) had a mac ophage shee hickness sco e o 1, ie. he e was only li le mac ophage in il a ion in hese issues. Eigh o he nine issue samples wi h hea y lymphocy e in il a- ion had g ade 4 nec osis and he nin h had g ade 3 ne- c osis. Median wea a e o hese issues was highe han ha o he issues wi h lowe numbe s o lympho- cy es (Table 3). Lymphocy e cu hickness co ela ed posi i ely wi h he g ade o nec osis (ρ= 0.41, p< 0.001) and in lamma o y in il a e sco e (ρ= 0.79, p< 0.001).. All i e issue samples wi h ge minal cen e s had g ade 4 nec osis. The wea olume o wea a e o hese cases did no di e om cases wi hou ge minal cen e s (me- dian wea olume in cubic millime e s 61 e sus 37.5, p = 0.94; median wea a e in cubic millime e s/yea 11.9 e sus 8.8, p= 0.86). Tissues wi h one o mo e g anu- lomas we e associa ed wi h highe o al wea olume and wea a e when compa ed o issues wi h no g anu- lomas (Table 4). G ade o nec osis co ela ed posi i ely wi h syno ial lining sco e (ρ= 0.86, p< 0.001) and issue o ganiza ion sco e (ρ= 0.80, p< 0.001). Co ela ions be ween his ological a iables, o al wea olume o he head and cup componen s, wea a e as well as WB and SF me al ion concen a ions a e lis ed in Table 5. To al wea olume co ela ed wi h mac ophage shee hickness, g ade o nec osis (Fig. 3), syno ial lining sco e, issue o ganiza ion sco e and o al ALVAL sco e. Wea a e had simila co ela ions bu he co ela ion wi h mac ophages did no qui e each s a is ical signi i- cance. WB cobal and ch omium ion concen a ions had Table 1 Median olume ic wea and ange o ace abula and emo al componen s and bo h combined Componen Median olume ic wea (mm 3 ) Range (mm 3 ) Ace abula cup 14 2–247 Femo al head 24 4–485 Bo h combined 39 7–541 Table 2 Median concen a ions (μg/l) and anges (μg/l) o ch omium and cobal ions in bo h whole blood and syno ial luid Me al ion Whole blood Range Syno ial luid Range Ch omium 9.7 0.5–93.9 701 7.0–52360 Cobal 15.4 0.7–224.7 281.5 27.0–14870 Leh o i a e al. BMC Musculoskele al Diso de s (2017) 18:523 Page 4 o 11 Fig. 1 Dis ibu ion o mac ophage shee hickness sco es among all pe ip os he ic issue samples Fig. 2 Dis ibu ion o lymphocy e cu hickness sco es among all pe ip os he ic issue samples Leh o i a e al. BMC Musculoskele al Diso de s (2017) 18:523 Page 5 o 11 simila co ela ions. SF ch omium ion concen a ion co ela ed wi h g ade o nec osis, syno ial lining sco e and o al ALVAL sco e. SF cobal ion concen a ion co - ela ed wi h all bu mac ophage shee hickness. Nei he wea olume, wea a e, WB me al ion concen a ions o SF ch omium ion concen a ion we e associa ed wi h lymphocy e cu hickness o p esence o ge minal cen- e s. Howe e , SF cobal ion concen a ion did co ela e wi h lymphocy e cu hickness. Discussion In he p esen s udy, a spec um o in lamma o y and nec o ic changes associa ed wi h ARMD we e seen – a iable mac ophage and lymphocy e in il a ion and nec osis in pe ip os he ic issues [2–4, 8, 21]. Mos pa ien s e inced low- o-mode a e mac ophage in il a- ion and li le o no lymphocy e in il a ion. A ew pa ien s e inced a e y p ominen lymphocy e in il a- ion wi h g ade 4 nec osis ypical o an ALVAL esponse i s p oposed by Wille e al. [20]. The hickness o lymphocy e cu s co ela ed posi i ely wi h he deg ee o nec osis. Bea ing wea and WB me al ion concen a ions co ela ed posi i ely wi h he numbe o mac ophages and he deg ee o nec osis. Ou s udy is no wi hou limi a ions. Fi s ly, due o he pe ip os he ic issue being sampled only a he ime o e ision su ge y, i is di icul o say abou he na u al his o y o ARMD. Secondly, issue samples we e analyzed by one obse e only. Howe e , mul iple mic o- ome sec ions o each sample we e made and analyzed by a senio musculoskele al pa hologis well acquain ed wi h ARMD his