Epidemiology of Periodontal DiseaseCriteria, Aline Mendes Silva de Pinho, Carolina Marques Borges, Mauro Henrique Nogueira Guimara es de Abreu, Eﬁgˆenia Ferreira e Ferreira, and

Epidemiology of Periodontal DiseaseGuest Editors: Fernando Oliveira Costa, Cristiano Susin, Jose Roberto Cortelli, and Isabela Almeida Pordeus

International Journal of Dentistry

Epidemiology of Periodontal Disease

International Journal of Dentistry


Guest Editors: Fernando Oliveira Costa, Cristiano Susin,Jose Roberto Cortelli, and Isabela Almeida Pordeus

Copyright © 2012 Hindawi Publishing Corporation. All rights reserved.

This is a special issue published in “International Journal of Dentistry.” All articles are open access articles distributed under the CreativeCommons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the originalwork is properly cited.

Editorial Board

Ali Abdalla, EgyptYahya Ail, GermanyJasim M. Albandar, USAEiichiro Ariji, JapanManal Awad, UAEAshraf F. Ayoub, UKSilvana Barros, USASema Belli, TurkeyM. A. Rabelo Buzalaf, BrazilGiuseppina Campisi, ItalyFrancesco Carinci, ItalyLim K. Cheung, HongKongBrian W. Darvell, KuwaitHugo De Bruyn, BelgiumDong Mei Deng, The NetherlandsShinn-Jyh Ding, TaiwanJ. D. Eick, USAAnnika Ekestubbe, SwedenCarla Evans, USAVincent Everts, The NetherlandsStefano Fedele, UKG. Nogueira Filho, CanadaRoland Frankenberger, GermanyGerald Glickman, USAValeria V. Gordan, USARosa H. Grande, BrazilYoshitaka Hara, Japan

James K. Hartsfield, USAYumiko Hosoya, JapanSaso Ivanovski, AustraliaChia-Tze Kao, TaiwanElizabeth Kay, UKHeidrun Kjellberg, SwedenKristin Klock, NorwayKee-Yeon Kum, Republic of KoreaManuel Lagravere, CanadaDaniel M. Laskin, USAClaudio R. Leles, BrazilLouis M. Lin, USAA. D. Loguercio, BrazilTommaso Lombardi, SwitzerlandMartin Lorenzoni, AustriaAdriano Loyola, BrazilMaria Machado, BrazilJukka H. Meurman, FinlandHendrik Meyer-Luckel, GermanyKonstantinos Michalakis, GreeceMasashi Miyazaki, JapanYasuhiro Morimoto, JapanCarlos A. Munoz-Viveros, USAHiroshi Murata, JapanRavindra Nanda, USAToru Nikaido, JapanJoseph Nissan, Israel

Chikahiro Ohkubo, JapanAthena Papas, USAPatricia Pereira, USARoberta Pileggi, USAA. B. M. Rabie, Hong KongMichael E. Razzoog, USAAndr Reis, BrazilStephen Richmond, UKGeorge E. Romanos, USAKamran Safavi, USATuula Salo, FinlandGilberto Sammartino, ItalyRobin Seymour, UKTimo Sorsa, FinlandGianrico Spagnuolo, ItalyAndreas Stavropoulos, DenmarkDimitris N. Tatakis, USAShigeru Uno, JapanJacques Vanobbergen, BelgiumMarcos Vargas, USAAhmad Waseem, UKIzzet Yavuz, TurkeyCynthia Yiu, Hong KongLi-wu Zheng, Hong KongQiang Zhu, USASpiros Zinelis, Greece

Contents

Epidemiology of Periodontal Disease, Fernando Oliveira Costa, Cristiano Susin, Jose Roberto Cortelli,and Isabela Almeida PordeusVolume 2012, Article ID 848641, 2 pages

Periodontal Status and Risk Factors among Adults of Sebha City (Libya), Syed Wali Peeran,A. J. A. Ranjith Singh, G. Alagamuthu, Syed Ali Peeran, and P. G. Naveen KumarVolume 2012, Article ID 787502, 5 pages

Effects of Forced Alcohol Intake Associated with Chronic Stress on the Severity of Periodontitis: AnAnimal Model Study, Alessandra Nogueira Porto, Alex Semenoff Segundo,Tereza Aparecida Delle Vedove Semenoff, Fabio Miranda Pedro, A’lvaro Henrique Borges,Jose Roberto Cortelli, Fernando de Oliveira Costa, and Sheila Cavalca CortelliVolume 2012, Article ID 465698, 6 pages

BANA-Positive Plaque Samples Are Associated with Oral Hygiene Practices and Not CD4+ T Cell Countsin HIV-Positive Patients, Cathy Nisha John, Lawrence Xavier Graham Stephen, and Charlene Wilma JoyceAfricaVolume 2012, Article ID 157641, 6 pages

Prevalence of Periodontal Bone Loss in Brazilian Adolescents through Interproximal Radiography,Benedicto Egbert Correa de Toledo, Eliane Maron Barroso, Alex Tadeu Martins, and Elizangela Partata ZuzaVolume 2012, Article ID 357056, 5 pages

Impact of Periodontal Disease on the Quality of Life of Diabetics Based on Different Clinical DiagnosticCriteria, Aline Mendes Silva de Pinho, Carolina Marques Borges,Mauro Henrique Nogueira Guimara es de Abreu, Efigenia Ferreira e Ferreira,and Andrea Maria Duarte VargasVolume 2012, Article ID 986412, 8 pages

Individual and Contextual Determinants of Periodontal Health in 12-Year-Old Schoolchildren in aBrazilian Capital City, Lidia Moraes Ribeiro Jorda o, Daniela Nobre Vasconcelos,Rafael da Silveira Moreira, and Maria do Carmo Matias FreireVolume 2012, Article ID 325475, 7 pages

Prevalence and Reasons for Tooth Loss in a Sample from a Dental Clinic in Brazil,Andreia Affonso Barretto Montandon, Elizangela Partata Zuza, and Benedicto Egbert Correa de ToledoVolume 2012, Article ID 719750, 5 pages

Hindawi Publishing CorporationInternational Journal of DentistryVolume 2012, Article ID 848641, 2 pagesdoi:10.1155/2012/848641

Editorial


Fernando Oliveira Costa,1 Cristiano Susin,2

Jose Roberto Cortelli,3 and Isabela Almeida Pordeus1

1 Federal University of Minas Gerais, Contorno Avenue 4849, 30110031 Belo Horizonte, MG, Brazil2 Laboratory for Applied Periodontal & Craniofacial Regeneration, Center for Clinical and Translational Craniofacial Research,College of Dental Medicine, Georgia Health Sciences University, USA

3 University of Taubate, Taubate, SP, Brazil

Correspondence should be addressed to Fernando Oliveira Costa, [email protected]

Received 3 December 2012; Accepted 3 December 2012

Copyright © 2012 Fernando Oliveira Costa et al. This is an open access article distributed under the Creative CommonsAttribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work isproperly cited.

Epidemiology is the study of health and disease in popu-lations and of how these states are influenced by biology,heredity, and physical and social environment, as well aspersonal behavior. Advances in research over recent yearshave led to a fundamental change in our understandingof the periodontal diseases. As recently as the mid-1960s,the prevailing model for the epidemiology of periodontaldiseases included the following precepts: (1) all individualswere considered more or less equally susceptible to severeperiodontitis; (2) gingivitis usually progressed to periodon-titis with consequent loss of bone support and eventuallyloss of teeth; (3) susceptibility to periodontitis increased withage and was the main cause of tooth loss after age 35–55.Since the development of this paradigm, advances in theunderstanding of periodontal diseases have led this diseasemodel to be reevaluated. Current knowledge has shown thatperiodontitis does not present a linear progression and isnot age dependent. Moreover, its distribution and severityare strongly influenced by host susceptibility and risk fac-tors. Several epidemiological studies evaluating destructiveperiodontal diseases have been pursuing associations in theincessant identification of risk factors for these diseases.Analytical epidemiology seeks to identify the risk factorsassociated with a disease, to quantify the strength of thoseassociations and to estimate whether an association is causal.An understanding of risk factors can lead to theories ofcausation and then to treatment protocols for cliniciansto use in their daily practice. The essential features ofepidemiology as a method of research, when comparedto clinical research and case studies, are that (1) groupsrather than individuals are the focus of study; (2) persons

with and without a particular disease (e.g., periodontaldiseases) and with and without the exposure of interest areincluded, rather than just patients. The study of populationgroups rather than individuals is to allow for valid estimateswhile accounting for normal biological variation (e.g., someindividuals form dental biofilm readily; others do not).Broadening a study to include those with and without adisease can provide a reference point against which a risk isquantified .

Under such background, this special issue was designedto summarize the current status of periodontal epidemi-ology. Thus, we have invited authors to submit originalresearch articles and analysis that seek to clarify aspects of theepidemiology of periodontal diseases and the relevant relatedfactors. As a result, various interesting and scientificallysignificant papers have been accepted. This special sup-plement presents manuscripts reporting information aboutthe prevalence and distribution of periodontitis, as wellas their associations with risk factors. In addition severalstudies reported tooth loss, one of the most visible results ofthe evolution of periodontitis that causes physiological andpsychological impacts on patient’s life.

These papers represent an exciting and insightful obser-vation into the state of art, as well as emerging future topics,in this important interdisciplinary field. We hope that thisspecial issue would attract a major attention of the peers.

Acknowledgment

I greatly appreciate all the Guest Editors shown below fortheir efforts and valuable contributions extended to the

2 International Journal of Dentistry

publication of this special issue. In addition, I thank all thestaff members in Hindawi Publishing Corporation for thekindest help to all of editorial members.

Fernando Oliveira CostaCristiano Susin

Jose Roberto CortelliIsabela Almeida Pordeus


Research Article

Periodontal Status and Risk Factors among Adults ofSebha City (Libya)

Syed Wali Peeran,1 A. J. A. Ranjith Singh,2 G. Alagamuthu,3

Syed Ali Peeran,4 and P. G. Naveen Kumar5

1 Department of Periodontology and Oral Implantology, Faculty of Dentistry, Sebha University, Sebha, Libya2 Department of Advanced Zoology and Biotechnology, Sri Paramakalyani College, Alwarkurichi, India3 Department of Chemistry, Sri Paramakalyani College, Alwarkurichi, India4 Faculty of Dentistry, Gezan University, Gezan, Saudi Arabia5 Department of Community Dentistry, Faculty of Dentistry, Sebha University, Sebha, Libya

Correspondence should be addressed to Syed Wali Peeran, [email protected]

Received 5 August 2012; Accepted 16 October 2012

Academic Editor: Jose Roberto Cortelli

Copyright © 2012 Syed Wali Peeran et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

The present study was aimed at assessing the periodontal status and risk factors like age, gender, tooth brushing habit, and smokingamong the adult population of Sebha city, Libya. 452 adults, aged 35–54 years, comprised the study sample. 266 (58.84%) werefemales and 186 (41.15%) were males. Data was collected by interview and clinical examination using CPI of CPITN index. Chi-square test and ANOVA were used for statistical analysis at 5% level of significance. Results indicate that 76.32% used toothbrushand paste; 8.84% were current smokers and were all males. Majority, 52.65% were, detected with shallow pockets followed by30.08% with calculus, 12.17% had deep pockets, 3.31% had bleeding, and only 1.33% were healthy. Age, gender, current smokingstatus and frequency of tooth brushing showed statistically significant difference with CPI codes. Health professionals can utilizethis data to identify individuals at risk and to target population level interventions.

1. Introduction

Available data indicates that the prevalence of periodontaldisease is very high in several African countries affecting allage groups [1]. Libya is one of the North African countryin the Mediterranean region and Sebha is the largest citysituated in the south of the Libyan Arab Jamahiriya andlocated in the middle of Libyan desert. People come to Sebhafrom the various urban areas surrounding it and thereforeit has a heterogeneous society [2]. Information on the oralhealth status of the Sebha population is scarce. To date,except for the report of Hassan [2] and Leous [3], there is apaucity in publication concerning the oral health in generaland the prevalence and severity of periodontal diseases inparticular.

Since 1993, WHO began collecting data on periodontalstatus from different countries. This data is stored in theWHO Global Oral Data Bank (GODB) [4]. The periodontalcountry profile for Libya is mentioned in the WHO data

base, which is an extract of the 1982-1983 pathfinder study[3], conducted to assess the oral health situation in SocialistPeople’s Libyan Arab Jamahiriya. It was conducted on 849children and 269 adults from 14 sites of the followinglocalities namely Tripoli, Sebha, Benghazi, Zwara, Ajelat, andKaddah. CPITN index was used to assess the periodontaldisease and its results indicate that the periodontal diseaseis present from a very young age. There was increased toothloss by the age of 55–64 years. High prevalence of periodontaldisease was the main cause of tooth loss in adulthood.Among the 35–44 years old, there was none with healthyperiodontium and bleeding. 13% had calculus, 53% hadshallow pockets, and 34% deep pockets. The average numberof sextants with bleeding and higher score was 5.7. Ever sincethe completion of this study, two decades have passed andthere is no new data which has been updated.

Age [5, 6], gender [7, 8], toothbrushing [9, 10], andsmoking [7, 11–13] are among the important risk factors forperiodontal disease. Information of the periodontal status


and additional information of risk factors can help thedental professional, to identify people in the high risk forperiodontal disease and undertake strategic planning for apreventive and therapeutic treatment program.

Against this background, the present study has beendesigned to assess the periodontal status and risk factors likeage, gender, frequency of toothbrushing, and smoking of theadult population in Sebha city, Libya.

2. Material and Methods

This study is a part of the extended project which wasundertaken to evaluate the periodontal status and its deter-minants on a large scale amongst the population of Sebha(details are published elsewhere). Here,data of the adultpopulation aged 35–54 years old is presented. One WHO,index age group of 35–44 years and other group of 45–54 years which are representative of adult population wasconsidered. For sampling, the WHO guidelines (1997) [14]were followed. 4 sites were identified and 50 subjects per sitefor each of the 2 age groups were selected. An additional10 subjects each were added to compensate for any non-participation and hence the sample was 480 (4×2×60). Thesurvey was conducted at the Primary Health Centre (PHC)in Al-Jadeed, Al-Mahdia, Gurda, and Sukkara. The medicalofficer posted at the PHC was responsible to collect theparticipants for the study in the respective areas. People fromhousehold, offices, and those visiting the PHC comprised thestudy sample. All those who agreed to voluntarily participatewere recruited for the survey. The participant was firstinterviewed to collect general information and data about theoral hygiene practices and smoking habits. The participantwas asked simple questions about the oral hygiene aidsused (toothbrush and others), use of toothpaste (yes, no),frequency of toothbrushing (once, twice, >twice), currentsmoker (yes, no), and frequency of smoking (5, 6–10, 11–20,>20/per day). After the interview, the participant underwenta clinical examination. ADA type 3 examination procedurewas followed. CPI of the CPITN was used to record theperiodontal status [14]. The main outcome measures of CPIare presence of gingival bleeding on gentle probing, dentalcalculus, and probing periodontal pocket depth (PPD): 4-5or ≥6 mm. The dentition was divided into six sextants. Theindex teeth were 16, 17, 11, 26, 27, and 36, 37, 31, 46, 47.When two index teeth were considered in a sextant, the toothwith the highest score was recorded. The overall CPI score ofthe participant represented the value of the highest recordedcode for that individual.

The examination was performed by a single, trained,calibrated examiner assisted by a recording clerk, using aCPI probe and mouth mirror. During calibration 10% of thesample was reexamined and kappa value of 0.8 was obtainedwhich indicates a good intra-examiner consistency. All thesubjects were well educated and signed the informed consentform. The ethical board, scientific committee of the Facultyof dentistry, Sebha University, Sebha, Libya approved theprotocol of the study.

Statistical analysis was done with the help of a statisticalpackage for social sciences (SPSS version 16). Mean standard

deviation and percentages were calculated. Chi-square testand ANOVA were used to compare between the groups at5% level of significance. The variables were age, gender,frequency of toothbrushing, and smoking habit, which werecompared with the CPI codes.

3. Results

The sample size calculated was 480, but the survey could becompleted among 452 people, of which 266 (58.84%) werefemales and 186 (41.16%) were males. 249 (55.08%) werein the age group of 35–44 years and 203 (44.92%) in 45–54years, respectively.

Overall, 344 (76.15%) reported use of toothbrush andpaste, 45 (9.96%) used finger, and 63 (13.93%) used siwak.225 (49.78%) brushed once daily, 96 (21.23%) brushedtwice, and 21 (4.64%) brushed more than twice. 40 (8.84%)were current smokers and were all males. 19 (4.2%) reportedto have smoked 5–10 cigarettes, while 12 (2.65%) smoked10–20/per day and 9 (1.95%) smoked more than 20.

Majority, 52.65% (n = 238), were detected with shallowpockets (CPI code 3) followed by 30.08% (n = 136) withcalculus (code 2), 12.17% (n = 55) had deep pockets(code 4), 3.31% (n = 15) had bleeding (code 1), and only1.33% (n = 6) were healthy (code 0). Figure 1 shows thedistribution of the study sample according to the CPI codes.

Table 1 shows the distribution of CPI codes in relation tothe different variables considered in the study. Age, gender,current smoking status, and frequency of toothbrushingshowed statistically significant difference with CPI codes.

Table 2 shows that for 35–44 years, the mean numberof sextant for calculus was significantly higher as comparedto the shallow pockets which was higher in 45–54 years agegroup.

The mean (SD) CPI score for the entire sample was 2.71(0.77). The mean CPI score was 2.57 (0.79) and 2.89 (0.72)for 35–44 years and for 45–54 years, respectively.

4. Discussion

The present study is the first population-based survey amongadult population in Sebha in the recent times after thepathfinder survey conducted in 1982-1983 [3]. Thoroughtraining and calibration of the examiner ensured reliablerecording of the CPI (community periodontal index). Themajor advantages of the CPI are simplicity, speed, repro-ducibility, and international uniformity [15]. Although thisindex has certain shortcomings when used as a stand-alonemeans of assessing the extent and severity of periodontal dis-ease [16] it has been widely used for descriptive periodontalepidemiologic studies. In addition, the CPI data has beenused for surveillance of periodontal health at country andintercountry levels [15].

For the age group 35–44 years, the most frequentlyobserved condition was shallow pockets of 4-5 mm among47.39% followed by calculus in 38.15%. Deep pockets ofmore than 6 mm were found in 8.84% of the subjects. Bleed-ing was found in only 3.61% of the population surveyed.

International Journal of Dentistry 3

Table 1: Number of subjects as per the CPI codes for different variables.

VariableHealthy Bleeding Calculus Shallow pockets Deep pockets Chi-sqaure test

(code 0) (code 1) (code 2) (code 3) (code 4) (P value)

Age (in years)35–44 5 9 95 118 22

22.05 (P = 0.0001)∗45–54 1 6 41 120 33

GenderMale 0 3 54 98 31

14.42 (P = 0.01)∗Female 6 12 82 140 24

Frequency of toothbrushing

None 2 2 18 65 21

42.01 (P = 0.0001)∗Once 0 9 69 127 20

Twice 4 4 41 35 12

>Twice 0 0 8 11 2

Current smoking statusYes 0 0 4 25 11

17.11 (P = 0.001)∗No 6 15 132 213 44

Frequency of smoking

0 6 15 132 213 44

18.28 (P = 0.2)5–10 0 0 2 12 5

10–20 0 0 2 7 3

>20 0 0 0 6 3∗

Statistically significant.

Table 2: Mean number (SD) of sextants affected by periodontal conditions in various age groups.

Age Group Code 0∗ Code 1∗∗ Code 2∗∗∗ Code 3# Code 4##

35–44 yrs 1.36 (1.45) 0.55 (1.02) 2.53 (1.66) 1.30 (1.57) 0.18 (0.69)

45–54 yrs 1.33 (1.29) 0.28 (0.74) 1.84 (1.55) 2.04 (1.67) 0.38 (0.96)

ANOVA: F∗ = 0.09, P = 0.7632; F∗∗ = 9.39, P = 0.0023; F∗∗∗ = 20.29, P < 0.0001; F# = 23.23, P < 0.0001; F## = 6.64, P = 0.0103. SD: Standarddeviation.

Healthy Bleeding Calculus Shallowpockets

Deeppockets

Excludedsextant

2.01 3.61 38.15 47.39 8.84 0

0.49 2.96 20.2 59.11 16.26 0.08

0

10

20

30

40

50

60

70

35–44 years

45–54 years

35–44 years45–54 years

Figure 1: Percentage distribution of the subjects as per the CPIcodes.

Only 2.01% had healthy periodontium. Figure 2 shows thecomparison of periodontal status for 35–44 years old fromEMRO countries with the present study [4]. For the countrieslike Sudan, Iran, Egypt, Morocco, and Libya shallow pockets(code 3) was the highest recorded CPI score. However,calculus (code 2) as the highest score was reported forcountries like Syrian Arab Rep., Cyprus, Pakistan, Lebanon,Saudi Arabia, and Iraq. The highest percentage of healthy

Deep pocketShallow pocketsCalculus

BleedingHealthy

Cyp

rus

Paki

stan

Leba

non

Sau

di A

rabi

a

Iraq

Suda

n

Iran

Egy

pt

Mor

occo

Sebh

a ci

ty, L

ibya

Liby

an A

rab

J.

0

10

20

30

40

50

60

70

80

90

100

Syri

an A

rab

Rep

.

(%)

Figure 2: Overview of periodontal status of 35–44 year old fromdifferent EMRO countries. Data obtained from WHO, Global oraldata bank. Sebha city, Libya-present study.

periodontium (code 0) was reported for Saudi Arabia (20%).34% Libyans and 26% Sudanese were among the ones to havehigh percentage of deep pockets (code 4).

In a recent study among Sudanese [17], 36.1% hadhealthy periodontal tissues, 10.9% bleeding, 42.0% calcu-lus, 8.5% 4-5 mm periodontal pocketing, 0.7% periodontal


pocketing of ≥6 mm, and 1.8% excluded sextants. TheSudanese have better periodontal health as compared to thepresent study population.

When compared with the 1982-1983 pathfinder surveyin Libya [3], a high percentage of people had shallow pocketsand deep pockets versus the present study where percentageof shallow pockets and calculus was more. Even after 20years, there is no improvement in the percentage of healthyperiodontium (0% versus 1.33% in the present study) butthe prevalence of severe form of periodontitis (deep pockets>6 mm) has been reduced from 34% to 12% in the presentstudy.

It is widely assumed that symptoms of periodontal dis-eases escalates with age [6]. In the present study, percent-age of subjects with healthy periodontium, bleeding, andcalculus was comparatively more in the 35–44 years agegroup, while signs of periodontal destruction, that is, shallowpockets, and deep pockets were on a higher side among 45–54 years old (Figure 1).

In the present study, there was a significant statisticaldifference observed between the CPI codes and gender.The percentage calculation showed that the percentage ofhealthy periodontium and bleeding were higher amongfemales and percentage of deep pockets were more preva-lent among males. The percentage of calculus and shal-low pockets were similar among the males and females.Studies indicate that prevalence of periodontal diseasewas slightly lower in females [18]. This may be becausethe females give more importance to aesthetics and havethe habit of toothbrushing and seeking dental care regu-larly.

Smoking can cause a number of changes within theperiodontium which can predispose an individual to theprogression of periodontal disease [19]. In a study, 5.5%of smokers had healthy periodontium compared to 44.2%among non-smokers and the periodontal condition ofthe smokers was very poor compared with that of non-smokers and this difference was statistically significant [19].Literature [11] indicates that predominantly males, due totheir smoking habit, have shown more susceptibility toperiodontal diseases. The smokers in the present study wereall males; significant difference in periodontal status wasobserved with the current smoking status, but when thedata was stratified according to the frequency of smokingno statistically significant difference was found with the CPIcodes.

In the present study, 21.23% people brushed twice dailyin comparison to 53% Sudanese [17] and 32% Chinese [20].The frequency of toothbrushing was significantly associatedwith CPI codes in this study. Lang et al. [21] showed thatfrequent tooth brushing and flossing as well as regular dentalattendance were found to be significantly associated withlower plaque, gingivitis, and calculus scores.

