Risk factors in implant procedures

Introduction

There are two main factors associated with failure of osseointegrated implants. They are, low density bone (type IV) ^{(Enquist et al., 1988} ; ^{Jaffin and Berman, 1991} ; ^{Johns et al., 1992)} and, to a lesser extent, a bone height of less than 10 mm (Friberg, 1996). Special surgical protocols have been developed to allow treatment of these extreme cases which include, amongst other things, guided bone regeneration (GBR) and the use of wide diameter implants. From the published literature, it seems that modification of the original surgical protocol with these two techniques has proved their efficacy. Nevertheless, a number of problems and complications have emerged from their use in routine clinical practice which have not been reported in the studies.

The aim of this article is to describe two types of risk factor : one is practical and is involved with the placing of wide diameter implants ; the other is biological and is associated with the selection of patients to be considered for guided bone regeneration.

Large diameter implants

Wide diameter implants were developed at the beginning of the 1990s to allow surgeons to treat extreme cases ^{(Langer et al., 1993)}. Several indications for their use were proposed : low bone density or volume, replacement of an implant which is mobile at the time of placement or one that has not osseointegrated, restoration of single molars by implant-supported prostheses ^{(Renouard and Riachi, 1994)}.

Despite extensive use of wide diameter implants there have been few papers on the subject. Nevertheless, these implants seem to have a success rate which is comparable to that of standard implants. ^{Davarpanah et al. (1995)} reported a success rate of 96 % over 2 years in 55 implants of 5 mm diameter (without polished necks) (Nobel Biocare^®). ^{Graves et al. (1994)} followed up 266 large implants (3I^®) in 196 patients over 2 years. Eleven implants had not osseointegrated. All were placed in bone type III or IV. Their success rate with this type of implant was 94 % in the mandible and 98 % in the maxilla. All these results are encouraging, considering that the implants were used in unfavourable situations such as low bone density or volume, extreme biomechanical situations (single molar implants). However, it should be noted that ^{Minsk et al. (1996)}, in a retrospective study of 1263 implants, reported a higher failure rate for wide diameter implants compared with those of small or standard diameter. The authors did not suggest an explanation for this.

For the majority of authors, the only criterion of success used was the state of osseointegration. No problem specific to wide implants was reported. Renouard and Arnoux (submitted to publication) have found a success rate of 90 % after one year in 98 consecutive implants (Nobel Biocare^®) in 78 patients. However they discovered crestal bone loss which was greater than with fixtures of standard diameter (3.75 mm). This bone loss occurred either while the implant was covered (8 of the 98 fixtures) (^{fig. 1a} and ^1b), or just after loading. After one year of loading, 21 of the 98 fixtures showed bone loss extending beyond the second thread (^{fig. 2a} and ^2b). When the bone loss occurs before loading, there are no clinical signs, except perhaps for the chance finding of a sinus at the time of second stage surgery. Of the 8 cases of bone loss occurring whilst the implant was covered, 7 were at type I sites, according to the classification of Lekholm et al. (1985).

The peculiarity of the 5 mm implants is the absence of the smooth flange and hence the need to use a countersink drill when preparing the site. It is this particular design of implant, together with the surgical technique which has led to the debate about the cause of bone loss.

In 1996 Wide Platform fixtures were introduced (Nobel Biocare). They have a smooth flange, provide the wide base necessary for a large crown and require a countersink for their placement. Nevertheless, a similar bone loss has been observed (^{fig. 3a}, ^3b and ^3c) whilst they are still covered or shortly after loading. It seems, therefore, that the problem resides in the concept of wide implants, rather than their shape. The insertion of wide diameter implants requires the use of extra drills. Thus, in the Brånemark system, after the 3 mm drill, it is advisable to use an intermediate sized drill of between 3 and 4.3 mm, prior to one of 4.3 mm. When the bone is very dense a drill of 3.7 mm can be used before 4.3 mm. The margins of the prepared bone cavities can be chamfered with a countersink. The fixture is then inserted, after tapping the thread.

