A Systematic Review of the Survival FDP_I_Implant Supported FDP

Bjarni E. Pjetursson Ken T an Niklaus P. Lang Urs Bra ¨gger Matthias Egger Marcel Zwahlen A systematic review of the survival and complication rates of fixed partial dentures (FPDs) after an observation period of at least 5 years I. Implant-supported FPDs Authors’ affiliations: Bjarni E. Pjetursson, Niklaus P. Lang, Urs Bra ¨gger, School of Dental Medicine, University of Berne, Berne, Switzerland Ken T an, National Dental Center, Singapore, Singapore Matthias Egger, Marcel Zwahlen, Division of E
of 18
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
  A systematic review of the survival andcomplication rates of fixed partialdentures (FPDs) after an observationperiod of at least 5 years I. Implant-supported FPDs Bjarni E. PjeturssonKen TanNiklaus P. LangUrs Bra¨gger Matthias Egger Marcel Zwahlen Authors’ affiliations: Bjarni E. Pjetursson, Niklaus P. Lang, Urs Bra¨gger  ,School of Dental Medicine, University of Berne,Berne, Switzerland Ken Tan , National Dental Center, Singapore,Singapore Matthias Egger, Marcel Zwahlen , Division ofEpidemiology and Biostatistics, Department ofSocial and Preventive Medicine, University ofBerne, Berne, Switzerland Matthias Egger  , MRC Health Services ResearchCollaboration, Department of Social Medicine,University of Bristol, Bristol, UK Correspondence to: Bjarni E. Pjetursson Department of Periodontology andFixed ProsthodonticsUniversity of BerneFreiburgstrasse7CH 3010 BerneSwitzerlandTel.: þ 41 31 632 2577Fax: +41 31 632 4915e-mail: Key words: biological complications, complication rates, failures, implant dentistry, long-itudinal, peri-implantitis, success, survival, systematic review, technical complications AbstractObjectives: The objective of this systematic review was to assess the 5- and 10-year survivalof implant supported fixed partial dentures (FPDs) and to describe the incidence ofbiological and technical complications. Methods: An electronic MEDLINE search complemented by manual searching wasconducted to identify prospective and retrospective cohort studies on FPDs with a meanfollow-up time of at least 5 years. Patients had to have been examined clinically at thefollow-up visit. Assessment of the identified studies and data abstraction was performedindependently by two reviewers. Failure and complication rates were analyzed usingrandom-effects Poisson regression models to obtain summary estimates of 5- and 10-yearsurvival proportions. Results: The search provided 3844 titles and 560 abstracts. Full-text analysis was performedfor 176 articles resulting in 21 studies that met the inclusion criteria. Meta-analysis of thesestudies indicated an estimated survival of implants in implant-supported FPDs of 95.4%(95percent confidenceinterval(95%CI):93.9–96.5%)after5 and92.8%(95%CI:90–94.8%)after 10 years. The survival rate of FPDs supported by implants was 95% (95% CI: 92.2–96.8%) after 5 and 86.7% (95% CI: 82.8–89.8%) after 10 years of function. Only 61.3% (95%CI: 55.3–66.8%) of the patients were free of any complications after 5 years. Peri-implantitisand soft tissue complications occurred in 8.6% (95% CI: 5.1–14.1%) of FPDs after 5 years.Technical complications included implant fractures, connection-related and suprastructure-related complications. The cumulative incidence of implant fractures after 5 years was0.4% (95% CI: 0.1–1.2%). After 5 years, the cumulative incidence of connection-relatedcomplications (screw loosening or fracture) was 7.3% and 14% for suprastructure-relatedcomplications (veneer and framework fracture). Conclusion: Despite a high survival of FPDs, biological and technical complications arefrequent.This, in turn, means that substantial amounts of chair time have to be accepted bythe clinician following the incorporation of implant-supported FPDs. More studies withfollow-up times of 10 and more years are needed as only few studies have described thelong-term outcomes. In earlier days, oral implants were mainlyused for the treatment of edentulous pa-tients, and numerous studies have reportedsuccessful outcomes for the rehabilitationofthesepatients (e.g., Adelletal. 