Assessing cutoff values of saliva parameters and oral microbes in dental students
Peter Germann Tutor: Nicklas Strömberg
Abstract: 249 Text: 3632
Tables and figures: 4 References: 23 Pages: 20
ABSTRACT
Background: The ability to identify individual risk factors for development of caries are important parts in the perspective of being able to control and limit caries disease. One such component is to establish, cut-off values for normal versus not normal values of known and putative risk factors.
Aim: The aim was to analyze if individual variations of the amount of stimulated saliva secretion, pH-level and buffer capacity as well as content of S. mutans and A.
actinomycetemcomitans could be used to form an alternative method to reach cut-off values based on the natural variation in a population of dental students.
Methods: This study was conducted by collecting saliva from 71 dental students at Umeå university. The saliva was collected three times over a period of five weeks. The collection of saliva was conducted pair vice by the students themselves. The saliva was analyzed regarding saliva secretion rate, buffer capacity, pH-level and quantity of S. mutans and A.
actinomycetemcomitans. Measured factors were plotted graphically to show their stability over time among the participants.
Results: Saliva secretion (and buffering, pH) showed a normal distribution that ranged widely but with stable intra-individual values, the bacterial factors showed a non-parametric distribution ranging from negative to positive values that varied largely.
Conclusions: The normal distribution of salivatory properties makes it possible to calculate cut-off values based on standard deviation. The non-parametric distribution of the microbial factors suggests cut-off values based on infected versus non-infected and intra-individually high numerical counts of bacteria on repeated analysis.
BACKGROUND
Dental caries is one of the most common chronic diseases in the world (Selwitz et al., 2007, Kassebaum et al., 2015, Schwendicke et al., 2015, Martinez-Mier et al., 2015, Oziegbe et al., 2013). Untreated caries in permanent dentition affects 2.5 billion people, and untreated caries in deciduous dentition 573 million children worldwide (Righolt et al., 2018). The cause of the disease is an interaction between the host saliva defenses, fermentable carbohydrates in the diet and acid producing bacteria, that over time will cause a demineralization of the enamel and dentin of teeth which in time can lead to cavities and decay (Fejerskov and Kidd, 2008).
The total cost of dental diseases worldwide is estimated to $544.41 billion, when both direct and indirect costs are considered in 2015 (Righolt et al., 2018). The major portion of this cost is accredited to severe tooth loss (Righolt et al., 2018), a condition often preceded by untreated dental caries (Fejerskov and Kidd, 2008). With this taken into account, it’s reasonable to assume that diagnosis of caries and its prevention and treatment at an early stage would be beneficial to a lot of people and potentially save a lot of money from a socioeconomic standpoint.
The factors associated with caries are clinically used either to identify potential causes of caries activity and how to treat the disease or to estimate the risk for the disease or chance to remain healthy. These so-called risk factors, regardless of using cut-off values or raw data, can accordingly be used to categorize patients into different groups based on cause and treatment needs and to estimate the individual risk to develop caries. To get a more accurate prediction a multimarker approach using 18 different markers to predict caries have shown to be more accurate than a single to five marker approach (Nordlund et al., 2009). Today the most accurate single marker when it comes to predicting future caries development is past caries experience (Fontana M, Zero DT. 2006, Zero et al., 2001, NIH Consens Statement. 2001, Brons-Piche et al., 2019, Alaluusua S. 1993).
Both saliva and bacteria are factors used to investigate the risk and cause of caries. The properties and quantity of the saliva for example its pH-level, buffert capacity, and salivatory secretion rate as these can affect the formation and development of caries (Fejerskov and Kidd, 2008). Microbiological parameters especially the presence of bacteria from the S. mutans species have been proven to have a correlation with the development of dental caries (Srilatha et al., 2018, O'Sullivan et al., 1996, Thibodeau et al., 1996). Measurements of the presence and quantities of S. mutans is clinically used to assess the risk for development of dental caries.
