Smoking Osteointegration - Systematic review

Influence of smoking on osseointegratedimplant failure: a meta-analysis

Daisuke HinodeShin-ichi TanabeMasaaki YokoyamaKenji FujisawaEiji YamauchiYouji Miyamoto

Authors’ affiliations:Daisuke Hinode, Shin-ichi Tanabe, MasaakiYokoyama, Department of Preventive Dentistry,Institute of Health Bioscience, The University ofTokushima Graduate School, Tokushima City,JapanKenji Fujisawa, Department of Oral andMaxillofacial Surgery, Tokushima UniversityHospital, Tokushima City, JapanEiji Yamauchi, Department of Oral Care andClinical Education, Tokushima UniversityHospital, Tokushima City, JapanYouji Miyamoto, Division of Dentistry and OralSurgery, School of Medicine, Akita University,Akita City, Japan

Correspondence to:D. HinodeDepartment of Preventive DentistrySubdivision of StomatologyDivision of Biosystem and Nutritional ScienceInstitute of Health BioscienceThe University of Tokushima Graduate SchoolTokushima 770-8504JapanTel.: þ81 88 633 7337Fax: þ81 88 633 7338e-mail: [email protected]

Key words: implant failure, meta-analysis, smoking

Abstract: The aim of this study was to examine the influence of smoking on osseointegrated

implant failure by performing a meta-analysis. A computerized literature search using

PubMed database (in English) and Japana Centra Revuo Medicina (in Japanese) was carried

out to identify all relevant studies. Among 175 studies identified and chosen for detailed

review, 19 were appropriate for inclusion in our meta-analysis. When smokers were

compared with non-smokers, odds ratio (OR) for osseointegrated implant failure was

significantly elevated (OR 2.17, 95% confidence intervals (CI), 1.67–2.83). Seven studies

were appropriate to examine the influence of intra-oral location (maxillary arch vs.

mandibular arch) of implant failure on smoking. The OR for implant failure occurring in the

maxillary arch was significantly elevated (OR 2.06, 95% CI, 1.61–2.65), whereas the OR in the

mandibular arch did not demonstrate a significant increased risk associated with smoking

(OR 1.32, 95% CI, 0.72–2.4). Our meta-analysis revealed a significant relationship between

smoking and the risk of osseointegrated implant failure, more particularly those implants

located in the maxillary arch.

The success of osseointegration between an

endosseous titanium implant and bone can

be expected. Failure to achieve osseointe-

gration and loss of acquired osseointegra-

tion has been related to several factors such

as the length, diameter and surface rough-

ness of the fixture, overloading and infec-

tion (Esposito et al. 1998a, 1998b). A

literature review revealed that information

related to the effect of smoking on endoss-

eous dental implants is poorly described.

Wallace (2000) has demonstrated the dele-

terious effect of tobacco use on implant

treatment, and proposed that smoking

should be considered a relative contraindi-

cation to implant placement when plan-

ning treatment. Johnson & Hill (2004)

reported that implant failure in smokers

is twice as much as those in non-smokers,

with a higher failure rate in maxillary arch

accounting for the majority of the differ-

ence.

On the contrary, Bain et al. (2002) re-

cently reported that no difference in im-

plant failure between smoking and non-

smoking groups was detected. Kumar et

al. (2002) also showed that the success rate

was not significantly different between

smokers and non-smokers when the suc-

cess was defined as no clinical signs or

symptoms of peri-implant pathosis, no

clinically visible mobility and no evidence

of bone loss.