1981). Astheyearspassedtheindicationsforimplanttherapy were broadened and today, themajority of patients receiving implants are Copyright r Blackwell Munksgaard 2004 Date: Accepted 30 June 2004 To cite this article: Pjetursson BE, Tan K, Lang NP, Bra¨gger U, Egger M,Zwahlen M. A systematic review of the survival andcomplication rates of fixed partial dentures (FPDs) afteran observation period of at least 5 years. I. Implant-supported FPDs. Clin. Oral Impl. Res. 15 , 2004; 625–642doi: 10.1111/j.1600-0501.2004.01117.x 625  partially edentulous (e.g., Lekholm et al.1994; Naert et al. 2002a, 2002b).Clinical research activities in implantdentistry have mainly focused on implantsurvival, and the incidence of biologicaland technical complications have been ad-dressed only to a minor extent. Besidesreportingonsurvivalproportions, occasion-ally success proportions were reported inthe studies. However, the latter depend onthe variety of criteria for implant successchosen by the investigators (Albrektssonet al. 1986; Buser et al. 1990; Karoussiset al. 2003) and hence, success is notidentically defined in all studies.In most studies, results of oral implantssupporting a variety of reconstructions,such as single crowns, fixed partial den-tures (FPDs), fixed total bridges and over-dentures, were usually reported withoutproviding detailed information on the nat-ure of the reconstructions.Longitudinal studies may be conductedin two different ways: (1) as prospective orretrospective cohort studies that are char-acterized by the fact that all patients arefollowed for the same observation periodand (2) as prospective or retrospective co-hort studies in which the follow-up time isnot the same for all patients. Such studiesmay include patients with various indivi-dual observation periods.Traditionally, life table analyses and Ka-plan–Maier statistics have been applied inimplant survival studies. Often patient co-horts were analyzed prospectively or retro-spectively with periods up to a certainobservation time. As an example, a long-term evaluation of non-submerged ITI s im-plants, reported an 8-year life table analysis(Buseretal.1997)inwhichthemeanfollow-uptimewas 3.1 yearsforindividualobserva-tion periods that ranged from 1 to 8 years.The outcome of implant therapy hasbeen presented in several traditional re-views (Cochran 1996; Fritz 1996; Espositoet al. 1998; Fiorellini et al. 1998; vanSteenberghe et al. 1999). In contrast toresults reported in traditional reviews,meaningful interpretation of survivalproportions and determination of the inci-dence of biological and technical com-plications would require a mean observa-tion period of at least 5 years. A recentsystematic review (Berglundh et al. 2002)excluded all studies not fulfilling the re-quirement of at least 5 years observationperiod of the reconstruction and concludedthatimplantlosspriortofunctionalloadingis to be expected on average in 2.5% of allimplants placed. However, implant lossduring function over 5 years occured in 2–3% of implants supporting fixed recon-structions, while in overdenture therapy,more than 5% of the implants may beexpected to be lost during a 5-year func-tional period. Hence, a 92–97% survival oforal implants may be expected over aperiod of at least 5 years.However, the longevity of the prostheticappliances on the implants has not yetbeen reviewed systematically. Therefore,the main objective of this systematicreview was to obtain robust estimates ofthe long-term survival of implant-sup-ported FPDs and of the incidence of speci-fic biological and technical complicationsin partially edentulous patients over anobservation period of at least 5 years. Material and methods Search strategy and study selection A MEDLINE (PubMed) search from 1966up to and including April 2004 was con-ducted for English-language articles pub-lished in the Dental Literature textsearching for ‘fixed partial dentures ORbridges’, ‘partial edentulism’, ‘implantsAND fixed