Articles
Mithridade DAVARPANAH * Henry MARTINEZ ** Frédéric CHICHE ***
*Département de Parodontologie,
Institut de Stomatologie et de Chirurgie maxillo-faciale,
Hôpital Pitié-Salpêtrière,
Paris-VI, France
**Université de Paris-VII (Denis-Diderot),
Service d'Odontologie de l'Hôtel-Dieu,
Unité d'Implantologie,
Paris, France
***Université de Paris-VII (Denis-Diderot),
Service d'Odontologie de l'Hôtel-Dieu,
Unité d'Implantologie,
Paris, France
The use of machined, commercial grade titanium for dental implants has considerably influenced the therapeutic arsenal for the treatment of various types of tooth loss. The majority of multicentre studies published in the 80s and early 90s evaluated Brånemark implants (machined titanium) and ITI implants (TPS surface coated titanium). The results of these studies confirmed their good medium and long term prognosis (Adell et al., 1981,
The 3i Implant-Innovation® system began at the end of the 80s with the aim of produ cing the most favourable prosthetic components to be used with the Brånemark system. In the 90s, 3i introduced a number of major modifications to their implants and prosthetic components. In the case of implants, technological innovations allowed the introduction of new sizes, implant designs and a new surface texture (Osseotite). This surface is the basis of new concepts: the early loading and the prosthetic predictability. Several teams have undertaken a number of major fundamental studies and longitudinal clinical trials demonstrating the reliability of the 3i system.
The use of machined, commercial grade titanium for dental implants has considerably influenced the therapeutic arsenal for the treatment of various types of tooth loss. The majority of multicentre studies published in the 80s and early 90s evaluated Brånemark implants (machined titanium) and ITI implants (TPS surface coated titanium). The results of these studies confirmed their good medium and long term prognosis (Adell et al., 1981, 1990 ; Albrektsson et al., 1988 ; Engquist et al., 1988 ; Jemt et al., 1989 ; Buser et al., 1990, 1997 ; Zarb et Schmitt, 1990 ; Astrand et al., 1996 ; Bernard et al., 1995). Numerous in vitro and in vivo studies have largely demonstrated the excellent biocompatibility of titanium and its perfect tissue integration (Quirynen et al., 1996). Several studies have confirmed the role of the peri-implant mucosa as a physiological barrier (Berglundh et al., 1991 ; Buser et al., 1992 ; Listgarten et al., 1992).
The 3i system (Implant Innovations) was introduced at the end of the 80s. In the course of the last decade, a number of major surgical and prosthetic innovations have been introduced. They made significant contributions to the technological and clinical developments in oral implantology during the 90s. The clinical application of these surgical ideas confirms the excellent results obtained from in vivo animal experiments. The aim of this article is to present the development of 3i implants and their clinical efficacy.
The 3i system began at the end of the 80s and was concerned with producing the most favourable prosthetic components to be used with the Brånemark implants. The first commercially available implants had a morphology and surface texture similar to the fixtures of the Brånemark system (standard diameter, machined titanium, screw-fitted). Soon, a range of implant diameters was developed (3.25, 3.75, 4.0, 5.0 and 6.0 mm). The objective was to be better able to adapt the choice of implant to the tooth being replaced, the volume of residual bone and the biomechanical conditions pertaining to the patient.
At the beginning of the 90 s, the original morphology of the 3i implant evolved towards the first generation of self-tapping (ST) implants and then rapidly towards the second generation, super-self-tapping (ICE). Successive modifications of implant design have the aim of simplifying the surgical protocol and to improve the primary anchorage of the implant. The apical part of the ICE implant has the shape of a truncated cone which allows a progressive engagement in the bone. The apex of the implant has four cutting flutes that facilitate the placement of the implant and improves primary stability. The elimination of the need to tap a thread in the majority of cases (except where there is very dense bone) simplifies the surgical process.
