applied osseointegration research - …...applied osseointegration research c/o professor tomas...

APPLIEDOSSEOINTEGRATIONRESEARCH

Volume 6: Clinical Performance and Enhanced StabilityFebruary 2008



Applied Osseointegration Researchc/o Professor Tomas Albrektsson

Department of BiomaterialsP.O. Box 412

SE-405 30 GothenburgSweden

Telephone +46 31-786-2945Fax: +46 31-786-2941

email: [email protected]





ISSN 1651-0070

Applied Osseointegration Research - Volume 6, 2008

1

Editor-in-ChiefTOMAS ALBREKTSSON

Sweden

Associate EditorsJAN GOTTLOW MAGNUS JACOBSSON LARS SENNERBY

Sweden Sweden Sweden

ANN WENNERBERG WARREN MACDONALD LARS RASMUSSON Sweden UK Sweden

Editorial Board Carlos Aparicio Burton Becker Sascha Jovanovic

Spain USA USA

Hugo de Bruijn Georg Watzek George A ZarbBelgium Austria Canada

Young Taeg Sul Michael Norton Antonio RocciKorea UK Italy


2

When are Implant Systems Really Safe for Clinical Use?

We are all concerned with the clinical safety of im-plant systems, be they dental implants, maxillofacial systems or orthopaedic components. In particular, the means by which an implant system might be proven safe before release for clinical use or even marketing continues to attract debate or criticism.

Computer modelling and laboratory testing are two well tried methods which can give useful indications in the search for safety of implants. Laboratory tests under accurately known and measured conditions, with carefully controlled variables and environments, can be helpful in comparing design solutions or com-petitive systems for similar applications. But systems which interface living bone and function within un-predictable human subjects can be subject to many factors for which we have little data and can provide insufficient predictions. Under such constraints, laboratory tests may not be realistic, and may give unreliable indications of implant performance in clinical practice (unreliably pessimistic or unreliably optimistic). In all implant fields one can recall stories of systems launched with claims of laboratory valida-tion, and complex and detailed test programmes in support of their clinical performance, which neverthe-less demonstrated poor relevance to clinical realities and ultimately gave poor results in clinical practice.

Computer modelling is similarly reliant on the input information and accuracy of conditions modelled. Simplifying assumptions are nearly always required, but do they undermine any relevance of the model-ling to the clinical situation? Marvellous complexity and incredible accuracy are possible with the com-puting power currently available, both for calcula-tions and graphical displays or simulations, but the basic question still remains – how realistic is the model or how true are the assumptions being made?

In the end, good clinical practice and good engineer-ing lead to the same conclusion – the ultimate test or proof is in the real situation; the patient. Laboratory and computer testing and modelling may or may not give confidence that every risk has been accounted for and every aspect of predicted service is accommodated in the system design and production. The real test (the “gold standard”) is performance in the clinical setting. Good engineering and thorough pre-clinical testing will reduce the risk to patients and the number of “clinical trial patients” required to be exposed to unproven components or systems, but no amount of careful pre-clinical testing can remove the need for carefully designed and monitored clinical trials.

So clinical results are the only complete and reli-able proof of a system’s suitability for clinical use.

Warren MacdonaldVisiting Lecturer, Department of Biomaterials,Institute of Clinical Sciences,University of Göteborg,GöteborgSWEDEN


3

The Neoss Implant System

Treatment with osseointegrated implants is an integral part of contemporary patient care. Many companies provide the dental community with the necessary products. Quite frequently they also provide us unnecessary products! What should I use? Which implant system is the best? Naturally, there is no single or easy answer to such a question. It’s a mat-ter of preferences. These preferences will of course vary from individual to individual, but also reflect different needs of the surgeon, the prosthodontist and the dental technician. Not to forget the staff member responsible for the inventory of the clinic!

The founders of the Neoss company, Dr. Neil Meredith and Engineer Fredrik Engman, have cre-ated an implant system that should fulfil all needs of versatility, flexibility and ease-of-use, but through a minimum of components and instruments.

I admit to finding the thinking and the design fea-tures of the Neoss implant system attractive. But does it work? What’s the scientific evidence? In this issue of Applied Osseointegration Research you will find a mix of clinical studies, animal experiments and laboratory tests evaluating the Neoss implant system.

The osseointegration process is influenced by the mechanical stability of the implant at surgery and the following biological response. The mechanical stability is related to the macro-design of the implant whereas the surface properties will influence the biological response. The modern moderately rough surfaces have demonstrated strong bone responses, allegedly shortening the healing time. However, the Neoss is a modern implant system that is, actually minimally rough (as demonstrated in surface topographical studies). Despite this the Neoss system obviously achieves quite positive results. Still, when using im-

mediate or early loading protocols, the system is challenged by loading the bone simultaneously with the healing process and development of stability.

The papers accepted for publication in this issue have assessed the mechanical stability of Neoss implants at surgery by insertion torque values and ISQ values (resonance frequency analysis). The biological response during healing is documented using histological analy-ses and biomechanical tests such as removal torque and ISQ values. The clinical outcome of two-stage, one-stage and immediate loading protocols has been evalu-ated at one-year follow-up by clinical and radiographic examinations. The biocompatibility of PEEK healing abutments is evaluated by means of bacterial coloniza-tion and real-time PCR (polymerase chain reaction) tests. The outcomes are, in my mind, excellent. Enjoy!

Jan Gottlow, DDS, Ph DGuest-Editor-in-Chief of Volume 6Department of Biomaterials,Institute of Clinical Sciences,University of Göteborg,GöteborgSWEDEN


4


5

Table of Contents

A REVIEW OF IMPLANT DESIGN, GEOMETRY AND PLACEMENTNeil Meredith 6

HISTOLOGICAL EVALUATION OF A BIMODAL TITANIUM IMPLANT SURFACE . A PILOT STUDY IN THE DOG MANDIBLE.Luiz A Salata1, Paulo EP Faria1, Marconi G Tavares1, Neil Meredith2,3 and Lars Sennerby4 13

HISTOLOGICAL AND BIOMECHANICAL ASPECTS OF SURFACE TOPOGRAPHY AND GEOMETRY OF NEOSS IMPLANTS. A STUDY IN RABBITS.Lars Sennerby1 , Jan Gottlow1, Fredrik Engman2 and Neil Meredith2,3 18

A ONEYEAR CLINICAL, RADIOGRAPHIC AND RFA STUDY OF NEOSS IMPLANTS USED IN TWOSTAGE PROCEDURESPeter Andersson1 , Damiano Verrocchi1, Rauno Viinamäki1, Lars Sennerby1,2 23

IMMEDIATE/EARLY LOADING OF NEOSS IMPLANTS. PRELIMINARY RESULTS FROM AN ONGOING STUDYPeter Andersson1 , Damiano Verrocchi1, Luca Pagliani2, Lars Sennerby3 27

A RETROSPECTIVE FOLLOWUP OF 50 CONSECUTIVE PATIENTS TREATED WITH NEOSS IMPLANTS WITH OR WITHOUT AN ADJUNCTIVE GBRPROCEDUREThomas Zumstein

1 and Camilla Billström

2 31

INSERTION TORQUE MEASUREMENTS DURING PLACEMENT OF NEOSS IMPLANTSLuca Pagliani1, Lars Sennerby2, Peter Andersson3, Damiano Verrocchi3 , Neil Meredith4 36

STRESS EVALUATION OF DENTAL IMPLANT WALL THICKNESS USING NUMERICAL TECHNIQUESRudi C. van Staden1, Hong Guan1, Yew-Chaye Loo1, Newell W. Johnson2, Neil Meredith3, 39

COMPARATIVE ANALYSIS OF TWO IMPLANTCROWN CONNECTION SYSTEMS A FINITE ELEMENT STUDYRudi C. van Staden1, Hong Guan1, Yew-Chaye Loo1, Newell W. Johnson2, Neil Meredith3, 48

COMPARISON OF EARLY BACTERIAL COLONIZATION OF PEEK AND TITANIUM HEALING ABUTMENTS USING REALTIME PCRStefano Volpe1, Damiano Verrocchi2, Peter Andersson2, Jan Gottlow3, Lars Sennerby2,3 54

SURVIVAL RATE, FRACTURE RESISTANCE AND MODE OF FAILURE OF TITANIUM IMPLANTS IN CLINICAL FUNCTION AND DYNAMIC LOADING.Neil Meredith1,2 and Fredrik Engman2 57


6

A Review of Implant Design, Geometry and Placement

Neil MeredithUniversity of Bristol, UK

INTRODUCTIONBone is a unique structural material. Its physical and mechanical properties mimic both natural materials such as wood and man-made materials including poly-mers. Its mechanical properties can be directly corre-lated to its complex structure and it should therefore be described as an anisotropic material (Carter & Beaupre, 2001). In contrast many man-made polymers have a uniform structure in all directions and can thus be con-sidered homogenous materials. Bone’s unique property is its ability to form new bone and to remodel exist-ing bone. This is especially important in its response to applied mechanical stresses (Carter et al, 1998). B o n e h e a l i n g c o n s i d e r a t i o n sIn the complete absence of stress bone can and will form and remodel; as with the bone formation occur-ring during osseointegration in the submerged phase of a two-stage implant (Lekholm & Zarb,1985). There is also evidence to show that the application of light mechanical loads can induce favourable stresses within the bone, which may induce acceler-ated and enhanced bone formation (Pilliar & Ma-niatopoulos, 1986) Much research has been carried out to analyse the influence of mechanical stress on bone formation, but it is extremely difficult to model these parameters in a laboratory or even an animal model (Brunski, 1999, Brunski et al, 1993). It is clear that the application to bone of mechanical stresses in excess of a certain threshold level will not induce bone formation but can lead to the formation of a fibrous tissue capsule and malunion, or in the case of dental implants a failure of osseointegration (Szmukler-Moncler et al. 1998). This threshold is not clearly known or defined and is likely to vary in individual cases and sites in relation to bone quality and quantity, the loading conditions applied, and the systemic and regenerative capacities of the patient (Brunski et al, 2001). There is clearly a need to op-timise the factors associated with implant placement and design which can optimise the conditions for bone remodelling or formation under immediate and early loading conditions (Sennerby & Roos, 1998)

Biomechanics of implant placement and loadingThere are three main biomechanical parameters that influence the stress distribution and optimal stability of an implant in bone; these are the placement pro-cedures including the drilling of the osteotomy site and the use of compression techniques to increase local stability (O’Sullivan et al. 2004a). Secondly, the design features of the implant itself (O’Sullivan et al, 2000) and thirdly, the loading conditions to which the implant is subjected (Friberg et al, 1991). Loading conditions may differ due to a single or two-stage surgical placement technique (Sennerby L Roos J;1998). In a two-stage technique, the implant fixture is placed, typically level with the crest of the bone and submerged beneath the soft tissue, for a healing period which may vary but has historically been rec-ommended as three months in the mandible and six months in the maxilla. A two-stage protocol effectively eliminates any dynamic functional loads during this healing and osseointegration period. A single-stage technique is different in that the implant is exposed to the oral environment at the time of initial surgery and placement. In this case, a number of options exist in that the implant may be loaded (immediate loading) or may remain unloaded by the provision of a relieved prosthesis over the implant site for a healing period, which may typically vary from six to eight weeks. Loading protocolsThe alternative is early or immediate loading, in which a prosthesis or temporary restoration is placed directly at the time of surgery or a short period thereafter; perhaps a week. Immediate loading and delayed one-stage loading are very different. In delayed one-stage loading it is likely that the effectively unloaded implant is subjected to small dynamic loads, applied through the soft tissue and through intermittent contact with a prosthesis (Orenstein et al.; 1998). The clinical evidence for this protocol is that it is highly successful and that the dynamic loads if any are small enough and of appropriate frequency that they do not lead to a failure of osseointegration and


7

formation of a fibrous tissue capsule. However, there is no strong evidence to suggest that this early load-ing one-stage technique will actually accelerate bone formation or enhance the quality of bone formed. Immediate and early loading, attachment of a pros-thesis at the time of implant placement or shortly, thereafter relies on the principal that the dynamic functional loading applied will be below the thresh-old which can induce failure of osseointegration and formation of a fibrous tissue capsule, and at a level whereby bone remodelling and formation may progress unhindered or even accelerated in the early healing stage. The ranges of clinical, anatomical and surgical parameters are very wide and therefore at present the selection, use and success of immediate and early-loading techniques are generally founded on the experience, knowledge and understanding of the clinician on an individual case by case basis.

Primary Implant StabilityIt is clear that one of the keys to successful osseointe-gration is the primary stability of an implant. It is considered highly desirable that this level of stability should be as high as possible. In experienced hands it is commonly measured by the insertion torque necessary to place the implant. This is clearly a subjective feeling, but can give an experienced operator a level of confi-dence in the prospects of a successful outcome for a case.

Electronic drill controllers displaying graphs of inser-tion torque are available, but the interpretation of such data is difficult (Friberg B, Sennerby L, Roos J; 1995). An alternative technique is resonance frequency analy-sis (RFA)(Meredith N, Cawley P, Alleyne D; 1996), which has been available for some ten years. It utilises a non-destructive test method to measure the local in-terfacial stiffness of an implant and surrounding bone.

Secondary Implant StabilityIt is evident in the first six to eight weeks following placement that there is more new bone formation in poorer quality bone, typically in the anterior max-illa. Here the blood supply can be good but there is an open trabecular network and primary stability is typically lower than in the mandible for example. In the mandible, there are smaller changes in stability and these may be accompanied by local remodel-ling rather than new bone formation (Meredith et al. 1997). This can result in a measurable increase

in stability in bone qualities 3 and 4, more marked than in bone quality 1 and 2 (Andersson et al. 2007)

Clinical findings indicate that a large part of the healing and remodelling process, which is termed os-seointegration, is probably completed within the first eight weeks of placement, under normal conditions. Higher primary stability at placement is also measured (Meredith N, Bok, K, Friberg, Sennerby ; 1997) in better quality and denser bone. O’Sullivan (2001) measured implant stability as a function of the changes in strain following implant placement for a period of two hours following placement . He observed a sharp initial fall in stability and decrease in interfacial strain.

This is interesting because it is not a biological or physi-ological phenomenon, what is occurring is mechanical stress relaxation in the bone following placement. This suggests that although a high level of stability and compression may occur at the time of insertion, this stability may decrease very rapidly thereby creating a higher risk situation. This could be especially impor-tant where implants are placed in poor bone quality, or where the bone has been artificially compressed to a high level by the use of osteotomes for example.

It is clear therefore that there is a very complex and subtle inter-relationship between bone qual-ity, stability, geometry and placement technique. Implant GeometryA range of geometries has been available for dental implants for a number of years and their variation and relation to success do warrant some observations. It is interesting to note that historically the cylindrical implant has been associated with a relatively high in-

Figure 1. illustrates the changes in implant stability at diferent clinical stages following implant placement. (After Andersson et al. 2007)


8

cidence of implant failure (Albrektsson T, Sennerby L; 1991). This has also born some relationship to specific systems and it is important to be certain that the aeti-ology of this failure is due to the geometry alone; the evidence suggests that this geometry could play a part. However, by relatively small modifications in geom-etry and placement technique it is possible to achieve a highly successful implant system, the Straumann (Basle, Switzerland) systemwas essentially cylindri-cal but with a small widely spaced thread super-im-posed. The drilling protocol for this implant uses a small modification with the implant being 0.05mm larger than the preparation site. The idea is that this creates localised compression around the implant. In general, this seems to be successful, although the screw pattern used is that of a Thorpe screw, which was designed for use in orthopaedic surgery to pull bone plates together (Ansell R, Scales J; 1968). Threaded implants with a close pitch and a deep profile (Brånemark; Nobelbiocare, Gothenburg, Sweden) are quite typical of another design of im-plant which has been clinically highly successful. The variations in stress distribution between different implant systems under applied loading would there-fore appear to be quite substantial. However, both these geometry types (the threaded cylinder and the threaded implant) are equally successful clinically.

Figure 2. Implant stability for different implant types and bone quality in the maxilla (After O’Sullivan, Sennerby, Meredith; 2000)

Implant geometries have not been routinely designed for use in specific bone or quality types. However, implants having a slightly tapered geometry (approxi-mately 4° from parallel) have been introduced to create compression in poorer bone qualities and optimise sta-bility (Brånemark MkIV; Nobelbiocare, Gothenburg, Sweden) . Figure 2 Illustrates the stability at placement for a number of implant types in the human ca-daver maxilla (O’Sullivan, Sennerby, Meredith; 2000)

Static and dynamic stressesWhat is apparent is that two separate issues need to be considered in the successful placement and loading of a dental implant. These are the dynamic stresses experienced by an implant in function and the static stresses encountered during implant place-ment. Dynamic stresses and applied loads can be considerable once an implant has reached an equilib-rium position in bone and healing has taken place.

Prior to this the nature, magnitude and direction of applied dynamic loads can influence the treatment outcome (Goodman et al. 1993). Static loads at the time of implant placement, however, are a con-sequence of those techniques or geometry that are designed to contribute to the maximal stability of an implant in bone. The use of slightly tapered threaded implants is one such way of seeking to increase stability in poor bone qualities (O’Sullivan 2001). Is compression and the application of high levels of primary stress a clinical issue? The answer is potentially yes, both at the time of placement in high initial bone qualities and at the time of abutment connection and


9

loading in two-stage implants (Misch C E,1990). A number of implant systems use a drilling sequence that creates a preparation site that is only very slightly small-er than the implant, being inserted. This will quite obviously create quite low levels of insertion stress. The evidence is that these implants are highly successful in average to good bone qualities where compression and stability are adequate and over-compression is avoided. In systems inducing a high level of compression, pos-sibly by use of a tapered implant or a drilling sequence with a drill size much smaller than the implant diam-eter there are occasions when the consequences of over-compression may be early failure of the implant itself (Hobkirk JA, Rusiniak K,1977) (Ivanoff C-J 1999).Figure 3. illustrates the variation in insertion torque for 3.75 mm implants placed in final osteotomies of differing diameter. This clearly demonstrates a relationship between compression of the oste-otomy site and insertion torque. (O’Sullivan, 2001)

Most implant designs do not directly address the op-timisation of stability in poor bone qualities. Some implants have become available with a slightly tapered geometry, which creates local compression and thereby achieves good stability; these have been very successful.

The advantage of having a geometrical feature on the implant is that it does not rely on a complex placement or drilling protocol to create a level of compression, which may be variable in relation to its outcome. The slightly tapered implant thus provides a simple way of repeatedly and consistently offering an increase in sta-bility in poor bone qualities (O’Sullivan et al. 2004b). The disadvantages of modified geometry implants of this type are that they can lead to over-compression when used in good bone qualities and they may either fail to seat or on some occasions may lead to failure. What is needed therefore is an implant geometry and a placement method which can achieve a high level of sta-

Figure 3. Variation in insertion torque with drilling depth as a function of time for 3.75mm implants placed in final osteotomy diameters 2.85-3.7mm. (O’Sullivan, 2001)

bility without over-compressing good bone and yet can optimise the compression and stability for placement in poor bone qualities. This would thereby optimise the success of implant placement in higher risk areas. Implant DesignSuch considerations have not been made hitherto in relation to implant design and placement tech-niques. This proposes a combination of geometry, surface characteristics and placement procedures which will optimise implant placement and stress distribution in all bone qualities. To achieve this it is evident that a tapered geometry will offer benefits in bone compression and there is good evidence that only a small degree of taper achieves a substan-tial increase in stability (O’Sullivan et al., 2004b). It is therefore desirable to have an implant with a positive tolerance (taper) which will cause optimal compression in poor bone qualities. In order to address the seating, stability and compression levels within good bone qualities further consideration needs to be paid to the surface geometry. Many of today’s implant systems use a thread-cutting geom-etry to tap a thread into the bone during insertion into a cylindrical hole. This works very effectively and creates a high level of bone-to-implant contact. A possible thread refinement from current implants is altering the implant to bone volume ratio within the threads by reducing the thickness of the implant threads. An important feature of the geometry of the thread cutting face is that there is adequate volume in the relief chambers for bone clearance (Figure 4.). It is important that bone clearance chambers are designed to maximise the volume for bone chip entrapment but also to provide maxi-

Figure 4. Bone relief chambers and cutting face


10

mal bone to implant contact on the threaded area. Historically some implant designs have been less suc-cessful because the bone collection chambers were very wide and very openly spaced (Friberg et al. 1997). It is also important that the cutting face of the self-tap-ping implant feature is sharp and without burrs. In order to optimise the placement of a threaded implant, having a positive tolerance in good bone qualities, a secondary cutting feature can be introduced along the side of the implant (Figure 4 and 5.). This secondary cutting face is much shallower than the apical cutting face and actually does not engage in soft bone qualities. In dense bone however, when the implant is inserted there will be elastic recovery during the insertion process such that the bone will engage the secondary cutting faces and a small amount will be removed.