To clarify the relationships between smok-

ing habit and osseointegrated implant failure,

we conducted a meta-analysis of published

epidemiological reports. In addition, we ex-

amined the influence of intra-oral location

(maxillary arch vs. mandibular arch) of os-

seointegrated implant failure in smokers.Copyright r Blackwell Munksgaard 2006

Date:Accepted 5 October 2005

To cite this article:Hinode D, Tanabe S-i, Yokoyama M, Fujisawa K,Yamauchi E, Miyamoto Y. The influence of smoking onosseointegrated implant failure: a meta-analysis.Clin. Oral Impl. Res. 17, 2006; 473–478doi: 10.1111/j.1600-0501.2005.01244.x

473

Material and methods

Literature search

Studies with potential relevance for our

meta-analysis were identified by using

both exploded MeSH heading and text

words in search of MEDLINE (online

PubMed 1993 to August 2004) and Japana

Centra Revuo Medicina (1993–2003) data-

bases. The main search terms were ‘smok-

ing’, ‘smoke’ and ‘tobacco’ in combination

with ‘implant’. Additional articles of po-

tential relevance were identified in current

journals (Dentistry in Japan, 1993–2004,

Japan Printing Co., Ltd, Tokyo, Japan and

Clinical Research in Dentistry, 2004,

Quintessence Publishing, Tokyo, Japan).

Inclusion criteria

Data extraction was performed by two of

the authors and the selected data were

entered into a standard evidence table.

Selected studies were limited to case–con-

trol and cohort studies that included data in

which smoking was examined as a risk

factor for the failure of osseointegrated

implant and variance for that estimate

were provided. Failures of implant can be

divided into biological failure, mechanical

failure, iatrogenic failure, and inadequate

patient adaptation as described by Esposito

et al. (1998a, 1998b). Several kinds of

definition of implant failure have been

reported by researchers. We chose the defi-

nitions of implant failure according to the

criteria of Bain & Moy (1993) with slight

modification. In brief, an implant was

considered to be a failure when it had

been removed for any reason, and showed

progressive bone loss assessed by radio-

graph. Furthermore, the data extracted

were entered into each table of the max-

illary arch or the mandibular arch when

this information was available. Studies

were excluded if they were case reports,

case series or reviews. When studies over-

lapped, only the largest data set, or the data

set with the most relevant data, was in-

cluded.

Statistical analysis

Figure 1 enumerates the key steps of the

statistical analysis of this study. The

strength of the relationship between smok-

ing and osseointegrated implant failure was

assessed by odds ratio (OR). All calcula-

tions for the meta-analysis were performed

using Microsoft Excel 2000 (Microsoft

Corporation, Redmond, WA, USA) or the

SPSS 11 for Windows statistical software

package (SPSS Inc., Chicago, IL, USA). We

used both random-effects and fixed-effects

models to derive the pooled OR from com-

binations of studies. Heterogeneity among

the studies was assessed using the Q-sta-

tistic test and by comparing random-effects

and fixed-effects estimates. A random-ef-

fects model (DerSimonian–Laird method)

of meta-analysis was used if the P-value

was less than 0.05. When there was no

significant difference, we used a fixed-ef-

fects model (general variance-based

method). Statistical significance was

judged by using the OR and 95% confi-

dence interval (CI), and there was statisti-

cal significance if the 95% CI did not

include one.

A sensitivity analysis was conducted to

assess the friability of our findings with

respect to different assumptions as shown

in Fig. 1. It was performed on pooled out-

comes in terms of study design (cohort vs.

case–control study), year of publication

(before 2000 vs. after 2000), case number

(less than 200 vs. 200 or more) and the

ratio of implant failure (less than 10% vs.

10% or more). Estimates of OR and the

95% CI were calculated. To explore the

potential for publication bias, data were

examined by means of the formal tests

proposed by Begg’s method (Begg & Ma-

zumdar 1994) in addition to the Fail-Safe n

statistical measure (Rosenthal 1979). The

Fail-Safe n is the number of unpublished

studies finding no association that would

be required to change a significant meta-

analysis result (based on the published

studies) to a non-significant result. If n is

more than 5 times the number of studies

plus 10, then in general publication bias is

not considered to be a problem.

Results

Of the 175 studies identified by our litera-

ture search in MEDLINE and Japana Cen-

tra Revuo Medicina databases, 19 were

chosen for detailed review. The number

of cases and selected study design charac-

teristics are shown in Table 1. Among

these, 12 case–control and seven cohort

studies were appropriate for inclusion in

our analysis. A significant heterogeneity

was demonstrated by the Q-statistic test

(Po0.05); therefore, a random-effects

model was used. The synthesized OR for

overall studies was 2.17 (95% CI: 1.67–

2.83) with a range of 0.64 to 23.1 (Fig. 2).