13.93% used siwak for cleaning their teeth in the presentstudy. Studies have reported that in Saudi Arabia, 65% urbanpopulation used siwak on a daily basis [22]. Miswak (siwak)is known to have antibacterial effect on pathogens causingperiodontal disease and caries.

5. Conclusion

This study shows that more than half of the adult populationin Sebha are detected with the signs of destructive periodon-titis. If this trend continues, in the coming years the severityof periodontal disease is bound to increase enormously.In view of the potential association of periodontal diseasewith systemic disease (atherosclerosis, myocardial infarction,stroke, pneumonia, diabetes mellitus, and pregnancy) [23],this current health situation is of public health significanceespecially among adults and needs urgent attention. Hence,the study results can be utilized to identify individuals atrisk of periodontal disease and to target population levelinterventions.

Acknowledgment

The authors wish to thank the Medilinkers Research consul-tancy for providing help in statistical analysis and editing thepaper.

References

[1] C. F. Ndiaye, “Oral Health in the African Region: progress andperspectives of the Regional Strategy,” African Journal of OralHealth, vol. 2, pp. 2–9, 2005.

[2] A. K. Hassan, “Reasons for tooth extraction among patientsin Sebha, Libyan Arab Jamahiriya: a pilot study,” EasternMediterranean Health Journal, vol. 6, no. 1, pp. 176–178, 2000.

[3] P. A. Leous, “Assignment report. Oral health situation,Socialist People’s Libyan Arab Jamahiriya,” 1982-1983, http://whqlibdoc.who.int/emro/-1993/DENT 44 HMD 469.pdf.

[4] The WHO Global Oral Health Data Bank for periodontaldisease, 2012, http://www.dent.niigata-u.ac.jp/prevent/perio/perio.xls.

[5] D. A. Perry and M. G. Newman, “Occurrence of periodontitisin an urban adolescent population,” Journal of Periodontology,vol. 61, no. 3, pp. 185–188, 1990.

[6] J. Baginska, M. Wilczynska-Borawska, and W. Stokowska,“The evaluation of CPITN index among adults living inPodlasie region,” Advances in Medical Sciences, vol. 51, pp.119–121, 2006.

[7] J. J. Hyman and B. C. Reid, “Epidemiologic risk factorsfor periodontal attachment loss among adults in the UnitedStates,” Journal of Clinical Periodontology, vol. 30, no. 3, pp.230–237, 2003.

[8] P. Bouchard, P. Boutouyrie, C. Mattout, and D. Bourgeois,“Risk assessment for severe clinical attachment loss in an adultpopulation,” Journal of Periodontology, vol. 77, no. 3, pp. 479–489, 2006.

[9] H. Loe, “Oral hygiene in the prevention of caries andperiodontal disease,” International Dental Journal, vol. 50, no.3, pp. 129–139, 2000.

[10] R. M. Davies, G. M. Davies, R. P. Ellwood, and E. J. Kay,“Prevention—part 4: toothbrushing: what advice should begiven to patients?” British Dental Journal, vol. 195, no. 3, pp.135–141, 2003.

[11] G. K. Johnson and J. M. Guthmiller, “The impact of cigarettesmoking on periodontal disease and treatment,” Periodontol-ogy 2000, vol. 44, no. 1, pp. 178–194, 2007.

[12] P. E. Petersen, D. Bourgeois, H. Ogawa, S. Estupinan-Day, andC. Ndiaye, “The global burden of oral diseases and risks to oral


health,” Bulletin of the World Health Organization, vol. 83, no.9, pp. 661–669, 2005.

[13] P. Corraini, V. Baelum, C. M. Pannuti, A. N. Pustiglioni, G.A. Romito, and F. E. Pustiglioni, “Risk indicators for increasedprobing depth in an isolated population in Brazil,” Journal ofPeriodontology, vol. 79, no. 9, pp. 1726–1734, 2008.

[14] World Health Organization, Oral Health Surveys, Basic Meth-ods, World Health Organization, Geneva, Switzerland, 4thedition, 1997.

[15] P. E. Petersen and H. Ogawa, “Strengthening the preventionof periodontal disease: the WHO approach,” Journal ofPeriodontology, vol. 76, no. 12, pp. 2187–2193, 2005.

[16] P. N. Papapanou, “Epidemiology of periodontal diseases: anupdate,” Journal of the International Academy of Periodontol-ogy, vol. 1, no. 4, pp. 110–116, 1999.

[17] N. Khalifa, P. F. Allen, N. H. Abu-bakr, M. E. Abdel-Rahman,and K. O. Abdelghafar, “A survey of oral health in a Sudanesepopulation,” BMC Oral Health, vol. 12, pp. 1–9, 2012.

[18] S. P. Jacob, “Global prevalence of periodontitis: a literarurereview,” International Arab Journal of Dentistry, vol. 3, no. 1,pp. 26–30, 2012.

[19] P. Basavaraj and N. Khuller, “Periodontal Health amongSmokers and Non-smokers—a comparative study,” Journal ofIndian Dental Association, vol. 5, no. 1, pp. 92–93, 2011.

[20] L. Zhu, P. E. Petersen, H. Y. Wang, J. Y. Bian, and B. X. Zhang,“Oral health knowledge, attitudes and behaviour of adults inChina,” International Dental Journal, vol. 55, no. 4, pp. 231–241, 2005.

[21] W. P. Lang, D. L. Ronis, and M. M. Farghaly, “Preventivebehaviors as correlates of periodontal health status,” Journalof Public Health Dentistry, vol. 55, no. 1, pp. 10–17, 1995.

[22] M. Al-Otaibi, M. Zimmerman, and B. Angmar-Mansson,“Prevailing oral hygiene practices among urban Saudi Ara-bians in relation to age, gender and socio-economic back-ground,” Acta Odontologica Scandinavica, vol. 61, no. 4, pp.212–216, 2003.

[23] F. A. Scannapieco, “Systemic effects of periodontal diseases,”Dental Clinics of North America, vol. 49, no. 3, pp. 533–550,2005.


Research Article

Effects of Forced Alcohol Intake Associated withChronic Stress on the Severity of Periodontitis: An AnimalModel Study

Alessandra Nogueira Porto,1, 2 Alex Semenoff Segundo,3

Tereza Aparecida Delle Vedove Semenoff,3 Fabio Miranda Pedro,3 Alvaro Henrique Borges,3

Jose Roberto Cortelli,1 Fernando de Oliveira Costa,4 and Sheila Cavalca Cortelli1

1 Nucleus of Periodontal Research, University of Taubate, 12020-340 Taubate, SP, Brazil2 Dental School, University of Cuiaba, 78020-590 Cuiaba, MT, Brazil3 University of Cuiaba, 78020-590 Cuiaba, MT, Brazil4 Department of Periodontology, Federal University of Minas Gerais, 31270-901 Belo Herizonte, MG, Brazil

Correspondence should be addressed to Alessandra Nogueira Porto, [email protected]

Received 27 June 2012; Revised 13 October 2012; Accepted 14 October 2012

Academic Editor: Isabela Almeida Pordeus

Copyright © 2012 Alessandra Nogueira Porto et al. This is an open access article distributed under the Creative CommonsAttribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work isproperly cited.

This study histometrically evaluated the effect of forced alcohol intake by stressed animals on the severity of ligature-inducedperiodontitis in rats. Thirty-two rats were randomly divided in four groups: group GAL—alcohol and ligature; group GASL—alcohol, chronic physical stress, and ligature; GNC—negative control; GPC—positive control. GAL and GASL received 20%ethanol ad libitum, and GNC received water ad libitum for 60 days. After 24 hours of exposition to alcohol intake—by GAL andGASL—immobilization was applied as a chronic stressor in the GASL group for a two-month period, six times a week, in randomhours. The means of the respective groups were statistically compared (Analysis of Variance and Tukey tests, P < 0.05). The mostsevere periodontal breakdown was observed in nonstressed animals which drank alcohol (GAL), followed by stressed animalsexposed to alcohol (GASL). GASL did not differ from the positive control group (GPC). The negative control group showed thelowest values of periodontal breakdown (P < 0.05). Conclusions. Non-stressed alcohol consumer animals showed the most severepattern of periodontal breakdown. Although stressed animals which were forced to drink alcohol showed poorer periodontal statusthan the negative controls, their results were similar to those of positive controls.

1. Introduction

Periodontal disease is a multifactorial infectious disease trig-gered by the development of dental biofilm, which can harmperiodontal tissues. There are several factors which influ-ence periodontitis: personal and social characteristics, behav-ioral, systemic, genetic and dental factors, and the microbialcomposition of the dental biofilm [1, 2].

Stress may be defined as the combination of physicaland mental responses caused by certain external stimuli(stressors) which allow an individual (human or animal) toovercome certain environmental demands, and the physicaland mental weariness caused by the stress process itself [3].

Stress has been related not only to a higher occurrence ofperiodontal disease but also to its severity [4]. Other socialand behavioral factors that could be related to periodontalstatus are tobacco use, social economic status, nutritionalstatus, psychological aspects, and abusive alcohol intake [1].Among these factors, alcohol has become relevant because itis one of the few socially accepted psychotropic drugs. Thissocial acceptance has contributed to a rise in alcoholism,which has become a serious public health problem [5].According to Silva Furtado Amaral et al. [6] the WorldHealth Organization has reported that the rates of death andfunctional limitations associated with abusive alcohol intakehave surpassed those associated with tobacco smoking.


Alcoholism per se imposes a state of stress because, besides itsphysiological demand, it acts as a chemical aggressor to thebody. All the alcoholic drinks consumed by an individual donot remain stored in the body and, before being eliminated,it is first metabolized in the liver, which results in high levelsof reactive oxygen species [7, 8].

Epidemiological studies have related periodontal diseaseto the consumption of alcoholic drinks. Several cross-sec-tional studies have verified an association between the abu-sive consumption of alcohol and poor periodontal health [9–11]. Abusive consumption of alcohol has also been associatedwith the severity of periodontitis [2]. However, many ques-tions on this subject are still not answered and laborato-rial research could be helpful to clarify these doubts. Animalmodel studies may more specifically help elucidate questionsrelated to alcohol intake, stress, and periodontal disease[12, 13].

As a lifestyle factor, alcohol intake could vary understressful situations. As isolate factors, drinking, and stressseem to contribute to poorer periodontal status. In the pre-sent study it was hypothesized that alcohol consumption bystressed animals could be accompanied by an increase in theseverity of periodontitis. Therefore, aiming to confirm orreject this hypothesis we evaluated whether forced alcoholintake by stressed animals affects histometrical parameters ofperiodontal breakdown.

2. Materials and Methods

Thirty-two animals of the Rattus norvegicus species Lewisbreed (two-month old), weighing average 250 g, were sel-ected for the present study. All the selected animals wentthrough an environmental adaptation period of four weeks.The animals were divided in groups of four and each groupwas kept in a cage (polyethylene 16 × 40 × 30) [14, 15].They all received standard rat feed and water or alcohol (eth-anol 20%) ad libitum. They were kept in light/dark cyclesof 12 hours; the temperature was controlled at 24◦C andhumidity at >50%. The experiment took place at theUniversity Center of Varzea Grande—UNIVAG and it wasapproved and registered by the ethics in animal researchcommittee (CEP/UNIC-2009 no. 307-321) of the Institution.

2.1. Experimental Groups. Initially, a research assistant ran-domly divided the animals in four experimental groups:

(1) group GAL: alcohol + ligature (n = 8);

(2) group GASL: alcohol + stress + ligature (n = 8);

(3) group GPC: stress + ligature, positive control (n = 8);

(4) group GCN: negative control (n = 8).

After the division, the animals in the GAL and GASLgroups were offered alcohol (ethanol 20%) ad libitum for 60days [11, 16] while negative control (GCN) received water adlibitum for the same period of time.

2.2. Experimental Periodontal Disease. Because alcoholintake is very limited on the first day, ligatures were insertedonly in the second experimental day (GAL, GASL, and GPC).

The animals received general anesthesia through the intra-muscular administration of 0.05 mL xylazine hydrochloride(Rompun, Bayer. Animal Health, Sao Paulo, SP. Brazil)and 0.1 mL ketamine (Dopalen, Agribrands. Animal Health,Paulınia, SP, Brazil) per 100 grams of body weight.

After anesthesia, a sterile silk suture number 4 (Ethicon,Johnson and Johnson, Sao Paulo, SP, Brazil) was placed in thegingival crevice of the second right upper molar [17].

The animals in the positive control group were kept inthe same environment in the first experimental day, drinkingwater, but without suffering any kind of intervention. Theanimals in the negative control group (GNC) did not receiveany type of intervention but were kept in their cages in thesame environment during all the study.

2.3. Stress Induction. The model of physical stress inductionchosen for this study was immobilization done during 60days, six times a week, in different times of the day as previ-ously described [18, 19]. At room temperature, the animalsin the GASL group were placed in polyvinyl chloride tubescompatible to their body sizes. The tubes were closed on bothends with metallic wire, which enabled the animals to breathewhile they were immobilized for four consecutive hours.

2.4. Histological Examination. After sixty days all the animalswere euthanized. Right maxilla was removed and fixed informaldehyde 10% for 48 hours. After that process, the jawsegment was decalcified in EDTA for approximately fiveweeks (EDTA was renewed six times); following dehydrationin graded alcohol, the samples were embedded in paraffin.The tissue blocks included in paraffin allowed 6 µm histolog-ical slices that followed the long axis of the teeth in the mesio-distal plane and were stained with hematoxylin and eosin.

For the histometric analysis, ten serial sections contain-ing the 1st and 2nd upper molars and the following struc-tures (a) dental pulp, (b) the mesial cementoenamel junctionof the 2nd molar, (c) proximal bone crest, and (d) connectiveattachment had to be observed. Histometry was performedin the mesial aspect from first molars, by the same blindedexaminer, through the capturing of images in the microscopeand measured in millimeters (software ImageLab 2000-Diracon Bio Informatica Ltda., SP, Brazil).

2.5. Statistical Analysis. The intraexaminer reproducibilitywas calculated before and during the histometry. In the cali-bration previous to the study, the error was calculated bydouble measurements of 10% of the specimens with a one-week interval. During the study, after every 10 examined spe-cimens, one was reexamined. Paired t-test statistics were runand no differences were observed in the mean values for com-parison (P > 0.5). Additionally, Pearson’s correlation coef-ficient was obtained between the 2 measurements in eachtime point and revealed a very high correlation (0.98 and0.99; P < 0.01).

Two histometrical parameters were considered as peri-odontal status indicators. The distance between the cemen-toenamel junction and the bottom of the junctional epithe-lium (CEJ-JE) in the mesial site of the first molars wasdefined as the loss of histological attachment. In addition,


Figure 1: Distance between the CEJ in the mesial side of the secondmolar and the most apical portion of the junctional epithelium(CRJ-JE) and the distance between the CEJ and the alveolar bonecrest (CEJ-BC). 4x magnificence. Image captured from the alcoholassociated with stress group (GASL).

the distance between the cementoenamel junction and thealveolar bone crest was also analyzed and it was considered asthe indicator for bone loss (CEJ-BC). The mean values werecalculated for each of the groups for further statistical com-parisons. Analysis of Variance (ANOVA) and Tukey (P <0.05) tests were applied.

3. Results

There were no statistically significant differences among thebody weights of the groups at any of the evaluation periods(ANOVA, P < 0.05). Loss of histological attachment (CEJ-JE) and bone loss (CEJ-BC) (P < 0.05) were statistically dif-ferent in the animals exposed to alcohol (GAL) and the onesexposed to alcohol and stress (GASL). The most severe per-iodontal breakdown was observed among the nonstressedanimals exposed to alcohol, followed by the group of stressedanimals exposed to alcohol. This last group (GASL) didnot differ from the positive control group (GPC). Figures 1and 2 illustrate results from GASL group at two differentexperimental times. Finally, the negative control groupshowed the lowest values of CEJ-JE and CEJ-BC when com-pared to the animals in the other groups (Table 1).

4. Discussion

The severity and rate of the progression of periodontitisare influenced by a great variety of determinants and riskfactors. Oral hygiene, tobacco smoking, diabetes, age, andthe presence of specific microorganisms can be considered asrisk factors, since individuals under these conditions seem to

Figure 2: Image captured from the same animal illustrated inFigure 1 (alcohol associated with stress group–GASL) before per-iodontitis induction.

Table 1: Mean values (mm) of loss of histological attachment (CEJ-JE) and bone loss (CEJ-BC) for the different groups.

CEJ-JE CJE-BC

(mean ± SD) (mean ± SD)

Stress + alcohol 0.33 ± 0.09a 0.61 ± 0.03a

Nonstress + alcohol 0.77 ± 0.27b 1.09 ± 0.32b

Positive control 0.25 ± 0.16a 0.92 ± 0.20a

Negative control 0.06 ± 0.14c 0.21 ± 0.02c

Different letters within the same column indicate statistically significantdifferences between the groups (P < 0.05). CEJ-JE: loss of histologicalattachment; CEJ-BC: bone loss.

be more likely to develop periodontitis [20]. Consistent datain the literature suggests that they represent serious effectsover periodontal diseases [21]. However, the combination ofthese factors is not enough to explain the difference in theprogression rate and severity of the disease. Other systemicconditions such as stress [22], osteoporosis [23], obesity [24],and abusive alcohol drinking [6] have been considered aspartially responsible for this variability as well.

The effects of alcohol to the body and the functionalimbalances associated with its consume have been constantlyinvestigated [25]. Such interest is justified by the wide rangeof injuries and toxic effects that alcohol drinking causes theorgans and tissues, which can reflect in different systemicdiseases [26]. Besides, The World Health Organization esti-mates that two billion people around the world are alcoholconsumers, and that 76 million show some kind of systemicadverse event related to that. In Brazil, it is estimated that68% of the population consumes alcohol and that 11.2% arealcoholics.


De Souza et al. [27] demonstrated that alcohol consump-tion directly framed the progression of bone loss in aninduced periodontitis model. On the other hand, instead ofa direct action of alcohol over the periodontal tissues, theseeffects are based on poor oral hygiene, which could betterexplain a higher occurrence and severity of periodontal dis-ease in people who drink alcohol [25]. In this context, animalstudies are important as they may help understand the bio-logical events once they allow the elimination of certain biasby the control of some variables. Therefore, animal studieshave observed that alcohol affected the periodontal tissuesthrough different mechanisms [28]. Reduced host responses,due to deficiency and functional changes in neutrophils, aswell as changes in the protein metabolism and in the healingof tissues have been previously reported [29]. In the presentstudy alcohol drinking negatively impacted the progressionof periodontitis. Similarly De Souza et al. [27] also observeda direct effect of alcohol drinking in the bone loss progressionin a ligature-induced periodontitis model.

One common question is in which level the alcohol asa water substitute under experimental conditions reflectshuman alcohol drinking. The answer to that is not simple,especially considering that the metabolism of rats is fasterthan that of humans. According to The World Health Org-anization mild drinkers are men who drink 21 alcohol units(1 unit = 10 grams) per week. For women, this number is 14units per week. For example, a glass of red wine (90 mL) has12% or 1.7 unit of alcohol. A beer has 5% (1.1 unit) whiledistilled drinks such as vodka and whisky have 40% or 2.0alcohol units [5]. Supporting dose dependence, De Souzaet al. [30] reported a significantly higher alveolar bone lossin rats receiving 20% ethanol in comparison to rats receiv-ing 10% ethanol or water. Tezal et al. [11] carried an inter-esting cross-sectional study to evaluate the effect of the con-sumption of different alcoholic drinks (wine, beer, andliquor) on the severity of periodontal disease. They observedthat the effects over the periodontal tissues did not changedepending on the kind of drink taken. These same authorsalso verified contrary activity of alcohol over some periodon-tal pathogens such as Actinobacillus actinomycetemcomitansand Porphyromonas gingivalis. Clinical results suggested thatalcohol seems to affect more severely the gums, followed bythe periodontal ligament and the alveolar bone and that theeffect of alcohol on periodontal disease may depend on thedose, frequency, and time of drinking [29].

Shimazaki et al. [9] also demonstrated that individualsthat consumed more than 15 grams of alcohol a day hada significant increase in the progression rate of periodontaldisease and a higher inflammatory infiltrate, besides a highernumber of periodontal pockets when compared to those whodid not consume alcohol.

Although, dental biofilm acts besides some recognizedrisk factors such as tobacco use, as previously mentioned,more recently, researchers are considering whether and inwhat degree lifestyle could affect the periodontium. In thiscontext, stress seems to affect not only periodontal statusbut also personal lifestyle. Individuals exposed to stress-ful situations tended to smoke a higher number of cigarettes[31] and become alcohol consumers [32]. Also, a decrease in

oral hygiene was observed in a group of stressed individuals[10]. Furthermore, direct tissue damage could be attributedto the stress process [4]. Due to the frequent combinedoccurrence of the studied factors, this study evaluated theeffect of alcohol intake by stressed animals. The fact thatthe stressed animals exposed to alcohol showed betterperiodontal conditions than the nonstressed animals wasonly a partially unexpected result of this study. Yaroslavskyand Tejani-Butt [33] evaluated the relation between stressand alcohol consumption and observed that rats exposedto chronic stress consumed a greater amount of alcohol. Inaddition, changes in central dopamine type-2 receptor siteswere found indicating an altered dopamine neurotransmis-sion following stress and alcohol exposure. Since stressedanimals consumed more alcohol, the authors concluded thatit is possible that the consumption of alcohol reverses thesealterations related to dopamine, suggesting a self-medicatingphenotype. Generally, alcohol intake tends to be confinedto the weekends. Jimenez-Ortega et al. [34] compared theimmunological effect of the discontinuous feeding of aliquid diet containing a moderate amount of ethanol to thatof continuous ethanol administration or a control diet. Thediscontinuous alcohol group received the ethanol diet for 3days and the control liquid diet for the remaining 4 days ofeach week (for a total of 4 weeks). The authors concludedthat discontinuous drinking of a moderate amount of eth-anol can be more harmful for the immune system than a con-tinuous ethanol intake, as the model applied in the presentstudy. Those authors suggested that probably the discon-tinuous alcohol intake induces a greater stress as indicatedby the searched immune indicators.

Therefore, future studies must be carried as to elucidatethe isolate and combined effects of alcohol and stress overperiodontal tissues. Our data suggest that although stressand alcohol may be harmful separately, their association pro-bably triggered different processes. The evaluation of thesefactors is relevant because the combination of stress andalcohol drinking is a frequent condition in human beingswith periodontitis. Although the association did not result asexpected, the effects of alcohol drinking confirmed to be animportant factor for the development of a periodontal dis-ease.

5. Conclusions

Stressed animals which drank alcohol showed worse peri-odontal status than negative controls. However, nonstressedanimals who also consumed alcohol for the same periodshowed a pattern of periodontal breakdown even moresevere. Therefore, the factors of alcohol drinking and stressdo not seem to show any synergic effect.

Acknowledgments

The present study was partially supported by Foundation ofResearch Support of Minas Gerais (FAPEMIG; Brazil-APQ-00207-11) and National Research Council (CNPq; Brazil-474235/2010-6).


References

[1] M. E. Nunn, “Understanding the etiology of periodontitis: anoverview of periodontal risk factors,” Periodontology 2000, vol.32, pp. 11–23, 2003.

[2] T. W. Stanford and T. D. Rees, “Acquired immune suppressionand other risk factors/indicators for periodontal diseaseprogression,” Periodontology 2000, vol. 32, pp. 118–135, 2003.

[3] H. Selye, “Thymus and adrenals in the response of theorganism to injuries and intoxication,” British Journal ofExperimental Pathology, vol. 17, pp. 234–248, 1936.

[4] F. N. Hugo, J. B. Hilgert, M. C. Bozzetti et al., “Chronic stress,depression, and cortisol levels as risk indicators of elevatedplaque and gingivitis levels in individuals aged 50 years andolder,” Journal of Periodontology, vol. 77, no. 6, pp. 1008–1014,2006.

[5] E. Hornecker, T. Muuss, H. Ehrenreich, and R. F. Mausberg, “Apilot study on the oral conditions of severely alcohol addictedpersons,” The Journal of Contemporary Dental Practice, vol. 4,no. 2, pp. 51–59, 2003.

[6] C. Da Silva Furtado Amaral, R. R. Luiz, and A. T. T. Leao, “Therelationship between alcohol dependence and periodontaldisease,” Journal of Periodontology, vol. 79, no. 6, pp. 993–998,2008.

