Beruflich Dokumente
Kultur Dokumente
Shin-ichi Tanabe
Masaaki Yokoyama
implant failure: a meta-analysis
Kenji Fujisawa
Eiji Yamauchi
Youji Miyamoto
473
Hinode et al . Smoking and implant failure
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 (19932003) 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 Beggs method
journals (Dentistry in Japan, 19932004,
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
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 1.743.72 5.93 Casecontrol nr r, bn
De Bruyn & Collaert (1994) 114/338 5.46 1.5719.02 2.43 Casecontrol nr r
Gorman et al. (1994) 646/1420 2.03 1.333.11 4.31 Cohort nr r
Minsk et al. (1996) 157/570 1.21 0.682.16 9.49 Casecontrol nr r
Wang et al. (1996) 13/70 0.98 0.195.02 15.66 Casecontrol nr r
Yamada et al. (1997) 389/347 2.08 1.044.18 5.3 Casecontrol nr r, bnn
Keller et al. (1999) 32/216 2.05 0.805.2 13.31 Casecontrol Current vs. never or former r
De Bruyn et al. (1999) 30/32 0.64 0.22.08 24.19 Cohort nr r
Jones et al. (1999) 126/217 4.06 1.3811.96 4.66 Casecontrol nr r
Wallace (2000) 72/115 2.68 1.046.91 10.70 Casecontrol nr r
Lambert et al. (2000) 959/1928 1.53 1.152.05 6.93 Cohort Current vs. never or former r
Kuroyama et al. (2001) 1522/2994 1.22 0.931.59 5.43 Casecontrol nr r
Geurs et al. (2001) 62/279 2.6 0.996.83 5.87 Casecontrol Current vs. never or former r
Widmark et al. (2001) 67/131 5.3 2.5311.12 20.2 Cohort nr r
Schwartz-Arad et al. (2002) 380/579 1.86 0.794.34 2.29 Cohort nr r
Kan et al. (2002) 70/158 2.76 1.166.62 10.09 Casecontrol Current vs. never or former r
Karoussis et al. (2003) 28/84 2.08 0.3313.12 4.46 Cohort nr r
Leonhardt et al. (2003) 31/13 3.5 0.3931.81 18.18 Cohort nr r
Shiratori & Isokawa (2003) 303/592 23.1 5.3998.9 2.68 Casecontrol 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 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 1.784.34 194/921 1.99 0.904.39
Minsk et al. (1996) 84/285 1.41 0.732.72 73/285 0.55 0.122.45
Yamada et al. (1997) 175/131 2.36 0.846.68 214/216 1.77 0.684.6
De Bruyn et al. (1999) 30/32 0.64 0.22.08
Wallace (2000) 42/84 4.31 1.3413.83 30/31 1.04 0.195.59
Lambert et al. (2000) 478/793 1.78 1.192.66 481/1135 1.23 0.801.9
Shiratori & Isokawa (2003) 117/302 16.27 1.94136.66 186/290 27.20 3.58206.93
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: 1.612.65). Beggs 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 2.17). Several casecontrol smokers and non-smokers. Smoking might
maxillary analysis was 100 (45 7 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
0.893.09). No significant difference was found only seven studies that met the effects of smoking in osseointegrated im-
found by Beggs 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 6 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: 0.722.4)). 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
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: 721764. Radiology and Endodontics 87: 649652.
Maxillofacial Implants 8: 609615. 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: 1623. Dentistry 87: 485489.
Dentistry 23: 695699, 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: 10881101. tional Journal of Oral & Maxillofacial Implants dontitis: a 10-year prospective cohort study of
Berge, T.I. & Gronningsaeter, A.G. (2000) Survival 6: 142146. 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: 329339.
mandibular overdentures. Clinical Oral Implants M.K. (2001) Retrospective radiographic analysis of Keller, E.E., Tolman, D.E. & Eckert, S.E. (1999)
Research 11: 154162. 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: 295302. 21: 517523. 707721.
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: 260264. 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 165168. plants 17: 816819.
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.,
139148. crobiota. Journal of Clinical Periodontology 28: Ueda, A., Furumoto, K. & Satoh, T. (2001)
Esposito, M., Hirsh, J.M., Lekholm, U. & Thom- 377388. 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: 196209. habits, intra-oral findings and clinical results
of Oral Sciences 106: 527551. 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 218236.
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: 7989. Implants Research 12: 589594. Dentistry 8: 103106.
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: 2232. Journal of Long Term Effects of Medical Implants
man. Journal of Periodontology 74: 14151422. Rosenthal, R. (1979) The file drawer problem and 6: 103117.
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: 638641. vall, A.M. & Ivanoff, C.J. (2001) Rehabilitation of
254261. 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 35-
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: 153157. Journal of Oral & Maxillofacial Implants 16:
from a clinical training center. Compendium of Shiratori, K. & Isokawa, R. (2003) The study of 7379.
Continuing Education in Dentistry 17: 848850, the risk of implant treatment by smoking and Yamada, Y., Niimi, A., Sawai, T., Watanabe, K.,
852854, 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: 5766. 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: 489496. 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: 239240. Implantology 10: 163166.