In 1995, 3i developed a new surface texture (Osseotite) by double-etching with acid (HCl/H2SO4) in order to meet the inadequacies of the smooth surface in unfavourable bony conditions. There were many publications reporting significant failure rates, in some cases above 20 %, where machined implants were placed in low density bone (Engquist et al., 1988 ; Jaffin and Berman, 1991 ; Johns et al., 1992 ; Hutton et al., 1995). Moreover, lower rates of success were reported when smooth surface implants of 10 mm or less were placed in cases of insufficient bone (Quirynen et al., 1991 ; Bahat, 1993 ; Lekholm et al., 1994 ; Wyatt and Zarb, 1998). The Osseotite implant has a dual surface finish in order to meet the soft tissue requirements as well as those of the hard tissues. The coronal part of the implant, from the platform as far as the third thread (3.0 mm) has a smooth surface. From the third thread to the apex, the surface is rough (double acid-etched with HCl/H2SO4). This hybrid implant, with a dual surface finish, ensures the health of the soft tissues in the cervical region adjacent to the smooth part of the implant and optimal bone healing, thanks to the etched surface (Davies, 1998 ; Davarpanah et al., 1999). Indeed, Könönen et al. (1992) observed better attachment of fibroblasts to a smooth surface compared with that to a rough surface. In contrast, osteoblasts have a better cellular attachment to a rough rather than a machined surface (Bowers et al., 1992). The morphology of the implant is similar to the ICE implant. The advantages of this new implant are:
- increase in the area of bone/implant contact;
- increase in the force needed to dislodge the implant;
- absence of an added surface layer;
- elimination of the risk of particles becoming detached from, or erosion, of the surface;
- no surface contamination;
- optimal bone healing against the bone surface;
- reduced time for tissue healing;
- long term compatibility of the smooth surface part of the implant with the peri-implant mucosa;
- secondary failures almost non-existent;
- prosthetic restoration more predictable.
More recently (1997), the non-submerged TG-Osseotite has been developed. This new design of implant has the advantages of being non-submerged and of the hybrid Osseotite implant. Two types of implant design are commercially available according to the thickness of soft tissue (collar height 1.8 or 2.8 mm). The non-submerged part of the implant is flared and the platform is 4.8 mm diameter. There is an 8° morse cone in the internal part of the collar that allows perfect fit for the screwed post. This implant is available in three different diameters, 3.25, 4.0 and 5.0 mm.
In 1998, a truncated cone implant was developed (Osseotite XP). This design is narrow at the apical part (2.6 mm), has a body of 4.0 mm diameter and a collar of 5.0 mm. This design is also available in smaller sizes (body 3.25 and collar 4.1 mm) as well as a larger size (body 5.0 mm and collar 6.0 mm). XP implants have a hybrid surface, identical to Osseotite. There are many clinical situations where this shape has both surgical and prosthetic advantages.
From the surgical point of view, a large collar allows optimal implant stability and meets the prosthetic requirements for the emergence profile. Where there is a buccal concavity at the edentulous site, the correct choice of diameter for the body of the implant avoids having to manage fenestration of the implant or the requirement for a bone graft. The cervical flaring of the XP implant improves the primary mechanical anchorage within the crestal bone. Where the bone is of poor quality, this mechanical characteristic reduces the amount of micro-movement of the implant and complements the biological advantages provided by the Osseotite surface.
From the prosthetic viewpoint, the large collar of the XP implant reduces the biomechanical risk factors (such as screw fracture, unscrewing of components) thanks to the increased seating area of the prosthetic component and improved aesthetics (emergence profile). This implant exploits the emergence profile concept developed by 3i in the early 90s. The aim of this concept is to adapt the diameter of the implant to the diameter of the future prosthetic tooth, by judicious choice of the diameter of the body of the implant, its collar and the abutment. This new implant allows a progressive flaring of the various implant components. In the posterior segments, the 5.0 to 6.0 mm collar of the XP implants restricts those prosthetic reconstructions where the embrasures are poorly aligned. These latter cases show an anatomy that is not very natural and often lead to problems of food packing.