Figure 6 illustrates an insertion torque profile for the Neoss implant demonstrating a near linear increase in insertion torque with a modulated plot attribut-able to secondary cutting (Luca et al., 2007). This will not impair the stability of the implant in the site but will create a different fits in dense bone quality or weak bone, thereby optimising stability in all bone qualities but without over compressing good bone. The combination of features relating implant geometry with forming and thread cutting in bone of different quality has been combined into a single implant design and the concept is called TCF® which represent a Thread Cutting and a thread Forming implant leading to cutting for optimal seating and forming for optimal stability.

Screw taps work differently from the use of progres-sively increasing drill sizes to match implant diam-eters. They are available for use on rare occasions where there is uniformly dense cortical bone along the whole implant length. In such cases, the com-pression occurring at the implant tissue interface is different from that under typical conditions where there is a combination of cortical and trabecular bone.

Under normal conditions the TCF feature of the Neo implant is designed to create an optimal level of compression, starting from the apex of the threads as the implant is inserted into a cylindrically prepared site. In uniformly dense bone, the use of a screw tap to pre-tap the site will create a level of compression on implant insertion that is uniformly applied to both the peaks and troughs of the thread, thereby achieving a comparable overall level of compression in a structur-ally different quality of bone. Screw-taps are therefore

Figure 5. Implant design geometry for the Neoss implant

recommended on occasions where bone density is extremely high and bone quality is very homogenous. Insertion TorqueInsertion torque is a commonly assessed qualitative parameter during implant placement. Placements and procedures vary considerably between systems and between operators. Some clinicians favour a very high level of final insertion torque and other operators work with a very gentle insertion technique, rarely encountering an insertion torque greater than 30Ncm. Historically, a high final peak insertion torque may be variously affected by three phenomena. The implant may bottom out in the site, so that the final tightening torque is actually inducing very high levels of shear stresses at the implant tissue interface, as the implant butts against the bottom of the preparation site. A second cause of a high insertion torque may be contact at the flange of the implant with the cre-stal cortical plate, thereby achieving a high level of compression and static stresses. The third reason is a very high level of interfacial stresses leading to a high level of implant stability. On occasion, this may or may not accompany a level of over-compres-sion and very rarely it may lead to implant failure.

The Neoss implant system has been designed to achieve the optimal level of compression and stability, for implant placements of bone in all qualities. This does not rely on a very high level of final insertion torque. Excellent results can be obtained using a gentle tech-nique where the insertion torque at any period dur-


11

Figure 6. Insertion torque plot for Neoss implant.

ing placement does not exceed 30ncm. The optimal tissue response can be expected to be achieved if an insertion torque between 25 and 30Ncm is obtained.

Historically, if implants fail to seat or encounter very high insertion torques, surgeons typically remove the implant and use a screw-tap to aid the preparation. In the case of a Neo implant, if the insertion torque dur-ing placement reaches high levels then the implant can simply be anti-rotated a few turns and then re-inserted. This clears bone swarf from the cutting faces of the im-plant and reduces friction at the interface, thereby allow-ing smooth insertion without any risk of over-heating. A second feature that is important during implant place-ment is the static stresses obtained between the inter-face of the implant flange and the crestal, cortical bone. Perifixtural bone lossA clear and well-recognised characteristic of the ex-ternal hexed Brånemark implant is the loss of bone in the period following abutment connection and early loading from the level of the abutment-implant interface down to the first thread of the implant. The aetiology of this is not clear, but a number of hypoth-eses have been put forward. It has been proposed for example that there is micro-leakage between the abutment and the implant, of the implant abut-ment interface, resulting in the release of bacterial toxins causing a local peri-implant reaction and bone loss within a localised zone around this interface. A second cause proposed is the surface characteristics of the implant; it has been suggested that one im-plant system with micro grooves has a more favour-able coronal stress response and encourages bone formation by the presence of these micro grooves.

A third, but not commonly discussed reason may well be related to mechanical stresses. This is particularly prevalent with the earlier designs of implants using commercially pure, type I titanium, which is rela-tively soft. In such cases the combination of a static load at implant placement around the flange between the implants and the cortical bone and then the su-perimposed dynamic stresses at the time of implant loading will cause a high level of mechanical stress and bending within the implant body, around the neck, between the flange and the threaded body of the implant. The bone response to these localised stresses is likely to be resorption; the high levels of stress at the implant neck can be visualised by a stress profile, superimposed on an implant in bone in this region.

It is therefore highly desirable to have a flange and neck design that minimises both the static stresses at place-ment and the dynamic stresses on the functional load-ing. The Neoss implant has been designed specifically without a neck region, as in the external hex implant designs, and the thread leads directly into the flange. The relationship between the flange and the threaded portion of the implant is slightly different for the 3.5, 4 and 4.5mm diameters. This enables the use of one common abutment connection without impairing the fit of seating of the implant in the surrounding bone.

In the flange region, it is therefore possible to assess the level of compression occurring in the flange dur-ing implant insertion. The Neo system has great For a two-stage implant with a high level of surrounding bone a counter sink can be used to provide the opti-mal seating, minimising static stresses and optimising the interface between the flange and the surrounding bone, thereby providing optimal conditions for direct bone formation. The Neo implant system is there-fore designed with a number of features in geometry, preparation technique and material properties that jointly result in the optimal biomechanical relationship between a dental implant and the surrounding bone.

REFERENCESAlbrektsson T, Sennerby L (1991). State of the art in oral implants. J. Clin. Periodontol.; 18:474-81.

Andersson P ,Verrocchi D, Viinamäki R, Sennerby L. (2008) A One-Year Clinical, Radiographic and RFA Study of Neoss Implants Used in Two-Stage Procedures. Appl. Oss. Res. 7;pp

Ansell R, Scales J (1968). A study of some factors which affect the


12

strength of screws and their holding power in bone. J. Biomechanics; 1:279-302.

Brunski, JB, Skalak, R. Biomechanics of osseointegration and dental prostheses. In: Naert I, van Steenbcrghe I) Worthington P, eds. Os-seointegration in Oral Rehabilitation, Quintessence Pub. Co., London, 1993:133-156.

Brunski, JB. In vivo bone response to biomechanical loading at the bone-dental-implant interface. Adv Dent Res 1999; 13:99119.

Brunski J, Antonio N, Helms J Implant Stability and the Bone-Implant Interface. An Editorial for Applied Osseointegration Research. Applied Osseointegration Research –Volume 2, Number 1, 2001

Carter DR, Beaupe GS, Giori NJ, Helms. Mechanobiology of skeletal regeneration. Clin Orthop Rel Res 1998;355:s4l-55.

Carter DR, Beaupre GS. Skeletal Form and Function. Cambridge Univ. Press, Cambridge UK, 2001

Friberg B, Jemt T, and Lekholm I Early failures in 4641 consecutively placed Brånemark dental implants. A study from stage I surgery to the connection of completed prostheses. Int. J. Oral Maxillofacial. Implants; 1991:6:142 – 146.

Friberg B, Sennerby L, Roos J, et al Evaluation of bone density using cutting resistance measurements and microradiography: an in vitro study in pig ribs. Clin Oral Impl Res 1995;5:164-171.

Friberg B, Nilson H, Olsson M, Palmquist C (1997). Mk II: the self-tap-ping Brånemark implant. 5-year result of a prospective 3-center study. Clin. Oral Impl. Res.; 8:279-285.

Goodman S, Wang JS, Doshi A, Aspenberg P (1993). Difference in bone ingrowth after one versus two daily episodes of micromotion: experiments with titanium chambers in rabbits. J. Biomed. Mater. Res.; 27:1419-1424.

Hobkirk JA, Rusiniak K (1977). Investigation of variable factors in drill-ing bone. J. Oral Surgery; 35:968-973

Ivanoff CJ, Sennerby L, Lekholm U. Influence of initial implant mobility on the integration of titanium implants. An experimental study in rab-bits. Clin Oral Impl Res 1996; 7 120-127

IvanoffC-J(1999). On surgical and implant related factors influencing integration and function of titanium implants – expenmental and clinical aspects. Thesis. University of Goteborg, Sweden.

Johansson P, Strid KG. Assessment of bone quality from cutting resist-ance during implant surgery. Int J Oral Maxillofac Impl 1994;9:279-288

Lekholm U, Zarb GA. Patient selection and preparation. In: Brane-mark Pi , Zarb (iA, Albrektsson T (eds): Tissue-integrated prostheses: Osseointcgration in clinical dentistry. Quintessence, Chicago 1985, pp 199-209

Meredith N, Alleyne D, Cawley P. Quantitative determination of the sta-bility of the implant-tissue interface using resonance frequency analysis. Clin Oral Impl Res 1996; 7: 261-267

Meredith N, Shagaldi F, Alleyne D, et al. The application of resonance frequency measurements to study the stability of titanium implants dur-ing healing in the rabbit tibia. Clin Oral Impl Res 1997; 8:234-243.

Meredith N, Book K, Friberg B, Jemt T, Sennerby L. Resonance fre-quency measurements of implant stability in vivo. A cross-sectional and longitudinal study of resonance frequency measurements on implants in the edentulous and partially dentate maxilla. Clin Oral Impl Res

1997;8:226-33.

MischCE(1990). Density of f bone: Effect on treatment plans, surgi-cal approach, healing and progressive bone loading. Int. J. Oral Impl.; 6:23-31.

Orenstein IH, Tarnov/ DP, Morris HF, Ochi S. Factors affecting implant mobility at placement and integration of mobile implants at uncovering. J Periodontol 1998; 69: 1404-1412

Pilliar RM, Maniatopoulos C. Observations on the effect of movement on bone ingrowth into porous-surfaced implants. Clin Orthop Rel Res 1986;208:108-113.

Sennerby L, Thomsen P, Ericsson L (1992). A morphometric and bio-mechanic comparison of titanium implants inserted in rabbit cortical and cancellous bone. Int. J. Oral Maxillofac. Implants; 7:62-71.

Sennerby L, Roos J. Surgical determinants of clinical success of osseointe-grated oral implants: a review of the literature. Int J Prosthodont 1998; 11: 408-420

Szmukler-Moncler S, Salama H; Reingewirtz Y, Dubruille JH. Timing of loading and effect of micromotion on bone-dental implant interface: review of experimental literature. J Biomed Mater Res Appl Biomat 1998;43:192-203.

O’Sullivan D, Sennerby L, Meredith N. (2000) Measurements compar-ing the initial stability of five designs of dental implants: a human cadaver study. Clin Implant Dent Relat Res. 2000;2(2):85-92.

O’Sullivan (2001) The effect of implant geometry upon the primary stability of dental implants. PhD Thesis, University of Bristol.

O’Sullivan D, Sennerby L, Jagger D, Meredith N.A (2004a) Comparison of two methods of enhancing implant primary stability. Clin Implant Dent Relat Res.;6:48-57.

O’Sullivan D, Sennerby L, Meredith N. Influence of implant taper on the primary and secondary stability of osseointegrated titanium implants.Clin Oral Implants Res. 2004b;15:474-80.


13

The aim of the present pilot study was to analyse the bone tissue response to a novel implant surface (Bimo-dal surface, Neo Implant System™) in comparison with two well-documented and commercially available im-plant surfaces (TioBlast™ surface and TiUnite™ surface).

MATERIALS AND METHODS

ImplantsA total of eight test implants, 9 mm long and 3.5 mm in diameter (Neo Implant System™, Neoss Ltd, Harrogate, UK) (NE implants) were used in the study. These implants had a bimodal surface created by blast-ing with 100 to 300 μm diameter ZrO2 spheres and subsequent blasting with irregularly shaped TiO-based particles, 75 to 150 μm wide (Fig. 1a). Eight control implants were also used; four implants with a TiO2-blasted surface, 9 mm long and 3.5 mm in diameter (MicroThread™, AstraTech AB, Mölndal, Sweden) (AT implants)(Fig. 1b), four implants with an oxi-dized surface, 10 mm long and 3.75 mm in diameter (Brånemark System™, MKIII, TiUnite, Nobel Biocare Ab, Gothenburg, Sweden) (NB implants)(Fig 1c).

Animals and anaesthesiaFour mongrel male dogs weighing between 20 and 25 kg were used in the study. The animals were pre-anaesthetized with xilazine (Ronpum®, Brazil, 20 mg/Kg I.M.) and ketamine 1g (Dopalen®, Brazil, 0,8 g/Kg I.M.) and anaesthetized with thionembutal 1 g (Tiopental®, Brazil, 20 mg/Kg I.V.). The animals were kept on intravenous infusion of saline during surgery, all of which was carried out under sterile

Histological Evaluation of a Bimodal Titanium Implant surface . A Pilot Study in the Dog Mandible.

Luiz A Salata1, Paulo EP Faria1, Marconi G Tavares1, Neil Meredith2,3 and Lars Sennerby4

1Dept. Oral & Maxillofacial Surgery, Faculty of Dentistry of Ribeirao Preto, The University of Sao Paulo, Brasil2University of Bristol, Bristol, UK3Neoss Ltd, Harrogate, UK4Dept Biomaterials, Institute of Clinical Sciences, Sahlgrenska Academy, Göteborg University, Sweden

This histological pilot study revealed a favourable bone tissue response to a novel bimodal titanium implant surface after four months of healing, with no apparent differences from TiO-blasted and oxidized control implants.

INTRODUCTION

New implant systems with different geometries and surface topographies are continually being launched on the market. It is important to evaluate critically each implant surface in both experimental models and in clinical follow-up studies. One prerequisite for a suc-cessful clinical outcome with osseointegrated titanium implants is secure bone integration immediately fol-lowing surgery (Albrektsson et al. 1981). In essence, the surgical trauma initiates a healing process which includes the formation of a blood clot, migration and differentiation of cells, formation of a granulation tissue and, finally, bone formation and remodelling. In the presence of a titanium surface, healing results in formation of direct bone-implant contacts (BICs) and the number and extent of BICs increase with time (Johansson et al. 1987, Sennerby et al. 1993). The peak torque required to achieve implant removal increases with time in parallel with increased BICs (Johansson et al. 1987). The first generation of osseointegrated implants had either a minimally rough surface (ma-chined/turned surface) or a very rough surface pro-duced by titanium plasma spraying (TPS)(Brånemark et al. 1969, Schroeder et al. 1976). On the basis of further clinical and experimental research it is currently believed that moderately rough implant surfaces are preferable (Albrektsson & Wennerberg 2006); such as surfaces produced by blasting, anodic oxidation, acid etching or combinations of these techniques. Experi-mental research has generally demonstrated a stronger bone tissue response to surface modified implants than to smoother control surfaces, indicating more rapid integration (Albrektsson & Wennerberg 2006).


14

Figure 1: Scanning electron micrography (SEM) of the three surfaces evaluated;a/ Bimodal surface,b/ TiO-blasted surface, c/ Oxidized surface

conditions. After surgery the animals received intra-venously vitamin compound (Potenay®, Brazil); an anti-inflammatory/analgesic (Banamine®, Brazil) and antibiotic (Pentabiótico®, Brazil). The antibiotic was administered in single doses immediately after sur-gery, and then 48 and 96 hours postoperatively. The study protocol had been approved by the University of Sao Paulo´s Animal Research Ethics committee.

Exprimental protocolThe mandibular premolars were extracted five months prior to commencement of the experiment. At the time of implant placement, crestal incisions were made and mucoperiosteal flaps were raised bilaterally. Two implant cavities were prepared on each site, in accordance with the manufacturers guidelines. Two NE implants were placed on one side and one each of AT and NB implants on the contra lateral side (Figs. 2a and b). Cover screws were placed and the flaps were closed and sutured.

After 4 months of healing, the dogs were sacrificed and the implants with their surrounding tissues were harvest-ed and fixed by immersion in buffered formaldehyde.

HistologyThe fixed specimens were dehydrated in a graded series of ethanol and embedded in light curing methacrylate (Technovit® 7200 VCL, Kulzer, Friedrichsdorf, Germa-ny). Ground sections approximately 10 μm thick were prepared using a sawing and grinding technique (Exakt Apparatebau®, Norderstedt, Germany) and stained with toluidine blue. One central section was taken from each implant site in the bucco-lingual direction.

The sections were examined under a Leitz micro-scope equipped with a Microvid system for morpho-metrical measurements. The degree of bone-implant contact (BIC) was measured from the first bone contact and expressed as a percentage mean total BIC. The bone area (BA) within the implant threads was measured and expressed as a mean total BA.


15

Figure 2. Clinical photographs showing a/ TiO-blasted (left) and oxidized (right) implants.b/ Bimodal surface implants.

RESULTS AND DISCUSSION

Healing was uneventful in all dogs. Some mar-ginal bone resorption, especially on the buccal aspect, was seen for all implant types at the mar-ginal portion which may be due to remodelling continuing after tooth extraction (Arajou et al. 2005).

Histology revealed close contact between mature bone and test implants (Figs 3). There were no apparent differences between test and control implants and new bone was observed to fill the threads of all three surfaces (Figs. 4a-c). In some areas, the presence of a thin layer of bone and an osteoblast seam facing the bone marrow indicated bone formation directly at the test implant surface (Fig. 5); as previously described for the control implant surfaces used in the study, i.e. blasted and oxidized surfaces (Ivanoff et al. 2001, Rocci et al. 2003). It has been speculated that such direct bone formation seen at surface modified implants is the result of an adherent blood clot through which mes-enchymal cells can migrate and differentiate to form bone directly on the implant surface (Davies 2003).

The morphometric measurements showed no apparent differences with regard to BIC or BA as similar mean values were obtained (Fig. 5). However, few animals were used and statistics could not be applied.

CONCLUSION

The present pilot study revealed bone integration of the novel bimodal implant surface, with no ap-parent differences from TiO-blasted and oxidized control implants after four months of healing.

Figure 3. Light micrograph showing overview of a test implant (left) in contact with mature bone (B). The implant is covered by oral mucosa (OM) lined by an epithelium (E). The arrow points to the first bone contact.


16

Figure 4. Light micrographs showing bone formation towards all three implant surfaces.: a/ Bimodal surface, b/ TiO-blasted surface, c/ Oxidized surface

REFERENCES

Albrektsson T, Brånemark PI, Hansson H, Lindström J. Osseointegrated implants. Requirements for ensuring a long-lasting direct bone-to-im-plant anchorage in man. Acta Orthop Scand 1981;52:155-170

Albrektsson T, Wennerberg A. Oral implant surfaces: Part 1--review focusing on topographic and chemical properties of different surfaces and in vivo responses to them. Int J Prosthodont. 2004;17:536-43.

Araujo MG, Sukekava F, Wennstrom JL, Lindhe J. Ridge alterations following implant placement in fresh extraction sockets: an experimental study in the dog. J Clin Periodontol. 2005;32:645-52.

Branemark PI, Adell R, Breine U, Hansson BO, Lindstrom J, Ohlsson A. Intra-osseous anchorage of dental prostheses. I. Experimental studies. Scand J Plast Reconstr Surg. 1969;3(2):81-100.

Davies JE, Hosseini MM. Histodynamics of endosseous wound healing. In Davies, JE (ed) Bone engineering. Em squared incorporated, Toronto 2000, pp1-14

Davies JE. Understanding peri-implant endosseous healing. J Dent Educ 2003;67:932-949

Ivanoff CJ, Hallgren C, Widmark G, Sennerby L, Wennerberg A. His-tologic evaluation of the bone integration of TiO(2) blasted and turned titanium microimplants in humans.Clin Oral Implants Res. 2001;12:128-34.


17

Figure 5. Light micrograph of a bimodal implant surface (I). Bone (B) is formed from the surface and towards the bone marrow (BM). Os = osteoid, V= vessel

Figure 6. Results from morphometrical measurements of BIC and BA.

Johansson C, Albrektsson T. Integration of screw implants in the rabbit: a 1-year follow-up of removal torque of titanium implants. Int J Oral Maxillofac Implants. 1987;2:69-75.

Rocci A, Martignoni M, Burgos PM, Gottlow J, Sennerby L. Histol-ogy of retrieved immediately and early loaded oxidized implants: light microscopic observations after 5 to 9 months of loading in the posterior mandible. Clin Implant Dent Relat Res. 2003;5 Suppl 1:88-98.

Schroeder A, Pohler O, Sutter F. [Tissue reaction to an implant of a tita-nium hollow cylinder with a titanium surface spray layer] SSO Schweiz Monatsschr Zahnheilkd. 1976;86:713-27.