Research indicates that the S. mutans adhesion type SpaP (A, B or C) in combination with collagen binding (Cnm or Cbm) play a role in the development of caries (Esberg et al., 2017).
One problem with predicting future caries lesions is that the tools used are very “blunt”. For example: if a saliva test gets sent into a lab for analysis for presence and quantity of S. mutans,
a species of acid producing bacteria that has been proven to have a strong correlation with dental caries (Loesche et al., 1975). The laboratory result would typically only list the quantity of S. mutans, with no regards to specific sub-types. The problem with this is that it has been proven that S. mutans with different SpaP type (A, B or C) and individual subtypes of those, coupled with if the strain is of the type high acid tolerant or low acid tolerant, have an impact on the bacteria’s ability to cause dental caries (Esberg et al., 2017). A lower number of a high acid tolerant strain of a high virulent SpaP sub type could be a valid risk indicator that’s likely to go unnoticed if the lab report only contains a total quantity of S. mutans with no quantities of individual sub types.
Another problem with measuring salivatory properties (secretion rate, pH-level, buffer capacity) and for presence of microbes (S. mutans, Lactobacillus), is that some of these values are not stable over time. The levels of S. mutans for example is highly dependent on the type of diet (Dennis et al., 1975) and levels can vary between meals. The variation in levels of different parameters (especially the microbial ones) can give a less than optimal reading, which makes an argument that some of the parameters might be better tested more than once, in order to give a more fair value. Another thing to consider is that some of these values are something that we all have, pH-value of the saliva for example, however not everyone will be S. mutans positive. This means that these parameters should be interpret differently, pH-value of the saliva will have a natural variation as everyone has it and a standard deviation value could be calculated, this means that the results over a population would follow a normal distribution curve which would be best analyzed through quantitative measures. In the example with S. mutans as not everyone is positive, the result over a population would instead follow a non-parametric distribution, which would likely mean it could be better analyzed through qualitative means.
So-called cut-off values that indicate when saliva and bacteria factors may consitute an increased risk for development of caries are used i) in the education of dental students, ii) in the clinical risk and cause investigation of trained dentists and iii) in the insurance system to decide if the patient is entitled to reimbursement of costs. In order to establish guidelines for cut-off values that indicates potential harmful levels it is important to know the distribution of the factor among healthy individuals, normal or non-parametric distribution, and how stable the distribution pattern is in shape and spread over time and how stable each individual is regarding the given factor. Not before these factors are known can it be decided what levels of risk factors that are likely to coincide with health or disease. The overall distribution of events forms the basis for today’s statistics, either handling data following a normal distribution (parametric statistic´s) or non-parametric distribution (non-parametric statistics) (Stephenson, 2010).
The primary aim of the present study was to analyze the distribution of putative salivatory, and microbial risk factors associated with development of dental caries. A secondary aim was to suggest cut-off values based on the results.
METHODS
Study participants
The only inclusion criteria for the study was being a student at the dental program at Umeå university at the time on their fourth semester. Exclusion criteria was limited to people not wanting to participate.
The collection of data for these experiments was conducted in several different steps.
Collecting saliva. The students worked pair-vice, and sampled the saliva from each other according to a standardized procedure. The instruction was to chew on a piece of paraffin for 1 min end then swallow the produced saliva. After the saliva had been swallowed, a timer started, and the subject continued to chew on the paraffin. The saliva was collected in a test tube, until a total amount of 5 ml had been collected. When 5 ml had been produced, the test continued until the strike of the next minute (so the total time of the test would be in even minutes). After completion of the sampling the subject carefully blow his/her breath into the tube and then sealed it. The test tube was placed on ice and sent to a lab together with the information of the total time the test took to complete. The sampling was repeated three times within a five-week period with at least one week in between.
Saliva secretion was calculated by weighing the test tubes filled with saliva and subtracting the weight of the test tube (6.3 g). The weight in grams was then multiplicated with an assumed density 1 g/ml in order to get the value for volume of saliva collected in ml. By then dividing that value by the time it was collected in min a value for saliva secretion measurement in ml/min was given.
