Smoking Osteointegration - Systematic review


Published on

Published in: Education, Health & Medicine
  • Be the first to comment

  • Be the first to like this

Smoking Osteointegration - Systematic review

  1. 1. Daisuke Hinode Influence of smoking on osseointegrated Shin-ichi Tanabe Masaaki Yokoyama implant failure: a meta-analysis Kenji Fujisawa Eiji Yamauchi Youji Miyamoto Authors’ affiliations: Key words: implant failure, meta-analysis, smoking Daisuke Hinode, Shin-ichi T anabe, Masaaki Yokoyama, Department of Preventive Dentistry, Institute of Health Bioscience, The University of Abstract: The aim of this study was to examine the influence of smoking on osseointegrated Tokushima Graduate School, Tokushima City, implant failure by performing a meta-analysis. A computerized literature search using Japan Kenji Fujisawa, Department of Oral and PubMed database (in English) and Japana Centra Revuo Medicina (in Japanese) was carried Maxillofacial Surgery, Tokushima University out to identify all relevant studies. Among 175 studies identified and chosen for detailed Hospital, Tokushima City, Japan review, 19 were appropriate for inclusion in our meta-analysis. When smokers were Eiji Yamauchi, Department of Oral Care and Clinical Education, Tokushima University compared with non-smokers, odds ratio (OR) for osseointegrated implant failure was Hospital, Tokushima City, Japan significantly elevated (OR 2.17, 95% confidence intervals (CI), 1.67–2.83). Seven studies Youji Miyamoto, Division of Dentistry and Oral Surgery, School of Medicine, Akita University, were appropriate to examine the influence of intra-oral location (maxillary arch vs. Akita City, Japan mandibular arch) of implant failure on smoking. The OR for implant failure occurring in the Correspondence to: maxillary arch was significantly elevated (OR 2.06, 95% CI, 1.61–2.65), whereas the OR in the D. Hinode mandibular arch did not demonstrate a significant increased risk associated with smoking Department of Preventive Dentistry (OR 1.32, 95% CI, 0.72–2.4). Our meta-analysis revealed a significant relationship between Subdivision of Stomatology Division of Biosystem and Nutritional Science smoking and the risk of osseointegrated implant failure, more particularly those implants Institute of Health Bioscience located in the maxillary arch. The University of Tokushima Graduate School Tokushima 770-8504 Japan Tel.: þ 81 88 633 7337 Fax: þ 81 88 633 7338 The success of osseointegration between an accounting for the majority of the differ- e-mail: endosseous titanium implant and bone can ence. be expected. Failure to achieve osseointe- On the contrary, Bain et al. (2002) re- gration and loss of acquired osseointegra- cently reported that no difference in im- tion has been related to several factors such plant failure between smoking and non- as the length, diameter and surface rough- smoking groups was detected. Kumar et ness of the fixture, overloading and infec- al. (2002) also showed that the success rate tion (Esposito et al. 1998a, 1998b). A was not significantly different between literature review revealed that information smokers and non-smokers when the suc- related to the effect of smoking on endoss- cess was defined as no clinical signs or eous dental implants is poorly described. symptoms of peri-implant pathosis, no Wallace (2000) has demonstrated the dele- clinically visible mobility and no evidence Date: terious effect of tobacco use on implant of bone loss. Accepted 5 October 2005 treatment, and proposed that smoking To clarify the relationships between smok- To cite this article: should be considered a relative contraindi- ing habit and osseointegrated implant failure, Hinode D, Tanabe S-i, Yokoyama M, Fujisawa K, Yamauchi E, Miyamoto Y. The influence of smoking on cation to implant placement when plan- we conducted a meta-analysis of published osseointegrated implant failure: a meta-analysis. ning treatment. Johnson & Hill (2004) epidemiological reports. In addition, we ex- Clin. Oral Impl. Res. 17, 2006; 473–478 doi: 10.1111/j.1600-0501.2005.01244.x reported that implant failure in smokers amined the influence of intra-oral location is twice as much as those in non-smokers, (maxillary arch vs. mandibular arch) of os- Copyright r Blackwell Munksgaard 2006 with a higher failure rate in maxillary arch seointegrated implant failure in smokers. 473