The clinical interest in wide diameter implants lies in the fact that the periphery of the implant is more often in contact with cortical bone which is more favourable for primary stability but, equally, makes it more susceptible to mechanical forces. The peripheral speed of a 4.3 mm drill is much greater than that of 3 mm drill. The angular speed of a drill rotating at 1500 rpm is 157 rd/s. The peripheral speed of a 3 mm drill is 471 m/s whilst that of a 4.3 mm drill is 675 m/s, an increase of 30 %. It is possible that this large difference causes a not inconsiderable trauma, especially in dense bone. This could well lead to the development of slowly progressing bone sequestration during the time that the implant is covered or shortly after loading. It would seem to be essential to undertake a modified drilling procedure.

There have been no published studies on this subject. However, it would be prudent not to use a 4.3 mm drill when faced with high density bone. After using the 3 mm drill, a tapered reamer should be used prior drilling out to 3.8 mm. (^{fig. 4a} and ^4b). If the bone is very dense, a tap is used. It is not always necessary to tap a thread to the full depth, only the marginal area need be prepared in this way, in order to facilitate the insertion of the implant. Finally the implant is screwed into place using a hand applicator. This procedure appears to be less traumatic to the bone and may reduce the amount of marginal bone loss which is sometimes observed with wide diameter implants. Unfortunately, there is no drill intermediate in size between 3.0 mm and 3.85 mm which means that the larger size must be used with great care. Whereas high density bone is a favourable factor for standard implants, it is a risk factor for wide diameter implants. Prospective studies are needed to evaluate the reliability of these implants, taking into account not only the rates of osseointegration, but also other factors, such as the stability of the crestal bone. Different surgical protocols need to be developed in order to deal with different clinical situations. In the meantime, wide diameter implants need to be used with great care, especially when placed in dense bone.

Influence of tobacco on guided bone regeneration

Guided bone regeneration (GBR) has its origin in Guided Tissue Regeneration (GTR) used in periodontics. After having been the subject of much controversy, GBR now appears to be a reliable technique, both for the augmentation of alveolar ridges before implant placement as well as for the treatment of bone defects around osseointegrated implants. Many authors have shown that it is possible to obtain new bone formation under a membrane as long as the correct protocol is strictly adhered to ^{(Dalhin et al., 1989} ; ^{Becker and Becker, 1990} ; ^{Buser et al., 1990)}. However, GBR is still considered by the majority of practitioners to be a risky procedure with few indications. There is a large discrepancy between the small number of reported complications and failures and the problems actually encountered in daily practice. This difference can be explained by a lack of appreciation of the risk factors associated with this technique. In ^{1997, Fugazzoto et al}. published a study involving 1507 sites treated with GBR with a success rate of 97 %, which is comparable to the majority of other studies. The criteria for success and failure are clearly described but, in contrast, the criteria for patient inclusion are not given. However, one can imagine that because of their extensive experience, the authors only applied the techniques of GBR to those patients who they considered to be at minimum risk of complications. It is possible that smoker patients were not integrated in their evaluation. Of all the risk factors, smoking is certainly a negative influence for successful bone regeneration. That is why we have undertaken a retrospective study the influence of smoking on the success rates of Guided Bone Regeneration.

Materials and methods

A retrospective study was undertaken on patients treated in a private practice, between January 1990 and September 1993. Sixty nine patients (26 male and 43 female) with a mean age of 47 years (18-75 years) were followed up every 3 months for 3 years. Of the 69 patients, 13 were smokers. The following observations can be made from the records of preliminary clinical examinations : there was no selection of patients and they maintained the follow-up visits every 3 months from the surgical procedure to the end of the study. The Gore Tex^® membranes were placed according to the precise protocol for their use in GBR. The membranes remained buried for a mean period of 23 weeks (4-60 weeks). Thirteen patients were categorised as smokers as their consumption was in excess of 15 cigarettes per day ^{(Ah et al., 1994)}. The mean size of the defects was 5.04 mm (2-13 mm) and was measured by the number of exposed threads not covered by bone after implants placement. These data were confirmed by photographs enlarged to 1 : 1. The membranes were placed in order to treat the defects with dehiscences in 45 cases and fenestrations in 24 cases.

