Original Paper Received: March 12, 2013 Accepted after revision: September 10, 2013 Published online: January 3, 2014
Caries Res 2014;48:147–153 DOI: 10.1159/000355614
Relationship between Plaque pH and Different Caries-Associated Variables in a Group of Adolescents with Varying Caries Prevalence E.M. Aranibar Quiroz a, b T. Alstad c G. Campus d, e D. Birkhed a P. Lingström a
Departments of a Cariology, b Endodontics, and c Prosthetic Dentistry, Institute of Odontology, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden; d Department of Surgery, Microsurgery and Medicine Sciences, Dental Section, University of Sassari, Sassari, and e WHO Collaborating Centre of Milan for Epidemiology and Community Dentistry, University of Milan, Milan, Italy
Abstract The pH response of the dental biofilm after a sugar challenge can be considered to mirror the acidogenic potential and thereby the caries risk of an individual. The aim of this crosssectional study was to evaluate the relationship between plaque pH and different caries variables in adolescents with varying caries prevalence. One hundred individuals, aged 14–15 years, were examined regarding different caries-related variables: (i) caries score (DSm, DSi, DSm + i, DTm), (ii) salivary secretion rate and buffer capacity, (iii) oral microflora of plaque and saliva, (iv) plaque amount, (v) plaque pH and (vi) dietary intake, oral hygiene habits and fluoride use. Plaque pH was assessed using the microtouch method before and after a 1-min mouthrinse with 10 ml 10% sucrose. Depending on the minimum pH, the participants were divided into three groups: low pH (≤5.3), medium pH (>5.3–6.3) and high pH (>6.3). Statistically significant differences between the three groups (p < 0.01) were found for initial caries (DSi) and combined manifest and initial caries (DSm + i). A statistically significant difference was also found in the log values for salivary lactobacilli (p = 0.02) within the three groups, and for the total number of bacteria in plaque (p = 0.04); for both variables, the low-pH group had the highest values. The only
© 2014 S. Karger AG, Basel 0008–6568/14/0482–0147$39.50/0 E-Mail [email protected] www.karger.com/cre
covariate significantly associated was the Cariogram score in the medium-pH group (p < 0.01) and the number of meals per day in the high-pH group (p = 0.02). To conclude, plaque pH measured by the microtouch method is a method that can be used for discriminating between individuals with varying caries prevalence. © 2014 S. Karger AG, Basel
Dental caries is prevalent in people of all ages worldwide [Petersen et al., 2005]. The assessment of caries activity is therefore an important step towards identifying subjects who are at risk of developing the disease. Due to the multifactorial complexity of dental caries, it is essential to have as much information as possible about the different variables directly or indirectly (primary and secondary factors) involved in the caries process [Fontana and Zero, 2006]. Apart from aetiological factors, a number of biological and environmental factors need to be taken into account [Selwitz et al., 2007]. The use of fluorides, especially in toothpaste, the level of oral hygiene and dietary habits are parameters that are included when assessing the caries risk, but they do not give any updated information about the current caries activity. It is therefore important to identify other parameters that mirror the complex interaction between all the factors underlying the disease. E. Miguel Aranibar Quiroz Department of Endodontics Institute of Odontology SE–405 30 Gothenburg (Sweden) E-Mail miguel.aranibar @ odontologi.gu.se
Downloaded by: Universitätsbibliothek Giessen 134.176.129.147 - 5/15/2015 2:49:59 PM
Key Words Adolescents · Caries risk · Cariogram · Diet · Plaque pH
148
Caries Res 2014;48:147–153 DOI: 10.1159/000355614
Materials and Methods Subjects and Study Design In Sweden, all individuals up to the age of 19 years have access to free, regular dental care. A total of 100 adolescents, aged 14–15 years at the start of the study, were asked to participate in this cross-sectional study. The individuals came to the clinic for their annual dental check-up at two of the Public Dental Clinics in the Gothenburg area in 2009. All the participants were healthy and without any systemic disease. Informed consent was obtained prior to the start of the study. The study was approved by the Ethics Committee at the University of Gothenburg (Ö 370–02). All data collection was carried out under regular conditions. Dental caries was identified by clinical and radiographic examination. Plaque pH was then measured, followed by collection of plaque and saliva samples and registration of the amount of plaque. Finally, the participants answered a questionnaire. Clinical Variables Caries registration was carried out according the criteria described by WHO [1997], including both initial and manifest lesions. Two bitewing radiographs were taken for assessment of enamel and dentine caries. Calculation of DSm (decayed surfaces, manifest caries), DSi (decayed surfaces, initial caries), DSm + i (decayed surfaces, both manifest and initial caries), and DTm (decayed teeth, manifest caries) was based on the data collected from the combined clinical and radiographical examination. A pooled plaque sample was collected with four sterile toothpicks, according to Kristoffersson and Bratthall [1982], from one interproximal area in both front and premolar regions on each side of the upper jaw. It was transferred to transport medium for estimation of the number of MS in relation to the total number of cultivatable microorganisms. A paraffin-stimulated whole saliva sample was collected during 3 min, after which the salivary secretion rate was calculated in millilitres per minute. One millilitre of saliva was then transferred to a transport medium for the determination of the number of MS and lactobacilli (LB). The buffer capacity was determined using a chairside kit (Dentobuff Strip®, Orion Diagnostica, Espoo, Finland); this test discriminates between low (yellow, ≤4.0), medium (green, 4.5–5.5) and high (blue, ≥6.0) buffer capacity. All samples were transported to the laboratory at the Department of Cariology on the same day. At the laboratory, the plaque samples were dispersed by sonication for 10 s and the saliva samples were shaken on a mechanical mixer for half a minute. After serial dilutions in 0.05 M phosphate buffer (pH 7.3), 25-μl portions of the plaque and saliva samples were plated in duplicate on MSB agar for the growth of MS. The MSB agar plates were incubated in candle jars at 37 ° C for 2 days and the blood agar plates in 95% H2 and 5% CO2 for 7 days. For the growth of LB, the saliva samples were plated on Rogosa SL agar and incubated aerobically at 37 ° C for 3 days. For the total growth of microorganisms in the plaque samples, blood agar was used. The number of colony-forming units of MS was counted on the MSB agar and identified by their characteristic colony morphology. All colony-forming units in the Rogosa SL agar were considered to be LB. The proportion of MS in plaque was calculated as the percentage of the total count of colony-forming units on blood agar. The amount of plaque was determined visually and without staining, according to Silness and Löe [1964]. This was carried out
Aranibar Quiroz/Alstad/Campus/ Birkhed/Lingström
Downloaded by: Universitätsbibliothek Giessen 134.176.129.147 - 5/15/2015 2:49:59 PM
A relationship between plaque pH and caries activity was observed by Stephan [1944] already in 1944. He reported that although caries-free individuals could acquire acidogenic plaque, the pH fall after sugar exposure was greater in caries-active than in caries-inactive subjects. These findings have been followed by more recent studies [Shimizu et al., 2008]. The acidogenicity of the dental biofilm is considered as a crucial factor in the caries process. Acid production in plaque affects the solubility of the underlying enamel/dentine surface and may consequently lead to the development of a caries lesion. Furthermore, a difference in plaque acidogenicity has been found when comparing individuals with normal and low salivary secretion rates [Lingström and Birkhed, 1993], as well as in relation to the number of mutans streptococci (MS) in the oral cavity [Aranibar Quiroz et al., 2003]. As the plaque pH graph is considered to summarise most of the factors involved in the caries process [Lingström et al., 1993], its individual pattern is therefore believed to indicate the current caries risk. Different information, including clinical, anamnestic, dietary and salivary tests, is collected when assessing caries risk; this is time-consuming and the data are often difficult to interpret. A chairside method that could be used to perform a quick screening of the actual caries risk is therefore desirable. It is important for a method of this kind to take into account as many parameters as possible involved in the caries process; it should also been able to provide reproducible results under standardized conditions. The Cariogram is a computer-based model that has been developed to describe and calculate the individual caries risk profile [Bratthall and Hänsel Petersson, 2005]. It has been found to be a suitable tool when used in dental practice as it allows a more objective data interpretation than the clinical examination [Hänsel Petersson and Bratthall, 2000; Hänsel Petersson et al., 2002; Sonbul et al., 2008; Al Mulla et al., 2009; Campus et al., 2009; Almosa et al., 2012; Campus et al., 2012]. However, the Cariogram is a theoretical model and it does not measure the biological activity in the ecological environment surrounding the tooth. The measurement of plaque pH may be a better alternative in the diagnosis of the caries risk. The hypothesis was that the in vivo plaque pH method has the potential to assess caries risk. As a first step in this direction, the aim of this study was to evaluate the relationship between plaque pH and different caries-associated variables in a group of adolescents with varying caries prevalence, and as a second step to compare the findings with the caries risk profile obtained with the Cariogram.