Sennerby, L., Thomsen, P. and Ericson, L.E. Early bone tissue response to titanium implants inserted in rabbit cortical bone. I. Light microscopic observations. J Mat Sci: Mat Med 1993; 4:240-250

0

10

20

30

40

50

60

70

80

90

100

Bone area BIC

(%)

AstraNobelNeoss


18

Histological and Biomechanical Aspects of Surface Topography and Geometry of Neoss Implants. A Study in Rabbits.

Lars Sennerby1 , Jan Gottlow1, Fredrik Engman2 and Neil Meredith2,3

1Department of Biomaterials, Institute of Clinical Sciences, Sahlgrenska Academy, Göteborg University, Sweden2Neoss Ltd, Harrogate, UK3University of Bristol, Bristol, UK

This experimental study showed evidence of surface mediated bone formation on the bimodal titanium surface as previously described for other commercially available surface modified implants. Removal torque tests showed increased stability by adding a modified surface and vertical flutes as compared to turned control implants without flutes.

INTRODUCTIONOsseointegrated implants are clinically successful if a direct bone-implant contact can be established and maintained (Albrektsson et al. 1981). The bone-im-plant interface is biomechanically challenged in rota-tional, axial and lateral directions during healing, the prosthetic phase and clinical function. The ability to withstand loading is decisive for the clinical outcome and factors of importance are (i) type and magnitude of loading, (ii) the quality of the bone-implant integration and (iii) the mechanical properties of the surrounding bone. Implant integration is time dependent and the biomechanical properties of the bone-implant interface improve with time (Johansson et al. 1987, Sennerby et al. 1993, Friberg et al. 1999). Therefore, the use of a two-stage procedure with three to six months of heal-ing usually ensures a mature bone-implant interface and good clinical results. However, the trend today is to use immediate/early loading protocols, which make great demands on the bone-implant interface since the implants will be loaded during initial healing.

The first generation of osseointegrated implants had a relatively smooth (machined, turned) surface (Brånemark et al. 1969). Good long-term clini-cal outcomes have been reported on all indications when used in good bone qualities and using a two-stage procedure (Albrektsson & Sennerby, 1991). However, in more challenging situations such as low bone densities, bone grafting and immediate loading, increased failure rates have been reported (Friberg et al. 1991, Becktor et al. 2004, Glauser et al. 2001).

Surface modification is one way of improving implant

integration and stability, as shown in numerous experi-mental studies (Albrektsson & Wennerberg 2004). It is believed that the surface irregularities ensure a firm contact with the blood clot allowing primitive cells to migrate to the interface, differentiate to osteoblasts and form bone directly on the surface (Davies 2003). For a smooth surface, shrinkage of the blood clot will create a gap at the interface and cells cannot reach the surface (Miranda-Burgos et al. 2007). Thus, implants with a moderately rough surface integrate more rapidly and with more bone contacts than smooth surfaced implants (Ivanoff et al. 2001, Zechner et al. 2003).

A second means of improving implant integration is by geometric features. Most dental implants are self-tap-ping and have an apical configuration including cutting edges and bone chambers. Bone ingrowth into such voids is most likely to increase the rotational stability of the implant. Recent research has indicated that grooves at the thread flank may lead to improved healing by guided bone formation as well as to an improved inter-lock with bone (Hall et al, 2005). The Neoss implant is a self-tapping implant with apical bone chambers and vertical flutes. It has a bimodal surface topogra-phy which is produced by blasting with two different sizes of ZrO and Ti-based particles. The influence of the surface topography and geometry on the bone tissue response and stability is presently not known.

The present study was conducted to examine the early tissue responses to the bimodal titanium surface. The aim was also to evaluate the effect of surface topogra-phy and geometrical features on rotational stability.


19

Figure 3a. Close up of an implant with bimodal surface and vertical flutes (B+). 3b. Non-fluted bimodal implant (B-)


ImplantsA total of 96 implants, 7 mm long and 4.0 mm in diameter were implanted in the study. These were both original and modified Neoss implants (Neoss Ltd, Harrogate, UK) as follows (fig. 1);

48 implants with a bimodal surface created by blasting with 100 to 300 μm wide ZrO2 spheres followed by irregularly shaped Ti-based particles, 75 to 150 μm wide (Fig 2).

- 36 with vertical flutes (original surface and geometry)(B+)(Fig 3a)

- 12 without flutes (B-)(Fig 3b)24 implants with turned surfaces

- 12 with two vertical flutes (original geometry) (T+) - 12 without flutes (T-)

24 other implants used for another investigation.

each of T+, T-, B+ and B-implants were placed in the femora of each animal. Two B+ and two B- im-plants were placed in the tibiae of each animal. Cover screws were placed and the flaps were closed with resorbable sutures. The animals were allowed to heal for three (n=4) and six (n= 8) weeks after surgery.

Removal torque After six weeks of healing, the femoral implants in eight animals were subjected to removal torque (RTQ) test. The tests were performed in a specially designed rig using a motor-driven device. A linearly increasing torque was applied until failure of integra-tion occurred; the peak value in Newton-centimeters (Ncm) was recorded. A mean value was calculated

Animals, anaesthesia and experimental protocolA total of 12 female New Zealand white rabbits were used in the study, after the protocol had been approved by the animal ethics committee of the Gothenburg University.. General anesthesia was induced by intra-muscular injections of fluanisone-fentanyl (0.7 mL, Hypnorm™, Helsingborg, Sweden) and intraperitoneal injection of diazepam (0.25 mg/kg, Apozepam™, Al-pharma AB, Stockholm, Sweden). Local anesthesia was induced at both proximal tibial metaphyses and distal femoral condyles by injections of lidocaine (about 2 mL, Xylocaine®,Astra Zeneca AB, Södertälje, Sweden).

The distal femoral condyles and proximal tibial metaphyses were used as experimental sites. The bone was exposed via incisions through skin and fascia. Two implant sites were prepared in each femur and tibia; each animal receiving 8 implants. Implants were selected by a rotational scheme, so that one

FIgure 2. Three-dimensional view of the bimodal surface topography at a thread flank.

FIgure 1. The four types of implants used in the study; (L to R) bimodal surfaced fluted and non-fluted and two with turned surfaces.


20

for each of the four implant types. The percentage difference from that of turned implants without vertical flutes was calculated for each of the other groups. The Wilcoxon Signed Rank test was used for statistics and a difference was considered if p < 0.05.

HistologyAll implants and surrounding bone tissues were retrieved and fixed by immersion in a 4% buffered formaldehyde solution. The specimens were then de-hydrated in a graded series of ethanol and embedded in light curing methacrylate (Technovit® 7200 VCL, Kulzer, Friedrichsdorf, Germany). Ground sections, approximately 10 μm thick, were prepared using a sawing and grinding technique (Exakt Apparatebau®, Norderstedt, Germany). One central section was taken from each implant site, stained with Toluidine Blue and examined in a Nikon light microscope.

RESULTS

Light microscopy of the three-week specimens of the bimodal implant surface revealed bone forma-tion directly onto the implant surface (Fig. 4a). This could be seen as thin rims of bone following the contour of the implant threads (Figs. 4a and b) and as solitary islets with no obvious connection to exist-ing bone surfaces (Fig. 5). Osteoblastic seams were often seen on the bone rims, facing the adjacent soft tissues (Figs. 4b and 6). Non-bone areas consisted of a loose connective tissue rich in cells and vessels and devoid of signs of inflammation (Figs. 4a-b, 5 and 6). A more mature bone-implant interface was seen in the six-week specimens and a larger proportion of the implant surface was in contact with bone (Fig. 7).

The removal torques were found to be correlated with surface topography and the absence or pres-ence of vertical flutes (Fig. 8); the lowest torques were recorded for machined implants without flutes and the highest for bimodal surface im-plants with vertical flutes (p<0.05)(TABLE 1).

DISCUSSION

The present study indicated that the bimodal topog-raphy induced bone formation directly at the implant surface as previously described for other commercially available surface modified implants (Piattelli et al. 1996, Ivanoff et al. 2001, Rocci et al. 2003, Berglundh

FIgure 4. Light micrographs showing Neoss implants after three weeks of healing. a/ Overview showing thin rims of bone following the contour of the threads (Arrows), I = implant, LCT = loose connective tissue. b/ Close up of a. Bone (B) has been formed directly on the implant (I) surface. Osteoblasts (Os) and osteoid can be observed on the surface of the bone facing a loose connective tissue (LCT).


21

et al. 2003). This was seen as seen thin rims or solitary islets of bone at the surface with osteoblastic seams fac-ing the adjacent bone marrow. Previous descriptions of the healing of smooth, turned implants have reported bone formation from the adjacent tissues and towards the implant surface (Sennerby et al 1992, Palma et al. 2006, Miranda-Burgos et al. 2007). The mechanisms behind the different integration pathways are probably related to the integrity of the blood clot-implant in-terface during the early events of bone healing (Davies 2003). With a rough surface, the clot can maintain a firm contact in spite of shrinkage, whereas a gap may be formed at a smooth implant surface. In the former case mesenchymal cells can migrate to the implant surface, differentiate and start to produce bone matrix. The influence of surface modification on the clinical outcome is not clear. For instance, clinical studies com-paring titanium-plasma sprayed and turned surfaces or TiO2 blasted and turned surfaces could not find any statistically significant differences with regard to sur-vival rate and marginal bone loss (Åstrand et al. 2004a, 2004 b). However, other non-comparative studies have indicated better survival rates for surface modi-fied than for turned implants in challenging clinical situations such as bone grafting (Brechter et al. 2005) and in immediate loading (Glauser et al 2001, 2003). Since the trend today is to use immediate/early load-ing, the use of surface modified implants is preferable.

Figure 5. Light micrograph after three weeks showing solitary bone formation (B) at the apical part of the implant (I) facing a loose connective tissue (LCT) rich of cells.

Figure 6. Light micrograph showing bone formation at the bottom of a thread after three weeks of healing. New bone (NB) is formed by osteoblasts (Os) on previously formed bone (PB) and separated by a cement line (white arrow). LCT = loose connective tissue

FIgure 7. Light micrographs after 6 weeks of healing showing almost complete bone filling of a thread with mature bone. Arrows point at newly formed secondary osteons indicative of remodelling.

Turned

(T-)

Turned with flute

(T+

Bimodal

(B-)

Bimodal with flute

(B+)42.4

(15.4)44.4

(15.0)46.9

(13.2)58.5

(13.2)*

Table 1. Results from removal toruqe measurements.

* P<0.05 compared with T- implants

The removal torque tests revealed an improved resist-ance to torque with modified topography and a verti-cal flute added. This is best explained by ingrowth of bone into micro- and macroscopic undercuts at the surface. Hall et al (2005) showed that a macroscopic groove added to the thread flank can stimulate bone formation over the implant surface. The relatively


22

small effect of surface topography may be explained by the design of the test implants, since they all had apical undercuts, i.e. cutting edges and bone chambers.

CONCLUSION

The present experimental study showed evidence of surface mediated bone formation at the bimodal surface as previously described for other commer-cially available surface modified implants. Removal torque tests showed increased stability with the modified surface and adding vertical flutes when compared to turned control implants without flutes.

REFERENCESAlbrektsson T, Brånemark PI, Hansson H, Lindström J. Osseointegrated implants. Requirements for ensuring a long-lasting direct bone-to-im-plant anchorage in man. Acta Orthop Scand 1981;52:155-170

Albrektsson T & Sennerby L.State of the art in oral implants. J Clin Peri-odontol. 1991;18:474-81.

Albrektsson T & Wennerberg A. Oral implant surfaces: Part 1--review focusing on topographic and chemical properties of different surfaces and in vivo responses to them. Int J Prosthodont. 2004;17:536-43.

Åstrand P, Engquist B, Anzen B, Bergendal T, Hallman M, Karlsson U, Kvint S, Lysell L, Rundcranz T. A three-year follow-up report of a com-parative study of ITI Dental Implants and Branemark System implants in the treatment of the partially edentulous maxilla.Clin Implant Dent Relat Res. 2004a;:130-41.

Åstrand P, Engquist B, Dahlgren S, Grondahl K, Engquist E, Feldmann H. Astra Tech and Branemark system implants: a 5-year prospective study of marginal bone reactions.Clin Oral Implants Res. 2004b Aug;15:413-20.

Berglundh T, Abrahamsson I, Lang NP, Lindhe J. De novo alveolar bone formation adjacent to endosseous implants. A model study in the dog. Clin Oral Implants Res 2003;14:251-262


Brechter M, Nilson H, Lundgren S. Oxidized titanium implants in reconstructive jaw surgery. Clin Implant Dent Relat Res. 2005;7 Suppl 1:S83-7.

Davies JE. Understanding peri-implant endosseous healing. J Dent Educ 2003;67:932-949

Friberg B, Jemt T, Lekholm U. Early failures in 4,641 consecutively placed Branemark dental implants: a study from stage 1 surgery to the connection of completed prostheses. Int J Oral Maxillofac Implants. 1991 Summer; 6(2):142-6.

Friberg B. (1999) On bone quality and implant stability measurements. Thesis. Dept of Biomaterials/Handicap Research, Göteborg University, Sweden.

Glauser R, Ree A, Lundgren A, Gottlow J, Hammerle CH, Scharer P. Immediate occlusal loading of Branemark implants applied in various jawbone regions: a prospective, 1-year clinical study. Clin Implant Dent Relat Res. 2001;3(4):204-13.

Glauser R, Lundgren AK, Gottlow J, Sennerby L, Portmann M, Ruhstaller P, Hammerle CH. Immediate occlusal loading of Branemark TiUnite implants placed predominantly in soft bone: 1-year results of a prospective clinical study. Clin Implant Dent Relat Res. 2003;5 Suppl 1:47-56.

Ivanoff CJ, Hallgren C, Widmark G, Sennerby L, Wennerberg A. His-tologic evaluation of the bone integration of TiO(2) blasted and turned titanium microimplants in humans. Clin Oral Implants Res. 2001; 12:128-34.

Johansson C, Albrektsson T. Integration of screw implants in the rabbit: a 1-year follow-up of removal torque of titanium implants. Int J Oral Maxillofac Implants. 1987;2:69-75.

Miranda-Burgos P, Rasmusson L, Meirelles L, Sennerby L. Early bone tissue responses to oxidized and machined titanium implants in the rab-bit tibia. Clin Implant Dent Relat Res, Accepted for publication

Palma VC, Magro-Filho O, de Oliveria JA, Lundgren S, Salata LA, Sen-nerby L. Bone reformation and implant integration following maxillary sinus membrane elevation: an experimental study in primates. Clin Implant Dent Relat Res. 2006;8:11-24.

Piattelli A, Scarano A, PIattelli M, Calabrese L. Direct bone formation on sand-blasted titanium implants: an experimental study. Biomaterials 1996;17:1015-8


Sennerby, L., Thomsen, P. and Ericson, L.E. Early bone tissue response to titanium implants inserted in rabbit cortical bone. I. Light micro-scopic observations. J Mat Sci: Mat Med 1993;4:240-250

Zechner W, Tangl S, Furst G, Tepper G, Thams U, Mailath G, Watzek G. Osseous healing characteristics of three different implant types. Clin Oral Implants Res. 2003a Apr;14(2):150-7.

Figure 8. Graph showing the percentage change of removal torque in comparison with turned implants without vertical flutes (T-). * p< 0.05


23

Peter Andersson1 , Damiano Verrocchi1, Rauno Viinamäki1, Lars Sennerby1,2

1Private Practice, Fiera di Primiero and Feltre, Italy2Dept Biomaterials, Inst Clinical Sciences, Sahlgrenska Academy, Gothenburg University, Sweden

This study reports a survival rate of 98.1% for 102 Neoss implants in 44 patients with a mean bone loss of 0.7 mm after one year. The reduced component inventory and innovative designs enhancing primary stability and facilitating laboratory technology/prosthetics offer practical advantages whilst the clinical results with the implant system tested compare favourably with existing systems

INTRODUCTION

The use of implant-supported bridges is a routine treatment modality for the edentulous patient with documented good long-term results (Albrektsson & Sennerby 1991, Esposito et al. 1998). From initially a rather complicated and restricted procedure, the tech-niques of implant-supported dentistry have improved and simplified, at least from a surgical point of view. The introduction of self-tapping and surface modified implants, surgical guides, shortened healing periods and immediate loading are some examples. However, improvements could still be made in dental laboratory and prosthetic techniques in order to reduce further treatment times and the overall cost of the treatment.

The aim of the present study was to evaluate a new implant system (Neoss) for one year using clinical and radiographic examinations and implant stability measurements by resonance frequency analysis (RFA).


Patient inclusionConsecutive patients requiring implant treatment for total or partial loss of teeth were included in the study until at least 100 implants had been inserted. Only two-stage procedures with 3 months of healing from placement to abutment connection were performed.

Preoperative examinations included intraoral and panoramic radiographs. Computerized tomography was used if required. The implant sites should have sufficient bone for at least 7 mm long implants. Pa-

tients should be over 18 years old and present no con-traindications for oral surgery under local anaesthesia.

Clinical techniquesPatients were administered 2 gr of amoxicillin (Aug-mentin

®, Roche, Milan, Italy) and sedation if required

(Valium®, Roche, Milan, Italy, 5 mg) prior to surgery. Anaesthesia was induced by infiltration with articain/epinephrine (Septocain™, Specialites Septodont, Saint-Maur-Des-Fosses, France). Crestal incisions were used for flap elevation. Implant sites were prepared with a 2.2 mm twist drill followed by a 3 mm (3.5 mm wide implant), 3.4 mm (4.0 mm wide implant) and 3.9 mm drills (4.5 mm wide implant) (Neoss ltd, Harrogate, UK). Full countersink preparation was made for all implants. Implants were inserted with the torque driver set to 40 Ncm and final seating was performed with a hand wrench. Cover screws were applied and the flaps were replaced and sutured. Bone quality and quantity according to Lekholm and Zarb (1985) were registered.

Abutment connection was performed 3 months later, usually with a punch technique and no sutures. Bite registration and impressions were taken with closed tray and then healing abutments were con-nected. Neolink™ abutments (Neoss ltd, Harrogate, UK) of gold or titanium (to match the metal of the framework) were used. Abutments were either cast or welded into the metal frameworks. Porcelain and acrylic veneers were used. Constructions were screw-retained if implant angulation permitted; otherwise, individual abutments (Neo Matrix, Ne-oss ltd, Harrogate, UK) were used for cementation.

A One-Year Clinical, Radiographic and RFA Study of Neoss Implants Used in Two-Stage Procedures


24

Follow-upImplant stability was assessed by RFA (Mentor®, Integration Diagnostics AB, Gothenburg, Sweden) at implant surgery, abutment connection and after one year of loading (after unscrewing the construc-tions ). Cemented constructions were not removed.

Digital or conventional intraoral radiographs were taken at abutment connection and after one-year of service. Conventional radiographs were pho-tographed with a digital camera on a light desk. Measurements were made using a personal computer (Image J 1.34S, National Institutes of Health, USA) at mesial and distal aspects. Each radiograph was calibrated using the known width of the coronal cyl-inders of the implants. The upper platform was used as a reference point for measurements (Figure 1) .

RESULTS

Forty-four (44) patients (16 male and 28 female, mean age 54 years) were treated according to the protocol. Forty-eight (48) prosthetic construc-tions were delivered including five full cross-arch bridges, 28 partial bridges and 17 single crowns. A total of 102 implants were placed, 34 in the max-illa and 68 in the mandible, in bone quality and quantity as presented in Table 1. Implant lengths and diameters are presented in Tables 2 and 3.

At writing all patients have completed one year in function. All constructions except one pa-t ient with a cemented bridge on f ive im-plants were removed for stability checking.

All patients maintained a fixed bridge or crown during the one year study. Two failures were experienced, giv-ing a survival rate of 98.1% after one year. One 13 mm long by 4.0 mm wide implant placed in a defect in a mandible was removed at impression due to mobility. In this case, a new implant was placed and a temporary bridge was made on the remaining three implants whilst awaiting healing. One 7 mm long by 4.5 mm wide im-plant in Q4 bone in the1st molar region in a maxilla was lost after one year. The implant was removed and the three-unit bridge was supported on the remain-ing two implants. No other prosthetic complications such as fractures or screw-loosening were experienced.

A total of 92 implants with readable radiographs at

Figure 1. Clinical photographs showing a/ two implants in the posterior mandible after insertion, b/ healing of soft tissues at the time of bridge delivery, c/ final bridge.

