Relationship of microbial and salivary parameters with dental caries in Brazilian pre-school children
Waller A. Bretr\ Carlos Djahjah', Rui S. Almeida^ Philippe P. Hujoel^ and Walter J. Loesche' University of Michigan. Schooi of Dentistry, Ann Arbor. Ml. ' Llniversity of Washingfon. Schooi of Public Heaith. Seattie. WA; '^Ambulaf6rio Praia do Pinto. Rio de Janeiro. Brazii
Bretz WA, Diuh.jah C, Almeida RS, Htijocl PP, Locschc W.I: Relationship of microbial and salivary parameters with dental caries in Brazilian pre-school children. Cciinmunily Dent Otal Epidetniol 1992; 20: 261 4. Abstraet Caries e.xamiiiatioii and eollcetioii of paraffiti wax-stitnulated saliva samples were performed in }1 ehildren. ?< 6 years old, in a child-care facility at the Vidigal shim, Rio de ,laneii(.i, Btazil. The levels of mutatis stteptocoeei and lactobacilli in saliva were estimated by the Catieseteen and by the Dentoeult tests and the saliva secretion rate was determined. Statistical analysis was perfonned on surfacebased and patient-based caries prevalence rates (SBCPR and PIJCPR), and related to baeterial and salivary parameters, l^he tesults show that 31 ofthe .^7 ehildren were caries active. The SBCPR for the primary dentition was 6.7%+ I.O"i>. Oeclusal surfaces were the most affeeted by deeay. Regression analysis revealed that mutans streptococci .salivary levels were significantly associated with the SBCPR (7^ = 0.0001). Similarly, lactobacilli salivary levels were sigtiificantly associated with the SBCPR (/' = 0.0001). No significant association could be found belween the saliva secretion rate and the SBCPR. When regression analysis was used to model dependence of the SBCPR on both organistns, the tnutaus streptococci and laclobacilli salivaty levels were significantly associated with the SBCPRs (/' = 0.0021 and 0.01 18. te.spectively), and salivary levels of these organisms aeeouiued for 57"/o ofthe SBCPR variability. These Hnditigs indicate that the levels of muUius streptococci and lactobacilli in saliva are signincantly related to the SBCPRs on the primary dentition oi' these children.
In developing countries the occutrenee of dental decay is tiot deelitiing. eontraiy to wlial has been observed in developed countries (1). This is well demonstrated in Brazil, where a recent national survey has indicated that dental caries and periodontal diseases are prevalent in high levels throtighout the sociocconotnic strata (2). In a separate survey (3) in the State of Rio de .laneiro. approxitnately 6000 schoolchildren in the age range of 6-18 yv were randomly chosen, and Uie number of decayed, tnissing and filled teeth (DMFT) detemiincd. The lesult.s indicated thai a ehild at the age of 12 had on average a DMFT of 9.2, of which about 5yV» was dtie lo decay, .suggesting that the delivery of dental eare to this segment of the po|itilation has been inadequate. The delivery of dental tieatment and of preventive programs in developing countries has been hampered by econotnic coiistiaitits. Heallh care providers are thus forced to eonsider eosl-
Key words: dentai caries: lacfobacilii: mutans streptococci: pre-school chiidren: saliva secreflon rate Waiter A, Brelz. tjniversity of Michigan, Schooi of Dentistry, Department of Cariology and General Dentistry. Ann Arbor. Mi. 48109, tJSA Accepfed for publication 9 January 1992
benefil and efficiency in the de\elopnient cay, although they should not be eotisidof dental preventive and treatment pro- ered absolute ri.sk factors. grams. It would be preferable to identify There is compelling evidence in the litsegtnents of the poptilatioti at an early erature suggesting a strong association age who ate at high or low risk for devel- between the presence and the levels of oping dental decay, so thai adequate pre- mutans streptococci and/or lactobacilli vetitive strategies ean be itnpletiiented at with dental decay on the primary dentia reasonable eost. The need to seatch tioti of pre-school children in developed lor risk indicators for dental earies and eounlries {\5 18). but little information periodontal disease.s in order lo target is ax'ailable about similar studies on ehilri,sk groups has been one ofthe subjects ilreii of developing countries. The purof a veeenl eonferenee on the assessment pose of this study was to seek correlaof risk in dentistry (4). tions of .salivary mierobiologic paraThe role ofthe tnulans slieptococci in meter.s (mutatis streptococci and human dental deeay, and to a lesser extent laetobaeilli) and salivary seeretion rate of lactobaeilli. has been well documented with the occurrence of dental decay on (5 S), A nutnber of rapid and simple mi- the primary dentitioti of Brazilian preerobiologieal-based tests for tbe enumera- sehool ehildren who reside in a welltion of iiuUans stteptocoeei and laetoba- known slum area. eilli are now available (9-14). These assays could be used lo discriminate highly infected subjeets from non-infeeted sub- Material and method jects, atid therefore may be indicators of Study population The satiiple for this an increased or reduced risk for dental de- study consisted of 37 ehildren. ."^-6 \r old
262
BRHTZ I;T AI..
