biological factors contributing jan-michae´l hirsch2,

8/12/2019 Biological factors contributing Jan-Michae´l Hirsch2,

http://slidepdf.com/reader/full/biological-factors-contributing-jan-michael-hirsch2 1/25

Eur J Oral Sci 1998; 106: 527–551 Copyright © Eur J Oral Sci 1998

Printed in UK. All rights reserved EUROPEAN JOURNAL OF

ORAL SCIENCES

ISSN 0909-8836

Review

Marco Esposito1,2,Jan-Michae l Hirsch2,Biological factors contributing

Ulf Lekholm3, Peter Thomsen1

1Institute of Anatomy and Cell Biology,Goteborg University, Goteborg, 2Departmentto failures of osseointegrated

of Oral and Maxillofacial Surgery, UppsalaUniversity Hospital, Uppsala, and GoteborgUniversity, Goteborg, 3The Branemark Clinic, Faculty of Odontology, Go ¨ teborgoral implantsUniversity, Go ¨ teborg, Sweden

(I). Success criteria and epidemiology

Esposito M, Hirsch J-M, Lekholm U, Thomsen P: Biological factors contributing

to failures of osseointegrated oral implants. (I) Success criteria and epidemiology.

Eur J Oral Sci 1998; 106: 527–551. © Eur J Oral Sci, 1998

The aim of this review was to off er a critical evaluation of the literature and toprovide the clinician with scientifically-based diagnostic criteria for monitoring

the implant condition. The review presents the current opinions on definitions of

osseointegration and implant failure. Further, distinctions between failed and

failing implants are discussed together with the presently used parameters to

assess the implant status. Radiographic examinations together with implant

mobility tests seem to be the most reliable parameters in the assessment of the

prognosis for osseointegrated implants. On the basis of 73 published articles, the

rates of early and late failures of Branemark implants, used in various

anatomical locations and clinical situations, were analyzed using a metanalytic

approach. Biologically related implant failures calculated on a sample of 2,812

implants were relatively rare: 7.7% over a 5-year period (bone graft excluded).

The predictability of implant treatment was remarkable, particularly for partially

edentulous patients, who showed failure rates about half those of totally

edentulous subjects. Our analysis also confirmed (for both early and late failures)

the general trend of maxillas, having almost 3 times more implant losses than

mandibles, with the exception of the partially edentulous situation which

displayed similar failure rates both in upper and lower jaws. Surgical trauma

together with anatomical conditions are believed to be the most important

etiological factors for early implant losses (3.6% of 16,935 implants). The low

prevalence of failures attributable to peri-implantitis found in the literature

together with the fact that, in general, partially edentulous patients have less Marco Esposito, Institute of Anatomy andresorbed jaws, speak in favour of jaw volume, bone quality, and overload as the Cell Biology, Goteborg University,

Medicinaregatan 3, S-413 90 Go ¨ teborg,three major determinants for late implant failures in the Bra nemark system.SwedenConversely, the ITI system seemed to be characterized by a higher prevalence of Telefax: +46–317733308losses due to peri-implantitis. These diff erences may be attributed to the diff erentE-mail: [email protected] designs and surface characteristics. On the basis of the published

literature, there appears to be a number of scientific issues which are yet not Key words: dental implants;osseointegration; medical device failure;fully understood. Therefore, it is concluded that further clinical follow-up and

metanalysisretrieval studies are required in order to achieve a better understanding of the

mechanisms for failure of osseointegrated implants. Accepted for publication August 1997

Oral implants have revolutionized the practice of design. Reviews on currently used oral implantshave been presented (1, 2). Due to the long-termdentistry. Over the years, a large number of diff er-

ent implant systems have been introduced. As clinical use of osseointegrated oral implants, thescientific literature is extensive and continuouslyshown in Table 1, a classification of oral implants

can be based on placement modality and implant increasing. The present review will exclusively focus



528 Esposito et al.

Table 1

Classification of oral implants

Subperiosteal implants

Ramus frame implants a ‘‘hybrid’’ between subperiosteal and endosseous implants

Mpins and needles

Endosseous implants

blades and disksS

Qroot-formed analogues (screws, cylinders, hollow baskets, truncated cones andPcombination forms)

Transosteal implants mandibular staple bone plate and transmandibular implants

on osseointegrated systems (root-formed analogues; tion. These authors described osseointegration as‘‘a process in which a clinically asymptomatic rigidTable 1). Relevant articles dealing with various

osseointegrated implant systems are reviewed. fixation of alloplastic material is achieved andmaintained in bone during functional loading’’.However, the epidemiological data on implant fail-

ures are essentially based on the Branemark system, This is in contrast to implants surrounded byfibrous connective tissue (fibrointegration), whichdue to the fact that too little scientific information

is available regarding the long-term performances have shown a clinically discernible mobility whenloaded (11).of other implant systems (3, 4).

Many experimental and clinical studies have Obviously, it is of great interest to be able todefine osseointegration. According to the Dorlandfocused on the mechanisms of tissue integration

and the possibilities to secure long-term success. Illustrated Dictionary (1994, p. 1198) , osseointegr-ation is the direct anchorage of an implant by theClosely related to these issues are, however, the less

well-known biological processes which may con- formation of bony tissue around the implants with-out the growth of fibrous tissue at the bone-implanttribute to the failure of an implant.

The aim of the present article was to identify, interface. On the basis of previous suggested defini-tions, a provision of definitions from various view-the biological factors contributing to failures of

osseointegrated oral implants. The first part of this points was recently given by B (12)( Table 2 ).review includes: definitions of implant failures,

parameters for success/ failure evaluation, reliable Success, in general terms, can be defined as thegaining of what is aimed at. Therefore, to bediagnostic tools for the evaluation of implant condi-

tions and the epidemiology of implant losses. In considered successful, an osseointegrated oralimplant has to meet certain criteria in terms of the second part of the review, which will be pub-lished separately, the etiology and the patho- function (ability to chew), tissue physiology (pres-

ence and maintenance of osseointegration, absencephysiology of failing implants will be discussed.Only the English language literature was evaluated. of pain and other pathological processes) and user

satisfaction (aesthetics and absence of discomfort).Obviously, every single implant has to fulfill and

Definitions of osseointegration, successbe tested for all the defined success criteria, other-

and failurewise it should be considered as surviving . This termapplies to those implants which are still in function,The concept of osseointegration was developed bybut which have not been tested with respect toB (5) in the middle of the 1960s and ledsuccess criteria, or where neither the criteria forto the predictable long-term success of oralsuccess or failure are met (13, 14).implants. Osseointegration has been defined from

Consequently, a failure may be defined as thevarious viewpoints, including the description of first instance at which the performance of long-term clinical results, a numeric evaluation of the implant, measured in some quantitative way,interfacial mechanical capacity, and the morpholo-falls below a specified, acceptable level (15). Thisgical appearance of the tissue-implant interface ( 6) .definition of implant failure includes a great dealOne of the first definitions of osseointegration,of arbitrariness and encompasses a large variety of given by A et al . (7), was a ‘‘directclinical situations, ranging from all symptomaticfunctional and structural connection between livingmobile implants to implants showing more thanbone and the surface of a load-bearing implant’’.0.2 mm of peri-implant bone loss after the first yearSubsequently, several other definitions of osseoin-of loading (14, 16), or bleeding pockets exceedingtegration were presented (8, 9). According to Z5 mm of probing depth (17).& A ( 10), the only acceptable definition

of osseointegration is based on a clinical examina- As shown in Table 3, failures can be divided into



Factors contributing to implant failures 529

Table 2 A biological failure can be defined as the inad-equacy of the host tissue to establish or to maintainDefinitions of osseointegration (12)

osseointegration. Such failures can further be(a) From the viewpoint of the patient divided according to chronological criteria in earlyAn implant fixture is osseointegrated if it provides a stable and

(or primary) failures (failure to establish osseoin-apparently immobile support of a prosthesis under functional

tegration, i.e., an interference with the healingloads, without pain, inflammation or loosening over the lifetimeof the patient. process) and late (or secondary) failures (failure to

maintain the established osseointegration, i.e., pro-(b ) From a viewpoint of macro- and microscopic biology and medicine cesses involving a breakdown of osseointegration).Osseointegration of a fixture in bone is defined as the close One way to chronologically discriminate early fromapposition of new and reformed bone in congruence with the late failures may be to confine all implants removedfixture, including surface irregularities, so that at light micro-

before bridge insertion (or after an ‘‘appropriate’’scopic level, there is no interpositioned connective or fibroushealing period) to the group of early failures.tissue and that a direct structural and functional connection is

established, capable of carrying normal physiological loads Similarly, all failures occurring after the prostheticwithout excessive deformation and without initiating rejecting rehabilitation may belong to the group of latemechanisms.

failures. Obviously, this subdivision is not sufficient(c) From a macroscopic biomechanical point of view because there is not yet any non-invasive methodA fixture is osseointegrated if there is no progressive relative

which may accurately determine if and to whatmotion between the fixture and surrounding living bone andextent an implant is osseointegrated. Recently, newmarrow under functional levels and types of loading for the

entire life of the patient and exhibits deformations of the same techniques for a non-invasive evaluation of theorder of magnitude as when the same loads are applied directly bone-implant interface were described (18, 19),to the bone.

however, being still at experimental and clinical(d ) From a microscopic biophysical point of view trial stage levels.Osseointegration implies that at light microscopic and electron The division of biological failures into early andmicroscopic levels, the identifiable components of tissue within

late failures is an attempt to provide a simple anda thin zone of a fixture surface are identified as normal bonepractical sub-classification. In fact, this classifica-and marrow constituents which continuously grade into a

normal bone structure surrounding the fixture: that mineralized tion has been employed in most clinical studies andtissue is found to be in contact with the fixture surface over will also be used in the present review.most of the surface within nanometers so that no functionally

The loss of osseointegration is clinically mani-significant intervening material exists at the interface.fested by a peri-implant radiolucency and implantmobility. These signs are considered to arise fromthe replacement of the highly specialized bone tissue

biological failures (related to biological processes) with a fibrous connective tissue capsule, unable toand mechanical failures of the components (includ-contribute to the functional capacity of the bone-ing fractures of implants, coatings, connectingimplant unit.screws and prostheses). An iatrogenic failure can

From both theoretical and practical points of be defined as one characterized by a stable andview, the distinction between successful and failedosseointegrated implant, but due to malpositioning,implants on the one hand, and failing and failedit is prevented from being used as part of theimplants on the other appears crucial, as it isanchorage unit. This group also includes implantspossible that implants might fail in a slow andwhich have to be removed due to violation of gradual way. Therefore, if reliable symptoms oranatomical structures such as the inferior alveolarsigns distinguishing the early changes preceding thenerve. Another group of failures can be related toloss of osseointegration could be identified togetherinadequate or insu fficient patient adaptation (psy-with etiologic factors, a treatment modality mightchological, aesthetical and phonetical problems).

be provided.In the following, the review will only deal withbiological failures. In the literature, terms currently used to indicate

Table 3

Classification of oral implant failures according to the osseointegration concept

Biological:Early or primary (before loading): failure to establish osseointegrationLate or secondary (after loading): failure to maintain the achieved osseointegration

Mechanical: fracture of implants, connecting screws, bridge frameworks, coatings, etc.Iatrogenic: nerve damages, wrong alignment of the implants, etc.Inadequate patient adaptation: phonetical, aesthetical, psychological problems, etc.



530 Esposito et al.

failing implants or complications are: peri-implant thickness of the mucoperiosteum and consequentlymight not be correlated to implant success ordisease, peri-mucositis and peri-implantitis (14 ).

Peri-implant disease is a collective term for inflam- failure. Recently, at the 1st European Workshopon Periodontology (14), it was agreed that anmatory reactions in the soft tissues surrounding

functioning implants. Peri-implant mucositis is a absence of mobility, an average radiographic mar-ginal bone loss of less than 1.5 mm during the firstterm describing reversible inflammatory reactions

in the soft tissues surrounding a functioning year of function and less than 0.2 mm annually

thereafter, and absence of pain and/ or paresthesia,implant, whereas peri-implantitis refers to inflam-matory reactions with loss of supporting bone in were to be considered success criteria for osseointe-

grated implants. It was also suggested that probingthe tissue surrounding a functioning implant (14).These definitions may be regarded as working depths related to a fixed reference point and bleed-

ing on probing should be measured.definitions. It is likely that, if the causes andprocesses which lead to implant failure are unrav- Periodontal parameters were introduced to better

describe the state of peri-implant tissues in orderelled, the various definitions of failure may bemore precise. to provide clinical tools to identify failing implants.

Obviously, the discrimination between failing andFinally, a clear distinction should be madebetween failures and complications. Complications failed implants can be relevant not only for scientific

purposes (e.g., mechanisms of failure), but alsomight indicate an increased risk for a failure, butare either of temporary significance or amenable to because a cause-related therapy could thereby be

attempted, if a failing implant and its causativetreatment. In this paper, only established failuresand failing implants will be discussed. etiology can be identified.