Quantity Number Quality Number

A 3 1 3

B 47 2 68

C 52 3 26

D - 4 5

Table 1. Bone Quality and Quanitity encountered

Diameter Number

3.5 mm. 17

4.0 mm. 57

4.5 mm. 28

Diameter Number

7 9

9 3011 2313 1915 21

Table 2. Implant Sizes placed Table 3. Implant lengths placed


25

baseline and follow-up have been evaluated to date. On average, the marginal bone level was located 0.6 mm (SD 0.7) below the platform at baseline and 1.3 mm (SD 1.0) below after one year; giving an average bone loss of 0.7 mm (SD 1.0) at one year. X implants showed more than 3 mm bone loss.

Stability measures by RFA gave a mean of 75.0 (SD 6.5) ISQ at placement, 74.4 (SD 6.8) at abutment connection and 76.2 (SD 6.6) after one-year. There seems to be a relation between bone quality and ISQ at implant placement (Fig. 4). Implant stability increased with time for implants in Q4 bone whilst stability decreased for implants in very dense bone, resulting in a similar stability for all implants after one year.

DISCUSSION

Early experience with the Neoss implant system used in a two-stage procedure with 3 months of healing in the present study shows satisfactory results. Two of 102 implants were lost during one year which compares well with the results reported from other implant systems (Albrektsson & Wennerberg 2004). From a surgical point of view, the implant was stable during insertion and good primary stability was generally achieved. When poor primary stability was encoun-tered, a wider implant was used to enhance stability. Since the different implant diameters have the same prosthetic platform, an implant could be replaced with a wider diameter without requiring changes to the other prosthetic components, as is commonly necessary with other implant systems. Abutment connection was easier with this implant than previ-ously experienced with external connection implants, which often require a flap procedure and removal of bone tissue in order to fit an abutment. A punch technique without suturing could frequently be used in the present study and in most cases impressions were taken in conjunction with abutment connection. The use of integrated abutments with the framework facilitated prosthetics and reduced the overall costs .

The RFA measurements revealed a high average pri-mary stability of these implants which is most probably due to the geometric features of the implant. Firstly, it has a two-start thread, a twin helix, with two threads running parallel but on diametrically opposite sides of the cylindrical substrate. Thus, the actual thread pitch is double the “apparent” thread spacing, so the implant

advances twice as far with one turn as for a single-helix thread. This feature reduces the insertion time and counteracts wobbling during insertion. Secondly, the implant has a positive tolerance, meaning that the implant is slightly conical in the coronal direction and as the implant screws home its diameter increases slightly and there is a slight tendency to tighten the thread base against the bone. Previous research has shown higher stability for slightly tapered implants in soft bone than for parallel-sided implants, without jeopardizing the integration process (O’Sullivan et al 2000, O´Sullivan et al 2004a, 2004b). Decreased primary stability was seen with decreased bone den-sity, which is in line with the findings of Östman et al (2006). Implant stability increased slightly with time and was similar for all bone qualities after one year of function, which indicates a favourable bone tissue response to the implants. The greatest increase of stabil-ity was seen for implants in Q4 bone which is in line with the findings of Friberg et al. (1999).. This can be explained by stiffening of the bone-implant interface with time due to bone formation and remodelling.

Figure 2. Radiographs from a single tooth case at a/ baseline and b/ after one year


26

In contrast, decreased stability was observed in dense bone, which may be explained by mechanical relaxa-tion and/or bone remodelling as a response to the pre-sumably high stresses induced by implant placement.

The radiographic analysis revealed that about 0.7 mm of bone was lost during the first year of loading, which is in the range of previously reported results with other systems (Oh et al. 2002). X implants showed more than 3 mm bone loss during one year. The fact that the bone level was located 0.6 mm be-low the platform at abutment connection indicates some remodelling during healing as generally the implants were placed flush with the surrounding bone. This accords with the findings of Engquist et al. (2001) who compared the marginal bone tissue response at Astra Tech and Brånemark system implants.

It is concluded that prosthetic rehabilitation of the edentulous patient with the Neoss implant system results in good short-term clinical and radiographic outcomes. The innovative design solutions do seem to enhance primary stability and facilitate labora-tory technology/prosthetics, and with the reduced component inventory this system offers advantages over our experiences with other implant systems.REFERENCESAlbrektsson T & Sennerby L.State of the art in oral implants. J Clin Peri-odontol. 1991;18:474-81.

Albrektsson T, Wennerberg A.Oral implant surfaces: Part 2--review focusing on clinical knowledge of different surfaces. Int J Prosthodont. 2004;17:544-64.

Engquist B, Astrand P, Dahlgren S, Engquist E, Feldmann H, Grondahl K. Marginal bone reaction to oral implants: a prospective comparative study of Astra Tech and Branemark System implants. Clin Oral Implants Res. 2002;13:30-7.

Esposito M, Hirsch JM, Lekholm U, Thomsen P. Biological factors con-tributing to failures of osseointegrated oral implants. (I). Success criteria and epidemiology. Eur J Oral Sci. 1998;106):527-51.

Friberg B, Sennerby L, Meredith N, Lekholm U. A comparison between cutting torque and resonance frequency measurements of maxillary implants. A 20-month clinical study.Int J Oral Maxillofac Surg. 1999;28:297-303.

Lekholm U, Zarb GA. Patient selection. In: Brånemark P.-I, Zarb GA, Albrektsson T, eds. Tissue integrated prostheses. Osseointegration in clinical dentistry. Chicago: Quintessence, 1985:199–209.

O’Sullivan D, Sennerby L, Meredith N. Measurements comparing the initial stability of five designs of dental implants: a human cadaver study.Clin Implant Dent Relat Res. 2000;2(2):85-92.

O’Sullivan D, Sennerby L, Jagger D, Meredith N.A comparison of two methods of enhancing implant primary stability. Clin Implant Dent

Figure 3. Radiographs from a partial edentulous mandible at a/ baseline and b/ after one year

Figure 4. Results from RFA measurements in relation to bone quality and time.

Relat Res. 2004a;6:48-57.


Ostman PO, Hellman M, Wendelhag I, Sennerby L. Resonance fre-quency analysis measurements of implants at placement surgery.Int J Prosthodont. 2006;19:77-83;

50

55

60

65

70

75

80

85

90

95

100

Placement Abutment One year

Imp

lan

t s

tab

ilit

y (

ISQ

)

_

Q 1Q2Q3Q4


27

Peter Andersson1 , Damiano Verrocchi1, Luca Pagliani2, Lars Sennerby3

1Private Practice, Fiera di Primiero and Feltre, Italy2Private Practice, Milano and Legnano, Italy3Dept Biomaterials, Inst Clinical Sciences, Sahlgrenska Academy, Gothenburg University, Sweden

This interim report of an ongoing study on immediate/early loading of Neoss implants reports a survival rate of 96.5 % for 141 Neoss implants in 33 patients after 6 months to three years of loading. All patients received and maintained a fixed bridge in spite of five failures in three patients

Immediate/Early loading of Neoss Implants. Preliminary Results from an Ongoing Study

INTRODUCTION

The use of immediate/ early loading protocols for implant-supported crowns and bridges has obvious advantages for the patients since only one surgical procedure and no healing periods are needed. Both function and aesthetics can be immediately restored with a temporary crown or bridge. Concerns have been raised about increased failure rates, since the original concept of osseointegration prescribed a submerged and unloaded healing period of 3 to 6 months before loading (Brånemark et al. 1969, Albrektsson et al. 1981). Today, histology from experimental and clinical studies has demonstrated that implants can integrate under the influence of functional loads (Piattelli et al. 1998, Rocci et al. 2003). Moreover, clinical follow-up studies have reported similar good clinical outcomes as for two-stage procedures (Attard & Zarb 2003). In fact, immediate/early loading is now a clinical reality and a commonly used procedure. In many studies, authors have identified primary implant stability as a critical factor for success (Östman et al 2006). Inclusion criteria based on insertion torque values and RFA measurements have been used (Calandriello et al, 2003, Rocci et al 2003, Vanden Bogaerde et al 2004, Östman et al 2005). It has been shown that the use of self-tapping implants, reduced final drill diameters and tapered implants may improve pri-mary stability (O’Sullivan et al 2001, 2004).

The aim of the present study was to report on the early experiences of an immediate/early loading protocol us-ing a new implant designed to give firm primary stability.


Patient groupThe study group consisted of 33 patients (18 female and 15 male) from three clinics, representing consecu-tive treatments with immediately/early loaded implant-supported provisional bridges or crowns. The patients were totally edentulous (n= 18), partially edentulous (n=12) or treated for single tooth loss (n=3) (Table 1). The possibility of placing long implants with good primary stability was assessed in each patient on the basis of radiography and clinical examinations. A final decision was taken after discussions with the patient.

Implant surgery was performed under local anaes-thesia and a total of 141 implants were inserted, 79 in the maxilla and 62 in the mandible (Neoss Ltd,

Quantity Number of

sites

Quality Number of

sitesA 10 1 3

B 104 2 81

C 26 3 48

D 1 4 9

E -

Table 1. Bone quantity and quality of implant sites according to Lekholm and Zarb (1985).


28

Harrogate, UK) using an insertion torque of at least 40 Ncm. Primary stability was ensured by reducing the final drill diameter in soft bone qualities. The majority of implants were 4 mm in diameter and 13-15 mm long (Table 2 and 3). Most implants had been placed in bone quantity b and quality 2 ac-cording the Lekholm & Zarb index (Table 4 and 5).

Sterile impression copings were attached directly to the implants (n=75) or to straight or angulated abutments (n=66) (Southern Implants™, Protera AB, Gothenburg, Sweden). Impressions and bite registration were taken after suturing. Healing abutments were then connected to the implants. A screw-retained laboratory-made crown (n=3) or metal-reinforced bridge (n=31) was delivered one to three days after surgery. In two patients treated in the posterior mandible, a provisional bridge was made in the mouth the same day. Altogether, 34 provisional prosthetic constructions were delivered to the 33 patients. Single and partial constructions were not in occlusion. Total cross-arch bridges were carefully adjusted with regard to occlusion, striving to-wards group function and contacts over implant sites.

After a period of 3 to 6 months, the provisional construc-tions were replaced with permanent crowns or bridges made of gold or titanium with acrylic or porcelain veneers.

Follow-upThe patients were monitored with clinical exami-nations controlling the occlusion during the first weeks and months. The protocol includes RFA measurements (Mentor®, Integration Diagnostics AB, Gothenburg, Sweden) at implant surgery, on delivery of the final prosthesis and after one year of loading.

Diameter Number

3.5 18

4.0 111

4.5 12

Total 141

Diameter Number

7 2

9 7

11 13

13 33

15 83

17 3

Total 141

Table 3. Implant lengths placed in this study

Table 2. Implant diameters placed in this study

Digital or conventional intraoral radiographs were taken at baseline and after one-year of service.

RESULTS AND DISCUSSION

All 33 patients have passed 6 months of loading and 13 have exceeded one year. A total of five failures were experienced, giving a survival rate of 96.5% after 6

Figure 1. Immediate loading in a mandible after extraction of three teeth.a. Six implants have been placedb. Provisional bridge from the laboratory.c. Showing the connected bridge


29

months to 3 years. All patients received and main-tained a fixed prosthesis during the follow-up period. The failures all occurred in the maxilla within three months of surgery (Table 4). Implant stability was 73.5 (SD 7.0) ISQ at placement and 74.4 (SD9.2) ISQ at final prosthesis delivery after 4 months on average.

DISCUSSION

The use of immediate/early loading protocols conveys obvious advantages for the patient since only one surgery and no healing periods are needed. Moreo-ver, there is no need for removable appliances; which can be difficult and uncomfortable to wear. In the present study, a fixed provisional bridge or crown was fabricated chair-side or delivered within a few days. In this interim report of an ongoing study, five of 141 implants failed after a follow-up of at least 6 months. All patients received and maintained a fixed construction throughout the study period in spite of five failed implants. The 3.5 % failure rate of this study is similar to those reported by other authors in studies on immediate loading. For instance, Glauser et al lost about 3 % of implants when used on all indications as in the present study. All failures in the present study occurred in the maxilla; a failure rate of 6.3 % which corroborates the findings of Olsson et al (2003). Consequently, the results were better in man-dibular bone as no implant failures were experienced; which is also in line with the findings in the literature (Attard & Zarb 2005). It can be speculated that dif-ferences in bone density may explain this finding as implants get better primary stability in mandibular bone (Östman et al 2006). However, all but one im-plant showed high stability values and other factors such as unfavourable loading need to be considered.

The implants used in the present study have a surface topography modified by blasting. It is known from animal experiments that such surfaces seem to integrate

faster and with more bone contacts than implants with a smooth surface. It is also possible that this contrib-uted to the good outcome of the present study. The RFA measurements revealed firm primary stability

This interim report of an ongoing study on im-mediate/early loading of Neoss implants reports a survival rate of 96.5 % for 141 Neoss implants in 33 patients after 6 months to three years of load-ing. All patients received and maintained a fixed bridge in spite of five failures in three patients.

Patient Position Case Diameter/Length

Quantity/Quality

Primary stability (ISQ) Time of failure

1 16 Partial 4/15 C/3 44 3 months

2 13 Total 4/13 B/3 75 3 months

3 14 Total 4/13 B/2 75 3 months

3 11 Total 4/13 B/2 66 3 months

3 26 Total 4/15 B/3 77 3 months

Figure 2. Immediate loading in a maxilla.a. Six implants have been installed in an arch-formb. The laboratory-made provisional bridge has been connected to the implants

Table 4. Characteristics of implant failures.


30

REFERENCESAlbrektsson T, Brånemark PI, Hansson H, Lindström J. Osseointegrated implants. Requirements for ensuring a long-lasting direct bone-to-im-plant anchorage in man. Acta Orthop Scand 1981;52:155-170

Attard NJ, Zarb GA.Immediate and early implant loading protocols: a literature review of clinicalstudies. J Prosthet Dent. 2005 Sep;94(3):242-58.


Calandriello R, Tomatis M, Rangert B. Immediate functional loading of Branemark System implants with enhanced initial stability: a prospective 1- to 2-year clinical and radiographic study. Clin Implant Dent Relat Res. 2003;5 Suppl 1:10-20.

Glauser R, Lundgren AK, Gottlow J, Sennerby L, Portmann M, Ruhstaller P, Hammerle CH. Immediate occlusal loading of Branemark TiUnite implants placed predominantly in soft bone: 1-year results of a prospective clinical study. Clin Implant Dent Relat Res. 2003;5 Suppl 1:47-56.


Olsson M, Urde G, Andersen JB, Sennerby L.Early loading of maxil-lary fixed cross-arch dental prostheses supported by six or eight oxidized titanium implants: results after 1 year of loading, case series.Clin Implant Dent Relat Res. 2003;5 Suppl 1:81-7.

Ostman PO, Hellman M, Sennerby L. Direct implant loading in the edentulous maxilla using a bone density-adapted surgical protocol and primary implant stability criteria for inclusion.Clin Implant Dent Relat Res. 2005;7 Suppl 1:S60-9.

Ostman PO, Hellman M, Wendelhag I, Sennerby L. Resonance fre-quency analysis measurements of implants at placement surgery.Int J Prosthodont. 2006;19:77-83;O’Sullivan D, Sennerby L, Meredith N. Measurements comparing the initial stability of five designs of dental implants: a human cadaver study.Clin Implant Dent Relat Res. 2000;2(2):85-92.

O’Sullivan D, Sennerby L, Jagger D, Meredith N.A comparison of two methods of enhancing implant primary stability. Clin Implant Dent Relat Res. 2004a;6:48-57.


Piattelli A, Corigliano M, Scarano A, Costigliola G, Paolantonio M. Immediate loading of titanium plasma-sprayed implants: an histologic analysis in monkeys. J Periodontol. 1998 Mar;69(3):321-7.


Vanden Bogaerde L, Pedretti G, Dellacasa P, Mozzati M, Rangert B, Wendelhag I. Early function of splinted implants in maxillas and poste-rior mandibles, using Branemark System Tiunite implants: an 18-month prospective clinical multicenter study. Clin Implant Dent Relat Res. 2004;6(3):121-9.

Figure 3. Immediate loading in a posterior mandible using a chair-side made bridgea. Three implants have been placedb. Temorary PEK abutments connected to the implants c. The pre-fabricated bridge is tried on the abutments and adjusted. d. The bridge after setting of the resin.e. The finalized provisional bridgef. The bridge in position.


31

Thomas Zumstein1 and Camilla Billström

2

1 Dr. med. dent., Private practice, Luzern, Switzerland 2 Neoss AB, Mölnlycke, Sweden

This retrospective study of 50 patients and 183 Neoss implants showed a survival rate of 98.2% for routine cases and 94.4 % for GBR cases after a follow-up of 1 to 2 years. The marginal bone loss was 0.3 mm the first and 0.1 mm the second year. GBR procedures involving short implants and soft bone seemed to increase the risk of implant failure.

INTRODUCTION

The use of osseointegrated implants has been proven to result in good clinical outcomes in the prosthetic rehabilitation of the edentulous patient, whether used in totally or partially edentulous jaws, including single tooth replacements (Esposito et al. 1998). Since the introduction some 40 years ago, the osseointegration technique has been continuously developed and refined in order to simplify and shorten implant treatment. For instance, the introduction of self-tapping implants, surface modifications and immediate/early loading protocols has markedly facilitated implant treatment. The presence of sufficient bone volumes was originally an absolute criterion for using implants. Today, num-bers of reconstructive techniques including the use of bone grafts, bone substitutes, osteodistraction devices and membranes are available to increase the load-bear-ing volume of the jaw bone. A localized defect such as that due to incomplete healing after extraction may complicate the placement of an implant by exposure of parts of the implant. In such cases, the use of a bovine bone substitute and a resorbable membrane is commonly used to achieve complete bone cover-age (Hurzeler et al. 1998, Hammerle & Lang 2001).

The Neoss implant system (Neoss Ltd, Harrogate, UK) is a novel design which according to the manu-facturer was developed to provide simple and ef-fective solutions for all kinds of cases with minimal components. Special attention has been paid to the technical/prosthetic phase and only one prosthetic platform is used irrespective of implant diameter.

The aim of this retrospective study was to report on the experiences with Neoss implants from the first consecutive 50 patients treated in one private office.


Patients and surgeryFifty consecutive patients (20 males, 30 females, mean age 57 years) needing implant treatment were enrolled in the study. Three patients were treated in both maxilla and mandible resulting in 53 treatment areas (jaws) included in the study. Intraoral and panoramic radio-graphs as well as CT scans were used for presurgical evaluations. Nine patients were totally edentulous (two in both jaws), 21 were partially edentulous (one in two areas) and 20 patients were treated for single tooth loss.

The patients were administered antibiotics prior to surgery (Dalacin® C 300mg, Pfizer). Surgery was per-formed under sterile conditions and local anaesthesia with Ultracain® DS Forte. Crestal incisions were used and implant sites were drilled in accordance with the guidelines given by the manufacturer for the appro-priate implant diameter. Implants were inserted into position with a drilling unit. A total of 183 Neoss implants (Neoss Ltd, Harrogate, UK) were placed; 116 in the maxilla and 67 in the mandible (Table 1). Implant lengths and diameters are shown in Table 2 and 3. Bone quality and quantity according to the Lekholm and Zarb index (1985) were registered (Ta-bles 4 and 5). The implant collar was either fully sub-merged in bone (n=1087) or to half its length (n=75).

Thirty of the treatment areas with 126 implants underwent GBR using BioOss™ and a resorb-able BioGide™ membrane (Geistlich, Switzer-land) simultaneously with implant placement.

Healing abutments were connected after a healing period of 3 to 6 months in 32 treatment areas (89 implants). In 21 areas (94 implants) healing abut-

A Retrospective Follow-up of 50 consecutive patients treated with Neoss Implants with or without an Adjunctive GBR-Procedure


32

ments were placed in conjunction with implant surgery and 8 of these (57 implants) were loaded with a crown/bridge for immediate function.

Prosthetics Impressions were taken on implant level for screw-retained prosthetics using Neolink abutments (Neoss Ltd, Harrogate, UK) and gold-ceramic or gold-acrylic frameworks. All crowns and partial bridges were gold-ceramic reconstructions and all but one of the full jaw bridges were gold-acrylic reconstructions. The crowns/bridges were attached with gold screws using a preload of 32 Ncm. Follow-upThe patients were scheduled for annual check-ups with clinical and radiographic examinations using intraoral or panoramic radiographs. Marginal bone level meas-urements were performed by an independent radiolo-gist in one baseline and one follow-up radiograph.