  2. 2. Hinode et al . Smoking and implant failure Material and methods 1. Test of heterogeneity (Q statistic test) Literature search Studies with potential relevance for our Random effects model Fixed effects model meta-analysis were identified by using (DerSimonian-Laird method) (general variance-based method) both exploded MeSH heading and text words in search of MEDLINE (online 2. Sensitivity analysis PubMed 1993 to August 2004) and Japana 1) study design (cohort versus case-control study) Centra Revuo Medicina (1993–2003) data- 2) year of publication (before 2000 versus after 2000) bases. The main search terms were ‘smok- 3) case number (less than 200 versus 200 or more) ing’, ‘smoke’ and ‘tobacco’ in combination 4) success rate (less than 90% versus 90% or more) with ‘implant’. Additional articles of po- 3. Analysis of the potential for publication bias tential relevance were identified in current 1) formal tests by Begg’s method journals (Dentistry in Japan, 1993–2004, 2) Fail-Safe N statistical measure Japan Printing Co., Ltd, Tokyo, Japan and Clinical Research in Dentistry, 2004, Quintessence Publishing, Tokyo, Japan). Synthesized odds ratio Adjust synthesized odds ratio Inclusion criteria Fig. 1. Key steps of statistical analysis. Data extraction was performed by two of the authors and the selected data were entered into a standard evidence table. using Microsoft Excel 2000 (Microsoft Fail-Safe n is the number of unpublished Selected studies were limited to case–con- Corporation, Redmond, WA, USA) or the studies finding no association that would trol and cohort studies that included data in SPSS 11 for Windows statistical software be required to change a significant meta- which smoking was examined as a risk package (SPSS Inc., Chicago, IL, USA). We analysis result (based on the published factor for the failure of osseointegrated used both random-effects and fixed-effects studies) to a non-significant result. If n is implant and variance for that estimate models to derive the pooled OR from com- more than 5 times the number of studies were provided. Failures of implant can be binations of studies. Heterogeneity among plus 10, then in general publication bias is divided into biological failure, mechanical the studies was assessed using the Q-sta- not considered to be a problem. failure, iatrogenic failure, and inadequate tistic test and by comparing random-effects patient adaptation as described by Esposito and fixed-effects estimates. A random-ef- Results et al. (1998a, 1998b). Several kinds of fects model (DerSimonian–Laird method) Of the 175 studies identified by our litera- definition of implant failure have been of meta-analysis was used if the P-value ture search in MEDLINE and Japana Cen- reported by researchers. We chose the defi- was less than 0.05. When there was no tra Revuo Medicina databases, 19 were nitions of implant failure according to the significant difference, we used a fixed-ef- chosen for detailed review. The number criteria of Bain & Moy (1993) with slight fects model (general variance-based of cases and selected study design charac- modification. In brief, an implant was method). Statistical significance was teristics are shown in Table 1. Among considered to be a failure when it had judged by using the OR and 95% confi- these, 12 case–control and seven cohort been removed for any reason, and showed dence interval (CI), and there was statisti- studies were appropriate for inclusion in progressive bone loss assessed by radio- cal significance if the 95% CI did not our analysis. A significant heterogeneity graph. Furthermore, the data extracted include one. was demonstrated by the Q-statistic test were entered into each table of the max- A sensitivity analysis was conducted to (Po0.05); therefore, a random-effects illary arch or the mandibular arch when assess the friability of our