Plaque pH No specific instructions regarding oral hygiene or food intake were given prior to the test. Measurements were made at two approximal sites in the front and premolar regions on both sides in the upper jaw using an iridium microelectrode (Beetrode®, MEPH-1; W.P. Instruments, New Haven, Conn., USA), according to the method described in detail by Lingström et al. [1993]. The electrode was connected to an Orion SA 720 pH/ISE Meter (Orion Research, Boston, Mass., USA), equipped with a porous glass reference electrode (MERE 1; W.P. Instruments). A salt bridge was created in a 3 M KCl solution between the reference electrode and one of the subject’s fingers. The measurement was carried out before (0 min) and at five different time points (2, 5, 10, 15 and 20 min) after a mouth rinse with 10 ml of 10% sucrose for 1 min. The electrode was calibrated against standard buffers before and during the measurements. Questionnaire After the clinical examination and sampling, the participants answered a questionnaire consisting of 14 closed questions related to general health, dietary habits, oral hygiene and the use of fluorides. Cariogram The Cariogram is an interactive computer-based program used to illustrate the individual caries risk profile [Petersson et al., 1998]. Data from ten parameters related to caries disease are entered into the computer program in order to assess the current caries risk. A score is given according to a predetermined scale for each factor. The result is expressed as a pie chart, in which the caries-associated factors are shown in different sectors and ‘the chance of avoiding caries’ is presented as a figure from 0 to 100%, where 0% (high risk) means that caries lesions will definitely develop over time [Petersson and Bratthall, 2000]. Statistical Analyses The mean plaque pH from the four sites for each individual was determined, after which the mean values for plaque pH, minimum-pH and maximum-pH decrease were calculated for each curve. The area under the curve at pH 5.7 (AUC5.7) and at pH 6.2 (AUC6.2) was calculated using a computer program [Larsen and Pearce, 1997]. The individuals were arranged in three groups based on their minimum-pH values: (1) low (≤5.3), (2) medium (>5.3– 6.3) and (3) high (>6.3). The selection of these three values was based on the fact that the mean value of the minimum pH ±1 standard deviation (5.8 ± 0.5) for the whole population included up to 70% of the participants (medium-pH group, n = 72), leaving the outermost groups on each side (low-pH group, n = 15 and high-pH group, n = 13). The difference between the clinical variables in the three pH groups was tested with one-way ANOVA. The microbial data from saliva and plaque were transformed into logarithmic values. Correlation coefficients between plaque pH variables and the different clinical caries variables were assessed. The sugar intake frequency was regarded as ‘low’, when the subject reported a consumption of a maximum intake of twice a week, ‘medium’
Plaque pH and Caries in Adolescents
when more than twice a week but not daily, and ‘high’ when daily consumption of sugar was reported. The number of initial (DSi) and manifest and initial (DSm + i) caries lesions was grouped as follows: DSi, DSm + i = 0, DSi, DSm + i 1–3, DSi, DSm + i ≥4. Cariogram scores were grouped into four different classes: 81–100 (n = 39), 61–80 (n = 22), 41–60 (n = 27) and 0–40% (n = 12) ‘chance of avoiding caries’. The association between the plaque pH groups and the Cariogram scores was assessed using a χ2 test. A non-parametric test for trend across ordered groups developed by Cuzick [1985] was performed. The reference group included individuals with the most favourable exposure level (i.e. 81–100% chance of avoiding caries; DSi = 0; DSm + i = 0 and low sugar frequency). A multiple linear regression model stratified for plaque-pH categorisation using initial decay surface (DSi) as the dependent variable was created. All the data were analysed using STATA® software (Stata Corp., College Station, Tex., USA, version 10.1 for Macintosh, http://www.stata. com); p values of 5.3–6.3