[7] C. S. Lieber, “Metabolism of alcohol,” Clinics in Liver Disease,vol. 9, no. 1, pp. 1–35, 2005.

[8] M. Valko, D. Leibfritz, J. Moncol, M. T. D. Cronin, M. Mazur,and J. Telser, “Free radicals and antioxidants in normal phy-siological functions and human disease,” International Journalof Biochemistry and Cell Biology, vol. 39, no. 1, pp. 44–84, 2007.

[9] Y. Shimazaki, T. Saito, Y. Kiyohara et al., “Relationshipbetween drinking and periodontitis: the Hisayama study,”Journal of Periodontology, vol. 76, no. 9, pp. 1534–1541, 2005.

[10] S. Shizukuishi, N. Hayashi, H. Tamagawa et al., “Lifestyle andperiodontal health status of Japanese factory workers,” Annalsof Periodontology, vol. 3, no. 1, pp. 303–311, 1998.

[11] M. Tezal, S. G. Grossi, A. W. Ho, and R. J. Genco, “The effectof alcohol consumption on periodontal disease,” Journal ofPeriodontology, vol. 72, no. 2, pp. 183–189, 2001.

[12] T. Breivik, Y. Gundersen, H. Osmundsen, F. Fonnum, and P.K. Opstad, “Neonatal dexamethasone and chronic tianeptinetreatment inhibit ligature-induced periodontitis in adult rats,”Journal of Periodontal Research, vol. 41, no. 1, pp. 23–32, 2006.

[13] R. T. Khisti, K. N. Boyd, S. Kumar, and A. L. Morrow, “Sys-temic ethanol administration elevates deoxycorticosteronelevels and chronic ethanol exposure attenuates this response,”Brain Research, vol. 1049, no. 1, pp. 104–111, 2005.

[14] H. A. Hogan, F. Argueta, L. Moe, L. P. Nguyen, and H. W.Sampson, “Adult-onset alcohol consumption induces osteope-nia in female rats,” Alcoholism: Clinical and ExperimentalResearch, vol. 25, no. 5, pp. 746–754, 2001.

[15] K. Irie, T. Tomofuji, N. Tamaki et al., “Effects of ethanol con-sumption on periodontal inflammation in rats,” Journal ofDental Research, vol. 87, no. 5, pp. 456–460, 2008.

[16] D. M. De Souza, L. H. Ricardo, M. D. A. Prado, F. D. A.Prado, and R. F. Da Rocha, “The effect of alcohol consumptionon periodontal bone support in experimental periodontitis inrats,” Journal of Applied Oral Science, vol. 14, no. 6, pp. 443–447, 2006.

[17] A. S. Segundo, K. Hennemann, V. R. Fontanella, and C. K.Rosing, “The role of psychoneuroimmune interactions in thepathogenesis of ligature-induced periodontal disease in Wistarrats,” Journal of the International Academy of Periodontology,vol. 9, no. 1, pp. 26–31, 2007.

[18] L. Shapira, I. Frolov, A. Halabi, and D. Ben-Nathan, “Exper-imental stress suppresses recruitment of macrophages butenhanced their P. gingivalis LPS-stimulated secretion of nitricoxide,” Journal of Periodontology, vol. 71, no. 3, pp. 476–481,2000.

[19] A. Semenoff-Segundo, A. N. Porto, T. A. Semenoff et al.,“Effects of two chronic stress models on ligature-inducedperiodontitis in Wistar rats,” Archives of Oral Biology, vol. 57,pp. 66–72, 2012.

[20] J. M. Albandar, “Global risk factors and risk indicators for per-iodontal diseases,” Periodontology 2000, vol. 29, no. 1, pp. 177–206, 2002.

[21] M. Razali, R. M. Palmer, P. Coward, and R. F. Wilson, “Aretrospective study of periodontal disease severity in smokersand non-smokers,” British Dental Journal, vol. 198, no. 8, pp.495–498, 2005.

[22] R. J. Genco, A. W. Ho, S. G. Grossi, R. G. Dunford, and L.A. Tedesco, “Relationship of stress, distress, and inadequatecoping behaviors to periodontal disease,” Journal of Periodon-tology, vol. 70, no. 7, pp. 711–723, 1999.

[23] B. J. Edwards and C. A. Migliorati, “Osteoporosis and itsimplications for dental patients,” Journal of the AmericanDental Association, vol. 139, no. 5, pp. 545–552, 2008.

[24] A. L. Dumitrescu and M. Kawamura, “Involvement of psy-chosocial factors in the association of obesity with periodon-titis,” Journal of Oral Science, vol. 52, no. 1, pp. 115–124, 2010.

[25] G. Novacek, U. Plachetzky, R. Potzi et al., “Dental and perio-dontal disease in patients with cirrhosis—role of etiology ofliver disease,” Journal of Hepatology, vol. 22, no. 5, pp. 576–582, 1995.

[26] J. C. Crabbe, “Use of genetic analyses to refine phenotypesrelated to alcohol tolerance and dependence,” Alcoholism:Clinical and Experimental Research, vol. 25, no. 2, pp. 288–292,2001.

[27] D. M. De Souza, L. H. Ricardo, M. D. A. Prado, F. D. A.Prado, and R. F. Da Rocha, “The effect of alcohol consumptionon periodontal bone support in experimental periodontitis inrats,” Journal of Applied Oral Science, vol. 14, no. 6, pp. 443–447, 2006.

[28] T. Tomofuji, T. Sanbe, D. Ekuni et al., “Oxidative damage of ratliver induced by ligature-induced periodontitis and chronicethanol consumption,” Archives of Oral Biology, vol. 53, no. 12,pp. 1113–1118, 2008.

[29] M. Tezal, S. G. Grossi, A. W. Ho, and R. J. Genco, “Alcohol con-sumption and periodontal disease: the third national healthand nutrition examination survey,” Journal of Clinical Peri-odontology, vol. 31, no. 7, pp. 484–488, 2004.

[30] D. M. de Souza, L. H. Ricardo, K. Z. Kantoski, and R. F. daRocha, “Influence of alcohol consumption on alveolar bonelevel associated with ligature-induced periodontitis in rats,”Brazilian Oral Research, vol. 23, no. 3, pp. 326–332, 2009.

[31] Q. M. Biggs, C. S. Fullerton, J. J. Reeves, T. A. Grieger, D.Reissman, and R. J. Ursano, “Acute stress disorder, depression,and tobacco use in disaster workers following 9/11,” AmericanJournal of Orthopsychiatry, vol. 80, no. 4, pp. 586–592, 2010.

[32] M. F. Lopez, T. L. Doremus-Fitzwater, and H. C. Becker,“Chronic social isolation and chronic variable stress duringearly development induce later elevated ethanol intake in adultC57BL/6J mice,” Alcohol, vol. 28, pp. 102–114, 2010.

[33] I. Yaroslavsky and S. M. Tejani-Butt, “Voluntary alcohol con-sumption alters stress-induced changes in dopamine-2 recep-tor binding in Wistar-Kyoto rat brain,” Pharmacology Bio-chemistry and Behavior, vol. 94, no. 3, pp. 471–476, 2010.


[34] V. Jimenez-Ortega, M. P. Fernandez-Mateos, P. C. Barquilla, D.P. Cardinali, and A. I. Esquifino, “Continuous versus discon-tinuous drinking of an ethanol liquid diet in peripubertal rats:effect on 24-h variation of lymph node and splenic mitogenicresponses and lymphocyte subset populations,” Alcohol, vol.45, no. 2, pp. 183–192, 2011.


Research Article

BANA-Positive Plaque Samples Are Associated with Oral HygienePractices and Not CD4+ T Cell Counts in HIV-Positive Patients

Cathy Nisha John,1 Lawrence Xavier Graham Stephen,1 and Charlene Wilma Joyce Africa2

1 Department of Periodontics and Oral Medicine, University of the Western Cape, Bellvile 7535, South Africa2 Anaerobe Group, Department of Medical Biosciences, University of the Western Cape, Modderdam Road, Bellville 7535, South Africa

Correspondence should be addressed to Charlene Wilma Joyce Africa, [email protected]

Received 6 July 2012; Revised 3 October 2012; Accepted 4 October 2012


Copyright © 2012 Cathy Nisha John et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

Background. The “red complex” microorganisms, namely, Porphyromonas gingivalis, Treponema denticola, and Tannerella forsythiaare considered as potential pathogens causing HIV-associated periodontal diseases. Moreover, it has been recognized that anassociation exists between CD4+ T cell counts and periodontal disease progression. Objective. To establish whether CD4+ T cellcounts or oral hygiene plays a greater role in producing BANA-positive results in HIV-associated periodontal disease. Materials andMethods. One hundred and twenty HIV-positive patients participated in the study, and their CD4+ T cell counts were obtainedfrom their medical records. The six Ramfjord teeth were used for evaluating periodontal clinical indices and subgingival plaquesampling. BANA test was used for the detection and prevalence of the “red complex” bacteria in plaque samples. Results. A majorityof 69.17% HIV-positive patients were BANA-positive. No significant associations were found between BANA and CD4+ T cellcounts. A highly significant association was found between BANA with probing depth and clinical attachment level (P ≤ 0.0001)and between BANA and the use of interdental aids (P = 0.0168). Conclusion. HIV-associated periodontal diseases are stronglyrelated to oral hygiene practices rather than the effect of CD4+ T cell counts, and the use of interdental aids was marked as asignificant predictor of BANA-negative plaque samples.

1. Introduction

Progressive human immunodeficiency virus infection (HIV)results in loss of immune response, especially those coordi-nated by CD4+ T lymphocyte cells [1]. Depletion of CD4 cellcount renders the host vulnerable to opportunistic infectionsof the mouth [2]. A decrease in CD4+ T cell count aswell as inadequate oral hygiene practices results in BANA-positive plaque accumulation and ultimately periodontalinflammation with destruction of the supporting toothstructures. The gradual increase in the proportions of gram-negative, anaerobic periodontopathogens, and their toxinscan be released into the blood stream, thus activating theimmune system [3, 4]. Inadequate oral hygiene can beconsidered as a risk indicator and/or risk predictor in thedevelopment of periodontal manifestations [5] which maybe exacerbated by HIV infection.

The “red complex” microorganisms, namely, the potentperiodontopathogenic bacteria Porphyromonas gingivalis,

Tannerella forsythia, and Treponema denticola are consideredas risk factors in HIV patients [6]. A rapid and simplechair-side bacteriological enzymatic test, BANA (N-benzoyl-DL-arginine-2-naphthylamide), could be reliably used ineveryday practice to identify the presence or prevalenceof “red complex” microorganisms [7]. Porphyromonas gin-givalis, Tannerella forsythia, and Treponema denticola arecultivable BANA-positive plaque species which possess anenzyme capable of hydrolyzing the synthetic peptide of theBANA reagent [8]. BANA has shown to correlate well withthe depth of the periodontal pocket and clinical indiceswhich are used to diagnose periodontal disease [9]. Thesynergistic mechanisms between the “red complex” speciesresult in an immune-inflammatory response [10].

In clinical periodontal practice, the BANA test has provedto be efficient in the diagnosis of periodontal disease and alsoin the bacteriologic monitoring of periodontally involvedpatients during the different phases of periodontal treatment[11]. The test provides equal accuracy in detecting these “red


complex” species comparable to that obtained with DNAprobes and immunology [7].

Microbial complexes of HIV-associated periodontal dis-eases are reported to be similar to those in HIV-negativeindividuals with periodontal diseases [12, 13]. No overtpathogen has been implicated for periodontal disease inHIV-seropositive individuals [14], although certain specificmicrobial complexes not commonly found in HIV-negativeindividuals may be responsible for chronic periodontaldiseases in HIV-positive patients [15]. Studies conducted byBrady et al. [16], based on selective media, culture methods,and microscopy, did not demonstrate any characteristicdifference in the microflora of HIV-infected and non-infected individuals with or without periodontal disease. Theinteraction of the “red complex” members with the epithelialcells in the subgingival environment leads to the destructionof host tissues.

The aim of the present study was to establish whetherimmunosuppression determined by the level of the CD4+T cell counts in HIV-positive patients, or their oral hygienepractices play a greater role in the colonization of BANA-positive microorganisms such as Porphyromonas gingivalis,Treponema denticola, and Tannerella forsythia, implicated inthe aetiology of HIV-associated periodontitis.


2.1. Study Population. A cohort descriptive study wasconducted involving a randomized selection of 120 maleand female HIV-positive patients attending the infectiousdiseases clinic at Tygerberg Medical Hospital, South Africa.HIV-positive patients with various CD4+ T cell counts irre-spective of their antiretroviral therapy were included in thestudy. Exclusion criteria included pregnancy, tuberculosis,diabetes, cardiovascular disease, autoimmune diseases, andalso patients who were on antibiotic treatment and thosewho had undergone dental treatment 3 months prior tothe study. Ethical committee clearance from the Universityof the Western Cape and written informed consent wereobtained from all the participants prior to the start of thestudy. Information regarding demographic features, generalhealth, HIV infection history, and most recent CD4+ T cellcounts (usually within the last 3 months) were obtained fromthe patients’ medical records. Viral loads were not recordedsince the information was not always available. Factorspredisposing to periodontal disease and oral hygiene habitswere obtained from questionnaires and included questionspertaining to frequency of visits to the dentist (never, once ayear, twice a year), frequency of tooth brushing (how manytimes a day), and the use of interdental aids (floss, interdentalbrushes, toothpicks).

2.2. Periodontal Procedures. Periodontal clinical measure-ments such as plaque index (PI), gingival index (GI), probingdepth (PD), and clinical attachment level (CAL) were takenat the mesial aspect of six Ramfjord teeth. These mea-surements were performed by a single calibrated examinerusing Williams periodontal probe. Intraexaminer Kappa

index and intraclass coefficient correlation of agreement forperiodontal pocket depth measurements ranged from 0.765to 0.985. No oral hygiene instructions were given prior to theclinical examination.

2.3. Enzyme Assay. The BANA test was done as describedby Loesche et al. [8]. The BANA reagent strip consisted ofa plastic strip to which upper and lower reagent matriceswere attached. After the removal of supragingival plaque,the subgingival plaque was collected from the site with themaximum pocket depth using a Gracey curette and dispersedon the lower reagent matrix of the BANA test strip. Aftermoistening the upper matrix, the strip was folded to meet theupper and lower reagents and was incubated for 15 minutesat 55◦C. If BANA-positive species were present when the teststrip was opened, a permanent blue colour was obtained onthe upper matrix.

2.4. Statistical Analysis. Data was analyzed using statisticalprogrammes such as SAS (SAS Institute Inc., Cary, NC,USA). The Wilcoxon rank sum test was used to determinethe association between CD4+ T cell counts and each of theclinical periodontal parameters with BANA. Pearson’s Chi-square test was applied to determine the association between“red complex” and periodontal status. Cochran-Mantel-Haenszel test (CMH test) was used to test the relationshipbetween a pair of variables adjusting for the effect of othervariables. Logistic regression model was used to determinethe association between BANA and oral hygiene adjustingfor ungrouped CD4+ T cell counts. A significance level of<0.01 rather than the usual <0.05 level indicated a highlysignificant result, while results with a P value between 0.01and 0.05 were referred to as marginally significant.

3. Results

Of the 120 participants included in the study, 55 (45.83%)were males and 65 (54.17%) were females with a mean ageof 33.25 years (range: 20–55). A majority of 83 (69.17%)HIV-positive patients were reported as BANA-positive and37 (30.83%) were BANA-negative (Table 1).

The mean CD4+ T cell count levels were 299.07cells/mm3 and 280.79 cells/mm3 for BANA-positive andBANA-negative patients, respectively. There was no sig-nificant association found between BANA with CD4+ Tcell counts (Wilcoxon test, P = 0.7075); however, highlysignificant associations were observed between BANA withprobing depth and clinical attachment level, while marginalsignificances were observed between BANA with plaque andgingival indices (Table 1).

By comparing BANA results with CD4+ T cell countsgrouped into <200, 200–500 and >500 cells/mm3, 30.12% ofthe <200 cells/mm3 group, 59.04% of the 200–500 cells/mm3

group and 10.84% of the >500 cells/mm3 group were BANA-positive. No significant associations were found betweenBANA with grouped CD4+ T cell counts (Table 2). Theresults revealed that the presence of BANA-positive microor-ganisms were not related to CD4+ T cell counts as a whole


Table 1: CD4+ T cell counts and clinical indices relative to frequency of BANA.

VariablesBANA

P valuePositive (N = 83) Negative (N = 37)

Mean (SD) Median Minimum–Maximum Mean (SD) Median Minimum–Maximum

CD4+ T cell countsa 299.07 (157.16) 300 61–859 280.79 (137.6) 256 36–562 0.7075

Plaque indexb 2.64 (0.46) 2.9 1.3–3 2.35 (0.64) 2.5 0.8–3 0.0248

Gingival indexc 2.83 (0.35) 3 1.5–3 2.59 (0.59) 2.9 0.5–3 0.0348

Probing depthd 5.06 (1.02) 5 2.9–6.8 4.12 (0.76) 4 2.9–5.8 <0.0001

Clinical attachment levele 5.6 (1.1) 5.9 3–7.3 4.59 (0.83) 4.4 3–6.1 <0.0001aNo statistical significance observed between CD4+ T cell counts and BANA test (Wilcoxon rank sum test).

b,cMarginally statistical significance observed between BANA and plaque and gingival indices (Wilcoxon rank sum test).d,eHighly statistical significance observed between BANA and probing depth and clinical attachment level (Wilcoxon rank sum test).

Table 2: Grouped CD4+ T cell counts relative to BANA.

BANAGrouped CD4+ T cell counts (P = 0.9989)

Total<200 200–500 >500

Negative 11 (29.73%) 22 (59.46%) 4 (10.81%) 37 (30.83%)

Positive 25 (30.12%) 49 (59.04%) 9 (10.84%) 83 (69.17%)∗

No statistical significance observed between BANA and grouped CD4+ T cell counts (Pearson’s Chi-square test).

or CD4+ T cell counts grouped into <200, 200–500 or>500 cells/mm3.

Of 83 (69.17%) BANA-positive patients, 62 (74.70%)brushed once a day and about 21 (25.30%) brushed twice aday while among the 37 (30.83%) BANA-negative patients,22 (59.46%) admitted brushing once a day and only 15(40.54%) admitted brushing twice a day. There were no sig-nificant associations found between BANA and the frequencyof brushing (Table 3). A marginally significant associationwas apparent when comparing positive BANA results withthe use of interdental aids (Chi-square test, P = 0.0168). Amajority of 70 (84.34%) of BANA-positive patients neverused interdental aids while 13 (15.66%) of them reportedusing interdental aids. Among the BANA-negative patients,24 (64.86%) never used interdental aids and 13 (35.14%)used interdental aids.

HIV-positive patients who brushed their teeth once aday and those who did not use any interdental aids to cleantheir teeth showed a greater prevalence of “red complex” asdetected by the BANA test. Table 4 shows the associationbetween BANA and oral hygiene practices such as brushingand the use of interdental aids adjusted for grouped as wellas ungrouped CD4+ T cell counts. The findings showedsignificant associations between BANA and interdental aidsbut not with brushing frequency. This explains the fact thatthe use of additional oral cleaning methods was essentialand merely brushing was not enough to obtain a disease-freemouth.

Moreover, CMH tests failed to demonstrate signifi-cant relationships between BANA and grouped as well asungrouped CD4+ T cell counts even when adjusted for oralhygiene practices (Table 5).

From the results it can be suggested that oral hygienemaintenance plays a greater role in the aetiology of peri-odontal disease in HIV-positive patients than the level of

immunosuppression. However, one must take into accountthe fact that this study included HIV patients whether or notthey were on therapy. This may have influenced the resultsobtained.

4. Discussion

Studies have suggested an increase in the severity andprevalence of periodontal diseases in HIV-positive pop-ulations. The primary aetiology of periodontal diseasesis the formation and adhesion of dental plaque on thesurfaces of the tooth. Dental plaque involves a communityof microorganisms found as a biofilm on the tooth surfaceand develops under different conditions and environments.In the periodontal disease progression, “red complex” isconsidered as the most significant cluster among the bacterialclusters [17], as the “red complex” pathogens increase innumbers and prevalence with increasing clinical parametersof the disease [18].

Since periodontal disease is a polymicrobial infection, acoaggregation of subgingival microorganisms exists in HIV-positive patients. Among them Porphyromonas gingivalis,Treponema denticola, and Tannerella forsythia can be con-sidered as true pathogens in HIV progression. However,other studies reported a lower prevalence of these puta-tive bacteria in HIV-positive compared with HIV-negativeindividuals [19, 20]. The clinical signs of HIV-associatedperiodontitis are always related to gingival inflammation,profound gingival bleeding, increased probing depth, andloss of clinical attachment and hence periodontal disease maybe determined as one of the first clinical representations of anundiagnosed HIV infection [21].

In the present investigation, BANA test showed signif-icant associations with all the periodontal indices. Clinical


Table 3: BANA relative to oral hygiene practices.

BANABrushing (P = 0.0925) Interdental aids (P = 0.0168)

Once a day Twice a day No Yes

Negative (n = 37) 22 (59.46%) 15 (40.54%) 24 (64.86%) 13 (35.14%)

Positive (n = 83) 62 (74.70%) 21 (25.30%) 70 (84.34%) 13 (15.66%)∗

No statistical significance observed between BANA positivity and brushing (Pearson’s Chi-square test).∗Statistical significance observed between BANA positivity and interdental aids (Pearson’s Chi-square test).

Table 4: Association between BANA with oral hygiene adjusted for grouped and ungrouped CD4+ T cell counts.

Ungrouped CD4+ T cell counts

Independent variables Grouped CD4+ T cell counts (P values) (CMH test) (Logistic regression model)

Odds ratio 95% CI P values

BANABrushinga 0.0897 0.50 0.22–1.14 0.1154

Interdental aidsb 0.0174 0.35 0.14–0.85 0.0244aNo statistical significance observed between BANA and brushing with grouped and ungrouped CD4+ T cell counts.

bStatistical significance observed between BANA and interdental aids with grouped and ungrouped CD4+ T cell counts.

Table 5: Association between BANA with grouped and ungroupedCD4+ T cell counts adjusted for oral hygiene.

Independent variables P values

BANAGrouped CD4+ T cell counts 0.9423

Ungrouped CD4+ T cell counts 0.9943∗

No statistical significance observed between BANA and grouped andungrouped CD4+ T cell counts (CMH test).

attachment level and probing depth were found highly sig-nificant (P ≤ 0.0001) and plaque (P = 0.0248), and gingival(P = 0.038) indices were found to be marginally significantwith BANA. This investigation’s results clearly indicate therelationship between BANA-positive species and their rolein periodontal disease. A study by Puscasu et al. [22], on61 adult patients with gingivitis or periodontitis, observeda statistical correlation with the severity of periodontaldestruction and BANA but they failed to determine a statis-tical correlation between plaque index and BANA. However,their study used both supragingival and subgingival plaquedeposits for BANA test, where supragingival plaque harboursless anaerobic BANA-positive species which could thusprevent their detection. On the contrary, in the present study,only subgingival plaque samples were taken to detect the“red complex,” where they colonize abundantly. This couldaccount for their positive response to the BANA test.

It has been suggested that the level of immunosuppres-sion or the decrease in the CD4+ T cell count determines theprogression of periodontal disease. Highly severe immuno-suppression in HIV patients may exacerbate preexistingperiodontitis, considering HIV infection as a modifier ofperiodontal disease [23]. However, in the present investiga-tion, BANA test was not found to be correlated with absoluteCD4+ T cell counts nor with grouped CD4+ T cell countseven when adjusted for oral hygiene. The interesting findingwas that an increase in the “red complex” microorganisms

in HIV infections was not clearly demonstrated to be dueto the decrease in CD4+ T cell counts, per se, but it couldbe due to the lack of oral hygiene maintenance that leadsto the colonization of “red complex” pathogens. Most ofthe HIV patients examined brushed only once a day (70%)and never used interdental aids (78.33%) thus indicating lesseffort on the patient’s part to maintain better oral hygiene,ultimately leading to gingival inflammation and periodontaldiseases. Lack of proper oral hygiene may lead to theaccumulation of periodontal microorganisms in the mouthresulting in a positive response to the BANA test. BANA wasnot significantly related to the frequency of brushing (Chi-square test, P = 0.0925) but there was a marginally significantrelationship between BANA and the use of interdental aids(Chi-square test, P = 0.0168). Inadequate brushing leadsto the accumulation of both supragingival and subgingivalplaque making oral hygiene difficult. Moreover, it wasevident that association of periodontal clinical parameterssuch as plaque and gingival indices, probing depth and lossof clinical attachment with BANA positivity in HIV patientssuggests improper maintenance of oral hygiene.

