Short Stories

A retrospective analysis of factors associated with multiple implant failures in maxillae

A retrospective analysis of factors associated with multiple implant failures in maxillae
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
   Anders Ekfeldt  A retrospective analysis of factors Ulf Christiansson associated with multiple implant Torbjo¨  rn ErikssonUlf Linde´  n  failures in maxillae Sture LundqvistTorgil RundcrantzLars-Åke JohanssonKrister NilnerCamilla Billstro¨ m Authors’ affiliations:  A. Ekfeldt, L.-Å. Johansson,  Department ofProsthetic Dentistry, Center of Medical and OralHealth, Halmstad. U. Christiansson,  Department of ProstheticDentistry, Public Dental Service of Skåne, Lund. T. Eriksson,  Department of Prosthetic Dentistry,Public Dental Service of Blekinge, Lyckeby. U. Linde´n,  Clinic of Prosthetic Dentistry,Community Dentistry of Skåne, Malmö. S. Lundqvist,  Department of Prosthetic andStomatognathic Physiology Dentistry, Växjö. T. Rundcrantz,  Department of ProstheticDentistry, Public Dental Service of Skåne,Kristianstad. K. Nilner,  Department of Prosthetic Dentistry,Center of Oral Health Service, MalmöUniversity, Malmö. C. Billstro¨ m,  Nobel Biocare Norden AB,Go¨teborg, Sweden Correspondence to:  Anders Ekfeldt Mun-H-centerOdontologen GöteborgMedicinaregatan  12  A 413 90  GöteborgSwedenTel: π 46 31 773 38 20 Fax: π 46 31 773 38 19 e-mail: anders.ekfeldt / Date: Accepted  30  October  2000 To cite this article: Ekfeldt A, Christiansson U, Eriksson T, Linde´n U,Lundqvist S, Rundcrantz T, Johansson L Å, NilnerK, Billström C. A retrospective analysis of factorsassociated with multiple implant failures in maxillae Clin. Oral Impl. Res.  12 ,  2001 ;  462 – 467 Copyright C Munksgaard  2001 ISSN  0905 - 7161 462 Key words:  osseointegrated implant, implant failure, risk factors, cluster phenomenon Abstract:  This retrospective study was designed to verify the factors that influenceimplant failures. Six prosthodontic clinics in Sweden participated in the study, andtogether they included a total of 54 patients treated between January 1988 andDecember 1996. All patients were completely edentulous in the maxilla, and receivedeither a fixed prosthesis or an overdenture supported by at least 4 implants (BrånemarkSystem A ). Half of the patients belonged to the study group, and an inclusion criterionfor this group was that they had lost at least half of their implants. To reduce bias, thepatients in the control group were matched to the study group, i.e. they were selectedso that both groups were as identical as possible. The results of the study indicate thatthe control group had a better initial bone support than the study group. Furthermore,the patients in the study group suffered from circumstances that could induce implantfailure, such as bruxism, personal grief, depression, as well as addictions to cigarettes,alcohol and/or narcotics. On the study form the clinicians were asked to give their ownopinion of the reason for implant failure. The answers given could easily be grouped into5 different topics, and this experience can be useful to improve patient selection. Thisstudy suggests that there are certain factors of importance to consider to prevent acluster phenomenon of implant failures i.e. lack of bone support, heavy smoking habitsand bruxism. Numerous studies, both retrospectiveandprospective,haveshownthatedentu-lous patients treated with osseointe-grated implants to support a fixed pros-thesis can do remarkably well over time(Adell et al.  1990 ; Zarb & Schmitt  1990 ;Quirynen et al.  1991 a). However, in al-mostallpublishedstudies thereareafewpatients with multiple implant failures,who experience a so-called ‘‘cluster phe-nomenon‘‘. The study of such clusterphenomena might help to detect whichfactors may contribute to implant failure(Quirynen et al.  1991 b; Weyant & Burt 1993 ).The purpose of this retrospective in-vestigation was to test the hypothesisthat certain factors are associated with acluster phenomenon of oral implant fail-ures. Therefore, we chose to study agroup of patients that had lost at leasthalf of their implants in the maxilla, andcompare them to a group without im-plant failures. Another important ques-tion that was raised was: is it possible toidentify patients that could experiencemultiple implant losses prior to theirtreatment?All clinics participating in this multi-center study are prosthodontic clinicsat Public Dental Health Service Centersin Sweden, and all have had severalyears of experience with implant treat-ment.  