High (n = 13) min pH >6.3
mean ± SD range
mean ± SD range
mean ± SD range
mean ± SD range
Clinical variables DSm 0.9±1.7 DSi 3.2±4.5 DSm + i 4.1±5.4 DTm 0.8±1.6 Saliva and microflora Secretion rate, ml/min 2.2±0.9 Plaque index, % 19.5±14.2 MS saliva, log CFU/ml 4.2±1.8 LB saliva, log CFU/ml 3.6±1.3 MS plaque, log CFU/sample 1.9±1.8 Total count in plaque, log CFU/sample 6.9±0.5 Plaque pH Maximum pH decrease 0.9±0.4 Minimum pH 5.8±0.5 AUC5.7 1.2±2.7 AUC6.2 4.1±5.3 Cariogram Chance to avoid caries, % 67.1±21.3
0–9 0–25 0–29 0–7
1.1±1.8 7.0±7.8 8.1±9.1 1.1±1.7
15
pH
0.4
Caries Res 2014;48:147–153 DOI: 10.1159/000355614
Relationship between Plaque pH and Different Caries-Associated Variables in a Group of Adolescents with Varying Caries Prevalence E.M. Aranibar Quiroz a, b T. Alstad c G. Campus d, e D. Birkhed a P. Lingström a
Departments of a Cariology, b Endodontics, and c Prosthetic Dentistry, Institute of Odontology, Sahlgrenska Academy, University of Gothenburg, Gothenburg, Sweden; d Department of Surgery, Microsurgery and Medicine Sciences, Dental Section, University of Sassari, Sassari, and e WHO Collaborating Centre of Milan for Epidemiology and Community Dentistry, University of Milan, Milan, Italy
Abstract The pH response of the dental biofilm after a sugar challenge can be considered to mirror the acidogenic potential and thereby the caries risk of an individual. The aim of this crosssectional study was to evaluate the relationship between plaque pH and different caries variables in adolescents with varying caries prevalence. One hundred individuals, aged 14–15 years, were examined regarding different caries-related variables: (i) caries score (DSm, DSi, DSm + i, DTm), (ii) salivary secretion rate and buffer capacity, (iii) oral microflora of plaque and saliva, (iv) plaque amount, (v) plaque pH and (vi) dietary intake, oral hygiene habits and fluoride use. Plaque pH was assessed using the microtouch method before and after a 1-min mouthrinse with 10 ml 10% sucrose. Depending on the minimum pH, the participants were divided into three groups: low pH (≤5.3), medium pH (>5.3–6.3) and high pH (>6.3). Statistically significant differences between the three groups (p < 0.01) were found for initial caries (DSi) and combined manifest and initial caries (DSm + i). A statistically significant difference was also found in the log values for salivary lactobacilli (p = 0.02) within the three groups, and for the total number of bacteria in plaque (p = 0.04); for both variables, the low-pH group had the highest values. The only
© 2014 S. Karger AG, Basel 0008–6568/14/0482–0147$39.50/0 E-Mail [email protected] www.karger.com/cre
covariate significantly associated was the Cariogram score in the medium-pH group (p < 0.01) and the number of meals per day in the high-pH group (p = 0.02). To conclude, plaque pH measured by the microtouch method is a method that can be used for discriminating between individuals with varying caries prevalence. © 2014 S. Karger AG, Basel