RESULTS

Forty-nine of the patients attended the first annual check up and 26 the second. A total of eight implant failures were registered, all during the first year in service, giving an overall Cumulative Success Rate (CSR) of 95.6% (Table 6). All but one failure oc-curred in the GBR group, giving a CSR of 94.4% for the GBR-group and 98.2 % for the non-GBR group after 1 year (Table 6). The characteristics of failed implants are shown in Table 7. Failures occurred

MAXILLA 28 27 26 25 24 23 22 21 11 12 13 14 15 16 17 18 TOTAL

0 1 11 16 17 9 6 4 4 7 9 12 12 7 1 0 116

0 3 9 5 3 4 5 1 2 6 5 5 6 11 2 0 67

MANDIBLE 38 37 36 35 34 33 32 31 41 42 43 44 45 46 47 48 TOTAL

Table 1. Distribution of implants in relation to position

IMP

LAN

T LE

NG

TH

PLA

CE

D

IMP

LAN

TS

FAIL

ED

IM

PLA

NTS

IN

TOTA

L

FAIL

ED

IM

PLA

NTS

IN

G

BR

PAT

IEN

TS

FAIL

ED

IM

PLA

NTS

IN

NO

N-G

BR

7 mm 7 2 2 0

9 mm 51 4 3 1

11 mm 75 2 2 0

13 mm 44 0 0 0

15 mm 6 0 0 0

17 mm 0 0 0 0

Total 183 8 7 1

Table 2. Placed and failed implants in relation to length

IMP

LAN

T D

IAM

ETE

R

PLA

CE

D

IMP

LAN

TS

FAIL

ED

IM

PLA

NTS

IN

TOTA

L

FAIL

ED

IM

PLA

NTS

IN

G

BR

PAT

IEN

TS

FAIL

ED

IM

PLA

NTS

IN

NO

N-G

BR

PA

TIE

NTS

3,5 mm 58 5 4 1

4,0 mm 112 2 2 0

4,5 mm 13 1 1 0

5,5 mm 0 0 0 0

Total 183 8 7 1

Table 3. Placed and failed implants in relation to diameter

BO

NE

Q

UA

LITY

PLA

CE

D

IMP

LAN

TS

FAIL

ED

IM

PLA

NTS

IN

TO

TAL

FAIL

ED

IM

PLA

NTS

IN

GB

R

PATI

EN

TS

FAIL

ED

IM

PLA

NTS

IN

NO

N-

GB

R

1 0 0 0 0

2 47 2 2 0

3 115 4 4 0

4 21 2 1 1

Total 183 8 7 1

Table 4. Placed and failed implants in relation to bone quality

BO

NE

Q

UAN

TITY

PLA

CE

D

IMP

LAN

TS

FAIL

ED

IM

PLA

NTS

IN

TO

TAL

FAIL

ED

IM

PLA

NTS

IN

GB

R

PATI

EN

TS

FAIL

ED

IM

PLA

NTS

IN

NO

N-

GB

R

A 44 0 0 0

B 103 4 4 0

C 26 4 3 1

D 10 0 0 0

E 0 0 0 0

Total 183 8 7 1

Table 5. Placed and failed implants in relation to bone quantity


33

Table 6. Life-tables

TOTAL

SU

RV

IVIN

G

IMP

LAN

TS

FAIL

ED

IM

PLA

NTS

CS

R (%

)

Implant placement – Prosthesis delivery 183 2 0 98.9%

Prosthesis delivery – 1 year 181 6 0 95.6%

1 year – 2 years 175 0 14 95.6%

2 years – 3 years 93 0 6 95.6%

3 years 13 - - -

GBR GROUP

SU

RV

IVIN

G

IMP

LAN

TS

FAIL

ED

IM

PLA

NTS

WIT

HD

RAW

N

IMP

LAN

TS

CS

R (%

)

Implant placement – Prosthesis delivery 126 2 0 98.4%


1 year – 2 years 119 0 8 94.4%

2 years – 3 years 60 0 8 94.4%

3 years 11 - - -

NON-GBR GROUP

SU

RV

IVIN

G

IMP

LAN

TS

FAIL

ED

IM

PLA

NTS

WIT

HD

RAW

N

IMP

LAN

TS

CS

R (%

)Implant placement

– Prosthesis delivery 57 0 0 100%


1 year – 2 years 56 0 6 98.2%

2 years – 3 years 33 0 0 98.2%

3 years 2 - - -

more often for short implants and in soft bone. In spite of the failures, all patients received and main-tained a fixed crown/ bridge during the follow-up.

A total of 270 radiographs could be used for meas-urements of marginal bone levels. On average, the bone was situated 1.6 (S.D. 0.75) mm below the top of the collar at baseline and 1.9 (S.D. 0.73) mm at the follow-up one year later. Thus, the total bone loss amounted to 0.3 (S.D. 0.88) mm (Table 8) during this year. There were only small differ-ences between the GBR and non-GBR groups.

DISCUSSION

The experiences with the Neoss implant system from the first 50 patients treated in one clinic are reported. A survival rate of 98.2 % was achieved after one to two years in routine cases with no need for bone augmentation procedures. This is in accordance with recent studies on other implant systems (Albrektsson & Wennerberg, 2004). The implant that failed in this group was of 3.5 mm diameter with length 9 mm placed in quality type 4 bone in the maxilla. It was one of 8 implants placed with an immediate loading protocol for a full bridge reconstruction. A higher failure rate was seen when implants were placed with a simultaneous GBR procedure using bovine bone and a resorbable membrane, as 7 of 126 implants were lost (5.6%). The results indicate that osseointegration of the exposed parts of the implants was not achieved for these implants during healing prior to loading, resulting in an unfavourable biomechanical situation. This notion is further supported by the fact that short implants (7 and 9 mm) and implants in soft bone qual-ity failed more often. It is possible that a prolonged healing period is needed, as also suggested for sinus lift procedures with bovine bone (Hallman et al. 2005).

PATIENT NUMBER POSITION LENGTH DIAMETER BONE

QUALITYBONE

QUANTITY GBR TIME OF FAILURE

2865 22 9 3.5 3 C Yes Prosthesis

1896 23 9 3.5 3 C Yes Prosthesis

47 15 11 4.5 4 C Yes 1 year

154 36 7 3.5 2 B Yes 1 year

154 44 7 3.5 3 B Yes 1 year

1454 35 11 4.0 2 B Yes 1 year

1454 25 9 4.0 3 B Yes 1 year

2782 26 9 3.5 4 C No 1 year

Table 7. Characteristics of failed implants


34

Due to the retrospective character of the present study, radiographs could not be provided for all implants. Prospective studies with radiographs of consecutive implants are needed to properly evalu-ate marginal bone levels. Nevertheless, the marginal bone measurements indicated an acceptable degree of bone loss during follow-up. The average mar-ginal bone level at follow-up was still situated at the implant collar. For the Brånemark implant design, the marginal bone level usually ends up at the first thread some 1.5 to 2 mm below the platform (Oh et al, 2002). Other implant designs have shown only minor bone loss which may be due to micro threads on the collar (Shin et al. 2006) Too few radiographs were available to evaluate the influence of countersink depth of the implant collar on marginal bone loss.

The clinical experiences with the present implant sys-tem from a surgical, prosthetic and laboratory techni-cian point of view were positive. The implant design resulted in firm primary stability in all bone qualities. This is probably due to the geometry of the implant, which has a positive tolerance and is thereby slightly tapered. During insertion the bone is compressed in a lateral direction which increases the stability of the implant. Internal connection is an advantage as it makes abutment connection easy. Impressions are taken on implant level and the technician chooses the type of abutment, which in case of screw-re-tained prosthetics is integrated with the framework.

CONCLUSIONS

Within the limitations of the present retrospective study, it is concluded that the Neoss implant system results in good clinical outcomes in routine cases as evidenced by survival rate and marginal bone loss. GBR procedures involving short implants and soft bone seem to increase the risk of implant failure.

BONE LEVEL BASELINE

BONE LEVEL 1 YEAR FOLLOW-

UP VISIT

BONE LEVEL 2 YEAR FOLLOW-

UP VISIT

BONE LOSS BASELINE TO 1

YEAR

BONE LOSS 1 YEAR TO 2 YEAR

Mean 1.60 1.90 2.28 0.30 0.09

S.D. 0.75 0.73 0.70 0.88 0.74

N 80 76 60 56 22

Table 8. Marginal bone level measurements

Figure 1. Periapical radiographs taken after delivery of a two-unit bridge and after one year of loading.

REFERENCES

Albrektsson T, Wennerberg A.Oral implant surfaces: Part 2--review focusing on clinical knowledge of different surfaces. Int J Prosthodont. 2004 Sep-Oct;17(5):544-64.

Esposito M, Hirsch JM, Lekholm U, Thomsen P. Biological factors con-tributing to failures of osseointegrated oral implants. (I). Success criteria and epidemiology. Eur J Oral Sci. 1998 Feb;106(1):527-51.

Hallman M, Hedin M, Sennerby L, Lundgren S.A prospective 1-year clinical and radiographic study of implants placed after maxillary sinus floor augmentation with bovine hydroxyapatite and autogenous bone.J Oral Maxillofac Surg. 2002 Mar;60(3):277-84;

Hammerle CH, Lang NP.Single stage surgery combining transmucosal implant placement with guided bone regeneration and bioresorbable materials. Clin Oral Implants Res. 2001 Feb;12(1):9-18.


35

Hurzeler MB, Kohal RJ, Naghshbandi J, Mota LF, Conradt J, Hut-macher D, Caffesse RG. Evaluation of a new bioresorbable barrier to facilitate guided bone regeneration around exposed implant threads. An experimental study in the monkey. Int J Oral Maxillofac Surg. 1998 Aug;27(4):315-20.


Oh TJ, Yoon J, Misch CE, Wang HL. The causes of early implant bone loss: myth or science? J Periodontol. 2002 Mar;73(3):322-33.

Shin YK, Han CH, Heo SJ, Kim S, Chun HJ. Ra-diographic evaluation of marginal bone level around implants with different neck designs after 1 year.

Figure 2. Periapical radiographs of a single tooth restoration at impression taking and after one year of loading


36

Insertion Torque Measurements During Placement of Neoss Implants

Luca Pagliani1, Lars Sennerby2, Peter Andersson3, Damiano Verrocchi3 , Neil Meredith4

This clinical study demonstrates that the Neoss implant design develops continuously increasing insertion torque (IT) during placement; as expected for a tapered implant design. This indicates that the total implant length laterally compresses the adjacent bone, providing firm primary stability of the implant. Moreover, the study demonstrates a correlation between insertion torque and resonance frequency analysis measurements.

1Private Practice, Milano and Legnano, Italy2Dept Biomaterials, Inst Clinical Sciences, Sahlgrenska Academy, Gothenburg University, Sweden3Private Practice, Fiera di Primiero and Feltre, Italy4 Bristol Dental Hospital and University, Bristol, UK

INTRODUCTION

Primary stability is considered a key factor for the clinical success of dental implants. It is determined by the density of the bone at the site, the surgical technique and the design of the implant (Sennerby & Meredith 1998). Historically, increased failure rates have been reported in sites of low bone den-sity (Jaffin & Berman 1991, Friberg et al 1991). Modified drilling protocols have been proposed with the final drill diameter reduced in an attempt to increase compression and thereby the stability of the implant during insertion (Friberg et al 1999).

Tapered implant designs have been introduced on the market in order to improve primary stability. Experi-mental and clinical studies with insertion torque (IT) measurements and/or resonance frequency analysis (RFA)have demonstrated higher primary stability for tapered implants compared with parallel-walled im-plants (O’Sullivan et al 2000, O´Sullivan et al 2004a, 2004b). Parallel-sided implants may show a high peak insertion torque which indicates a high degree of stabil-ity. However, this is mainly due to the clamping effect when the implant head reaches the marginal bone, whilst the threaded part of the implant shows little resistance during insertion. Tapered implants demon-strate continuously increasing insertion torque due to lateral compression of the bone from the whole im-plant length during insertion (O’Sullivan et al (2000)). Then stresses would be distributed along the tapered implant surface and not concentrated to a few spots.

According to the manufacturer, the Neoss implant has a positive tolerance, meaning that it is slightly coni-cal in the coronal direction, like a tapered implant. However, the insertion torque characteristics of that implant design are not known at present. The objec-tive of this clinical study was to evaluate the primary stability of the Neoss implant using IT and RFA meas-urements. The aim was also to look for correlations between IT and factors such as bone quality and RFA.


A total of 118 implants (Neoss ltd, Harrogate, UK) of different lengths and diameters (Table 1) placed in both jaws (59 mandibular, 59 maxillary) of 38 pa-tients were evaluated at placement surgery. Insertion torque in Ncm was measured at 20 rpm and 8 Hz to a maximum of 50 Ncm using a drilling unit specially de-signed for implant surgery (Elcomed, W&H, Milano, Italy)(Figure 1). A torque/time curve was obtained, saved and extracted. The final drill diameter and the degree of countersinking were noted. After final seat-ing, the stability of each implant was measured with resonance frequency analysis in ISQ units (Mentor, Osstell AB, Gothenburg, Sweden). Bone density and quantity were assessed by the Lekholm & Zarb index.

The torque/time curves were examined for mean inser-tion torque over the total curve and for the apical (E1), mid (E2) and coronal thirds (E3). The E1 and E3 parts always included 40 measurements (corresponding to 5 seconds). The registration time for the E2 part varied due to the different implant lengths accounted for.


37

Spearman’s rho test was used to test for pos-s ible corre lat ions. A stat i s t ica l ly s ignif i -cant correlation was considered if p<0.05.

RESULTS The torque/time curves displayed continuously increas-ing torque during insertion (Figure 2). Thus, the im-plants behaved as expected for a tapered implant design as previously described by O’Sullivan et al (2000). The total mean insertion torque was 15.1 Ncm (SD 7.2) and for regions was 4.8 Ncm (SD 6.5) in the E1, 13.3 Ncm (SD 6.5) during E2 and 26.9 Ncm (SD 11.9) in E3 (Figure 3). Except for the E1 region, these torques are higher than those previously reported by Friberg et al (1999a) in 523 self-tapping parallel-sided implants; used in both jaws of 105 patients. The data indicate higher primary stability for the present implant design as described by IT measurements. This accords with previous experimental research which showed higher stability for tapered implants than for parallel-sided ones, without jeopardizing the integration process (O’Sullivan et al 2000, O´Sullivan et al 2004a, 2004b).

A statistically significant correlation between IT and RFA was demonstrated for the total implant lengths as well as for the E1-E3 regions individually (Table 2). The same correlation was reported by Friberg et al (1999b) in measurements of 47 parallel-sided implants placed in maxillary bone of nine patients. Since RFA measures stability as a function of stiffness, the results indicate that the insertion torque technique can be used to measure bone density and to predict primary stability as measured with RFA. There was no correla-

Figure 1. The Elcomed drilling unit used in the study. The handpiece was calibrated before each use (right).

2. Showing a typical torque/time curve for a 4 x 13 mm implant placed in the second premolar region in the maxilla. This implant showed an ISQ value of 76.

tion between total mean IT and bone density as as-sessed using the Lekholm and Zarb index (1985). This may reflect the subjective nature of the latter method which is purely based on the surgeon’s perception of bone density during drilling of the implant site.

It is concluded that the Neoss implant design de-velops a continuously increasing insertion torque during placement as expected for a tapered implant design. This indicates that the total implant length is involved in lateral compression, providing firm primary stability for the implant. Moreover, there is a correlation between IT and RFA measurements.

REFERENCES

Friberg B, Jemt T, Lekholm U.Early failures in 4,641 consecutively placed Brånemark dental implants: a study from stage 1 surgery to the connection of completed prostheses. Int J Oral Maxillofac Implants. 1991;6:142-6.

Friberg B, Sennerby L, Gröndahl K, Bergström C, Bäck T, Lekholm U. On cutting torque measurements during implant placement: a 3-year clinical prospective study. Clin Implant Dent Relat Res. 1999a;1:75-83.

3. Graphic presentation of the total mean insertion torque and for E1-E3 regions


38

Friberg B, Sennerby L, Meredith N, Lekholm U. A comparison between cutting torque and resonance frequency measurements of maxillary implants. A 20-month clinical study.Int J Oral Maxillofac Surg. 1999b;28:297-303.

Jaffin RA, Berman CL. The excessive loss of Branemark fixtures in type IV bone: a 5-year analysis. J Periodontol. 1991;62:2-4.


O’Sullivan D, Sennerby L, Meredith N. Measurements comparing the initial stability of five designs of dental implants: a human cadaver study.Clin Implant Dent Relat Res. 2000;2(2):85-92.

O’Sullivan D, Sennerby L, Jagger D, Meredith N.A comparison of two methods of enhancing implant primary stability. Clin Implant Dent Relat Res. 2004a;6:48-57.


Sennerby L, Meredith N. Resonance frequency analysis: measuring implant stability and osseointegration. Compend Contin Educ Dent. 1998;19:493-8.


39

Stress Evaluation of Dental Implant Wall Thickness using Numerical Techniques

Using the Finite Element Method, four implant diameters were evaluated for their effect on the stress distribution at the implant wall. A two-dimensional model of the implant and mandibular bone used triangular and quadrilateral plane strain elements to compute the von Mises stress in the bone with varied masticatory forces and abutment screw preloads. The masticatory force is found to be more influential than abutment screw preload, and implant wall thickness significantly influences the stress magnitude within the implant.

Rudi C. van Staden1, Hong Guan1, Yew-Chaye Loo1, Newell W. Johnson2, Neil Meredith3,

1Griffith School of Engineering, Griffith University Gold Coast Campus, Australia;2School of Dentistry and Oral Health, Griffith University Gold Coast Campus, Australia;3Neoss Ltd, Harrogate, United Kingdom

INTRODUCTION

Development of an ideal substitute for missing teeth has been a major aim of dental practitioners for mil-lennia (Irish 2004). Dental implants are biocompat-ible threaded titanium ‘fixtures’ surgically inserted into the mandible or maxilla to replace missing teeth. The establishment of a good biomechanical link between implant and jawbone is called osseointegra-tion (Branemark et al. 1969, 1977). The success of osseointegration depends on many factors including: medical status of the patient, smoking habits, bone quality, bone grafting, disturbance, sensation or its disruption, haematoma, haemorrhage, tooth necrosis, irradiation therapy, parafunctions, operator experience, degree of surgical trauma, bacterial contamination, lack of preoperative antibiotics, immediate loading, nonsubmerged procedure and implant surface char-acteristics (Esposito et al. 1998). Excessive surgical trauma and impaired healing ability, premature loading and infection are likely to be the most common causes for early implant losses, whereas progressive chronic marginal infection (peri-implantitis) and overload in conjunction with host characteristics are the major reasons for delayed failures (Esposito et al. 1998). For both early and late implant failures, loading is considered an important factor (Geng et al. 2001).

The distribution and magnitude of stresses within the implant are influenced by the implant dimensions, as documented by (Huang et al. 2005, Capodiferro et al. 2006, Winklere et al. 2003, Naert et al. 1992, Tolman and Laney 1992). Catastrophic mechanical failure of the implant may occur by implant fatigue (Huang et al. 2005, Capodiferro et al. 2006), implant fractures, veneering resin/ceramic fractures or other mechanical retention failures (Winklere et al. 2003, Naert et al. 1992, Tolman and Laney 1992). From an engineering perspective, an important criterion in designing an implant is to have a geometry that can minimize mechanical failure caused by an extensive range of loading. As part of the implantation process, the torque is applied to the abutment screw causing an equivalent preload or clamping force between the abutment and implant. This is to ensure that the various implant components are perfectly attached to each other. However, to date no published research appears to have investigated the influence of mastica-tory forces and abutment screw preloading on stresses in implants of various diameters. Therefore, the aim of this study is to evaluate the stress within an imme-diately loaded implant under a range of masticatory forces, abutment preloads and implant diameters,.


40


Modelling and simulation were performed using the Strand7 (2004) Finite Element Analysis (FEA) System. First the implant and bone geometry were defined, then material properties for the implant and various bone components were assigned in terms of Young’s modulus, Poisson’s ratio and density. The loading and restraint conditions were applied and the effect on the stress profile within the implant was evaluated.

ModellingA two-dimensional (2D) representation of the implant and mandibular bone was analysed because this was considered to be equally accurate and more efficient in computation time, as three-dimensional (3D) model-ling. Data were acquired for the bone dimensions by CT scanning. Two different types of bone, cancellous and cortical, were distinguished and the boundaries were identified in order to assign different material properties within the finite element model. Figure 1 shows an implant and mandible section with the loading and restraint conditions. Also shown in the figure are the parameters investigated in this study.