Ekfeldt et al . Factors associated with multiple implant failures in maxillae Material and methods Study design Six prosthodontic clinics in Sweden par-ticipated in this retrospective study, andtogether they enrolled a total of  54  pa-tients that had been treated between Jan-uary  1988  and December  1996 . Oral sur-geons belonging to the treatment teaminstalled the implants.The initial discussion among the clini-cians was whether it was possible, priorto treatment, to identify patients thatwould suffer from multiple implantlosses. However, to use a possible futureimplant failure as an inclusion criterionwas not feasible. Therefore, we chose toconduct a controlled, retrospective studywhere half of the included patients hadsuffered from multiple failures, i.e. a lossof at least half of their implants (studygroup), while the other half did not loseany implants (control group).All patients were completely edentu-lous in the maxilla, and had receivedeither a fixed prosthesis or an overden-ture supported by at least  4  implants(Brånemark System A , Nobel Biocare AB,Göteborg, Sweden). To reduce bias, thepatients in the control group were ageand gender matched to the study group,i.e. they were selected so that bothgroups were as identical as possible re-garding age, gender, number of insertedimplants, and time of implant insertion.However, it was not considered possibleto match the two groups to the same oralsurgeon.Both the patients in the study groupand the control patients have been iden-tified through the recall system used atthe different clinics. In order to securethat the clinicians’ subjective opinionsand/or guesses would not influence theresults, only data that were clearly re-corded in the patients’ files were notedon the study forms. There was only oneinclusion criterion for the study group,namely that they had lost at least half oftheir implants. For the control group,the criteria were as follows: O  the control patients should have thesame amount of implants as the studypatients, and these implants shouldhave been placed within the sametime frame; O  no implant losses should have oc-curred during this time period; 463  |  Clin. Oral Impl. Res.  12 ,  2001  /  462 – 467 O  the gender ratio of the control patientgroup should be equal to that of thestudy patient group; O  the control patients should be ap-proximately the same age as the studypatients ( ∫ 5  years).All patients that had been treated withimplants in combination with bone aug-mentation were excluded from thisstudy. Records Data were collected from the patientrecords kept at the prosthodontic andsurgical departments. Only data thatwere clearly recorded in the files werenoted on the study forms. If any infor-mation was missing in the records, this Table 1.  Study population Study group Control group Total Number of patients 27 27 54Number of male patients 11 11 22Number of female patients 16 16 32...........................................................................................................................................Total number of implants 151 150 301Patient age: ∞ 40 years 0 0 041–50 years 6 5 1151–60 years 7 10 1761–70 years 8 10 18 ± 70 years 6 2 8 Table 2.  Implants inserted in the study and control groups; length of implants, and time offailures Fixture length (mm) 7 8.5 10 13 15 18 Missing Total Study groupIn function 0 0 9 6 5 0 2 22Failed before/at abutmentconnection 4 0 10 14 7 0 2 37Failed between abutment andprosthesis 1 0 3 10 3 0 1 18Failed during first year of loading 5 4 15 17 2 0 3 46Failed after first year of loading 1 0 8 9 6 0 3 27Missing 0 0 0 0 1 0 0 1Total 11 4 45 56 24 0 11 151...........................................................................................................................................Control groupIn function 1 4 39 61 42 2 0 149Unconnected 0 0 0 1 0 0 0 1Total 1 4 39 62 42 2 0 150 was stated on the form as ‘‘not evident‘‘.The clinicians also checked if any abnor-mal incidents had occurred during theimplant insertion or healing period. Ifnothing special had been noted in therecords, this was interpreted as ifnothing abnormal had taken place.The age and gender distribution of thestudy population was recorded, and thisis given in Table  1 .Apart from recording smoking habits,and alcohol or other drug abuse, someother health conditions were also re-corded, according to a review of theliterature (Esposito et al.  1998 b), such asdiabetes mellitus, thyroid dysfunction,general cortisone treatment, osteopor-osis; ongoing, or