findings with model was used. The synthesized OR for this information was available. Studies respect to different assumptions as shown overall studies was 2.17 (95% CI: 1.67– were excluded if they were case reports, in Fig. 1. It was performed on pooled out- 2.83) with a range of 0.64 to 23.1 (Fig. 2). case series or reviews. When studies over- comes in terms of study design (cohort vs. No publication bias was confirmed by the lapped, only the largest data set, or the data case–control study), year of publication analysis of Begg’s method (P40.1) and the set with the most relevant data, was in- (before 2000 vs. after 2000), case number Fail-Safe n statistical measure (n ¼ 6954 cluded. (less than 200 vs. 200 or more) and the 5 Â 19 þ 10). Therefore, the result of the ratio of implant failure (less than 10% vs. synthesized OR for overall studies was not Statistical analysis 10% or more). Estimates of OR and the adjusted. Figure 1 enumerates the key steps of the 95% CI were calculated. To explore the By investigating dependence of OR on statistical analysis of this study. The potential for publication bias, data were study characteristics, we have already im- strength of the relationship between smok- examined by means of the formal tests plicitly examined sensitivity of results to ing and osseointegrated implant failure was proposed by Begg’s method (Begg & Ma- these characteristics (study design, year of assessed by odds ratio (OR). All calcula- zumdar 1994) in addition to the Fail-Safe n publication, case number and rate of im- tions for the meta-analysis were performed statistical measure (Rosenthal 1979). The plant failure) as shown in Table 2. The 474 | Clin. Oral Impl. Res. 17, 2006 / 473–478
  3. 3. Hinode et al . Smoking and implant failure Table 1. Characteristics of studies selected in meta-analysis aimed to examine the influence of smoking on osseointegrated implant failure Author, the year of publish No. cases/ OR 95% CI Rate of Study Cigarette Diagnosis no. controls (confidence implant design smoking of implant interval) failure (%) failure Bain & Moy (1993) 390/1804 2.54 1.74–3.72 5.93 Case–control nr r, bn De Bruyn & Collaert (1994) 114/338 5.46 1.57–19.02 2.43 Case–control nr r Gorman et al. (1994) 646/1420 2.03 1.33–3.11 4.31 Cohort nr r Minsk et al. (1996) 157/570 1.21 0.68–2.16 9.49 Case–control nr r Wang et al. (1996) 13/70 0.98 0.19–5.02 15.66 Case–control nr r Yamada et al. (1997) 389/347 2.08 1.04–4.18 5.3 Casecontrol nr r, bnn Keller et al. (1999) 32/216 2.05 0.80–5.2 13.31 Case–control Current vs. never or former r De Bruyn et al. (1999) 30/32 0.64 0.2–2.08 24.19 Cohort nr r Jones et al. (1999) 126/217 4.06 1.38–11.96 4.66 Case–control nr r Wallace (2000) 72/115 2.68 1.04–6.91 10.70 Case–control nr r Lambert et al. (2000) 959/1928 1.53 1.15–2.05 6.93 Cohort Current vs. never or former r Kuroyama et al. (2001) 1522/2994 1.22 0.93–1.59 5.43 Case–control nr r Geurs et al. (2001) 62/279 2.6 0.99–6.83 5.87 Case–control Current vs. never or former r Widmark et al. (2001) 67/131 5.3 2.53–11.12 20.2 Cohort nr r Schwartz-Arad et al. (2002) 380/579 1.86 0.79–4.34 2.29 Cohort nr r Kan et al. (2002) 70/158 2.76 1.16–6.62 10.09 Case–control Current vs. never or former r Karoussis et al. (2003) 28/84 2.08 0.33–13.12 4.46 Cohort nr r Leonhardt et al. (2003) 31/13 3.5 0.39–31.81 18.18 Cohort nr r Shiratori & Isokawa (2003) 303/592 23.1 5.39–98.9 2.68 Case–control nr r vs., versus; nr, not reported; current, current smoker; quit, subject who quit smoking; never, subjects who had never smoked; r, removal; b, progressive bone loss assessed by the radiograph; bn, bone loss in excess of 50% of the fixture length; bnn, progressive bone loss with mobility or pain. Table 2. Distribution of odds ratio between smoking and non-smoking Group Studies OR (95% CI) (n) All study 19 2.17 (1.67–2.83) Study design Cohort 7 1.98 (1.31–3) Case–control 12 2.36 (1.63–3.42) Year of publication Before 2000 9 2.05 (1.65–2.55) 2000 or more 10 2.50 (1.64–3.82) Case numbers Less than 200 6 2.57 (1.61–4.12) 200 After 13 2.12 (1.61–2.8) Implant failure rate Less than 10% 12 2.10 (1.56–2.82) 