The implant is based on that used by Neoss (2006), it is conical with 2 degrees of taper and a helical thread. For the finite element model with D = 4.5mm, L = 11mm, Tcor = 1.2mm, 3314 plate elements and 3665 nodal points were used for the implant, 3804 elements and 4079 nodes for cancellous bone, and 1216 elements and 1453 nodes for cortical bone.

Implant

Cancellous bone

Cortical bone

Crown

Abutment

Abutment screw torque, T = 110, 320, 550Nmm

(Diameter, D = 3.5, 4.0, 4.5, 5.5)

(Young’s modulus = 1GPa)

(Young’s modulus = 13.7GPa)(Thickness = 1.2mm)

(Length, L = 11mm)

Fixed constraints

Masticatory force, FM = 200, 500, 1000N

FM

x

y45o

VV1-2

VV2-3

VV3-4

VV1

VV2

VV3

VV4

(Equivalent abutment screw preload, FP = 201.93, 587.44, 1009.67N)

FP

FP

Implant

Cancellous bone

Cortical bone

Crown

Abutment

Abutment screw torque, T = 110, 320, 550Nmm

(Diameter, D = 3.5, 4.0, 4.5, 5.5)

(Young’s modulus = 1GPa)

(Young’s modulus = 13.7GPa)(Thickness = 1.2mm)

(Length, L = 11mm)

Fixed constraints


FM

x

y45o

VV1-2

VV2-3

VV3-4

VV1

VV2

VV3

VV4

(Equivalent abutment screw preload, FP = 201.93, 587.44, 1009.67N)

FP

FP

The effect on stress in the implant wall under different mastication forces (FM) and abutment preload (FP) was evaluated. For all the possible parametric combi-nations, the von Mises stress along the line VV in the implant wall was investigated and is believed to play a crucial role in the probability of implant fracture.

As indicated in Figure 1 the von Mises stresses along the lines VV1-2, VV2-3 and VV3-4 are reported for all possible combinations of loading and diameters. The relative locations of these lines are also detailed in the figure. Each line is identified by its start and end points, so, for example, line VV1-2 begins at VV1 and ends at VV2. These locations were sug-gested by clinicians to be prone to micro fractures.

The range of implant diameters (3.5, 4.0, 4.5 and 5.5mm) is shown in Figure 2. Note that the cor t i ca l bone i s cons t ra ined d i s ta l ly to represent normal function of the mandible.

Figure 1. Finite element model of implant, components, implant/bone interface and bone

Figure 2 . Finite element model showing different implant diameters

3.5mm 4.0mm 4.5mm 5.5mm3.5mm 4.0mm 4.5mm 5.5mm


41

Plane strain elementsThe model represents a cross-sectional slice from the mandible (Figure 3). The “arc length” of the mandible is comparable to the width and depth of the slice. When the slice is subjected to in-plane (x-y) masti-catory forces (FM), it is restrained from deforming out-of-plane (in the z-axis) and it was assumed that all strains are confined in the z-axis. To accurately represent the mechanical behaviour of the bone and implant, 3-node triangular (Tri3) and 4-node quadri-lateral (Quad4) plane strain elements were therefore used for the construction of the finite element models. Note that for plane strain elements each node has a complete set of spatial degrees of freedom includ-ing u and v. The constraints meant that nodes could only translate in the x- (u) and y- (v) directions.

Material propertiesThe material properties adopted were specified in terms of Young’s modulus, Poisson’s ratio and density for the implant and all associated com-ponents (Table 1). All materials were assumed to exhibit linear, homogeneous elastic behaviour.

x

z

2D Slice

x

z

x

z

2D Slicez

x

y

2D Slice

Mandible

FM

45o

-x

FM

z

x

y

2D Slice

Mandible

FM

45o

-x

FM

Figure 3. Location of 2D slice in a mandiblea) Top view b) Isometric view

Component Description Young’s Modulus, E (GPa) Poissons ratio, v Density, ρ (g/cm3)

Implant, abutment, washer Titanium (grade 4) 105.00 0.37 4.51

Abutment screw Gold (precision alloy) 93.00 0.30 16.30

Crown Zirconia (Y-TZP) 172.00 0.33 6.05

Cancellous bone 1.00 0.30 0.74

Cortical bone Cortical thickness = 1.2mm 13.70 0.30 2.19

Table 1. Material properties

Loading conditionsLoading conditions included the masticatory force, FM, applied to the occlusal surface of the crown set at 200, 500 and 1000N, as shown in Figure 1, and applied at 45o inclination in the x y plane (re-fer to Figure 1). A preload, FP, was applied to the abutment screw of magnitude 201.93, 587.44 or 1009.67 N through the use of temperature sensi-tive elements. The technique for applying the masticatory forces to the crown, and the preload applied to the abutment screw, are discussed below.

Masticatory force, FMDuring normal function the crown is subjected to oblique loads applied to the occlusal surface of the crown. These loads are a result of normal mastica-tory function. The theoretical study by Choi et al. (2005) suggests that this loading condition can be considered to represent all applied loads. The implant was restrained in the x, y and z directions within the jawbone, assuming complete osseointegration at the bone and implant interface. Figure 1 shows the positions of the applied loading and restraints.


42

Preload, FP

The torque applied to the abutment screw causes the preload or clamping forces between the im-plant and abutment. The procedure for calculating FP and applying a temperature sensitive element, functioning throughout the abutment screw devel-oping the preload (or torque), is described below. Note that for the purpose of this article only the Neoss system calculations for FP and tempera-ture sensitive elements are shown as an example.

A dental implant system typically consists of a crown, abutment and abutment screw. The abut-ment screw is screwed with a manual screwdriver (Figure 4) into the internal thread of the implant. Finally, the crown is placed on to the abutment us-ing cement at the crown and abutment interface.

The nature of the forces clamping implant com-ponents together, and how they are generated and sustained, are not comprehensively covered in the literature. Preload was considered in finite element modelling by Haack et al. 1995, Lang et al. 2003 and Byrne et al. 2006. These studies were either based on complicated 3D modelling or did not specify the techniques used for replicating FP. For the pur-pose of this study the calculation used to determine the preload is based on findings by Dekker (1995).

Implant Abutment Abutment screw Manual screwdriver

Torque applied to top of abutment screw

x

y

z





x

y

z

The re la t ionship between the torque ap-plied to the abutment screw and the preload achieved was expressed by Dekker (1995) as:

nntt

Pin rri

FTcos2 … . . ( 1 )

u s ing the cond i t ion s l i s t ed in Tab l e 2 ;

Note that the effective radius (rt and rn) is the distance between the geometric centre of the part (implant, abutment or abutment screw) and the circle of points through which the resultant con-tact forces between mating parts (implant, abut-ment or abutment screw) pass (refer to Figure 4).

Calculations of the preload are shown in equation 2 below, using information derived in equation 1.

)1125.120.0()72.28cos(

)8725.026.0(2

4.0320 PF

. . . . . . . . . . . . . . . . . … … … … … … . . ( 2 )

Rearranging for FP;

)1125.120.0()72.28cos(

)8725.026.0(2

4.0

320PF

FP = 587.44N

Abbreviation Description Magnitude ReferenceTin = torque applied to the abutment screw (Nmm) = 320 Neoss (2006)FP = preload created in the abutment screw (N) = Unknown factori = the pitch of the abutment screw threads (mm) = 0.40 Neoss (2006)

miut

= the coefficient of friction between abutment screw thread surfaces and internal implant screw thread surfaces (dimensionless)

= 0.26 Lang et al. (2003)

rt = the effective contact radius between the inner implant thread and the abutment screw threads. (mm)

= (r3 + r4)/2 = (0.99 + 0.755)/2 = 1.745 / 2 = 0.87 Neoss (2006)

= the half-angle of the threads (degree) = 28.72ْ Neoss (2006)

miun= the coefficient of friction between the face of the abutment screw and the upper surface of the abutment (dimensionless) = 0.20 Lang et al.

(2003)rn

= the effective radius of contact between the abutment and implant surface (mm)

= (r1 + r2)/2 = (1.275 + 0.95)/2 = 2.225 / 2 = 1.11 Neoss (2006)

Table 2. Conditions applied in equation 1

Fig. 4. Implant system


43

The preload clamping the abutment to the implant is transferred through two interfaces. The first inter-face (SA1) is between the abutment and abutment screw and the second (SA2), between the abutment screw threads and the inner threads of the implant (Figure 5). The calculated preload, FP, is assumed to act equally, as a pressure, q, across the first and second interfaces. Due to equilibrium, only the pressure q, acting on SA1 is considered in this study.

SA1 = (π × r12) - (π × r2

2) = (π × 1.2752) - (π × 0.952)SA1 = 2.27mm2

T h e p r e s s u r e a c t i n g o n S A 1, w h e n F P = 5 8 7 . 7 7 N , w a s c a l c u l a t e d a s f o l l ow s :

2748984.244.587

1SA

Fq P

. . . . . … … . … … ( 3 )

q = 258.22 N/mm2

The clamping pressure, q, is a result of the applied torque and is a means of replicating the 3D torque in a 2D man-ner. For the present study, q, is calculated as above for the applied torques of 110, 320 and 550Nmm (Table 3).

In the analysis a negative temperature (-10 Kelvin, K) is applied to all the nodal points within the abut-ment screw, causing each element to shrink. A trial and error process is applied to determine the tem-perature coefficient, C, that can yield an equivalent q. The corresponding C is also presented in Table 3.

T (Nmm) FP (N) q (N/mm2) C3.5mm (×10-12) C4.0mm (×10-12) C4.5mm (×10-12) C5.5mm (×10-12)

110.00 201.93 88.76 -2.62 -2.54 -3.52 -4.14

320.00 587.44 258.22 -7.61 -7.39 -10.20 -12.71

550.00 1009.67 443.83 -13.08 -12.67 -17.64 -20.78

Table 3. Abutment screw torque (T), preload FP, and surface pressure q

Abutment

r1r2

x

y

z

x

z

SA1

Abutment

r1r2

x

y

z

x

z

SA1

Abutment screw

SA2

x

y

CL

Abutment screw

SA2

x

y

CLCL

Figure 5. Effective radius of abutment and abutment screw

a) Abutment

b) Abutment screw


44

RESULTS

The distribution of von Mises stresses in the implant is discussed for all combinations of masticatory and preload forces. Each parameter is discussed in a sepa-rate section for all four implant diameters. The von Mises stresses are reported between locations VV1-2 (0 - 1.51mm in length), VV2-3 (1.51 - 2.24mm) and NN3-4 (2.24 - 3.77mm), as shown in Figure 1.

Masticatory Force, FMThe distributions of von Mises stresses along the lines VV1-2, VV2-3 and VV3-4 for all values of FM are shown in Figure 6. Note that the pre-load, FP, is set at its medium value, i.e. 587.44N.

In general, when the applied masticatory force, FM, is increased, the von Mises stresses also increase propor-tionally, because the system being analysed is linear elastic. As expected the 3.5mm implant shows higher stresses than all other diameters (refer to Figures 6 a and b). The 3.5mm also indicates stress peaks along the lines VV1-2 and VV2-3 where all other parameters only have elevated stress peaks along the line VV1-2 (Figures 6 a, c, e and g). This is because the implant wall thickness for the 3.5mm implant is significantly reduced in the region corresponding to VV2-3, hence causing stress concentration. The 4.0 and 4.5mm diameter implants have similar stress distribution characteristics but the stresses are lower in magnitude at VV1-2, VV2-3 and VV3-4 than with the 3.5mm im-plant because of the greater wall thickness (Figures 6 d and f ). The 5.5mm implant displays greatly reduced stresses at all locations (Figures 6 g and h), with peak stresses occurring close to point VV1 (see Figure 1).

Preload Force, FP

To investigate the effect of different preload FP, FM is kept as a constant and its medium value, i.e. 500N is considered herein. The distributions of von Mises stresses along the lines VV1-2, VV2-3 and VV3-4 for all values of FP are shown in Figure 7. Similar stress distribution characteristics were found when varying FP as with FM. Note that when FP increases, the von Mises stresses also increase. However, the increase is not proportional to the increase of FP.. This is because FP, as an internal force, is a function of the abutment screw and implant diameters. This suggests that fail-ure of the crown is more likely to be caused by FM.

DISCUSSION

FEA has been used extensively to predict the bio-mechanical and mechanical performance of implant designs as well as the effect of clinical factors on the success of implantation (DeTolla et al. 2000, Geng et al. 2001). The principal difficulty in simu-lating the mechanical behaviour of implants is the modelling of the living human bone tissue and its response to applied mechanical forces (van Staden et al. 2006). A few studies have considered the influ-ence of such factors as the torque (Lang et al.2003) with which the abutment is connected, and the effect of masticatory forces on the probability of implant failure or loosening (Byrne et al. 2006).

This paper considers various combinations of loading parameters and their influence on the stress produced within implants of diameters 3.5, 4.0, 4.5 and 5.5mm. As expected, increased masticatory forces lead to great-er stresses within the implant. The preload applied to the abutment has less influence on the stresses than the masticatory forces. The 3.5mm implant shows higher stresses than all other diameters and indicates stress peaks along the lines VV1-2 and VV2-3 where all other parameters only have elevated stress peaks along the line VV1-2. The 4.0 and 4.5mm diameter implants have similar stress distribution characteristics and the 5.5mm implant displays greatly reduced stresses at all loca-tions, with peak stresses occurring close to point VV1.

CONCLUSION

Overall, it was found that the masticatory force is more influential on implant stresses than the abut-ment screw preload. As expected the implant wall thickness significantly influences the stress magnitude within the implant. Note also that when the wall thickness is decreased (especially for the 3.5mm) stress concentration occurs at the internal and external threads as well as at sharp corners of the implant wall.

Characteristically the stress showed an increase at the top of the implant thread and the top of the implant (line VV1-2). It should be noted that this research was a pilot study aimed at offering an initial understanding of the complicated stress distribution characteristics due to the various parameters. Other parameters which may be evaluated for their influence on implant stress profiles include the implant taper, pitch and design of


45

FM = 200N, FM = 500N, FM = 1000N,

VV1-2

VV2-3

VV3-4

FM = 200N, FM = 500N, FM = 1000N,

VV1-2

VV2-3

VV3-4

VV1-2

VV2-3

VV3-4

FM = 200N,

VV1-2

VV2-3

VV3-4

FM = 500N, FM = 1000N, FM = 200N,

VV1-2

VV2-3

VV3-4

FM = 500N, FM = 1000N,

FM = 200N,

VV1-2

VV2-3

VV3-4

FM = 500N, FM = 1000N, FM = 200N,

VV1-2

VV2-3

VV3-4

FM = 500N, FM = 1000N,

FM = 200N,

VV1-2

VV2-3

VV3-4

FM = 500N, FM = 1000N, FM = 200N,

VV1-2

VV2-3

VV3-4

FM = 500N, FM = 1000N,

Figure 6. Stress characteristics when varying FM.

Figure 6. Stress characteristics when varying FM.

a) Stress profile

c) Stress profile

e) Stress profile

g) Stress profile

b) Stress contour

d) Stress contour

f) Stress contour

h) Stress contour


46

FP = 201.93N,

VV1-2

VV2-3

VV3-4

FP = 587.44N, FP = 1009.67N, FP = 201.93N,

VV1-2

VV2-3

VV3-4

FP = 587.44N, FP = 1009.67N,

VV1-2

VV2-3

VV3-4

FP = 201.93N, FP = 587.44N, FP = 1009.67N,

VV1-2

VV2-3

VV3-4

FP = 201.93N, FP = 587.44N, FP = 1009.67N,

VV1-2

VV2-3

VV3-4

FP = 201.93N, FP = 587.44N, FP = 1009.67N,

VV1-2

VV2-3

VV3-4

FP = 201.93N, FP = 587.44N, FP = 1009.67N,

VV1-2

VV2-3

VV3-4

FP = 201.93N, FP = 587.44N, FP = 1009.67N,

VV1-2

VV2-3

VV3-4

FP = 201.93N, FP = 587.44N, FP = 1009.67N,

a) Stress profile b) Stress contour

c) Stress profile d) Stress contour

e) Stress profile f) Stress contour

g) Stress profile h) Stress contour

Figure 7. Stress characteristics when varying FP


47

implant thread, implant neck offset, different stages of bone remodelling and implant orientation within the bone. It is important that clinicians understand the methodology, applications, and limitations of FEA in implant dentistry, and become more confident in inter-preting results from FEA studies to clinical situations.

ACKNOWLEDGEMENTS

This work was made possible by the collaborative support from Griffith’s School of Engineering and School of Dentistry and Oral Health. A special thank you goes to Messer John Divitini and Fredrik Engman from Neoss Limited for their continual contribution.

REFERENCES

Branemark PI, Adell R, Breine U, Hansson BO, Lindstrom J, Ohls-son A. Intra-osseous anchorage of dental prostheses. I. Experimental studies. Scandinavian Journal of Plastic and Reconstructive Surgery. 1969;3(2):81-100.

Branemark PI, Hansson BO, Adell R, Breine U, Lindstrom J, Hallen O, Ohman A. Osseointegrated implants in the treatment of the edentulous jaw. Experience from a 10-year period. Scandinavian Journal of Plastic and Reconstructive Surgery. 1977;16:1-132.

Byrne D, Jacobs S, O’Connell B, Houston F, Claffey N. Preloads gener-ated with repeated tightening in three types of screws used in dental implant assemblies. The Journal of Prosthetic Dentistry. 2006;15(3):164-171.

Capodiferro S, Favia G, Scivetti M, De Frenza G, Grassi R. Clinical management and microscopic characterisation of fatique-induced failure of a dental implant. Case report. Head & Face Medicine. 2006;22(2):18.

Choi AH, Ben-Nissan B, Conway RC. Three-dimensional modelling and finite element analysis of the human mandible during clenching. Austral-ian Dental Journal. 2005;50(1):42-48.

Dekker M. An introduction to the design and behavior of bolted joints / John H. Bickford. 1995;3rd ed.

DeTolla DH, Andreana S, Patra A, Buhite R, Comella B. Role of the finite element model in dental implants. Journal of Oral Implantology. 2000;26(2):77-81.

Esposito M, Hirsch JM, Lekholm U, Thomsen P. Biological factors contributing to failures of osseointegrated oral implants. (II). Etiopatho-genesis. European Journal of Oral Sciences. 1998;106(3):721-764.

Geng JP, Tan KB, Liu GR. Application of finite element analysis in im-plant dentistry: a review of the literature. Journal of Prosthetic Dentistry. 2001;85:585-598.

Haack JE, Sakaguchi RL, Sun T, Coffey JP. Elongation and preload stress in dental implant abutment screws. International Journal of Oral & Maxillofacial Implants. 1995;10(5):529-536.

Huang HM, Tsai CM, Chang CC, Lin CT, Lee SY. Evaluation of load-ing conditions on fatigue-failed implants by fracture surface analysis. International Journal of Oral & Maxillofacial Implants. 2005;20(6):854-

859.

Irish JD. A 5,500 year old artificial human tooth from Egypt: a his-torical note. International Journal of Oral & Maxillofacial Implants. 2004;19(5):645-647.

Lang LA, Kang B, Wang RF, Lang BR. Finite element analysis to determine implant preload. The Journal of Prosthetic Dentistry. 2003;90(6):539-546.

Naert I, Quirynen M, van Steenberghe D, Darius P. A study of 589 consecutive implants supporting complete fixture prostheses. Part II: prosthetic aspects. Journal of Prosthetic Dentistry. 1992;68:949-956.

Neoss Limited (2006), Neoss Implant System Surgical Guidelines, UK.

Strand7 Pty Ltd (2004) Strand7 Theoretical Manual, Sydney, Australia.

Tolman DE, Laney WR. Tissue integrated prosthesis complications. International Journal of Oral & Maxillofacial Implants 1992;7:477-484.

van Staden RC, Guan H, Loo YC. Application of the finite element method in dental implant research. Computer Methods in Biomechani-cal and Biomedical Engineering. 2006;9(4):257-270.

Winkler S, Ring K, Ring JD, Boberick KG. Implant screw mechan-ics and the settling effect: overview. Journal of Oral Implantology. 2003;29(5):242-245.


48

Comparative Analysis of Two Implant-Crown Connection Systems - A Finite Element Study

Rudi C. van Staden1, Hong Guan1, Yew-Chaye Loo1, Newell W. Johnson2, Neil Meredith3,

1Griffith School of Engineering, Griffith University Gold Coast Campus, Australia;2School of Dentistry and Oral Health, Griffith University Gold Coast Campus, Australia;3Neoss Ltd, Harrogate, United Kingdom

The internal and external-hex connections of the Neoss and 3i implant systems, were compared in a three-dimensional Finite Element Analysis. Chewing forces of 200, 500 and 1000N and abutment screw preloads of 110, 320 and 550Nmm were studied. The connection type strongly influences the stress profile within the crown, with the external-hex connection exhibiting greater stresses than the internal-hex.

