Clin Oral Impl Res 2000: 11: 144–153 Copyright C Munksgaard 2000Printed in Denmark ¡ All rights reserved

ISSN 0905-7161

A 5-year prospective clinical and radiographicstudy of non-submerged dental implants

Weber HP, Crohin CC, Fiorellini JP. A 5-year prospective clinical and Hans-Peter Weber,radiographic study of non-submerged dental implants. Constant C. Crohin,Clin Oral Impl Res 2000: 11: 144–153. C Munksgaard 2000. Joseph P. Fiorellini

Osseointegrated implants as anchors for various prosthetic reconstruc- Department of Restorative Dentistry,tions have become a predictable treatment alternative. It was expected that Department of Periodontology, Harvardimplants required submucosal placement during the healing period for School of Dental Medicine, Boston,successful tissue integration. However, it has been demonstrated that Massachusetts, USAhealing and long-term health of implants could be achieved with equalpredictability in a 1-stage, non-submerged approach. This prospective 5-year study not only calculates implant success by life table analysis, butalso evaluates the correlation between observed bone level changes withclinical parameters as measured by suppuration, plaque indices, bleedingindices, probing depth, attachment level and mobility. A total of 112ITIA dental implants were inserted in different areas of the jaws. Clinicaland radiographic parameters were evaluated annually for 5 years,whereas a portion of the study group for which 6-year evaluations wereavailable were included in the life-table analysis. The overall success rateafter 5 years in service was 99.1%, while after 6 years it was reduced to

Key words: dental implants –95.5% due to the fracture of 3 implants in 1 patient. The mean crestalosseointegrated – non-submergedbone loss experienced during the first year was 0.6 mm followed by animplants – prospective study – clinical

annual yearly loss of approximately 0.05 mm. No significant differences parameters – radiographic evaluationcould be found between the amount of bone loss measured at each of theyearly follow-up visits. This suggests that statistically the followed im- Hans-Peter Weber, Department ofplants did not show any radiographically measurable bone loss following Restorative Dentistry, Harvard School ofthe initial period of bone loss associated with implant placement and Dental Medicine, 188 Longwood Avenue,osseointegration. Low levels of correlation between the individual and Boston, MA 02115, USA

e-mail: hpweber/hms.harvard.educumulative clinical parameters with radiographically measured bone losssuggests that these parameters are of limited clinical value in assessing

Accepted for publication 22 March 1999and predicting future peri-implant bone loss.

Since landmark studies by Brånemark et al. (1977)and Schroeder et al. (1976, 1978), osseointegratedimplants as anchors for various prosthetic recon-structions have become a predictable treatmentalternative. It was proposed early on, that implantsrequired submucosal placement during the healingperiod for successful tissue integration (Bråne-mark et al. 1977; Adell et al. 1981). As a result, asecond surgery after an appropriate time of heal-ing was necessary. Later, it was demonstrated thathealing and long-term health of implants could beachieved with equal predictability in a 1-stage,non-submerged approach (Buser et al. 1990; Weberet al. 1996).

Longitudinal radiographic bone level measure-

144

ments in clinical studies have identified a meanbone loss pattern for 2-stage Brånemark type im-plants of 1 to 1.5 mm in the first year followingimplant insertion and less than 0.2 mm in sub-sequent years (Adell et al. 1981; Albrektsson et al.1986; Lindquist et al. 1988). Similar observationswere made for one-stage ITI implants with 0.75mm–1 mm crestal bone loss in year 1 and less than0.1 mm in the follow-up years (Weber et al. 1992;Pham et al. 1994; Bragger et al. 1998).

Although implants are predictable, long-termclinical studies indicate that failures do occur. Bio-logical implant failures defined by the loss or fail-ure to achieve osseointegration have been observed(Brånemark et al. 1977). The etiological factors

Prospective study of non-submerged dental implants

contributing to implant failure are not clearlyunderstood, however two etiological concepts havebeen proposed (Rosenberg et al. 1991; Tonetti &Schmid 1994). The first, loss of biomechanicalequilibrium by excessive load is evidenced by suchsigns as peri-implant bone loss, peri-implant radio-lucency, peri-implant suppuration and mobility.These are hypothesized to arise from the replace-ment of the highly specialized bone with a fibrouscapsule unable to contribute to the functional ca-pacity of the bone–implant interface. Second, in-fection has been proposed as an etiological agentinvolved in implant failure. An imbalance in thehost–parasite equilibrium can manifest itself in aseries of inflammatory changes leading to 2 dis-tinct clinical syndromes: peri-implant mucositisand peri-implant bone loss (Mombelli & Lang1998).