10% or more 7 2.53 (1.72–3.73) synthesized OR of the subgroup by cohort study and by case–control study was 1.98 (95% CI: 1.31–3) and 2.36 (95% CI: 1.63–3.42), respectively. Furthermore, the synthesized OR of the subgroup (case num- ber 4200, n ¼ 13) and the subgroup (im- plant failure rate o10%, n ¼ 12) was 2.12 (95% CI: 1.61–2.8) and 2.1 (95% CI: 1.56– 2.82), respectively. In subgroup based on the year of publication, the synthesized OR before 2000 (n ¼ 9) and after 2000 (n ¼ 10) was 2.05 (95% CI: 1.65–2.55) and 2.5 Fig. 2. Random effects of the odds ratios of osseointegrated implant failure on smoking for individual studies (95% CI: 1.64–3.82), respectively, and and overall. The diamond symbol shows the odds ratio and the horizontal line represents the 95% confidence both of them revealed a significant differ- intervals. ence between smokers and non-smokers. 475 | Clin. Oral Impl. Res. 17, 2006 / 473–478
  4. 4. Hinode et al . Smoking and implant failure Table 3. Studies selected in meta-analysis aimed to examine the influence of intra-oral location on osseointegrated implant failure Author, the year In the maxillary arch In the mandible arch of publication No. cases/ OR 95% CI No. cases/ OR 95% CI no. controls no. controls Bain & Moy (1993) 196/883 2.78 1.78–4.34 194/921 1.99 0.90–4.39 Minsk et al. (1996) 84/285 1.41 0.73–2.72 73/285 0.55 0.12–2.45 Yamada et al. (1997) 175/131 2.36 0.84–6.68 214/216 1.77 0.68–4.6 De Bruyn et al. (1999) 30/32 0.64 0.2–2.08 – – – Wallace (2000) 42/84 4.31 1.34–13.83 30/31 1.04 0.19–5.59 Lambert et al. (2000) 478/793 1.78 1.19–2.66 481/1135 1.23 0.80–1.9 Shiratori & Isokawa (2003) 117/302 16.27 1.94–136.66 186/290 27.20 3.58–206.93 Sensitivity analysis for all studies did not smoking by the influence of intra-oral loca- ence. Factors contributing to higher im- show much difference between study de- tion indicated an increased OR in the max- plant failure in maxillary arch than in sign, case number and the rate of implant illary arch but not in the mandibular arch. mandibular arch are not yet understood. failure. Our meta-analysis of 19 studies The high implant failure rate observed in revealed a statistically significant associa- the maxilla suggests that bone density may tion between smoking and osseointegrated Discussion have some bearing upon early implant fail- implant failure by the results of the synthe- ure (Friberg et al. 1991). In addition, smok- sized OR. Sweet (1992) and Bain & Moy (1993)were ing has been reported to be the strongest The studies selected to investigate the the first authors to report that smoking is a risk factor for periodontal bone loss (Pau- influence of intra-oral location (placed in risk factor for osseointegrated implant fail- lander et al. 2004). On the other hand, it the maxillary or mandibular arch) are listed ure, and some prospective studies sup- has been reported that the short-term prog- in Table 3. The data of De Bruyn et al. ported this relationship. However, several nosis is greatly affected by peri-implant (1999) were entered into the table of the recent studies have indicated no effect of infection related to smoking (Mau 1993; maxillary arch, because the number of im- smoking on the survival of dental im- De Bruyn & Collaert 1994; Gorman et al. plant failures in smokers and non-smokers plants (Esposito et al. 1998a, 1998b; Berge 1994). Reibel (2003) suggested that the was only available in the maxillary arch. & Gronningsaeter 2000; Carlsson et al. increase of implant failures in smokers is Tests of homogeneity of implant failure in 2000; Lambert et al 2000; Quirynen et al. due to exposure of peri-implant tissue to seven papers selected were examined, and 2001). Although a meta-analysis study was tobacco smoke, possibly linking the smok- no significant difference was found in the previously reported (Bain et al. 2002), no ing effects on implant survival to the smok- maxillary arch. Then, the fixed-effects evaluation using the synthesized OR esti- ing effects on peri-implantitis. Haffajee & model was used to calculate OR. The mates has been published. Therefore, we Socransky (2001) showed