previous, multiple bone  Ekfeldt et al . Factors associated with multiple implant failures in maxillae fractures; epilepsy or cerebral damage,i.e. Parkinson’s disease or cerebralhemorrhage; cytostatic treatment; on-going, or previously carried out radio-therapy in the maxillary region; gener-ally impaired state of health and/or im-paired immune defense, personal grief,depression or other diagnosed psycho-logical illnesses, or bruxism. Results Fixture stability In the study group, a total of  151  im-plants were inserted and out of these  128 were lost –  55  before loading,  46  duringthe first year after loading, and  27  duringthe second year or later. The incidenceof failure was  43 % before loading (earlyfailures), and  57 % after loading (late fail-ures). Of the  73  late failures, the ma- Table 3.  Notes about abnormal incidents in surgical records (number of implants andpercentage of all inserted implants) Exposed Insufficientimplant initial AbnormalInfection threads stability swelling Total Study group ( n Ω 151) 9 (6.0%) 6 (4.0%) 19 (12.6%) 6 (4.0%) 40 (26.5%)Control group ( n Ω 150) 3 (2.0%) 7 (4.7%) 9 (6.0%) 0 (0%) 19 (12.7%)Total 12 13 28 6 59 Table 4.  Smoking habits (number of patients) Moderate Heavy smokerNon-smoker smoker ( ± 10 cig/day) Unknown Study group 14 3 9 1Control group 17 6 2 2Total 31 9 11 3 Table 5.  Reason for implant failure according to the clinician (number of patients) and numberof lost implants Number of lostimplantsNumber ofpatients Before AfterStudy group loading loading Group A Lack of bone support and/or bad bone quality 8 20 18Group B Heavy smoking habits 7 9 22Group C Bruxism or other overload problems 4 5 14Group D Influences by other illnesses 2 3 4Group E Psychological reasons 2 6 6Missing – 4 12 9 464  |  Clin. Oral Impl. Res.  12 ,  2001  /  462 – 467 jority ( 63 %) were lost during the firstyear of loading. Three patients lost all oftheir implants. In the control group, atotal of  150  implants were inserted, andsubsequently, according to the inclusioncriteria, none of these were lost.The mean length of the inserted im-plants was  10 . 9  mm for the study groupand  12 . 7  mm for the control group. Thismeans that approximately  1 . 8  mm( 16 . 5 %) longer implants were inserted inthe control group compared to the studygroup. One implant ( 13  mm in length)was left unconnected in the controlgroup. None of the implants in the studygroup were left unconnected. Table  2 presents the distribution of implantlength for the study and control groups,as well as the time of the failures in thestudy group.The notes made during the implant in-sertion and/or healing period revealedthat a few abnormal incidents had oc-curred. The total frequency of incidentsduring this time was  26 . 5 % for the studygroup, and  12 . 7 % for the control group.For the majority of the implants, onlyone abnormality was observed on eachimplant. Table  3  summarizes the notesmade in the surgeons records.In the study group,  9  ( 6 %) out of the 151  inserted implants showed signs ofinfection during the healing period and  7 of these implants were lost. Insufficientinitial stability was registered for  19 ( 12 . 6 %) implants in the study group atimplant insertion, and  17  of these werelost –  9  were early and  8  were late fail-ures. Six implants ( 4 %) in the studygroup had exposed implant threads atthe time of insertion, and  5  of themfailed later on. Abnormal swelling wasregistered for  6  implants, whereof allwere lost. Comparatively few compli-cations were registered for the controlgroup.Three patients in the study groupshowed signs of addiction to alcohol ornarcotics. According to the records,there were no patients with these prob-lems registered in the control group. Case history and clinical examination beforeprosthesis insertion Signs of bruxism were recorded beforeprosthesis insertion in  7  patients in thestudy group. According to the case his-tory,  3  patients were aware of their prob-lem. Another six patients in this groupreported that they were depressed or suf-fered from some form of personal griefsuch as grief after the death of a husbandor child, and it was verified that two pa-tients suffered from psychological ill-nesses. At the clinical examination, atotal of six cases of complicated inter-maxillary relations were registered, onein the control group and the rest in thestudy group.Apart from the parameters mentionedabove, there were no obvious registereddifferences between the two groups, ex-cept that the number of patients classi-fied as heavy smokers (those thatsmoked more than  10  cigarettes/day)was obviously larger in the study group( 9 ) compared to the control group ( 2 ).Otherwise there was an equal number ofsmokers in the two groups (Table  4 ).  