Several periodontal diagnostic parameters havebeen proposed as markers of health or disease(Tonetti et al. 1994; Bragger et al. 1996). Modifiedplaque and bleeding indices may be used to evalu-ate oral hygiene and muscosal inflammation. Prob-ing depth and mobility are also frequently con-sidered clinical parameters. The primary diagnos-tic tool used was radiographic assessment of boneheight changes to characterize the peri-implantconditions (Brånemark et al. 1977; Adell et al.1981). However, many studies did not obtain stan-dardized radiographs immediately after surgery asthere was concern that the radiation exposurecould adversely affect the degree of osseointegr-ation. Therefore, the bone loss measurements ob-tained in these studies have done little to describeearly bone changes around the implant.

Continuous evaluation of patients treated withosseointegrated implants is necessary to determinethe long-term success of the dental implant systemused, to ascertain factors affecting the success oftherapy and to identify method-specific problems.Clinical evaluation is important to optimize main-tenance, detect early signs of disease and plan cor-rective interventions. In addition, current academicand regulatory standards require a 3-year study

Table 1. Clinical parameters evaluated

Index Description Reference

Suppuration Presence or absence of suppurationModified plaque (mPI) Mombelli & Lang 1994Modified sulcus bleeding (mBI) Mombelli & Lang 1994Probing depth The depth of peri-implant probe penetration, recorded in mm Fiorellini & Weber 1994Attachment level The sum of the measured peri-implant probing depth and the distance from the Fiorellini & Weber 1994

implant crown margin to the coronal border of the peri-implant mucosa in mmKeratinized mucosa The width in mm of the keratinized tissue surrounding the implantPeriotestA (Siemens, Bensheim, Germany) value (PTV) A measure of the damping characteristics of the tissues surrounding the implant Buser et al. 1990

145

with an extended follow-up of 5 years as the mini-mum which allows the establishment of crestalbone loss patterns from surgical placement, initialfunctional loading and after 5 five years in func-tion.

The purpose of this study was to allow corre-lation of observed bone level changes with clinicalparameters as measured by suppuration, plaqueindices, bleeding indices, probing depth, attach-ment level and mobility as well as to contributeinformation on the value of these parameters aspredictors of long-term peri-implant tissue sta-bility. A hypothesis was used that the alveolarcrestal bone levels surrounding non-submergeddental implants will confirm the observed greatestpercent change in the first year post-surgical place-ment, but then exhibit little or no bone loss in sub-sequent years in function. Furthermore, it washypothesized that the clinical parameters: suppu-ration, plaque indices, bleeding indices, probingdepth, attachment level and mobility will correlatewith each other and with the radiographic boneloss observed.

Materials and methodsPatient enrollment and clinical proceduresAfter the study protocol was approved by the Har-vard Medical School Committee on HumanStudies, 46 patients were consecutively enrolled inthe prospective longitudinal trial. Patient’s in-formed consent, medical and dental history, dentalradiographs were taken and reviewed. A thoroughclinical examination was performed. Each patientwas then offered a comprehensive dental treatmentplan, with dental implants as option for the re-placement of missing teeth.

ITIA dental implants (Institut Straumann AG,Waldenburg, Switzerland) were inserted in differ-ent areas of the jaws, generally divided into groupsof anterior (inter-canine) mandible and maxilla,and posterior (distal to the canines) mandible andmaxilla. Out of a total of 112 implants placed, 77were of the hollow screw design while the remain-

Weber et al.

der were hollow cylinders. A strict surgical proto-col designed to minimize heat induced trauma tothe surgical site was followed. All implants wereplaced in a non-submerged manner (1-stage surgi-cal concept). After a healing period of 3 monthsfor cortical bone sites and 4 months for spongybone sites, the implants were loaded with eithersingle tooth replacements or short span fixed par-tial dentures of 3–4 units.

Clinical evaluation parametersStandardized intra-oral radiographs of the 112 im-plants placed were obtained at the following timepoints: immediately after surgery, 3–6 months aftersurgery at the time of functional loading and atyearly intervals following surgical placement.

Periapical radiographs as well as panoramicradiographs were exposed and developed understandardized conditions. The implants were clin-ically evaluated at yearly intervals at the mesial,distal, buccal and lingual/palatal aspect of eachimplant using the indices in Table 1. For the pur-poses of this study, the year 1, 3 and 5 clinical andradiographic data were analyzed.