opa hology. Thi dly, we did no pe o m a p io i sample size calcula ion. Ou s udy was e o- spec i e o na u e and pa ien s we e included on an “all-come ”basis. Howe e , a pos e io i powe analysis e ealed ha ou s udy has 90% powe (10% be a) o de- ec 0.35 co ela ion (medium e ec size) wi h a ype I e o p obabili y o 5% (al a). Fou hly, al hough we con- secu i ely ec ui ed pa ien s, no all pa ien s who unde - wen su ge y because o ARMD du ing he ec ui men pe iod we e included due o e used consen , missing issue samples and missing wea da a on some pa ien s. Thus, ou pa ien se ies is no comple ely consecu i e. Howe e , he numbe o excluded pa ien s was low in compa ison o he numbe o consecu i e pa ien s in- cluded. Indeed, he la ge numbe o pa ien s included is a majo s eng h o ou s udy. Ano he s eng h is ha we only included pa ien s e ised o ARMD and wi h iden ical hip esu acing implan s. Thus, ou da a speci - ically desc ibes pa ien s wi h ARMD while minimizing he con ounding e ec om ha ing di e en implan designs o ailu e modes o he han ARMD. Also, he e was no con ounding e ec om possible unnion wea deb is as in he case o THRs since we only in es iga ed he e ec s o bea ing wea deb is. Me al deb is o igina ing om he bea ing su aces and/o unnion has been shown o ha e cy o oxic e - ec s [31–34]. I has been sugges ed ha he cy o oxici y o me al deb is u he leads o issue des uc ion and mac ophage ec ui men o clea he issue and me al deb is [3, 21]. In suppo o his, we obse ed a co el- a ion be ween implan wea and he numbe o mac o- phages as well as he deg ee o nec osis bu no wi h he numbe o lymphocy es. Simila indings we e made in a s udy by G amma opoulos e al. [13]. Lang on e al. howe e did no ind co ela ion be ween wea and he amoun o mac ophages o nec osis [6]. We also ob- se ed ha he p esence o g anulomas was associa ed wi h inc eased o al wea olume. G anulomas a e hough o o m in esponse o high numbe o wea pa icles and ou esul s suppo his idea [35]. High wea , o high WB me al ion concen a ions, ha e been associa ed o mac ophage-domina ed issue esponses in o he s udies as well [8, 16, 17]. Me al hype sensi i i y leading o ype IV esponse wi h s ong lymphocy ic in- il a ion has been sugges ed as a cause o ailu e in hose pa ien s wi h low wea [8, 16, 17]. Con a y o ou hypo hesis, his was no obse ed in ou s udy. In ac , pa ien s wi h hea y lymphocy ic in il a ion had highe wea a e han pa ien s wi h lowe numbe s o lympho- cy es. These indings sugges ha excessi e me al deb is accumula ion was he cause o lymphocy ic in il a ion in hese pa ien s, no me al hype sensi i i y. G amma o- poulos e al. also did no ind co ela ion be ween wea and lymphocy ic in il a ion, bu no ed he p esence o a pa ien subg oup wi h hype sensi i i y- ela ed his o- pa hological indings and simul aneous low bea ing wea , sugges ing me al hype sensi i i y as a cause o ailu e in hose pa ien s [13]. Table 3 Wea a es acco ding o lymphocy e cu hickness Lymphocy e cu <2 Lymphocy e cu 2 o 3 P- alue Median wea a e (mm3/yea ) 8.1 14.5 0.054 Range (mm3/yea ) 1.1 …99.8 3.7 …48.0 Table 4 Wea olume and a e: compa ison be ween pa ien s wi h one o mo e g anulomas and hose wi hou g anulomas G anuloma p esen G anuloma absen P- alue Median o al wea olume (mm3) 106.5 31.0 0.016 Range (mm3) 10…378 7…541 Median wea a e (mm3/yea ) 16.3 8.1 0.035 Range (mm3/yea ) 1.6…99.8 1.1…86.4 Leh o i a e al. BMC Musculoskele al Diso de s (2017) 18:523 Page 6 o 11 The e a e se e al easons ha likely con ibu e o he inconsis ency o indings be ween di e en his opa ho- logical s udies. I is possible and p obable ha some pa- ien s ha e bo h high-wea ing implan s and an unde lying hype sensi i i y esponse ha would ha e e oked e en in he p esence o a low-wea ing