Evolution of parameters used for success/ Parameters used for evaluating failedimplantsfailure evaluation

One of the first attempts to evaluate in an objective The evolution from anecdotal clinical observationsto scientific evidence requires quantification basedway the clinical performance of osseointegrated

oral implants was made by the Swedish National on the availability of parameters able to convertsubjective impressions into objective data ( 17). TheBoard of Health and Welfare in 1975 (20).

Periodontal (gingival index, plaque index and parameters, which have been employed clinicallyto evaluate implant conditions, will be discussedpocket depths), prosthetic (type of occlusion) and

radiographic parameters (absence of peri-implant with the attempt to identify the most reliable ones.

Ideally, it would be of great interest to diff erentiateradiolucency), together with the patient’s opinionon the treatment, were used. At the Harvard between parameters indicating failed and failingimplants, respectively. Some variations in theConsensus Conference in 1978 (21), success/ failure

criteria for all types of oral implants were assessment methodology have to be considered,since implants are characterized by diff erent designsestablished. Subsequently, A et al . (16)

proposed stricter criteria which have been generally and subjected to diff erent surgical procedures andloading regimens.accepted. At that time conventional periodontal

indices were not included, because they were not The most common diagnostic criteria employedfor the evaluation of established implant failuresconsidered to be correlated to implant success, and

this opinion was supported by S & Z (22) are the following:who did not consider it necessary to include anymeasure of mucosal health in their proposed criteria

Clinical signs of early infectionfor implant success. In 1988, at the second

Consensus Conference (23), it was concluded that During the healing period ( 3 to 9 months), com-plications such as swelling, fistulas, suppuration,diff erent success criteria should be applied to

diff erent implant systems. Unfortunately, no agree- early/ late mucosal dehiscences, and osteomyelitis,can occasionally be present and may indicatement was reached, and the success criteria of the

first conference in 1978 were maintained. implant failure. The most rational and commonexplanation for this is infection. However, earlyPeriodontal indices ( lack of redness, bleeding on

probing, probing from a fixed reference point) were wound dehiscences of submerged implants can alsobe seen in relation to poorly retained sutures,again proposed as clinical criteria to be used in

evaluating implant conditions (24). Further, inadequate flap adaptation, or premature wearingof a denture (25). Early signs may represent aspecific criteria for diff erent implant designs were

suggested. Consensus was expressed on excluding much more critical finding than if the same com-plication occurs later, because the healing process,probing depths, as that might be related to the




leading to the integration of the implant in the mobility of the underlying implant, the implantsurrounding bone, can be disturbed. Late super- must be suspected to be surrounded by a fibrousficial postoperative infection of soft tissues can tissue capsule. Mobility is always a clear signsometimes be attributed to a retained piece of of failure.suture material (25) or to insufficient tightening Several diff erent kinds of mobility have beenof the cover screw, and are generally uncomplicated, recognized: (1 ) rotation mobility; ( 2) lateral orwhile late deep infections are more serious. Late horizontal mobility; (3) axial or vertical mobility;

exposures of the implants, occurring after some as well as diff erent degrees of mobility (31).months of healing, are in most cases attributed Further, it has been suggested that slight changesto the pressure of dentures (25) in combination of horizontal mobility can be better evaluated bywith a thin alveolar crest, superficially buccally or means of electronic devices such as the Periotestlingually installed fixtures, and a thin mucosa.

(32–39). This procedure provides an objectiveIn a study monitoring 509 implants, a total of

method to measure the degree of mobility. It has36 soft tissue perforations over implants were

been shown that implant and abutment length, therecorded (26). 4 of these implants (11%) were not

type of jaw treated, as well as the bone densityintegrated at abutment connection. Other signs of have a major influence on Periotest values (40–42).infections were present around about 20 implants,However, despite some claims (35, 43), it remainsof which 6 (30%) turned to be early failures. Theseto be demonstrated if changes in Periotest scorespercentages were much higher than those based onare an early sign of implant failure, precedingimplants showing uneventful healing and calculated

radiographic changes.within the same study (2.8%). It has been suggested that implants should alsoConsequently, clinical signs of infectionsbe evaluated for possible rotational movementsobserved during the postoperative submerged(44, 45). Recently, it was, e.g., proposed to applyperiod may lead to an increased risk for implanta reverse-torque test, with forces not exceedingfailure, which does not seem to be as high as could

be feared. Therefore, signs of infection cannot be 10 Ncm, to every single implant at abutment con-used alone to determine the fate of an implant, but nection to discover mobile implants (28) . With thisshould be evaluated in conjunction with other procedure, an incidence of 4.7% of early failuresparameters such as radiolucency and mobility. In was reported. Interestingly, as will be describedthe absence of these signs of implant failure, clinical later, this figure slightly exceeded the average ratesigns of infection represent a complication, which, of 3.6% of early failures calculated with a metana-if left untreated, might lead to an implant failure. lytic approach (Table 8). Further, the incidence of

failures diagnosed by the reverse torque test wasPain or sensitivity 3× higher than that observed by F et al .

(46). In the report by S (28), an increasePain or discomfort is often associated with mobilityof the reverse-torque test to 20 Ncm was shown toand could be one of the first signs which indicatesreduce the number of late failures. However, thean implant failure (27–29). Pain or sensitivity whenfigures for early failures were not reported. It is yetchewing, tightening the abutment screw, or at per-unclear if this method is advisable or not due tocussion can be experienced, even under local anaes-the risk of inducing an iatrogenic fracture at thethesia. No studies have identified the structuralbone-implant interface. In addition, clinical obser-correlation to pain around implants. Interestingly,

failed implants can also be completely asympto- vations (26, 47) indicate that an increase in bonematic (30). Further, pain may reflect adverse tissue apposition may occur over time. Therefore, it isreactions not primarily related to implant mobility. not recommendable to apply a 20 Ncm reverseFor instance, in cases of implants placed over the torque to screen implants for rotational mobility.

mandibular canal, after nerve transposition or lat- Rotational mobility, in absence of vertical anderalization procedures, pain can reflect intraosseous horizontal mobility, may not necessarily be associ-oedema and pressure on the inferior alveolar nerve ated with the presence of a soft tissue capsule, butand may therefore be associated with perfectly might reflect a weak or immature bone/ implantstable implants. It has been suggested that persist- interface, particularly for implants placed in sitesent discomfort may be evident long before any

with limited bone volume or bone of low mineralradiographic changes (25). However, scientific

content. In contrast, horizontal and vertical mobil-evidence is still lacking.

ity invariably reflects bone loss and the presence of a peri-implant soft capsule (30 ). Consequently,

Clinically discernible mobility even though a simplified dichotomous (yes/ no)scale for mobility assessment might be preferableOnce the clinician has distinguished between the

mobility of a poorly connected abutment and the because it eliminates subjective interpretations,



532 Esposito et al.

some weakly osseointegrated implants might yet be radiolucency can, occasionally, be noted even incases of successful implants. In an evaluation of iatrogenically rotated and removed.

Occasionally, clinically discernible mobility can the accuracy and precision in the radiographicdiagnosis of clinical mobility, it was concluded thatbe present without distinct radiographic bone

changes (29 ). Therefore, mobility is the cardinal despite the relatively good diagnostic accuracy, theprobability of predicting fixture instability can besign of implant failure.low in a population showing a low prevalence of

mobile fixtures (62). However, in this latter study,Radiographic signs of failureimplant losses were analyzed over a 5-year period,and the positive predictive values ( 17%) were calcu-In general, intraoral radiographs are taken after

the abutment connection, in order to control that lated with a far too optimistic prevalence of earlyfixture loss of 1.5% according to F et al .abutments are properly seated. On the basis of

measurements on these radiographs, the baseline (46 ). If recalculating positive predictive values at aprevalence of implant failure of 10%, the predictab-value for future marginal bone changes can be

established. Standardized periapical radiographs ility of standardized intraoral radiographs definitelyimproves (about 61%). The above mentioned con-should be taken at regular follow-up intervals to

detect peri-fixtural radiolucency and/ or progressive clusions were modified by recent findings fromanother radiographic investigation ( 29). In fact,marginal bone loss or ‘‘saucerization’’ (48–52).

Although panoramic radiographs have been used the positive predictive value (i.e., the probabilitythat an implant was clinically mobile, when the(35, 53–56), this technique is of limited value in

monitoring implant conditions due to inferior radiographic evaluation had demonstrated signs of loss of osseointegration) was found to be muchimage resolution and the inability to modify the

angulation of the X-ray beam (51). In fact, while higher ( 83%). From a selected population of 482implants, 114 of the 138 implants which werepanoramic films are able of resolving about 5 line

pairs per mm, periapical films can clearly delineate radiographically reported as suspicious for loss of osseointegration, were found to be clinically mobile.at least 10 line pairs per mm (57).

There can be 2 well-distinct radiographic pic- Interestingly, the remaining 24 were regarded asstable. Unfortunately, as the authors pointed out,tures: a thin peri-fixtural radiolucency surrounding

the entire implant, suggesting the absence of a the study design did not permit evaluation of thetrue negative predictive value (the number of direct bone-implant contact and possibly a loss of

stability (29, 47, 58, 59), and an increased marginal implants which were mobile despite the absence of radiographic signs of implant failure), becausebone loss. In the first case, the implant is usually

found mobile when tested, whereas in the latter, patients without radiographic signs of lack of osseo-integration were not tested for implant stability.the fixture can be stable. It should be consideredthat an abnormal rate of marginal bone loss can Therefore, the radiographic examination remains

one of the primary tools for detection of failedalso be a sign of a mechanical failure (fracture of the implant) (26, 47, 60). Since the distinction implants in clinical routine, even though it is not

as accurate as the mobility test.between these 2 radiographic pictures is not alwaysclear, when a suspected peri-fixtural radiolucency The most important factors for making a proper

radiographic evaluation of the implant conditionsor excessive marginal bone loss is observed, it isrecommendable to remove the prosthetic construc- are the quality of the radiographs (29, 62–65)

together with the examiner’s experience (29, 62).tion and check the implants for stability. Clinicallydiscernible mobility after bridge removal can con-firm the presumptive radiographic diagnosis of

Dull sound at percussionimplant failure.

The resolution of the radiographic technique It has been suggested (22, 44) that a subdued soundupon percussion is indicative of soft tissue encap-together with the projection of anatomical struc-

tures could limit the detection of a thin soft tissue sulation, whereas a clear crystalline sound indicatessuccessful osseointegration. Once the clinician haslayer surrounding the fixture. This could explain

why, occasionally, barely perceptible clinical mobil- verified that the abutment is properly attached tothe implant, the test is conducted by hitting theity does not correspond to appreciable radiographic

changes. Conversely, due to the Mach band eff ect abutment with a loosely held metallic instrument.Although it is a rather subjective test without a(a visual phenomenon, where the borders of adja-

cent areas of diff erent photographic densities solid scientific background, it can provide a usefulindication to the examiner. It has also been sug-appear to have larger density diff erences than really

exist, i.e., the presence of soft tissue around the gested that a dull tone on percussion might bepresent long before radiographic signs of implantimplant is simulated) (49, 52, 61), peri-implant




failure (25). Unfortunately, no scientific evidence possible to verify an annual progression in thehas been presented yet to substantiate this range of 0.1 mm on radiographs (50 ).hypothesis. Some in vitro studies based on the Branemark

In conclusion, the cardinal sign of implant failure implant design provided information on the accu-is mobility. However, the necessity of removing the racy of periapical radiographs in the assessment of fixed prosthesis, which can be a cumbersome and marginal bone levels around implants. It was foundtime-consuming procedure, has made intraoral con- that stereoscopic radiographs were superior to

ventional radiography a valuable aid in determining single radiographs, which could not detect a gapthe success of oral implants in clinical routine. of 0.2 mm between the bone and implant, and that

the vertical angulation of the X-ray beam influ-enced, to a great extent, the accuracy of bone level

Parameters used for evaluating failing assessments (63). In particular, the deviation fromimplants a perpendicular projection relative to the long axis

of the implant should not exceed 9°. The impor-The clinical signs previously discussed emerge onlytance of a strict parallelism was stressed in anotherwhen the failure process reached an irreversibleinvestigation, and it was concluded that singlestate. However, the ideal parameter for monitoringradiographs were unreliable due to the great diffi-implant conditions should be sensitive enough toculties of obtaining serially identical radiographs indistinguish early signs of implant failure. The fol-clinical cases (64). However, the threads of thelowing parameters have thus been proposed.