INTRODUCTION

Dental implants are a well accepted treatment for partially or totally edentulous subjects. Innovations through research have led to advancements in sur-gical and restorative techniques, improved surface features and restorative components. Dental implants typically use either internal-hex or external-hex connections with the crown (Figure 1 a). Although both connections are extensively used clinically, distinctly different stress distributions are produced within the crown. Clinicians have reported implant components linked to mechanical failure of crown and implant (Maeda et al. 2006, Merz et al. 2000, Khraisat et al. 2002). Two major factors may be implicated in crown and implant failure. These are;

- over tightening of the abutment screw lead-ing to failure of the crown for internal-hex and external-hex connection type implant systems.

- excessive masticatory loads transferred from the oc-clusal plane of the crown to a stress concentration at the interface between the abutment and implant body.

Theoretical techniques such as the Finite Element Method (FEM) can be used to evaluate mechanical factors that could affect implant performance and success (Capodiferro et al. 2006, Gehrke et al. 2006, Huang et al. 2005, Khraisat et al. 2005). This study was undertaken using Finite Element Analysis (FEA) to aid understanding of the stresses in both internal-hex and external-hex implant systems under different loading conditions.

Studies by Maeda et al. (2006), Khraisat et al. (2002) and Merz et al. (2000) have all considered the stress within the abutment screw but disregarded the stress within the crown. To date no published research appears to have investigated the stress profile in the crown due to an internal-hex or external-hex connection. Ultimately, the outcome of this study will help dental practitioners to identify locations within the implant system susceptible to stress concentrations.


The modelling and simulation herein are performed using the Strand7 FEA System (2004). The first step of the modelling is to define the geometry of the implant system. Then the material behaviour of the model is specified in terms of Young’s modulus, Poisson’s ratio and density for all components. The appropriate load-ing and restraint conditions are applied and the indi-vidual parameters and their contribution to the stress profile within the crown and implant is evaluated.

ModellingGeometry for the implant systems were obtained from the manufacturers (Figure 1a). Section AA is set at a location where maximum compres-sive stresses occur within the crown, positioned at 30o from the x-axis towards the negative z-axis.

The parameters investigated are shown in Figure 1b. The implant is conical with 2 degrees of taper, a heli-cal thread, outside diameter of 4.5mm and length of


49

11mm. Half the implant is modelled and symmetrical constraints are applied along the plane of symmetry as indicated (Figure 1b). For the Neoss (2006) and 3i (2006) finite element models, the total numbers of elements are respectively 13464 and 30420 for the implant, 3564 and 9108 for the abutment, 17424 and 25956 for the abutment screw, and 38484 and 47052 for the crown. The total number of nodal points for the entire Neoss model are 122688 and 82547 for the 3i.

Crown

Abutment screw

Abutment

Implant

External connectionInternal connectionNeoss 3i

Section AA Section AA

Crown

Abutment screw

Abutment

Implant

External connectionInternal connectionNeoss 3i

Section AA Section AA

Abutment screw torque, T = 110, 320, 550Nmm (Equivalent abutment screw preload, FP = 201.93, 587.44, 1009.67N)

FM

z

x

y

45o


90o

Fixed restraint, Rz, Dx and Dy

Fixed restraint, Rxyzand Dxyz

y-axis

x-axis

Abutment screw torque, T = 110, 320, 550Nmm (Equivalent abutment screw preload, FP = 201.93, 587.44, 1009.67N)

FM

z

x

y

45o


90o

Fixed restraint, Rz, Dx and Dy

Fixed restraint, Rxyzand Dxyz

y-axis

x-axis

Fig. 1. Finite element model of external and internal-hex system a) Implant system

External connectionInternal connection

Section AA

NN1-2

NN3-4

NN1

NN3

NN2

NN4

NN2-3

II1

II2II3II4II5II6II7

II2-3

II1-2

II3-4

II6-7

II5-6

II4-5

External connectionInternal connection

Section AA

NN1-2

NN3-4

NN1

NN3

NN2

NN4

NN2-3

II1

II2II3II4II5II6II7

II2-3

II1-2

II3-4

II6-7

II5-6

II4-5

b) Loading and restraint conditions (with detailed parameters) c) Locations for measuring stress profile and contour

The main focus of this study is to examine the stress characteristics within the crown and next to the crown-abutment interface. Therefore an as-sumption is made to restrain the outer edge of the implant (Figure 1b) when the mandibular bone is not included in the finite element model. Note that these loading and restraint conditions are the same for both internal and external-hex systems.


50

Stress reportingThe von Mises stresses along the lines NN (NN1-2, NN2-3 and NN3-4) and II (II1-2, II2-3, II3-4, II4-5, II5-6 and II6-7) for both systems are reported for all possible combinations of loading (Figure 1 c). The relative locations of these lines are also detailed in the figure, which are identified by their start and end points. So, for example, the line II1-2 begins at II1 and ends at II2. The von Mises stresses along the lines NN and II, in the crown, are believed to play a crucial role in the probability of crown fracture. On Section AA lines NN and II were chosen because the greatest stresses due to the masticatory loading (compres-sive is prominent over tensile) occur on this plane.

Material propertiesThe material properties in the model are specified in terms of Young’s modulus, Poisson’s ratio and density for the implant and all associated com-ponents (Table 1). All materials are assumed to exhibit linear, elastic and homogeneous behaviour.

Loading conditionsMasticatory force, FM, was applied to the occlusal surface of the crown at 100, 250 or 500N, inclined at 45o to the x and y-axes (Figure 1 b). The preload, FP, of 100.97, 293.72 or 504.84N is applied to the abutment screw through the use of temperature sensitive ele-ments (Figure 1 b). Note that FM and FP are set to half of the total magnitude because only half of the implant system is modelled. Therefore the total FM modelled is 200, 500, 1000N and FP is 201.93, 587.44, 1009.67N. The manner of modelling the masticatory forces and the preload applied to the abutment screw was de-scribed by van Staden et al. (2008). For the purposes of this study both the abutment screw preload, FP, and the surface area between abutment and abut-ment screw, SA1, are halved when compared to those used by van Staden et al. (2008) due to the modelling assumption (half model). Calculations for the abut-ment screw surface pressure, q, confer identical results to those found by van Staden et al. (2008).

For the present study a negative temperature (-10

Kelvin, K) was applied to all the nodal points within the abutment screw, causing each element to shrink. A trial and error process was applied to determine the temperature coefficient, C, for the Neoss and 3i systems (i.e. CNeoss and C3i) that can yield an equivalent q. It was found that when FP = 201.93, 587.44 and 1009.67N then CNeoss = -3.51×10-4, -9.28×10-4 and -15.60×10-4 /K, and C3i = -0.98×10-

4, -1.80×10-4 and -2.68×10-4 /K, respectively.

RESULTS

Zirconia, typically used as a dielectric material, has proven adequate for application in dentistry. With its white appearance and high Young’s modulus it is ideal for use in sub frames for dental restorations such as crowns and bridges, which can then be veneered with conventional feldspathic porcelain. Zirconia has a frac-ture strength greater than titanium; therefore it may be considered as a high strength material. However with masticatory and preload forces the compressive strength of 2.1 GPa (Curtis et al. (2005)) can easily be exceeded; especially for implant systems with exter-nal-hex connections, as confirmed during this study.

The distribution of von Mises stresses in the crown is discussed for all parametric combinations of mastica-tory and preload forces. Each parameter is discussed in a separate section. For the Neoss system, the von Mises stresses are reported between locations NN1-

2 (0-1.76mm in length), NN2-3 (1.76-1.87mm) and NN3-4 (1.87-3.96mm). For the 3i system the von Mises stresses are reported between locations II1-2 (0-2.38mm), II2-3 (2.38-2.78mm), II3-4 (2.78-3.67mm), II4-5 (3.67-4.06mm), II5-6 (4.06-4.65mm) and II6-7 (4.65-5.27mm), as shown in Figure 1c.

Masticatory Force, FM

The distributions of von Mises stresses along the lines NN and II for all values of FM are shown in Figure 2. Note that the preload, Fp, is set at its medium value, i.e. 587.44N.

In general, when the applied masticatory force, FM, is increased, the von Mises stresses also increase propor-

Component DescriptionYoung’s Modulus, E

(GPa)

Poissons

ratio, v

Density, ρ

(g/cm3)

Implant, abutment, washer Titanium (grade 4) 105.00 0.37 4.51

Abutment screw Gold (precision alloy) 93.00 0.30 16.30

Table 1. Material properties


51

NN1-2

NN3-4

NN1

NN3

NN2

NN4

Crown

FM = 200N FM = 1000N

NN2-3

NN1-2

NN3-4

NN1

NN3

NN2

NN4

Crown

FM = 200N FM = 1000N

NN2-3

II1

Crown

II2

II3

II4

II5

II6

II7

II2-3

II1-2

II3-4

II6-7

II5-6

II4-5

FM = 200N FM = 1000N

II1

Crown

II2

II3

II4

II5

II6

II7

II2-3

II1-2

II3-4

II6-7

II5-6

II4-5

FM = 200N FM = 1000N

tionally, because the system being analysed is linear elastic. When FM increases the stress along the line NN increases showing two peaks along the line NN3-4 (refer to Figure 2a). The larger of these two peaks occurs at a distance of ±3.8mm in length from NN1. This stress peak (as can be identified in Figure 2 b) is caused by a sharp corner and sudden change in section at this point.

Elevated stress concentrations are identified at the be-ginning of the line II3-4 (Figures 2 c and d). This stress peak, as can be identified in Figure 2 c, is caused by a sharp corner at this point. For the 3i system the volume of the crown exceeds that of the Neoss system, thereby suggesting that the 3i crown may show greater resis-tance to the applied masticatory forces. However, even though the Neoss crown has a thinner wall along the line NN3-4, reduced stresses are still evident due to the abutment’s high Young’s modulus. Overall, the design differences between the Neoss and 3i systems result in

Figure2. Stress characteristics when varying FM

a) Stress profile

c) Stress profile

b) Stress contour

d) Stress contour

higher stresses in the 3i system when FM is increased.

Preload Force, FP

To investigate the effect of different preload FP, FM is kept as a constant and its medium value, i.e. 500N is considered herein. The distributions of von Mises stresses along the lines NN and II for all values of FP are shown in Figure 3.

As found in Section 3.1 for FM, when FP increases the stresses calculated along the line NN increase, showing two peaks along the line NN3-4 (refer to Figures 3 a and b). Also, as found for FM, elevated stress peaks are identified at the beginning of the line II3-4 (Figures 3 c and d). Overall, all values of FM cause greater stresses along lines NN and II, than do varying values of FP.


52

NN1-2

NN3-4

NN1

NN3

NN2

NN4

Crown

FP = 201.93N FP = 1009.67N

NN2-3

NN1-2

NN3-4

NN1

NN3

NN2

NN4

Crown

FP = 201.93N FP = 1009.67N

NN2-3

II1

Crown

II2

II3

II4

II5

II6

II7

II2-3

II1-2

II3-4

II6-7

II5-6

II4-5

FP = 201.93N FP = 1009.67N

II1

Crown

II2

II3

II4

II5

II6

II7

II2-3

II1-2

II3-4

II6-7

II5-6

II4-5

FP = 201.93N FP = 1009.67N

Figure3. Stress characteristics when varying FP

c) Stress profile d) Stress contour

a) Stress profile b) Stress contour

DISCUSSION

FEA has been used extensively to predict the biome-chanical performance of the jawbone surrounding a dental implant (DeTolla et al. 2000, Geng et al. 2001). Previous research considered the influence of the implant dimensions and the bone-implant bond on the stress in the surrounding bone. However, to date no research has been conducted to evaluate the stress produced by different implant to crown connections (ie. internal-hex and external-hex).

The analysis completed in this paper uses the FEM to replicate internal-hex and external-hex connections for

loading parameters of FM and FP. As shown in Table 2, two stress peaks were revealed along the lines NN and II at locations 3.76 and 2.89mm from the top. The stress values shown were calculated with the other parameter (i.e. FM or FP) set to its average. The mastication force FM is applied on the occlusal surface of the crown, evenly distributed along 378 nodal locations (Figure 1 b), and orientated at 45o in the x-y plane. This induces compressive stresses in the right hand side of the crown and tensile in the left. Varying FM from 200 to 1000N for the internal-hex and external-hex systems results in a change in von Mises stress of 545.64 (818.47-272.82MPa) and 698.09MPa (1047.14-349.05MPa) respectively. The geometrical design of the external-hex system tends to induce stress concentrations, located

Parameter-------------FM (N)-------------- -------------FP (N)--------------

Line 200 500 1000 201.93 587.44 1009.67

NN (3.76mm) 272.82 545.64 818.47 231.55 545.64 891.83

II (2.89mm) 349.05 698.09 1047.14 466.21 698.09 951.67

Table 2. von Mises stress (MPa) in the crown (location of stress reporting in brackets)


53

2.89mm from the apex in this study. For this system, a stress concentration at this point is also induced by FP, increasing the compressive stresses on the right hand side of the crown. Increasing FP from its minimum to maximum values, for the external-hex system, increases the stress by 485.46MPa (951.67-466.21MPa).

The internal-hex system has reduced stress concentrations, demonstrating that this design is less susceptible to stress concentrations within the crown. However, because of the transfer of the preload through the abutment screw to abutment contact, changing FP is more influential on this hex system than FM. Overall FM is more influential on the stress within the crown for the external-hex system and FP is more influential on the internal-hex system.

CONCLUSION

This research is a pilot study aimed at offering an initial understanding of the stress distribution characteristics in the crown under different loading conditions. Realistic geometries, material properties, loading and support conditions for the implant system were used.

The geometrical design of the external-hex system tends to induce stress concentrations in the crown at a distance of 2.89mm from the apex. At this location FP also affects the stresses, so that the compressive stresses on the right hand side of the crown are increased. The internal-hex system has reduced stress concentrations in the crown. However, because the preload is transferred through the abutment screw to the abutment contact, changing FP has greater effect on this hex system than FM. Overall FM is more influential on the stress within the crown for the external-hex system and FP is more influential on the internal-hex system.

Future recommendations include the evaluation of other implant parameters such as the implant wall thickness and thread design. Ultimately, all implant components can be understood in terms of their influence on the stress produced within the implant itself. ACKNOWLEDGEMENTSThis work was made possible by the collaborative sup-port from Griffith’s School of Engineering and School of Dentistry and Oral Health. A special thank you goes to Messires John Divitini and Ian Kitchingham from Neoss Limited for their continual contribution.

REFERENCESCapodiferro S, Favia G, Scivetti M, De Frenza G, Grassi R. Clini-cal management and microscopic characterisation of fatique-induced failure of a dental implant. Case report. Head and Face Medicine. 2006;22(2):18.

Curtis AR, Wright AJ, Fleming GJ. The influence of simulated mastica-tory loading regimes on the bi-axial flexure strength and reliability of a Y-TZP dental ceramic. Journal of Dentistry. 2006;34(5):317-325.

DeTolla DH, Andreana S, Patra A, Buhite R, Comella B. Role of the finite element model in dental implants. Journal of Oral Implantology. 2000;26(2):77-81.

Gehrke P, Dhom G, Brunner J, Wolf D, Degidi M, Piattelli A. Zir-conium implant abutments: fracture strength and influence of cyclic loading on retaining-screw loosening. Quintessence International. 2006;37(1):19-26. Geng JP, Tan KB, Liu GR. Application of finite element analysis in im-plant dentistry: a review of the literature. Journal of Prosthetic Dentistry. 2001;85(6):585-598.

http://www.3i-online.com.htm (accessed 12th July 2006).

Huang HM, Tsai CM, Chang CC, Lin CT, Lee SY. Evaluation of load-ing conditions on fatigue-failed implants by fracture surface analysis. International Journal of Oral & Maxillofacial Implants. 2005;20(6):854-859.

Khraisat A, Stegaroiu R, Nomura S, Miyakawa O. Fatigue resistance of two implant/abutment joint designs. Journal of Prosthetic Dentistry. 2002;88:604-610.

Khraisat A. Stability of implant-abutment interface with a hexagon-me-diated butt joint: failure mode and bending resistance. Clinical Implant Dentistry Related Research. 2005;7(4):221-228.

Maeda Y, Satoh T, Sogo M. In vitro differences of stress concentrations for internal and external hex implant-abutment connections: a short communication. Journal of Oral Rehabilitation. 2006;33:75-78.

Merz BR, Hunenbart S, Belser UC. Mechanics of the implant-abutment connection: an 8-degree taper compared to a butt joint connection. In-ternational Journal of Oral & Maxillofacial Implants. 2000;15:519-526.

Neoss Limited (2006) Neoss Implant System Surgical Guidelines, UK.

Strand7 Pty Ltd (2004) Strand7 Theoretical Manual, Sydney, Australia.

van Staden RC, Guan H, Loo YC, Johnson NW, Meredith N. Stress Evaluation of Dental Implant Wall Thickness using Numerical Tech-niques. Applied Osseointegration Research. 2008 (In Press).


54

Comparison of early bacterial colonization of PEEK and titanium healing abutments using real-time PCR

Stefano Volpe1, Damiano Verrocchi2, Peter Andersson2, Jan Gottlow3, Lars Sennerby2,3

1Private Practice, Rome, Italy2Private Practice, Fiera di Primiero and Feltre, Italy3Dept Biomaterials, Inst Clinical Sciences, Sahlgrenska Academy, Gothenburg University, Sweden

Thise study found no differences in total bacterial load at PEEK or titanium Neoss healing abutment surfaces or their adjacent peri-implant pockets/sulci. The number of periodontal pathogens was low and there was no significant difference between the abutment materials. The findings support the use of PEEK as abutment material in dental implant care.

INTRODUCTION

PEEK (polyetheretherketone) is a synthetic polymer with high biomechanical strength and inert chemical properties, which make it attractive for use in industrial and medical applications (for review see Kurtz & De-vine 2007). Healing abutments made of commercially pure titanium have been the “gold standard” in dental implant care for many years. Wound healing studies have documented the formation of a healthy mucosal barrier adhering to the titanium abutment surface after abutment connection (e.g. Berglundh et al 1991). Yet, it has been shown that peri-implant pockets are colonized by bacteria, including periodontal pathogens, already within 2 weeks after abutment connection (Koka et al 1993, Quirynen et al 2006). The Neoss implant system offers both titanium and PEEK healing abutments.

The aim of the present study was to evaluate the bio-compatibility of PEEK abutments to that of titanium abutments by comparing the bacterial coloniza-tion of the abutment surfaces and the surrounding peri-implant pocket/sulcus using real-time PCR.


Patient and sampling sitesFourteen (14) partially edentulous patients (9 women and 5 men with a mean age of 58 years) with need of im-plant treatment volunteered for the study. All patients were of good general health and had not received any antibiotic therapy within the previous 3 months. Every patient received 2 submerged Neoss implants in one edentulous area. Second stage surgery was performed after 3-6 months of healing. Each patient received one

titanium and one PEEK healing abutment. A chlo-rhexidine rinse regimen was instituted during the first postoperative week. Sutures were removed after 7 days.

Bacterial sampling was performed two weeks after abutment connection using a commercial test system (Meridol® Perio Diagnostics, GABA International, Münchenstein, Switzerland). Each test kit contains 4 paperpoints for sampling. All samples were taken at the distal surface of the abutments in order to stay as far as possible away from the neighbouring anterior teeth. Two paperpoints were used at each titanium and PEEK abutment site. The first paperpoint was positioned in close contact with the abutment surface at the mucosal margin. The second paperpoint was placed in the peri-abutment sulcus/pocket area after removal of the abutment. Each paperpoint was held in position for 10 seconds and immediately thereafter placed in a test tube (accompanying the test kit). The samples were sent to a specialized microbiological laboratory (Carpagen GmbH, Münster, Germany).

FIgure 1. Clinical photograph showing the PEEK (left) and titanium (right) healing abutments


55

Real-time PCRReal-time polymerase chain reaction (PCR) by the Meridol® Perio Diagnostics (GABA International, Münchenstein, Switzerland) detects and quantifies six periodontal pathogens (A. actinomycetemcomitans, F. nucleatum ssp., P. gingivalis, P. intermedia, T. for-sythensis and T. denticola) and the total bacterial load. Statistical analysis The Wilcoxon signed rank test was used to calculate the significance of the differences found between numbers of bacteria at titanium and PEEK abutment sites.

RESULTS

The study involved 28 healing abutments (14 PEEK and 14 titanium abutments) in 14 patients. At the day of bacterial sampling all sites showed clinically healthy tissue conditions (Fig. 1 and 2).