In the chapter « Selection of the metal and surface characteristics » in the book by Brånemark et al. (1988), Kesemo and Lausmaa stated: « The surface roughness of implants will have different consequences according to the geometry involved. […] a surface that is rough or porous may be advantageous because, from a biomechanical point of view, it will provide a good distribution of forces. […] the texture of the surface may influence the biology at the interface because from the point at which the curvature of the roughness corresponds to the size of the cells and large macromolecules, the latter can penetrate the zone concerned. » The significance of the surface texture has been studied for the last three decades. Various research studies have analysed various surface irregularities showing improved tissue healing. Predecki et al. (1972) observed rapid bone growth and good mechanical adhesion to an irregular implant surface. Kirsch and Donath (1984) obtained more rapid bone apposition on the TPS surface. A better blood clot forms on a rough surface compared with a smooth one. Several teams have shown evidence of improved adhesion of osteoblasts on rough surfaces (Buser et al., 1991 ; Keller et al., 1994; Larsson et al., 1996 ; Mustafa et al., 1998 ; Wennerberg et al., 1998). Thomas et al. (1987) believe that a rough surface considerably increases the surface area of the implant and consequently the amount of bone-implant contact. Brunette (1988) found an increased cellular interaction in the presence of a modified implant surface. Bowers et al. (1992), in a histological study, confirmed a significant increase in the attachment of bone cells to a rough surface. The inadequate adhesion of the fibrin mesh to a smooth titanium surface increased the formation of cavities adjacent to the implant. These spaces determined the percentage of direct bone-implant contact. Carlsson et al. (1988) reported that the forces needed to dislodge implants was higher with implants having a rough surface compared with those with a smooth surface. Several teams have recently confirmed that higher torsional forces are needed to dislodge implants with a rough surface (Wennerberg et al., 1995, 1996 ; Klokkevold et al., 1997 ; Pebé et al., 1997 ; Buser et al., 1998 ; Cordioli et al., 2000 ; Gotfredsen et al., 2000). Over the last decade, many researchers have evaluated the percentage of bone-implant contact with different surface textures in animals. Buser et al. (1991) analysed, in pigs, the percentage of direct bone-implant contact for different types of surface: smooth, sandblasted Al2O3, hydroxyapatite, TPS, acid etched with HF/HNO3 and with HCl/H2SO4. After hydroxyapatite, the highest percentage of bone-implant contact was found with SLA (sandblasted and acid etched with HCl/H2SO4), where there was 52 % to 58 % bone-implant contact after 3 to 6 weeks. Levels of 60 % to 70 % have been obtained with hydroxyapatite coatings, however, evidence of resorption of the coating have been reported. Subsequently, several teams have confirmed in animal studies an increase in bone-implant contact in the presence of a rough surface (Ericsson et al., 1994 ; Gotfredsen et al., 1995 ; Pebé et al., 1997 ; Wennerberg et al., 1995, 1996, 1997, 1998). Recently, Trisi et al. (1999) have observed better bone healing adjacent to rough surfaces in humans (in cases with low bone density). They obtained bone-implant contacts of 6.2 %, 3.55 % and 6.7 % in contact with smooth surfaces after 3, 6 and 12 months healing. Adjacent to rough surfaces the bone-implant contact was 58.9 %, 72.9 % and 76.7 % at the same time intervals.
These various in vivo and in vitro studies, in animals and humans, allow us to conclude that in comparison to smooth surfaces, a rough surface enables: an increase in the bone-implant contact, greater resistance to dislodging forces and more rapid bone healing.
The surgical protocol for self-tapping and ICE implants (smooth surface) is similar to those of Brånemark implants. The clinical results with these implants confirm the results obtained with the reference system. Very satisfactory rates of success are obtained with machined, commercial grade, titanium implants, depending on the volume and quality of the bone.
Graves et al. (1994) reported success rates after two years of 95.9 % in a prospective study. Two hundred and sixty-eight wide diameter implants (5.0 and 6.0 mm) were placed in 196 patients. Seventy-eight implants were for single teeth and 129 were for the support of partial prostheses. There were 11 failures up to the re-entry stage which were explained as being due to poor bone quality. Unfortunately, the authors did not publish the statistical analysis.
Lazzara et al. (1996) presented a retrospective multicentre study of 1,969 implants placed in 653 patients with different types of tooth loss over a period of 5 years (evaluation period from 6 to 60 months). The success rates for a total of 1,277 implants were 97 % in the mandible (710 implants) and 93.8 % in the maxilla (567 implants). Vigolo and Givani (2000) reported the results of a retrospective study of 52 implants over 5 years, placed in 44 patients between 1992 and 1994. Two implants were lost by the re-entry stage and a third after the construction of a provisional prosthesis. The success rate 5 years after construction of the prostheses was 94 %.