Dental caries is prevalent in people of all ages worldwide [Petersen et al., 2005]. The assessment of caries activity is therefore an important step towards identifying subjects who are at risk of developing the disease. Due to the multifactorial complexity of dental caries, it is essential to have as much information as possible about the different variables directly or indirectly (primary and secondary factors) involved in the caries process [Fontana and Zero, 2006]. Apart from aetiological factors, a number of biological and environmental factors need to be taken into account [Selwitz et al., 2007]. The use of fluorides, especially in toothpaste, the level of oral hygiene and dietary habits are parameters that are included when assessing the caries risk, but they do not give any updated information about the current caries activity. It is therefore important to identify other parameters that mirror the complex interaction between all the factors underlying the disease. E. Miguel Aranibar Quiroz Department of Endodontics Institute of Odontology SE–405 30 Gothenburg (Sweden) E-Mail miguel.aranibar @ odontologi.gu.se
Downloaded by: Universitätsbibliothek Giessen 134.176.129.147 - 5/15/2015 2:49:59 PM
Key Words Adolescents · Caries risk · Cariogram · Diet · Plaque pH
148
Caries Res 2014;48:147–153 DOI: 10.1159/000355614
Materials and Methods Subjects and Study Design In Sweden, all individuals up to the age of 19 years have access to free, regular dental care. A total of 100 adolescents, aged 14–15 years at the start of the study, were asked to participate in this cross-sectional study. The individuals came to the clinic for their annual dental check-up at two of the Public Dental Clinics in the Gothenburg area in 2009. All the participants were healthy and without any systemic disease. Informed consent was obtained prior to the start of the study. The study was approved by the Ethics Committee at the University of Gothenburg (Ö 370–02). All data collection was carried out under regular conditions. Dental caries was identified by clinical and radiographic examination. Plaque pH was then measured, followed by collection of plaque and saliva samples and registration of the amount of plaque. Finally, the participants answered a questionnaire. Clinical Variables Caries registration was carried out according the criteria described by WHO [1997], including both initial and manifest lesions. Two bitewing radiographs were taken for assessment of enamel and dentine caries. Calculation of DSm (decayed surfaces, manifest caries), DSi (decayed surfaces, initial caries), DSm + i (decayed surfaces, both manifest and initial caries), and DTm (decayed teeth, manifest caries) was based on the data collected from the combined clinical and radiographical examination. A pooled plaque sample was collected with four sterile toothpicks, according to Kristoffersson and Bratthall [1982], from one interproximal area in both front and premolar regions on each side of the upper jaw. It was transferred to transport medium for estimation of the number of MS in relation to the total number of cultivatable microorganisms. A paraffin-stimulated whole saliva sample was collected during 3 min, after which the salivary secretion rate was calculated in millilitres per minute. One millilitre of saliva was then transferred to a transport medium for the determination of the number of MS and lactobacilli (LB). The buffer capacity was determined using a chairside kit (Dentobuff Strip®, Orion Diagnostica, Espoo, Finland); this test discriminates between low (yellow, ≤4.0), medium (green, 4.5–5.5) and high (blue, ≥6.0) buffer capacity. All samples were transported to the laboratory at the Department of Cariology on the same day. At the laboratory, the plaque samples were dispersed by sonication for 10 s and the saliva samples were shaken on a mechanical mixer for half a minute. After serial dilutions in 0.05 M phosphate buffer (pH 7.3), 25-μl portions of the plaque and saliva samples were plated in