No publication bias was confirmed by the

analysis of Begg’s method (P40.1) and the

Fail-Safe n statistical measure (n¼69545 � 19þ 10). Therefore, the result of the

synthesized OR for overall studies was not

adjusted.

By investigating dependence of OR on

study characteristics, we have already im-

plicitly examined sensitivity of results to

these characteristics (study design, year of

publication, case number and rate of im-

plant failure) as shown in Table 2. The

1. Test of heterogeneity (Q statistic test)

Random effects model(DerSimonian-Laird method)

Fixed effects model(general variance-based method)

2. Sensitivity analysis 1) study design (cohort versus case-control study)2) year of publication (before 2000 versus after 2000)3) case number (less than 200 versus 200 or more)4) success rate (less than 90% versus 90% or more)

3. Analysis of the potential for publication bias1) formal tests by Begg’s method2) Fail-Safe N statistical measure

Adjust synthesized odds ratioSynthesized odds ratio

Fig. 1. Key steps of statistical analysis.

Hinode et al . Smoking and implant failure

474 | Clin. Oral Impl. Res. 17, 2006 / 473–478

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, n¼ 13) and the subgroup (im-

plant failure rate o10%, n¼ 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 (n¼ 9) and after 2000 (n¼ 10)

was 2.05 (95% CI: 1.65–2.55) and 2.5

(95% CI: 1.64–3.82), respectively, and

both of them revealed a significant differ-

ence between smokers and non-smokers.

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/no. controls

OR 95% CI(confidenceinterval)

Rate ofimplantfailure (%)

Studydesign

Cigarettesmoking

Diagnosisof implantfailure

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 rGorman et al. (1994) 646/1420 2.03 1.33–3.11 4.31 Cohort nr rMinsk et al. (1996) 157/570 1.21 0.68–2.16 9.49 Case–control nr rWang et al. (1996) 13/70 0.98 0.19–5.02 15.66 Case–control nr rYamada 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 rDe Bruyn et al. (1999) 30/32 0.64 0.2–2.08 24.19 Cohort nr rJones et al. (1999) 126/217 4.06 1.38–11.96 4.66 Case–control nr rWallace (2000) 72/115 2.68 1.04–6.91 10.70 Case–control nr rLambert et al. (2000) 959/1928 1.53 1.15–2.05 6.93 Cohort Current vs. never or former rKuroyama et al. (2001) 1522/2994 1.22 0.93–1.59 5.43 Case–control nr rGeurs et al. (2001) 62/279 2.6 0.99–6.83 5.87 Case–control Current vs. never or former rWidmark et al. (2001) 67/131 5.3 2.53–11.12 20.2 Cohort nr rSchwartz-Arad et al. (2002) 380/579 1.86 0.79–4.34 2.29 Cohort nr rKan et al. (2002) 70/158 2.76 1.16–6.62 10.09 Case–control Current vs. never or former rKaroussis et al. (2003) 28/84 2.08 0.33–13.12 4.46 Cohort nr rLeonhardt et al. (2003) 31/13 3.5 0.39–31.81 18.18 Cohort nr rShiratori & 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.

Fig. 2. Random effects of the odds ratios of osseointegrated implant failure on smoking for individual studies

and overall. The diamond symbol shows the odds ratio and the horizontal line represents the 95% confidence

intervals.

Table 2. Distribution of odds ratio betweensmoking and non-smoking

Group Studies(n)

OR (95% CI)

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 publicationBefore 2000 9 2.05 (1.65–2.55)2000 or more 10 2.50 (1.64–3.82)

Case numbersLess than 200 6 2.57 (1.61–4.12)200 After 13 2.12 (1.61–2.8)

Implant failure rateLess than 10% 12 2.10 (1.56–2.82)10% or more 7 2.53 (1.72–3.73)



Sensitivity analysis for all studies did not

show much difference between study de-

sign, case number and the rate of implant

failure. Our meta-analysis of 19 studies

revealed a statistically significant associa-

tion between smoking and osseointegrated

implant failure by the results of the synthe-

sized OR.