Branemark implants were found useful in control-ling the identity of serial radiographs (67), butagain it was concluded that the accuracy obtainedRadiographic observed progressive marginal bonefrom the ideal in vitro situation might be difficultloss

to reproduce in the clinical situation. Another studyThere seems to be unanimous consensus that pro- confirmed that is extremely difficult to achieve validgressive marginal bone loss is a pathological sign

bone loss measurements below 0.2 mm (65).which can lead to implant failure. However, to

Several techniques for taking optimal intraoralwhat extent the marginal bone resorption should

radiographs have been described (48, 49, 52, 68).progress in order to advocate treatment, and which

However, the satisfaction of the requirements foris the most appropriate treatment procedure,

identical exposure geometry are very difficult toremains to be decided.

meet in clinical practice, particularly when com-One of the most commonly used success criteria

pared to the in vitro situation.

for the evaluation of marginal bone loss was Another important limitation of the radiographicproposed by A et al . (16, 22). Theseexamination is that only the interproximal aspects

authors suggested using less than 1.5 mm of mar-of the implant can be evaluated.

ginal bone loss during the 1st year of loading andBone loss around the neck portion of Branemarkthereafter less than 0.2 mm yearly as success cri-

implants has been subjected to numerous radio-teria. This concept was probably developed fromgraphic investigations (26, 27, 60, 66, 69–87).the radiographic findings on the mean marginalMarginal bone loss has been observed around otherbone loss around Branemark implants (60).implant systems as well (88–101), although withHowever, this parameter was meant to be employedsome diff erences. In fact, more marginal bone lossto evaluate individual implants (13). Recently, thiswas described among some other implant systems,statement has been slightly modified (14), addingwhen compared to Branemark implants (53–55,the word ‘‘average’’. It is interesting to observe that92, 102).despite the majority of clinical follow-up studies

It can be argued that marginal bone loss aroundreferring to the A et al . success criteria,the neck of osseointegrated implants is probablyfew authors actually apply the bone loss criterioninfluenced by the implant design, both in the short-to individual implants.(96, 103) and long-term period (54, 55, 92).Some authors doubt that a firm limit for anHowever, due to the complex interactions betweenacceptable annual bone loss can be established (26,surgically-induced trauma, stress distribution, mic-66). This doubt was based on reported higherrobiota and host response on the marginal boneamounts of bone loss, which stabilized after 2 or 3loss, the exact role played by the various implantyears without leading to loss of the implant. It hasdesigns and surface characteristics remains to betherefore been proposed that an implant should beunderstood. In this context, it should also beconsidered failed when the marginal bone loss hasobserved that even the diff erences in radiographicreached the apical 1/ 3 of the implant (27 ).

Moreover, from a technical point of view, it is not image characteristics of diff erent implant designs




Probing cannot easily be performed around all PAL is the sum of PPD and REC. It has beensuggested that increased measures of 2 mm or moreimplant or abutment designs. For example, probing

a cylinder is practical when the neck is continuous should be interpreted as resorption of alveolar bone(68, 114). The same practical limitations related towith the cylinder circumference, but can be difficult

around screws, particularly when bone resorption probe penetration and implant shape, as discussedabove, are evident. Even though this is a morehas reached the level of the threads (117, 118). It

has therefore been suggested that the sensitivity of accurate method to monitor implant health (123),

it still provides less precise information than radio-PPD might be low at implants compared to naturaldentition, due to the possible underestimation of graphs, particularly when tissues are inflamed or in

the presence of infra-bony craters (119). However,the true PPD values (117).The superstructure, in most of the cases, has to additional information on the tissue conditions at

buccal and lingual surfaces can be gained whenbe removed to permit probing as well. It has alsobeen suggested (17, 116) that implants should be comparing PAL with radiographs.designed in a way to permit probing.

Probing penetrates the peri-implant tissues withresultant damage (110, 111). The magnitude and

Crevicular fluid analysisthe clinical significance of such damage has yet tobe determined. Another direction pursued has been The analysis of crevicular fluids (composition and

flow rate), though able to distinguish betweenthe application of standardized forces. However,preliminary studies have indicated that the reprodu- healthy and inflamed sites, has not been demon-

strated as being capable of diff erentiation betweencibility was not necessarily increased (119).It can be concluded that increased pocket depth destructive and non-destructive inflammation.

Therefore, this technique cannot yet be used tocan be correlated with a higher degree of inflamma-tion of the peri-implant mucosa (110, 113, 119, identify failing implants (113, 124–136). However,

recent findings from a preliminary report seem to120), but not necessarily to bone loss (55).However, absolute PPD alone cannot be used as indicate that increased levels of interleukin-1b

might be associated with failing implants ( 134).an indicator of a pathological condition, sinceadditional factors, such as tissue thickness and These findings were not confirmed by another

controlled study (113).diff erent abutment lengths (68, 117, 121), mayinfluence the PPD assessments around implantswhen compared with teeth. Progressive PPD overtime might therefore be a better indicator for failing

Microbiological sampling

implants than absolute probe measurements.Conversely, other authors consider a deep pocket There is not yet a complete agreement on perio-dontal pathogens, even though some speciesa protected habitat for putative pathogens, which

is a clear indication of peri-implant pathology (for example Porphyromonas gingivalis andActinobacillus actinomycetemcomitans) are heavily(15, 116).

It remains also to be established if probing can associated with periodontal breakdown (137–143).A similar flora as that present in gingival pocketsdistinguish failed implants, characterized radio-

graphically by a thin peri-fixtural radiolucency. of teeth in health and disease has been observedaround implants, though with some diff erences inpartially- and fully-edentulous patients (68, 69, 109,

Mucosal recession (REC)110, 113, 124, 125, 144–159). However, there is notany convincing evidence yet that laboratory testsREC can be defined as the linear distance between

the location of the free mucosal margin and a fixed or commercially-available kits for identification of

suspected periodontal pathogens are of any use inlandmark (68, 122). When threads or a roughimplant surface become exposed, it might be diffi- the implant situation. In fact, it is yet unknown

whether these suspected pathogens induced bonecult to maintain the area clean from plaque andthe prognosis of the implant may become question- loss, or whether they simply found an ideal habitat

in infrabony pockets which may have been initiatedable. Otherwise, recession is mainly an aestheticalproblem and not an indication of failing implants. by overloading.

One exception might be the rare condition of osteomyelitis around implants, where the causative

Probing ‘‘attachment’’ levels (PAL)bacterial species must be identified, and an eff ectiveantibiotic treatment administered. The use of PAL are probing depths related to a fixed reference

point on the implant in order to monitor ‘‘attach- microbiological sampling is, otherwise, yet confinedto research situations.ment’’ loss over time (17, 114, 119). In other words,



536 Esposito et al.

diff erent implant characteristics (material, coating, Absence of keratinized mucosa

roughness, shape, surgical and prosthetic proced-ures, etc.). On the other hand, with this approach,A relationship in correlation between implant fail-the generalization to other implant systems is lim-ure and absence of an adequate band of keratinizedited. For a comparative analysis of failure patternsmucosa surrounding the abutment has been sug-among diff erent osseointegrated oral implant sys-gested ( 160). In particular, authors have directlytems, please refer to E et al . (4).attributed the reason for some late losses to the

Studies which did not include complete data onlack of keratinized mucosa (35, 161–164). Oneimplant losses were excluded or only partiallyhypothesis behind this idea is that the keratinizedemployed. From 159 follow-up investigations pub-tissue is more resistant to inflammatory destructivelished through February 1997 on the Branemarkprocesses induced by the oral microbiota. However,system, 73 articles were selected. The excludedthere is no scientific evidence supporting this hypo-articles did not provide detailed information onthesis (35, 60, 69, 70, 97, 110, 122, 124, 125,how many implants were placed in each jaw, did165–167). In conclusion, keratinized mucosa doesnot diff er between early and late implant losses,not appear to be related to implant failure.did not specify the follow-up periods properly, orHowever, the presence of keratinized tissue mightpresented the same patient materials already avail-facilitate the patient’s hygienic procedures.able in other publications. Failure data regardingIn conclusion, while it is possible to clearlythe ‘‘development group’’ (58, 60, 172) or thediff erentiate between a successful and a failed‘‘casuistry group’’ (173) were also excluded, sinceimplant, it still remains difficult to identify failingthese investigations were based on prototypes of implants. In this context, the distinction between atitanium implants having diff erent designs com-failed implant, characterized by a thin peri-fixturalpared to the Branemark implants used today.radiolucency and mobility, and a failing implant,However, failure rates of these studies, with onecharacterized by progressive marginal bone loss,exception (173), were much higher than thoseclinical signs of peri-implant infection and absencehereby presented. The outcome of implants placedof discernible mobility, might disclose a diff erentin irradiated patients will be discussed separatelyetiology (overload and peri-implantitis). This dis-in the 2nd part of the review.tinction seems justified in light of experimental

One rationale for analyzing failures instead of (168, 169) and clinical (30, 170) findings.success is that smaller diff erences are moreObviously, both etiological factors can interact withpromptly recognized. For example, if a 96% successeach other, resulting in a variety of intermediaterate of a certain technique is compared with a 98%situations.

success rate of another technique, almost no diff er-So far, there is substantial agreement and scient- ence may be noticed. Conversely, using a failureific evidence that bone level assessments are morerate approach, it is obvious that the former modal-reliable with a well-performed intraoral radio-ity fails twice as often than the second.graphic examination than with all the conventional

Fixtures removed for mucosal disorders (sus-periodontal indices (22, 31, 55, 111, 117, 121, 171).pected peri-implantitis cases), but still clinicallyHowever, radiographic findings should always bestable, were recorded as failures in our analysis.carefully evaluated in conjunction with other clin-Mechanical and iatrogenic failures, i.e., implantsical parameters.which had fractured, or were not used as abutments(‘‘sleeping implants’’), were not counted as biolo-

Frequencies of early and late failuresgical failures. It should be emphasized that thetopic of this analysis is the fate of the individualIn order to provide data on the prevalence and

incidence of biological failures in diff erent clinical implant rather than the outcome of prosthetic

reconstructions. Therefore, in terms of functioningsituations, a large number of clinical follow-upstudies were analyzed. In this section, only investi- bridges, failure rates are lower.

In Table 4, a summary of follow-up studies,gations dealing with the Branemark system wereevaluated. This does not mean that other implants providing data on implant losses in fully edentulous

patients, is shown. Data are subdivided in 2 groups:cannot achieve similar or even better results; how-ever, since the Branemark system can be considered patients wearing fixed bridges and overdentures,

respectively. The higher failure rates in the overden-the reference implant for osseointegrated oralimplants, more scientific reports are available on ture group, both for early and total number of

losses (5.9 and 12.8%, respectively), particularly inthis specific system (3, 4). Moreover, it was judgedthat it was easier to draw conclusions if only one Swedish studies (75, 80, 174–176), may be

explained by the fact that this treatment alterna-well-characterized implant system was analyzed,thus avoiding confounding factors derived from tive was mainly used in critical situations. Such



Table 4

Studies reporting on failure rates for Bra nemark implants in totally edentulous patients wearing fixed prost

Implants Implant location Authors (ref. no.) inserted/ failed* maxilla/ mandible

Fixed prosthesesA et al., 1981 (60) 230/ 37 (9.1) 100/ 130 21(1L et al ., 1987 (198) 42/ 2 (4.8) maxilla 2/ 0A et al ., 1990 (199) 269/ 9 (1.9) 82/ 187 5(5;J & P, 1990 (200) 286/ 25 (5.2) 112/ 174 15(1

A et al ., 1990 (201) 869/ 18 ( 1) mandible 9/ 9T et al ., 1991 (202) 45/ 2 (0) atrophic mand. 0/ 2J, 1991 (203) 125/ 4 (3.2) maxilla 4/ 0K & B, 1991 (204) 154/ 3 (2) mandible 3/ 0Q et al ., 1991 (66) 589/ 30 (2.5) 269/ 320 15(1T & L, 1992 (205) 1230/ 64 (2.2) 292/ 938 27(1E et al ., 1994 (78) 63/ 2 (3.2) mandible 2***J, 1994 (77) 449/ 31 (3.3) maxilla 15/ 1O et al ., 1995 ( 206) 563/ 27 (0.5) 200/ 363 3(3;J & L, 1995 (80) 655/ 64 (3.5) maxilla 23/ 4H et al ., ( 207) 837/ 0 (0) mandible 0/ 0L et al ., 1996 (87) 272/ 3 (0.7) mandible 2/ 1Z & S, 1996 (208) 259/ 32 ( 8.1) 26/ 233 21(0B et al ., 1997 (209)**** 425/ 31 (0.9) 370/ 55 4(4;

Subtotal fixed prostheses 6609/ 384 (2.6) 2722/ 3887 171(

(%) 100/ 5.8 41/ 59 45(6OverdenturesE et al ., 1988 (174) 339/ 67 (13.6) 191/ 148 46(3T et al ., 1991 (202) 85/ 6 (1.2) atrophic mand. 1/ 5S et al ., 1993 (75) 86/ 12 (8.1) maxilla 7/ 5J, 1993 (175) 211/ 70 (7.1) maxilla 15/ 5H et al ., 1994 (210) 68/ 6 (5.9) mandible 4/ 2P et al ., 1994 (176) 89/ 27 (21.3) maxilla 19/ 8J & L, 1995 (80) 127/ 36 (9.4) maxilla 12/ 2W et al ., 1995 (211) 98¤/ 3 (2) mandible 2/ 1J et al ., 1996 (81) 510¤¤/ 44 (3.1) 117/ 393 16(9Z & S, 1996 (212) 132/ 5 (2.3) 17/ 115 3(?)/H et al ., 1997 (213) 561¤¤¤/ 13 (2) 40/ 521 11(3