that a greater synthesized OR of the maxillary arch was constructed and performed this study. Our difference in prevalence of orange and red 2.06 (95% CI: 1.61–2.65). Begg’s method meta-analysis of the 19 studies examining bacterial complexes (both complexes in- was carried out to explore the potential this relationship revealed a significant in- clude periodontopathogens) exists at sites publication bias, and the results of the creased risk for the implant failure among with periodontal pocket in the maxilla in maxillary arch showed no significant dif- smokers compared with that among non- comparison with the mandibular in both ference. The Fail-Safe n calculation for the smokers (OR ¼ 2.17). Several case–control smokers and non-smokers. Smoking might maxillary analysis was 100 (45 Â 7 þ 10) studies and cross-sectional studies with thus modulate the implant failure by and publication bias was not considered to different case numbers and ratio of implant influencing bacterial infections in the max- be a large factor; therefore, the synthesized failure have found similar associations. illary and the mandibular arches differ- OR of the maxillary arch was not adjusted. Therefore, it confirmed a significant rela- ently. On the other hand, tests of homogeneity of tionship between smoking and the risk of Using meta-analysis, it has been demon- implant failure in the mandibular arch osseointegrated implant failure. strated in this study that smoking has found a significant difference (Po0.05), In this report, we also attempted to significant negative effects on the survival and the random-effects model was used. evaluate the location of the implant failure rate of dental implants. Dentists have an The synthesized OR was 1.66 (95% CI: on smoking by using a meta-analysis, and important role to play in preventing the 0.89–3.09). No significant difference was found only seven studies that met the effects of smoking in osseointegrated im- found by Begg’s method whereas the Fail- inclusion criteria. There has been consider- plant failure and tobacco-related oral dis- Safe n calculation for the maxillary analysis able discussion regarding implant location eases. Reibel (2003) suggested in his review was 28 (o5 Â 6 þ 10); therefore publica- and failure. In our analysis, the maxillary that practitioners should pursue more for- tion bias was considered to be a factor and arch showed a statistically significant in- mal training in smoking cessation counsel- the synthesized OR of the mandibular arch crease in the implant failure in smokers ing, which should be as much a part of was adjusted (1.32 (95% CI: 0.72–2.4)). compared with non-smokers; however, the their job as plaque control and dietary Finally, analysis of implant failure on mandibular arch did not show any differ- advice. The evidence presented in this 476 | Clin. Oral Impl. Res. 17, 2006 / 473–478
  5. 5. Hinode et al . Smoking and implant failure study may contribute to develop tobacco tween smokers and non-smokers, how- intervention habits by dental practitioners. ever, only 0.3% difference in the success Several investigators have reported that rate of rough-surface implant was found. the surface roughness of implants (fixture) Among pooled data used in this study, no also represents a factor of implant failure study had investigated the effect of the (Kumar et al. 2002; Feldman et al. 2004). surface roughness of fixture between smo- Feldman et al. (2004) reported that there kers and non-smokers. Further research is was a 9% difference in 5-year cumulative needed to clarify the influence against sur- survival rates in the maxilla between ma- face-modified dental implants in the max- chined-surfaced short-length (86.8%) and illary arch on smoking leading to the dual acid-etched short-length implants results observed in this study. (95.8%). It has recently been reported that no significant difference of the effect of smoking on achieving initial osseointegra- Acknowledgements: We thank Dr tion exists when surface-modified dental Makoto Fukui, General Dentistry, implants were used (Kumar et al. 2002). In Tokushima University Hospital, who the systematic review of Bain et al. (2002), gave excellent support and assistance. 