The studies selected to investigate the

influence of intra-oral location (placed in

the maxillary or mandibular arch) are listed

in Table 3. The data of De Bruyn et al.

(1999) were entered into the table of the

maxillary arch, because the number of im-

plant failures in smokers and non-smokers

was only available in the maxillary arch.

Tests of homogeneity of implant failure in

seven papers selected were examined, and

no significant difference was found in the

maxillary arch. Then, the fixed-effects

model was used to calculate OR. The

synthesized OR of the maxillary arch was

2.06 (95% CI: 1.61–2.65). Begg’s method

was carried out to explore the potential

publication bias, and the results of the

maxillary arch showed no significant dif-

ference. The Fail-Safe n calculation for the

maxillary analysis was 100 (45 � 7þ 10)

and publication bias was not considered to

be a large factor; therefore, the synthesized

OR of the maxillary arch was not adjusted.

On the other hand, tests of homogeneity of

implant failure in the mandibular arch

found a significant difference (Po0.05),

and the random-effects model was used.

The synthesized OR was 1.66 (95% CI:

0.89–3.09). No significant difference was

found by Begg’s method whereas the Fail-

Safe n calculation for the maxillary analysis

was 28 (o5 � 6þ 10); therefore publica-

tion bias was considered to be a factor and

the synthesized OR of the mandibular arch

was adjusted (1.32 (95% CI: 0.72–2.4)).

Finally, analysis of implant failure on

smoking by the influence of intra-oral loca-

tion indicated an increased OR in the max-

illary arch but not in the mandibular arch.

Discussion

Sweet (1992) and Bain & Moy (1993)were

the first authors to report that smoking is a

risk factor for osseointegrated implant fail-

ure, and some prospective studies sup-

ported this relationship. However, several

recent studies have indicated no effect of

smoking on the survival of dental im-

plants (Esposito et al. 1998a, 1998b; Berge

& Gronningsaeter 2000; Carlsson et al.

2000; Lambert et al 2000; Quirynen et al.

2001). Although a meta-analysis study was

previously reported (Bain et al. 2002), no

evaluation using the synthesized OR esti-

mates has been published. Therefore, we

constructed and performed this study. Our

meta-analysis of the 19 studies examining

this relationship revealed a significant in-

creased risk for the implant failure among

smokers compared with that among non-

smokers (OR¼ 2.17). Several case–control

studies and cross-sectional studies with

different case numbers and ratio of implant

failure have found similar associations.

Therefore, it confirmed a significant rela-

tionship between smoking and the risk of

osseointegrated implant failure.

In this report, we also attempted to

evaluate the location of the implant failure

on smoking by using a meta-analysis, and

found only seven studies that met the

inclusion criteria. There has been consider-

able discussion regarding implant location

and failure. In our analysis, the maxillary

arch showed a statistically significant in-

crease in the implant failure in smokers

compared with non-smokers; however, the

mandibular arch did not show any differ-

ence. Factors contributing to higher im-

plant failure in maxillary arch than in

mandibular arch are not yet understood.

The high implant failure rate observed in

the maxilla suggests that bone density may

have some bearing upon early implant fail-

ure (Friberg et al. 1991). In addition, smok-

ing has been reported to be the strongest

risk factor for periodontal bone loss (Pau-

lander et al. 2004). On the other hand, it

has been reported that the short-term prog-

nosis is greatly affected by peri-implant

infection related to smoking (Mau 1993;

De Bruyn & Collaert 1994; Gorman et al.