Subtotal overdentures 2306/ 295 (5.9) 878/ 1428 136((%) 100/ 12.8 38/ 62 46(7

Total full edentulism 8915/679 (3.4) 3600/5313 307((%) 100/7.6 40/60 45(6

* In parenthesis the percentage of early failures.** The first figure in parenthesis refers to maxillary and the second to mandibular losses.*** Both failed fixtures were in the 1-step procedure group.**** The majority of the implants were inserted in fully edentulous patients.¤46, ¤¤177 and ¤¤ ¤22 fixtures were originally kept sleeping at initial overdenture placement. Failures occurring in these groups wer



Table 5

Clinical studies reporting on failure rates for Bra nemark implants in partially edentulous patients wearing partial fix

Implants Implant location Authors (ref. no.) inserted/ failed* maxilla/ mandible

Partial bridgesE et al ., 1986 (214) 41/ 0 (0) 29/ 12 0/ 0J et al ., 1989 (173) 448/ 13 (0.9) 238/ 210 4(4;Q et al ., 1991 (66) 509/ 29 (3.3) 304/ 205 17(1

T & L, 1992 (205) 302/ 4 (0) 106/ 196 0/ 4(J & L, 1993 (76) 259/ 7 (0.5) 101/ 158 (p)*** 4(3;N & L, 1993 (215) 1203/ 56 (2.6) 652/ 551 (p) 31(2Z & S, 1993 (216) 94/ 7 ( 4.2) 50/ 44 4(0;Z & S, 1993 (217) 105/ 2 (1.9) 41/ 64 (p) 2(0;L et al ., 1994 (27) 558/ 36 (3.6) 220/ 338 20(9O et al ., 1995 ( 79) 69/ 8 (1.4) mandible 1/ 7N & L, 1995 (218)**** 309/ 7 (2.3) 177/ 132 7(3;B et al ., 1996 (180) 50/ 0 (0) 4/ 46 (p) 0/ 0B & H, 1996 (181) 446/ 8 (0.7) 68/ 378 (p) 3(0;

Subtotal partial bridges 4393/ 177 (2.1) 1990/ 2403 93(5(%) 100/ 4.0 45/ 55 53(5

Single implantsJ et al ., 1990 (145) 23/ 1 (0) 21/ 2 0/ 1(J & P, 1993 (219) 70/ 1 (0) 59/ 11 0/ 1(

E et al ., 1994 (220) 93/ 2 ( 1.1) 84/ 9 1(0;T et al ., 1994 ( 221) 27/ 0 (0) 19/ 8 0/ 0A et al ., 1995 (222) 65/ 1 (0) 62/ 3 0/ 1(B & B, 1995 (179) 24/ 1 (0) 7/ 17 0/ 1(E et al ., 1995 (177) 82/ 2 ( 2.4) 74/ 8 2(1;H et al ., 1995 (182)***** 76/ 2 (1.3) 59/ 17 1(1;M et al ., 1996 (83) 84/ 2 ( 1.2) 75/ 9 1(0;A-A & Z, 1996 (223) 49/ 1 (2) 35/ 14 1(1;B et al ., 1996 (180) 22/ 1 (4.5) 4/ 18 1(?)/B & H, 1996 (181) 59/ 2 (0) 21/ 38 0/ 2(H et al ., 1996 (84) 107/ 3 (0.9) 88/ 19 1(1;

Subtotal single implants 781/ 19 (1) 608/ 173 8(4;(%) 100/ 2.4 78/ 22 42(5

Total partial edentulism 5174/196 (2) 2598/2576 101((%) 100/3.8 50/50 52(5

* In parenthesis the percentage of early failures** The first figure in parenthesis refers to maxillary and the second to mandibular losses.*** (p)=posterior area.**** Both total and partial edentulous ‘‘recalcitrant’’ periodontal patients are included.***** Immediate implants and guided bone regeneration techniques were also included.All late failures in single implant investigations, but one (145), occurred during the first year of loading.In studies (179–182) single implants replaced molars.



Table 6

Studies reporting on failure rates for Branemark implants used in complicated situations (GBR procedures, nerve transposition, immtuberosity implants) and bone grafts

Implants Implant location FAuthors (ref. no.) inserted/ failed maxilla/ mandible ear

Complicated situationsP & T, 1990 (224) 63/ 1 (i+a)* (1.6)** mandible 1/ 0T & K, 1991 (225) 303/ 2 (i) (0.7) 44/ 259 2(0;2)/K & B, 1991 (204) 41/ 3 (i+a) (7.3) mandible 3/ 0F et al ., 1992 (183) 23/ 3 (n) (13) mandible 3/ 0B, 1992 ( 226) 72/ 5 (0) maxillary tuber 0/ 5T, 1992 (187) 43/ 3 (2.3) maxillary tuber 1/ 2B et al ., 1994 (227) 49/ 3 (i+GBR) (6.1) 33/ 16 3(?)/ 0R, 1994 (184) 250/ 36 (n) (4.8) mandible 12/ 24D et al ., 1995 (228) 55/ 6 (GBR) (7.3) 35/ 20 4(3;1)/B et al ., 1995 ( 229) 51/ 7 (11.8) maxillary tuber 6/ 1H & B, 1995 (185) 63/ 5 (n) (7.9) mandible 5/ 0

Subtotal complicated situations 1013/ 74 (3.9) 278/ 735 40(10;2(%) 100/ 7.3 27/ 73 54(25;6

Bone graftsA et al ., 1990 (188) 140/ 39 (o) (7.7) maxilla 11/ 28I & A, 1992 (230) 46/ 8 (o) (13.4) maxilla 6/ 2T & L, 1992 (205) 109/ 9 (o; s; ng) (8.3) 92/ 17 9(7;2)/N et al ., 1993 (231) 177/ 40 (o) (9) maxilla 16/ 24I et al ., 1993 (232) 59/ 14 (ig) (24) maxilla 14/ 0R et al ., 1993 (233) 95/ 5 (s) (5.3) maxilla 5/ 0K et al ., 1994 (234) 83/ 6 (s; ng) (1.2) maxilla 1/ 5J et al ., 1994 (235) 291/ 19 (s; ng; o) (4.1) maxilla 12/ 7S et al ., 1994 (236) 55/ 2 (o) (3.6) 42/ 13 2(2;0)/J & L, 1995 (80) 83/ 12 (o) (3.6) maxilla 3/ 9K, 1995 (237) 187/ 29 (ig) (15.5) maxilla 29/ 0B et al ., 1996 (238) 182/ 37 (s) (16.5) maxilla 30/ 7K et al ., 1996 (239) 85/ 24 (o) (22.3) maxilla 19/ 5L et al ., 1996 (240) 30/ 0 (s) (0) maxilla 0/ 0A et al ., 1996 (241) 92/ 22 (o) (7.6) maxilla 7/ 15

D et al ., 1997 (242) 121/ 8 (s) (4.1) maxilla 5/ 3

Subtotal bone grafts 1833/ 274 (9.2) 1803/ 30 169(16(%) 100/ 14.9 98/ 2 62(99;

* (i)=immediate placement; (GBR)=guided bone regeneration; (n)=nerve transposition; (a)=alveolectomy; (o)=onlay bone graftsgrafts after Le Fort I osteotomy.** In parenthesis the percentage of early failures.*** The first figure in parenthesis refers to maxillary and the second to mandibular losses.



540 Esposito et al.

Table 7

Detailed analysis of failure rates for Bra nemark implants inserted in the maxillary tuberosity and in bone grafts placed in the atrophicmaxilla; data from Table 6

Surgical Implants Failures Loading timeprocedures inserted/ failed* early/ late** (range)

maxillary tuber 166/ 15 (4.2) 7/ 8 (9) 0–63 months

sinus and nasal lifts 583/ 53 (7.4) 43/ 10 (9.1) 0–80 monthsonlay grafts 741/ 153 (9.4) 70/ 83 (20.6) 1 month–10 yrinlay grafts (Le Fort I ) 246/ 43 (17.5) 43/ 0 (17.5) 1 month–4 yr

* In parenthesis the percentage of early failures.** In parenthesis the total percentage of implant losses (early+late failures).

situations included the following cases: insufficient transposition, and installation of fixtures in themaxillary tuberosity. Apart from the occurrencebone volume present for implant placement in

number and positions necessary for a fixed restora- of transient or permanent inferior alveolar nerveparesthesia/ anesthesia (183–185), the averagetion; the prognosis for a fixed implant-supported

bridge was questionable; and when the treatment failure rate of 7.3% can be regarded as an accept-able outcome in relation to the complexity and thewas an emergency solution for patients planned to

receive a fixed restoration, but who had lost one risks of these procedures.From the 2nd part of Table 6, it is evident thator more implants. If the small group of patients,

where a cluster of failures occurred, was not consid- the prevalence of failed implants in bone grafts ishigher (14.9%) than in the other situations. Aered, according to Q et al . (171), implants

used for supporting overdentures did not seem to comprehensive review on this subject has recentlybeen presented by T (186). However, infail more frequently than those supporting full

fixed bridges. order to analyze the influence of diff erent surgicalprocedures on failure rates, Table 7 was createdTable 5 presents the outcome in partially edentul-

ous patients. Data are again divided into 2 categor- using data already presented in Table 6. It can beobserved that implants placed in the maxillaryies: subjects wearing a partial bridge and single

crowns, respectively. It is evident that implants tuberosity seem to behave equally well as thoseplaced in a grafted sinus. Therefore, maxillaryplaced in partially dentate patients perform some-

what better than those placed in totally edentulous tuberosity implants might be a valid alternative tosinus lift procedures (187), local anatomical condi-patients ( 2% of early and 3.8% of total losses versus3.4% and 7.6%, respectively). In particular, both tions permitting. Both onlay and inlay grafting

procedures resulted in the highest failure rates ( 20.6the early and total failure rates for single implantswere very low (1 and 2.4%, respectively), even in and 17.5%, respectively).

It is generally agreed that implants in maxillasgroups representing the development period (145,177); in clinical centers with little experience (178); do not perform as well as in the mandibles, and

this trend can be noticed in all clinical situationsor when a single implant was used to replace amolar (179–182). The high success rate might be ( Tables 4–7) with one exception: the partially eden-

tulous situation ( Table 5), where similar failurepartly explained by a more accurate patient selec-tion in terms of age, health and anatomical condi- rates can be noticed for the mandible and the

maxilla. Fully edentulous mandibles (Table 4) aretions, together with a more favourable loaddistribution (molar areas excluded). Even though slightly over-represented and, obviously, this has

influenced the final result, lowering the total fail-the observation period was still too short to drawfinal conclusions, this treatment modality seemed ure rate.

Table 8 summarizes the data obtained by addingto be highly predictable in terms of biologicalsuccess outcome. the total failure rates of the diff erent clinical situ-

ations described in Tables 4–6. The average incid-Table 6 presents a rather heterogeneous group of reports on failures of implants used in complicated ence of early failures is approximately 3.6%. The

failure rate for late losses ( 3.6%, Table 8) mightsituations, including bone grafts. The 1st part of the table includes studies dealing with barrier mem- actually be an underestimation, as the prosthetic

reconstructions were not removed to test implantbranes to enhance bone regeneration aroundimplants in the presence of wide bony defects or mobility in all the studies presented. In addition,

only a minority of the inserted fixtures were fol-fenestrations, immediate placement of implants inextraction sockets, technically demanding nerve lowed up to 5 years or longer. In this respect,




clinical studies with prolonged follow-up periodscan provide more precise information.

Table 9 summarizes data on 3013 implants fol-lowed for up to 5 years, with adjustments for drop-outs when possible, but without distinction for thediff erent clinical situations. The results show aaverage prevalence of implant losses over a 5-year

period of 8.6% and, when bone grafts are excluded(201 implants), of 7.7%. In this sample, maxillasare slightly over-represented; therefore, this can beconsidered the worst scenario.

Unfortunately, these data do not provideinformation on failing implants. In fact, at leastfrom a theoretical point of view, a few additionalloaded fixtures, not yet failed, might be failing dueto progressive bone loss and mucosal tissue irrita-tion (peri-implantitis group). It is very difficult, atthe present time, to provide exact figures forimplant losses due to peri-implantitis. However, thetotal number of implants failed for peri-implantitisseems to be extremely low in the few existing long-term studies providing this information (76, 79, 84,165, 173, 188, 189). An attempt to calculate theprevalence of implant losses due to peri-implantitiscan be found in Table 10. The average prevalenceof implant losses attributable to peri-implantitiswas 2.8% in relation to the total number of biolo-gical failures, and 10.3% in relation to implantlosses occurring after the 1st year of loading.Therefore, after approximately 5 years of function,1 implant out of every 533 inserted seemed to belost due to recurrent infection (peri-implantitis).

However, the data presented must be interpretedwith caution, because the distinction betweendiff erent causes of implant losses could only befound in a minority of the studies analyzed.