(19 boys and 18 girls), enrolled in a childcare facility at Vidigal Slum, Rio de Janeiro, Brazil. These children usually spend weekdays at the facility. There is no fluoride in the water supply of this slum, and the ehildren had no prior history of dental treatment. Thus, it was possible to study the natural occurrenee of dental deeay in this group of ehildren. The children were of a low socioeconomic status, and white, black and mulatto boys and girls constituted the study group. Exatnination procedures Detnographie data (age, sex, raee) were obtained from these children. The date of birth was obtained from the birth eertificate attached to the children's enrolment files. Prior to the cotnmencement of the study, ihe examiner (CD.) utidcrwcnt a 2-day ealibration exercise according to WHO dental caries diagnostic criteria (19). All ehildren were given a standardized visual-tactile examination for dental caries. Dental explorers, plane mirrors, artificial light, and a dental chair were used to perform the dental examinations. No radiographs were taken. Microbial and salivary analysis - paraffin wax-stimulated saliva satnples were collected frotn all ehildren for 5 min, and the amount of stiinulated whole saliva secreted per minute was cotnputed. Two ml of saliva were used to estimate the levels of mutans streptococci and lactobacilli by the Cariescreen (APO Diagnostics Inc. Toronto, Canada) and the Dentoeult (Orion Diagnostiea, Espoo, Finland) dip-slide tests, respeetively. After incubation for 2 days at 37 C, the colony density, i.e., the number of eolony-forming units (CFU)/ml, on the dip-slide was determined by individuals who were unaware of the dental examinations. The colony density on the dip-slide was
scored under a dissecling microscope with the aid of a seoring chart provided by the tnanufaclurers. The range of scores appearing on the chart for mutans streptococci (IO'' to 10'' CFU/ml) and for laetobaeilli (lOMo 10'' CFU/ml) do not reflect the number of CFU/ml of saliva obtained by conventional plating techniques, but rather the number of CFU/ dip-slide. However, sludies have shown that there is a strong linear association between salivary mutans streptococci and laetobaeilli recovery by conventional plating techniques and colony density on the dip-slide (13, 14, 20). Statistical anatysis - Dental caries prevalenee was estimated both at a palienl level and at a surface level. Prevalenee of earies al a patient level was estimated as the petcentage of patients exhibiting at least one carious lesion (PBCPR-patient-based caries prevalence rate). Prevalence of caries at a surface level was estiinated as the percentage of surfaces exhibiting caries of all surfaces at risk (SBCPR-surface-based caries prevalence rate). The SBCPR was estiS xi/ni , ., . , muted as p = , where N is the patient sample size, ni is the number of surfaces al risk within patient i, and xi is the number of carious lesions in patient i. Since ni was constant from patient to patient, it was not neeessary to weigli this estimator by the number of surfaces at risk. The average +s.e. rates per subject were calculated by age, gender, race, tooth surface type, tnutans slreptocoeei and lactobacilli levels. Since L xi/ni had a skewed distribution, the Freeman-Tukey binomial arc-sine transformation was employed for studying the associations between SBCPRs and bacterial and salivary parameters. The SBCPRs wete related to bacterial and salivary parameters
by tncaiis of regression analysis, analysis of variance and co-vari techniques. The appropriateness of h regression tnodels was investigated using residual plots and regression diagnostics
Results
Results itidicaled that 31 out of 37 ] dren had at least one earious surfaee oi Ihcir primary dentition (PBCP|< ^ M.Q'Va). Tliete were no particular atje gender or race trends in Brazilian pre.school children of a low socioeconomi ^ status when the SBCPRs were analyzed The average SBCPR for all children was 6.7'^± 1%, indicating that these children had on average 6.7'^ of their surface*; decayed. There was a certain degree or clustering of caries occurrence among the teeth within children (rho = 0.()52). Analysis of surface caries attack itidi~ cated that occlusal surfaces were th^ most affected surfaces by dental caries (Table I). On average, 36% of the occlul sal surfaces were found lo be decayed At the most, only 4"-2.5xlO' CFU/dip-slide), and that about 30"/i of the children had le)\v (o tnoderate levels of this organism (io-* to 10^ CFU/dip-slide), SBCPRs i,,.. ereased with increasing levels of mutatis streptococci. Moreover, the adjusted odds ratio indicated that surface caries increased 1.6 times with an increase of;) log unit of mutans stteptocoeei salivary levels. Saliva samples with moderate to high laetobaeilli levels (> 10' CFU/dip..