partial dentures OR bridges’,‘implants’ AND ‘complications’, ‘im-plants’ AND ‘failures’, ‘implants’ AND‘longitudinal’.Manual searches of the bibliographies ofall full-text articles and related reviews,selected from the electronic search werealso performed. Furthermore, manualsearching was applied to the followingjournals for the years 2001–2003: ClinicalImplant Dentistry & Related Research,Clinical Oral Implants Research, Interna-tional Journal of Oral & MaxillofacialImplants, International Journal of Perio-dontics & Restorative Dentistry, Interna-tional Journalof Prosthodontics, Journalof Clinical Periodontology, Journal of Perio-dontology  and Journal of Prosthetic Den-tistry. From this extensive search, it was ob-vious that there were no randomized con-trolled clinical trials (RCTs) availablecomparing implant therapy with conven-tional reconstructive dentistry. Inclusion criteria In the absence of RCTs, this systematicreview was based on prospective or retro-spective cohort studies. The additional in-clusion criteria for study selection were:  that the studies had a mean follow-uptime of 5 years or more,  that the publications reported in Eng-lish and in the Dental literature,  that the included patients had beenexamined clinically at the follow-upvisit, i.e., publications based on patientrecordsonly, onquestionnairesorinter-views were excluded,  that the studies reported details on thecharacteristics of the suprastructures,  thatpublicationswhichcombined find-ings of both fixed partial dentures andsingle crowns described at least 2/3 ofthe reconstruction as FPDs. Selection of studies Titles and abstracts of the searches wereinitially screened by two independent re-viewers (B. E. P., K. T. or U. B.) forpossibleinclusion in the review. The full text of allstudies of possible relevance was then ob-tained for independent assessment by thetwo reviewers. Any disagreement was re-solved by discussion. Agreement betweenexaminers and reviewers was determinedusing k statistics (Fig. 1).Figure 1 describestheprocessofidentify-ing the 21 studies selected from an initialyield of 3844 titles. Data were extractedindependently by two reviewers using adata extraction form. Disagreement regard-ing data extraction was resolved by con-sensus. Excluded studies Of the 176 full-text articles examined, 155were excluded from the final analysis (seereference list).The main reasons for exclusion were amean observation period of less than 5years, no distinction between the type ofreconstructions or between totally and par-tially edentulouspatients, nodataavailablewith respect to FPDs, multiple publica-tions on the same patient cohorts withthe same observation period, and case de-scriptions of implant losses without rele-vant information on the entire patientcohort. Pjetursson et al . Systematic review of FPDs 626 | Clin. Oral Impl. Res. 15 , 2004 / 625–642  Data extraction Of the included 21 studies, information onthe survival proportions of the reconstruc-tions and on biological and technical comp-lications was retrieved. Biological complica-tions included disturbances in the functionof the implant characterized by a biologicalprocess affecting the supporting tissues.‘Peri-implantitis’ and ‘soft tissue complica-tions’ were included in this category.Technical complications denoted me-chanical damage of implants, implantcomponents and/or the suprastructures.Among these, ‘fractures of the implants,screws or abutments’, ‘fractures of thelutingcement’ (lossofretention), ‘fracturesor deformations of the framework or ve-neers’, ‘loss of the screw access holerestoration’ and ‘screworabutmentloosen-ing’ were included. From the includedstudies the number of events for all ofthese categories were abstracted and thecorresponding total exposure time of thereconstruction was calculated.Results on implant supported and com-bined tooth–implant-supported FPDs havebeen analyzed separately and have beenreported elsewhere (Lang et al. 2004). Statistical analysis By definition, failure and complication ratesare calculated by dividing the number ofevents (failures or complications) in thenumerator by the total exposure time (FPDtime and/or implant time) in the denomi-nator.The numerator could usually be ex-tracted directly from the publication. Thetotal exposure time was calculated by tak-ing the sum of:(1) Exposure time of FPDs/implants thatcould be followed for the wholeobservation time.