Although comparisons with other oral implantsystems are beyond the scope of this review, othersystems such as the ITI (Table 10) seem to be moreoften aff ected by soft-tissue problems and progress-ive marginal bone loss (peri-implantitis) leading toimplant failure (32, 35, 54, 56, 88, 99, 165, 190).With the ITI system, this type of peri-implantinfection seemed to be a site-specific infectionresembling chronic adult periodontitis ( 109), which

responded favourably to antimicrobial treatment(191, 192). Reports regarding the IMZ system (161,193) also suggested a higher incidence of implantfailure due to peri-implantitis. Unfortunately, thesedata were not presented in detail to permit a properevaluation. Diff erent implant designs and surfacecharacteristics, favouring plaque retention (22,194), might, at least partly, explain these diff er-ences. Longer follow-up periods in conjunctionwith controlled, accurate radiographic studies andmonitoring of the soft tissue conditions are neededto confirm this hypothesis (4).

T a b l e 8

S u m m a r y o f f a i l u r e r a t e s f o r

B r a ˚ n e m a r k i m p l a n t s u s e d i n v a r i o u s c l i n i c a l s i t u a t i o n s

C l i n i c a l

I m p l a n t s

I m p l a n t l o c a t i o n

F a i l u r e s

L o

a d i n g t i m e

s i t u a t i o n s

i n s e r t e d / f a i l e d *

m a x i l l a / m a n d i b l e

e a r l y / l a t e * *

( r a n g e )

t o t a l e d e n t u l i s m

( T a b l e 4 )

6 6 0 9 / 3 8 4

( 2 . 6

)

2 7 2 2 / 3 8 8 7

1 7

1 ( 1 0 7 ; 6 4 ) / 2 1 3 ( 1 6 3 ; 5 0 )

1 m o n t h – 1 5 y r

o v e r d e n t u r e s ( T a b l e 4 )

2 3 0 6 / 2 9 5

( 5 . 9

)

8 7 8 / 1 4 3 8

1 3

6 ( 1 0 3 ; 3 0 ) / 1 5 9 ( 1 3 8 ; 1 9 )

3 m o n t h s – 1 3 y r

p a r t i a l e d e n t u l i s m

( T a b l e 5 )

3 9 3 / 1 7 7

( 2 . 1

)

1 9 9 0 / 2 4 0 3

9 3

( 5 1 ; 4 2 ) / 8 4 ( 3 8 ; 4 6 )

1 m o n t h – 9 y r

s i n g l e i m p l a n t s ( T a b l e 5 )

7 8 1 / 1 9

( 1 . 0

)

6 0 8 / 1 7 3

8 (

4 ; 3 ) / 1 1 ( 8 ; 3 )

3 m o n t h s – 8 y r

c o m p l i c a t e d s i t u a t i o n s ( T a b l e 6 )

1 0 1 3 / 7 4

( 3 . 9

)

2 7 8 / 7 3 5

4 0

( 1 0 ; 2 7 ) / 3 4 ( 1 0 ; 2 4 )

0 – 6

y r

b o n e g r a f t s ( T a b l e 6 )

1 8 3 3 / 2 7 4

( 9 . 2

)

1 8 0 3 / 3 0

1 6

9 ( 1 6 7 ; 2 ) / 1 0 5 ( 1 0 5 ; 0 )

0 – 1 0 y r

t o t a l

1 6 9 3 5 / 1 2 2 3

( 3 . 6

)

8 2 7 9 / 8 6 5 6

6 1

7 ( 4 4 2 ; 1 6 8 ) / 6 0 6 ( 4 6 2 ; 1 4 2 )

1 0 0 / 7 . 2

4 9 / 5 1

5 0

( 7 2 ; 2 7 ) / 5 0 ( 7 6 ; 2 3 )

* I n p a r e n t h e s i s t h e p e r c e n t a g e o f e a r l y f a i l u r e s .

* * T h e fi r s t fi g u r e r e f e r s t o m a x i l l a r y

a n d t h e s e c o n d t o m a n d i b u l a r l o s s e s .



542 Esposito et al.

Table 9

Summary of studies reporting on failure incidence for Bra nemark implants used in various clinical situations followed up to 5 yearsafter loading

Detection time of failureImplants

Authors (ref. no.) inserted/ failed before loading 1 yr 2 yr 3 yr 4 yr 5 yr

A et al ., 1990 (188) 45/ 12 ( BG )* 4 4 2 0 1 1Z & S, 1990 ( 243 ) 262/ 29 ( TE ) 21 1 5 1 1 0T & K, 1991 ( 225 ) 24/ 0 (II ) 0 0 0 0 0 0J & L, 1993 (76) 259/ 7 (PE ) 4 1 0 0 0 2K et al ., 1994 (234) 20/ 2 (BG ) 0 0 0 2 0 0L et al ., 1994 (27) 558/ 36 (PE ) 20 3 8 2 0 3P et al ., 1994 (176) 8/ 0 (OV ) 0 0 0 0 0 0J, 1994 (77) 449/ 31 ( TE ) 15 7 6 2 0 1J & L, 1995 (80) 801/ 111 ( TE; BG; OV ) 38 40 19 9 3 2O et al ., 1995 ( 79) 69/ 8 (PE ) 1 4 3 0 0 0H et al ., (207) 83/ 0 ( TE ) 0 0 0 0 0 0H et al ., 1996 (84) 107/ 3 (SI ) 1 2 0 0 0 0K et al ., 1996 (239) 52/ 18 (BG ) 15 0 2 0 0 1L et al ., 1996 ( 87) 272/ 2 ( TE ) 2 0 0 0 0 0

total 3013/ 259 121 62 45 16 5 10(%) 100/ 8.6 47 (early losses) 53 ( late losses)

Subtotal (bone grafts excluded) 2812/ 216 99 52 41 13 3 8(%) 100/ 7.7 46 (early losses) 54 ( late losses)

Data for maxillas only 1528/ 155 63 46 29 11 2 4(bone grafts excluded)(%) 100/ 10.1 40 (early losses) 60 ( late losses)

* PE=partial edentulism; BG=bone grafts; TE=total edentulism; II=immediate implants; OV=overdentures; SI=single implants.

An analysis of the possible causes and their bone quality and volume ( 90%) and to chronicmarginal infections due to bacterial plaque (10%;relative importance for implant failure is desirable

for several reasons. Such information will draw data from Table 10). Clearly, both these etiologicfactors may coexist in the same case.attention to the areas where diagnostic, preventiveand therapeutic tools are needed. Unfortunately, Another factor which might have influenced the

absolute number of late failures is the problem of little information is found in the literature onpossible mechanisms for failures. The fact that drop-outs, which is unavoidable in long-term

follow-up studies. However, complications in thethere is not yet agreement on the etiologic role of overload and bacterial plaque (peri-implantitis), low compliance group (once excluded, deceased

and change of address) seem not likely to haveparticularly for late failures, may be one reason.In Table 11, we attempted to divide failures of jeopardized bridge function to a level requiring

dentist intervention. Considering that implant treat-Branemark implants according to chronologicaldistribution and their possible etiologic factors. Out ment has mainly been provided by specialists in

most of the investigations, one may assume thatof the total number of failures, early fixture lossesaccounted for 47% (data from Table 9) which might the patient would return to the center responsible

for the treatment in case of subjective problems.have been caused by iatrogenic (e.g., infections orbone overheating) and host-related factors, includ- Finally, the reader should be aware that the

positive results listed in this section are valid onlying anatomical factors (e.g., poor bone quality orextremely thin bucco-lingual bone crests). for experienced specialists and may be influenced

by the strict protocol of clinical investigations. TheAccordingly, late failures amounted to 53%, of which about 1/ 2 were detected during the 1st year importance of this aspect will be further discussed

in the 2nd part of the review.of loading. It seems reasonable to attribute mostof these late failures to overload in relation to poor Another interesting observation was the consist-

ent presence of several failed fixtures in the samebone quality and insufficient bone volume. Theremaining late failures (after the 1st year of loading) patient. In other words, few patients accounted for

most of the failures. This has been described in themight be attributed to progressive or ‘‘dramatic’’changes of the loading conditions in relation to literature as the ‘‘cluster phenomenon’’ or the ‘‘clus-



Table 10

Summary of studies reporting on implant loss prevalence attributable to peri-implantitis regarding Bra nemark and ITI syst

Implants Failures RemoAuthors (ref. no.) inserted/ failed* early/ late** peri-imp

Branemark SystemJ et al ., 1989 (173) 448/ 13 (PE ) 4/ 9 (4) 1 (2) A et al ., 1990 (188) 140/ 39 (BG) 11/ 28 (18) 0 Q et al ., 1992 (26) 509/ 29 (PE ) 23/ 6 (2) 0 J & L, 1993 (76) 259/ 7 (PE ) 4/ 3 (2) 1 (5) M-S & Z, 1993 (165) 68/ 5 (OV mand.) 4/ 1 (1) 1 (?) O et al ., 1995 (79)**** 69/ 8 (PE ) 1/ 7 (3) 0 H et al ., 1996 (84) 107/ 3 (SI ) 1/ 2 ( 0) 0

total 1600/ 104 48/ 56 (29) 3 (%) 100/ 6.5 46/ 54 2.8/ 10.3*

ITI SystemD’ et al ., 1989 (32) 95/ 15 (OV; PE; TE) 9/ 6 (6) 6 (2–4)B et al ., 1991 (88) 54/ 2 (PE, SI ) 0/ 2 ( 2) 2 (2–3)W et al ., 1992 (190) 461/ 39 (OV; PE; SI ) 30/ 9 ( 9?) 9 (?) S et al ., 1993 (35) 201/ 14 (OV, TE, PE ) 2/ 12 (8) 4 (2–4) M-S & Z, 1993 (165) 74/ 5 (OV mand.) 3/ 2 (2) 1 (3) L-V et al ., 1995 (56) 153/ 13 (OV mand.) 4/ 9 ( 9) 8 (4–8)V et al ., 1995 (54) 153/ 33 (OV mand.) 4/ 29 (29?) 29 (?) A et al ., 1996 (99) 216/ 8 (OV & TE mand.) 4/ 4 (4) 4 (2)

total 1407/ 129 56/ 73 (79) 63 (%) 100/ 9.7 43/ 57 48.8/ 91.3

* PE=partial edentulism; BG=bone grafts; TE=total edentulism treated with fixed bridges; OV=overdentures; SI=single implan** In parenthesis the number of failures after the first year of loading.*** In parenthesis the year of removal.

**** Data presented by A et al. (189)***** The first figure represents the % of implant losses due to peri-implantitis in relation to the total number of failed implants, w(suspected peri-implantitis) for implants failed after the first year of loading.The ITI implants were of diff erent designs: type C, E, F, K, H, TPS, Bonefit, hollow cylinders and hollow screws.



544 Esposito et al.

Table 11

Distribution and percentage of failed Bra nemark implants in relation to possible etiologic factors over a 5-year period; data fromTables 9 and 10

Early failures (47%) Late failures (53%)

before 1 yr of loading (45%) after 1 yr of loading (55%)iatrogenic/ host related factors overload/ host related factors overload ( 90%)/ peri-implantitis ( 10%)

tering eff ect’’ (149, 195, 196), and might indicate A fair comparison between diff erent implantsystems will only be possible when a general con-that also some genetic or systemic factors can play

a role in implant failure. sensus on the parameters for monitoring implantconditions will be obtained. So far, there is noIn conclusion, biological implant failures with

the Branemark system are relatively rare: 7.7% over agreement on the parameters to be employed. Datacollected from diff erent trials should be amenablea 5-year period ( Table 9, bone grafts excluded).

The predictability of implant treatment is remark- to comparison in order to permit pooling andanalysis. Therefore, standardization of clinical trialable, particularly in partially edentulous situations.

In fact, implants in partially edentulous patients designs dealing with oral implants is an urgentmust. In particular, success criteria must be well(Table 5) perform better than in fully edentulous

patients ( Table 4) in terms of early and total failure described, universally accepted, and should berelated to the goal of the treatment. Ideally, inrates (2% and 3.8% versus 3.4% and 7.6%, respect-

ively). Our analysis also confirms the general trend order to avoid possible bias, clinical performancesof diff erent oral implants should be evaluated byof maxillas to have almost 3× more implant losses

than mandibles (Table 8), both in relation to early independent bodies. It is definitively recommendedto concentrate more emphasis on the aspects of and late failures, with the exception of the partially

edentulous situation, in which practically no diff er- failure providing adequate figures, so that studiescould be more easily compared and precious clinicalence could be observed. The low prevalence of peri-

implantitis found in the literature together with the guidelines retrieved. Moreover, only a limitednumber of clinical studies have been specificallyfact that, in general, partially edentulous patients

have less resorbed jaws, speak in favour of lack of designed to identify the prognostic factors influen-cing implant failure. As is evident from the resultssufficient bone volume, deficient bone quality, and

overload as the 3 major determinants for late presented in the reviewed literature, there is also

an apparent need for an analysis of the clinicalimplant failures for Branemark implants. Thesefindings are in agreement with a recent immuno- course with new/ improved (non-invasive) tech-niques. Further, in order to understand the mechan-histochemical investigation on late implant failures

(30). However, the relationships between these isms of failure and their clinical correlates, failedimplants and their surrounding tissues have to benegative prognostic factors and infections remain

to be established in prospective clinical studies. collected and analyzed.