An alternate analysis using the logistic regression modelgave virtually identical results. Not adjusting for CD4+T cell counts, the use of interdental aids was significantwith P = 0.0168 (Chi-square test). Adjusting for CD4+ Tcell counts in a logistic regression model, the P value forinterdental aids was still significant with P = 0.0244. Theodds ratio was 0.35 for interdental aids, and the 95%confidence interval was 0.14 to 0.85.

Even when only a minority of the study population usedinterdental aids as an additional cleaning device, it marked amajor difference in the oral hygiene maintenance comparedto the frequency of brushing. As one would expect, morefrequent brushing and interdental use tends to provide betterplaque control and improved gingival health. Clinicians mustmotivate the patients and highlight the benefits of additionalcleaning devices to clean their teeth. Overall, the oral hygiene


status among the HIV-positive patients who participated inthe study was poor which may explain the presence of BANA-positive pathogens.

The immunosuppression of HIV patients influencesthe prevalence of periodontal diseases which in turn isexacerbated by their poor oral hygiene. Since the currentinvestigation was strictly restricted for the detection of “redcomplex” subgingival pathogens only, knowledge of anyrelation with one or a combination of other periodonto-pathic bacteria causing destruction of the periodontiumcannot be ignored. This could be considered as a limitationof the study since other bacterial species that might havebeen present subgingivally were not detected. The presentstudy included patients who had not received any formof dental treatment 3 months prior to the study, therebyeliminating the bias of those visiting dental clinics fororal complaints. Some of the HIV patients in the studywere on antiretroviral therapy, which could explain thereason of not finding any association between BANA-positivepathogens and their CD4+ T cell count levels. Earlierintervention to antiretroviral therapies and the widespreaduse of antiretroviral therapy could limit the exposure toimmunosuppression in HIV patients. The regular use ofantimicrobials in the form of antiretroviral therapy mayreduce the virulence of the subgingival microbiota in HIV-positive patients, even under severe immunosuppression.

This study employed the Ramfjord teeth instead of a full-mouth assessment for the periodontal status of the studygroup. While some researchers have demonstrated the epi-demiological validity of the Ramfjord teeth in representingthe periodontal status of the whole mouth [24–29], othershave reported the introduction of a bias (either positively ornegatively), thus making it unsuitable for determining peri-odontal disease severity or prevalence [30–32]. We elected touse the Ramfjord teeth since it was found to reduce time,cost, patient, and examiner fatigue, while also providinga practical alternative to the 168 measurements for eachclinical parameter required to characterise the prevalenceand severity of periodontal disease in a single whole mouth[28]. However, we are mindful of the limitations of partialmouth measurements. Measurement of sites on the buccalside of the tooth have been reported to show better reliabilitythan measurement on the lingual side because of bettervisibility to the examiner. A better representation maytherefore be achieved by measuring several sites on a toothinstead of just one aspect as we have done. Furthermore,partial recording techniques were shown to reduce theseverity of periodontal destruction scores to almost half ofthat demonstrated in full mouth examinations, resulting inan underestimation of both the extent and prevalence ofperiodontal disease in some studies [28, 33].

5. Conclusion

The presence of “red complex” pathogens (as demonstratedby the BANA test) was not associated with the level ofimmunosuppression nor CD4+ Tcell counts in HIV-positivepatients. Although immunosuppression plays a major role in

the periodontal inflammation, improper oral hygiene mayaggravate its severity and progression. The “red complex”subgingival pathogens were closely related to the clinicalparameters of inflammation and periodontal destructionand appeared meaningful in periodontal diagnosis. Theresults reported in this investigation would imply that anassociation existed between the presence of BANA-positiveplaques and the use of interdental aids, and virtually noevidence of an association with CD4+ T cell counts. Thehigh prevalence of periodontal manifestations underlines theneed for routine dental visits and treatment with a propermaintenance of oral hygiene. A greater emphasis on oralhealth and prevention of periodontal disease is necessaryespecially among HIV-infected individuals.

Acknowledgments

The authors thank all their subjects who volunteered theirtime and staff at the infectious disease clinic as well asProfessor Richard Madsen for the statistical assistance. Thispaper is based upon work supported financially by theNational Research Foundation (NRF). Any opinion, findingsand conclusions, or recommendations expressed in thispaper are those of the authors and therefore the NRF doesnot accept any liability in regards thereto.

References

[1] M. Alves, R. Mulligan, D. Passaro et al., “Longitudinal evalu-ation of loss of attachment in HIV-infected women comparedto HIV-uninfected women,” Journal of Periodontology, vol. 77,no. 5, pp. 773–779, 2006.

[2] J. M. Coffin, S. H. Hughes, and H. E. Varmus, “Pathogenesisof HIV and SIV,” in Retroviruses, J. M. Coffin, S. H. Hughes,and H. E. Varmus, Eds., pp. 189–248, Cold Spring HarborLaboratory Press, 1997.

[3] E. B. Fowler, L. G. Breault, and M. F. Cuenin, “Periodontaldisease and its association with systemic disease,” MilitaryMedicine, vol. 166, no. 1, pp. 85–89, 2001.

[4] V. John and S. J. Kim, “Periodontal disease and systemicdisease. Clinical information for the practicing dentist,”Journal of the Indiana Dental Association, vol. 81, no. 2, pp.15–18, 2002.

[5] B. Bakdash, “Oral hygiene and compliance as risk factors inperiodontitis,” Journal of Periodontology, vol. 65, no. 5, pp.539–544, 1994.

[6] L. de Souza Goncalves, S. M. Ferreira, A. Silva Jr. et al.,“Association of T CD4 lymphocyte levels and subgingivalmicrobiota of chronic periodontitis in HIV-infected Braziliansunder HAART,” Oral Surgery, Oral Medicine, Oral Pathology,Oral Radiology, and Endodontics, vol. 97, no. 2, pp. 196–203,2004.

[7] W. J. Loesche, D. E. Lopatin, J. Giordano, G. Alcoforado,and P. Hujoel, “Comparison of the benzoyl-DL-arginine-naphthylamide (BANA) test, DNA probes, and immunologi-cal reagents for ability to detect anaerobic periodontal infec-tions due to Porphyromonas gingivalis, Treponema denticola,and Bacteroides forsythus,” Journal of Clinical Microbiology, vol.30, no. 2, pp. 427–433, 1992.

[8] W. J. Loesche, J. Giordano, and P. P. Hujoel, “The utilityof the BANA test for monitoring anaerobic infections due


to spirochetes (Treponema denticola) in periodontal disease,”Journal of Dental Research, vol. 69, no. 10, pp. 1696–1702,1990.

[9] M. F. M. Grisi, T. A. Correa Filho, C. L. Fanganiello,W. Martins Junior, C. R. Silva-Neto, and S. L. Salvador,“Relationship between the presence or absence of gingivalbleeding and the enzymatic BANA test,” Brazilian DentalJournal, vol. 12, no. 1, pp. 23–26, 2001.

[10] L. Kesavalu, S. Sathishkumar, V. Bakthavatchalu et al., “Ratmodel of polymicrobial infection, immunity, and alveolarbone resorption in periodontal disease,” Infection and Immu-nity, vol. 75, no. 4, pp. 1704–1712, 2007.

[11] M. F. M. Grisi, A. B. Novaes, I. Y. Ito, and S. L. Salvador,“Relationship between clinical probing depth and reactivityto the BANA test of samples of subgingival microbiota frompatients with periodontitis,” Brazilian Dental Journal, vol. 9,no. 2, pp. 77–84, 1998.

[12] C. S. P. Tsang and L. P. Samaranayake, “Predominant cultivablesubgingival microbiota of healthy and HIV-infected ethnicChinese,” Acta Pathologica, Microbiologica et ImmunologicaScandinavica, vol. 109, no. 2, pp. 117–126, 2001.

[13] T. Alpagot, N. Duzgunes, L. F. Wolff, and A. Lee, “Risk factorsfor periodontitis in HIV+ patients,” Journal of PeriodontalResearch, vol. 39, no. 3, pp. 149–157, 2004.

[14] P. A. Marray, M. Grassi, and J. R. Winkler, “The microbiologyof HIV-associated periodontal lesions,” Journal of ClinicalPeriodontology, vol. 16, no. 10, pp. 636–642, 1989.

[15] M. Patel, M. Coogan, and J. S. Galpin, “Periodontal pathogensin subgingival plaque of HIV-positive subjects with chronicperiodontitis,” Oral Microbiology and Immunology, vol. 18, no.3, pp. 199–201, 2003.

[16] L. J. Brady, C. Walker, G. E. Oxford, C. Stewart, I. Magnusson,and W. McArthur, “Oral diseases, mycology and periodontalmicrobiology of HIV-1-infected women,” Oral Microbiologyand Immunology, vol. 11, no. 6, pp. 371–380, 1996.

[17] S. S. Socransky, A. D. Haffajee, M. A. Cugini, C. Smith, and R.L. Kent, “Microbial complexes in subgingival plaque,” Journalof Clinical Periodontology, vol. 25, no. 2, pp. 134–144, 1998.

[18] S. C. Holt and J. L. Ebersole, “Porphyromonas gingivalis,Treponema denticola, and Tannerella forsythia: the “red com-plex”, a prototype polybacterial pathogenic consortium inperiodontitis,” Periodontology 2000, vol. 38, no. 1, pp. 72–122,2005.

[19] L. de Souza Goncalves, S. M. S. Ferreira, C. O. Souza, R. Souto,and A. P. V. Colombo, “Clinical and microbiological profiles ofhuman immunodeficiency virus (HIV)-seropositive Braziliansundergoing highly active antiretroviral therapy and HIV-seronegative Brazilians with chronic periodontitis,” Journal ofPeriodontology, vol. 78, no. 1, pp. 87–96, 2007.

[20] L. de Souza Goncalves, A. P. V. Colombo, and R. Souto,“Detection of Helicobacter pylori, Enterococcus faecalis, andPseudomonas aeruginosa in the subgingival biofilm of HIV-infected subjects undergoing HAART with chronic periodon-titis,” European Journal of Clinical Microbiology and InfectiousDiseases, vol. 28, no. 11, pp. 1335–1342, 2009.

[21] H. C. Tenenbaum, A. E. Simor, and D. Mock, “Periodontitisas an early presentation of HIV infection,” Canadian MedicalAssociation Journal, vol. 144, no. 10, pp. 1265–1269, 1991.

[22] C. G. Puscasu, A. S. Dumitriu, and H. T. Dumitriu, “Thesignificance of BANA test in diagnosis of certain formsof periodontal disease,” Journal of Oral Health and DentalManagement, vol. 5, no. 3, p. 31, 2006.

[23] D. F. Kinane, “Periodontitis modified by systemic factors,”Annals of Periodontology, vol. 4, no. 1, pp. 54–64, 1999.

[24] R. J. Hunt, “The efficiency of half-mouth examinations inestimating the prevalence of periodontal disease,” Journal ofDental Research, vol. 66, no. 5, pp. 1044–1048, 1987.

[25] P. Goldberg, L. Matsson, and H. Anderson, “Partial recordingof gingivitis and dental plaque in children of different ages andin young adults,” Community Dentistry and Oral Epidemiology,vol. 13, no. 1, pp. 44–46, 1985.

[26] J. D. Owens, S. A. Dowsett, G. J. Eckert, D. T. Zero, and M.J. Kowolik, “Partial-mouth assessment of periodontal diseasein an adult population of the United States,” Journal ofPeriodontology, vol. 74, no. 8, pp. 1206–1213, 2003.

[27] E. G. S. Mumghamba, W. Pitiphat, M. I. N. Matee, E. Simon,and A. T. Merchant, “The usefulness of using Ramfjordteeth in predicting periodontal status of a Tanzanian adultpopulation,” Journal of Clinical Periodontology, vol. 31, no. 1,pp. 16–18, 2004.

[28] M. V. Vettore, G. de Almeida Lamarca, A. T. T. Leao, M. doCarmo Leal, and A. Sheiham, “Partial recording protocols forperiodontal disease assessment in epidemiological surveys,”Cadernos de Saude Publica, vol. 23, no. 1, pp. 33–42, 2007.

[29] A. Najah, S. Seham, and F. Raghad, “The usefulness ofRamfjord teeth to represent the full-mouth pocket depth inepidemiological study,” Mustansiria Dental Journal, vol. 7, no.2, pp. 272–275, 2010.

[30] J. L. Fleiss, M. H. Park, N. W. Chilton, J. E. Alman, R. S.Feldman, and H. H. Chauncey, “Representativeness of the“Ramfjord teeth” for epidemiologic studies of gingivitis andperiodontitis,” Community Dentistry and Oral Epidemiology,vol. 15, no. 4, pp. 221–224, 1987.

[31] A. Kingman and J. M. Albandar, “Methodological aspects ofepidemiological studies of periodontal diseases,” Periodontol-ogy 2000, vol. 29, no. 1, pp. 11–30, 2002.

[32] S. A. Dowsett, G. J. Eckert, and M. J. Kowolik, “Theapplicability of half-mouth examination to periodontal dis-ease assessment in untreated adult populations,” Journal ofPeriodontology, vol. 73, no. 9, pp. 975–981, 2002.

[33] A. Kingman, E. Morrison, H. Loe, and J. Smith, “Systematicerrors in estimating prevalence and severity of periodontaldisease,” Journal of Periodontology, vol. 59, no. 11, pp. 707–713,1988.


Research Article

Prevalence of Periodontal Bone Loss in Brazilian Adolescentsthrough Interproximal Radiography

Benedicto Egbert Correa de Toledo, Eliane Marcon Barroso,Alex Tadeu Martins, and Elizangela Partata Zuza

Department of Masters in Dental Sciences, the University Center of Educational Foundation of Barretos (UNIFEB),Avenida Professor Roberto Frade Monte, 389, 14783-226 Barretos, SP, Brazil

Correspondence should be addressed to Elizangela Partata Zuza, [email protected]

Received 24 July 2012; Accepted 8 September 2012


Copyright © 2012 Benedicto Egbert Correa de Toledo et al. This is an open access article distributed under the Creative CommonsAttribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work isproperly cited.

Purpose. The aim of this study was to verify the prevalence of alveolar bone loss in Brazilian adolescents through the interproximalX-rays analysis. Methods. Bilateral and standardized interproximal (bitewing) X-rays were performed in 15-year-old adolescents(n = 326), and the processing of films and measurements of alveolar bone levels were accomplished by a single examiner. A distancebetween the cementoenamel junction (CEJ) and the alveolar bone crest more than 2 mm was considered as periodontal bone loss.Results. The results showed percentage of bone loss of 10.4% with predominance of horizontal defects (8.9%) over the vertical types(1.5%). It was verified higher individual distribution of one lesion (67.6%) than two (26.5%) or three lesions (5.6%), and higheroccurrence was detected in men (14.95) than in women (8.21). Conclusion. It can be concluded that the interproximal radiographywas an efficient method for the detection of alveolar bone loss, revealing low prevalence in adolescents and predominance ofhorizontal bone defects.

1. Introduction

Periodontal diseases are among the most frequent diseasesthat may affect children and adolescents [1]. Epidemiologicalstudies have been supporting that the gingivitis is practicallyuniversal in children and adolescents, but destructive formsof periodontal disease may also occur although this is lowerin young people than in adults [2, 3]. Thus, the AmericanAcademy of Periodontology withdrew the age factor of theclassification system of periodontal disease [2]. Althoughprevious data revealed a more surface problem in childrenand adolescents, some studies pointed out the presence ofmore advanced forms of periodontal disease, since severegingivitis [1, 4–6] until chronic periodontitis [7, 8] withperiodontal pocket, clinical attachment loss [9–11], andalveolar bone loss [12–15].

Some studies also showed that periodontal attachmentloss and alveolar bone loss are uncommon in children, but itsprevalence increases in 12- to 17-year-old adolescents whencompared with children from 5 to 11 years [13, 16–20].

A multinational study in 8.730 subjects from 15 to 17 yearsold was accomplished by Hansen et al. [19], which repre-sented a global sample (18 centers of 16 countries), and theauthors stated that the early periodontal destruction in 15-year-old young people seems to be a common phenomenonin the whole world. Although these studies have found thatthe alveolar bone loss was a common finding in adolescents,the diagnosis of bone loss in pediatric patients has beenneglected [15]. Sjodin et al. [21] demonstrated that 40%of patients diagnosed in adulthood with periodontitis hadalready presented alveolar bone loss in deciduous dentition.

Several studies showed the prevalence of periodontaldisease in young and adolescent patients using X-rays as amethod of diagnosis [17, 18, 22–27]. Although there arelimitations in the X-ray conventional technique [28], itrepresents an available and acceptable method when thetechnical standardization can be obtained because it is veryuseful [29]. The studies that use X-rays to check the alveolarbone loss aim to obtain information about periodontal dis-ease to diagnose if the images indicate a disease condition or


a normalcy variation. In this way, some studies reported var-ious findings considering the subjectivity of the parametersand analyses, focusing on some of the problems about thismethodology [13]. However, the X-rays have shown a strongcorrelation between the clinical and radiographic findings[30] and a good correlation with the clinical attachmentloss [31], especially using the interproximal (bitewing) X-rays, which have shown to be a valuable method for thedetection of incipient bone loss [18]. Thus, the interproximal(bitewing) radiography has been the most used method tocheck bone loss in young people [17–19, 22, 25, 32].

The radiographic examination can be a useful methodfor the determination of the early reabsorption of alveolarbone crest, considering the extent of the gingival problemin children and young people. The aim of this study was toverify the prevalence of alveolar bone loss in adolescents bymeans of the interproximal X-rays.


2.1. Sample Selection. This research was approved by theResearch Ethics Committee of University Center of Edu-cational Foundation of Barretos—UNIFEB. A survey wascarried out at Elementary Public Schools in Barretos city(Sao Paulo, Brazil) to check the availability of 15-year-oldstudents for voluntary participation in the research. Someschools were randomly chosen according to their locationthrough the method of stratified sampling to obtain a rep-resentative sample of the city. The adolescents were taken tothe dental clinics of the University Center of the EducationalFoundation of Barretos (UNIFEB) by bus, kindly loanedby the Road Company Rio Grande (Barretos, Sao Paulo,Brazil) for safety transportation and without any cost forthe students and institution (UNIFEB). Parents or guardiansshould sign a written consent form for authorization of theadolescents for the participation in the study. Thus, initially380 patients from the 400 obtained were examined. Thecriterion for exclusion of the subject in the sample was basedon the absence of at least one proximal surface of a first uppermolar and a first lower molar for the readings [32].

2.2. Radiographic Examination. The radiographic examina-tion was obtained through two X-rays shots per patient(one on each side), using the interproximal technique inthe region of premolars and molars [16, 19, 25]. An X-raymachine was used at 70 KV 8 mA (Dabi Atlante, RibeiraoPreto, SP, Brazil), and the X-ray shots were standardized bythe use of positioners to the interproximal technique (bitew-ings, Indusbello, Londrina, Brazil). The shots were carriedout within all the standards of biosecurity, using lead apronand rapid exposure Ektaspeed films (Kodak, Sao Paulo,Brazil), packed in PVC plastic film. The films were evaluatedby an investigator (EMB) that evaluated the radiographs inan appropriate negatoscope in a dark room, with the screencovered by paper and dark suitable window for accommoda-tion of the film for the examinations.

The measurements were carried out by a single examiner(EMB) on different days with two days apart, evaluating

Table 1: Distribution of the subjects in the study.

GenderExaminedN (%)

ExcludedN (%)

IncludedN (%)

Man 129 (33.95) 22 (5.8) 107 (28.2)

Woman 251 (60.05) 32 (8.4) 219 (57.6)∗

Total 380 (100) 54 (14.2) 326 (85.8)∗P < 0.0001 (binomial test for comparison of proportions; P < 0.05 indicatesstatistically significant difference).

a maximum of 40 X-rays per day. The measurements weremade from the distance from the cementoenamel junctionuntil the alveolar bone crest, with the support of a magni-fying glass and precision caliper (Mitutoyo, Tokyo, Japan).A distance greater than 2 mm between the cementoenameljunction and the alveolar bone crest was considered as boneloss [17–19, 25]. The bone crest was defined as the mostcoronal level in which the periodontal membrane retainsits normal thickness [5]. Subsamples of three radiographswere read again, with one-week interval for the proof anddetermination of reproducibility and reliability of the data.The intraexaminer diagnostic confidence was evaluated byKappa statistics [33], reaching a concordance level of 88%.

The measurements were performed from the distalsurface of the canine teeth up to the mesial surface of thesecond molars fully erupted. Surfaces with superposition ofdental elements, presence of dental braces or cases with nopossibility of determining the cementoenamel junction wereexcluded, as well as teeth in eruption process, and films withprocessing problems or that presented errors in the X-raysshot. Then, 54 patients were excluded from the 380 patientsinitially examined, and the study was followed with 326subjects.

2.3. Statistical Analysis. The statistical analysis was per-formed in some data by the binomial test for proportions.The Cochran test (Q) was used for comparison between thenumbers of lesions. P < 0.05 was considered as statisticallysignificant difference. The program used was BioEstat 5.0(Belem, PA, Brazil).

3. Results

Table 1 shows the distribution of young people radiographedin relation to gender. The results showed that the proportionof women was higher than men in the sample (P < 0.0001).Table 2 shows the distribution of types of alveolar boneloss (horizontal and vertical) according to the gender. Thedata showed higher proportion of bone loss in men thanin women (P = 0.03). There was greater predominance ofhorizontal bone loss than vertical lesions (P < 0.0001). Thedistribution of the number of horizontal and vertical lesions(1, 2 or 3) can be verified in Table 3. It is verified that menshowed higher proportion in the total number of lesions thanwomen (P = 0.03), and that the majority showed one or twolesions.


Table 2: Distribution of bone loss types according to the gender.

GenderBone loss

HorizontalN (%)

VerticalN (%)

Horizontal + VerticalBone loss N (%)

Man (n = 107) 14 (13.1) 2 (1.87) 16 (14.95)∗

Woman (n = 219) 15 (6.8) 3 (1.4) 18 (8.2)

Total (n = 326) 29 (8.9)# 5 (1.5) 34 (10.4)∗P = 0.03 (comparison between genders); #P < 0.0001 (comparisonbetween horizontal versus vertical bone loss). Binomial test for comparisonof proportions (P < 0.05 indicates statistically significant difference).

Table 3: Distribution of the lesions’ number according to the gen-der.

GenderLesion

1 lesionN (%)

2 lesionsN (%)

3 lesionsN (%)

TotalN (%)

Man (n = 107) 11 (68.75) 4 (25) 1 (6.25) 16 (14.95)∗

Woman (n = 219) 12 (66.7) 5 (27.8) 1 (5.55) 18 (8.2)

Total (n = 326) 23 (67.6)# 9 (26.5)# 2 (5.6) 34 (10.4)∗P = 0.03 (binomial test for comparison between genders; P < 0.05 indicatesstatistically significant difference). #P < 0.01 (Cochran test, comparisonbetween 1 lesion versus 2 lesions; 1 lesion versus 3 lesions; 2 lesions versus 3lesions; P < 0.05 indicates statistically significant difference).

4. Discussion

Some authors reported that the distance considered asnormal in young individuals would be from 1 to 2 mmbetween the cementoenamel junction and the bone crest[34, 35]. It has also been recommended that this distancecould be greater than 3 mm [36]; however, in the presentstudy, it was considered a distance greater than 2 mmbetween the cementoenamel junction and the alveolar bonecrest, following the guidelines of some studies [17–19, 25].Moreover, there was uniformity in the age of sample to avoidits influences, because the distance from cementoenameljunction to alveolar bone crest seems to increase with age[37].