The data recorded were : sex, age, periodontal condition, hygiene, type of edentulism, number of threads exposed, temporary treatment given, possible exposure of membrane, presence of suppuration, period of burial of the membrane.

Two parameters were used to describe the tissue response in GBR technique : a quantitative evaluation giving the rate of coverage of the implant by new formed bone and the qualitative evaluation of the regenerated bone (solid or soft bone). All these data were collected when the implants were placed or at the second step procedure. In some cases it was possible to remove a sample of the new bone from above the cap for histological examination.

Criteria of success (^{fig. 5a}, ^5b and ^5c) :

- Quantitative : Complete coverage of all the threads which were previously exposed.

- Qualitative : Regenerated bone with a hard and pearly-white or corticated aspect and free of suppuration.

Criteria of failure (^{fig. 6a}, ^6b, ^6c and ^6d) :

- Quantitative : Two or more threads uncovered by newly formed tissue.

- Qualitative : Newly formed tissue with the appearance of a fibrous tissue or a granulation tissue.

All the quantitative and qualitative data collected in this study can be expressed as percentages of frequency of occurrence. Statistical analysis was done with the χ² test.

Results

Of the 69 implants placed, 53 were considered as successful and 16 were failures, using our qualitative and quantitative criteria. The need to consider the alignment of the prosthesis was the principal cause of the initial exposure of the threads on some of the Brånemark implants. The mean length of implant exposure at the time of surgery was 5.04 mm. The membranes used were GTAM (Gore Tex^®). A standard technique was used for placing the membranes.

The data are provided in two distribution tables.

In ^{table I} it can be seen that in the non-smokers (NS), 84 % showed the characteristics of a successful result, whereas 54 % of smokers were categorised as failures. Similarly, the risk of suppuration was three times greater in smokers compared with non-smokers (38 % against 11 %) ^{(Buser et al., 1990} ; ^{Jovanovic et al., 1992} ; ^{Jovanovic and Giovannoli, 1992)}.

By comparison, it seems that in the cases of smokers, 54 % of the membranes became exposed compared with 21 % of non-smokers (p = 0.0001). Equally, it should be noted that failure occurred in 75 % of cases where the membranes were exposed in smokers and non-smokers (^{table II}).

Discussion

The retrospective study involved a collection of patients treated over three years in private practice in order to test the tissue response in smokers and non-smokers. Our criteria of success were new bone which was hard, pearly-white and/or corticated, and total coverage of all the threads previously exposed. It has been possible to show that in non-smokers there was complete coverage of the membrane in 79 % of cases after healing. As long as the membrane remained covered for the whole of the healing period, 87 % of cases were successful, according to our criteria. In comparison, there was exposure of the membrane in 54 % of the smokers which led to failure in three quarters of them. It was usual to find suppuration in all cases of exposure because it was the suppuration which led to the exposure (p = 0.0001).