Analysis of radiographic parametersThe developed radiographs were scanned on aPolaroid Sprintscan 35 PLUS (Polaroid, Cam-bridge, MA, USA) model digital optical slide scan-ner with an optical density of 3.4 Units OD. Thesize of the image was standardized at 675 d.p.i.with a resulting size on average of 940 by 620 pixelsat 8 bits per plane and 256 scales of gray. The re-sulting digital images of the radiographs were ana-lyzed using a validated method (Crohin 1998, the-

Table 2. Life table analysis of 112 ITI implants

Interval (years) Implants at start Drop-outs during Implants under risk Failures during interval Success rate Cumulative success rateof interval interval within group

0–1 112 1 111 0 100% 100%1–2 111 4 107 0 100% 100%2–3 107 0 107 1 99.1% 99.1%4–5 106 0 106 0 100% 99.1%5–6 56 1 55 2 96.4% 95.5%

Table 3. Implant failures

Implant . Design Location (WHO) Time to loss Bone loss (M) Bone loss ( D) Diagnosis

84 HC, 12 mm 11 3 years 7 months 4.45 mm 5.84 mm Peri-implant bone loss31 HC, 10 mm 24 5 years 2 months 1.57 mm 1.51 mm Fracture36 HS, 10 mm 46 5 years 7 months NA NA Fracture14 HS, 10 mm 36 6 years 5 months 2.46 mm 1.84 mm Fracture

146

sis). The mesial and distal marginal bone levelswere calculated and recorded for each implantbeing studied. All assessments of peri-implantbone levels were scored by one calibrated exam-iner.

Statistical analysisDescriptive statistics, paired and unpaired Stu-dent’s t-tests and regression analysis were per-formed (StatviewA, SAS Institute, Cary, NC,USA). Alveolar crestal bone heights were com-pared for the proximal aspects of all implants aswell as the clinical parameters measured for thedifferent time periods. Unpaired Student’s t-testswere used to compare the alveolar crestal boneheight in the maxilla and mandible as well as inposterior and anterior regions. Simple andmultiple regression analyses were used to comparethe utility of clinical parameters in predicting fu-ture peri-implant bone loss. The cumulative im-plant success rates were calculated using the lifetable method described by Cutler & Ederer (1958).The life table analysis included the 6-year follow-up, as a number of the patients had passed thatevaluation with 1 patient exhibiting a dramaticchange (3 implant fractures in year 5–6 timeperiod).

ResultsThe life table analysis is shown in Table 2. An over-all 6-year cumulative success rate of 95.5% and acorresponding value of 99.1% at 5 years was ob-served. Of the 112 implants placed, 4 implantswere lost. Of the 4 lost, 3 were lost due to implantfracture. One was lost due to peri-implant bone

Prospective study of non-submerged dental implants

Fig. 1. Radiographic peri-implant bone loss.

loss. The mean crestal bone loss experienced by theimplants lost prior to fracture was 1.85 mm. Thedemographics of the failed implants are summar-ized in Table 3. Six patients with 6 implantsdropped out of the study.

Peri-implant bone level analysisUpon analysis of the peri-implant bone loss dataobtained for mesial and distal sites, no signifi-cant differences were detected between the valuesfor the mesial and distal sites. When peri-implantbone loss data obtained was analyzed by un-paired Student’s t-test, no significant differenceswere detected between years 1, 3 and 5 (Fig. 1).Peri-implant bone loss data obtained for man-dibular and maxillary sites were analyzed by un-paired Student’s t-test and revealed no significant

Table 4. Coefficient of determination (r 2) of year 1 parameters (simple regression)

As determinant of Bone loss AL PD BI KM PTV PI Suppuration

Y1 bone loss 0.015 0.026 0.004 0.002 i 0.0002 0.002 i 0.018Y3 bone loss 0.285 0.025 0.017 0.003 i 0.00001 0.008 0.041 i 0.034Y5 bone loss 0.393 0.032 0.004 0.022 0.007 0.019 0.03 NA

i: indicates an inverse relationship.

Table 5. Coefficient of determination (r 2) of year 3 parameters (simple regression)

As determinant of Bone loss AL PD BI KM PTV PI Suppuration

Y3 bone loss 0.011 0.14 i 0.014 i 0.004 i 0.044 0.002 NAY5 bone loss 0.679 0.017 0.004 0.042 i 0.0001 0.057 0.003 NA

i: indicates an inverse relationship.

147

differences in bone loss pattern between the jaws.Comparing peri-implant bone loss data for an-terior (intercanine) versus posterior areas, anANOVA statistical analysis revealed a significantdifference in year 3 in which anterior areas ex-perienced approximately twice the bone loss rateof posterior areas. However, the small size of theavailable anterior dataset (nΩ20) makes thepower of this observation less significant. Thesignificant increase in the bone loss in the year 3anterior group can be accounted for by the in-clusion of data from the single anterior implantlost due to peri-implant bone loss (mesial lossΩ4.45 mm, distal lossΩ5.84 mm), as well as thesmall size of the available anterior dataset.