T;ihlc 1, Dental caries in Brazilian pre-school children by surface type No. of children
37
10'
5x10'
10'
2.5x10'
5x10'
10'
(N.4)
(N.2)
(N.S)
IN.5)
(N.7]
(M,.H|
mutans streptococci CFU/Dlp-Sllde
Fif>. I. Mutans streptococci salivary levels iuitl surface-based curies prevalence rule (.SBCPR) in Brazilian pre-school children.
Surface type
No. of surfaces
.SBCPR' niciin ±s.c.
Approximal mesial distal
740 740
3.5% ±0.9% 3.9% ±0.9%
Smooth facial lingual
740 740
4,0% ±0.9%
I-issure occlusal
256
.36..S%±3..3%
2.f-i%±0.8%
" SBCPR-surface-based caries prevalence rate (rates shown as percentages).
Risk indicators of dentat caries in Brazilian children
ND (N.7I
10' IN.13)
10' (N.^1
10' (N.7)
10' (N-6)
Lactobacilli CFU/Dlp-Sllde ND .• nol delecled
Fii^. 2. L;ictob:icilli salivary levels and surfacebased caries prevalence rale (SBCPR) in l$razilian pre-school children.
laetobaeilli salivary levels were significantly associated with the SBCPRs (P = 0.002\ and 0.0118. respectively). The effects of tnutans streptococci on the SBCPR tended to beapptoximately three limes higher than the lactobacilli effects, and this difference was marginally significant (/'= 0.0.^1). The coefficient of multiple determination showed that .'57% ofthe total variation present in the SBCPR can be explained by the salivary levels of both otganisms.
Discussion slide) were observed in abotit 46'/
Waller A. Bretr\ Carlos Djahjah', Rui S. Almeida^ Philippe P. Hujoel^ and Walter J. Loesche' University of Michigan. Schooi of Dentistry, Ann Arbor. Ml. ' Llniversity of Washingfon. Schooi of Public Heaith. Seattie. WA; '^Ambulaf6rio Praia do Pinto. Rio de Janeiro. Brazii
Bretz WA, Diuh.jah C, Almeida RS, Htijocl PP, Locschc W.I: Relationship of microbial and salivary parameters with dental caries in Brazilian pre-school children. Cciinmunily Dent Otal Epidetniol 1992; 20: 261 4. Abstraet Caries e.xamiiiatioii and eollcetioii of paraffiti wax-stitnulated saliva samples were performed in }1 ehildren. ?< 6 years old, in a child-care facility at the Vidigal shim, Rio de ,laneii(.i, Btazil. The levels of mutatis stteptocoeei and lactobacilli in saliva were estimated by the Catieseteen and by the Dentoeult tests and the saliva secretion rate was determined. Statistical analysis was perfonned on surfacebased and patient-based caries prevalence rates (SBCPR and PIJCPR), and related to baeterial and salivary parameters, l^he tesults show that 31 ofthe .^7 ehildren were caries active. The SBCPR for the primary dentition was 6.7%+ I.O"i>. Oeclusal surfaces were the most affeeted by deeay. Regression analysis revealed that mutans streptococci .salivary levels were significantly associated with the SBCPR (7^ = 0.0001). Similarly, lactobacilli salivary levels were sigtiificantly associated with the SBCPR (/' = 0.0001). No significant association could be found belween the saliva secretion rate and the SBCPR. When regression analysis was used to model dependence of the SBCPR on both organistns, the tnutaus streptococci and laclobacilli salivaty levels were significantly associated with the SBCPRs (/' = 0.0021 and 0.01 18. te.spectively), and salivary levels of these organisms aeeouiued for 57"/o ofthe SBCPR variability. These Hnditigs indicate that the levels of muUius streptococci and lactobacilli in saliva are signincantly related to the SBCPRs on the primary dentition oi' these children.