Some authors also emphasize the possibility of an under-estimation in the verification of the incipient periodontitisby X-rays methods, whether by deflection of angulation,difficulties of identification of the cementoenamel junction,evaluation only in posterior teeth, and even by other aspectsof the radiographic technique [13]. In the present study,all the procedures were standardized, such as X-rays shots,revelation, and readings of the sample in the attempt to avoiderrors.

Our results are in agreement with the observations byHansen et al. [19] that revealed that the alveolar bone lossshowed an average of 10.2%, nearly to the percentageobtained in our findings (10.4%). From the three Braziliancenters that participated in the multinational study ofHansen et al. [19], two of them named as Brazil 1 and Brazil 3(Araraquara and Porto Alegre) demonstrated average valuesof 8.95% and 8.70%, respectively. These values were also

close to 10.4% verified in the present study, as well as thepercentage of 10.2% obtained in the global sample, and in theresult obtained by Hansen et al. [17] of 11.3%. Conversely,some findings showed higher percentage of alveolar boneloss in younger individuals [18, 19, 22, 25]. Gjermo et al.[18] examined a sample of 304 schoolchildren with lowsocioeconomic conditions in Belo Horizonte (Brazil), from13 to 16 years old, and found 27.6% of alveolar bone loss.In the same way, similar results had already been observed inthe global study by Hansen et al. [19] in the sample of Brazil2, and a small sample was discarded due the high rate of boneloss (26.6%). The highest values presented in populations oflow socioeconomic conditions had been observed previouslyby Albandar et al. [22] and Aass et al. [25]. Thus, the samplewas composed by several socioeconomic segments in thepresent study.

The horizontal bone defects were more prevalent thanthe vertical types. Our results showed a percentage of 8.9%and 1.5% for horizontal and vertical defects, respectively.These findings are similar to those found by Hansen et al.[19], which demonstrated a prevalence of 8.85% for hor-izontal types and 1.35% for vertical lesions, and even byHansen et al. [17] that found 10.8% and 0.5% for horizontaland vertical defects, respectively. This proportion was alsoverified in the study by Gjermo et al. [18] that found 51.8 forhorizontal defects and 16.2 for the vertical types, in a sampleof low socioeconomic condition.

Moreover, our results showed higher prevalence of onelesion defects (67.7%), followed by two (26.5%) and threelesions (5.6%), which was similar to the findings by Hansenet al. [19] that found percentages of 64.80% (one lesion),24.68% (two lesions), and 7.01% (three lesions), respectively.Aass et al. [25] also found high prevalence of one and twolesions in younger individuals (97.5%), as well as Gjermoet al. [18] that showed higher prevalence of one lesion(40.7%) in people with low socioeconomic condition. Addi-tionally, the men showed higher prevalence of bone loss(14.95%) than women (8.2%), whose tendency had alreadybeen revealed by some authors [17–19, 25].

Considering that the alveolar bone loss may increase withage [25], it is important to focus in the young population toprevent the development of periodontal diseases. Thus, otherstudies in young people are encouraged for evaluating theprogression of alveolar bone loss in prospective studies, aswell as for verification of their relationship with the clinicalaspects.

5. Conclusions

It can be concluded that the interproximal radiography wasan efficient method for the detection of alveolar bone loss,revealing low prevalence in adolescents and predominanceof horizontal bone defects.

Conflict of Interests

The authors declare that they have no conflict of interests.


References

[1] T. J. Oh, R. Eber, and H. L. Wang, “Periodontal diseases in thechild and adolescent,” Journal of Clinical Periodontology, vol.29, no. 5, pp. 400–410, 2002.

[2] G. C. Armitage, “Development of a classification system forperiodontal diseases and conditions,” Annals of Periodontology,vol. 4, no. 1, pp. 1–6, 1999.

[3] American Academy of Pediatric Dentistry (AAPD), Endorse-ments Reference Manual, “Periodontal Disease of Childrenand Adolescents,” vol. 30, pp. 156–163, 2003/2004.

[4] G. J. Parfitt, “Five year longitudinal study of gingival conditionof a group of children in England,” Journal of Periodontology,vol. 1, pp. 26–32, 1957.

[5] A. Bjorby and H. Loe, “Gingival and oral hygienic conditionsof school children in Goteborg,” Sveriges TandlakarforbundTidning, vol. 61, no. 11, pp. 561–572, 1969.

[6] G. Vadiakas, C. J. Oulis, K. Tsinidou, E. Mamai-Homata, andA. Polychronopoulou, “Socio-behavioural factors influencingoral health of 12 and 15 year old Greek adolescents. A nationalpathfinder survey,” European Archives of Paediatric Dentistry,vol. 12, no. 3, pp. 139–145, 2011.

[7] E. Bimstein, “Seven-year follow-up of 10 children with peri-odontitis,” Pediatric Dentistry, vol. 25, no. 4, pp. 389–396,2003.

[8] V. Clerehugh, “Periodontal diseases in children and adoles-cents,” British Dental Journal, vol. 204, no. 8, pp. 469–471,2008.

[9] D. A. Perry and M. G. Newman, “Occurrence of periodontitisin an urban adolescent population,” Journal of Periodontology,vol. 61, no. 3, pp. 185–188, 1990.

[10] V. Clerehugh, H. V. Worthington, M. A. Lennon, and R.Chandler, “Site progression of loss of attachment over 5years in 14- to 19-year-old adolescents,” Journal of ClinicalPeriodontology, vol. 22, no. 1, pp. 15–21, 1995.

[11] M. D. Petit and U. van der Velden, “Periodontitis in children,”Nederlands Tijdschrift voor Tandheelkunde, vol. 104, no. 2, pp.67–70, 1997.

[12] J. E. Stoner and A. S. Prophet, “Early periodontal disease inchildren and young adults,” The Dental Practitioner and DentalRecord, vol. 20, no. 5, pp. 173–176, 1970.

[13] S. M. Jenkins, P. M. Dummer, and M. Addy, “Radiographicevaluation of early periodontal bone loss in adolescents. Anoverview,” Journal of Clinical Periodontology, vol. 19, no. 6, pp.363–366, 1992.

[14] B. Sjodin and L. Matsson, “Marginal bone loss in the primarydentition. A survey of 7-9-year-old children in Sweden,”Journal of Clinical Periodontology, vol. 21, no. 5, pp. 313–319,1994.

[15] B. E. C. Toledo, C. K. Rosing, N. C. Jarrim, and R. C. L.Cordeiro, “Analise radiografica da reabsorcao da crista osseaem criancas de 6 a 12 anos,” Stomatos, vol. 4, pp. 29–33, 1997.

[16] N. L. Latcham, R. N. Powell, and J. D. Jago, “A radiographicstudy of chronic periodontitis in 15 year old Queenslandchildren,” Journal of Clinical Periodontology, vol. 10, no. 1, pp.37–45, 1983.

[17] B. F. Hansen, P. Gjermo, and K. R. Bergwitz-Larsen, “Peri-odontal bone loss in 15-year-old Norwegians,” Journal ofClinical Periodontology, vol. 11, no. 2, pp. 125–131, 1984.

[18] P. Gjermo, H. T. Bellini, V. Pereira Santos, J. G. Martins,and J. R. Ferracyoli, “Prevalence of bone loss in a groupof Brazilian teenagers assessed on bite-wing radiographs,”

Journal of Clinical Periodontology, vol. 11, no. 2, pp. 104–113,1984.

[19] B. F. Hansen, P. Gjermo, H. T. Bellini, K. Ihanamaki, and L.Saxen, “Prevalence of radiographic alveolar bone loss in youngadults, a multinational study,” International dental journal, vol.45, no. 1, pp. 54–61, 1995.

[20] R. Shafer, “Periodontal diseases of children and adolescents,”Journal of Periodontology, vol. 74, no. 11, pp. 1696–1704, 2003.

[21] B. Sjodin, L. Matsson, L. Unell, and J. Egelberg, “Marginalbone loss in the primary dentition of patients with juvenileperiodontitis,” Journal of Clinical Periodontology, vol. 20, no.1, pp. 32–36, 1993.

[22] J. M. Albandar, Y. A. Buischi, and M. F. Barbosa, “Destructiveforms of periodontal disease in adolescents. A 3-year longitu-dinal study,” Journal of Periodontology, vol. 62, no. 6, pp. 370–376, 1991.

[23] C. Kallestal, L. Matsson, and A. K. Holm, “Periodontal con-ditions in a group of Swedish adolescents. (I). A descriptiveepidemiologic study,” Journal of Clinical Periodontology, vol.17, no. 9, pp. 601–608, 1990.

[24] A. L. Neely, “Prevalence of juvenile periodontitis in a circum-pubertal population,” Journal of Clinical Periodontology, vol.19, no. 6, pp. 367–372, 1992.

[25] A. M. Aass, T. Tollefsen, and P. Gjermo, “A cohort study ofradiographic alveolar bone loss during adolescence,” Journalof Clinical Periodontology, vol. 21, no. 2, pp. 133–138, 1994.

[26] E. M. Barroso, B. E. C. Toledo, and P. H. R. Souza, “Avaliacaoda reabsorcao ossea periodontal atraves de radiografias inter-proximais em escolares de 11 a 14 anos,” Jornal Brasileiro DeOdontopediatria & Odontologia Para Bebe, vol. 3, pp. 215–218,2000.

[27] A. T. Merchant, W. Pitiphat, J. Parker, K. Joshipura, M. Keller-man, and C. W. Douglass, “Can nonstandardized bitewingradiographs be used to assess the presence of alveolar boneloss in epidemiologic studies?” Community Dentistry and OralEpidemiology, vol. 32, no. 4, pp. 271–276, 2004.

[28] M. K. Jeffcoat, “Radiographic methods for the detection ofprogressive alveolar bone loss,” Journal of Periodontology, vol.63, no. 4, pp. 367–372, 1992.

[29] C. K. Rosing and R. V. Opperman, “Periodontology—Scienceand Clinics,” Sao Paulo: Artes Medicas, pp. 3–17, 2001 (Portu-guese).

[30] P. N. Papapanou and J. L. Wennstrom, “Radiographic andclinical assessments of destructive periodontal disease,” Jour-nal of Clinical Periodontology, vol. 16, no. 9, pp. 609–612, 1989.

[31] E. Hausmann, K. Allen, J. Norderyd, W. Ren, O. Shibly, andE. Machtei, “Studies on the relationship between changes inradiographic bone height and probing attachment,” Journal ofClinical Periodontology, vol. 21, no. 2, pp. 128–132, 1994.

[32] J. M. Albandar, “Prevalence of incipient radiographic peri-odontal lesions in relation to ethnic background and dentalcare provisions in young adults,” Journal of clinical periodon-tology, vol. 16, no. 10, pp. 625–629, 1989.

[33] M. Ayres and M. Ayres Jr, BioEstat—Statistics Application onBiological Sciences and Medicine, vol. 11, 5th edition, 2007.

[34] J. E. Stoner, “An investigation into the accuracy of measure-ments made on radiographs of the alveolar crests of driedmandibles,” Journal of Periodontology, vol. 43, no. 11, pp. 699–701, 1972.

[35] C. Kallestal and L. Matsson, “Criteria for assessment of inter-proximal bone loss on bite-wing radiographs in adolescents,”Journal of Clinical Periodontology, vol. 16, no. 5, pp. 300–304,1989.


[36] R. Blankenstein, J. J. Murray, and O. P. Lind, “Prevalenceof chronic periodontitis in 13-15-year-old children. A radio-graphic study,” Journal of Clinical Periodontology, vol. 5, no. 4,pp. 285–292, 1978.

[37] L. Shapira, E. Tarazi, L. Rosen, and E. Bimstein, “The rela-tionship between alveolar bone height and age in the primarydentition. A retrospective longitudinal radiographic study,”Journal of Clinical Periodontology, vol. 22, no. 5, pp. 408–412,1995.


Research Article

Impact of Periodontal Disease on the Quality of Life ofDiabetics Based on Different Clinical Diagnostic Criteria

Aline Mendes Silva de Pinho,1, 2 Carolina Marques Borges,1

Mauro Henrique Nogueira Guimaraes de Abreu,1

Efigenia Ferreira e Ferreira,1 and Andrea Maria Duarte Vargas1

1 Department of Social and Preventive Dentistry, School of Dentistry, Federal University of Minas Gerais,32170-901 Belo Horizonte, MG, Brazil

2 Rua Maria Heilbuth Surette, 528 /102-Buritis, 30575-100 Belo Horizonte, MG, Brazil

Correspondence should be addressed to Aline Mendes Silva de Pinho, [email protected]

Received 10 June 2012; Accepted 21 August 2012


Copyright © 2012 Aline Mendes Silva de Pinho et al. This is an open access article distributed under the Creative CommonsAttribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work isproperly cited.

The aim of this study was to determine the impact of periodontal disease on the quality of life of individuals with diabetes accordingto different clinical criteria (I-AAP, II-Beck, III-Machtei, IV-Lopez, V-Albandar, VI-Tonetti, and VII-CPI). This cross-sectionalstudy sampled 300 individuals in Belo Horizonte, Brazil. The Oral Health Impact Profile was used to measure the impact ofperiodontal disease on quality of life. Prevalence of periodontal disease was 35.3%, 30.7%, 35.0%, 9.7%, 92.3%, 25.3%, and 75.3%using criteria I, II, III, IV, V, VI, and VII, respectively. The III-Machtei (P = 0.043) and IV-Lopez (P < 0.001) criteria were associatedwith OHIP-14; functional limitation was associated with IV-Lopez (P = 0.006) and V-Albandar (P = 0.018) criteria. Pain was onlyassociated with V-Albandar criteria (P < 0.001). Psychological discomfort was associated with the IV-Lopez (P = 0.018) criteria.Physical disability was associated with the IV-Lopez (P = 0.047) and V-Tonetti (P = 0.046) criteria. Being handicapped wasassociated with the I-AAP (P = 0.025) and II-Beck (P = 0.041) criteria. Concepts of health and disease determined by clinicaldiagnostic criteria may influence the assessment of the impact of periodontal disease on diabetics’ quality of life.

1. Introduction

Diabetes mellitus is one of the primary public healthproblems. Its chronic nature and the limitations it imposescontribute significantly to the increase in hospitalization,disability, and mortality rates. The occurrence of diabetes inpopulation groups is linked mainly to socioeconomic factors,cultural factors, stress, and family predisposition [1]. In theyear 2000, the prevalence of diabetes in the world was 171million, and it is estimated that this number will reach 366million by 2030 [2]. In Brazil, there are approximately 10million people with diabetes.

Studies suggest that diabetes is a risk factor for periodon-tal disease, pointing out that the prevalence, incidence, andseverity of periodontal disease are higher among individualswith diabetes in comparison to healthy individuals [10,11]. The mechanisms by which diabetes influences peri-odontal disease include vascular abnormalities, neutrophil

dysfunction, abnormalities in collagen synthesis, and geneticpredisposition [12].

The most commonly employed clinical parameters forthe diagnosis and classification of periodontal disease aremeasurements of the depth of periodontal pockets, clinicalinsertion loss, and bleeding upon probing, which oftengenerate different information regarding the prevalence ofthe disease. There is no consensus in the scientific literatureregarding a gold standard clinical diagnostic criterion forperiodontal disease, which limits comparison among differ-ent studies [13].

The issue of accuracy of periodontal protocols of theNational Health and Nutrition Examination Survey(NHANES) was discussed. The authors verified that theperiodontal disease case used could lead to an underestimateof the prevalence of periodontal disease by 50% or more[14].


The change in the medical paradigm from focus onthe presence or absence of disease to a broader view thatconsiders the subjective experience of individuals regardingphysical, social, and psychological wellbeing has allowedmore detailed knowledge of the living conditions of thepopulation [15]. Beside the need for the convergence of clin-ical diagnostic criteria, the concomitant use of nonnormativemeasures is important to determining whether periodontaldisease causes an impact on people’s daily lives.

One of the first efforts to document this issue in Brazilwas made by Leao et al., who used the Dental Impact onDaily Living Index to measure periodontal health impacts ondaily living.We made the highlighted change according to thelist of references. Please check Roughly 60% of interviewedindividuals reported some dissatisfaction with their qualityof life [16]. It has been observed that studies relating theimpact of periodontal disease on quality of life of diabeticsare very scarce. Among these few studies, Drumond-Santanaet al. evaluated the potential impact of periodontal diseaseon quality of life in diabetics. They found that diabetics withmild-to-moderate and advanced periodontal disease suffereda more negative impact on quality of life than those who wereperiodontally healthy or had gingivitis [17]. The Oral HealthImpact Profile (OHIP) and its shortened version, the OHIP-14, are indexes that can be used for this purpose [18, 19].

Therefore, based on the hypothesis that the impact ofperiodontal disease on quality of life may vary depending onthe clinical diagnostic criteria used for its diagnosis, the aimof this study was to test associations between different clinicaldiagnostic criteria for the determination of periodontaldisease and its impact on the daily life of individuals withdiabetes.

2. Methods

A cross-sectional study was carried out in 2005 in the city ofBelo Horizonte, state capital of Minas Gerais, Brazil. The cityhas 2,412,937 inhabitants and occupies a strategic positionon the geopolitical maps of Brazil and Latin America [20].The study was approved by the Ethics Committee of theFederal University of Minas Gerais (Protocol no. 012/2004).

Sample calculation considered both type I and II errors.For this, we assumed a 95% confidence interval, 80% powerof test, and parameters of diabetes and quality of life fromDrumond-Santana et al. [17]. The sample calculation bycomparison of proportions presented a higher number thanthose found by comparison of means. Thus, we conducteda different sample size calculation for each dimension ofOHIP-14. This procedure assured, for a sample size of 300individuals, a type I error = 0.05 and a power of test =80% for all dimensions of OHIP-14, except for psychologicaldisability, social disability, and handicap.

We sampled 300 men and women over 30 years of agewith a diagnosis of either type (I and II) of diabetes, andwho were registered and assisted by the Belo Horizontehealth care system. Diagnosis of diabetes was confirmedby municipality primary care physicians. In all participantsthe disease was under control. Sampling procedures werepublished elsewhere [21]. The city of Belo Horizonte has

nine administrative districts with differences in the numberof diabetic patients registered. Simple random samplingwas performed for each district. Thus, the number ofrandomly selected patients was proportional to the numberof diabetic individuals registered in each district. Thenumber of individuals examined was determined by a samplecalculation using a formula recommended by the WorldHealth Organization [22] for the anticipated populationproportion with the occurrence of disease.

The inclusion criteria were individuals who had diabetesdiagnose (either type I or II), up to 30 years old, were regis-tered in the Belo Horizonte public health care system, weredentate, and had signed the informed consent. Additionally,teeth with gingival morphology damage, ill-adapted dentalrestorations, large cavitation, or third molars were excluded.

2.1. Definition of Variables. The outcome was the impact oforal health on the quality of life of adults with diabetes mea-sured by the modified Oral Health Impact Profile (OHIP-14),which has been validated for use in Brazil [23]. As the aimwas to investigate the impact of periodontal disease, the word“gums” was added to 13 questions on the OHIP-14. Thisissue was adopted in accordance with Drumond-Santanaet al. [17]. Responses were dichotomized as “no impact”,which only considered the option “never”, and “impact”,which considered the options “sometimes” and “always”.The administration of the OHIP-14 was carried out at basichealth units of the public healthcare system. Periodontaldisease was assessed by the following clinical criteria: gingivalbleeding upon probing, probing depth, and clinical insertionloss. For the determination of gingival bleeding, a probewas introduced into the gingival pocket until the limit ofits base, waiting 30 to 60 seconds for the determinationof bleeding. Bleeding was recorded (presence or absence)using the modified Ainamo and Bay index [24]. Probingdepth was determined from the measurement of the distancefrom the gingival margin to the bottom of the gingival sulcusor periodontal pocket. The values for each of the surfaces(mesial, distal, vestibular, and lingual) of all the teeth presentwere recorded. Clinical attachment loss was determined fromthe distance between the enamel-cementum junction andthe bottom of the gingival sulcus or periodontal pocket.Attachment loss was measured using the enamel-cementumjunction as the base, when visible.

2.2. Clinical Diagnostic Criteria for the Definition of Peri-odontal Disease. Seven sets of clinical criteria were used todiagnose periodontal disease and are described in Table 1.All exams were performed by one researcher (A.M.S.), whowas submitted to the Kappa test (κ) for the determinationof reliability (intraexaminer agreement). For the Kappa test,all clinical exams were performed at the College of Dentistry,Federal University of Minas Gerais. A total of 588 dentalsurfaces were clinically examined for the Kappa test (κ) cal-culation, with a one-week interval between each participant’sfirst and second examinations. Results were satisfactory forall the clinical criteria: bleeding upon probing (0.79), probingdepth (0.81), and clinical insertion loss (0.91).


Table 1: Description of clinical diagnostic criteria for periodontal disease.

Clinical diagnostic criteria Description

(I) AAPPresence of disease when a person has, at least, one tooth probing depth ≥ 4 mm andclinical insertion loss ≥ 4 mm at the same site [3].

(II) BeckAttachment loss ≥ 5 mm in four or more sites and at least one of these sites with probingdepth ≥ 4 mm [4].

(III) MachteiAttachment loss ≥ 6 mm in two or more teeth and probing depth ≥ 5 mm in one or moresites [5].

(IV) LopezFour or more teeth with at least one site with probing depth ≥ 4 mm and insertion loss ≥3 mm [6].

(V) Albandar

Classifies periodontal disease as mild (1 or more teeth with probing depth ≥ 3 mm),moderate (1 or more teeth with probing depth ≥ 5 mm or 2 or more teeth with probingdepth ≥ 4 mm), and severe (2 or more teeth with probing depth ≥ 5 mm or 4 or moreteeth with probing depth ≥ 4 mm) [7].

(VI) TonettiConsiders attachment loss ≥ 3 mm in two or more nonadjacent teeth or insertionloss ≥ 5 mm in 30% in of teeth [8].

(VII) Community PeriodontalIndex (CPI)

Considers the worst condition encountered in six sites evaluated and used the followingfour codes: 0 = healthy; 1 = absence of pockets, bacterial plaque retention factors, orbleeding following probing; 2 = depth as much as 3 mm and presence of bacterial plaqueretention factors; 3 = pockets with probing depth between 4 and 5 mm; 4 = probingdepth ≥ 6 mm [9]

Table 2: Sample characterization. Belo Horizonte, 2005.

Variable N (%)

GenderMale 120 (40)Female 180 (60)

Age30–54 years old 148 (49.3)≥55 years old 152 (50.7)

Marital statusWith partner 128 (42.7)Without partner 172 (57.3)

Income≤R$ 400 150 (50)>R$ 400 150 (50)

Educational backgroundBachelor/high School 48 (16)Elementary/none 252 (84)

Type of diabetesType I 86 (28.7)Type II 214 (71.3)

During the main study, clinical exams were performedin the dental offices of Brazilian National Health Care unitsunder artificial light, using Williams periodontal probes(Hu-Friedy), mouth mirrors, and gauze. All norms ofbiosafety were followed. To obtain a CPI score, data collectedwere transformed according to CPI criteria.

2.3. Data Analysis. The data were organized in a databaseusing the Statistical Package for Social Sciences (SPSS) ver-sion 15.0 for Windows. Processing involved data codification,entry and editing. Absolute and relative frequencies weredetermined for all variables. Univariate analysis was carriedout to determine the association of each set of clinical criteria

used to estimate periodontal disease and the OHIP-14, usingthe Mann-Whitney, Kruskall Wallis and Pearson chi-squaretests, with the significance level of P < 0.05.

3. Results

Of all study participants, the majority had diabetes type II(71.3%), were female (60%), age 55 or older (50.7%), livingwithout a partner (57.3%), and had low educational level(71.3%) (Table 2).

Table 3 presents the association between OHIP-14 overallresults and periodontal disease, according to each one of theclinical diagnostic criteria. Only III-Machtei (P = 0.043) andIV-Lopez (P < 0.001) were statistically associated with overallOHIP-14 (Table 3). The prevalence of periodontal diseasevaried depending on the clinical criteria employed: 35.3%(I-AAP), 30.7% (II-Beck), 35.0% (III-Machtei), 9.7% (IV-Lopez); 92.3% (V-Albandar), 25.3% (VI-Tonetti), and 75.3%(VII-CPI) (Figure 1).