Buffer capacity of the saliva was measured by pipetting 1 ml of undiluted saliva to a test tube containing 3 ml of 0.005 M of HCl. The sample was then aerated for 20 min by means of inserting a plastic tube that slowly released technical air into the samples. The air flow was carefully adjusted by the technician for each sample as to not form excessive foam on the thicker saliva samples. After being aerated for 20 min the pH-value of the sample would be measured by means of a Metrohm 744 pH-meter, the procedure for the measurement is exactly the same as for the pH value of collected saliva below.
pH values of the collected saliva, were obtained by means of a Metrohm 744 pH-meter.
The pH-meter remains accurate by a daily calibration regimen with pH 7.00 and pH 4.00
technical buffer solution from Mettler-Toledo AG. The electrodes of the pH-meter were dipped in the undiluted saliva samples and the value was written down. Between each sample the electrodes would be gently rinsed in MQ-solution and any excess solution would be gently dabbed off with a piece of paper.
S. mutans and A. actinomycetemcomitans was measured by first preparing the saliva samples by pipetting 200 μl of saliva to a centrifuge tube, then centrifuging the samples for 10 min at 12,470 xg using a Sigma 1-14 centrifuge. The supernatant was removed, and the pellet resuspended in 100 μl 0,2% SDS (sodium dodecyl sulfate), incubated at 98°C for 10 min. The samples were then centrifuged for 10 min at 12,470 xg. The supernatant was transferred to a new test tube. For all S. mutans and A. actinomycetemcomitans qPCR tests one μl of this suspension was used.
Measuring the levels of S. mutans was done by qPCR using the KAPA SYBR FAST Universal qPCR kit and Corbett Rotor-Gene 6000 with the use of S. mutans specific primers: 5′- AGCCATGCGCAATCAACAGGTT-3′ and 5′-CGCAACGCGAACATCTTGATCAG- 3′. To get calibration curves, serial dilutions of DNA prepared from the S. mutans strain Ingbritt were used.
Measuring the levels of A. actinomyctemcomitans were done by the same procedure as for the measurements for S. mutans with the exception for the primers and the DNA strain used for calibration. For this test A. actinomycetemcomitans‐specific ltxA primers: 5’- CTAGGTATTGCGAAACAATTTG-3’ and 3’-CCTGAAATTAAGCTGGTAATC-‘5 were used.
Statistics
Microsoft Excel was used to calculate mean value and standard deviation for the different tests. To calculate mean value all results from a test was highlighted and the excel command AVERAGE was used, for the standard deviation the excel command STDEV.P was used.
Ethical reflection
The subjects for this study were offered to participate in this study as part of their education, where they would get to practice collecting saliva samples for lab analyses, and also have the opportunity to see which variables that would remain stable and which would vary over time.
They would also get to follow up on the results of the study and get a deeper understanding of what the different results and cut-off values meant. The identity of the subjects was kept secret by way of assigning a number to each individual. The subjects would get access to the results
and if they wanted to, get to know which number was theirs. As the test results was kept secret from the other participants, and the collecting of samples was done in a non-invasive manner with little risk of negative impact on the participants health. It is fair to assume that the beneficial contribution to the science community and better understanding of caries risk factors and cut-off values in the participants of the study outweighs the possible negative.
RESULTS
Salivatory values had a normal distribution pattern but with a large inter-individual variation.
When saliva secretion was measured among the 71 study participants at three separate collections, a normal distribution pattern with a mean 2.0 – 2.1 ml/min value (standard deviation 0.87 – 0.96 ml/min) across the three runs was observed (Fig. 1, Supplemental 1).