Acknowledgements – This study was supported by grants fromConclusions the Swedish Medical Research Council (grant 9495), the

National Research Council of Italy (CNR), the Center forWhile surgical trauma together with bone volume

Biomaterial Research, the Faculty of Medicine and the Facultyand quality are generally believed to be the most of Odontology, Goteborg University, Sweden. The secretarial

help of Mrs Silvana Gandini Esposito is gratefully acknow-important etiological factors for early implant fail-ledged. No financial support for this study has been receivedures, the etiology of late failures is more controver-

from any oral implant company.sial. Loading conditions in relation to jaw volumeand bone quality together with chronic infection(peri-implantitis) are considered to play the pre- Referencesdominant role (197). This review of the literature

1. W P. Current implant usage. J Dent Educusing a metanalytic approach has shown that late

1988; 52: 692–695.failures caused by peri-implant infection are a rare 2. V S D. Implant systems. In: N I, Voccurrence, at least in relation to the Branemark S D, W P, eds. Osseointegration

in oral rehabilitation. London: Quintessence, 1993; 77–86.system. Further, our analysis indicates that there3. E SE, P A, M HL, P JL.might be major diff erences among oral implant

Validation of dental implant systems through a review of systems in the modality of failure which is in

literature supplied by system manufacturers. J Prosthetagreement with other preliminary recent findings Dent 1997; 77: 271–279.

4. E M, H J-M, L U, T P.(4).




Failure patterns of four osseointegrated oral implant rated endosseous implants. J Prosthet Dent 1989; 62:

567–572.systems. J Mat Sci: Mater Med 1998; in press.

5. B P-I, B U, A R, H BO, 23. NIH consensus conference statement: dental implants. Int

J Oral Maxillofac Implants 1988; 3: 290–293.L J, O A. Intra-osseous anchorage of

dental prostheses. Scand J Plast Reconstr Surg 1969; 24. L B, F ME, M RM, B A,

F S, G J, H JE, I H,3: 81–100.

6. B P-I, L J, E L, T P, L RJ, L SB, M RF, M DH,

S MS, V S D. Consensus report.B R, S R. Anatomy of osseointegration

and the transfer of load. In: F RJ, D WH, eds. Discussion section VIII. In: N M, B W,K K, eds. Proceedings of the World Workshop onReconstructive preprosthetic oral and maxillofacial surgery.

Philadelphia: W.B. Saunders Co., 1995; 165–224. Clinical Periodontics. Chicago: The American Academy

of Periodontology, 1989; (VIII)11–18.7. A T, B P-I, H H-A,

L J. Osseointegrated titanium implants. 25. WP,BCL,T TD. The Swedish

system of osseointegrated implants: Problems and com-Requirements for ensuring a long-lasting, direct bone-to-

implant anchorage in man. Acta Orthop Scand 1981; plications encountered during a 4-year trial period. Int

J Oral Maxillofac Implants 1987; 2: 77–84.52: 155–170.

8. B P-I. Introduction to osseointegration. In: 26. Q M, N I, V S D, D

C, C A. Periodontal aspects of osseointegratedBranemark P-I, Zarb GA, Albrektsson T, eds. Tissue-

integrated prostheses. Chicago: Quintessence Publishing fixtures supporting a partial bridge. An up to 6-year

retrospective study. J Clin Periodontol 1992; 19: 118–126.Co., Inc., 1985; 11–76.

9. S SG, E J, M PA, 27. L U, V S D, H I,

B C, F T, G J, H P, H K,SA. Adhesion of bone to titanium. In: C

P,M A, L AJC, eds. Biological and biomechanical L WR, L U. Osseointegrated implants in the

treatment of partially edentulous jaws: A prospective performance of biomaterials. Amsterdam: Elsevier, 1989;409–414. 5-year multicenter study. Int J Oral Maxillofac Implants

1994; 9: 627–635.10. Z GA, A T. Osseointegration: a requiem

for the periodontal ligament? Int J Periodont Rest Dent 28. S DY, S RL, C TA, K PHJ.

The reverse-torque test: A clinical report. Int J Oral 1991; 11: 88–91.

11. N JR, A JE, M MA, G Maxillofac Implants 1996; 11: 179–185.

29. G K, L U. The predictive value of radio-GW. Tissue response to dental implants protruding

through mucous membrane. Oral Sci Rev 1974; 5: 85–105. graphic diagnosis of implant instability. Int J Oral

Maxillofac Implants 1997; 12: 59–64.12. B R. A biomechanical study of osseointegration.

In-vivo measurements in rat, rabbit, dog and man. 30. E M, T P, M J, G C, E

LE, L U. Immunohistochemistry of soft tissuesGoteborg: Goteborg University, 1996.

13. A T, Z GA. Current interpretations of the surrounding late failures of Branemark implants. Clin

Oral Implants Res 1997; 8: 352–366.osseointegrated response: clinical significance. Int

J Prosthodont 1993; 6: 95–105. 31. S LB, R GS, S ED, K RL.

Evaluation in reconstructive implantology. Dent Clin14. A T, I F. Consensus report of session

IV. In: L NP, K T, eds. Proceedings of the 1st North Am 1986; 30: 327–349.32. D’H B, S W . A comparative study of resultsEuropean Workshop on Periodontology. London:

Quintessence Publishing Co., Ltd., 1994; 365–369. with various endosseous implant systems. Int J Oral

Maxillofac Implants 1989; 4: 95–105.15. M A . Criteria for success. Monitoring. In: L

NP, K T, eds. Proceedings of the 1st European 33. O J, A C. The periotest method as a measure

of osseointegrated oral implant stability. Int J Oral Workshop on Periodontology. London: Quintessence

Publishing Co., Ltd., 1994; 317–325. Maxillofac Implants 1990; 5: 390–400.

34. C H, O LF, D RL, M J.16. A T, Z G,W P, E AR.

The long-term efficacy of currently used dental implants. Assessment of oral implant mobility. J Prosthet Dent

1993; 70: 421–426.A review and proposed criteria of success. Int J Oral

Maxillofac Implants 1986; 1: 11–25. 35. S MAM, O K, V K, P H.

Failures in the osseointegration of endosseous implants.17. M A, L NP. Clinical parameters for the

evaluation of dental implants. Periodontol 2000 1994; Int J Oral Maxillofac Implants 1993; 8: 92–97.

36. N IE, R D, V S D, T4: 81–86.

18. M N, A D, C P. Quantitative deter- JA, N M . The influence of splinting procedures on the

periodontal and peri-implant tissue damping character-mination of the stability of the implant-tissue interfaceusing resonance frequency analysis. Clin Oral Implants istics. A longitudinal study with the PeriotestA device.

J Clin Periodontol 1995; 22: 703–708.Res 1996; 7: 261–267.

19. E JJ, B JB, S HA. A dynamic modal 37. M-S R, M D, M E, O A.

Periotest measurements and osseointegration of mandib-testing technique for noninvasive assessment of bone-

dental implant interfaces. Int J Oral Maxillofac Implants ular ITI implants supporting overdentures. A one-year

longitudinal study. Clin Oral Implants Res 1995; 6: 73–82.1996; 11: 728–734.

20. B B. Evaluation of the results of treatment with 38. T J, L P, V S D, Q

M, N I. Mechanical state assessment of the implant-osseointegrated implants by the Swedish National Board

of Health and Welfare. J Prosthet Dent 1983; 50: 114–115. bone continuum: A better understanding of the periotest

method. Int J Oral Maxillofac Implants 1995; 10: 43–49.21. S PA, S LB. Recommendations of the

consensus development conference on dental implants. 39. B U, H-P C, BW, B D, L

NP. Correlations between radiographic, clinical andJ Am Dent Assoc 1979; 98: 373–377.

22. S DE, Z GA. Criteria for success of osseointeg- mobility parameters after loading of oral implants with



546 Esposito et al.

fixed partial dentures. A 2-year longitudinal study. Clin 57. S C, M-H L. Microdensitometric

Oral Implants Res 1996; 7: 230–239. and visual evaluation of the resolution of dental films.40. B D, W HP, L NP. Tissue integration of Oral Surgery 1978; 45: 811–822.

non-submerged implants. 1-year results of a prospective 58. H BO. Success and failure of osseointegratedstudy with 100 ITI hollow-cylinder and hollow-screw implants in the edentulous jaw. Swed Dent J 1977 (Suppl.implants. Clin Oral Implants Res 1990; 1: 33–40. 1): 1–101.

41. T J, Q M, D P, V S 59. V S D. A retrospective multicenter evalu-D. Periotest: An objective clinical diagnosis of bone ation of the survival rate of osseointegrated fixtures sup-

apposition toward implants. Int J Oral Maxillofac porting fixed partial prostheses in the treatment of partialImplants 1991; 6: 55–61. edentulism. J Prosthet Dent 1989; 61: 217–223.42. V SD, T J,N I,NM. Damping 60. A R, L U, R B, B P-I. A

characteristics of bone-to-implant interfaces. Clin Oral 15-year study of osseointegrated implants in the treatmentImplants Res 1995; 6: 31–39. of the edentulous jaw. Int J Oral Surg 1981; 10: 387–416.

43. A C. The use of the periotest value as the initial 61. R F. Mach bands: Quantitative studies on neural success criteria of an implant: 8-year report. Int J Periodont networks in the retina. San Francisco: Holden-Day, Inc.,Rest Dent 1997; 17: 151–161. 1965.

44. A R, L U, B P-I. Surgical 62. S S, G K, G H-G. Accuracy andprocedures. In: B P-I, Z GA, A precision in the radiographic diagnosis of clinical instabil-T, eds. Tissue-integrated prostheses. Chicago: Quintessence ity in Branemark dental implants. Clin Oral Implants ResPublishing Co., Inc., 1985; 211–232. 1995; 6: 220–226.

45. A T, S L. Direct bone anchorage of 63. H L, R B. Radiographic evaluation of oral implants: Clinical and experimental consideration of osseointegrated implants of the jaws. Dentomaxillofacthe concept of osseointegration. Int J Prosthodont 1990; Radiol 1980; 9: 91–95.3: 30–41.

64. S IP. Errors in radiographic assessment of marginal46. F B, J T, L U. Early failures in 4641 bone height around osseointegrated implants. Scand consecutively placed Branemark dental Implants: a study J Dent Res 1990; 98: 428–433.from stage 1 surgery to the connection of completed 65. B DK. Estimating the validity of radiographic meas-prostheses. Int J Oral Maxillofac Implants 1991; 6:

urements of marginal bone height changes around osseoin-142–146.

tegrated implants. Implant Dent 1992; 1: 79–83.47. S K-G. Radiographic results. In: B P-I,

66. Q M, N I, V S D, SZ GA, A T, eds. Tissue-integrated pros-

E, C L, T G, G J, D Mtheses. Chicago: Quintessence Publishing Co., Inc., 1985;

G. The cumulative failure rate of the Branemark system187–198.

in the overdenture, the fixed partial, and the fixed full48. S K-G. Radiographic procedures. In: B

prostheses design: a prospective study on 1273 fixtures.P-I, Z GA, A T, eds. Tissue-integrated

J Head Neck Pathol 1991; 10: 43–53. prostheses. Chicago: Quintessence Publishing Co., Inc.,

67. S IP. Estimation of angulation of Branemark tita-1985; 317–327.

nium fixtures from radiographic thread images. Clin Oral 49. H L. Radiographic examination of endosseous

Implants Res 1991; 2: 20–23.implants in the jaws. In: W P, B P-

68. Q M. Tissue response to loading and microbiota.I, eds. Advanced osseointegration surgery. Applications inIn: N I , V S D, W P, eds.

the maxillofacial region. Chicago: Quintessence PublishingOsseointegration in oral rehabilitation. London:Co., Inc., 1992; 80–93.Quintessence Publishing Co., Ltd., 1993; 171–180.50. B U. Radiographic parameters for the evaluation

69. A R, L U, R B, B P-I,of peri-implant tissues. Periodontol 2000 1994; 4: 87–97.L J, E B, S L. Marginal tissue51. F NL. Diagnostic imaging in dental implantol-reactions at osseointegrated titanium fixtures. A 3-yearogy. Oral Surg Oral Med Oral Pathol Oral Radiol Endod longitudinal prospective study. Int J Oral Maxillofac1995; 80: 540–554.Implants 1986; 15: 39–52.52. G K, E A, G H-G.

70. C JF, Z GA. The longitudinal clinical efficacy of Radiography in oral endosseous prosthetics. Goteborg:osseointegrated dental implants: a 3-year report. Int J Oral Nobel Biocare AB, 1996.Maxillofac Implants 1987; 2: 91–100.53. M JP, S L. Clinical report on the success

71. C DV, Z GA, S A, L DW. Theof 47 consecutively placed Core-Vent implants followedlongitudinal eff ectiveness of osseointegrated dentalfrom 3 months to 4 years. Int J Oral Maxillofac Implantsimplants. The Toronto study: Bone level changes. Int1990; 5: 53–60.