Ekfeldt et al . Factors associated with multiple implant failures in maxillae Notes made after prosthesis insertion After rehabilitation, diagnostic signs ofbruxism were reported in  6  of the  7  pa-tients with reported bruxism beforeprosthesis insertion in the study group,compared to none in the control group.The seventh patient lost all of his im-plants before loading. He was one ofthree patients in the study group thatlost all implants. These patients were re-habilitated with complete dentures aftertheir losses.For all the patients in the study group,except four with missing values, the cli-nicians gave their own opinion of thereason for implant failure. The results ofthis were easily summarized into fivemain topics, Groups A-E (Table  5 ).Of the two patients in Group D, onehad a verified ‘‘dry mouth syndrome’’and asthmatic problems, and was under-going treatment with cortisone – theother had uncontrolled diabetes mel-litus, and was in a generally impairedstate of health. The two patients inGroup E suffered from personal grief dueto the loss of close family members (hus-band and child respectively).In Group C there were three timesmore late failures than early failures.The same relation was found in GroupB. These findings highlight the import-ance of correct loading and the bad in-fluence smoking has on the outcome ofthe treatment. For patients with lack ofbone support and/or bad bone quality(Group A), there seems to be an equaldistribution between early and late fail-ures. Discussion In a retrospective study it can be diffi-cult to know the actual number of pa-tients that will fulfil the study’s in-clusion criteria. All the participatingcenters, however, have a distinct patientrecall system that secures that most pa-tients come to annual follow-up visits,which made it possible to select and in-clude patients according to the inclusioncriteria.A total of  54  patients, divided into twogroups of equal age and gender – onewhere failures had occurred (studygroup) and one without implant failures(control group) – were included in this 465  |  Clin. Oral Impl. Res.  12 ,  2001  /  462 – 467 study. Furthermore, the study group wasselected according to the criterion thatthere should be a cluster loss of at leasthalf of the inserted implants.To reduce bias when comparing thetwo groups, the patients in the controlgroup were matched to the study group(Larsen & Marx,  1986 ). This means thatthe control patients were selected insuch a way that both groups were asidentical as possible regarding age, gen-der, number of inserted implants, andtime of implant insertion.In the study group,  43 % ( 128 ) of all theimplants lost were early failures, and 57 % ( 73 ) were late failures. This is in ac-cordance with the results demonstratedby Esposito et al. ( 1998 a) in their meta-analysis of several studies, indicating 40 % early and  60 % late losses in maxil-lae that had not been bone grafted. Ac-cording to the same meta-analysis, halfof the late failures were lost during theirfirst year in function. The correspondingfigure for the present study was that al-most  2 / 3  of the late failures were lostduring their first year of function.One of the most important factors re-sponsible for implant failures is probablythe local anatomic structure regardingbone quality and quantity, or rather thelack thereof. In this study it is obviousthat there were better anatomic con-ditions (bone volume) in the controlgroup compared to the study group. Onefactor supporting this is that  16 . 5 %longer implants were inserted in thecontrol group compared to the studygroup (Table  2 ). Another supporting factis the finding that there were twice asmany implants with insufficient initialstability in the study group (Table  3 ).Furthermore, eleven  7  mm implantswere inserted in the study group com-pared to only one in the control group.Although some good results can be ob-tained with the use of shorter implants(Balshi et al.  1997 ), they seem to failmore often than longer ones – at leastin the maxilla (Jemt  1991 ). All of thesethings combined suggest that there wasbetter bone support in the control groupthan the study group, and that maybethere were better loading conditions inthe control group.Etiologic factors caused by surgicalfailures, such as bone overheating or