Davarpanah et al. (2001a), in a multicentre study, evaluated the reliability of self-tapping and ICE implants for the treatment of various types of tooth loss. The overall survival rate 3 years after the prostheses were fitted was 93 % for 602 smooth-surfaced self-tapping implants (92 % for self-tapping implants and 94 % for ICE implants). It is important to indicate that these authors had higher secondary failure rates compared with primary failures. The majority of failures were associated with poor bone density. It was noted that the marginal bone was stable at the level of the first and second threads in 477 (87.8 %) and 25 (4.7 %) of the implants respectively.
Recently, Andersen et al. (2001) reported the results of a 3-year comparative evaluation between 28 standard diameter (3.75 mm) implants placed in 27 patients and 32 implants of 3.25 mm diameter in 28 patients. All were self-tapping, single tooth implants in the maxilla. After 3 years in function, the success rates were 93.8 % for the small diameter implants and 100 % for those of standard diameter. The pattern of marginal bone loss and its level of stability (0.4 mm) was similar for the two groups (table I).
The use of hydrochloric and sulphuric acids to treat the surfaces of machined titanium implants allows the production of a uniform surface texture. Acid-etching is a subtractive process that removes microscopic particles from the implant surface, creating also an irregular surface morphology to the titanium (Davarpanah et al., 1999). The probability of surface contamination and the dissemination of micro-particles in the tissue environment are very much reduced. An analysis of the surface topography of the Osseotite surface after etching with HCl/H2SO4 shows distances of 1 to 3 microns horizontally between the high points of the irregularities and 5 to 10 microns vertically. Awareness of these dimensions is very important. In effect, the bone matrix inserts itself into the 1 to 2 micron pores (Wong et al., 1995). According to Davies (1998), two biological phenomena are at work after placing an implant : osteogenesis in the area and osteogenesis at the point of contact. The microstructure of the surface has a fundamental role in the quality of osseointegration that is achieved. The first phenomenon allows the formation of a network of connective tissue, a « plexus of fibrin », coming from the bony wall facing the implant. In contact osseogenesis, the newly-formed bone makes direct contact with the implant surface, thanks to adhesion of the fibrin coming from the initial blood clot. In this way, the migration of osteogenic cells is facilitated. Thanks to the micro-topography of the Osseotite implant, there is an excellent spreading of the blood (surface wetablility) when placing the implant. Park and Davies (2000) showed an elevated interaction between the red corpuscles and platelets on the Osseotite surface compared with a machined surface. At the same time, mineralization of the matrix enables a higher percentage of bone/implant contact to be obtained (Davies, 1998).
Histological studies by Dziedic et al. (1996) confirmed better bone anchorage with the Osseotite surface compared with a smooth surface. The force necessary to dislodge implants with an etched surface (Osseotite) was greater than that for smooth implants (Pebé et al., 1997; Baker et al., 1999; Cordioli et al., 2000; Klokkevold et al., 2001) (table II). Pebé et al. (1997) showed, in the dog, that dislodging forces are higher with Osseotite implants (66 Ncm) compared those with a smooth surface (50 Ncm). Baker et al. (1999) carried out a comparative study of the forces required to dislodge double-etched implants and smooth surface implants in the rabbit over periods of 1, 2, 3, 4, 5 and 8 weeks. The results confirmed more rapid healing on the Osseotite surface. After 3 weeks the dislodging forces on implants with an Osseotite surface (56.5 ± 12.1 Ncm) was greater than those at 6 weeks for smooth surfaced implants (33.3 ± 3.2 Ncm). Also in the rabbit, Klokkevold et al. (2001) obtained results showing higher dislodging forces after 1, 2 and 3 months for double acid etched (Osseotite) implants. This type of mechanical test (of dislodging forces) confirms to quality of osseointegration.