Soft tissue parametersSuppuration was noted in less than 4% of the im-plants during year 1. In subsequent years, no sup-puration was noted. The implant lost due to peri-implant bone loss showed signs of suppuration atremoval, however the data was not included due toits loss between years 3 and 4.

The mean modified plaque index in the first yearwas 0.3. A rise to a mean of 0.45 is evident in year3 and continues at that level through year 5.

No significant differences were discovered be-tween the modified bleeding levels observed be-tween the 3 time intervals. However, a difference issuggested (PΩ0.06) between year 1 and year 3. Nomajor shifts were seen in the frequency distributionof the bleeding levels recorded at the 3 time points.

The measured probing depths showed an in-crease from a mean of 2.4 mm to 3 mm betweenyears 1 and 3. In year 5 the probing depths essenti-ally remained the same. The attachment levelshowed a consistent increase in value from a meanof 2.25 mm in year 1 to 2.45 mm in year 5 at a

Weber et al.

Table 6. Multiple regression analysis of 6 clinical parameters* as determinants of boneloss. Adjusted coefficient of determination (adjusted r 2)

As determinant of Year 1 clinical Year 3 clinical Year 5 clinicalparameters parameters parameters

Year 1 bone loss 0.001.

Year 3 bone loss 0.038. 0.034Year 5 bone loss 0.028 0.044 0.143

* Clinical parameters used: attachment level, plaque index, bleeding index, probingdepth, keratinized mucosa and PerioTestA values.

. Clinical parameters used: plaque index, bleeding index, probing index, keratinizedmucosa.

mean rate of 0.05 mm/year. It is important to notethat this is the same rate change observed with themeasured radiographic bone loss. The width of themeasured keratinized mucosa around the study im-plant decreased from a mean of 3.5 mm in year 1to 2.8 mm in year 5 at a rate of approximately 0.2mm/year.

Lastly mobility as measured by PeriotestA values(PTV) increased from a mean of ª2.4 PTV in year1 to ª1.95 PTV in year 5 at an approximate rateof 0.11 PTV/year.

Clinical correlationSimple regression analysis results are summarizedin Tables 4 and 5. The results in general indicatedan overall very low rate of correlation between theindividual clinical parameters and the radiographi-cally measured bone loss. Multiple regressionanalysis summarized in Table 6, demonstrated thatthe cumulative predicative power of the 6 clinicalparameters in regards to bone loss ranged from2.8% to 14.3%.

DiscussionPrevious controlled 3 to 10 year prospective longi-tudinal trials of the ITIA implant system have re-ported failure rates ranging from 0% to 5.2% (Ni-shimura et al. 1997; Behneke et al. 1997; Buser etal. 1997; Mericske-Stern et al. 1994; Weber et al.1992; Buser et al. 1999).

In this present study, an actuarial analysis(Table 2) revealed an overall cumulative successrate of 95.5% after year 6 and a correspondingvalue of 99.1% at 5 years. This confirms thecumulative 6 year success rate of 95.4% reportedby Buser et al. (1997). No ‘‘early failures’’ wereobserved in this study, which is consistent withearlier short-term studies of ITIA implants (Mer-icske-Stern et al. 1994; Buser et al. 1990). Obvi-ously, the 3 implant failures were experienced dueto implant fracture in the year 5–6 period of ser-

148

vice. All 3 fractures occurred in 1 patient whowas diagnosed with severe bruxism only after theimplants had been placed and restored withsingle unit or short span fixed partial dentures.The mean bone loss experienced by these im-plants prior to fracture was 1.85 mm, whichcorresponds to bone loss to just below the firstset of threads. These failures can most likely beattributed to metal fatigue of the hollow implantafter advanced bone loss and would demonstratethat in severe masticatory loading conditions,hollow implants are contraindicated. It may beassumed that the greater bending tendency ofhollow implants lead to increased bone loss andthen implant fracture in this bruxing patient.

The implant lost due to peri-implant bone lossshowed approximately twice the amount of boneloss prior to surgical removal. The two implantfailures experienced by this cohort confirm the 2clinically distinct types of failures described in theliterature (Rosenberg et al. 1991; Tonetti & Schmid1994).

Previous clinical trials of the ITIA dental im-plant system have reported crestal peri-implantbone loss values ranging from 0.75 mm–1 mm inyear 1 and less than 0.1 mm in the follow-up years(Weber et al. 1992; Pham et al. 1994; Bragger et al.1998). In this study, using an improved measuringtechnique (Crohin 1998, thesis) a mean of 0.6 mmbone loss was measured during the first year fol-lowed by an annual yearly mean bone loss of ap-proximately 0.05 mm. No significant differencescould be found between the amount of bone lossmeasured at each of the yearly follow-up visits.This may suggest that statistically this implant sys-tem did not show any radiographically measurablebone loss following the initial period of bone lossassociated with implant placement and osseoin-tegration (Fig. 1).