Smokers present a risk of membrane exposure during the healing period which is clearly greater than in non-smokers (54 % for smokers compared with 21 % for non-smokers). According to ^{Mosely et al. (1978)}, tobacco use interferes with wound healing. There are complications and healing is compromised in surgical wounds following plastic or vascular surgery in smokers ^{(Sherwin et al., 1990} ; ^{Rees et al., 1984} ; ^{Siana et al., 1989)}. In the dental domain, smoking increases the risk of dry socket ^{(Sweet and Butler, 1978} ; ^{Meecham et al., 1988)}. ^{Clarke et al. (1981)} studied the effects of nicotine injection in rabbits and observed vasoconstriction with a severe restriction in gingival blood flow. ^{Johnson et al. (1989)} showed that in rats, morphological changes occurred in the microvasculature of the oral mucosa after the injection of nicotine. One can explain the diminution of gingival blood flow by the vasoconstrictive effect of nicotine ^{(Bergstrom and Floderus-Myrhe, 1983} ; ^{Preber and Bergstrom, 1985a}, ^b, ¹⁹⁸⁶ ; ^{Goultschin et al., 1990)}. Many authors have observed that the normal function of polymorphonuclear leucocytes is altered and compromised in smokers ^{(Kenney et al., 1977} ; ^{Mac Farlane et al., 1992)}. All these effects contribute to reduced healing. From the periodontal point of view, recent studies have highlighted the role of tobacco in affecting the severity of periodontitis ^{(Goultschin et al., 1990} ; ^{Bergstrom et al., 1991} ; ^{Bridges et al., 1991)}. According to ^{Mac Farlane (1992)} there is a correlation between tobacco consumption and the incidence of refractory periodontitis. Within the context of periodontal treatment, smoking compromises healing after muco-gingival surgery ^{(Miller, 1987)}. ^{Bain and Moy (1993)} studied the correlation between the failure of implants and tobacco consumption. Of 2 194 Brånemark implants placed in 540 patients between 1984 and 1990, the overall failure rate was 5.92 % which is comparable to other studies. By contrast they had a failure rate of 11.28 % in smokers compared with 4.76 % in non-smokers. At the present time, all studies demonstrate the adverse effect of smoking on healing, be it from a local or systemic point of view.

It seems, therefore, that suppuration which is secondary to the placement of an expanded polytetrafluoroethylene (e-PTFE) membrane can occur in both smokers and non-smokers. The excellent biocompatability of PTFE has been largely demonstrated in studies where the material has been totally buried in connective tissue ^{(Robert et al., 1993)}. It is likely that local biological changes occur which affect the integrity of gingival epithelium above the membrane, leading to the eventual colonisation of the exposed membrane by pathogenic bacteria from the oral cavity. These changes can include tissue necrosis provoked by a reduction in vascularity of the sub-epithelial tissues, due to smoking, and a thinning of the connective tissue above the membrane due to the presence of the Gore Tex^®.

Most of the failures in our study were related to premature exposure of the membrane. Furthermore, this explanation has frequently been implicated in cases treated by GTR, with membrane exposure occurring in 53 % (^{Nygaard-østby et al., 1996}) to 72.5 % ^{(Tonetti et al., 1993)} of cases. However the coverage of the membranes is better in those cases treated by GBR than it is in GTR.

It seems that, in smokers, one can clearly demonstrate greater exposure of the membranes during the healing period. This exposure of the membrane, due to biological changes in the local tissues, is usually accompanied by suppuration. There is also a distinct increase in failure rates (75 %) when the membrane is exposed (^{table II}). One can conclude that smokers present an increased risk of exposure during healing.

However all the membranes which remained covered after the normal period of healing in smokers were successful.

When considering the failures, it is interesting to note that the mean time for which the membranes were retained in place was 14 weeks and that most of the successful cases were retained for 24 weeks. This is confirmed since only 25 % of membranes which became exposed prematurely led to success, against 87 % for the membranes which remained covered until the second step procedure Premature exposure of the membrane seems, therefore, to lead to a clear reduction in success rates compared with those membranes which remain covered. This accords with the view of the majority of authors, that failure of GBR is usually due to exposure of the membrane during the healing phase. Premature exposure of the membrane impedes the process of formation of new bone and can even lead to bone loss ^{(Dahlin et al., 1989} ; ^{Becker and Becker, 1990)}. Also, ^{Lekholm et al. (1993)}, showed in dogs, that when the membrane was placed for a planned period of 4 weeks, there was less bone formation when compared with those cases where the membranes were placed for 16 weeks (42 % against 100 %). Inflammatory changes subsequent to premature exposure of the membrane compromises the process of bone regeneration, even leading to resorption of existing bone ^{(Warrer et al., 1991)}. During the healing phase, the unexposed membranes do not show cultivable microorganisms and are associated to new peri-implant bone formation. Membranes exposed prematurely during the healing phase, are associated with a significantly elevated percentage of cultivable microorganisms associated with bone loss (P. gingivalis, B. Forsythus, P. intermedia, Fusobacterium, staphylococci and streptococci) ^{(Nowzari and Slots, 1995)}. Also, this study has shown that there was a higher percentage of prematurely exposed membranes in patients suffering from periodontal diseases.