J Periodont Rest Dent 1991; 11: 113–125.54. V PAM, V B G-J, S AP,72. F B,G K, LU. A new self-tappingO J-P. Clinical evaluation of mandibular over-

Branemark implant: Clinical and radiographic evaluation.dentures supported by multiple-bar fabrication: a follow-Int J Oral Maxillofac Implants 1992; 7: 85–85.up study of two implant systems. Int J Oral Maxillofac

73. Q M, N I, V S D. FixtureImplants 1995; 10: 595–603.design and overload influence marginal bone loss and55. S H, J VK, R E-J. A 10-yearfixture success in the Branemark system. Clin Oral follow-up study of IMZ and TPS implants in the edentul-Implants Res 1992; 3: 104–111.ous mandible using bar-retained overdentures. Int J Oral

74. E M, E A, G K. RadiologicalMaxillofac Implants 1995; 10: 231–243.evaluation of marginal bone loss at tooth surfaces facing56. L-V R, P J, O E, Hsingle Branemark implants. Clin Oral Implants Res 1993;H, H R-P. ITI titanium plasma-sprayed screw4: 151–157.implants in the treatment of edentulous mandibles: A

75. S J-I, L E, B I, F A,follow-up study of 39 patients. Int J Oral Maxillofac

Implants 1995; 10: 373–378. N K. A clinical and radiological two-year follow-up




study of maxillary overdentures on osseointegrated 92. QM, N I, V S D, D

implants. Clin Oral Implants Res 1993; 4: 39–46. L, D P. Periodontal aspects of Branemark and IMZ76. J T, L U. Oral implant treatment in posterior implants supporting overdentures: a comparative study.

partially edentulous jaws: A 5-year follow-up report. Int In: L WR, T DE, eds. Tissue integration inJ Oral Maxillofac Implants 1993; 8: 635–640. oral, orthopedic and maxillofacial reconstruction.

77. J T. Fixed implant-supported prostheses in the eden- Proceedings of the 2nd International Congress on Tissuetulous maxilla. A 5-year follow-up report. Clin Oral Integration in the Oral, Orthopedic, and Maxillofacial Implants Res 1994; 5: 142–147. Reconstruction. Chicago: Quintessence Publishing Co.,

78. E I, R K, G P-O, L J, N Inc., 1992; 80–93.K. Clinical and radiographical features of submerged and 93. G K, H B, S I, H F,nonsubmerged titanium implants. Clin Oral Implants Res H-H E, P CS, C K.1994; 5: 185–189. Marginal tissue response adjacent to Astra dental implants

79. O M, G J, A P, B K. Bridges sup- supporting overdentures in the mandible. A 2-year follow-ported by free-standing implants versus bridges supported up study. Clin Oral Implants Res 1993; 4: 83–89.by tooth and implant. A 5-year prospective study. Clin 94. C G, C S, C E. Single-toothOral Implants Res 1995; 6: 114–121. implant rehabilitation: a retrospective study of 67

80. J T, L U. Implant treatment in edentulous implants. Int J Prosthodont 1994; 7: 525–531.maxillae: A 5-year follow-up report on patients with 95. F B, E S, A K. Mandibular singlediff erent degrees of jaw resorption. Int J Oral Maxillofac crystal sapphire implants: changes in crestal bone levelsImplants 1995; 10: 303–311. over three years. Clin Oral Implants Res 1995; 6: 181–188.

81. J T, C J, H J, H MR, H JE, 96. J Y-C, H C-H, L K-W. A 1-year radiographicJ RB, MK S, MN DC, V evaluation of marginal bone around dental implants. IntS D, T R, W RM, H I. J Oral Maxillofac Implants 1996; 11: 811–818.A 5-year prospective multicenter follow-up report on

97. F B, T T, S J, A K.overdentures supported by osseointegrated implants. Int Rehabilitation of mandibular edentulism by single crystalJ Oral Maxillofac Implants 1996; 11: 291–298. sapphire implants and overdentures: 3–12 year results in

82. P S, S K, S B, S B. 86 patients. A dual center international study. Clin Oral Marginal bone levels around maxillary implants sup-

Implants Res 1996; 7: 220–229.porting overdentures or fixed prostheses: A comparative

98. C MS, V R IP, D P C, W RPJ. Astudy using detailed narrow-beam radiographs. Int J Oral

clinical retrospective evaluation of FA/ HA coatedMaxillofac Implants 1996; 11: 223–227.

( Biocomp) dental implants. Results after 1 year. Clin Oral 83. M C, H M , D P. Marginal bone

Implants Res 1996; 7: 345–353.levels at Branemark system implants used for single tooth

99. A P, A I, B G, H S-E, Hrestoration. The influence of implant design and anatom-

S,K U. Non-submerged implants in the treatmentical region. Clin Oral Implants Res 1996; 7: 162–169.

of the edentulous lower jaw. A 2-year longitudinal study.84. H PJ, L WR, J T, H D, K PHJ,

Clin Oral Implants Res 1996; 7: 337–344.P G, Z GA, H I. Osseointegrated

100. F B, A K. Long-term evaluation of singleimplants for single-tooth replacement: A prospective

crystal sapphire implants as abutments in fixed prostho-5-year multicenter study. Int J Oral Maxillofac Implants

dontics. Clin Oral Implants Res 1997; 8: 58–67.1996; 11: 450–455.101. D L D, G AK, P GE, A S.

85. L U, S L, R J, B W. Soft tissueOsseointegration of rough acid-etched implants: one-yearand marginal bone conditions at osseointegrated implantsfollow-up of placement of 100 minimatic implants. Intthat have exposed threads. A 5-year retrospective study.J Oral Maxillofac Implants 1997; 12: 65–73.Int J Oral Maxillofac Implants 1996; 11: 599–604.

102. TER, SY,AY,K T. Correlation86. J T, B K. Prosthesis misfit and marginal bone lossbetween mucosal inflammation and marginal bone lossin edentulous implant patients. Int J Oral Maxillofacaround hydroxyapatite-coated implants: a 3-year cross-Implants 1996; 11: 620–625.sectional study. Int J Oral Maxillofac Implants 1997;87. L LW, C GE, J T. A prospective12: 74–81.15-year follow-up study of mandibular fixed prostheses

103. H CHF, B U, B W, L NP. Thesupported by osseointegrated implants. Clinical resultseff ect of subcrestal placement of the polished surface of and marginal bone loss. Clin Oral Implants Res 1996;ITI implants on marginal soft and hard tissues. Clin Oral 7: 329–336.Implants Res 1996; 7: 111–119.88. B D, W HP, B U, B C. Tissue

104. S IP. Comparison of radiographic image character-integration of one-stage ITI implants: 3-year results of a

istics of Branemark and IMZ implants. Clin Oral Implantslongitudinal study with hollow-cylinder and hollow-screwRes 1991; 2: 151–156.implants. Int J Oral Maxillofac Implants 1991; 6: 405–412.

105. B U, B W, L NP, B D. Digital89. W HP, B D, F JP, W RC.subtraction radiography for the assessment of changes inRadiographic evaluation of crestal bone levels adjacent toperi-implant bone density. Int J Oral Maxillofac Implantsnonsubmerged titanium implants. Clin Oral Implants Res1991; 6: 160–166.1992; 3: 181–188.

106. B U, BW, F I, L NP. Image90. A K, B H, F A, V K L,processing for the evaluation of dental implants.L E. A 3-year clinical study of Astra dentalDentomaxillofac Radiol 1992; 21: 208–212.implants in the treatment of edentulous mandibles. Int

107. F I, B U, BW, TM, LJ Oral Maxillofac Implants 1992; 7: 321–329.NP. Digital image processing (II). In vitro quantitative91. D B H, C B, L U, F L. Aevaluation of soft and hard peri-implant tissue changes.comparative study of the clinical efficacy of Screw VentClin Oral Implants Res 1994; 5: 105–114.implants versus Branemark fixtures, installed in a perio-

dontal clinic. Clin Oral Implants Res 1992; 3: 32–41. 108. R TE, R TW, T HJ, K PH. The




porting fixed bridges. A microbiological and histological 164. B MS, G D, K JN, M DJ, F

IM, G L. Hydroxyapatite-coated cylindricalstudy. J Clin Periodontol 1986; 13: 560–562.145. J T, L U, G K. A 3-year followup implants in the posterior mandible: 10-year observations.

Int J Oral Maxillofac Implants 1996; 11: 626–633.study of early single implant restorations ad modum

Branemark. Int J Periodont Rest Dent 1990; 10: 341 –349. 165. M-S R, ZGA. Overdentures: An alternat-ive implant methodology for edentulous patients. Int146. S M, N MG, N S, H R, S

R, F T. Characterization of the subgingival micro- J Prosthodont 1993; 6: 203–208.166. W JL, B F, L U. The influencebial flora around endosteal sapphire dental implants in

partially edentulous patients. Int J Oral Maxillofac of the masticatory mucosa on the peri-implant soft tissuecondition. Clin Oral Implants Res 1994; 5: 1–8.Implants 1990; 5: 247–253.147. Q M, L MA. The distribution of bac- 167. W K, B D, L NP, K T. Plaque-

induced peri-implantitis in the presence or absence of terial morphotypes around natural teeth and titaniumimplants ad modum Branemark. Clin Oral Implants Res keratinized mucosa. Clin Oral Implants Res 1995; 6:

131–138.1990; 1: 8–12.

148. B W, B B, N MG, N S . Clinical 168. I F. Loss of osseointegration caused by occlusal loadof oral implants. A clinical and radiographic study inand microbiologic findings that may contribute to dental

implant failure. Int J Oral Maxillofac Implants 1990; monkeys. Clin Oral Implants Res 1996; 7: 143–152.169. I F. Histological evaluation of peri-implant bone at5: 31–38.

149. R ES, T JP, S J. Microbial diff er- implants subjected to occlusal overload or plaque accumu-

lation. Clin Oral Implants Res 1997; 8: 1–9.ences in 2 clinically distinct types of failures of osseointeg-rated implants. Clin Oral Implants Res 1991; 2: 135 –144. 170. S M, A J, L P, C JL, Q M,

V S D. Histo-pathologic characteristics of 150. B GR, M M, R JW, H WW.Plaque-induced inflammation around implants. Int J Oral peri-implant soft tissues in Branemark implants with 2

distinct clinical and radiological patterns. A histometricMaxillofac Implants 1992; 7: 330–337.

151. G RR, C CM, K WJ. Microbial colon- and ultrastructural study. Clin Oral Implants Res 1991;2: 128–134.ization on natural tooth structure compared with smooth

and plasma-sprayed dental implant surfaces. Clin Oral 171. Q M, N I , V S D, T

J, D C, T G. Periodontal aspects of Implants Res 1993; 4: 53–64.152. L A , A B, L U, W M, osseointegrated fixtures supporting an overdenture. A

4-year retrospective study. J Clin Periodontol 1991; 18:DG. A longitudinal microbiological study on osseo-integrated titanium implants in partially edentulous 719–728.

172. B P-I, H BO, A R, B U,patients. Clin Oral Implants Res 1993; 4: 113–120.153. S LH, K D, G JJ, S L J, H O, O A. Osseointegrated

implants in the treatment of the edentulous jaw. ExperienceGS, S DE, M ME. The microbiota of the

peri-implant region in health and disease. Implant Dent from a 10-year period. Stockholm: Almqvist & WiksellInternational, 1977.1994; 3: 170–174.

154. G K, Z G-GK, M Y, H 173. J T, L U, A R. Osseointegrated implantsin the treatment of partially edentulous patients: A prelim-S, N J. Clinical and microbiological status of

osseointegrated implants. J Periodontol 1994; 65: 766–770. inary study on 876 consecutively placed fixtures. Int J Oral

Maxillofac Implants 1989; 4: 211–217.155. PW,QM,NM, V SD. The eff ect of periodontal parameters on the subgingival 174. E B, B T, K T, L U. A

retrospective multicenter evaluation of osseointegratedmicrobiota around implants. Clin Oral Implants Res 1995;6: 197–204. implants supporting overdentures. Int J Oral Maxillofac

Implants 1988; 3: 129–134.156. M A, M M, G T, G U,

L NP. The microbiota of osseointegrated implants in 175. J T. Implant treatment in resorbed edentulous upper jaws. A three-year follow-up study on 70 patients. Clinpatients with a history of periodontal disease. J Clin

Periodontol 1995; 22: 124–130. Oral Implants Res 1993; 4: 187–194.176. P S, S K, S B. Implant-supported157. PW, QM, V SD. The

influence of periodontitis on the subgingival flora around maxillary overdentures: Outcome in planned and emer-

gency cases. Int J Oral Maxillofac Implants 1994; 9:implants in partially edentulous patients. Clin Oral Implants Res 1996; 7: 405–409. 184–190.

177. E B, N H, A P. Single-tooth replace-158. M R, S M, H C, F-D-J L.Osseointegrated implants in patients treated for general- ment by osseointegrated Branemark implants. A retro-

spective study of 82 implants. Clin Oral Implants Resized severe adult periodontitis. An interim report.