duplicate on MSB agar for the growth of MS. The MSB agar plates were incubated in candle jars at 37 ° C for 2 days and the blood agar plates in 95% H2 and 5% CO2 for 7 days. For the growth of LB, the saliva samples were plated on Rogosa SL agar and incubated aerobically at 37 ° C for 3 days. For the total growth of microorganisms in the plaque samples, blood agar was used. The number of colony-forming units of MS was counted on the MSB agar and identified by their characteristic colony morphology. All colony-forming units in the Rogosa SL agar were considered to be LB. The proportion of MS in plaque was calculated as the percentage of the total count of colony-forming units on blood agar. The amount of plaque was determined visually and without staining, according to Silness and Löe [1964]. This was carried out
Aranibar Quiroz/Alstad/Campus/ Birkhed/Lingström
Downloaded by: Universitätsbibliothek Giessen 134.176.129.147 - 5/15/2015 2:49:59 PM
A relationship between plaque pH and caries activity was observed by Stephan [1944] already in 1944. He reported that although caries-free individuals could acquire acidogenic plaque, the pH fall after sugar exposure was greater in caries-active than in caries-inactive subjects. These findings have been followed by more recent studies [Shimizu et al., 2008]. The acidogenicity of the dental biofilm is considered as a crucial factor in the caries process. Acid production in plaque affects the solubility of the underlying enamel/dentine surface and may consequently lead to the development of a caries lesion. Furthermore, a difference in plaque acidogenicity has been found when comparing individuals with normal and low salivary secretion rates [Lingström and Birkhed, 1993], as well as in relation to the number of mutans streptococci (MS) in the oral cavity [Aranibar Quiroz et al., 2003]. As the plaque pH graph is considered to summarise most of the factors involved in the caries process [Lingström et al., 1993], its individual pattern is therefore believed to indicate the current caries risk. Different information, including clinical, anamnestic, dietary and salivary tests, is collected when assessing caries risk; this is time-consuming and the data are often difficult to interpret. A chairside method that could be used to perform a quick screening of the actual caries risk is therefore desirable. It is important for a method of this kind to take into account as many parameters as possible involved in the caries process; it should also been able to provide reproducible results under standardized conditions. The Cariogram is a computer-based model that has been developed to describe and calculate the individual caries risk profile [Bratthall and Hänsel Petersson, 2005]. It has been found to be a suitable tool when used in dental practice as it allows a more objective data interpretation than the clinical examination [Hänsel Petersson and Bratthall, 2000; Hänsel Petersson et al., 2002; Sonbul et al., 2008; Al Mulla et al., 2009; Campus et al., 2009; Almosa et al., 2012; Campus et al., 2012]. However, the Cariogram is a theoretical model and it does not measure the biological activity in the ecological environment surrounding the tooth. The measurement of plaque pH may be a better alternative in the diagnosis of the caries risk. The hypothesis was that the in vivo plaque pH method has the potential to assess caries risk. As a first step in this direction, the aim of this study was to evaluate the relationship between plaque pH and different caries-associated variables in a group of adolescents with varying caries prevalence, and as a second step to compare the findings with the caries risk profile obtained with the Cariogram.