When evaluating the measured bone loss withthe jaws as a unit, a higher mean amount of boneloss was observed in the maxilla as compared withthe mandible, though this was not statistically sig-nificant. The same was observed in the Pham et al.(1994) study which included a similar patient co-hort as the one reported here, but at an earlier timeperiod. Evaluation of measured bone loss by re-gion, i.e. anterior (intercanine) vs posterior in thisdataset revealed a greater mean amount of boneloss in the anterior region, though this was onlystatistically significant in the third year followingimplant placement. The power of this differenceshould be considered, as the size of the availabledataset for anterior implants was small. The sig-nificant difference seen in the third year may bedue to the disproportionately large amount ofbone loss experienced by the implant lost due to

Prospective study of non-submerged dental implants

peri-implant infection (Table 3). The results ofother studies comparing the anterior to posteriorwere also complicated by the small amount ofavailable data (Buser et al. 1997).

When clinical parameters were evaluated, sup-puration was reported for 2 implants (3% of thetotal). Both implants were successfully treated andsuppuration did not reoccur. Plaque index in-creased with time (Fig. 1). This suggests that thehigh level of oral hygiene maintained by the pa-tients in year 1 eroded with time. These obser-vations confirm the studies by Behneke et al.(1997) and Nishimura et al. (1997) who also found

Fig. 2. Frequency distribution for plaque index.

Fig. 3. Frequency distribution for modified sulcus bleeding.

149

Fig. 4. Probing depth in mm.

Fig. 5. Attachment level in mm.

an increase in plaque index over time. Linqvist etal. (1998) reported that an increase in plaque indexincreases the risk of peri-implant bone loss. How-ever, this study found that the correlation was veryweak (Æ4% or less).

For the majority of implants peri-implantbleeding was absent and remained so. However, asmall group showed shifts in their bleeding ten-dency (Fig. 2). None of the differences seen weresignificant. Mericske-Stern et al. (1994) andBehneke et al. (1997) reported the same shift inbleeding. Nishimura et al. (1997) confirmed thatthe majority of implants showed absence ofbleeding as seen in this present study (Fig. 3). Thepresent results could not confirm the partially sig-nificant decrease in plaque and bleeding indices

Weber et al.

during the 5 year period reported by Mericske-Stern et al. (1994). When the correlation betweenbleeding and plaque scores were examined, the as-sociation was weak, 1.8%, 14.7% and 11.5% inyears 1, 3 and 5, respectively.

Peri-implant probing values showed a significantincrease in depth after the first year, but then sta-bilized (Fig. 4). Behneke et al. (1997) showed asimilar significant increase, while Mericske-Sternet al. (1994) reported only a trend towards greaterprobing depths, but not a significant difference.

Attachment level (AL) values showed a steadyincrease with time, although no statistically sig-nificant differences could be measured (Fig. 5).Buser et al. (1991) also reported this increase. Wecould not confirm the statistically significant in-

Fig. 6. Width of keratinized mucosa in mm.

Fig. 7. Keratinized mucosa (buccal vs lingual).

150

Fig. 8. Periotest (r) values.

crease in AL observed by Behneke et al. (1997).The increase over 5 years was approximately 0.25mm. Due to its small magnitude, it is unlikely thatthis difference will be clinically observable withoutmore sophisticated measuring instrumentation.

The overall amount of peri-implant keratinizedmucosa decreased significantly (Figs 6, 7). We ob-served decreases in both buccal and lingual kera-tinized mucosa, although the rate of loss of lingualkeratinized mucosa was more than twice that ofthe buccal (Fig. 7). Behneke et al. (1997) reporteda similarly decreased width in oral keratinized mu-cosa, but also reported an increase in buccal width.Mericske-Stern et al. (1994) also reported this buc-cal increase.

Periotest values also increased with time thoughthe increase was not statistically significant (Fig.8). Studies by Behneke et al. (1997) and Nishimuraet al. (1997) confirm the lack of significant changein Periotest values over time. Isidor (1998) showedthat in clinically stable implants, the use of Perio-test values is of little additional value in assessingthe stability of an implant as compared to manualmobility assessments.

The usefulness of the measured clinical indicesin predicting future peri-implant bone loss(Tables 4 and 5) revealed that all the clinical par-ameters obtained showed very low levels of cor-relation. Bone loss in year 1 correlated moststrongly with bone loss seen in year 5 (rΩ0.393).However, AL, the next closest parameter, corre-lated a level of 10 orders of magnitude lowerwith bone loss. Bone loss in year 3 correlatedstrongly with bone loss seen in year 5 (rΩ0.679).Multiple regression analyses showed that thecumulative predictive power of the clinical par-

Prospective study of non-submerged dental implants

ameters could at best predict bone loss in 14.3%of the implants studied.