Throughout this study, we have attempted to show the incidence and the importance of risk factors in Guided Bone Regeneration. Tobacco is a major risk factor. The lack of appreciation of these factors and not taking them into account, will inevitably lead to higher failure rates than those reported in the literature.

Conclusion

The efficacy of wide diameter implants has been demonstrated in the treatment of patients where there is reduced bone volume. However, a number of failures and complications associated with the use of these implants have been revealed. It seems that these problems are associated with the techniques which have been developed for their placement, especially the use of the larger diameter drills. The increase in diameter of these instruments is associated with a higher peripheral rotational speed and can perhaps lead to a more important trauma. This phenomenon could explain the form of the lesions which have been observed around wide diameter implants.

The results of our retrospective analysis, which took place over 3 years, showed that failure of regeneration followed premature exposure of the membranes. This exposure occurred predominantly in smokers. There is a statistically significant correlation between smoking and membrane exposure, as well as between smoking and reduced regeneration. Therefore, it is apparent that, when considering Guided Bone Regeneration it is essential to take into account risk factors such as smoking which reduce the success rate is significantly, even if the surgical procedure is carried out perfectly.

BIBLIOGRAPHIE

• AH MKB, JOHNSON GK, KALDHAL WB, PATIL KD, KALKWARF KL. The effect of smoking on the response to periodontal therapy. J Clin Periodontol 1994;2:91-97.
• ALBREKTSSON T, ZARB GA, WORTHINGTON P, ERIKSSON A. The long-term efficacy of currently used dental implants : A review and proposed criteria of success. Int J Oral Maxillofac Implants 1986;1:11-25.
• ABRAHAMSSON I, BERGLUNDH T, WENNSTRÖM J, LINDHE J. The peri-implant hard and soft tissues at different implants systems. A comparative study in the dog. Clin Oral Impl Res 1996;7:212-219.
• BAHAT O. Osseointegrated implants in the maxillary tuberosity : report on 45 consecutive patients. Int J Oral Maxillofac Implants 1992;7:459-467.
• BAIN CA, MOY PK. The association between the failure of dental implants and cigarette smoking. Int J Oral Maxillofac Implants 1993;8:609-615.
• BATTAGLIA G, DI MARCO F, PICCOLI A, VIANELLO F, FIRANATI F, NACCARATO R. Clinical makers of slow healing and relapsing gastric ulcer. Gut 1987;28:210-215.
• BECKER W, BECKER BE. Guided tissue regeneration for implants placed into extraction sockets and for implant dehiscences : surgical techniques and case reports. Surgical and restorative advantages. Int J Periodont Rest Dent 1990;377-391.
• BERGSTROM J, ELIASSON S, PREBER H. Cigarette smoking and periodontal bone loss. J Periodontol 1991;62:242-246.
• BERGSTROM J, FLODERUS-MYRHE B. Co-twin study of the relationship between smoking and some periodontotal disease factors. Community Dent Oral Epidemiol 1983;11:113-116.
• BUSER D, BRAGGER U, LANG NP, NYMAN S. Regeneration and enlargement of jaw bone using guided tissue regeneration. Clin Oral Implant Res 1990;1:22-32.