Regarding OHIP-14 subscales, functional limitation wasassociated with IV-Lopez (14.9% of diabetics with peri-odontal disease suffered an impact on quality of life;P = 0.006) and V-Albandar (28.4%, 8.2%, and 53.7% ofdiabetics with mild, moderate, severe periodontal disease,resp. suffered an impact on quality of life; P = 0.018); Painwas only associated with the V-Albandar criteria (28.1%,14.5%, 49.2% of diabetics with mild, moderate, and severeperiodontal disease, resp. suffered an impact on quality oflife; P < 0.001). Psychological discomfort was associatedwith IV-Lopez (12.3% of diabetics with periodontal diseasesuffered an impact on quality of life; P = 0.018). Physicaldisability was associated with IV-Lopez (13.2% of diabeticswith periodontal disease suffered an impact on quality oflife; P = 0.047) and V-Tonetti (30.6% of diabetics withperiodontal disease suffered an impact on quality of life;P = 0.046) criteria. Social disability was only associated with


Table 3: Mean, standard deviation, median, and P value, OHIP-14 among diabetics. Belo Horizonte, 2005.

Diagnostic criteria Mean (± SD) Median P value

(I) AAPHealthy 7.56 (5.86) 6.45

0.120Periodontal disease 8.79 (6.43) 7.72

(II) BeckHealthy 7.57 (5.86) 6.70


(III) MatcheiHealthy 7.49 (5.99) 6.40


(IV) LopezHealthy 7.57 (5.92) 6.50

<0.001Periodontal disease 11.91 (6.34) 12.38

(V) Albandar

Healthy 7.05 (4.70) 5.72

0.265Mild PD 7.32 (6.22) 5.60

Moderate PD 7.65 (6.39) 7.00

Severe PD 8.64 (6.09) 8.07

(VI) TonettiHealthy 7.66 (5.94) 6.81


(VII) CPI

Healthy 7.87 (5.52) 8.06

0.470Bleeding 7.16 (5.83) 4.98

Pocket 4-5 mm 7.73 (6.17) 6.90

Pocket ≥ 6 mm 8.69 (6.09) 7.06

PD: Periodontal disease.

100

90

80

70

60

50

40

30

20

10

0AAP Beck Machtei Lopez Albandar Tonetti CPI

Prevalence

Figure 1: Prevalence of periodontal disease according to clinicaldiagnostic criteria. Belo Horizonte, 2005.

IV-Lopez (13.0% of diabetics with periodontal disease suf-fered an impact on quality of life; P = 0.056). A handicap wasassociated with I-AAP (47.0% of diabetics with periodontaldisease suffered an impact on quality of life; P = 0.025) andII-Beck (40.9% of diabetics with periodontal disease sufferedan impact on quality of life; P = 0.041) criteria (Tables 4 and5).

4. Discussion

The choice of one of the different sets of clinical criteriafor the diagnosis of periodontal disease is not a simple task,since there is no consensual recommendation regarding theuse of these measures [3–5]. The measurement of health ordisease does not go beyond the tooth or site and controversial

measures impede the determination of an individual as eitherhealthy or sick. For instance, in our study the prevalence ofperiodontal disease ranged from 9.7% (IV-Lopez) to 92.3%(V-Albandar), depending on which set of clinical diagnosticcriteria was used.

This disparity calls for reflection. Some of the clinicaldiagnostic criteria used in the present study consider theresult of the exam by the number of affected sites, whereasothers consider the number of affected teeth or a com-bination of teeth and sites. Moreover, depending on themanner in which the measurement is determined, a singletooth may have four or six sites. Manau et al. investigatedwhether the application of different definition criteria ofperiodontal disease had an influence on the significance ofthe association between periodontal disease and prematurityor low birth weight. They used fourteen definitions ofperiodontal disease and results showed that significance ofthe association between periodontal disease and pregnancyoutcomes may be determined by the periodontal diseasedefinition or measurement used [13]. A similar study wascarried out involving cases and controls in order to comparefour definitions of periodontal disease and respective asso-ciations with premature birth/low birth weight; the authorsconcluded that the magnitude of the association varieddepending on the definition of periodontal disease employed[25].

Two clinical criteria of periodontal disease (Machtei andLopez) were statistically associated with the OHIP-14 overallscore. The Machtei criteria may be considered the broadestfor determining the prevalence of periodontal disease, asit considers insertion loss ≥ 6 mm in two or more sitesand probing depth ≥ 5 mm in one or more sites for thediagnosis of the disease. The Lopez criteria appear to be morecautious regarding the extension of periodontal disease, as


Ta

ble

4:P

reva

len

ceof

peri

odon

tald

isea

seam

ong

diab

etic

sba

sed

onA

AP,

Bec

k,M

atch

ei,L

opez

,an

dTo

net

ticr

iter

iaan

dea

chof

the

seve

nO

HIP

subs

cale

s;P

valu

e;B

elo

Hor

izon

te,2

005.

OH

IPsu

bsca

les

Fun

ctio

nal

limit

atio

nPa

inPs

ych

olog

ical

disc

omfo

rtP

hysi

cald

isab

ility

Psyc

hol

ogic

aldi

sabi

lity

Soci

aldi

sabi

lity

Han

dica

pPe

riod

onta

ldis

ease

No

(%)

Yes

(%)

No

(%)

Yes

(%)

No

(%)

Yes

(%)

No

(%)

Yes

(%)

No

(%)

Yes

(%)

No

(%)

Yes

(%)

No

(%)

Yes

(%)

AA

PP=

0.87

4P=

0.88

0P=

0.57

9P=

0.78

7P=

0.28

6P=

0.19

1P=

0.02

5∗

No

impa

ct10

8(6

5)58

(34.

9)38

(65.

5)20

(34.

5)59

(67)

29(3

3)10

2(6

5.4)

54(3

4.6)

49(7

0)21

(30)

105

(68.

2)49

(31.

8)15

9(6

7.9)

74(3

2.1)

Impa

ct86

(64)

48(3

5.8)

156

(64.

5)86

(35.

5)13

5(6

3.7)

77(3

6.3)

92(6

3.9)

52(3

6.1)

145

(63)

85(3

7)89

(61)

57(3

9)35

(53)

31(4

7)B

eck

P=

0.63

1P=

0.70

0P=

0.41

1P=

0.47

7P=

0.46

5P=

0.19

0P=

0.04

1∗

No

impa

ct11

7(7

0.5)

49(2

9.5)

39(6

7.2)

19(3

2.8)

64(7

2.7)

24(2

7.3)

111

(71.

2)45

(28.

8)51

(72.

9)19

(27.

1)11

2(7

2.7)

42(2

7.3)

169

(72.

2)65

(27.

8)Im

pact

91(6

7.9)

43(3

2.1)

169

(69.

8)73

(30.

7)14

4(6

7.9)

68(3

2.1)

97(6

7.4)

47(3

2.6)

157

(69.

3)73

(31.

7)96

(65.

8)50

(30.

7)39

(59.

1)27

(40.

9)M

ach

tei

P=

0.21

4P=

0.25

7P=

0.31

2P=

0.52

9P=

0.31

7P=

0.09

5P=

0.39

7N

oim

pact

113

(68.

1)53

(31.

9)34

(58.

6)24

(41.

4)61

(69.

3)27

(30.

7)10

4(6

6.7)

52(3

3.3)

49(7

0)21

(30)

107

(69.

5)47

(30.

5)15

5(6

6.2)

79(3

3.8)

Impa

ct82

(61.

2)52

(38.

8)16

1(6

6.5)

81(3

3.5)

134

(63.

2)78

(36.

8)91

(63.

2)53

(36.

8)14

6(6

3.5)

84(3

6.5)

88(6

0.3)

58(3

9.7)

40(6

0.6)

26(3

9.4)

Lop

ezP=

0.00

6∗P=

0.76

4P=

0.01

8∗P=

0.04

7∗P=

0.08

2P=

0.05

6P=

0.85

8N

oim

pact

157

(94.

6)9

(5.4

)53

(91.

4)05

(8.6

)85

(96.

6)03

(3.4

)14

6(9

3.6)

10(6

.4)

67(9

5.7)

03(4

.3)

144

(93.

5)10

(6.5

)21

1(9

0.2)

23(9

.8)

Impa

ct11

4(8

5.1)

20(1

4.9)

218

(90.

1)24

(9.9

)18

6(8

7.7)

26(1

2.3)

125

(86.

8)19

(13.

2)20

4(8

8.7)

26(1

1.3)

127

(87)

19(1

3)60

(90.

9)06

(9.1

)To

net

tiP=

0.58

3P=

0.14

8P=

0.93

2P=

0.04

6∗P=

0.47

7P=

0.42

4P=

0.17

0N

oim

pact

126

(75.

9)40

(24.

1)39

(67.

2)19

(32.

8)66

(75)

22(2

5)12

4(7

9.5)

32(2

0.5)

50(7

1.4)

20(2

8.6)

118

(76.

6)36

(23.

4)17

9(7

6.5)

55(2

3.5)

Impa

ct98

(73.

1)36

(26.

9)18

5(7

6.4)

57(2

3.6)

158

(74.

5)54

(25.

5)10

0(7

4.7)

44(3

0.6)

174

(75.

7)56

(24.

3)10

6(7

2.6)

40(2

7.4)

45(6

8.2)

21(3

1.8)

∗ Stat

ical

lysi

gnifi

can

t.


Ta

ble

5:P

reva

len

ceof

peri

odon

tald

isea

seam

ong

diab

etic

sba

sed

onA

lban

dar

and

CP

Icr

iter

iaan

dea

chof

the

seve

nO

HIP

subs

cale

s;P

valu

e;B

elo

Hor

izon

te,2

005.

Alb

anda

rC

omm

un

ity

per

iodo

nta

lin

dex

Hea

lthy

(%)

Mild

(%)

Mod

erat

e(%

)Se

vere

(%)

Pva

lue

Hea

lthy

Ble

edin

gPo

cket

=4

or5

mm

Pock

et≥

6m

mP

valu

e∗

Fun

ctio

nal

limit

atio

nN

oim

pact

10(6

)48

(28.

9)34

(20.

5)74

(44.

6)0.

018∗

14(8

.4)

28(1

6.9)

69(4

1.6)

55(3

3.1)

0.50

6Im

pact

13(9

.7)

38(2

8.4)

11(8

.2)

72(5

3.7)

15(1

1.2)

21(1

5.7)

46(3

4.3)

52(3

8.8)

Pain N

oim

pact

03(5

.2)

18(3

1)10

(17.

2)27

(46.

6)<

0.00

1∗06

(10.

3)10

(17.

2)25

(43.

1)17

(29.

3)0.

726

Impa

ct20

(8.3

)68

(28.

1)35

(14.

5)11

9(4

9.2)

23(9

.5)

39(1

6.1)

90(3

7.2)

90(3

7.2)

Psyc

hol

ogic

aldi

scom

fort

No

impa

ct07

(8)

28(3

1.8)

13(1

4.8)

40(4

5.5)

0.87

007

(8)

15(1

7)38

(43.

2)28

(31.

8)0.

638

Impa

ct16

(7.5

)58

(27.

4)32

(15.

1)10

6(5

0)22

(10.

4)34

(16)

77(3

6.3)

79(3

7.3)

Phy

sica

ldis

abili

tyN

oim

pact

13(8

.3)

43(2

7.6)

21(1

3.5)

79(5

0.6)

0.77

714

(9)

25(1

6)61

(39.

1)56

(35.

9)0.

972

Impa

ct10

(6.9

)43

(29.

9)24

(16.

7)67

(46.

5)15

(10.

4)24

(16.

7)54

(37.

5)51

(35.

4)Ps

ych

olog

ical

disa

bilit

yIm

pact

05(7

.1)

24(3

4.3)

10(1

4.3)

31(4

4.3)

0.69

908

(11.

4)11

(15.

7)30

(42.

9)21

(30)

0.65

4N

oim

pact

18(7

.8)

62(2

7)35

(15.

2)11

5(5

0)21

(9.1

)38

(16.

5)85

(37)

86(3

7.4)

Soci

aldi

sabi

lity

No

impa

ct15

(9.7

)49

(31.

8)22

(14.

3)68

(44.

2)0.

231

15(9

.7)

29(1

8.8)

57(3

7)53

(34.

4)0.

638

Impa

ct08

(5.5

)37

(25.

3)23

(15.

8)78

(53.

4)14

(9.6

)20

(13.

7)58

(39.

7)54

(37)

Han

dica

pN

oim

pact

18(7

.7)

71(3

0.3)

34(1

4.5)

111

(47.

4)0.

675

25(1

0.7)

40(1

7.1)

89(3

8.8)

80(3

4.2)

0.53

4Im

pact

05(7

.6)

15(2

2.7)

11(1

6.7)

35(5

3)04

(6.1

)09

(13.

6)26

(39.

4)27

(40.

9)∗ St

atis

cally

sign

ifica

nt.


they consider four or more teeth with at least one site withprobing depth ≥ 4 mm and insertion loss ≥ 3 mm.

It is appropriate to highlight that Community Peri-odontal Index (CPI) was statistically associated with neitherthe overall OHIP-14 nor with any of its subscales. CPIis recommended as an epidemiological tool by the WorldHealth Organization (WHO), and it has been widely usedinternationally. In our study, we were able to classify theworst CPI criteria code 3 or 4 (with pockets). If the patienthad only one site probing ≥ 4 mm, he/she was diagnosedas having periodontal disease. This may explain why therewas no association between impact on quality of life andperiodontal disease.

The test of the association of each of the seven setsof criteria and the OHIP-14 dimensions showed that theLopez criteria were statistically associated with functionallimitation, psychological discomfort, physical disability, andsocial disability (limitrophe P value for this last one).This evidence may be explained considering that the morestrict a clinical criterion is, the higher the chances are ofit having an association with negative impact on qualityof life. The Albandar criterion was only associated withpain. Psychological disability was not associated with anyof sets of clinical criteria for the diagnosis of periodontaldisease. Depending on the clinical criteria used, there weredifferences in the impact of periodontal disease on quality oflife, even when measured by the same questionnaire (OHIP-14).

Different cut-off points for probing depth and insertionloss imply differences in prevalence rates and the resultsof tests on associated factors. Criteria for the definitionof the presence and absence of periodontal disease arecontroversial, and there is a clear need to establish a cut-offpoint between health and disease for periodontics, as well asestablish the meaning of the disease for individuals. Towardsthis end, some studies have adopted more rigorous criteriato minimize overestimated rates of periodontal disease andits association with quality of life. Severe periodontal disease,measured by Beck criterion, will negatively impact on thediabetic’s life, as measured by OHIP-14 [17].

It is worth noting that the majority of studies pub-lished so far comprise the issue of periodontal disease andquality of life in individuals without diabetes. Bernabe andMarcenes (2010) investigated the association between qualityof life and periodontal disease among 3,122 British adultsand found that they were inversely associated in spite ofsocioeconomic conditions, demographic factors, and clinicalconditions [26]. Among the few studies which focused onthe relation between diabetes and quality of life, Sandbergand Wikblad (2003) pointed out that having diabetes playedan important role in the domains physical functioning, rolefunctioning physical, general health and social functioning,when compared to patients without a history of diabetes[27].

Although our study included a representative sampleof diabetic patients in the public healthcare system ofBelo Horizonte, Brazil, it is important to point out somelimitations of this research. First, our study is cross-sectional,which does not allow any inference with regard to causality.

Second, our sample size does not give us reasonable valuesfor type I and II errors for the dimensions “psychologicaldisability,” “social disability” and “handicap.” The mainissue raised by the results of the present study is that thedefinition of a disease established by a given set of clinicaldiagnostic criteria may either overestimate or underestimatethe chances of an individual being considered sick or healthy.Likewise, the impact periodontal disease has on quality oflife may vary depending on the definition of the diseaseestablished by one of the seven sets of diagnostic criteriaanalyzed here. Additionally, we did not register the meanduration of diabetes, and therefore were not able to considerdevelopment of comorbidities that negatively impact qualityof life.

As health and sickness are terms that correspond tounique experiences laden with subjectivity, it seems unlikelythat these terms are recognized in the totality of theirmeanings [28]. The concrete phenomenon of becoming ill,self-perception of the individual, and professional diagnosismost often fail to coexist in harmony [29]. This means thatthe same disease may be seen from different standpoints,the consequences of which are the differences identified inthe impact that periodontal disease has on the daily life ofaffected individuals.

5. Conclusion

The concept of health or disease as determined by clinicaldiagnostic criteria may influence the results of the assessmentof the impact of periodontal disease on quality of life ofdiabetics, while the perception of affected individuals isunique. Studies directed at the establishment of standardizedcriteria for the diagnosis of periodontal disease are needed.Such studies should be focused on the health-disease process,including its social determination. A reference standard forepidemiological studies addressing periodontal disease couldcontribute to the organization of health actions in the field ofcollective health.

Acknowledgment

The authors whould like to thank FAPEMIG-Fundacao deAmparo a Pesquisa de Minas Gerais for financial support.

References

[1] Brasil, Ministerio da Saude, Secretaria de Polıticas de Saude,“Plano de reorganizacao da atencao a hipertensao arterial eao diabete mellitus,” Brasılia, Brazil, 2001, http://bvsms.saude.gov.br/bvs/publicacoes/miolo2002.pdf.

[2] World Health Organization, http://www.who.int/en/.[3] American Academy of Periodontology, “Parameter on chronic

periodontal disease with advanced loss of periodontal sup-port,” Journal of Periodontology, vol. 71, no. 5, pp. 852–855,2000.

[4] J. D. Beck, G. G. Koch, R. G. Rozier, and G. E. Tudor, “Preva-lence and risk indicators for periodontal attachment loss in apopulation of older community-dwelling blacks and whites,”Journal of periodontology, vol. 61, no. 8, pp. 521–528, 1990.


[5] E. E. Machtei, L. A. Christersson, S. G. Grossi, R. Dunford, J. J.Zambon, and R. J. Genco, “Clinical criteria for the definitionof ‘established periodontitis’,” Journal of Periodontology, vol.63, no. 3, pp. 206–214, 1992.

[6] N. J. Lopez, P. C. Smith, and J. Gutierrez, “Higher risk ofpreterm birth and low birth weight in women with periodon-tal disease,” Journal of Dental Research, vol. 81, no. 1, pp. 58–63, 2002.

[7] J. M. Albandar, “A 6-year study on the pattern of periodontaldisease progression,” Journal of Clinical Periodontology, vol. 17,no. 7, pp. 467–471, 1990.

[8] M. S. Tonetti and N. Claffey, “Advances in the progression ofperiodontitis and proposal of definitions of a periodontitiscase and disease progression for use in risk factor research:group C Consensus report of the 5th European workshop inperiodontology,” Journal of Clinical Periodontology, vol. 32, no.6, pp. 210–213, 2005.

[9] World Health Organization, Oral Health Surveys: Basic Meth-ods, WHO, Geneva, Switzerland, 4th edition, 1997.

[10] S. Bjelland, P. Bray, N. Gupta, and R. Hirsch, “Dentists, dia-betes and periodontitis,” Australian Dental Journal, vol. 47, no.3, pp. 202–207, 2002.

[11] A. Kasaj, G. G. Zafiropoulos, H. Tekyatan, A. Pistorius, and B.Willershausen, “Periodontal disease status of pregnant womenwith diabetes mellitus,” Collegium Antropologicum, vol. 32, no.1, pp. 115–118, 2008.

[12] R. C. Oliver and T. Tervonen, “Diabetes—a risk factor forperiodontitis in adults?” Journal of Periodontology, vol. 65, no.5, pp. 530–538, 1994.

[13] C. Manau, A. Echeverria, A. Agueda, A. Guerrero, and J. J.Echeverria, “Periodontal disease definition may determine theassociation between periodontitis and pregnancy outcomes,”Journal of Clinical Periodontology, vol. 35, no. 5, pp. 385–397,2008.

[14] P. I. Eke, G. O. Thornton-Evans, L. Wei, W. S. Borgnakke, andB. A. Dye, “Accuracy of NHANES periodontal examinationprotocols,” Journal of Dental Research, vol. 89, no. 11, pp.1208–1213, 2010.

[15] A. T. Leao and A. Sheiham, “The dental impact on dailyliving,” in Measuring Oral Quality of Life, G. D. Slade, Ed.,University of North Carolina, Dental Ecology, Chapel Hill,NC, USA, 1997.

[16] A. T. T. Leao, M. C. Cidade, and J. R. Varela, “Impactos dasaude periodontal na vida diaria,” Revista Brasileira de Odon-tologia, vol. 55, no. 4, pp. 238–241, 1998.

[17] T. Drumond-Santana, F. O. Costa, E. G. Zenobio, R. V. Soares,and T. D. Santana, “Impact of periodontal disease on qualityof life for dentate diabetics,” Cadernos de Saude Publica, vol.23, no. 3, pp. 637–644, 2007.

[18] G. D. Slade and A. J. Spencer, “Development and evaluation ofthe oral health impact profile,” Community Dental Health, vol.11, no. 1, pp. 3–11, 1994.

[19] G. D. Slade, “Derivation and validation of a short-form oralhealth impact profile,” Community Dentistry and Oral Epi-demiology, vol. 25, no. 4, pp. 284–290, 1997.

[20] “Prefeitura Municipal de Belo Horizonte,” Belo Horizonte,Brazil, 2008, http://portalpbh.pbh.gov.br/pbh/.

[21] A. M. Silva, A. M. D. Vargas, E. Ferreira, and M. H. N. G.de Abreu, “Periodontitis in individuals with diabetes treatedin the public health system of Belo Horizonte, Brazil,” RevistaBrasileira de Epidemiologia, vol. 13, no. 1, pp. 118–125, 2010.

[22] S. K. Lwanga and S. Lemeshow, Sample Size Determination inHealth Studies: A Pratical Manual, WHO, Geneva, Switzerland.

[23] B. H. de Oliveira and P. Nadanovsky, “Psychometric propertiesof the Brazilian version of the oral health impact profile—short form,” Community Dentistry and Oral Epidemiology, vol.33, no. 4, pp. 307–314, 2005.

[24] J. Ainamo and I. Bay, “Problems and proposals for recordinggingivitis and plaque,” International Dental Journal, vol. 25,no. 4, pp. 229–235, 1975.

[25] I. S. Gomes-Filho, S. S. Cruz, E. J. C. Rezende et al., “Exposuremeasurement in the association between periodontal diseaseand prematurity/low birth weight,” Journal of Clinical Peri-odontology, vol. 34, no. 11, pp. 957–963, 2007.

[26] E. Bernabe and W. Marcenes, “Periodontal disease and qualityof life in British adults,” Journal of Clinical Periodontology, vol.37, no. 11, pp. 968–972, 2010.

[27] G. E. Sandberg and K. F. Wikblad, “Oral health and health-related quality of life in type 2 diabetic patients and non-diabetic controls,” Acta Odontologica Scandinavica, vol. 61, no.3, pp. 141–148, 2003.

[28] D. Czeresnia, “O conceito de saude e a diferenca entre pre-vencao e promocao,” in Promocao da saude. Conceitos, reflex-oes, tendencias, D. Czeresnia and C. M. Freitas, Eds., Fiocruz,Rio de Janeiro, Brazil, 4th edition, 2008.

[29] S. Caponi, “A saude como abertura ao risco,” in Promocao dasaude. Conceitos, reflexoes, tendencias, D. Czeresnia and C. M.Freitas, Eds., Fiocruz, Rio de Janeiro, Brazil, 4th edition, 2008.


Research Article

Individual and Contextual Determinants ofPeriodontal Health in 12-Year-Old Schoolchildren ina Brazilian Capital City

Lidia Moraes Ribeiro Jordao,1 Daniela Nobre Vasconcelos,1 Rafael da Silveira Moreira,2

and Maria do Carmo Matias Freire1

1 School of Dentistry, Federal University of Goias, 74605-220 Goiania, GO, Brazil2 Department of Public Health, Aggeu Magalhaes Institute, Oswaldo Cruz Foundation, Ministry of Health, 50670-420 Recife, PE,Brazil

Correspondence should be addressed to Lidia Moraes Ribeiro Jordao, [email protected]

Received 12 July 2012; Accepted 21 August 2012


Copyright © 2012 Lidia Moraes Ribeiro Jordao et al. This is an open access article distributed under the Creative CommonsAttribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work isproperly cited.