This study confirms that the bone loss rates ofITIA implants placed in a 1-stage protocol are wellwithin the clinically acceptable parameters estab-lished as success by previous studies. Both the rateand distribution of the observed bone loss corre-late with other investigations. The low levels ofcorrelation between the individual and cumulativeclinical parameters with radiographically meas-ured bone loss, suggests that these measures areof limited clinical value in assessing and predictingfuture peri-implant bone loss.

AcknowledgementsThis study was supported by a grant from the ITI Foundation,Waldenburg, Switzerland.

ResumeDes implants osteointegres servant d’ancrages pour des re-constructions prothetiques variees sont devenus un traitementalternatif previsible. Les implants requeraient le placementsous-muqueux durant la periode de guerison pour une inte-gration tissulaire a succes. Cependant, il a ete demontre quela guerison et la sante a long terme d’implants pouvaient etreobtenues avec une prediction semblable lorsqu’une approchenon-enfouie et en une etape etait appliquee. Cette etude pros-pective de cinq annees non seulement calcule le succes del’implant par l’analyse de table de vie, mais evalue egalementla relation entre les variations des niveaux osseux observesavec les parametres cliniques mesures comme la suppuration,les indices de plaque et de saignement, la profondeur au son-dage, les niveaux d’attache et la mobilite. Un total de 112 im-plants ITIA ont ete places dans differents endroits desmachoires. Des parametres cliniques et radiographiques ontete releves chaque annee pendant cinq ans tandis qu’une par-tie de ce groupe d’etude a pu etre evaluer durant six annees.Le taux de succes general apres cinq ans etait de 99.1% maisil etait reduit a 95.5% apres six annees du a une fracture detrois implants chez un patient. La perte osseuse crestalemoyenne durant la premiere annee etait de 0.6 mm suivie parune perte osseuse annuelle d’environ 0.05 mm. Aucune diffe-rence significative n’a pu etre trouvee dans la quantite de per-te osseuse mesuree a chacune des visites de suivi. Ceci sugge-re que statistiquement les implants suivis n’ont pas montre deperte osseuse mesurable radiographiquement apres la periodeinitiale de perte osseuse associee avec le placement de l’im-plant et son osteointegration. De mauvais niveaux de correla-tion entre les parametres cliniques individuels et cumules etla perte osseuse mesuree radiographiquement suggerent queces parametres ont une valeur clinique limitee dans l’evalua-tion et la prevision de la perte osseuse paroımplantaire.

ZusammenfassungOsseointegrierte Implantate als Verankerung fur verschiedeneprothetische Rekonstruktionen stellen eine zuverlassige Be-handlungsvariante dar. Es wurde angenommen, dass Implanta-te wahrend der Einheilperiode unter die Schleimhaut plaziertwerden mussen, damit es zu einer erfolgreichen Gewebeintegra-tion kommt. Es konnte jedoch gezeigt werden, dass eine guteEinheilung und gesunde Verhaltnisse um Implantate uber einenlangen Zeitraum mit gleicher Voraussagbarkeit auch mit einem

151

einzeitigen, transmukosalen Verfahren erreicht werden konnen.Diese prospektive Untersuchung uber 5 Jahre berechnet nichtnur den Implantaterfolg mittels einer Analyse der Ueberlebens-tabelle, sondern es werden auch Korrelationen zwischen denbeobachteten Veranderungen der Knochenhohe und klinischenParametern wie Suppuration, Plaqueindizes, Blutungsindizes,Sondiertiefen, Attachmentniveaus und Beweglichkeit ausgewer-tet. Insgesamt wurden 112 dentale ITIA-Implantate in verschie-denen Regionen der Kiefer eingesetzt. Die klinischen und radio-logischen Parameter wurden uber einen Zeitraum von 5 Jahrenjahrlich ausgewertet. Ein Teil der untersuchten Gruppe, fur wel-che 6-jahres Resultate vorhanden waren, wurde in die Analyseder Ueberlebenstabelle einbezogen. Die Erfolgsrate nach 5 Jah-ren in Funktion betrug 99.1%, wahrend sie nach 6 Jahren auf95.5% reduziert wurde, weil in einem Patienten 3 Implantatefrakturierten. Wahrend des ersten Jahres betrug der Verlust anmarginalem Knochen 0.6 mm. In den folgenden Jahren betrugder Verlust ungefahr 0.05 mm pro Jahr. Zwischen den Kno-chenverlusten, welche bei jeder der jahrlichen Nachuntersu-chungen registriert wurden, konnten keine signifikanten Unter-schiede gefunden werden. Dies lasst vermuten, dass statistischbetrachtet keines der verfolgten Implantate einen radiologischmessbaren Knochenverlust aufweist. Einzig in der initialen Peri-ode nach dem Setzen der Implantate findet ein Knochenverluststatt, welcher durch das Setzen der Implantate und die Osseoin-tegration verursacht wird. Zwischen den individuellen undkummulativen klinischen Parametern und dem radiologisch ge-messenen Knochenverlust konnten nur schwache Korrelationengefunden werden. Dies lasst vermuten, dass diese Parameter nureinen beschrankten klinischen Wert bei der Ermittlung undVorhersage von zukunftigem periimplantaren Knochenverlusthaben.