in-fections, which were reported threetimes as often in the study group (Table 3 ), are reasons for early implant failures(Esposito et al.  1998 b). The present ma-terial did not allow any comparisons be-tween the different surgeons to evaluateif this might have influenced the results.However, all of the implants were in-serted by specialists in oral surgery, theprotocol for implant insertion has beenfollowed and abnormal incidences re-ported.Other host-related factors, such as thepatients’ general health conditions, mayhave played an important role in theearly implant failures witnessed in thestudy group. Examples of these healthconditions were diabetes, osteoporosis,ongoing medication etc. Even though areview of the literature according to Es-posito et al. ( 1998 b) failed to prove anevident relation between implant lossesand these factors, there seems to be aconsensus that some play an importantrole in the mechanism of implantfailure.Three patients in the study groupshowed signs of addiction to alcohol ornarcotics. This has earlier been pointedout as a reason for implant failure (Wey-ant  1994 ), but this study could not sub-stantiate this. According to the patientrecords, there were no patients withthese problems registered in the controlgroup.In this study, possible complicatingfactors were more common in the studygroup compared to the control group.There were, for example, four timesmore heavy smokers (more than  10  ciga-rettes/day) in the study group than in thecontrol group (Table  4 ). Only few studiesare presented in the literature pointingout the relation between heavy smokinghabits and implant failure (Bain & Moy 1993 ; De Bruyn & Collaert  1994 ), how-ever there seems to be a general consen-sus about the negative effects of smok-ing (Esposito et al.  1998 b).An extensive load over a long periodof time will often induce different com-plications in most fixed and removableprosthodontic reconstructions. Fracturesof abutment teeth and/or fractures of theprosthodontic material (Ekfeldt  1989 )are examples of these complications. Inan implant supported prosthesis it is ob-vious that an overload might induce im-plant failures, and therefore increase late  Ekfeldt et al . Factors associated with multiple implant failures in maxillae losses of implants (Balshi et al.  1997 ). Inthe literature there are indications of arelation between overload and late im-plant failure (Esposito et al.  1998 b). Thisstudy showed that where loading prob-lems were present, there were threetimes more late failures (Table  5 ).On the study form the clinicians gavetheir opinions of the reason for implantfailure, either based on their own experi-ence in treating the patient or fromstudying the patient’s record. This wasdone for all but four patients in the studygroup, and the answers could easily begrouped into five different topics. This isan interesting finding of the study, and itindicates the possibility that the treatingteam’s clinical experience could improvepatient selection. This, in turn, mightdecrease the number of implant failures.Furthermore, these five groups/factorsare all discussed in the literature, for re-view see Esposito et al.  1998 b.The experience gained from this andother studies indicates that there are cer-tain factors of importance in patientswhere a cluster phenomena of implantfailure occurs. Also the long-term clin-ical experience of the treating team isimportant for patient selection. How-ever, there is still a need for a proper andwell-designed multicenter, prospectivefollow-up study in order to analyze thesefactors and their impact on the outcomeof implant treatment before definite con-clusions can be drawn. Acknowledgements:  The authors wish toparticularly thank Ellen Lund for her as-sistance and support in reading and cor-recting the manuscript. Re´sume´ Cette e´tude re´trospective a e´te´ planifie´e pour e´tudierles facteurs influenc¸ant les e´ches implantaires. Six cli-niques de prothe`se en Sue`de ont participe´ a` cette e´tu-de apportant un total de  54  patients traite´s entre jan-vier  1988  et de´cembre  1996 . Tous ces patients e´taiente´dente´s supe´rieurs et ont rec¸u soit une prothe`se fixesoit une prothe`se amovible fixe´e a` au moins quatreimplants  ad modum  Brånemark A . La moitie´ des pa- 466  |  Clin. Oral Impl. Res.  12 ,  2001  /  462 – 467 tients e´taient du groupe d’e´tude et un crite`re d’inclu-sion pour ce groupe e´tait qu’ils avaient perdu aumoins la moitie´ de leur implants. Afin de diminuerles donne´es biaise´es les patients du groupe controˆlee´taient comparables a` ceux du groupe d’e´tude c.