implan . This combina ion may esul in a mixed- ype issue e- sponse ha has he cha ac e is ics o bo h wea - ela ed o eign-body esponse and hype sensi i i y- ela ed ype IV issue esponses and he e o e makes i di icul o dis inguish be ween he wo based on implan wea o WB me al ion le els alone. Also, h eshold o he onse o adap i e immune esponse is likely a iable be ween indi iduals [13]. This would explain why some pa ien s ole a e ex ensi e amoun o wea deb is and some pa- ien s de elop ALVAL in he p esence o a low wea ing MoM hip eplacemen . In addi ion o pa ien suscep i- bili y, di e ences in implan ypes among s udies may Table 5 Spea man ho co ela ion coe icien s and associa ed p- alues o co ela ions be ween o al wea olume, wea a e, indi ec ma ke s o wea (whole blood and syno ial luid me al ion concen a ions) and his opa hological g ading (Na u and ALVAL) Na u g ading ALVAL g ading Lymphocy ic cu ing Mac ophage shee hickness G ade o nec osis In lamma o y in il a e sco e Syno ial lining sco e Tissue o ganiza ion sco e To al ALVAL sco e To al wea olume ho = 0.11 p= 0.32 ho = 0.25* p= 0.020 ho = 0.35* p< 0.01 ho = 0.13 p= 0.23 ho = 0.37* p< 0.01 ho = 0.25* p= 0.023 ho = 0.31* p< 0.01 Wea a e ho = 0.17 p= 0.12 ho = 0.20 p= 0.069 ho = 0.42* p< 0.0001 ho = 0.16 p= 0.15 ho = 0.48* p< 0.0001 ho = 0.35* p< 0.01 ho = 0.40* p< 0.001 WB C ho = 0.089 p= 0.51 ho = 0.30* p= 0.024 ho = 0.45* p< 0.001 ho = 0.19 p= 0.26 ho = 0.54* p< 0.001 ho = 0.33* p= 0.015 ho = 0.48* p< 0.001 WB Co ho = 0.18 p= 0.18 ho = 0.29* p= 0.029 ho = 0.51* p< 0.001 ho = 0.26 p= 0.055 ho = 0.60* p< 0.001 ho = 0.40* p< 0.01 ho = 0.55* p< 0.001 SF C ho = 0.30 p= 0.12 ho = 0.16 p= 0.40 ho = 0.48* p< 0.01 ho = 0.25 p= 0.19 ho = 0.56* p< 0.01 ho = 0.34 p= 0.070 ho = 0.53* p< 0.01 SF Co ho = 0.49* p< 0.01 ho = 0.17 p= 0.37 ho = 0.54* p< 0.01 ho = 0.44* p= 0.017 ho = 0.47* p= 0.011 ho = 0.38* p= 0.045 ho = 0.57* p< 0.01 Values ha a e s a is ically signi ican a e lagged wi h * Fig. 3 The di e ence in median o al wea olume (head and cup) in pa ien s wi h low-g ade nec osis (g ades 1 and 2) e sus pa ien s wi h high-g ade nec osis (g ades 3 and 4), p< 0,001 Leh o i a e al. BMC Musculoskele al Diso de s (2017) 18:523 Page 7 o 11 play a ole. Subs an ially highe ailu e a es ha e been epo ed o ASR XL THRs in compa ison wi h ASR hip esu acings [5, 19]. ASR XL and ASR ha e simila bea ing couples, bu in he ASR XL he e is unnion- in e ace be ween he i anium s em and CoC head ha se es as an addi ional sou ce o me al deb is. I has been shown ha wea om he unnion is di e en in na u e compa ed o bea ing su ace wea and may lead o lymphocy ic and nec o ic issue esponses [14, 36, 37], possibly con ibu ing o he inconsis en- cies be ween o he ecen his opa hological s udies. Na u e al. sugges ed ha he lymphocy ic immune e- sponse in pa ien s wi h MoM implan s is a dynamic p ocess, beginning wi h pe i ascula lymphocy ic agg e- ga es and leading o o ma ion o lymphoid ollicles wi h ge minal cen e s, also e med as e ia y lymphoid o - gans (TLOs) [2]. TLOs a e capable o o ming new B and T cells locally. TLOs a e seen in a ec ed issues o pa- ien s wi h ch onic au oin lamma o y diseases such as heuma oid a h i is, Sjog ens synd ome and Hashimo o’s hy oidi is and a e conside ed o be o med as a esponse o a pe sis en an igen ha canno be elimina ed [38]. Mi al e al. demons a ed he p esence o TLOs and asso- cia ed chemokines in issues o pa ien s wi h ailed MoM hip implan s and sugges ed ha hese pa ien s o m a speci ic pa hological subse [39] in addi ion o he well- es ablished o eign-body esponse [3, 13] and ALVAL e- sponse [8, 20, 40]. In keeping wi h indings by Na u e al. and Mi al e al., we ound ha a mino i y o he pa ien s displayed lymphoid ollicles wi h ge