• BRIDGES RB, SAXE SR, MCKEAN HE. Cigarette smoking and diabets effects on loss of attachment (abstract 2601). J Dent Res 1991;70:591.
• CLARKE NG, SHEPHARD BC, HIRSCH RS. the effects of intra-arterial epinephrine and nicotine on gingival circulation. Oral Surgery 1981;52:577-582.
• DAHLIN C, LINDE A, GOTTLOW J, NYMAN S. Generation of new bone around titanium implants using a membrane technique. An experimental study in rabbits. Int J Oral Maxillofac Implants 1989;4:19-25.
• DAVARPANAH M, MARTINEZ H, TECCUCIANU JF. Les implants de large diamètre. Implant 1995;1:289-299.
• ENQUIST B, BERGENDAL T, KALLUS T, LINDEN U. A retrospective multicenter evaluation of osseointegrated implants supporting overdentures. Int J Oral Maxillofac Implants 1988;3:129-134.
• FRIBERG B, JEMT J, LEKHOLM U. Early failures in 4641 consecutively placed Brånemark dental implants : a study from stage 1 surgery to the connection of the completed prostheses. Int J Oral Maxillofac Implants 1991;6:142-146.
• FUGAZZOTTO P, SHANAMAN R, MANOS T, SHECTMAN R. Régénération osseuse guidée autour d'implants en titane : compte rendu du traitement de 1503 sites avec réentrées. Rev Int Parodont Dent Rest 1997;3:293-299.
• GIOVANNOLI JL, RENOUARD F. Choix du mode d'espacement de la membrane dans le traitement des pertes de substance par régénération osseuse guidée. J Parodontol Implant Orale 1995;14:179-192.
• GOULTSCHIN J, COHEN HDS, DONCHIN M, BRAYER L, SOSKHOLNE WA. Association of smoking with periodontal treatment needs. J Periodontol 1990;61:364-367.
• GRAVES SL, JANSEN CE, SIDDIQUI AA, BEATY KD. Wide diameter implant : indications, considerations and preliminary results over two-years period. Aust Prosthet J 1994;8:31-37.
• JAFFIN R, BERMAN C. The excessive loss of Brånemark fixtures in type IV bone : a 5-year analysis. J Periodontol 1991;62:2-4.
• JOHNS R, JEMT T, HEATH MR. A multicenter study of overdentures supported by Brånemark implants. Int J Oral Maxillofac Implants 1992;7:513-522.
• JOVANOVIC SA, GIOVANNOLI JL. Application des principes de la régénération tissulaire guidée au traitement des défauts osseux péri-implantaires. J Parodontol 1992;11:29-44.
• JOVANOVIC SA, SPIEKERMANN H, RICHTER EJ. Bone regeneration around titanium dental implants in dehisced defect sites. A clinical study. Int J Oral Maxillofac Implants 1992;7:233-245.
• JOHNSON GK, FUNG, YK, SQUIER CA. Effects of systemic administration of nicotine on capillaries in rat oral mucosa. J Oral Pathol Med 1989;18:230-232.
• KENNEY EB, KRAAL JH, SAXE SR, JONES J. The effects of cigarette smoke on human polymorphonuclear leukocytes. J Periodont Res 1977;12:277-234.
• LANGER B, LANGER L, HERMANN I, ERUG M. The wide fixture : a solution for special bone situations and rescue for the compromised implant. Part 1. Int J Oral Maxillofac Implants 1993;8:400-408.
• LEKHOLM U, BECKER W, DAHLIN C, BECKER BE, DONATH K, MORRISON E. The role of early versus late removal of GTAM membranes on bone formation at oral implants placed into immediate extraction sockets. An experimental study in dogs. Clin Oral Implant Res 1993;4:121-129.
• MACFARLANE GD, HERZBERG MC, WOLFF LF, HARDIE NA. Refractory periodontitis associated with abnormal PMN leukocyte phagocytosis and cigarette smoking. J Periodontol 1992;63:908-913.