J Periodontol 1996; 67: 782–787. 1995; 6: 238–245.178. L WR, J T, H D, H PJ, K PHJ,159. A M, C G. Microbial findings of deep

peri-implant bone defects. Int J Oral Maxillofac Implants P G, Z GA, H I. Osseointegratedimplants for single-tooth replacement: Progress report1997; 12: 106–112.

160. K G, S W, D J. Should the exit of from a multicenter prospective study after 3 years. Int

J Oral Maxillofac Implants 1994; 9: 49–54.the artificial abutment tooth be positioned in the regionof the attached gingiva? Int J Oral Surg 1985; 14: 504 –508. 179. B W, B BE. Replacement of maxillary and

mandibular molars with single endosseous implant res-161. K A, A KL. The IMZ osteointegratedimplant system. Dent Clin North Am 1989; 33: 733–791. torations. A retrospective study. J Prosthet Dent 1995;

74: 51–55.162. M RM, L B, F ME. Dental implants: a

review. J Periodontol 1992; 63: 859–870. 180. B TJ, H RE, P MC, R B.A comparative study of one implant versus two replacing163. A Z, T H, M O, K A. Mucosal

considerations for osseointegrated implants. J Prosthet a single molar. Int J Oral Maxillofac Implants 1996;

11: 372–378.Dent 1993; 70: 427–432.



550 Esposito et al.

181. B O, H M. Use of wide implants and integrated fixed prosthesis: A six-year longitudinal study.

Quintessence Int 1987; 18: 89–96.double implants in the posterior jaw: A clinical report.

Int J Oral Maxillofac Implants 1996; 11: 379–386. 199. A J, B K, G J, N H, OM, A P. Osseointegrated implants in edentulous182. H R, M N, M G, W G.

Branemark single tooth implants: A preliminary report of jaws. A 2-year longitudinal study. Int J Oral MaxillofacImplants 1990; 5: 155–163.76 implants. J Prosthet Dent 1995; 73: 274–279.

183. F B, I C-J, L U. Inferior alveolar 200. J G, P S. Complications, supplemen-tary treatment, and maintenance in edentulous archesnerve transposition in combination with Branemark

implant treatment. Int J Periodont Rest Dent 1992; 12

: with implant-supported fixed prostheses. Int J Prosthodont1990; 3: 89–92.441–449.184. R B. Implant placement in combination with 201. A R, E B, L U, B P-I, J

T. A long-term follow-up study of osseointegratednerve transpositioning: Experiences with the first 100cases. Int J Oral Maxillofac Implants 1994; 9: 522–531. implants in the treatment of totally edentulous jaws. Int

J Oral Maxillofac Implants 1990; 5: 347–359.185. H J-M, B P-I. Fixture stability and nerve

function after transposition and lateralization of the 202. T RG, M ME, A WF, MA JT.Endosseous cylinder implants in severely atrophic mand-inferior alveolar nerve and fixture installation. Br J Oral

Maxillofac Surg 1995; 33: 276–281. ibles. Int J Oral Maxillofac Implants 1991; 6: 264–269.203. J T. Failures and complications in 391 consecutively186. T DE. Reconstructive procedures with endosseous

implants in grafted bone: A review of the literature. Int inserted fixed prostheses supported by Branemark

implants in edentulous jaws: A study of treatment fromJ Oral Maxillofac Implants 1995; 10: 275–294.187. T JF. Osseointegrated fixtures in the pterygoid the time of prosthesis placement to the first annual

checkup. Int J Oral Maxillofac Implants 1991; 6: 270–276.region. In: W P, B P-I, eds.

Advanced osseointegration surgery. Applications in the max- 204. K JL, B BG. The immediate implant: Atreatment alternative. Int J Oral Maxillofac Implants 1991;illofacial region. Chicago: Quintessence Publishing Co.,

Inc., 1992; 182–188. 6: 19–23.205. T DE, L WR. Tissue-integrated prosthesis188. A R, L U, G K, B P-I,

L J, J M. Reconstruction of severely complications. Int J Oral Maxillofac Implants 1992; 7:477–484.resorbed edentulous maxillae using osseointegrated fix-

tures in immediate autogenous bone grafts. Int J Oral 206. O M, F B, N H, K K. MkII –A

modified self-tapping Branemark implant: 3-year resultsMaxillofac Implants 1990; 5: 233–246.189. A P, B K, G J, O M. Combination of a controlled prospective pilot study. Int J Oral

Maxillofac Implants 1995; 10: 15–21.of natural teeth and osseointegrated implants as prosthesisabutments: A 2-year longitudinal study. Int J Oral 207. H PJ, B RC, W CD. Rehabilitation of the

edentulous mandible with osseointegrated dental implants:Maxillofac Implants 1991; 6: 305–312.

190. W D, J KJ, C HA, 10 year follow-up. Aust Dent J 1995; 40: 1–9.208. Z GA, S A. The edentulous predicament I: aW AC, S JP, F JW , L WRE,

QAA. Experience with ITI osseointegrated implants prospective study of the eff ectiveness of implant-supportedfixed prostheses. J Am Dent Assoc 1996; 127: 59–65.at five centres in the UK. Br J Oral Maxillofac Surg 1992;

30: 377–381. 209. B TJ, E A, S T, V L. Three-

year evaluation of Branemark implants connected to191. M A, L NP. Antimicrobial treatment of peri-implant infections. Clin Oral Implants Res 1992; 3: angulated abutments. Int J Oral Maxillofac Implants 1997;

12: 52–58.162–168.192. L B, B U, H CHF, F I, 210. H KW, S A, Z GA. Complications

and maintenance requirements for fixed prostheses andLNP. Treatment of an early implant failure according

to the principles of guided tissue regeneration (GTR). overdentures in the edentulous mandible. A 5-year report.Int J Oral Maxillofac Implants 1994; 9: 191–196.Clin Oral Implants Res 1992; 3: 42–48.

193. K A, M PJ. The IMZ endosseous two phase 211. W PS, W RM, H MR. T he eff ects of prefabricated bar design on the success of overdenturesimplant system: a complete oral rehabilitation treatment

concept. J Oral Implantol 1986; 12: 576–589. stabilized by implants. Int J Oral Maxillofac Implants

1995; 10: 79–87.194. D DA, F B, W PA, P RM,

H TP, A D, M AH, S DC. A 212. Z GA, S A. The edentulous predicament II:

the longitudinal eff ectiveness of implant-supported overd-clinical and radiographic assessment of a porous-surfaced,titanium alloy dental implant system in dogs. Int J Oral entures. J Am Dent Assoc 1996; 127: 66–72.

213. H M, N I. Implant supported overdentures – Implantol 1987; 4: 31–37.

195. W RJ, B BA. An assessment of survival rates the Leuven experience. J Dent 1997; 25 (suppl.1): s25–s32.214. E I, L U, B P-I, L J,and within-patient clustering of failures for endosseous

oral implants. J Dent Res 1993; 72: 2–8. G P-O, N S. A clinical evaluation of fixed-bridge restorations supported by the combination of teeth196. H JE, H MR, C JY, H J, J T,

J RB, MK S, MN DC, V and osseointegrated titanium implants. J Clin Periodontol

1986; 13: 307–312.S D, T R, W RM, H I.Factors related to success and failure rates at 3-year 215. N M, L B. The successful application of

osseointegrated implants to the posterior jaw: A long-follow-up in a multicenter study of overdentures supportedby Branemark implants. Int J Oral Maxillofac Implants term retrospective study. Int J Oral Maxillofac Implants

1993; 8: 428–432.1995; 10: 33–42.

197. T MS, S J. Pathogenesis of implant failures. 216. Z GA, S A. The longitudinal clinical eff ect-iveness of osseointegrated dental implants in anteriorPeriodontol 2000 1994; 4: 127–138.

198. L LW, C GE, G P-O. partially edentulous patients. Int J Prosthodont 1993;

6: 180–188.Rehabilitation of the edentulous mandible with a tissue-




217. Z GA, S A. The longitudinal clinical eff ect- resorbed maxilla: A 2-year longitudinal study. Int J Oral Maxillofac Implants 1993; 8: 45–53.iveness of osseointegrated dental implants in posterior

partially edentulous patients. Int J Prosthodont 1993; 232. I S, E A, A P, B J-E.Early results from reconstruction of severely atrophic6: 189–196.

218. N M, L B. The successful use of osseointeg- (class VI) maxillas by immediate endosseous implants inconjunction with bone grafting and Le Fort I osteotomy.rated implants for the treatment of the recalcitrant perio-

dontal patient. J Periodontol 1995; 66: 150–157. Int J Oral Maxillofac Surg 1993; 22: 144–148.233. R GM, B TJ, R H, V O219. J T, P P. A 3-year follow-up study on single

implant treatment. J Dent 1993; 21: 203–208. RP. Augmentation of the maxillary sinus floor with auto-genous bone for the placement of endosseous implants: A220. E A, C GE, B G. Clinical evalu-

ation of single-tooth restorations supported by osseointeg- preliminary report. J Oral Maxillofac Surg 1993; 51:1198–1203.rated implants: A retrospective study. Int J Oral

Maxillofac Implants 1994; 9: 179–183. 234. K EE, E SE, T DE. Maxillary antraland nasal one-stage inlay composite bone graft:221. T B, O J, G K, F B.

Osseointegrated implants in adolescents. An alternative in Preliminary report on 30 recipient sites. J Oral MaxillofacSurg 1994; 52: 438–447.replacing missing teeth? Eur J Orthod 1994; 16: 84–95.

222. A B, O P, L A-M, L B. 235. J J, S-P S, O AJ. Varying treat-ment strategies for reconstruction of maxillary atrophySingle-tooth restorations supported by osseointegrated

implants: Results and experiences from a prospective with implants. Results in 98 patients. J Oral MaxillofacSurg 1994; 52: 210–216.study after 2 to 3 years. Int J Oral Maxillofac Implants

1995; 10: 702–711. 236. S H, N FW, S H, B KJ.Local ridge augmentation using bone grafts and osseointe-223. A-A L, Z GA. Clinical eff ectiveness of

implant-supported single-tooth replacement: The Toronto grated implants in the rehabilitation of partial edentulism:Preliminary results. Int J Oral Maxillofac Implants 1994;study. Int J Oral Maxillofac Implants 1996; 11: 311–321.

224. P SM, T RG. Immediate fixture placement: 9: 557–564.237. K L. A modified method of simultaneous boneA treatment planning alternative. Int J Oral Maxillofac

Implants 1990; 5: 337–345. grafting and placement of endosseous implants in theseverely atrophic maxilla. Int J Oral Maxillofac Implants225. T DE, K EE. Endosseous implant placement

immediately following dental extraction and alveoloplasty: 1995; 10: 682–688.238. B JE, A P, I S. RetrospectivePreliminary report with 6-year follow-up. Int J Oral

Maxillofac Implants 1991; 6: 24–28. analysis of one-stage maxillary sinus augmentation withendosseous implants. Int J Oral Maxillofac Implants 1996;226. BO. Osseointegrated implants in the maxillary tuber-

osity: Report on 45 consecutive patients. Int J Oral 11: 512–521.239. K PA , NA , M LE, E S,Maxillofac Implants 1992; 7: 459–467.

227. B W, D C, B BE, L U, V N B. Reconstruction of the resorbed edentulousmaxilla using autogenous rib grafts and osseointegratedS D, H K, K C. The use of

e-PTFE barrier membranes for bone promotion around implants. Clin Oral Implants Res 1996; 7: 286–290.240. L S, M P, J C, N H.titanium implants placed into extraction sockets: A pro-

spective multicenter study. Int J Oral Maxillofac Implants Augmentation of the maxillary sinus floor with partic-

ulated mandible: a histologic and histomorphometric1994; 9: 31–40.228. D C, L U, B W, B D, H study. Int J Oral Maxillofac Implants 1996; 11: 760–766.

241. A P, N PG, B P-I. TitaniumK, C A, V S D. Treatment of fenes-tration and dehiscence bone defects around oral implants implants and onlay bone graft to the atrophic edentulous

maxilla. A 3-year longitudinal study. Int J Oral Maxillofacusing the guided tissue regeneration technique: A prospect-ive multicenter study. Int J Oral Maxillofac Implants 1995; Surg 1996; 25: 25–29.

242. D P, H M, G F, M C.10: 312–318.229. B TJ, L HY, H RE. The use of pterygo- Autologous bone graft to augment the maxillary sinus in

conjunction with immediate endosseous implants: a retro-maxillary implants in the partially edentulous patient: Apreliminary report. Int J Oral Maxillofac Implants 1995; spective study up to 5 years. Int J Periodont Rest Dent

1997; 17: 27–39.10: 89–98.230. I S, A P. Maxillary alveolar ridge augmen- 243. Z GA, S A. The longitudinal clinical eff ect-

iveness of osseointegrated dental implants: The Torontotation with onlay bone-grafts and immediate endosseousimplants. J Craniomaxillofac Surg 1992; 20: 2–7. study. Part I: Surgical results. J Prosthet Dent 1990;

63: 451–457.231. N E, K K-E, G J. Bone grafts and

Branemark implants in the treatment of the severely

biological factors contributing jan-michae´l hirsch2,

Documents