1994). Reibel (2003) suggested that the

increase of implant failures in smokers is

due to exposure of peri-implant tissue to

tobacco smoke, possibly linking the smok-

ing effects on implant survival to the smok-

ing effects on peri-implantitis. Haffajee &

Socransky (2001) showed that a greater

difference in prevalence of orange and red

bacterial complexes (both complexes in-

clude periodontopathogens) exists at sites

with periodontal pocket in the maxilla in

comparison with the mandibular in both

smokers and non-smokers. Smoking might

thus modulate the implant failure by

influencing bacterial infections in the max-

illary and the mandibular arches differ-

ently.

Using meta-analysis, it has been demon-

strated in this study that smoking has

significant negative effects on the survival

rate of dental implants. Dentists have an

important role to play in preventing the

effects of smoking in osseointegrated im-

plant failure and tobacco-related oral dis-

eases. Reibel (2003) suggested in his review

that practitioners should pursue more for-

mal training in smoking cessation counsel-

ing, which should be as much a part of

their job as plaque control and dietary

advice. The evidence presented in this

Table 3. Studies selected in meta-analysis aimed to examine the influence of intra-oral location on osseointegrated implant failure

Author, the yearof publication

In the maxillary arch In the mandible arch

No. cases/no. controls

OR 95% CI No. cases/no. controls

OR 95% CI

Bain & Moy (1993) 196/883 2.78 1.78–4.34 194/921 1.99 0.90–4.39Minsk et al. (1996) 84/285 1.41 0.73–2.72 73/285 0.55 0.12–2.45Yamada et al. (1997) 175/131 2.36 0.84–6.68 214/216 1.77 0.68–4.6De 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 0.19–5.59Lambert et al. (2000) 478/793 1.78 1.19–2.66 481/1135 1.23 0.80–1.9Shiratori & Isokawa (2003) 117/302 16.27 1.94–136.66 186/290 27.20 3.58–206.93



study may contribute to develop tobacco

intervention habits by dental practitioners.

Several investigators have reported that

the surface roughness of implants (fixture)

also represents a factor of implant failure

(Kumar et al. 2002; Feldman et al. 2004).

Feldman et al. (2004) reported that there

was a 9% difference in 5-year cumulative

survival rates in the maxilla between ma-

chined-surfaced short-length (86.8%) and

dual acid-etched short-length implants

(95.8%). It has recently been reported that

no significant difference of the effect of

smoking on achieving initial osseointegra-

tion exists when surface-modified dental

implants were used (Kumar et al. 2002). In

the systematic review of Bain et al. (2002),

93.5% of success rate in smokers for the

implant group with smooth surface and

98.7% for the implant group with rough

surface was observed, a clinically relevant

difference between these two groups. Be-

tween smokers and non-smokers, how-

ever, only 0.3% difference in the success

rate of rough-surface implant was found.

Among pooled data used in this study, no

study had investigated the effect of the

surface roughness of fixture between smo-

kers and non-smokers. Further research is

needed to clarify the influence against sur-

face-modified dental implants in the max-

illary arch on smoking leading to the

results observed in this study.

Acknowledgements: We thank Dr

Makoto Fukui, General Dentistry,

Tokushima University Hospital, who

gave excellent support and assistance.

We are also grateful to Dr Daniel

Grenier, Universite Laval, Canada, for

critical discussion of the study and for

amending the paper.

References

Bain, C.A. & Moy, P.K. (1993) The association

between the failure of dental implants and cigar-

ette smoking. International Journal of Oral &

Maxillofacial Implants 8: 609–615.

Bain, C.A., Weng, D., Meltzer, A., Kohles, S.S. &

Stach, R.M. (2002) A meta-analysis evaluating

the risk for implant failure in patients who smoke.

Compendium of Continuing Education in

Dentistry 23: 695–699, 702. 704 passim; quiz

708.

Begg, C.B. & Mazumdar, M. (1994) Operating

characteristics of a rank correlation test for pub-

lication bias. Biometrics 50: 1088–1101.

Berge, T.I. & Gronningsaeter, A.G. (2000) Survival

of single-crystal sapphire implants supporting

mandibular overdentures. Clinical Oral Implants

Research 11: 154–162.

Carlsson, G.E., Lindquist, L.W. & Jemt, T. (2000)

Long-term marginal peri-implant bone loss in

edentulous patients. International Journal of

Prosthodontics 13: 295–302.

De Bruyn, H. & Collaert, B. (1994) The effect of

smoking on early implant failure. Clinical Oral

Implants Research 5: 260–264.

De Bruyn, H., Collaert, .B., Linden, U., Johansson,

C. & Albrektsson, T. (1999) Clinical outcome of

Screw Vent implants. A 7-year prospective follow-

up study. Clinical Oral Implants Research 10:

139–148.

Esposito, M., Hirsh, J.M., Lekholm, U. & Thom-

sen, P. (1998a) Biological factors contributing to

failures of osseointegrated oral implants. (I). Suc-

cess criteria and epidemiology. European Journal

of Oral Sciences 106: 527–551.

Esposito, M., Hirsh, J.M., Lekholm, U. & Thom-

sen, P. (1998b) Biological factors contributing to

failures of osseointegrated oral implants. (II). Etio-

pathogenesis. European Journal of Oral Sciences

106: 721–764.

Feldman, S., Boitel, N., Weng, D., Kohles, S.S. &

Stach, R.M. (2004) Five-year survival distribu-

tions of short-length (10 mm or less) machined-

surfaced and osseotite implants. Clinical Implant

Dentistry & Related Research 6: 16–23.

Friberg, B., Jemt, T. & Lekholm, U. (1991) Early

failures in 4,641 consecutively placed Branemark

dental implants: a study from stage 1 surgery to

the connection of completed prostheses. Interna-

tional Journal of Oral & Maxillofacial Implants

6: 142–146.

Geurs, N.C., Wang, I.C., Shulman, L.B. & Jeffcoat,

M.K. (2001) Retrospective radiographic analysis of

sinus graft and implant placement procedures

from the Academy of Osseointegration Consen-

sus Conference on Sinus Grafts. International

Journal of Periodontics & Restorative Dentistry

21: 517–523.

Gorman, L.M., Lambert, P.M., Morris, H.F., Ochi,

S. & Winkler, S. (1994) The effect of smoking on

implant survival at second-stage surgery: DICRG

Interim Report No. 5. Implant Dentistry 3:

165–168.

Haffajee, A.D. & Socransky, S.S. (2001) Relation-

ship of cigarette smoking to the subgingival mi-

crobiota. Journal of Clinical Periodontology 28:

377–388.

Johnson, G.K. & Hill, M. (2004) Cigarette smoking

and the periodontal patient. Journal of Perio-

dontology 75: 196–209.

Jones, J.D., Lupori, J., Van Sickels, J.E. & Gardner,

W. (1999) A 5-year comparison of hydroxyapatite-

coated titanium plasma-sprayed and titanium

plasma-sprayed cylinder dental implants. Oral

Surgery Oral Medicine Oral Pathology Oral

Radiology and Endodontics 87: 649–652.

Kan, J.Y., Rungcharassaeng, K., Kim, J., Lozada, J.L.

& Goodacre, C.J. (2002) Factors affecting the

survival of implants placed in grafted maxillary

sinuses: a clinical report. Journal of Prosthetic

Dentistry 87: 485–489.

Karoussis, I.K., Salvi, G.E., Heitz-Mayfield, L.J.,

Bragger, U., Hammerle, C.H. & Lang, N.P.

(2003) Long-term implant prognosis in patients

with and without a history of chronic perio-

dontitis: a 10-year prospective cohort study of

the ITI Dental Implant System. Clinical Oral


Keller, E.E., Tolman, D.E. & Eckert, S.E. (1999)

Maxillary antral-nasal inlay autogenous bone

graft reconstruction of compromised maxilla:

a 12-year retrospective study. International

Journal of Oral & Maxillofacial Implants 14:

707–721.

Kumar, A., Jaffin, R.A. & Berman, C. (2002) The

effect of smoking on achieving osseointegration of

surface-modified implants: a clinical report. Inter-

national Journal of Oral & Maxillofacial Im-

plants 17: 816–819.

Kuroyama, I., Osato, S., Ohtsuka, T., Kubota, R.,

Okudera, H., Okada, M., Saijoh, S., Ogawa, T.,

Ueda, A., Furumoto, K. & Satoh, T. (2001)

A retrospective study of 5,092 implants con-

secutively placed in both maxilla and mandible:

relationship between systemic history, life

habits, intra-oral findings and clinical results

[in Japanese, with English abstract]. Journal of

Japanese Society of Oral Implantology 14:

218–236.



Lambert, P.M., Morris, H.F. & Ochi, S. (2000) The

influence of smoking on 3-year clinical success of

osseointegrated dental implants. Annals of Perio-


Leonhardt, A., Dahlen, G. & Renvert, S. (2003) Five-

year clinical, microbiological, and radiological out-

come following treatment of peri-implantitis in

man. Journal of Periodontology 74: 1415–1422.

Mau, J. (1993) On statistics of success and loss for

dental implants. International Dental Journal 43:

254–261.

Minsk, L., Polson, A.M., Weisgold, A., Rose, L.F.,

Sanavi, F., Baumgarten, H. & Listgarten, M.A.

(1996) Outcome failures of endosseous implants

from a clinical training center. Compendium of

Continuing Education in Dentistry 17: 848–850,

852–854, 856 passim.

Paulander, J., Wennstrom, J.L., Axelsson, P. &

Lindhe, J. (2004) Some risk factors for perio-

dontal-bone loss in 50-year-old individuals. A

10-year cohort study. Journal of Clinical Perio-


Quirynen, M., Peeters, W., Naert, I., Coucke, W. &

van Steenberghe, D. (2001) Peri-implant health

around screw-shaped c. p. titanium-machined

implants in partially edentulous patients with or

without ongoing periodontitis. Clinical Oral


Reibel, J. (2003) Tobacco and oral diseases. Update

on the evidence, with recommendations. Medical

Principles and Practice 12: 22–32.

Rosenthal, R. (1979) The ‘‘file drawer problem’’ and

tolerance for null results. Psychological Bulletin

86: 638–641.

Schwartz-Arad, D., Samet, N., Samet, N. & Mam-

lider, A. (2002) Smoking and complications of

endosseous dental implants. Journal of Perio-


Shiratori, K. & Isokawa, R. (2003) The study of

the risk of implant treatment by smoking and

the support for smoking cessation [in

Japanese]. Quintessence Dental Implantology

10: 57–66.

Sweet, J.B. (1992) The relationship of cigarette

smoking to impaired intra-oral wound healing: a

review of evidence and implications for patient

care. Journal of Oral and Maxillofacial Surgery

50: 239–240.

Wallace, R.H. (2000) The relationship between ci-

garette smoking and dental implant failure. Eur-

opean Journal of Prosthodontics and Restorative

Dentistry 8: 103–106.

Wang, I.C., Reddy, M.S., Geurs, N.C. & Jeffcoat,

M.K. (1996) Risk factors in dental implant failure.

Journal of Long Term Effects of Medical Implants

6: 103–117.

Widmark, G., Andersson, B., Carlsson, G.E., Lind-

vall, A.M. & Ivanoff, C.J. (2001) Rehabilitation of

patients with severely resorbed maxillae by means

of implants with or without bone grafts: a 3–5-

year follow-up clinical report. International

Journal of Oral & Maxillofacial Implants 16:

73–79.

Yamada, Y., Niimi, A., Sawai, T., Watanabe, K.,

Hibino, Y., Nakai, H., Honda, M., Fujimoto, T.,

Murakami, I., Yanbe, M., Suzuki, E., Kasuya, Y.,

Matsuyama, M., Ozeki, K., Goto, Y., Takeuchi,

M., Kawai, M., Ohara, H., Hibi, H. & Ueda, M.

(1997) The influence of cigarette smoking on

osseointegrated implants [in Japanese, with Eng-

lish abstract]. Journal of Japanese Society of Oral

Implantology 10: 163–166.