Also, Davies and Dziedzic (1996) showed a statistically significant increase in osseointegration with the Osseotite surface compared with a smooth surface. This improved quality of osseointegration in explained by the surface topography of the titanium. Several teams have undertaken comparative studies to assess the percentage of direct bone-implant contact in different clinical models with various surface finishes ( table III ). Pebé et al. (1997) undertook the first comparative study in dogs with implants of different surfaces. Cordioli et al. (2000), in a study using rabbits, reported a higher percentage of bone-implant contact with the etched (Osseotite) implant. Abrahamsson et al. (2001), in dogs, obtained a higher percentage of bone-implant contact with Osseotite (72 %) compared with machined implants (58 %). Celletti et al. (2001) undertook a comparative (Osseotite surface and smooth surface) histomorphometric study in rabbits. The mean percentage of bone-implant contact at 15, 30 and 60 days was respectively 29.4 %, 37.7 % and 45.2 % for machined surfaces and for Osseotite surfaces was 39 %, 55.2 % and 67.4 %. Lazzara et al. (1999) confirmed these results in a human histomorphometric study. Implants with a diameter of 2 mm were specially prepared with an etched surface on one side and a smooth surface on the other. They were placed in the posterior maxilla in 11 patients and removed after 6 months' healing. The bone-implant contact was 72.9 % and 33.9 % respectively for the Osseotite and smooth surfaces. The percentage of bone-implant contact after early loading (1 and 2 months) has been studied in the baboon. Vernino et al. (2000) obtained a similar bone-implant contact at 1 month (76.6 %) and 2 months (77.2 %). No failures were reported after 3 months in function. It is interesting to note that there was a good functional and tissue response following early loading. Trisi et al. (2001) have confirmed these results in humans. A histological analysis of 11 double-surface implants (Osseotite and smooth) was undertaken after 2 months' healing in low density bone. A mean bone-implant contact of 18.9 % was obtained with the smooth surface and 47.8 % with the Osseotite surface. These authors considered that the results were due to the osseoconductive properties of the micro-structure of the Osseotite surface.
Since 1995 several European and American teams have undertaken a number of major clinical studies on Osseotite implants in various clinical situations (table IV).
Sullivan et al. (1997) reported on the first multicentre study of 147 Osseotite implants placed in 75 patients. After 3 years the level of success was 96.6 %. Recently, the same authors have published the results from the same study after 6 years. No further failure was reported and the level of crestal bone remained perfectly stable. The authors noted that amongst the 5 failures in the first part of the study, 4 were attributed to one patient with a complex systemic medical condition.
Grunder et al. (1999), in a multicentre study, indicate the predictability of prosthetic success. For these authors, secondary implant failures (after construction of the prosthesis) constitute a real problem of treatment management for the clinician and a psychological problem for the patient. They presented the 2-year results of 219 Osseotite implants placed in 74 patients. Three failures were reported before the second stage surgery where the bone was of normal density. A success rate of 98.6 % was obtained before and 100 % after construction of the prosthesis.
Davarpanah et al. (2001) have recently confirmed this predictability in a prospective multicentre study involving 5 centres. A total of 413 implants were placed in 142 patients. The implants were assessed 3 years after prostheses had been constructed. These authors reported an implant survival rate of 96 % and a prosthetic success of 98.7 %. It is interesting to note that there was a 98.4 % success for 187 short implants (≤ 10 mm). Testori et al. (2001a) evaluated the efficacy of 485 Osseotite implants in 181 patients over 4 years. A total of 355 implants were located in the posterior segments. Six implants were considered to be primary failures and none was lost after the prostheses had been constructed. An overall success rate of 98.7 % over the 4 years was achieved. In the upper and lower posterior segments, success rates of 98.4 % and 99.4 % respectively were obtained.
Khang et al. (1999), in a multicentre study, compared 147 smooth surface ICE implants and 201 Osseotite implants placed in bone of varying qualities in 75 patients. The two types of implant were placed in each subject. The success rate for the 184 Osseotite implants that were followed up was higher (at 96 %) compared with the 135 ICE implants (88 %). A significant difference was found for those placed in low density bone. The failure rate for those placed in the less favourable bone was 3 % for Osseotite implants and 14 % for ICE implants. Davarpanah (2001), in a multicentre study (involving 13 centres), presented 5-year results of 419 self-tapping, 619 ICE and 545 Osseotite implants placed in 528 patients with various types of tooth loss. The overall success rate for the 1,583 implants was 96.5 %. Amongst the 55 failures, 49 of them were those with machined surfaces and 6 were Osseotite implants. The success rate by type was: ICE, 94.8 %, self-tapping, 95.9 % and Osseotite, 98.9 %. These results confirm the importance of surface texture.