The study aimed to describe periodontal health status and its association with individual and contextual factors among 12-year-old schoolchildren in a midwest Brazilian capital city. This cross-sectional study included data from an oral health surveycarried out in 2010 in the city of Goiania, Brazil (n = 2, 075) and secondary data obtained from the local health authority.Data were collected through oral clinical examinations and interviews. For assessment of periodontal status two componentsof the community periodontal index (CPI) were used: calculus and bleeding after probing. Dependent variable was presence ofany periodontal condition. Independent individual variables were the children’s sex and color/race, and their mother’s level ofschooling. Contextual variables were related to the schools (type and existence of toothbrushing program) and its geographiclocation in the health districts. Rao-Scott test and multilevel Poisson analysis were performed. The prevalence of calculus and/orbleeding was 7%. Brown color, public schools, and those located in health district with intermediate socioeconomic indicatorswere associated to a higher prevalence of this condition. The prevalence of adverse periodontal condition was low and theinequalities in its distribution were determined by individual as well as contextual factors related to the schools and the geographicarea.

1. Introduction

Alongside caries, periodontal diseases remain as global oralhealth burdens. Periodontitis constitute a major cause oftooth loss in adults worldwide, and most children andadolescents exhibit signs of gingivitis [1].

This is particularly the case in underprivileged sub-populations in both developing and developed countries[2, 3]. Even in high-income countries with advanced publicoral health care, inequalities in periodontal health remaina major public health issue [4]. Likely reasons to accountfor these prevalent diseases include genetic, epigenetic, andenvironmental risk factors, as well as other individual andsocioeconomic determinants [1].

Adolescence can be a critical period in the individ-ual’s periodontal status. Epidemiological and immunolog-ical studies suggest that irreversible tissue damage fromperiodontal disease begins in late adolescence and earlyadulthood [5]. Gingivitis prevalence, severity and extentincrease with age, beginning in the deciduous dentition andreaching a peak at puberty, followed by a limited decline inadolescence [6–8].

At a population level, plaque and calculus deposition, aswell as occurrence of gingivitis and periodontitis, are slightlyhigher in boys than in girls [6]. Dentofacial anomalies,toothbrushing frequency, socioeconomic and psychologicalconditions experienced in early life and life course have alsobeen associated with gingival bleeding in adolescents [9–11].


Calculus and gingival bleeding are clinical indicators ofpoor oral hygiene and periodontal condition [12] and havebeen frequently used in epidemiological studies as part ofthe community periodontal index (CPI) [13] or separatelyanalyzed.

Both gingival bleeding and calculus in Brazilian ado-lescents have been significantly associated with race, familyincome, levels of schooling, and type of school attended[10, 12, 14, 15].

Current studies show that, regardless the socioeconomicindicator used, people who are socioeconomically disadvan-taged consistently have poorer periodontal outcomes [16].However, investigation of the influence of socioeconomicstatus on the etiological pathway of periodontal condition isstill required to better understand its determinants.

Moreover, global reports have emphasized that con-tinuing surveillance of levels and patterns of risk factorsis of fundamental importance to planning and evaluatingcommunity preventive activities and oral health promotionin all parts of the world [17].

In that context, multilevel analysis technique considersboth people and areas on the distribution and determinantsof population health, being an important approach tounderstand the significance of specific contexts for differentindividual health outcomes [18]. So far, there are veryfew studies investigating the influence of contextual vari-ables using multilevel analysis in the periodontal status ofschoolchildren [12, 19].

The present study aimed to describe periodontal healthstatus and its association with individual and contextualfactors among 12-year-old schoolchildren in a midwestBrazilian capital city.

2. Material and Methods

2.1. Type of Study and Source of Data. The present cross-sectional study was carried out in the city of Goiania, capitalcity of Goias State, located in the Midwest region of Brazil.The analysis included primary data from an epidemiologicalsurvey of oral health carried out in 2010. Secondary datawere obtained from the local health authority in the sameyear.

The oral health survey of 12-year-old schoolchildrenin the urban area followed the methodology of the 2010Brazilian National Survey of Oral Health [20]. Clinical exam-inations were according to diagnostic criteria establishedby the World Health Organization (WHO) [21]. Althoughdata on other oral conditions were collected, only data onperiodontal condition was included in the present study.

2.2. Ethical Aspects. The research protocol was approvedby the Ethics Committee of the Federal University ofGoias, Brazil (Report 226/2010) and only the schools andthe children whose parents signed an informed consentparticipated in the study.

2.3. Sample. The age 12 years was chosen following theWHO recommendations for the monitoring of oral health

status among children. It is generally the age at whichchildren leave primary school, and therefore it is the last ageat which a reliable sample may be easily obtained through theschool system [21].

Sample size was calculated to be representative of the12-year-old schoolchildren in Goiania. We used the clustersampling technique and the sample was randomly selected intwo stages. Initially, we draw the number of first stage unities(schools), followed by second stage unities (schoolchildren).

According to data obtained from the State and the City’sEducation Department, the total number of 12-year-oldschoolchildren enrolled in 2009 was 17,911 in 281 publicand private schools. Sample size was calculated using anequation for proportions for infinite populations based oncaries prevalence, using the Epi Info software, version 3.5.1.The minimum number of schoolchildren to take part in theresearch was 2,171, considering a confidence interval of 95%,sampling error of 2%, and caries prevalence of 65.3%. Foreffect of study sample design, a simplified and conservativecorrection was needed, multiplying the obtained sample sizeby 1.2 (an extra 20%). The final sample size was 2,605schoolchildren.

To calculate the amount of schools, we used a formulathat consisted of multiplication of the number of schools bythe number of schoolchildren of the sample, divided by thetotal number of 12-year-old schoolchildren in Goiania. Wethen obtained a sample of 41 schools.

The sample was equitably distributed in the seven healthdistricts in the city: Central, Eastern, Northwestern, North-ern, Western, Southwestern, and Southern. Each healthdistrict comprises a geographic area with specific population,well-defined health issues, and unique interaction withhealth care teams.

As the total sample was of 2,605 schoolchildren, and thenumber obtained from the list was approximately of 2,962schoolchildren, we opted for including all students attendingthe selected schools.

2.4. Data Collection. Data were collected through oralclinical examinations and interviews with the children bysix teams composed by one dentist and one recorder, whoworked in the public health service. They were previouslytrained and calibrated, according to the criteria used inthe 2010 Brazilian National Survey of Oral Health [20].Inter-examiner Kappa coefficient for periodontal conditionvaried from 0.68 to 1.00, showing a good to excellentreproducibility.

Intraoral examinations were carried out at the schoolsusing a mouth mirror and a WHO periodontal probe undernatural light, with the children seated in school chairs.

For the assessment of periodontal status, two compo-nents of the Community Periodontal Index (CPI) [20] wereused: calculus and bleeding. Periodontal pockets were notregistered. Each sextant of the mouth was examined forcalculus detected during probing and bleeding observed afterprobing. Index teeth were 16, 11, 26, 36, 31, and 46. Ineach tooth, six sites were examined, and the worst conditionwas recorded. Gingival bleeding was recorded from 10 to 30


o Type of school

o

Contextual level

• City,s health district

o Group I

o Group II

o Group III

(public/private)

Existence of toothbrushing program

• Schools

Individual level

Demographic variables:

Sex, color/race

Socioeconomic variable:

Mother,s level of schooling

Periodontal health status

Calculus and/or bleeding

∗City’s health districts: (group I) with the best socio-economicindicators, (groupII) with intermediate socioeconomic indicators,and (group III) with the worst socioeconomic indicators.

∗

Figure 1: Theoretical hierarchical proposed model of the associa-tion between individual and contextual variables and periodontalhealth status of 12-year-old schoolchildren. Goiania-GO, Brazil,2010.

seconds after probing. Recommended probe placement wasof approximately 60 degrees inclination in relation to thelongitudinal axis of the tooth.

Information on demographic and socioeconomic char-acteristics of the participating children was also collected:sex, self-rated skin color or race, and mother’s level ofschooling. Self-rated skin color or race followed criteriaproposed by the Brazilian Institute of Geography and Statis-tics (IBGE): white, black, yellow (Asian origin individuals),brown, or indigenous. Mother’s level of schooling was basedon completed years of study and was obtained from thechildren’s school records.

Secondary data were: type of school (public and private),existence of toothbrushing program at the schools, and thecity’s health districts where the schools were located.

The School Toothbrushing Program was created in 1992through a partnership between the local Health and Educa-tion Secretaries. Its aim is to improve the oral health statusof schoolchildren enrolled in public schools of Goiania via

implementing daily toothbrushing with fluoride toothpasteafter the school meal.

2.5. Data Analysis. The dependent variable was the preva-lence of periodontal condition, featured by the presence ofcalculus and/or bleeding (yes or no).

Independent variables were divided into two levels ofdata organization: individual (schoolchildren) and contex-tual (schools and health districts). In the individual level,we analyzed children’s demographic characteristics (sexand skin color or race), and one socioeconomic indicator(their mothers’ level of schooling). Self-rated color/race wascategorized in: white, black, and brown. Due to the smallnumber of individuals who rated themselves as yellow orindigenous, these categories were excluded from the analysis.Level of mother’s schooling was grouped as follows: less thaneight, from eight to eleven and more than eleven, years ofstudy.

Contextual variables were type of school (public andprivate), existence of toothbrushing program at the school,and city’s health district where schools were located. Theseven health districts were grouped according to theirsocioeconomic characteristics, as informed by the localhealth authorities. The Health District located more centrally(Central-Campinas) presented better-off socioeconomic andhealth indicators than the others, which were located inthe outskirts. Therefore, we have classified them in threecategories: (Group I) with the best indicators (Central-Campinas); (Group II) with intermediate indicators (North,South and East); (Group III) with the worst indicators(Southwest, West, and Northwest).

Individual and contextual variables were initiallydescribed according to the prevalence of periodontal status.Rao-Scott test [22], an equivalent to chi-square test for com-plex samples, was used to test dependence between variables.This analysis was performed using the Stata 12 software,considering the complex sample plan and sample weights.

After that, we performed a multilevel analysis withrandom intercept [23], considering three levels: schoolchil-dren, schools, and health districts (Figure 1). Poisson log-linear regression models were adopted, with robust varianceestimator, to estimate prevalence ratio as effect measure,and its confidence interval of 95% measured by Wald test.We carried out only unadjusted analysis due to collinearitybetween the independent variables that could jeopardize datain the multiple analysis.

In addition to considering the sample design throughvariance partition in each level, multilevel analysis allows forthe inclusion of contextual variables of higher levels thanthe individual. This allows for the simultaneous approach ofcontextual and individual factors in the analysis. Multilevelanalysis was performed using the MLwiN 2.02 software. Forall analysis, we considered the 5% significance level andsample weights derived from sample design (school weights).

3. Results

Of the 41 schools invited to participate, 39 accepted (24public and 15 private). Of the 2,962 schoolchildren invited


Table 1: Frequency distribution of periodontal conditions (calculus and/or bleeding), according to independent variables. 12-year-oldschoolchildren, Goiania, Brazil, 2010.

Presence of calculus and/or bleeding Total n (%) PYes % (CI 95%) No % (CI 95%)

Individual variables

Sex 0.121

Female 6.2 (4.3–8.9) 93.8 (91.0–95.7) 1,022 (49.1)

Male 7.7 (6.0–10.1) 92.3 (89.9–94.0) 1,053 (50.9)

Color/race 0.005∗

White 4.8 (3.2–7.0) 95.2 (92.9–96.8) 787 (36.4)

Black 4.9 (2.1–10.9) 95.1 (89.0–97.9) 192 (8.8)

Brown 8.9 (6.9–11.3) 91.1 (88.6–93.1) 1,089 (54.5)

Mother’s level of schooling (years) 0.117

More than 11 5.6 (3.5–9.0) 94.4 (90.1–96.5) 427 (21.3)

8 to 11 6.3 (5.1–7.9) 93.7 (92.1–94.9) 1,080 (51.2)

Less than 8 9.3 (5.7–14.8) 90.7 (85.2–94.2) 568 (27.5)

Contextual variables

Health district (group) 0.337

I 4.7 (2.0–10.5) 95.3 (89.5–97.9) 425 (18.4)

II 6.6 (4.5–9.5) 93.4 (90.5–95.5) 745 (37.9)

III 8.3 (5.7–12.0) 91.7 (87.9–94.3) 905 (43.7)

School toothbrushing program 0.640

Yes 6.4 (4.0–10.1) 93.6 (89.8–96.0) 632 (30.5)

No 7.3 (5.2–10.2) 92.7 (89.8–94.8) 1,443 (69.50%)

Type of school 0.039∗

Private 4.6 (2.9–7.2) 95.4 (92.8–97.1) 604 (28.8)

Public 8.0 (5.9–10.7) 92.0 (89.3–94.0) 1471 (71.2)

Total 7.0 (5.3–9.2) 93.0 (90.7–94.6)∗

Rao-Scott test. P < 0.05.95% CI: Confidence Interval at 95%.

to take part in the survey, 2,075 agreed and were examined(response rate = 70.0%).

Sample was composed mainly of males (n = 1, 053,50.9%), those who classified themselves as brown (n =1, 089, 54.5%), and whose mothers had studied from eightto eleven years (n = 1, 080, 51.2%). The majority of thestudents were from public schools (n = 1, 471, 71.2%).

Table 1 presents the frequency distribution of the studiedvariables and the results of bivariate associations between thedependent variable and the independent variables.

Adverse periodontal conditions, such as presence ofcalculus and/or bleeding, were present in 140 schoolchildren,a prevalence of 7% (95% CI = 5.3–9.2). A total of 80individuals (4.2%, 95% CI = 2.5–6.9) had bleeding and 85had calculus (4.1%, 95% CI = 2.9–5.7).

In regard to individual variables, only color/race showedstatistically significant association with periodontal condi-tion. Brown individuals had a higher prevalence of dentalcalculus and/or bleeding compared with the others (P =0.005).

Among the contextual variables analyzed, type of schoolshowed significant association with periodontal condition.Public schools, compared to private ones, had a higherprevalence of dental calculus and/or bleeding (P < 0.05).

Table 2 shows the results of unadjusted multilevelanalysis. Children from schools located in health districtsof group II presented a 36.9% higher probability of havingadverse periodontal conditions compared to those of group I(P < 0.001). Those from public schools showed a prevalenceof calculus and bleeding 1.54 (95% CI = 1.14 − −2.08)times higher than the children from private schools (P =0.004). None of the other contextual factors was significantlyassociated with the dependent variable.

At the individual level, schoolchildren who classifiedthemselves as brown showed a prevalence of adverse peri-odontal condition 1.68 times higher than those of white color(P < 0.05). Sex and mother’s level of schooling did notpresent association with the dependent variable.

4. Discussion

The occurrence of adverse periodontal conditions (bleedingand/or calculus) in our study was low (7.0%) compared toother Brazilian studies [7, 12] and lower than that foundin schoolchildren of the same age in the city of Goiania in2003 [15]. This may indicate an improvement on periodontalhealth for this age-group. The low levels of periodontalconditions generally found in Brazilian schoolchildren may


Table 2: Results of nonadjusted multilevel analysis model of poisson log-linear regression for prevalence of periodontal condition (calculusand/or bleeding). Goiania, Brazil, 2010.

Regression coefficient Standard error Prevalence ratio Confidence interval 95% P

Contextual variables

District I — — 1.000 — — —

District II 0.3140 0.0840 1.369 1.161 1.614 0.000∗

District III 0.4180 0.2850 1.519 0.869 2.655 0.143

Private school — — 1.000 — — —

Public school 0.4320 0.1540 1.540 1.139 2.083 0.004∗

School toothbrushing program

Yes — — 1.000 — — —

No 0.1260 0.1780 1.134 0.800 1.608 0.477

Individual variables

Female — — 1.000 — — —

Male 0.2660 0.1490 1.305 0.974 1.747 0.074

White — — 1.000 — — —

Black −0.0260 0.3550 0.974 0.486 1.954 0.943

Brown 0.5180 0.2560 1.679 1.016 2.773 0.043∗

Mother’s schooling (years)

More than 11 — — 1.000 — — —

8 to 11 0.1370 0.2340 1.147 0.725 1.814 0.558

Less than 8 0.3900 0.3770 1.477 0.705 3.092 0.301∗P < 0.05.

be a result of the population’s high awareness regardingpersonal hygiene, which is part of the national culture, and itis frequently associated with health and well being. The 2009National School-Based Health Survey found that 95.2% ofthe adolescents reported toothbrushing frequency of twice aday or more, being higher for females [24].

However, our findings show that periodontal conditionis associated with individual and contextual variables suchas color/race, type of school and its location in the city, andconfirm the persistent inequalities in the population’s oralhealth. Higher prevalence of calculus and/or bleeding wasfound in those from lower socioeconomic groups.

This study is one of the first to apply multilevel analysis toa set of data representative of adolescents attending schoolsin Brazil. This kind of analysis allows researchers to dealwith the micro-level of individuals and the macro-level ofgroups simultaneously. However, issues of defining relevantcontexts, specifying the relevant group-level variables, andcollecting the necessary data remain a challenge [25] and thatwas also true in our case.

Thus, one limitation of the study was the small numberof individual and contextual variables analyzed due totime constraints in the data collection period. Nevertheless,the individual and contextual factors studied here can beuseful for identifying vulnerable groups and consequentlycontribute to a more effective and focused planning for oralhealth interventions.

The results did not show a significant worse periodontalstatus for males than females, which was different from somestudies [12, 26] but in accordance with another Brazilianstudy [27]. The difference in findings on sex might be due

to a greater concern on males’ general health and bodyimage in the whole of society over the last decade [28].This trend could be influencing male adolescents’ attitudestowards themselves.

The higher prevalence of adverse periodontal conditionin brown individuals was different from previous studiesthat have found higher prevalence in black individuals inthe state of Sao Paulo [12]. It is important to highlightthat there are nearly twice more brown individuals in thepopulation of Goias than in Sao Paulo [29], and that bothgroups account for lower family income, less years of studyand higher illiteracy nationwide than white individuals.Therefore, both black and brown individuals are consideredvulnerable groups in Brazil.

The protection of families and children’s developmentare crucial points of attention in the public policies. InBrazil, a higher proportion of families led by black or brownindividuals are couples with children younger than 14 yearsof age [29], stressing the importance of promoting economicand educational equality as a mean to improve families’health status as a whole.

Public schools, compared to private ones, showed higherprevalence of adverse periodontal condition. This finding hasalso been reported by other researchers in Brazil [12, 15,30, 31]. Similarly, studying in rural government schools inNepal was reported as an indicator of unhealthy periodontalstatus for adolescents [32]. The type of school used as analternative indicator for socioeconomic status is seen as afeasible predictor for caries experience in epidemiologicaldental caries studies involving schoolchildren in the Braziliancontext [31].


Children from schools located in health districts ofgroup II (intermediate socioeconomic indicators) presenteda higher probability of having periodontal condition com-pared to group I (best indicators) (P < 0.000). That couldbe partly explained by the fact that schools located in healthdistricts with the worst oral health indicators have greaternumber of students covered by health and oral health publicprograms, which should provide prevention, education, andoral treatment. This result also shows that schools locatedin health districts of intermediate socioeconomic indicatorsshould also be seen as a priority by local health authorities.

No significant difference of periodontal condition wasfound between schools with the School ToothbrushingProgram and those without them. Systematic reviews haveshown that health education interventions may result inreductions in plaque and gingival bleeding in the short-term, but it is yet unknown if these beneficial changes aresustained in the long run [33]. It is therefore recommendedthat oral health programs should focus on raising awarenessabout the issues that affect oral health and on promotingempowerment [34] so that individuals develop autonomyto make healthier choices and adopt healthier lifestyles. Thepresent study did not aim to evaluate the effectiveness ofthe toothbrushing program carried out in the schools, soother studies using appropriate methodologies are needed toinvestigate its impact on the population.

5. Conclusions

The prevalence of adverse periodontal conditions in 12-year-old schoolchildren was low and was associated withindividual and contextual variables. The inequalities in itsdistribution were mainly determined by the adolescents’color/race, type of school, and its location in the city. Appro-priate strategies addressed to the areas of socioeconomicdeprivation and the monitoring of school population oralhealth status are needed to reduce the disparities.

References

[1] L. J. Jin, G. C. Armitage, B. Klinge, N. P. Lang, M. Tonetti, andR. C. Williams, “Global oral health inequalities: task group—periodontal disease,” Advances in dental research, vol. 23, no. 2,pp. 221–226, 2011.

[2] P. E. Petersen and H. Ogawa, “Strengthening the preventionof periodontal disease: the WHO approach,” Journal ofPeriodontology, vol. 76, no. 12, pp. 2187–2193, 2005.

[3] A. Hugoson and O. Norderyd, “Has the prevalence ofperiodontitis changed during the last 30 years?” Journal ofClinical Periodontology, vol. 35, no. 8, pp. 338–345, 2008.

[4] R. Watt and A. Sheiham, “Inequalities in oral health: a reviewof the evidence and recommendations for action,” BritishDental Journal, vol. 187, no. 1, pp. 6–12, 1999.

[5] United States Department of Health and Human Services,Oral Health in America: A Report of the Surgeon General—Executive Summary, United States Department of Health andHuman Services, National Institute of Dental and CraniofacialResearch, National Institutes of Health, Rockville, Md, USA,2000.

[6] W. M. M. Jenkins and P. N. Papapanou, “Epidemiology ofperiodontal disease in children and adolescents,” Periodontol-ogy 2000, vol. 26, no. 1, pp. 16–32, 2001.

[7] N. C. N. Santos, T. D. B. Alves, and V. S. Freitas, “Oral healthamong adolescents: aspects relating to hygiene, dental cavitiesand periodontal disease in the cities of Recife and Feira deSantana, Brazil,” Ciencia & Saude Coletiva, vol. 12, no. 5, pp.1155–1166, 2007.

[8] F. Varas, G. Zillmann, A. Munoz et al., “Periodontal statusand treatment needs of children from 6 to 8 years old inthe Santiago Metropolitan Region of Chile,” Revista OdontoCiencia, vol. 26, no. 1, pp. 10–15, 2011.

[9] F. P. Ashley, L. A. Usiskin, R. F. Wilson, and E. Wagaiyu, “Therelationship between irregularity of the incisor teeth, plaque,and gingivitis: a study in a group of schoolchildren aged 11–14 years,” European Journal of Orthodontics, vol. 20, no. 1, pp.65–72, 1998.

[10] P. Kallio and H. Murtomaa, “Determinants of self-assessedgingival health among adolescents,” Acta Odontologica Scan-dinavica, vol. 55, no. 2, pp. 106–110, 1997.

[11] B. Nicolau, W. Marcenes, R. Hardy, and A. Sheiham, “A life-course approach to assess the relationship between social andpsychological circumstances and gingival status in adoles-cents,” Journal of Clinical Periodontology, vol. 30, no. 12, pp.1038–1045, 2003.

[12] J. L. Antunes, M. A. Peres, G. M. Jahn, and B. B. Levy, “The useof dental care facilities and oral health: a multilevel approachof schoolchildren in the Brazilian context,” Oral Health &Preventive Dentistry, vol. 4, no. 4, pp. 287–294, 2006.

[13] J. Ainamo, D. Barmes, G. Beagrie, T. Cutress, J. Martin, andJ. Sardo-Infirri, “Development of the World Health Organi-zation (WHO) community periodontal index of treatmentneeds (CPITN),” International Dental Journal, vol. 32, no. 3,pp. 281–291, 1982.

[14] H. C. Gesser, M. A. Peres, and W. Marcenes, “Gingivaland periodontal conditions associated with socioeconomicfactors,” Revista de Saude Publica, vol. 35, no. 3, pp. 289–293,2001.

[15] M. C. M. Freire, S. C. G. B. Reis, M. M. Goncalves, P. L.Balbo, and C. R. Leles, “Oral health in 12 year-old studentsfrom public and private schools in the city of Goiania, Brazil,”Revista Panamericana de Salud Publica, vol. 28, no. 2, pp. 86–91, 2010.

[16] L. N. Borrell and N. D. Crawford, “Socioeconomic positionindicators and periodontitis: examining the evidence,” Peri-odontology 2000, vol. 58, no. 1, pp. 69–83, 2012.

[17] P. E. Petersen, “The World Oral Health Report 2003: con-tinuous improvement of oral health in the 21st century—the approach of the WHO Global Oral Health Programme,”Community Dentistry and Oral Epidemiology, vol. 31, no. 1,pp. 3–24, 2003.