Plaque pH No specific instructions regarding oral hygiene or food intake were given prior to the test. Measurements were made at two approximal sites in the front and premolar regions on both sides in the upper jaw using an iridium microelectrode (Beetrode®, MEPH-1; W.P. Instruments, New Haven, Conn., USA), according to the method described in detail by Lingström et al. [1993]. The electrode was connected to an Orion SA 720 pH/ISE Meter (Orion Research, Boston, Mass., USA), equipped with a porous glass reference electrode (MERE 1; W.P. Instruments). A salt bridge was created in a 3 M KCl solution between the reference electrode and one of the subject’s fingers. The measurement was carried out before (0 min) and at five different time points (2, 5, 10, 15 and 20 min) after a mouth rinse with 10 ml of 10% sucrose for 1 min. The electrode was calibrated against standard buffers before and during the measurements. Questionnaire After the clinical examination and sampling, the participants answered a questionnaire consisting of 14 closed questions related to general health, dietary habits, oral hygiene and the use of fluorides. Cariogram The Cariogram is an interactive computer-based program used to illustrate the individual caries risk profile [Petersson et al., 1998]. Data from ten parameters related to caries disease are entered into the computer program in order to assess the current caries risk. A score is given according to a predetermined scale for each factor. The result is expressed as a pie chart, in which the caries-associated factors are shown in different sectors and ‘the chance of avoiding caries’ is presented as a figure from 0 to 100%, where 0% (high risk) means that caries lesions will definitely develop over time [Petersson and Bratthall, 2000]. Statistical Analyses The mean plaque pH from the four sites for each individual was determined, after which the mean values for plaque pH, minimum-pH and maximum-pH decrease were calculated for each curve. The area under the curve at pH 5.7 (AUC5.7) and at pH 6.2 (AUC6.2) was calculated using a computer program [Larsen and Pearce, 1997]. The individuals were arranged in three groups based on their minimum-pH values: (1) low (≤5.3), (2) medium (>5.3– 6.3) and (3) high (>6.3). The selection of these three values was based on the fact that the mean value of the minimum pH ±1 standard deviation (5.8 ± 0.5) for the whole population included up to 70% of the participants (medium-pH group, n = 72), leaving the outermost groups on each side (low-pH group, n = 15 and high-pH group, n = 13). The difference between the clinical variables in the three pH groups was tested with one-way ANOVA. The microbial data from saliva and plaque were transformed into logarithmic values. Correlation coefficients between plaque pH variables and the different clinical caries variables were assessed. The sugar intake frequency was regarded as ‘low’, when the subject reported a consumption of a maximum intake of twice a week, ‘medium’
Plaque pH and Caries in Adolescents
when more than twice a week but not daily, and ‘high’ when daily consumption of sugar was reported. The number of initial (DSi) and manifest and initial (DSm + i) caries lesions was grouped as follows: DSi, DSm + i = 0, DSi, DSm + i 1–3, DSi, DSm + i ≥4. Cariogram scores were grouped into four different classes: 81–100 (n = 39), 61–80 (n = 22), 41–60 (n = 27) and 0–40% (n = 12) ‘chance of avoiding caries’. The association between the plaque pH groups and the Cariogram scores was assessed using a χ2 test. A non-parametric test for trend across ordered groups developed by Cuzick [1985] was performed. The reference group included individuals with the most favourable exposure level (i.e. 81–100% chance of avoiding caries; DSi = 0; DSm + i = 0 and low sugar frequency). A multiple linear regression model stratified for plaque-pH categorisation using initial decay surface (DSi) as the dependent variable was created. All the data were analysed using STATA® software (Stata Corp., College Station, Tex., USA, version 10.1 for Macintosh, http://www.stata. com); p values of 5.3–6.3
High (n = 13) min pH >6.3
mean ± SD range
mean ± SD range
mean ± SD range
mean ± SD range
Clinical variables DSm 0.9±1.7 DSi 3.2±4.5 DSm + i 4.1±5.4 DTm 0.8±1.6 Saliva and microflora Secretion rate, ml/min 2.2±0.9 Plaque index, % 19.5±14.2 MS saliva, log CFU/ml 4.2±1.8 LB saliva, log CFU/ml 3.6±1.3 MS plaque, log CFU/sample 1.9±1.8 Total count in plaque, log CFU/sample 6.9±0.5 Plaque pH Maximum pH decrease 0.9±0.4 Minimum pH 5.8±0.5 AUC5.7 1.2±2.7 AUC6.2 4.1±5.3 Cariogram Chance to avoid caries, % 67.1±21.3
0–9 0–25 0–29 0–7
1.1±1.8 7.0±7.8 8.1±9.1 1.1±1.7
15
pH
0.4