ResumenLos implantes osteointegrados como anclajes para varias re-construcciones prosteticas se han convertido en una alternati-va de tratamiento predecible. Se contaba con que los implan-tes requiriesen una colocacion submucosa durante el periodode cicatrizacion para una integracion tisular con exito. Decualquier modo, se ha demostrado que la cicatrizacion y lasalud a largo plazo de los implantes podrıa lograrse con igualpredictabilidad en un abordaje de una sola fase no sumergi-da. Este estudio prospectivo de 5 anos no solo calcula el exi-to de los implantes por un analisis de tabla de vida, sino quetambien evalua la correlacion entre los cambios del nivel dehueso observado con parametros clınicos medidos por supu-racion, indices de placa, indices de sangrado, profundidad desondaje, nivel de insercion y movilidad. Se insertaron un totalde 112 implantes dentales ITIA en diferentes areas de los ma-xilares. Se evaluaron los parametros clınicos y radiograficosanualmente durante 5 anos, mientras que una porcion del es-tudio de los que se disponıa una evaluacion de 6 anos se in-cluyeron en el analisis de tabla de vida. El indice de exitototal tras 5 anos fue de 99.1% mientras que tras 6 anos seredujo al 95.5% debido a fractura de 3 implantes en un pa-ciente. La perdida media de hueso crestal experimentada du-rante el primer ano fue de 0.6 mm seguida por una perdidaanual de aproximadamente 0.05 mm. No se encontraron dife-rencias significativas entre la cantidad de perdida osea medidaen cada una de las visitas anuales. Esto sugiere que estadısti-camente los implantes seguidos no mostraron perdida oseamensurable radiograficamente tras el periodo inicial de perdi-da osea asociado con la colocacion del implante y la osteoin-tegracion. Los bajos niveles de correlacion entre los parame-tros individuales y acumulativos con la perdida osea medidaradiograficamente sugieren que estos parametros son de limi-tado valor clınico en la evaluacion y la prediccion futura deperdida osea periimplantaria.

Weber et al.

ReferencesAdell, R., Lekholm, U., Rockler, B. & Brånemark, P.I. (1981)

A 15-year study of osseointegrated implants in the treatmentof the edentulous jaw. International Journal of Oral Surgery10: 387–416.

Albrektsson, T., Zarb, G., Worthington, P. & Eriksson, A.R.(1986) The long-term efficacy of currently used dental im-plants. A review and proposed criteria of success. Interna-tional Journal of Oral and Maxillofacial Implants 1: 11–25.

Behneke, A., Behneke, N., d’Hoedt, B. & Wagner, W. (1997)Hard and soft tissue reactions to ITI screw implants: 3-year longitudinal results of a prospective study. InternationalJournal of Oral and Maxillofacial Implants 12: 749–757.

Bragger, U. (1994) Radiographic parameters for the evaluationof peri-implant tissues. Periodontology 2000 4: 87–97.

Bragger, U., Hugel-Pisoni, C., Burgin, W., Buser, D. & Lang,N.P. (1996) Correlations between radiographic, clinical and

152

mobility parameters after loading of oral implants with fixedpartial dentures. Clinical Oral Implants Research 7: 230–239.

Bragger, U., Hafeli, U., Huber, B., Hammerle, C.H.F. & Lang,N.P. (1998) Evaluation of postsurgical crestal bone levels ad-jacent to non-submerged dental implants Clinical Oral Im-plants Research 9: 218–224.

Brånemark, P.I., Hansson, Bo., Adel, R., Breine, U., Lindstrom,J., Hallen, O. & Ohman, A. (1977) Osseointegrated implantsin the treatment of the edentulous jaw. Experience from the10-year period. Scandinavian Journal of Plastic and Recon-structive Surgery 11 (16 Suppl): 1–132.