[18] J. Merlo, B. Chaix, M. Yang, J. Lynch, and L. Rastam,“A brief conceptual tutorial of multilevel analysis in socialepidemiology: linking the statistical concept of clustering tothe idea of contextual phenomenon,” Journal of Epidemiologyand Community Health, vol. 59, no. 6, pp. 443–449, 2005.

[19] J. L. F. Antunes, M. A. Peres, A. C. Frias, E. M. Crosato, andM. G. H. Biazevic, “Gingival health of adolescents and theutilization of dental services, state of Sao Paulo, Brazil,” Revistade Saude Publica, vol. 42, no. 2, pp. 191–199, 2008.

[20] A. G. Roncalli, N. N. da Silva, A. C. Nascimento et al., “Rele-vant methodological issues from the SBBrasil 2010 project fornational health surveys,” Cadernos de Saude Publica, vol. 28,pp. S40–S57, 2012.


[21] World Health Organization, Oral Health Surveys, Basic Meth-ods, World Health Organization, Geneva, Switzerland, 1997.

[22] J. N. K. Rao and A. J. Scott, “On Chi-squared tests formultiway contingency tables with cell proportions estimatedfrom survey data,” Annals of Statistics, vol. 12, pp. 46–60, 1984.

[23] H. Goldstein, Multilevel Statistical Models, London, Arnold,2003.

[24] M. V. Vettore, S. J. Moyses, L. M. V. Sardinha, and B. P. M. Iser,“Socioeconomic status, toothbrushing frequency, and health-related behaviors in adolescents: an analysis using the PeNSEdatabase,” Cadernos de Saude Publica, vol. 28, pp. s101–s113,2012.

[25] A. V. Diez-Roux, “Multilevel analysis in public healthresearch,” Annual Review of Public Health, vol. 21, pp. 171–192, 2000.

[26] B. Jenkins and P. Heasman, “The prevention and controlof periodontal disease,” in Prevention of Oral Disease, J. J.Murray, J. H. Nun, and J. G. Steele, Eds., pp. 125–144, OxfordUniversity Press, Oxford, UK, 4th edition, 2003.

[27] E. L. Dini, “Changes in periodontal conditions of children andadolescents from Araraquara, Brazil: 1995–1998,” BrazilianDental Journal, vol. 12, no. 1, pp. 51–55, 2001.

[28] S. Grogan and H. Richards, “Body image: focus groups withboys and men,” Men and Masculinities, vol. 4, no. 3, pp. 219–232, 2002.

[29] Instituto Brasileiro de Geografia e Estatıstica, Sıntese de Indi-cadores Sociais: uma analise das condicoes de vida da populacaobrasileira, 2010, no. 27, Estudos e Pesquisas, InformacaoDemografica e Socioeconomica, 2010.

[30] M. Maltz and B. B. Silva, “Relationship among caries,gingivitis and fluorosis and socioeconomic status of schoolchildren,” Revista de Saude Publica, vol. 35, no. 2, pp. 170–176,2001.

[31] C. Piovesan, M. C. Padua, T. M. Ardenghi, F. M. Mendes, andG. C. Bonini, “Can type of school be used as an alternativeindicator of socioeconomic status in dental caries studies? Across-sectional study,” BMC Medical Research Methodology,vol. 11, article 37, 2011.

[32] R. Yee, J. David, and D. Lama, “The periodontal health ofNepalese schoolchildren,” Community Dental Health, vol. 26,no. 4, pp. 250–256, 2009.

[33] R. G. Watt and V. C. Marinho, “Does oral health promotionimprove oral hygiene and gingival health?” Periodontology2000, vol. 37, pp. 35–47, 2005.

[34] A. R. C. Pauleto, M. C. T. Pereira, and E. G. Cyrino, “Oralhealth: a critical review about educative programmes forstudents,” Ciencia & Saude Coletiva, vol. 9, no. 1, pp. 121–130,2004.


Research Article

Prevalence and Reasons for Tooth Loss in a Sample from a DentalClinic in Brazil

Andreia Montandon,1 Elizangela Zuza,2 and Benedicto Egbert Toledo1, 2

1 Department of Social Dentistry, Sao Paulo State University (UNESP), Rua Humaita, 1680, 14801-903 Araraquara, SP, Brazil2 Department of Master of Dental Sciences, Educational Foundation of Barretos (UNIFEB), Avenida Prof. Roberto Frade Monte 389,14783-226 Barretos, SP, Brazil

Correspondence should be addressed to Elizangela Zuza, [email protected]

Received 20 April 2012; Accepted 24 July 2012


Copyright © 2012 Andreia Montandon et al. This is an open access article distributed under the Creative Commons AttributionLicense, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properlycited.

Purpose. To evaluate the prevalence and reasons for teeth extractions in a sample from a dental clinic in Brazil. Methods. Theprevalence of teeth mortality was analyzed by gender, age, tooth type and reasons for extraction on 800 teeth of 439 subjects,whose data was collected in clinical records in a convenience sample. Results. The groups with range from 35 to 44 years, 45 to 54years and 55 to 64 years revealed significantly greater number of teeth extractions than other age groups (P < 0.0001). The anteriorteeth loss increased significantly with aging, while the tooth mortality of premolar and molar were higher in younger people.The caries was the more prevalent reason for tooth mortality among young and adults up to 44 years old, while the periodontaldisease was the main reason for extractions from 45 years old until range of 81 years (P < 0.0001). Conclusions. It can be suggestedthat some reasons for tooth loss were age-dependent, but the caries and the periodontal diseases were the main reasons for toothmortality in this Brazilian sample.

1. Introduction

Studies concerning the epidemiology in dentistry haveshowed that dental caries and periodontal diseases arethe most prevalent pathologies that affect the oral cavity.Previous studies performed by American researchers hadsuggested that dental caries was the main reason for teethextraction, and other studies accomplished in New Zealand,Sweden, and even in Brazil confirmed that caries may leadto tooth mortality [1–5]. On the other hand, some studiessuggested that periodontal disease was the most prevalentreason that leads to tooth loss. Thus, controversial findingscould be explained by differences in the characteristics ofthe study population, immunological and genetic factors,cultural beliefs, and socioeconomic characteristics, amongothers. Immunological and genetic reasons are some of thecontributory factors that may explain why some populationsexposed to the same bacterial etiologic factors did notdevelop similar pathological conditions [6, 7].

Populations with poorer socioeconomic conditions haveshown higher prevalence and extent of teeth mortality, whichincreases with aging [8–10]. The prevalence, extent, and riskindicators for tooth loss were studied in a representativeBrazilian population, showing that 94% of the subjects hadexperienced tooth loss [8]. The authors verified a toothmortality mean of 11.2, ranging from 5.5 to 20.2 teeth insubjects from 30 to 39 years old and 60+ years age groups,respectively; gender, socioeconomic status, cigarette smok-ing, caries experience, and attachment loss were importantrisk indicators.

On the other hand, a decline in the tooth loss can beverified in developed countries in the last years [7, 11, 12],which can be explained by preventive programs and higheraccessibility to the oral health care that have been decreasingthe extractions [13, 14]. Brazil is a country in developmentprogress (underdeveloped), and some studies verified theprevalence of tooth extractions in Brazilians [4, 8–10, 15, 16],but few of them have verified the reasons for the tooth


mortality in this population. Thus, the aim of this studywas to describe the epidemiological data of the prevalenceand reasons for teeth extractions in a convenience sample ofBrazilians.


2.1. Sample Selection. This study was approved by theEthics Committee in Human of Araraquara Dental School(UNESP—process number 45/99). A review of 800 clinicalrecords was performed to select 439 patients with tooth lossfrom 1999 up to 2002 in Dental Clinics at State University ofSao Paulo (UNESP), Araraquara city, Brazil (Sudeste region).From the 800 clinical records, 361 subjects were excludedby X-ray absence or absence of specifications of the reasonsfor tooth mortality, and 439 patients were enrolled in thisstudy. From these 439 patients, 800 teeth with X-ray andspecifications of reasons for tooth loss were selected.

2.2. Data Collection. The clinical records were evaluatedconsidering the gender and the age groups. Additionally,the prevalence of tooth loss was evaluated by reasons ofextractions and type of teeth (molars, premolars, incisorsand canines). The incisors, and canines were nominatedonly as anterior teeth. The reasons for teeth mortalitywere adapted based on Cahen et al. [17] as follows: (1)caries: it is properly caries; (2) endodontic problems, suchas pulp inflammation, necrosis, or even tooth fracture;(3) periodontal diseases: tooth extraction by periodontalattachment loss; (4) eruption problems: tooth loss bydental impacation; (5) prosthetics: indication for extractionby prosthetic reasons; (6) trauma: extractions by externaltrauma; (7) orthodontics: tooth extraction by orthodonticindications; (8) occlusal problems: extractions by occlusaldysfunction, such as extrusion; (9) other reasons: any otherdetermined reasons, such as iatrogenic factors. A singleexaminer accomplished the clinical records assessment, anddata was described in a proper form.

2.3. Statistical Analysis. Descriptive data were expressed bymeans of numbers and percentages. Chi-square statistics (χ2)were applied, considering P < 0.05 as a statistical significantdifference. Adjusted residual chi-square (χ2) was calculated,considering 1.96 (α = 0.05) and 2.573 (α = 0.01).

3. Results

Table 1 shows the data collected from 1999 to 2002, thenumber of subjects, and teeth included in this study. Thenumber of subjects according to age groups and gender ratiocan be verified in Table 2, that shows a similar prevalenceof tooth loss between gender in all age groups (P = 0.67).Table 3 shows that the groups with range from 35 to 44years, 45 to 54 years, and 55 to 64 years revealed significantlygreater number of teeth extractions than other age groups(P < 0.0001). The anterior teeth loss increased with agingand the extractions of premolars and molars decreased withaging, while the tooth mortality of premolar and molar were

Table 1: Data collected for the study in different years.

Years Subjects(number)

Teeth(number)

Mean number of teethper subject

1999 138 274 2.0

2000 151 239 1.6

2001 127 235 1.9

2002 23 52 2.3

Total 439 800 1.8

Table 2: Number of subjects according to age groups and genderratio (n = 439).

Age group (years) Gender∗ (male/female) Total

15–19 15/16 31

20–24 12/15 27

25–29 13/10 23

30–34 21/17 38

35–44 34/24 58

45–54 45/32 77

55–64 55/39 94

65–74 28/31 59

75–84 14/18 32

P value 0.67 —∗

Chi-square test for categorical data in different age groups (P value >0.0.5does not indicate statistically significant difference).

higher in younger people (Table 4). The results showed thatcaries and periodontal diseases were statistically significantreasons for tooth loss when compared to the other factors(P < 0.0001; Table 5). Table 6 shows that caries was themost prevalent reason for tooth mortality among young andadults up to 44 years old, while the periodontal disease wasthe main reason for extractions from 45 years old until rangeof 81 years. Other reasons for teeth extractions can be verifiedas follows: iatrogenic factors (9.9), eruption problems (6.4),orthodontics (5.7), prosthetic indication (3.6), trauma (2.6),and occlusion problems (1.1). It can be seen in Table 6that tooth mortality for some reasons such as orthodontic,occlusal problems, and other reasons was significantly higher(P < 0.0001) among younger, and by prosthetic reasons wassignificantly higher in older people (P < 0.0001).

4. Discussion

Our data showed that the number of male and femalesubjects with tooth loss was similar among all age groups.Conversely, the study by Barbato et al. [10] verified higherprevalence of tooth mortality in women than in men. Otherstudies assessed the tooth mortality among the urban andrural adult population of Dharwad district (India) in 1223subjects (685 urban and 538 rural) and found that femalescompared to males had higher tooth loss [18]. Moreover, ourfindings showed that the age group between 35 to 44 years oldpresented a mean tooth loss of 2.3, which was lower than the


Table 3: Number and percentage (%) of teeth extractions by agegroup.

Age group Numberof teeth

% of teethextracted

Mean number of

teeth extracted

15–19 64 8.0 2.1

20–24 51 6.4 1.9

25–29 24 3.0 1.0

30–34 45 5.6 1.2

35–44∗ 131 16.4 2.3

45–54∗ 163 20.4 2.2

55–64∗ 180 22.5 1.9

65–74 105 13.1 1.8

75–84 37 4.6 1.2

15–81 800 100.0 1.8

Chi-square test (χ2) = P < 0.0001. ∗ > 2.576 (α = 0.01) adjusted residualchi-square.

Table 4: Prevalence (%) of extractions by age group and type ofteeth.

Age groupTooth

Anterior Premolar Molar Total

N (%) N (%) N (%) N

15–19 — 33 (51.6)∗ 31 (48.4)∗ 64

20–24 3 (5.9) 22 (43.1) 26 (51.0)∗ 51

25–29 5 (20.8) 8 (33.3) 11 (45.9) 24

30–34 11 (24.4) 14 (31.1) 20 (44.5) 45

35–44 33 (25.2) 42 (32.1) 56 (42.7)∗ 131

45–54 62 (38.0) 54 (33.2) 47 (28.8) 163

55–64 84 (46.7)∗ 52 (28.9) 44 (24.4) 180

65–74 69 (65.7)∗ 25 (23.8) 11 (10.5) 105

75–84 25 (67.6)∗ 7 (18.9) 5 (13.5) 37

15–81 292 (36.5) 257 (32.1) 251 (31.4) 800


mean of 11 verified in the other study [10]. This differencecould be explained by the sample size of the studies, whichwas of 439 Brazilians in our sample and a major populationof 13.431 Brazilians in the study by Barbato et al. [10].

Our findings showed that among adolescents aged from15 to 19 years, the prevalence of tooth extraction was 8.0%,while in a greater population of 16.833 Brazilian adolescentsit was 38.9% [19]. Probably this difference could also berelated to the sample size of the studies that is too discrepant.The results in the present study demonstrated higher missingteeth for premolars (43.1 to 51.6%) and molars (48.4 to51.0%) among the youngest subjects (15 to 24 years), andthe dental caries was the main reason for teeth extractions(67.2%). Our findings are in agreement with previous studiesthat have demonstrated that the first molars were the mostprevalent missing teeth in Brazilian younger people [9, 15,19]. Barbato and Peres [19] found more than 55% of missing

Table 5: Reasons for teeth extractions (N : number; % percentage).

ReasonsFrequency

N %

Caries 307∗ 38.4

Periodontal diseases 258∗ 32.3

Eruption problems 51 6.4

Prosthetics 29 3.6

Trauma 21 2.6

Orthodontics 46 5.7

Occlusal problems 9 1.1

Others reasons—iatrogenic and accidental 79 9.9

Total 800 100.0


first molars, and dental caries experiences were found in92.71% of all teeth lost. Other studies also suggested thatthe first molars were the most frequently missing teeth insubjects aged from 14 to 29 [9] and from 13 to 16 years [15].

The caries (38.4%) and periodontal disease (32.3%)were the most prevalent conditions for tooth mortality inthe present study with similar prevalence. Conversely, otherstudy on Brazilians suggested that the main reason for toothextraction was the caries (70.3%) and not the periodontaldisease (15.1%), but this study showed findings of a smallsample with 404 teeth extracted [16]. In a Caribbeanpopulation—Antigua, Viagnarajah [20] also suggested thatcaries was the main reason for tooth extraction (61.6%), andthe periodontal disease was the next more frequent reasonfor tooth loss (29.9%).

Our results showed that periodontal disease was moreprevalent among older subjects over 40 years of age. Somestudies are in agreement with our findings, showing thatperiodontitis became the most prevalent reason for toothmortality in people near 40 years old [10, 20–26]. In addi-tion, other reasons for teeth extractions were found in thepresent study as follows: iatrogenic factors (9.9%), eruptionproblems (6.4%), orthodontics (5.7%), prosthetic indication(3.6%), trauma (2.6%), and occlusal problems (1.1%).Caldas [16] also evaluated other reasons for extractions in404 teeth, such as prosthetic reasons (6.4%), third molar-eruption disorders (3.7%), orthodontic reasons (2.5%), andtrauma or patient′s request (1%), which showed lowerprevalence than the caries and periodontal disease.

Considering that few studies evaluated reasons for teethlosses in Brazilians, our retrospective study may be useful inclinical dentistry and used as baseline data regarding preva-lence and reasons for teeth extractions in other populations;moreover, other prospective studies might be performed.

5. Conclusions

Within the limits of this study, it can be suggested that somereasons for tooth loss were age dependent, but the cariesand the periodontal diseases were the main reasons for toothmortality in this Brazilian sample.


Table 6: Reasons for tooth extraction by age groups.

Age groupReasons

C PD E P T O OP OthersTotal

N (%) N (%) N (%) N (%) N (%) N (%) N (%) N (%)

15–1943 — 18 — — — — 3

64(67.2)∗ (28.1)∗ (4.7)

20–2418 — 24 — — 4 — 5

51(35.3) (47.1)∗ (7.8) (9.8)

25–2912 — 5 — — — — 7

24(50.0) (20.8)∗ (29.2)∗

30–3413 6 2 — 2 8 4 10

45(28.9) (13.3) (4.4) (4.4) (17.8)∗ (8.9)∗ (22.2)∗

35–4473 34 — — 2 — 2 20

131(55.7)∗ (26.0) (1.5) (1.5) (15.3)∗

45–5457 63 — 7 8 10 — 18

163(35.0) (38.7)∗ (4.3) (4.9) (6.1) (11.0)

55–6456 83 2 8 6 13 3 9

180(31.1) (46.1)∗ (1.1) (4.5) (3.3) (7.2) (1.7) (5.0)

65–7423 54 — 10 3 11 — 4

105(21.9) (51.4)∗ (9.5)∗ (2.9) (10.5) (3.8)

75–8412 18 — 4 — — — 3

37(32.4) (48.7)∗ (10.8) (8.1)

15–81 38.4 32.3 6.4 3.6 2.6 5.7 1.1 9.9 800

C: caries; PD: periodontal disease; E: eruption problems; P: prosthetics; T: trauma; O: orthodontics; OP: occlusal problems; Others: other reasons (accidentsor iatrogenic factors). Chi-square test (χ2) = P < 0.0001. ∗ > 2.576 (α = 0.01) adjusted residual chi-square.

Conflict of Interests

The authors declare that they have no conflict of interests.

References

[1] B. D. Moen, “Bureau of Economic Research and Statistics:survey of needs for dental care. II. Dental needs accordingto age and sex of patients,” Journal of the American DentalAssociation, vol. 46, no. 2, pp. 200–211, 1953.

[2] L. J. Coxhead, “Reasons for extractions in general practice,”New Zealand Dental Journal, vol. 56, pp. 182–183, 1960.

[3] C. Lundqvist, “Tooth mortality in Sweden. A statistical surveyof tooth loss in the Swedish population,” Acta OdontologicaScandinavica, vol. 25, no. 3, pp. 298–322, 1967.

[4] F. A. Machado, H. A. Godoy, A. J. S. Terra, and C. Marzola,“Principais causas das exodontias no municıpio de CampoGrande, MT,” Arquivos do Centro de Estudos da Faculdade deOdontologia da Universidade Federal de Minas Gerais, vol. 10,pp. 129–151, 1973 (Portuguese).

[5] I. D. M. Macgregor, “The pattern of tooth loss in a selectedpopulation of Nigerians,” Archives of Oral Biology, vol. 17, no.11, pp. 1573–1582, 1972.

[6] M. C. Downer, “The improving dental health of UnitedKingdom adults and prospects for the future,” British DentalJournal, vol. 170, no. 4, pp. 154–158, 1991.

[7] S. E. Marcus, T. F. Drury, L. J. Brown, and G. R. Zion, “Toothretention and tooth loss in the permanent dentition of adults:United States, 1988–1991,” Journal of Dental Research, vol. 75,pp. 684–695, 1996.

[8] C. Susin, R. V. Oppermann, O. Haugejorden, and J. M.Albandar, “Tooth loss and associated risk indicators in an

adult urban population from south Brazil,” Acta OdontologicaScandinavica, vol. 63, no. 2, pp. 85–93, 2005.

[9] C. Susin, A. N. Haas, R. V. Opermann, and J. M. Albandar,“Tooth loss in a young population from south Brazil,” Journalof Public Health Dentistry, vol. 66, no. 2, pp. 110–115, 2006.

[10] P. R. Barbato, H. C. M. Nagano, F. N. Zanchet, A. F. Boing,and M. A. Peres, “Tooth loss and associated socioeconomic,demographic, and dental-care factors in Brazilian adults: ananalysis of the Brazilian Oral Health Survey, 2002-2003,”Cadernos de Saude Publica, vol. 23, no. 8, pp. 1803–1814, 2007.

[11] T. Hiidenkari, T. Parvinen, and H. Helenius, “Edentulousnessand its rehabilitation over a 10-year period in a Finnish urbanarea,” Community Dentistry and Oral Epidemiology, vol. 25, no.5, pp. 367–370, 1997.

[12] A. L. Suominen-Taipale, P. Alanen, H. Helenius, A. Nordblad,and A. Uutela, “Edentulism among Finnish adults of workingage, 1978–1997,” Community Dentistry and Oral Epidemiology,vol. 27, no. 5, pp. 353–365, 1999.

[13] H. Loe, “Ural hygiene in the prevention of caries andperiodontal disease,” International Dental Journal, vol. 50, no.3, pp. 129–139, 2000.

[14] O. Haugejorden and K. S. Klock, “Expectation of retainingnatural teeth for a lifetime and its predictors among Norwe-gian adults,” Community Dental Health, vol. 19, no. 2, pp. 73–78, 2002.

[15] P. Gjermo, M. I. Beldi, H. T. Bellini, and C. R. Martins,“Study of tooth loss in an adolescent Brazilian population,”Community Dentistry and Oral Epidemiology, vol. 11, no. 6,pp. 371–374, 1983.

[16] A. F. Caldas Jr., “Reasons for tooth extraction in a Brazilianpopulation,” International Dental Journal, vol. 50, no. 6, pp.267–273, 2000.


[17] P. M. Cahen, R. M. Frank, and J. C. Turlot, “A survey of thereasons for dental extractions in France,” Journal of DentalResearch, vol. 64, no. 8, pp. 1087–1093, 1985.

[18] R. Kalyanpur and K. V. Prasad, “Tooth mortality and pros-thetic treatment needs among the urban and rural adult pop-ulation of Dharwad district, India,” Oral Health & PreventiveDentistry, vol. 9, pp. 323–328, 2011.

[19] P. R. Barbato and M. A. Peres, “Tooth loss and associatedfactors in adolescents: a Brazilian population-based oral healthsurvey,” Revista de Saude Publica, vol. 43, no. 1, pp. 13–25,2009.

[20] S. Vignarajah, “Various reasons for permanent tooth extrac-tions in a Caribbean population—Antigua,” InternationalDental Journal, vol. 43, no. 3, pp. 207–212, 1993.

[21] C. M. Belting, M. Massler, and I. Schour, “Prevalence andincidence of alveolar bone disease in men,” Journal of AmericanDental Association, vol. 47, pp. 190–197, 1953.

[22] W. A. Bossert and H. H. Marks, “Prevalence and characteris-tics of periodontal disease in 12, 800 persons under periodicdental observation,” Journal of American Dental Association,vol. 52, pp. 429–442, 1956.

[23] R. C. Oliver and L. J. Brown, “Periodontal diseases and toothloss,” Periodontology 2000, vol. 2, pp. 117–127, 1993.

[24] E. Reich and K. A. Hiller, “Reasons for tooth extraction in thewestern states of Germany,” Community Dentistry and OralEpidemiology, vol. 21, no. 6, pp. 379–383, 1993.

[25] G. Ong, “Periodontal reasons for tooth loss in an Asianpopulation,” Journal of Clinical Periodontology, vol. 23, no. 4,pp. 307–309, 1996.

[26] L. A. Chambrone and L. Chambrone, “Tooth loss in well-maintained patients with chronic periodontitis during long-term supportive therapy in Brazil,” Journal of Clinical Peri-odontology, vol. 33, no. 10, pp. 759–764, 2006.

Documents

Epidemiology of Periodontal DiseaseCriteria, Aline Mendes Silva de Pinho, Carolina Marques Borges, Mauro Henrique Nogueira Guimara es de Abreu, Eﬁgˆenia Ferreira e Ferreira, and