minal cen e s. Fu - he , all issue samples displaying ge minal cen e s had g ade 4 nec osis. This sugges s ha he o ma ion o TLOs in pe ip os he ic issues is associa ed wi h issue de- s uc ion, possibly accele a ing he p ocess o implan ail- u e. In line wi h his is a inding ha he p esence o TLOs has been associa ed wi h issue des uc ion and loss o unc ion in au oimmune diseases [38]. In he p esen s udy, we also obse ed ha e en in he absence o ge mi- nal cen e con aining lymphoid ollicles, he hickness o lymphocy ic cu s co ela ed wi h he g ade o nec osis. Whe he he lymphocy ic immune esponse, also e med ALVAL, is a dynamic p ocess leading o o ma ion o TLOs as Na u e al. sugges ed [2] o whe he pa ien s wi h TLOs de ine hei own dis inc pa hological subse as Mi - al e al. sugges ed [39] equi es u he esea ch. Howe e , due o he c oss-sec ional na u e o his opa hological s udies, i is di icul o in es iga e he na u al his o y o ARMD. ALVAL g ading in oduced by Campbell e al. [8] has been used in se e al s udies [9, 16, 18, 41] bu o he g ad- ing sys ems ha e been used as well [2, 13, 21, 22, 36]. This makes compa ison be ween s udies di icul . In he p esen s udy, all issue samples we e analyzed acco ding o wo g ading c i e ia: ALVAL g ading and g ading p inciples es ablished by Na u e al. [2]. ALVAL g ading is ela i ely es ic ed compa ed o he Na u g ading as i only includes in lamma o y in il a e sco e, syno ial lining sco e and issue o ganiza ion sco e. Mo eo e , as dis- cussed by Riccia di e al., bo h syno ial lining sco e and issue o ganiza ion sco e e lec he deg ee o nec osis [21]. A s ong co ela ion be ween hese sco es and he Na u sco e o nec osis was obse ed in ou s udy. ALVAL sco e was o iginally designed o help dis inguish ailu es ela ed o high wea om ailu es ela ed o suspec ed hype sensi i i y (ALVAL) esponse. Al hough nec osis is o en seen wi h ALVAL esponse, i is no speci ic o ALVAL as i is also seen wi h mac ophage-domina ed o eign body eac ions wi h possible ela ed cy o oxici y [3, 13]. This lea es only he in lamma o y in il a e sco e in ALVAL g ading speci ic o ALVAL esponse. In he p esen s udy, a s ong co ela ion be ween lymphocy e cu hickness and in lamma o y in il a e sco e was ob- se ed. This indica es ha he in lamma o y in il a e sco e is use ul in dis inguishing lymphocy e-domina ed e- sponses om hose ha a e no lymphocy e-domina ed. Phillips e al. also concluded ha ALVAL sco ing is use ul o dis inguishing be ween mac ophage and lymphocy e esponses [42]. In lamma o y in il a e sco e in ol es e alua ion o bo h lymphocy ic and mac ophagic compo- nen s. Howe e , bo h lymphocy es and mac ophages a e o en seen in pe ip os he ic issues. G amma opoulos e al. sugges ed ha an easie me hod o iden i y ALVAL e- sponses om wea - ela ed esponses would be o measu e only he hickness o lymphocy ic cu ing, e med Ox o d- ALVAL sco e in hei s udy [13]. We ag ee wi h G amma- opoulos e al. and ind ha sepa a e sco es o e alu- a ion o mac ophage and lymphocy e in il a ion p o ide mo e in o ma ion abou he ailu e mechan- ism and lead o easie compa ison be ween his o- pa hological s udies. In he p esen s udy, WB me al ion le els had a s ong co ela ion wi h bea ing wea olume, wea a e and a mode a e co ela ion wi h se e al his opa hological ea u es. SF me al ion le els also co ela ed wi h bea ing wea olume, wea a e and some o he his opa ho- logical ea u es, bu he co ela ions we e weake . Wea a e had a s onge co ela ion wi h WB and SF me al ion le els han o al bea ing olume. Wea a e, WB and SF me al ion le els likely e lec he ecen bu den o me al deb is whe eas o al wea olume e lec s he amoun o o al wea accumula ed du ing implan a ion ime. In a s udy by De Sme e al. bo h WB and SF me al ion le els we