93.5% of success rate in smokers for the We are also grateful to Dr Daniel implant group with smooth surface and ´ Grenier, Universite Laval, Canada, for 98.7% for the implant group with rough critical discussion of the study and for surface was observed, a clinically relevant amending the paper. difference between these two groups. Be- References Bain, C.A. & Moy, P.K. (1993) The association failures of osseointegrated oral implants. (II). Etio- plasma-sprayed cylinder dental implants. Oral between the failure of dental implants and cigar- pathogenesis. European Journal of Oral Sciences Surgery Oral Medicine Oral Pathology Oral ette smoking. International Journal of Oral & 106: 721–764. Radiology and Endodontics 87: 649–652. Maxillofacial Implants 8: 609–615. Feldman, S., Boitel, N., Weng, D., Kohles, S.S. & Kan, J.Y., Rungcharassaeng, K., Kim, J., Lozada, J.L. Bain, C.A., Weng, D., Meltzer, A., Kohles, S.S. & Stach, R.M. (2004) Five-year survival distribu- & Goodacre, C.J. (2002) Factors affecting the Stach, R.M. (2002) A meta-analysis evaluating tions of short-length (10 mm or less) machined- survival of implants placed in grafted maxillary the risk for implant failure in patients who smoke. surfaced and osseotite implants. Clinical Implant sinuses: a clinical report. Journal of Prosthetic Compendium of Continuing Education in Dentistry & Related Research 6: 16–23. Dentistry 87: 485–489. Dentistry 23: 695–699, 702. 704 passim; quiz Friberg, B., Jemt, T. & Lekholm, U. (1991) Early Karoussis, I.K., Salvi, G.E., Heitz-Mayfield, L.J., 708. failures in 4,641 consecutively placed Branemark Bragger, U., Hammerle, C.H. & Lang, N.P. Begg, C.B. & Mazumdar, M. (1994) Operating dental implants: a study from stage 1 surgery to (2003) Long-term implant prognosis in patients characteristics of a rank correlation test for pub- the connection of completed prostheses. Interna- with and without a history of chronic perio- lication bias. Biometrics 50: 1088–1101. tional Journal of Oral & Maxillofacial Implants dontitis: a 10-year prospective cohort study of Berge, T.I. & Gronningsaeter, A.G. (2000) Survival 6: 142–146. the ITI Dental Implant System. Clinical Oral of single-crystal sapphire implants supporting Geurs, N.C., Wang, I.C., Shulman, L.B. & Jeffcoat, Implants Research 14: 329–339. mandibular overdentures. Clinical Oral Implants M.K. (2001) Retrospective radiographic analysis of Keller, E.E., Tolman, D.E. & Eckert, S.E. (1999) Research 11: 154–162. sinus graft and implant placement procedures Maxillary antral-nasal inlay autogenous bone Carlsson, G.E., Lindquist, L.W. & Jemt, T. (2000) from the Academy of Osseointegration Consen- graft reconstruction of compromised maxilla: Long-term marginal peri-implant bone loss in sus Conference on Sinus Grafts. International a 12-year retrospective study. International edentulous patients. International Journal of Journal of Periodontics & Restorative Dentistry Journal of Oral & Maxillofacial Implants 14: Prosthodontics 13: 295–302. 21: 517–523. 707–721. De Bruyn, H. & Collaert, B. (1994) The effect of Gorman, L.M., Lambert, P.M., Morris, H.F., Ochi, Kumar, A., Jaffin, R.A. & Berman, C. (2002) The smoking on early implant failure. Clinical Oral S. & Winkler, S. (1994) The effect of smoking on effect of smoking on achieving osseointegration of Implants Research 5: 260–264. implant survival at second-stage surgery: DICRG surface-modified implants: a clinical report. Inter- De Bruyn, H., Collaert, .B., Linden, U., Johansson, Interim Report No. 5. Implant Dentistry 3: national Journal of Oral & Maxillofacial Im- C. & Albrektsson, T. (1999) Clinical outcome of 165–168. plants 17: 816–819. Screw Vent implants. A 7-year prospective follow- Haffajee, A.D. & Socransky, S.S. (2001) Relation- Kuroyama, I., Osato, S., Ohtsuka, T., Kubota, R., up study. Clinical Oral Implants Research 10: ship of cigarette smoking to the subgingival mi- Okudera, H., Okada, M., Saijoh, S., Ogawa, T., 139–148. crobiota. Journal of Clinical Periodontology 28: Ueda, A., Furumoto, K. & Satoh, T. (2001) Esposito, M., Hirsh, J.M., Lekholm, U. & Thom- 377–388. A retrospective study of 5,092 implants con- sen, P. (1998a) Biological factors contributing to Johnson, G.K. & Hill, M. (2004) Cigarette smoking secutively placed in both maxilla and mandible: failures of osseointegrated oral implants. (I). Suc- and the periodontal patient. Journal of Perio- relationship between systemic history, life cess criteria and epidemiology. European Journal dontology 75: 196–209. habits, intra-oral findings and clinical results of Oral Sciences 106: 527–551. Jones, J.D., Lupori, J., Van Sickels, J.E. & Gardner, [in Japanese, with English abstract]. Journal of Esposito, M., Hirsh, J.M., Lekholm, U. & Thom- W. (1999) A 5-year comparison of hydroxyapatite- Japanese Society of Oral Implantology 14: sen, P. (1998b) Biological factors contributing to coated titanium plasma-sprayed and titanium 218–236. 477 | Clin. Oral Impl. Res. 17, 2006 / 473–478
  6. 6. Hinode et al . Smoking and implant failure Lambert, P.M., Morris, H.F. & Ochi, S. (2000) The around screw-shaped c. p. titanium-machined Wallace, R.H. (2000) The relationship between ci- influence of smoking on 3-year clinical success of implants in partially edentulous patients with or garette smoking and dental implant failure. Eur- osseointegrated dental implants. Annals of Perio- without ongoing periodontitis. Clinical Oral opean Journal of Prosthodontics and Restorative dontology 5: 79–89. Implants Research 12: 589–594. Dentistry 8: 103–106. Leonhardt, A., Dahlen, G. & Renvert, S. (2003) Five- Reibel, J. (2003) Tobacco and oral diseases. Update Wang, I.C., Reddy, M.S., Geurs, N.C. & Jeffcoat, year clinical, microbiological, and radiological out- on the evidence, with recommendations. Medical M.K. (1996) Risk factors in dental implant failure. come following treatment of peri-implantitis in Principles and Practice 12: 22–32. Journal of Long T erm Effects of Medical Implants man. Journal of Periodontology 74: 1415–1422. Rosenthal, R. (1979) The ‘‘file drawer problem’’ and 6: 103–117. Mau, J. (1993) On statistics of success and loss for tolerance for null results. Psychological Bulletin Widmark, G., Andersson, B., Carlsson, G.E., Lind- dental implants. International Dental Journal 43: 86: 638–641. vall, A.M. & Ivanoff, C.J. (2001) Rehabilitation of 254–261. Schwartz-Arad, D., Samet, N., Samet, N. & Mam- patients with severely resorbed maxillae by means Minsk, L., Polson, A.M., Weisgold, A., Rose, L.F., lider, A. (2002) Smoking and complications of of implants with or without bone grafts: a 3–5- Sanavi, F., Baumgarten, H. & Listgarten, M.A. endosseous dental implants. Journal of Perio- year follow-up clinical report. International (1996) Outcome failures of endosseous implants dontology 73: 153–157. Journal of Oral & Maxillofacial Implants 16: from a clinical training center. Compendium of Shiratori, K. & Isokawa, R. (2003) The study of 73–79. Continuing Education in Dentistry 17: 848–850, the risk of implant treatment by smoking and Yamada, Y., Niimi, A., Sawai, T., Watanabe, K., 852–854, 856 passim. the support for smoking cessation [in Hibino, Y., Nakai, H., Honda, M., Fujimoto, T., Paulander, J., Wennstrom, J.L., Axelsson, P. & Japanese]. Quintessence Dental Implantology Murakami, I., Yanbe, M., Suzuki, E., Kasuya, Y., Lindhe, J. (2004) Some risk factors for perio- 10: 57–66. Matsuyama, M., Ozeki, K., Goto, Y., Takeuchi, dontal-bone loss in 50-year-old individuals. A Sweet, J.B. (1992) The relationship of cigarette M., Kawai, M., Ohara, H., Hibi, H. & Ueda, M. 10-year cohort study. Journal of Clinical Perio- smoking to impaired intra-oral wound healing: a (1997) The influence of cigarette smoking on dontology 31: 489–496. review of evidence and implications for patient osseointegrated implants [in Japanese, with Eng- Quirynen, M., Peeters, W., Naert, I., Coucke, W. & care. Journal of Oral and Maxillofacial Surgery lish abstract]. Journal of Japanese Society of Oral van Steenberghe, D. (2001) Peri-implant health 50: 239–240. Implantology 10: 163–166. 478 | Clin. Oral Impl. Res. 17, 2006 / 473–478