• MEECHAN JG, MACGREGOR DM, HOBSON RS, BATES JPC, DENNISON M. The effects of smoking on immediate pest-extraction socket filling with blood and on the incidence of painful sockets. Br J Oral Maxillofac Surg 1988;26:402-409.
• MILLER PD. Root coverage with the free gingival graft. Factors associated with incomplete coverage. J Periodontol 1987;58:674-681.
• MOSELY LH, FINSETH F, GOODY M. Nicotine and its effects on wound healing. Plastic Reconstr Surg 1978;61:570-575.
• MINSK L, POLSON, WEISGOLG A, ROSE L, SANAVI F, BAUMGARTEN H, LISTGARTEN M. Outcome failures of endosseous implants from a clinical training center. Comp Contin education 1996;17:848-858.
• NOWZARI H, SLOTZ J. Microbiologic and clinical study of polytetrafluoro-ethylene membrandes for guided bone regeneration around implants. Int J oral Maxillofac Implants 1995;10:67-73.
• NYGAARD-ØSTBY P, TELLEFSEN G, SIGURDSSON T, ZIMMERMAN, WIKESJÖ U.M.E. Periodontal healing following reconstructive surgery : effect of guided tissue regeneration. J Clin Periodontol 1996;23:1073-1079.
• PREBER H, BERGSTROM J. Occurence of gingival bleeding in smoker and non-smoker patients. Acta Odontol Scand 1985a;43:315-320.
• PREBER H, BERGSTROM J. The effects of non-surgical treatment on periodontal pockets in smokers and non-smokers. J Clin Periodontol 1985b;12:319-323.
• PREBER H, BERGSTROM J. Effects of non-surgical treatment on gingival bleeding in smokers and non-smokers. Acta Odontol Scand 1986;44:85-89.
• KENNEY EB, KRAAL JH, SAXE SR, JONES J. The effects of cigarette smoke on human polymorphonuclear leukocytes. J Periodont Res 1977;12:277-234.
• REES TD, LIVERETT DM, GUY CL. The effects of cigarette smoking on skin flap survival in the face lift patient. Plastic Reconstr Surg 1984;73:911-915.
• RENOUARD F, RIACHI F. Apport des implants de 5 mm de diamètre en implantologie orale. Inf Dent 1994;24:2069-2076.
• RENOUARD F, ARNOUX JP, SARMENT D. 5 mm diameter implants without a smooth surface collar : report on 98 consecutive placements. Int J Oral Maxillofac Implants (soumis à publication).
• ROBERT P, MAUDUIT J, FRANK R, VERT M. Biocompatibility and resorbability of polylactic acid membranee for periodontal guided tissue regeneration. Biomaterials 1993;14:353-364.
• TONETTI M, PINI PRATO G, CORTELLINI P. Periodontal regeneration of human intrabony defect IV. Determinants of healing response. J Periodontol 1993;64:934-940.
• SCHENK RK. Bone regeneration : biological basis in guided bone regeneration in implant dentistry. Buser D, Dahlin C, Schenk RK : eds. Chicago, Quintessence Publish. Co, 1994.
• SHERWIN MA, GASWIRTH CM. Detrimental effects of cigarette smoking on lower extremity wound healing. J Foot Surg 1990;29:84-87.
• SIANNA JE, REX S, GOTTRUP F. The effects of cigarette smoking on wound healing. Scand J Plastic Reconstr Surg Hand Surg 1989;23:207-209.
• SWEET JB, BUTLER DP. Effect of smoking on the incidence of localized osteitis following mandibular third molar sugery. Quint Intern 1978;2:9-10.
• WARRER K, GOTFREDSEN K, HJORTING-HANSEN E, KARRING T. Guided tissue regeneration ensures osseointegration of dental implants placed into extraction sockets. An experimental study in monkeys. Clin Oral Implant Res 1991;2:166-171.