(2) Exposure time up to a failure of theFPDs/implants that were lost due tofailure during the observation time.(3) Exposure time up to the end of obser-vation time for FPDs/implants thatdid not complete the observation per-iod due to reasons such as death,change of address, refusal to partici-pate, non-response, chronic illnesses,missed appointments and work com-mitments.For the evaluation of complication-freeFPDs the patient was used as unit ofanalysis. For these outcomes total patientexposure time was calculated by applyingthesameprocedures as for FPD- orimplantexposure time.Foreachstudy, eventratesfor FPDsand/or implants were calculated by dividing thetotalnumberofeventsby thetotal FPDsorimplant exposure time in years. For furtheranalysis, the total number of events wasconsidered to be Poisson distributed for agiven sum of implant exposure years andPoisson regression with a logarithmic linkfunction and total exposure time per studyas an offset variable were used (Kirkwood& Sterne 2003a, 2003b).Robust standard errors were calculatedto obtain 95 percent confidence intervals(95% CI) of the summary estimates of theevent rates. To assess heterogeneity of thestudy-specific event rates, the Spearmangoodness-of-fit statistics and associated P -value were calculated. If the goodness-of-fit P -value was below 0.05, indicating hetero-geneity, random-effects Poisson regression(with g -distributedrandomeffects)wasusedto obtain a summary estimate of the eventrates. Five-yearand 10-yearsurvivalpropor-tions were calculated via the relationshipbetween event rate and survival function S , S ( T  ) ¼ exp( À T  Â event rate), by assumingconstant event rates (Kirkwood & Sterne2003a, 2003b). The 95% CI forthesurvivalproportions were calculated by using the95% confidence limits of the event rates.Multivariable Poisson regression was usedto investigate formally whether event ratesvariedby study design (retrospective versusprospectivecohortstudies), yearofpublica-tion (1999 or before vs. 2000 or after), ormaterial (ceramicvs. gold–acrylicasveneermaterial). All analyses were performedusing Stata s (Stata Corporation, CollegeStation, TX, USA), version 8.2. First electronic search3844 Titles Independently selected by 2 reviewers640 titlesAgreed by both560 titlesAbstracts obtainedDiscussionAgreed on 151 abstractsFull text obtainedTotal full text articles176 Final number of studies included21 DiscussionDiscarded80 titlesKappa score 0.62Further handsearching 25 studies 67: Mean follow-up time less than 5 years.42: No detailed information on FPDs.18: No information on the reconstructions.14: Reporting on totally edentulous patients.5: Multiple publications on the samepatient cohort2: Reporting on failures without consideringthe entire patient cohort7: Reporting only on combined tooth-implant supported FPDs (Lang et al. 2004) Fig.1 . Search strategy. FPDs, fixed partial dentures. Pjetursson et al . Systematic review of FPDs 627 | Clin. Oral Impl. Res. 15 , 2004 / 625–642  Results Included studies A total of 21 studies of implant-supportedFPDs were included in the analysis. Thecharacteristics of the selected studies areshown in Table 1.All studies were published within thepast 10 years. These studies reported onlyon 16 differentpatientcohortsas fiveofthecohortswereexaminedafteranobservationperiod of both five and 10 years. Instead ofexcluding the 5-year data, the studies weredivided into two groups: A group witharound 5-year follow-up and a secondgroup with a 10-years follow-up time.Therefore, the same patients were not in-cluded twice in the same analysis. Sixteenof the studies were prospective and thefive remaining were retrospective studies(Table 1).The studies included a