e ound o co ela e well wi h linea wea o he emo al componen [43]. Lang on e al. also no ed a co ela ion be ween wea olume and WB me al ion le els [6]. In e es ingly, WB me al ion le els had s on- ge co ela ion wi h his opa hological indings compa ed o SF me al ion le els in he p esen s udy. This is Leh o i a e al. BMC Musculoskele al Diso de s (2017) 18:523 Page 8 o 11 suppo ed by a ecen s udy by Rei o e al. who ound ha SF me al ion le els had ela i ely poo co ela ion wi h his opa hological indings [12]. SF aspi a ion is an in asi e p ocedu e and he measu emen does no seem o p o ide any addi ional in o ma ion compa ed o WB measu emen . Measu emen o WB me al ion le els is a eliable, indi ec way o gain in o ma ion o he in-si u wea p ocess, in lamma ion and issue des uc ion. Clini- cians should use WB me al ion le els in he ollow-up o pa ien s wi h MoM hip implan s and closely moni o hose pa ien s wi h ele a ed me al ion le els. A schema ic diag am o he ela ionships be ween wea , his ology o he pe ip os he ic issues, WB me al ion le els and ollow-up o he pa ien s is p esen ed in Fig. 4. Conclusions In he p esen s udy wi h ailed ASR hip esu acings, o al wea olume, wea a e and WB me al ion concen- a ions co ela ed wi h he numbe o mac ophages and he deg ee o nec osis, bu no wi h he amoun o lym- phocy es. Mos issue samples e inced mac ophages bu li le o no lymphocy es ypical o a non-speci ic o eign- body esponse. A mino i y o he samples e inced s ong lymphocy e in il a ion combined wi h high amoun o nec osis, ypical o an ALVAL esponse. Howe e , con a y o ou hypo hesis, his ype o esponse was no associa ed wi h low implan wea in he p esen s udy. The signi icance o pa ien suscep ibili y in he de elop- men o ARMD is poo ly unde s ood and i is no cu en ly known which ac o s lead o he adap i e lymphocy ic esponse seen in some pa ien s. Fu u e s udies should be di ec ed o unde s and he pa ho- physiological mechanisms behind di e en ypes o issue esponses seen in pa ien s wi h MoM hips. WB me al ion le els co ela ed wi h o al wea olume, wea a e and his opa hological indings. Measu e- men o WB me al ion le els is use ul in he ollow- up o pa ien s wi h hip esu acings as i p o ides in o ma ion o he wea p ocess, in lamma o y e- sponse and issue des uc ion. Abb e ia ions ALVAL: Asep ic lymphocy ic asculi is-associa ed lesion; ARMD: Ad e se eac ion o me al deb is; ASR: A icula su ace eplacemen ; CoC: Ce amic-on-ce amic; MoM: Me al-on-me al; MoP: Me al-on-polye hylene; SF: Syno ial luid; THR: To al hip eplacemen ; TLO: Te ia y lymphoid o gan; WB: Whole blood Acknowledgemen s We wish o hank Ms. Ella Leh o o main aining ou s udy da abase. Funding The s udy was suppo ed by he compe i i e esea ch unds o Pi kanmaa Hospi al Dis ic , Tampe e, Finland (g an 9 N044, ep esen ing go e nmen unding). The sou ce o unding had no ole a any s age o his s udy. A ailabili y o da a and ma e ials The da ase s gene a ed and/o analysed du ing he cu en s udy a e no publicly a ailable due o indi idual p i acy o he pa ien s. Da a may be a ailable upon eques by email o he i s au ho . Au ho s’con ibu ions LL o med and analyzed he da a and w o e he ini ial d a o he manusc ip . AR, JP, HH, AH, JH and AE assis ed in in e p e a ion o he da a and edi ing he manusc ip . All au ho s ead and app o ed he inal manusc ip . E hics app o al and consen o pa icipa e All pa ien s ga e in o med consen and he s udy was app o ed by he e hical commi ee o Pi kanmaa Hospi al Dis ic (R11006). Consen o publica ion No applicable. Fig. 4 A schema ic diag am o he ela ionships be ween wea , his ology o he pe ip os he ic issues, whole blood me al ion le els and ollow-up o he pa ien s Leh o i a e al. BMC Musculoskele al Diso de s (2017) 18:523 Page 9 o 11