-a`.-d.qu’ils e´taient se´lectionne´s pour que les deux groupessoient aussi semblables que possible. Les re´sultats decette e´tude ont indique´ que le groupe controˆle avaitun meilleur support osseux initial que le test. De plusles patients du groupe test souffraient des circonstan-ces qui pouvaient induire l’e´chec implantaire commele bruxisme, des proble`mes personnels, la de´pressionainsi que l’abus de cigarettes, d’alcool et/ou de narco-tiques. Sur le questionnaire clinique, les cliniciens ontduˆ donne´ leur opinion sur la cause de l’e´chec implan-taire. Les re´ponses pouvaient facilement eˆtre groupe´esen cinq raisons diffe´rentes et cette expe´rience peutdonc eˆtre utile pour ame´liorer la se´lection des pa-tients. Cette e´tude sugge`re qu’il y a certains facteursd’importance a` conside´rer pour pre´venir les phe´nome`-nes d’e´checs implantaires de masse soit l’absence dusupport osseux, le tabagisme important et lebruxisme. Zusammenfassung Diese retrospective Studie wurde entworfen, um Fak-toren, welche Implantatmisserfolge beeinflussen, zuidentifizieren. Sechs prothetische Kliniken in Schwe-den nahmen an der Studie teil. Insgesamt konnten  54 Patienten, welche zwischen Januar  1988  und Dezem-ber  1996  versorgt worden waren, in die Untersuchungeinbezogen werden. Alle Patienten waren im Oberkie-fer zahnlos und erhielten entweder eine festsitzendeBru¨ckenrekonstruktion oder eine Hybridprothese,welche auf mindestens  4  Implantaten (Brånemark-Sy-stem A ) verankert wurde. Die Ha¨lfte der Patienten ge-ho¨rte zur Untersuchungsgruppe. Ein Einschlusskrite-rium fu¨r diese Gruppe war, dass die Probanden minde-stens die Ha¨lfte ihrer Implantate verloren hatten. UmVorurteile zu vermeiden, wurden die Patienten derKontrollgruppe einem entsprechenden Patienten derTestgruppe zugeordnet, d.h. sie wurden so ausgewa¨hlt,dass die beiden Gruppen so identisch wie mo¨glich wa-ren. Die Resultate der Studie zeigten, dass die Kon-trollgruppe initial ein besseres Knochenangebot alsTestgruppe aufwies. Des weiteren bestanden bei denPatienten der Untersuchungsgruppe Umsta¨nde wieetwa Bruxismus, perso¨nlicher Kummer, Depressionenund auch Konsum von Zigaretten, Alkohol und/oderBeruhigungsmitteln, welche Implantatmisserfolge in-duzieren ko¨nnen. Auf dem Untersuchungsblatt wur-den die Kliniker nach ihrer Meinung bezu¨glich der Ur-sache der Implantatmisserfolge gefragt. Die Antwor-ten konnten leicht in  5  verschiedene Gebieteeingeteilt werden. Diese Erfahrungen ko¨nnen hilfreichbei der Patientenauswahl sein. Die Ergebnisse dieserStudie lassen vermuten, dass gewisse wichtige Fakto-ren wie z.B. mangelndes Knochenangebot, starkesRauchen und Bruxismus existieren, welche in Be-tracht gezogen werden mu¨ssen, um einer Anha¨ufungvon Implantatmisserfolg vorzubeugen. Resumen Este estudio retrospectivo se disen˜o´ para verificar losfactores que influyen en los fracasos de los implantes.Seis clı´nicas de pro´tesis en Suecia participaron en elestudio, y en conjunto incluyeron un total de  54  pa-cientes tratados entre enero de  1988  y diciembre de 1996 . Todos los pacientes estaban completamenteede´ntulos en el maxilar, y recibieron una pro´tesis fijao una sobredentadura soportada por al menos  4  im-plantes (Sisterma Brånemark A ). La mitad de los pa-cientes pertenecı´an al grupo de estudio, y el criteriopara inclusio´n en este grupo fue que habı´an perdidoal menos la mitad de sus implantes. Para reducir latendencia, los pacientes en el grupo de control se equi-pararon al grupo de estudio, esto es, se seleccionaronde tal manera que los dos grupos fueran lo ma´s ide´nti-cos posible. Los resultados del estudio indican que elgrupo de control tuvo un mejor soporte o´seo inicialque el grupo de estudio. Mas aun, los pacientes delgrupo de estudio sufrieron circunstancias que pudie-ron inducir el fracaso de los implantes, tales comobruxismo, percances personales, depresiones, adema´sde adiccio´n a los cigarrillos, alcohol y/o narco´ticos. En el formulario de estudio se les pregunto´ a los clı´nicossu propia opinio´n de la razo´n del fracaso del implante.Las resupestas dadas pueden ser agrupadas en  5  to´pi-cos diferentes, y esta experiencia puede ser u´til paramejorar la seleccio´n de pacientes. Este estudio sugiereque hay ciertos factores de importancia a considerarpara prevenir un feno´meno en racimo de fracaso deimplantes, esto es, falta de soporte o´seo, tabaquismointenso y bruxismo.
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks