Community Dent Oral Epidemiol 2012; 40: 230238 All rights reserved

2011 John Wiley & Sons A/S

Oral health condition of French elderly and risk of dementia: a longitudinal cohort study
E, Letenneur L, Matharan F, Laporte C, Helmer C, Barberger-Gateau P, Arrive Miquel JL, Dartigues JF. Oral health condition of French elderly and risk of dementia: a longitudinal cohort study. Community Dent Oral Epidemiol 2011. 2011 John Wiley & Sons A S Abstract Objectives: Oral condition could be associated with cognitive impairment, but this is not yet well documented. We therefore hypothesized that people with poor oral condition would be more at risk to develop dementia. The objective of this study thus was to describe the oral condition of French community-dwelling elderly persons and to assess its relationship with the occurrence of dementia. Methods: Oral examination was conducted on a sample of individuals aged 6680 years followed-up prospectively for screening of dementia over 15 years in Gironde, France. Univariate and multivariate analyses of the risk of dementia were performed using a Cox proportional hazard model with delayed entry. Results: Data from 405 individuals were analyzed; 45.4% men; median age at baseline: 70 years [interquartile range (IQR): 6875]. The median number of decayed, missing, and lled teeth was 18 (IQR: 1324) and was higher in women (median: 20 versus 17, P = 0.004) and in persons with lower school level (median: 21 versus 17, P = 0.003). Among 348 persons with sextant eligible for periodontal assessment, 2 3 required periodontal care: 5.2% had bleeding observed, 44.8% calculus, 17.8% 45 mm pockets, and 2.9% 6 mm pockets. The incidence of dementia during a median follow-up of 10 years (IQR: 6.513.7) was 19 per 1000 person-years. The adjusted hazard ratio for a number of missing teeth11 (median) on the risk of dementia was 1.13 (95% condence interval, CI = [0.602.12]) in people with higher education (n = 312) and 0.30 (CI = 0.110.79) in persons with lower school level (n = 93) (P for modication effect = 0.0002). Conclusions: Having eleven or more missing teeth seemed to be associated with a lower risk of dementia in people with lower education possibly owing to the suppression of source of chronic inammation.

1,2,3, L. Letenneur2,3, F. E. Arrive Matharan2,3, C. Laporte1, C. Helmer2,3, P. Barberger-Gateau2,3, J. L. Miquel1 and J. F. Dartigues2,3
1 Department of Oral Health, Universite Victor Segalen, Bordeaux, France, 2INSERM, VictorU897, Bordeaux, France, 3Universite Segalen, Bordeaux, France

Key words: cohort; dementia; France; geriatrics; oral health , UFR dOdontologie, 16-20 Elise Arrive Cours de la Marne, 33082 Bordeaux Cedex, France Tel.: 33 (0) 5 57 57 48 10 Fax: 33 (0) 5 57 57 56 30 e-mails: elise.arrive@u-bordeaux2.fr This study has been presented at the IADR meeting held in 2010 in Barcelona, Spain, Poster # 15 21. Submitted 22 December 2010; accepted 24 September 2011

Alzheimers disease is a neurodegenerative cerebral disorder dened as a progressive cognitive impairment and loss of autonomy. It was estimated that 24.3 million people in the world had dementia in 2005, with 4.6 million new cases of dementia every year (one new case every 7 s) (1). The number of people affected should double every 20 years to 81.1 million by 2040, most of them living in industrialized countries (60% in 2001, rising to 71% by 2040). Although knowledge of the physiopathology of Alzheimers disease has greatly progressed over

the past decades, its causal mechanisms and associated factors are far from clear while of high importance to be able to identify preventive interventions. Oral condition could be associated with the cognitive impairment in different ways. First, dietary habits have been shown to play a role in dementia risk (25). The decrease in the masticatory performance owing to missing teeth or not functional denture or oral pain could lead to dietary changes which could therefore modify the risk of dementia. Second, researchers showed that
doi: 10.1111/j.1600-0528.2011.00650.x

230

Elderlys oral health and risk of dementia

mercury was accumulated in the brain and hypothesized that smallest amounts were able to cause nerve cell changes, such as in the Alzheimers disease (6). Mercury is a part of dental amalgams, the main lling used for dental restoration during decades, and can be released at low levels in the body. However, no clinical data demonstrated so far a causal relationship between mercury exposure and dementia (79). Finally, it has been suggested that bad oral condition, particularly periodontal disease, creating chronic peripheral inammation could be associated with an increase in cognitive decline (1013). We therefore hypothesized that people with poor oral condition (dental llings, tooth loss, or bas periodontal status) would be more at risk to develop dementia. The aim of our study was thus to examine the relationship between oral condition and occurrence of dementia over a 15-year period in French older community dwellers.

obtained from all patients participating in the study.

Assessment of dementia and Alzheimers disease

At the baseline screening, subjects were seen at home by a psychologist who completed a questionnaire to collect sociodemographic information and performed neuropsychological testing to assess the criteria for dementia according to the Diagnostic and Statistical Manual of Mental Disorders, Third Edition, Revised (DSM-III R) (15). When the participant was suspected of having dementia on a clinical basis by the psychologist or had declined since a previous examination, (s)he was examined at home by a neurologist to conrm and complete the DSM-III R criteria for dementia and to apply the National Institute of Neurological and Communication Disorders and Stroke Alzheimers Disease and Related Disorders Association criteria for Alzheimers disease (16) and the Hachinski score (17) for vascular dementia. Subjects were then classied as demented or cognitively normal. The same procedure was applied at each follow-up visit at 1, 3, 5, 8, 10, 13, and 15 years.

Methods
Study population
es QUID (PAQUID) Study is a The Personnes Age prospective, community-based cohort study of the epidemiology of dementia and Alzheimers disease in the elderly population of France described elsewhere (14). To summarize, the study initially included 3777 people aged 65 years or older randomly selected from electoral rolls of Gironde and Dordogne in the South West of France and followed-up with an active detection of dementia between 1988 and 2003. A substudy addressing oral health issues was also implemented in Gironde, called PAQUIDENT (PAQUID DENTal study). At the 1-year follow-up visit of the PAQUID Study (19891990), eligible participants underwent dental examination and answered a specic questionnaire. Inclusion criteria were (i) living in Gironde; (ii) aged 6680 years; and (iii) consenting to participate in this substudy. Data from patients having no dementia, complete oral examination at baseline, and follow-up visits were analyzed.

Assessment of potential confounders

Sociodemographic variables consisted of gender, educational level, and living condition. Tobacco and alcohol intake were also recorded. Health status was ascertained by measures of vascular risk factors, depressive symptoms, and body mass index. Vascular risk factors consisted of reported diabetes, hypertension, and history of vascular disease (stroke, angina pectoris, and myocardial infarction). Depressive symptoms were assessed by the Center for Epidemiologic StudiesDepression scale with a threshold at 17 in men and at 23 in women (18).

Assessment of oral health condition

Oral examinations were performed at the patients home, by ten trained and calibrated dentists using mirrors, explorers, and intra-oral light. Findings were recorded on a standardized screening sheet. Data collected included the number of decayed, lled, or missing but not replaced teeth, presence of dentures if they had missing teeth, and presence of temporo-mandibular joint disease. The DMFT [number of decayed (D), missing (M), or lled (F) teeth (T)] index, calculated for 28 teeth, described

Standard protocol approvals, registrations, and patient consents

We received approval from the Consultative Committee for the Protection of Persons participating in Biomedical Research of the Bordeaux teaching hospital. A written informed consent was

231

 et al. Arrive

the burden related to dental caries in an individual. The F-component included usually teeth with lling but, in this population from industrialized setting, we thought that the DMFT would be more accurate if it included also crowned teeth. The number of posterior occluding pairs (POP) was calculated for functional natural teeth only (sound, lled, or crowned) and for all posterior teeth (also decayed not treated or replaced with removable denture). It consisted in the number of pairs of opposing premolars (coded 1) plus molars (coded 2), wisdom teeth excluded (maximum score: 12). A periodontal examination through the Community Periodontal Index (CPI) as per the recommendations of the World Health Organization (19) allowed for the record of gingival bleeding, calculus, and periodontal pockets. A questionnaire addressed the frequency of dental visits, oral hygiene habits, pain or discomfort, and self-reported treatment needs.

Results
Among 2792 persons enrolled in PAQUID in Gironde, 450 persons underwent dental examination (16.1%) between 10 03 1989 and 28 06 1990 (Fig. 1). After exclusion of three subjects with incomplete oral examination and 42 without follow-up, 405 persons were included in the analyses [45.4% men (n = 184); median age at baseline: 70 years (IQR: 6875; min: 66; max: 80); 23% with low school level (n = 93)]. Compared with the rest of the sample (n = 2387), there was a higher proportion of men in the subsample of persons included in the PAQUIDENT analyses; more people had primary school diploma and they were younger (data not shown). Interviewed persons declared, for 161 of them (39.7%), that they felt discomfort to eat with dental origin, for 198 (48.9%), that they needed dental care, and for 109 (26.9%), that they underwent dental consultation at least once a year, whereas 246 (60.7%) declared to undergo dental consultation only when emergency. However, 237 (58.5%) declared to have a dental visit within the past year. Most of the patients (97.5%) beneted from the social security system, 325 (80.2%) subscribed a mutual insurance company, and 94 (23.2%) subscribed a private insurance company. The median DMFT was 18 teeth and was higher in women and in persons with lower school level (Table 1). Most of the DMFT burden was because of the M component (67.5%). Only one patient was quasi edentulous with no denture available; 43 (10.6%) persons had not replaced missing teeth at maxillary only; 79 (19.5%) at mandibulary only, and 99 (24.4%) at both jaws. One hundred and eighty-four persons (45.4%) had no POP of functional natural teeth (sound, lled, or crowned), but only 18 patients (4.4%) had no POP of teeth at all. Persons with lower school level were more likely to have six or less POP of functional natural teeth (94.6% versus 78.2%, P = 0.003). However, there was no difference in terms of overall POP of teeth (including teeth decayed not treated or replaced with removable denture) (25.8% versus 23.4%; P = 0.633). Among 348 persons with sextant eligible for periodontal assessment, 102 (29.3%) had healthy periodontal tissues, while 18 (5.2%) had bleeding observed, 156 (44.8%) calculus, 62 (17.8%) 45 mm pockets, and 10 (2.9%) 6 mm pockets. Persons with lower school level had a greater prevalence of

Statistical analysis
Quantitative variables were described with their median, interquartile range (IQR), minimum (min), and maximum (max), while percentage was calculated for qualitative variables. To facilitate analyses, continuous variables were categorized according to the values of the quartiles or medians. Subjects were classied in the higher school level group if they had passed the French primary school certicate (20). Comparisons were made using Chi-square and KruskalWallis tests. Survival without dementia was estimated using the KaplanMeier method. Analyses of the risk of dementia were performed using a Cox proportional hazard model with delayed entry (21, 22). Age was used as the time scale in the analysis, so the risk of dementia was adjusted nonparametrically for age. To assess the inuence of oral condition-related variables on the risk of dementia, theses variables were used in univariate and multivariate analyses as well as other explanatory variables that had a P-value <0.25 in the univariate analysis. The crude and adjusted hazard ratios (HR) and 95% condence intervals (CI) for each variable were estimated. Dental and periodontal variables were studied in separate analyses because they were too correlated (periodontal assessment possible only where tooth present). All analyses were conducted using SAS software, version 9.1 (SAS Institute, Inc., Cary, NC, USA).

232

Elderlys oral health and risk of dementia

3777 participants in the PAQUID study 985 not living in Gironde

2792 living in Gironde 942 not followed at 1 year and 58 with confirmed dementia

1792 with no dementia and followed at one year

411 participants aged 81 years or more

1381 aged between 66 and 80 years 931 refused to undergo oral examination

450 enrolled in PAQUIDENT 3 with incomplete oral examination and 42 with no follow-up visit 405 in the survival analysis 72 developed dementia during the follow-up

Fig. 1. Flow chart of the patients participating in the PAQUIDENT Study (n = 405), Gironde, France 19882003.

333 diagnosed with no dementia at the end of the follow-up

pockets of 4 mm or more, but not statistically signicant (24.3% versus 19.7%, P = 0.38), particularly those with eleven or more missing teeth (Fig. 2). Seventy-three (18.0%) reported hyposialia, 29 (7.2%; one missing data) pain at the temporomandibular joint, and 24 (5.9%; one missing data) bruxism. During a median follow-up of 10 years (IQR: 6.5 13.7; min = 1; max = 14.7), 72 persons developed a dementia (19 cases per 1000 person-years), including 61 (85%) Alzheimers diseases. The univariate analyses demonstrated that only school level and diabetes were associated with the incidence of dementia (Table 2). To facilitate the interpretation of the results, we chose to focus on missing teeth better than the global DMFT because it constituted the major part of the index. A signicant modication effect

(P = 0.0024) was found between school level and high number of missing teeth (11, the median value), resulting in differential survival curves without dementia according to these variables (Fig. 3). The hazard ratio of missing teeth 11 for the risk of dementia was 1.27 (95% CI: 0.702.31; P = 0.429) in persons with higher school level and 0.40 (95% CI: 0.170.94; P = 0.03) in the other group (Table 3). The signicant results found in univariate analyses remained in multivariate analyses (Table 3). After controlling for known risk factors for dementia, the risk of dementia was signicantly lower in people with a lower school level presenting with eleven or more missing teeth. These ndings could not be explained by differences in terms of deaths or follow-up. Indeed, there were not more deaths in the group of lower school level (50.4% versus 45.2%, P = 0.4), and they were followed until the

233

 et al. Arrive Table 1. Description of dental condition participants of the PAQUIDENT Study (n = 405), Gironde, France 19882003 Index DMFT Db Mc Fd Mc not replaced by denture DMFT In men In women DMFT In persons with lower school level In persons with higher school level M In persons with lower school level In persons with higher school level Posterior occluding pairs of functional natural teeth (sound, lled or crowned) Overall posterior occluding pairs of teeth (including teeth decayed not treated or replaced with removable denture)
a b a

Minimum Quartile 1 Median Quartile 3 Maximum % DMFT P-value 3 0 0 0 0 3 4 3 3 2 0 0 0 13 0 4 1 0 11 13 16 12 10 4 0 5 18 0 11 4 1 17 20 21 17 16 9 1 8 24 1 19 8 3 23 25 26 24 24 17 4 8 28 9 28 23 27 28 28 28 28 28 28 10 10 3.9 67.5 28.6 0.004e 0.003e P < 10)4
e

Number of decayed, missing and lled teeth. Number of decayed teeth. c Number of missing teeth. d Number of lled teeth. e Comparison of medians using KruskalWallis test.

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% HE-HM HE-LM
Pockets

Discussion
This paper was based on a unique set of data on oral condition of elderly persons in France. The DMFT index was high, with a high proportion of missing teeth particularly in persons with lower school level and in women. However, the percentage of decayed teeth was quite low and most of the missing teeth were replaced, whatever the school level and even more in women than in men. This is to compare with a previous study conducted in another region of France among a similar population who had a mean DMFT index of 23.3 4.0 with a low percentage of untreated decayed teeth (17.4%) (23). This revealed the important dental burden of these people in terms of pain, discomfort, and associated costs, reecting the lack of prevention measures but a good system of care in the last decades of the twentieth century in France. In the same way, we showed that the periodontal condition was quite poor with about half of the people requiring dental plaque and calculus removal and one in ve, a tooth scaling and root planning. The questionnaire revealed that most of the people seek dental consultation only in emergency, even if they feel they require oral care and despite the fact that all the people beneted from at least one health insurance support.

LE-HM
Sound

LE-LM

Bleed/calculus

Fig. 2. Periodontal condition by school level and number of missing teeth in the PAQUIDENT Study (n = 405), Gironde, France 19882003. HE-HM, Higher school level High number of missing teeth 11 (n = 102; 38 persons had no eligible sextant); HE-LM, Higher school level Low number of missing teeth <11 (n = 172); LE-HM, Lower school level High number of missing teeth 11 (n = 153; 19 persons had no eligible sextant); LE-LM, Lower school level Low number of missing teeth <11 (n = 27).

same age in both groups (median: 83.9 years versus 82.8 years; P = 0.1). The periodontal condition was not associated with the risk of dementia, neither in univariate analysis nor in multivariate analysis.

234

Elderlys oral health and risk of dementia Table 2. Demographic and oral variables associated with dementia in univariate analyses using Cox proportional hazard model. PAQUIDENT Study (n = 405), Gironde, France 19882003 Variable Demographics Gender (ref: female) Higher school level (ref: no) Living alone (ref: no) General health condition Depressive symptoms (ref: no) Hypertension (ref: no) History of vascular disease (ref: no) Diabetes (ref: no) Consumption Tobacco (ref: no) Wine (ref: no) Dental status DMFTa Continuous <13 [1318] [1824] 24 Filled teeth 0 [14] [48] 4 Missing teeth <4 [411] [1119] 19 Posterior occluding pairs of functional natural teeth (sound, lled or crowned) Continuous 0 [14] 4 Periodontal status Pocket 4 mm Bleeding or calculus Healthy No eligible sextants Other oral conditions Hyposialia (ref: no) Discomfort to eat with dental origin (ref: no) Temporomandibular pain (n = 404) (ref: no) Bruxism (n = 404) (ref: no)
a

Hazard ratio 0.74 0.45 1.48 1.52 0.69 1.56 1.90 1.01 0.80 1.01 1 1.12 1.01 1.38 1 1.10 0.80 0.73 1 1.92 1.77 1.12 0.97 1 0.80 0.84 1 0.79 0.85 1.59 1.16 1.10 0.88 0.71

95% condence interval 0.441.23 0.280.73 0.703.12 0.832.78 0.391.20 0.922.64 1.003.63 0.601.70 0.501.29 0.971.05 0.532.36 0.492.06 0.722.65

P-value 0.211 0.001 0.297 0.175 0.184 0.099 0.050 0.961 0.362 0.616 0.693

0.575 0.582.09 0.411.55 0.371.45 0.459

0.901.05 0.421.50 0.481.46

0.450 0.717

0.135 0.401.56 0.411.75 0.773.25 0.652.06 0.681.76 0.382.05 0.222.28 0.607 0.699 0.772 0.570

Ref, reference. Number of decayed, missing and lled teeth.

This study also explored oral health factors associated with dementia. The number of lled teeth was not associated with risk of dementia although we could estimate that most of the llings at this period consisted in dental amalgams. This did not support the hypothesis that amalgams could increase the risk of dementia owing to released mercury. However, it does not inrm neither because the dental assessment was cross-

sectional, and it is not known whether the people with missing teeth had dental amalgams before they were extracted. Similarly, no relation was found between the presence of amalgam restorations and dementia in a sample of 300 people from the Berlin Aging Study (BASE) (24). We did not show any relationship between the masticatory function, represented by the number of POP of teeth, and the occurrence of dementia. We

235

 et al. Arrive
Higher school level

a 1.0
0.9

Survivor Function Estimate

0.8

Fig. 3. Survival without dementia by school level and number of missing teeth in the PAQUIDENT Study (n = 405), Gironde, France 19882003. (a) Higher school level. (b) lower school level.

0.7

0.6

0.5

0.4

0.3 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 Age (Years)

Lower school level

b 1.0
0.9 Survivor Function Estimate

0.8

0.7

0.6

0.5

0.4 0.3 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 Age (Years)

therefore cannot conrm that impaired masticatory function, by changing dietary habits, could be associated with a higher risk of dementia or mild memory impairment as suggested by Korean and Japanese studies (13, 25). Impaired dental condition was surprisingly found to be associated with a lower risk of dementia, but only in people with lower school level. This could not be explained by higher rate of death or earlier death in the higher school level group because they were similar. We could hypothesize that, in the subsample with lower school level, the absence of teeth may have led to better periodontal condition, avoiding important chronic inammation. Indeed, recent publications showed that acute and chronic inammation was

associated with an increased cognitive decline (10), including inammation immune responses induced by periodontal disease (11, 26). In addition, a serological marker of periodontitis (Porphyromonas gingivalis immunoglobulin G) was found to be associated with impaired delayed memory and calculation (12). However, we could not demonstrate any inuence of periodontal condition, measured with CPI, on the risk of dementia. This may be due to the limitations of such index, which was shown (after the conduction of the PAQUIDENT Study oral examinations) to result in underestimation of adult subjects with pockets and an overestimation of those with healthy periodontium (27). Impaired dental condition was not associated with a lower risk of dementia in the higher education

236

Elderlys oral health and risk of dementia Table 3. Inuence of oral variables on occurrence of dementia in univariate and multivariate analyses stratied on the school level: PAQUIDENT Study (n = 405), Gironde, France 19882003 Higher school level (n = 312) Crude HR Missing teeth <11 11 Periodontal status Sound Bleed Calculus Pockets No eligible sextant
a

Lower school level (n = 93)

a

Adjusted 95% CI HR 1 1.07 1 0.71 0.42 1.51

Crude HR 1 0.40 1 1.20 0.90 1.41 95% CI

Adjusteda HR 1 0.30 1 1.24 0.97 1.02 95% CI

95% CI

1 1.27 1 0.75 0.51 1.81

0.702.31 0.341.64 0.201.35 0.804.12

0.572.02 0.311.63 0.151.15 0.633.57

0.170.94 0.403.57 0.302.71 0.444.48

0.110.79 0.393.88 0.293.19 0.283.66

HR, hazard ratio; CI, condence interval. Controlled for gender, body mass index, diabetes, depression, hypertension and ischemic cardiopathy history of brain stroke.

group likely because they would have a better periodontal condition throughout the life, so the presence of teeth would not be a source of chronic inammation, contrary to the other group. A recent study illustrated this in the Finnish population and showed that low education level was associated higher level of periodontal disease (28). The other main limit of our study is that the oral assessment could not be conducted on all the participants of the PAQUID Study for reasons of feasibility and acceptability which could limit the statistical power and the representativeness of our sample from the whole PAQUID cohort and from the elderly living in Gironde. The generalization of the results to this population should thus be carried out with caution. In conclusion, this study brought valuable data underlining the disparities in oral health among the elderly according to the education level and the need for an equitable lifelong dental and periodontal care and prevention system. Our study did not show any global relationship between oral condition and occurrence of dementia. However, having eleven or more missing teeth was associated with a lower risk of dementia in people with lower education, possibly due to the suppression of source of chronic inammation owing to periodontal disease.

References
1. Ferri CP, Prince M, Brayne C, Brodaty H, Fratiglioni L, Ganguli M et al. Global prevalence of dementia: a Delphi consensus study. Lancet 2005;366:21127. 2. Larrieu S, Letenneur L, Helmer C, Dartigues JF, Barberger-Gateau P. Nutritional factors and risk of incident dementia in the Paquid longitudinal cohort. J Nutr Health Aging 2004;8:1504. 3. Gillette Guyonnet S, Abellan Kan Van G, Andrieu S, Barberger Gateau P, Berr C, Bonnefoy M et al. IANA task force on nutrition and cognitive decline with aging. J Nutr Health Aging 2007;11:13252. 4. Cunnane SC, Plourde M, Pifferi F, Begin M, Feart C, Barberger-Gateau P. Fish, docosahexaenoic acid and Alzheimers disease. Prog Lipid Res 2009;48:23956. 5. Feart C, Samieri C, Rondeau V, Amieva H, Portet F, Dartigues JF et al. Adherence to a Mediterranean diet, cognitive decline, and risk of dementia. JAMA 2009;302:63848. 6. Mutter J, Naumann J, Sadaghiani C, Schneider R, Walach H. Alzheimer disease: mercury as pathogenetic factor and apolipoprotein E as a moderator. Neuro Endocrinol Lett 2004;25:3319. 7. Bates MN, Fawcett J, Garrett N, Cutress T, Kjellstrom T. Health effects of dental amalgam exposure: a retrospective cohort study. Int J Epidemiol 2004;33:894902. 8. Saxe SR, Snowdon DA, Wekstein MW, Henry RG, Grant FT, Donegan SJ et al. Dental amalgam and cognitive function in older women: ndings from the Nun Study. J Am Dent Assoc 1995;126:1495501. 9. Saxe SR, Wekstein MW, Kryscio RJ, Henry RG, Cornett CR, Snowdon DA et al. Alzheimers disease, dental amalgam and mercury. J Am Dent Assoc 1999;130:1919. 10. Holmes C, Cunningham C, Zotova E, Woolford J, Dean C, Kerr S et al. Systemic inammation and disease progression in Alzheimer disease. Neurology 2009;73:76874. 11. Kamer AR, Craig RG, Pirraglia E, Dasanayake AP, Norman RG, Boylan RJ et al. TNF-alpha and antibodies to periodontal bacteria discriminate between

Acknowldgements
We would like to acknowledge the nancial support of Novartis, Ipsen, and CNSA and thank all the persons who agreed to participate in this study.

237

 et al. Arrive Alzheimers disease patients and normal subjects. J Neuroimmunol 2009;216:927. Noble JM, Borrell LN, Papapanou PN, Elkind MS, Scarmeas N, Wright CB. Periodontitis is associated with cognitive impairment among older adults: analysis of NHANES-III. J Neurol Neurosurg Psychiatry 2009;80:120611. Okamoto N, Morikawa M, Okamoto K, Habu N, Hazaki K, Harano A et al. Tooth loss is associated with mild memory impairment in the elderly: the Fujiwara-kyo study. Brain Res 2010;1349:6875. Dartigues JF, Gagnon M, Barberger-Gateau P, Letenneur L, Commenges D, Sauvel C et al. The Paquid epidemiological program on brain ageing. Neuroepidemiology 1992;11(Suppl 1):148. American Psychiatric Association, ed. Diagnostic and statistical manual for mental disorders. DSM III-R, 3rd revised edn. Washington, DC: American Psychiatric Association; 1987. McKhann G, Drachman D, Folstein M, Katzman R, Price D, Stadlan EM. Clinical diagnosis of Alzheimers disease: report of the NINCDS-ADRDA Work Group under the auspices of Department of Health and Human Services Task Force on Alzheimers Disease. Neurology 1984;34:93944. Hachinski VC, Iliff LD, Zilhka E, Du Boulay GH, McAllister VL, Marshall J et al. Cerebral blood ow in dementia. Arch Neurol 1975;32:6327. chelle Fuhrer R, Rouillon F La version franc aise de le C.E. S. -D. (Center for Epidemiology Studies-Depres chelle sion Scale). Description et traduction de le valuation. Psychiatr Psychobiol 1989;4: 1636. dautoe World Health Organization, Community Periodontal Index (CPI) in Oral Health Surveys Basic methods, World Health Organization, Editor. 1997: Geneva. pp. 368. 20. Letenneur L, Gilleron V, Commenges D, Helmer C, Orgogozo JM, Dartigues JF. Are sex and educational level independent predictors of dementia and Alzheimers disease? Incidence data from the PAQUID project J Neurol Neurosurg Psychiatry 1999;66:177 83. 21. Commenges D, Letenneur L, Joly P, Alioum A, Dartigues JF. Modelling age-specic risk: application to dementia. Stat Med 1998;17:197388. 22. Lamarca R, Alonso J, Gomez G, Munoz A. Lefttruncated data with age as time scale: an alternative for survival analysis in the elderly population. J Gerontol A Biol Sci Med Sci 1998;53:M33743. 23. Bourgeois D, Berger P, Hescot P, Leclercq MH, Doury J. Oral health status in 6574 years old adults in France, 1995. Rev Epidemiol Sante Publique 1999;47:559. 24. Nitschke I, Muller F, Smith J, Hopfenmuller W. Amalgam llings and cognitive abilities in a representative sample of the elderly population. Gerodontology 2000;17:3944. 25. Kim JM, Stewart R, Prince M, Kim SW, Yang SJ, Shin IS et al. Dental health, nutritional status and recentonset dementia in a Korean community population. Int J Geriatr Psychiatry 2007;22:8505. 26. Stewart R. Mouths and brains. Could oral infection be a risk factor for dementia? J Neurol Neurosurg Psychiatry 2009;80:1184. 27. Holmgren CJ. CPITN interpretations and limitations. Int Dent J 1994;44:53346. 28. Bernabe E, Suominen AL, Nordblad A, Vehkalahti MM, Hausen H, Knuuttila M et al. Education level and oral health in Finnish adults: evidence from different lifecourse models. J Clin Periodontol 2011;38:2532.

12.

13.

14.

15.

16.

17. 18.

19.

238

This document is a scanned copy of a printed document. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material.