In developing countries the occutrenee of dental decay is tiot deelitiing. eontraiy to wlial has been observed in developed countries (1). This is well demonstrated in Brazil, where a recent national survey has indicated that dental caries and periodontal diseases are prevalent in high levels throtighout the sociocconotnic strata (2). In a separate survey (3) in the State of Rio de .laneiro. approxitnately 6000 schoolchildren in the age range of 6-18 yv were randomly chosen, and Uie number of decayed, tnissing and filled teeth (DMFT) detemiincd. The lesult.s indicated thai a ehild at the age of 12 had on average a DMFT of 9.2, of which about 5yV» was dtie lo decay, .suggesting that the delivery of dental eare to this segment of the po|itilation has been inadequate. The delivery of dental tieatment and of preventive programs in developing countries has been hampered by econotnic coiistiaitits. Heallh care providers are thus forced to eonsider eosl-
Key words: dentai caries: lacfobacilii: mutans streptococci: pre-school chiidren: saliva secreflon rate Waiter A, Brelz. tjniversity of Michigan, Schooi of Dentistry, Department of Cariology and General Dentistry. Ann Arbor. Mi. 48109, tJSA Accepfed for publication 9 January 1992
benefil and efficiency in the de\elopnient cay, although they should not be eotisidof dental preventive and treatment pro- ered absolute ri.sk factors. grams. It would be preferable to identify There is compelling evidence in the litsegtnents of the poptilatioti at an early erature suggesting a strong association age who ate at high or low risk for devel- between the presence and the levels of oping dental decay, so thai adequate pre- mutans streptococci and/or lactobacilli vetitive strategies ean be itnpletiiented at with dental decay on the primary dentia reasonable eost. The need to seatch tioti of pre-school children in developed lor risk indicators for dental earies and eounlries {\5 18). but little information periodontal disease.s in order lo target is ax'ailable about similar studies on ehilri,sk groups has been one ofthe subjects ilreii of developing countries. The purof a veeenl eonferenee on the assessment pose of this study was to seek correlaof risk in dentistry (4). tions of .salivary mierobiologic paraThe role ofthe tnulans slieptococci in meter.s (mutatis streptococci and human dental deeay, and to a lesser extent laetobaeilli) and salivary seeretion rate of lactobaeilli. has been well documented with the occurrence of dental decay on (5 S), A nutnber of rapid and simple mi- the primary dentitioti of Brazilian preerobiologieal-based tests for tbe enumera- sehool ehildren who reside in a welltion of iiuUans stteptocoeei and laetoba- known slum area. eilli are now available (9-14). These assays could be used lo discriminate highly infected subjeets from non-infeeted sub- Material and method jects, atid therefore may be indicators of Study population The satiiple for this an increased or reduced risk for dental de- study consisted of 37 ehildren. ."^-6 \r old
262
BRHTZ I;T AI..
(19 boys and 18 girls), enrolled in a childcare facility at Vidigal Slum, Rio de Janeiro, Brazil. These children usually spend weekdays at the facility. There is no fluoride in the water supply of this slum, and the ehildren had no prior history of dental treatment. Thus, it was possible to study the natural occurrenee of dental deeay in this group of ehildren. The children were of a low socioeconomic status, and white, black and mulatto boys and girls constituted the study group. Exatnination procedures Detnographie data (age, sex, raee) were obtained from these children. The date of birth was obtained from the birth eertificate attached to the children's enrolment files. Prior to the cotnmencement of the study, ihe examiner (CD.) utidcrwcnt a 2-day ealibration exercise according to WHO dental caries diagnostic criteria (19). All ehildren were given a standardized visual-tactile examination for dental caries. Dental explorers, plane mirrors, artificial light, and a dental chair were used to perform the dental examinations. No radiographs were taken. Microbial and salivary analysis - paraffin wax-stimulated saliva satnples were collected frotn all ehildren for 5 min, and the amount of stiinulated whole saliva secreted per minute was cotnputed. Two ml of saliva were used to estimate the levels of mutans streptococci and lactobacilli by the Cariescreen (APO Diagnostics Inc. Toronto, Canada) and the Dentoeult (Orion Diagnostiea, Espoo, Finland) dip-slide tests, respeetively. After incubation for 2 days at 37 C, the colony density, i.e., the number of eolony-forming units (CFU)/ml, on the dip-slide was determined by individuals who were unaware of the dental examinations. The colony density on the dip-slide was