Buser, D., Weber, H.P., Bragger. U. & Balsiger, C. (1991) Tissueintegration of one-stage ITI implants: 3-year results of alongitudinal study with Hollow-Cylinder and Hollow-Screwimplants. International Journal of Oral and Maxillofacial Im-plants 6(4): 405–412.

Buser, D., Weber, H.P. & Lang, N.P. (1990) Tissue integrationof non-submerged implants. Clinical Oral Implants Research1: 33–40.

Buser, D., Mericske-Stern, R., Bernard, J.P., Behneke, A.,Behneke, N., Hirt, H.P., Belser, U.C. & Lang, N.P. (1997)Long-term evaluation of non-submerged ITI Implants. Clin-ical Oral Implants Research 8: 161–172.

Buser, D., Mericske-Stern, R., Dula, K. & Lang, N.P. (1999)Clinical experience with one-stage, non-submerged titaniumdental implants. Advances of Dental Research (in press).

Crohin, C. (1998) A five year prospective study of non-sub-merged dental implants. D.M.D.-thesis, Harvard School ofDental Medicine, Boston, MA.

Cutler, S.J. & Ederer, F. (1958) Maximum utilization of the lifetable method in analyzing survival. Journal of Chronic Dis-ease 6: 699–712.

Fiorellini, J.P. & Weber, H.P. (1994) Clinical trials on the prog-nosis of dental implants. Periodontology 2000 4: 98–108.

Isidor, F. (1998) Mobility assessment with the periotest systemin relation to histologic findings of oral implants. Interna-tional Journal of Oral and Maxillofacial Implants 3: 377–383.

Jeffcoat, M.K. & Williams, R.C. (1984) Relationship betweenlinear and area measurements of radiographic bone levelsutilizing simple computerized techniques. Journal of DentalResearch 19: 191–198.

Lindquist, L.W., Rockler, B. & Carlsson, G.E. (1988) Bone re-sorption around fixtures in edentulous patients treated withmandibular fixed tissue-integrated prostheses. Journal ofProsthetic Dentistry 59: 59–63.

Mericske-Stern, R., Steinlin-Schaffner, T., Marti, P. & Geering,A.H. (1994) Peri-implant mucosal aspects of ITI implantssupporting overdentures. Clinical Oral Implants Research 5:9–18.

Mombelli, A. & Lang, N.P. (1994) Clinical parameters for theevaluation of dental implants. Periodontology 2000 4: 81–86.

Mombelli, A. & Lang, N.P. (1998) The diagnosis and treatmentof peri-implantitis. Periodontology 2000 17: 63–76.

Nishimura, K., Itoh, T., Takaki, K., Hosokawa, R., Naito, T. &Yokota, M. (1997) Periodontal parameters of ossointegrateddental implants. Clinical Oral Implants Research 8: 272–278.

Pham, A.N., Fiorellini, J.P., Paquette, D., Williams, R.C. &Weber, H.P. (1994) Longitudinal radiographic study ofcrestal bone levels adjacent to non-submerged dental im-plants. Journal of Oral Implantology 1: 26–34.

Quirynen, M., Naert, I., van Steenberghe, D. & Nys, L. (1991)A study of 589 consecutive implants supporting completefixed prostheses. Part I: Periodontal aspects. Journal of Pros-thetic Dentistry 68: 655–663.

Rosenberg, E.S., Torosian, J.P. & Slots, J. (1991) Microbial dif-ferences in 2 clinically distinct types of failures of osseointe-grated implants. Clinical Oral Implants Research 2(3): 135–144, July–Sept.

Schroeder, A., Pohler, O. & Sutter, F. (1976) Gewebsreaktion

Prospective study of non-submerged dental implants

auf ein Titan-Hohlzylinderimplantat mit Titan-Spritzschichtober-flache. Schweizerische Monatsschrift furZahnheilkunde 86: 713.

Schroeder, A., Stich, H., Straumann, F. & Sutter, F. (1978) Uberdie Anlagerung von Osteozement an einen belasteten Im-plantatkoerper. Schweizerische Monatsschrift fur Zahnheil-kunde 88: 1051.

Tonetti, M.S. & Schmid, J. (1994) Pathogenesis of implant fail-ures. Periodontology 2000 4: 127–138.

153

Weber, H.P., Buser, D., Fiorellini, J.P. & Williams, R.C. (1992)Radiographic evaluation of crestal bone levels adjacent tonon-submerged titanium implants. Clinical Oral ImplantsResearch 3: 181–188.

Weber, H.P., Buser, D., Donath, K., Fiorellini, J.P., Doppalapu-di, V., Paquette, D.W. & Williams, R.C. (1996) Comparisonof healed tissues adjacent to submerged and non-submergedunloaded titanium dental implants. A histometric study inbeagle dogs. Clinical Oral Implants Research 7: 11–19.