Feldman (2000) evaluated the reliability of 71 single-tooth Osseotite implants placed in 59 patients. Fifty-five percent were in posterior segments and 45 % anterior. A conventional two-stage technique was used in this study. A success rate of 98.6 % was found after periods up to 4 years (mean 36 months). Roy and Head (2000) presented results after 3 years of 688 Osseotite mandibular implants placed in 172 patients. Five implants were considered to have failed prior to prosthetic reconstruction. A survival rate after 36 months was reported to be 99.3 %.
In 1998, Lazzara et al. put forward a technique involving early loading of Osseotite implants (fig. 1, 2, 3, 4, 5 and 6).
Their multicentre pilot study analyzed the 1-year results following early loading (2 months) of 429 Osseotite implants placed in normal bone in 155 patients (table V). An implant success rate of 98.5 % was reported for the 429 implants. Amongst the 7 reported failures, only one occurred after the prosthesis was constructed. A success rate of 99.8 % was obtained after 10.5 months of loading. In a 3-year multicentre study, Testori et al. (2002) confirmed the very good prognosis for loading at 2 months, in the posterior segments. A total of 405 Osseotite implants was placed in the molar and premolar regions in 175 patients. Nine failures were reported (6 primary and 3 secondary). After 3 years the authors obtained success rates of 97.5 % in the mandible and 98.4 % in the maxilla. Success rates after construction of the prostheses were 98.9 % and 100 % respectively. As a result, Testori et al. (2002) proposed early loading (after 2 months) in cases of low bone density. Freiberger (2000) presented a multicentre study of 175 TG-Osseotite implants that had been loaded at 2 months, in 80 patients. A survival rate of 97.1 % was obtained after 16 months.
In other clinical situations (implantation following sinus graft), the Osseotite implant shows very good results (table VI, fig. 7, 8, 9, 10, 11, 12 and 13). Wallace (2001) presented the results of a clinical evaluation after 6 years of 594 implants (104 machined and 621 Osseotite) placed in sinus grafts. All 594 were used to support prostheses. The success rates obtained were 95.4 % for the double acid etched implants and 71.2 % for those with a smooth surface. Reich and Lückerath (2001) obtained similar results in a randomised split mouth study comparing 40 Osseotite and 43 Mk IV machined implants in the maxilla. Seventeen patients needing a double sinus lift were grafted with iliac bone + Beta-TCP (50:50). Implants were placed during the same surgical procedure and left to heal for 7-8 months before being used as abutments. All implants had been in function for at least 3 months. A success rate of 85 % was obtained for the Mk IV implants and 95 % for the Osseotite implants.
These pilot studies, analysing the behaviour Osseotite implants with immediate loading with either provisional or definitive prosthetic restoration, show very interesting preliminary results (fig. 14, 15, 16, 17, 18, 19 and 20) (Testori et al., 2001b ; Ibanez, 2002).
The results of basic and clinical research on the Osseotite surface enables us to establish a new protocol : the early loading. Analysis of the available results provides evidence of excellent implant and prosthetic outcomes. The various medium term (3-6 year), multicentre, longitudinal studies have demonstrated that this implant allows:
- better overall success rates than with smooth-surfaced implants;
- a high level of success in low density bone;
- better prosthetic success because of very much reduced levels of secondary failure;
- the possibility of being brought into early function (at 8 weeks);
- a higher percentage of anchorage and a better prognosis for short implants;
- better results in grafted sites.
The 3i system (Implant Innovations) was created in 1988 in order to improve the function and aesthetics of implant-supported prostheses. In the 90s, a considerable number of surgical and prosthetic innovations were proposed. They contributed to major clinical developments in clinical oral implantology. Amongst these innovations, a new surface texture (Osseotite) was developed in 1995. This surface has enabled improvements in the clinical reliability and the prognosis where less favourable bone conditions exist. Thanks to the results obtained with Osseotite implants, new concepts such as early loading and prosthetic predictability are proposed. All the recent medium and long term studies confirm the reliability of 3i implants.
Demande de tirés à part
Mithridade DAVARPANAH, 36, rue de Lubeck, 75116 PARIS - FRANCE.