scored under a dissecling microscope with the aid of a seoring chart provided by the tnanufaclurers. The range of scores appearing on the chart for mutans streptococci (IO'' to 10'' CFU/ml) and for laetobaeilli (lOMo 10'' CFU/ml) do not reflect the number of CFU/ml of saliva obtained by conventional plating techniques, but rather the number of CFU/ dip-slide. However, sludies have shown that there is a strong linear association between salivary mutans streptococci and laetobaeilli recovery by conventional plating techniques and colony density on the dip-slide (13, 14, 20). Statistical anatysis - Dental caries prevalenee was estimated both at a palienl level and at a surface level. Prevalenee of earies al a patient level was estimated as the petcentage of patients exhibiting at least one carious lesion (PBCPR-patient-based caries prevalence rate). Prevalence of caries at a surface level was estiinated as the percentage of surfaces exhibiting caries of all surfaces at risk (SBCPR-surface-based caries prevalence rate). The SBCPR was estiS xi/ni , ., . , muted as p = , where N is the patient sample size, ni is the number of surfaces al risk within patient i, and xi is the number of carious lesions in patient i. Since ni was constant from patient to patient, it was not neeessary to weigli this estimator by the number of surfaces at risk. The average +s.e. rates per subject were calculated by age, gender, race, tooth surface type, tnutans slreptocoeei and lactobacilli levels. Since L xi/ni had a skewed distribution, the Freeman-Tukey binomial arc-sine transformation was employed for studying the associations between SBCPRs and bacterial and salivary parameters. The SBCPRs wete related to bacterial and salivary parameters
by tncaiis of regression analysis, analysis of variance and co-vari techniques. The appropriateness of h regression tnodels was investigated using residual plots and regression diagnostics
Results
Results itidicaled that 31 out of 37 ] dren had at least one earious surfaee oi Ihcir primary dentition (PBCP|< ^ M.Q'Va). Tliete were no particular atje gender or race trends in Brazilian pre.school children of a low socioeconomi ^ status when the SBCPRs were analyzed The average SBCPR for all children was 6.7'^± 1%, indicating that these children had on average 6.7'^ of their surface*; decayed. There was a certain degree or clustering of caries occurrence among the teeth within children (rho = 0.()52). Analysis of surface caries attack itidi~ cated that occlusal surfaces were th^ most affected surfaces by dental caries (Table I). On average, 36% of the occlul sal surfaces were found lo be decayed At the most, only 4"-2.5xlO' CFU/dip-slide), and that about 30"/i of the children had le)\v (o tnoderate levels of this organism (io-* to 10^ CFU/dip-slide), SBCPRs i,,.. ereased with increasing levels of mutatis streptococci. Moreover, the adjusted odds ratio indicated that surface caries increased 1.6 times with an increase of;) log unit of mutans stteptocoeei salivary levels. Saliva samples with moderate to high laetobaeilli levels (> 10' CFU/dip..
T;ihlc 1, Dental caries in Brazilian pre-school children by surface type No. of children
37
10'
5x10'
10'
2.5x10'
5x10'
10'
(N.4)
(N.2)
(N.S)
IN.5)
(N.7]
(M,.H|
mutans streptococci CFU/Dlp-Sllde
Fif>. I. Mutans streptococci salivary levels iuitl surface-based curies prevalence rule (.SBCPR) in Brazilian pre-school children.
Surface type
No. of surfaces
.SBCPR' niciin ±s.c.
Approximal mesial distal
740 740
3.5% ±0.9% 3.9% ±0.9%
Smooth facial lingual
740 740
4,0% ±0.9%
I-issure occlusal
256
.36..S%±3..3%
2.f-i%±0.8%
" SBCPR-surface-based caries prevalence rate (rates shown as percentages).
Risk indicators of dentat caries in Brazilian children
ND (N.7I
10' IN.13)
10' (N.^1
10' (N.7)
10' (N-6)
Lactobacilli CFU/Dlp-Sllde ND .• nol delecled
Fii^. 2. L;ictob:icilli salivary levels and surfacebased caries prevalence rale (SBCPR) in l$razilian pre-school children.
laetobaeilli salivary levels were significantly associated with the SBCPRs (P = 0.002\ and 0.0118. respectively). The effects of tnutans streptococci on the SBCPR tended to beapptoximately three limes higher than the lactobacilli effects, and this difference was marginally significant (/'= 0.0.^1). The coefficient of multiple determination showed that .'57% ofthe total variation present in the SBCPR can be explained by the salivary levels of both otganisms.
Discussion slide) were observed in abotit 46'/