total of 1123patientsbetweentheageof 15 and 86 years.The proportion of patients with implantswho could not be followed for the completestudy period was available for 16 of the 21studies and ranged from 0% to 26%.Thestudiesreportedonfivecommerciallyavailable implant systems: Astra s TechImplants Dental System (Astra s Tech AB,Mo¨ldal, Sweden), Bioceram s sapphire im-plants (Kyocera America, Inc., San Diego,CA, USA), The Bra˚nemark s System (No-bel Biocare AB, Go¨teborg, Sweden), ITI s Dental Implant System (Straumann AG,Waldenburg, Switzerland) and Minimatic s implants (Minimatic Implants Technol-ogy, Boca Raton, FL, USA).The studies were mainly conducted inan institutional environment, such as uni-versity or specialized implant clinics. Fiveof the studies were multicenter studies.In four of the studies, the patients wererandomized into test and control groups tocompare implant-supported FPDs withcombined tooth–implant-supported FPDs(Olsson et al. 1995; Gunne et al. 1999),machined implant surfaces with TiO 2 -blasted implant surfaces (Gotfredsen &Karlsson 2001), titanium frameworkswith conventional cast frameworks (Jemtet al. 2003) and ceramic abutments totitanium abutments (Andersson et al.2003). In this review all the data from theincluded studies on implant-supportedFPDs were used irrespective of treatmentmodalities.The 21 studies included a total of 1336FPDs that were supported by 3578 oralimplants, and 299 of the FPDs were ana-lyzed both after a follow-up time of 5 and10 years (Table 2). Of the nine studies thatreportedonbridgedesign, 61% ofthe FPDs Table1 . Study and patient characteristics of the reviewed studies Study (year of publication) Implant systemSampling methodStudy design Plannedno. ofpatientsin studyActualno. ofpatientsin studyAgerangeMeanageSetting Drop-out(%) 5-year follow-up Preiskel & Tsolka (2004) Bra˚nemark implants Retrospective 44 44 64.2 University 0Andersson et al. (2003) Bra˚nemark implants Prospective 32 30 15–71 53 Universityand specialist6Multicenter, 3 centersRandomizationJemt et al. (2003) Bra˚nemark implants Prospective 42 35 25–74 53 Specialist clinics 17Multicenter, 4 centersRandomizationNaert et al. (2002a, 2002b) Bra˚nemark implants Retrospective n.r. $ 333 15–83 50.5 University n.a.Gotfredsen & Karlsson (2001) Astra implants Prospective 50 50 53 Universityand specialist0Multicenter, six centersRandomizationBra¨gger et al. (2001) ITI implants Prospective n.r. 33 23–83 55.7 University n.r.Mengel et al. (2001) Bra˚nemark implantstreated for GAPProspective 5 5 35–42 University 0Behneke et al. (2000) ITI implants Prospective 55 50 17–81 44.2 University 9Hosny et al. (2000) Bra˚nemark implants Retrospective n.r. 18 37–65 49.5 University n.r.Wennerberg & Jemt (1999) Bra˚nemark implants Retrospective 137 114 23–86 60.8 Specialist clinic 17De Leonardis et al. (1999) Minimatic implants Prospective 33 32 University 3O¨rtorp & Jemt (1999) Bra˚nemark implants Retrospective 58 52 28–77 59 Specialist clinic 10Wyatt & Zarb (1998) Bra˚nemark implants Prospective 88 77 15–72 45.1 University 13Olsson et al. (1995) Bra˚nemark implants Prospective 23 22 58 University 4RandomizationLekholm et al. (1994) Bra˚nemark implants Prospective 159 132 17–70 50 Universityand private17Multicenter, nine centers 10-year follow-up Bra¨gger et al. (2004) ITI implants Prospective n.r. 29 19–78 49.3 University n.r.Zarb & Zarb (2002) Bra˚nemark implants Prospective 30 25 University 17Attard & Zarb (2002) Bra˚nemark implants Prospective n.r. 35 20–65 45.1 University n.r.Lekholm et al. (1999) Bra˚nemark implants Prospective 127 94 18–70 50 Universityand private26Multicenter, six centersGunne et al. (1999) Bra˚nemark implants Prospective 23 20 58 University 13RandomizationFartash & Arvidson (1997) Bioceram sapphire implants Prospective 13 13 20–73 51.2 Specialist clinic 0 GAP, generalized aggressive periodontitis; n.r., not reported; n.a., not available. Pjetursson et al . Systematic review of FPDs 628 | Clin. Oral Impl. Res. 15 , 2004 / 625–642
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