The total bacterial load at the mucosal margin was on the average 13*106 (median 5*106) at ti-tanium abutments and 14*106 (median 7*106) at PEEK abutments (table 1). The correspond-ing value for the peri-implant pocket/sulcus was 9*106 (median 2*106) at titanium abutments and 3*106 (median 1*106) at PEEK abutments (table 2). The differences were not statistically significant.

The total number of the 6 periodontal pathogens detectable by the PCR test was on the average 18*103 (median 1*103) at titanium abutments and 14*103 (median 2*103) at PEEK abutments (table 3). The corresponding value for the peri-implant pocket/sulcus was 23*103 (median 2*103) at titanium abutments and 75*103 (median 8*103) at PEEK abutments (table 4). No sample showed ≥106 of periodontal pathogens. The differences between titanium and PEEK abutments were not statistically significant.

DISCUSSION

The Neoss PEEK abutments showed similar amounts of bacterial adhesion as the Neoss titanium abutments when evaluated using real-time PCR. This method is very sensitive since both living and dead bacteria are detected, which makes it more accurate than cultiva-tion. The detection limit for each periodontal pathogen is as low as 100 bacteria (Jervøe-Storm et al 2005).

Hultin and coworkers (2002) described microbio-logical findings and host response in partially eden-tulous patients with peri-implantitis. DNA-probe analysis sensitive for A. actinomycetemcomitans, P. gingivalis, P. intermedia, T. forsythensis and T. den-ticola found presence of periodontal pathogens at healthy as well as at diseased implant sites. However, only around implants with peri-implantitis were all 5 species recovered in amounts ≥106 of the target bacterial cells in each sample. No sample in the present study showed ≥106 of periodontal pathogens indicating healthy conditions at the tested sites.

Figure 2. Clinical photograph of the peri-implant mucosa after removal of the healing abutments

Patient Titanium PEEK

1 470 000 2 000 000

2 21 000 000 720 000

3 56 000 000 60 000 000

4 2 100 000 47 000 000

5 50 000 000 16 000 000

6 180 000 190 000

7 12 000 000 3 000 000

8 6 000 000 24 000 000

9 4 600 000 12 000 000

10 2 300 000 790 000

11 6 300 000 9 800 000

12 4 200 000 5 000 000

13 12 000 000 12 000 000

14 1 500 000 2 200 000

Mean 12 760 714 13907 143

Median 5 300 000 7 400 000

Min 180 000 190 000

Max 56 000 000 60 000 000

Sum 178 650 000 194 700 000

p = 0,507

Table 1. Total bacterial load - surface


56


1 4 400 000 220 000

2 27 000 000 7 800 000

3 1 500 000 16 000 000

4 10 000 000 3 000 000

5 39 000 000 550 000

6 370 000 1 300 000

7 650 000 310 000

8 1 600 000 430 000

9 1 400 000 2 600 000

10 1 700 000 170 000

11 1 000 000 6 700 000

12 1 600 000 190 000

13 150 000 1 200 000

14 31 000 000 4 300 000

Mean 8669286 3 197 857

Median 1 600 000 1 250 000

Min 150 000 170 000

Max 39 000 000 16 000 000

Sum 121 370 000 44 770 000

p=0,158

Table 2. Total bacterial load - sulcus The f ind ings o f the p re s en t s tudy sup-por t cont inued invest igat ion of PEEK as an abutment material in dental implant care.

ACKNOWLEDGEMENTThe authors want to thank Dr. Emanuele Leoncini , Medica l Facul ty, Univers i ty La Sapienza, Rome, Italy, for the statistical analyses.

REFERENCESBerglundh T, Lindhe J, Ericsson I, Marinello CP, Liljenberg B, Thomsen P. The soft tissue barrier at implants and teeth. Clin Oral Implants Res 1991;2:81-90.

Hultin M, Gustafsson A, Hallström H, Johansson LA, Ekfeldt A, Klinge B. Microbiological findings and host response in patients with peri-implantitis. Clin Oral Implants Res. 2002;13:349-58.

Jervøe-Storm P-M, Koltzscher M, Falk W, Dörfler A, Jepsen S. Compari-son of culture and real-time PCR for detection and quantification of five putative periodontopathogenic bacteria in subginvial plaque samples. J Clin Periodontol 2005;32:778-783.

Koka S, Razzoog M, Bloem T, Syed S. Microbial colonization of dental implants in partially edentulous subjects. J Prosthetic Dentistry 1993;70:141-144.

Kurtz S, Devine J. PEEK biomaterials in trauma, orthopedic, and spinal implants. A review. Biomaterials 2007;28:4845-4869.

Quirynen, M; Vogels, R; Peeters, W; van Steenberghe, D; Naert, I; Haffajee, A. Dynamics of initial subgingival colonization of ‘pristine’ peri-implant pockets. Clin Oral Implants Res. 2006;17:25-37.Patient Titanium PEEK

1 1 300 1 490

2 1 450 1 310

3 177 040 145 600

4 2 710 7 400

5 750 750

6 600 600

7 1 380 900

8 920 5 590

9 600 1 050

10 750 3 330

11 1 800 11 790

12 1 170 1 700

13 2 100 6 500

14 53 630 2 410

Mean 17 586 13 601

Median 1 340 2 055

Min 600 600

Max 177 040 145 600

Sum 246 200 190 420

p=0,388

Table 3. Total amount of periodontal pathogens - surface


1 6 960 6 350

2 2 000 8 700

3 15 400 895 550

4 20 760 25 530

5 1 340 13 950

6 600 850

7 600 600

8 900 750

9 4 350 67 750

10 5 210 600

11 900 750

12 1 100 10 220

13 750 14 130

14 261 390 1 950

Mean 23 019 74 834

Median 1 670 7 525

Min 600 600

Max 261 390 895 550

Sum 322 260 1 047 680

p = 0,133

Table 4. Total amount of periodontal pathogens - sulcus


57

Survival rate, fracture resistance and mode of failure of titanium implants in clinical function and dynamic loading.

Neil Meredith1,2 and Fredrik Engman2

1University of Bristol, Bristol, UK2Neoss Ltd, Harrogate, UK

The possible causes of implant fracture under dynamic loading conditions are discussed with regard to the in-vivo clinical behaviour and in-vitro testing.

INTRODUCTIONFracture of endosseous dental implants during place-ment or function will have serious implications for pa-tients. Fracture of implants during insertion may occur as the insertion load exceeds the fracture strength of the implant. Such a failure is most unlikely to be the result of clinical misuse and is most probably due to an error in design or material selection. Errors in manufacturing and flaws in materials may also contribute to failure. A design error is likely to involve a number of components whereas manufacturing and material errors may be lim-ited to a single component or batch of raw material.

Failures at placement are generally noticed clinically at the time. However, should the clinician fail to notice a crack or flaw and the component was then incorporat-ed in a structure then potentially serious bone loss and clinical complications may arise which may not be ap-parent for some months or even years post restoration.

The other mechanism of failure for implant compo-nents; fixtures, abutments and screws is fatigue failure. This occurs as a result of cyclic functional loading; the magnitude of which may be well below the ultimate strengths of the components. Good clinical practice and adherence to sound biomechanical principles of prosthesis design should minimise the risks of fatigue failure, although component design may play a role.

Catastrophic failure can be modelled in-vitro with some confidence by the application of a single load cycle applied by a calibrated testing system until fail-ure occurs. Fatigue loading is much more complex to model in-vivo and it can be difficult to extrapolate to clinical behaviour from such findings. Clinical study of failure specimens may be helpful in identify-

ing possible contributory factors. An international standard (ISO 14801:2003(E)) exists for fatigue test-ing of endosseous dental implants. It is most useful for comparing implants of different designs or sizes.

The published standard carries an important caveat not included in the similar standard for the testing of orthopaedic prostheses: ‘ Whilst it simulates the functional loading of an endosseous dental implant body and its premanufactured prosthetic compo-nents under ‘worst case’ conditions, the standard is not applicable for predicting the in-vivo perform-ance of an endosseous dental implant or prosthesis.’

A number of workers have identified the potential risks of fatigue failure, particularly in single tooth restorations (Marinello et al. 1997; Schmitt and Zarb 1993). Henry et al. (1996), however, reported a very low incidence of fatigue fracture in a prospective five year study. Ac-curate seating of the abutment on an implant is critical (Binon, 2000a). A loss of preload in the abutment screw can result in loosening with wear and fracture (Binon 2000b)(Haak et al. 1995) in the flat-on-flat systems.

If failure of an implant component occurs as a result of fatigue it is essential to understand the mode of failure. The loss of preload in an abutment screw (pos-sibly due to an incorrect torque at placement) may result in loosening of the implant abutment connec-tion and the subsequent development of unexpected bending loads causing fracture of the implant. Clear interpretation of the steps leading to failure by making a thorough examination of fractured components is the only way to correct diagnosis. It therefore becomes clear that simulation and modelling in-vitro of multi-factorial dynamic fracture modes is a great challenge.


58

A number of workers have developed sophisticated and elegant loading machines designed specifically to model masticatory function (Sakaguchi et al. 1986; Bates, Stafford and Harrison, 1975) for the study of wear and less frequently fracture. However, a simple uniaxial loading regimen can be useful in compar-ing implant performance (ISO 14801:2003(E)).

This investigation was prompted by the only reported clinical failure of a specific implant (Neoss system; Neoss Limited, Harrogate, England). The fracture of the im-plant flange occurred and was identified some 3.5 years after placement. The implant was one of three com-prising a bar retained overdenture with electroformed denture. Figure 1 illustrates the fractured implant in-situ. The fractured element was removed and examined under scanning electron microscopy (SEM; Figure 2).As a result of these investigations it was deduced that the most likely aetiology of failure was fatigue fracture . In order to better understand the events contributing to this single failure an in-vitro investi-gation was carried out according to the international standard test method ISO14801. The National Swed-ish Test House (Sveriges Tekniska Forskningsin-stitut) was commissioned to undertake this study.

METHOD

Twenty four implants (Neoss Limited; Harrogate UK) in two groups of diameter 3.5mm and 4.0mm were embedded in a specimen holder with a heat curing resin (Epotek 353ND, CA, Modulus 4.0GN/m2) and at a depth in accordancewith ISO14801. Abut-ments (identification number 10547) were attached to each implant with a gold abutment screw (iden-tification number E8140) and torqued to 32Ncm-1 using a calibrated torque wrench. A hemispherical loading member was attached to each abutment.The fatigue tests were performed in a servo hy-draulic test machine (Instron 1341; Instron Ltd, Uxbridge, England) as constant load amplitude tests with R=0.1 at 15 Hz. R is defined as Fmin divided by Fmax. Fracture was defined as a visible

Figure 1. a.) fractured implant in-situ and b.) overdenture bar construction

Figure 2. Fractured part of implant flange..

Figure 2b. SEM of fracture surface in implant flange..


59

unrecovered deformation, i.e. when a deflection of approximately five millimetres of the loading point occurred. The initial position of the deflection is measured at Fmax. The test set-up is shown in figure 3.

The aluminium cylinder of the test specimen was mounted in a holder of stainless steel so that the force was applied at 30° to the centre axis of the specimen top. The force was applied to the polished loading point of the specimen top using a flat ended cylindri-cal rod with a concave recess mounted onto the load cell. The holder was mounted on the base plate of a cylindrical stainless steel test reservoir and positioned to ensure vertical load application. The horizontal position of the specimen holder was adjusted at the start of the test to ensure the load application was in a vertical line beneath the load cell The loading pa-rameters are listed in Tables 1 and 2. Load cycling was carried out on each specimen for 5million cycles (‘run out’) or until fracture occurred. Fracture was classified as A – of the abutment screw, B – of the abutment screw and crack on the fixture or C- fracture on the fixture and epoxy (Epotek 353ND; Epotek Corp. Ca).

RESULTS

The results from the fatigue tests are tabulated in Table 1 for 3.5mm implants and Table 2 for 4.0mm implants.Results from fatigue tests are plotted as a Wohler graph (Fmax vs. Numbers of cycles in logarithmic scale) for the 3.5mm implants (Figure 4) and 4.0mm implants (Figure 5). In the Wohler graph the formula N = A x Fmax

B describes the relation between Fmax and the numbers of cycles to

Specimen No.

Fmax (N)

Fmin (N)

No. of cycles

Fracture location/remarks

1 240 24 5000000 Runout

2 420 42 66832 FractureB

3 380 38 45705 FractureC

4 330 33 925828 FractureA

5 280 28 5000000 Runout

6 280 28 5000000 Runout

7 280 28 5000000 Runout

8 420 42 39477 FractureB

9 380 38 190138 FractureB

10 330 33 2369670 FractureA

11 330 33 847538 FractureB

Figure 3. Schematic of Test Set Up1. Loading Device 2. Nominal Bone Level3. Abutment 4. Hemispherical Loading Member5. Dental Implant Body 6. Specimen Holder

Table 1. Test results from fatigue fracture of 3.5mm diameter implants

fracture. The constants A and B are calculated as a least square fit in a log-log plot where Fmax is the controlled variable and N (the numbers of cycles to fracture) is the obtained variable. The results are tabulated in Table 3.

Specimen No.

Fmax(N)

Fmin(N)

No. of cycles

Fracture location/remarks

1 350 35 4744655 Fracture A

2 420 42 97422 Fracture A

3 380 13 1712227 Fracture A

4 330 33 5000000 Runout

5 330 33 2106541 Fracture A

6 330 33 5000000 Runout

7 330 33 5000000 Runout

8 350 35 2653842 Fracture A

9 420 42 47775 Fracture B

10 380 38 106771 Fracture B

11 380 38 725278 Fracture A

12 420 42 90652 Fracture A

Table 2. Test results from fatigue fracture of 4.0mm diameter implants


60

Figure 4. Wohler graph for the fatigue test on fixture 3.5 mm diameter

Figure 5. Wohler graph for the fatigue test on fixture 4.0 mm diameter

Fixture Constant A Constant B R2

3.5mm dia. 1.433 x 1043 -14.72 -0.75924

4.0mm dia. 1.524 x 1050 -17.25 -0.76404

Table 3. Linear regression parameters for fractured specimens


61

DISCUSSION

Fatigue testing is the repeated application of cyclic mechanical loading to a specimen with the intention of simulating accelerated function leading to possible fracture. The complex loading conditions and vari-ables in occlusal function make accurate simulation of such loading extremely difficult(De Boever, 1978; Kraus, Jordan and Abrams, 1969; Haraldson, Carls-son and Ingervall, 1979). Indeed the ISO standard (14801:2003(E)) states that whilst the standard is designed to simulate the functional loading of an endosseous dental implant body and its premanu-factured prosthetic components under worst case conditions the standard is not applicable for predicting the in-vivo performance of an endosseous implant.

The purpose and value of such testing may therefore be open to question. However such testing is useful in giving indicators of possible weaknesses in design and production and for comparison between prod-ucts and implant systems (Strub & Gerds, 2003).

The most useful parameter in the results is the lowest value of Fmax at which an implant abutment complex ceases to fail and performs a repeatable runout be-tween specimens. In this study the value was 280N for 3.5mm diameter implants Published data is relatively scarce, however Straumann (2007) have published data comparable data for 3.5mm implants obtained by the ISO 14801 standard ( Figure 6). It is also important to understand the mode and pattern of failure that occurs. Clinically, for example, abutment screw fracture may have less serious implications in screw retained system than in cemented restorations (Drago 1995). Fracture of an implant will nearly al-ways have serious consequences. Clinical examination of the case described previously with the one and only fractured Neoss illustrates a number of useful points.

Figure 7 illustrates the over denture which was be-ing supported by the fractured implant. There were originally four implants placed, two each in the premolar regions. One implant was lost shortly after prosthetic reconstruction and not replaced. Three and a half years later the remaining implant was the one that fractured on the left side. This illustrates significant mesial and distal cantilevers and the clinician described the aetiology as biomechanical overload. Furthermore examination of the fractured implant margin (Figure 8.) indicates burnishing of the top face (Cibirka et al. 2001). This would only occur if the abutment screw had loosened allowing unnoticed movement of the implant and abutment at the interface resulting in high loads being applied to the implant margin leading to the failure (Gratton, Aquilino and Stanford, 2001; Khraisat et al. 2004).

In conclusion fatigue testing of dental implants and components alone is of limited value. In com-bination with clinical data and an understanding of the aetiology of failure useful data can be ob-tained to predict implant performance. The fatigue data for 3.5 and 4.0mm diameter Neoss implant compare favourably with other industry standard implant systems of unspecified size or function.

Figure 7. Overdenture supported by fractured implant in clinical case illustrating biomechanical overload

Figure 6. Comparison of fixture fatigue strengths


62

Figure 8. SEM fractured implant face (a) of burnished implant/abutment interface (b)

ACKNOWLEDGEMENTS

The authors would like to thank Klas Johansson, Technical Manager and Jukka Holappa, Technical Officer of SP Sveriges Tekniska Forskningsinstitut, Building Technology and Mechanics – Solid Mechan-ics and Structures for performing the fatigue testing.

REFERENCESBates JF, Stafford GD, Harrison A. Masticatory function review of the literature. 2. Speed of movement of the mandible, rate of chewing and forces developed in chewing. J Oral Rehabil. 1975;2:281-301.

Binon PP. The external hexagonal interface and screw-joint stability: a primer on threaded fasteners. Qintessence Dent Technol. 2000(a);23:91-105.

Binon PP. Implants and components: entering the new millennium. Int J Oral Maxillofac Implants. 2000;(b)15:76-94.

Cibirka RM, Nelson SK, Lang BR, Rueggeberg FA. Examination of

the implant-abutment interface after fatigue testing. J Prosthet Dent. 2001;85:26&-275.

De Boever JA, McCall WD Jr, Holden S, Ash MM Jr. Functional occlu-sal forces: s; an investigation by telemetry. J Prosthet Dent. 1978;40:326-333.

Drago CJ. A clinical study of the efficacy of gold-tite square abutment screws in cement-retained implant restorations. Int Maxillofac Implants. 1995;10:529-536.

Gratton DG, Aquilino SA, Stanford CM. Micromotion and dynamic fatigue properties of the dental implant-abutment interface. J Prosthet Dent. 2001;85:47-52.

Haack JE, Sakaguchi RL, Sun T, Coffey JP. Elongation and preload stress in dental implant abutment screws. Int J Oral Maxillofac Implants. 1995;10:529-536.

Haraldson T, Carlsson GE, Ingervall B. Functional state, bite force and postural muscle activity in patients with osseointegrated oral implant bridges. Acta Odontol Scand. 1979;37:195-206.

Henry PJ, Laney WR, Jemt T et al. Osseointegrated implants for single-tooth replacement: a prospective 5-year multicenter study. Int J Oral Maxillofac Implants. 1996;11:450-455.

INTERNATIONAL STANDARD; Dentistry — Fatigue test for endos-seous dental implants ISO 14801:2003(E)

Khraisat A, Hashimoto A, Nomura S, Miyakawa O. Effect of lateral cy-clic loading on abutment screw loosening of an external hexagon implant system. J Prosthet Dent. 2004;91:326-334.

Kraus BS, Jordan EJ, Abrams LA. The dentition: its alignment and articulation. In: Kraus BS, Jordan EJ, Abrams LA, eds. A study of the masticatory system. Dental anatomy and occlusion. Baltimore: Williams & Wilkins; 1969:223-237

Marinello CP, Meyenberg KH, Zitzmann N, Luthy H, Soom U, Im-oberdorf M. Single-tooth replacement: some clinical aspects. ,J Prosthet Dent. 1997;9:169-178.

Sakaguchi RL, Douglas WH, DeLong R, Pintado MR. The wear of a posterior composite in an artificial mouth: a clinical correlation. Dent Mater. 1986;2:235-240.

Schmitt A, Zarb GA. The longitudinal clinical effectiveness of osseointe-grated dental implants for single-tooth replacement. Int J Prosthodont. 1993;6:197-202.

Straumann. STARGET 2007;4 : 24-25

Strub JR, Gerds T. Fracture strength and failure mode of five differ-ent single-tooth implant-abutment combinations. Int JProsthodont. 2003;16:167-171.

Applied Osseointegration Researchc/o Professor Tomas Albrektsson

Department of BiomaterialsP.O. Box 412

SE-405 30 GothenburgSweden

Telephone +46 31-786-2945Fax: +46 31-786-2941

email: [email protected] ISSN

165

1-00

70 P

rinte

d in

Sw

eden

. SG

Zet

terq

vist

203

624

applied osseointegration research - …...applied osseointegration research c/o professor tomas...

Documents