However, the inter-individual values ranged from about 0.5 ml/min (n=6) to about 4.5 ml/min (n=6). The individual secretion values between the three measurements had an average difference of 0.46 ml/min between the tests, meaning that the intra-individual variations were low. The measurements of the buffer capacity and pH-value of the saliva also had a normal distribution with stable intra-individual values and larger inter-individual variations (Supplemental 2, 3, 4.)
Microbial risk factors had a non-parametric distribution pattern.
The measurements of S. mutans had a non-parametric distribution with a median of 0.015 – 0.03765 pg DNA between the different samplings. On an intra-individual level there where variations with a range of 0 – 4.12 pg DNA across the three measurements (Fig 2). The S.
mutans subjects with a score below 0.05 pg DNA were likely to have stable values across the three runs, the subjects with a score of 0.5 pg DNA or higher were more likely to have variations of several hundred percent between the tests (Fig 2). The tests for A.
actinomycetemcommitans largely followed the same pattern regarding stability between the different tests (Fig 2).
DISCUSSION
This study shows that different risk factors measured from the saliva followed two different patterns. Salivatory secretion, pH-level of the saliva and salivatory buffer capacity had a normal distribution curve. S. mutans and A. actinomycetemcomitans had a non-parametric distribution curve. It is, thus, important to separate the two when it comes to risk assessment.
In a normal distribution you have 95.5 % within two standard deviations and 99.7 % of the population within three standard deviations. This gives values of what should be regarded as
normal values (within two standard deviations),as well as when a value is between two and three standard deviations When assessing caries risk from tests following a normal distribution pattern it is essential to know what is to be considered normal and what would be considered deviating from the norm. The shape of the curve is also very informative, and its importance should not be disregarded. In our test of saliva secretion, the values ranged from 0,5 ml/min to 5,2 ml/min which affects the shape of the curve in such a way that it is shallow and wide. With this pattern follows a large standard deviation, in this case of almost 1 ml/min.
If the mean values instead would have ranged between 2,0 ml/min and 2,5 ml/min the curve would appear high and narrow with a lower standard deviation value. This would mean a narrow natural variation and even small deviations from the mean could signify a heightened risk for development of disease.
In our analyzes of bacteria we found a non-parametric distribution curve indicating that a large part of the population have a value of zero or close by, and a few extreme outliers often with several thousand times higher values than the majority, and in between these, often closer to the majority group there would be another group of individuals. These represent the different cut-offs that we were trying to find. The majority are often negative or have a minor well controlled infection which is represented by the large group close to the zero value. The few with extreme values represent those individuals with a clearly heightened risk to develop disease, the so-called high-risk group. And the ones in between represent those that are infected and could have a slightly increased risk of developing disease.
Another important finding was the stability over time of the different measured risk factors.
The amount of the analyzed risk factors were in most cases stable at an intra-individual level.
The risk factors with a more dichotomized distribution on the other hand often fluctuated several hundred percent between the measurements. This could indicate that tests following a non-parametric distribution, in a clinical environment should be monitored in a different way than those following a normal distribution pattern. A drop of a couple of hundred percent between measurements when testing for S. mutans would not necessarily mean that the current treatment is working, it could be the same natural fluctuation over time observed in this study. To get an accurate measurement of the progression of a parameter following such a pattern, a regime of tests close to each other in time over an extended period could be warranted. The good of such a course would have to be weighed against the vast increase in cost and manpower such a test regime would demand.
The cut-off values used in the student clinic (Umeå Universitet, 2008) can be found in Table 1. Using these cut-off values on our test population would yields some interesting results. The mean for our subject’s buffer capacity ranges from pH 5.5 – 6.1 between the three tests. This would in turn put roughly half of our subjects below the pH 5.5 cut-off for low salivatory buffer,
meaning that those individuals would be considered having an increased risk of developing caries. Having a cut-off value within one standard deviation from the mean, means that a substantial part of the population would be considered at risk. If the cut-off value was at for example two to three standard deviations from the mean, the marker would gain value as less of the population would be considered at risk and of those considered at risk, could be more likely to be infected with caries as their buffer value would be further away from what is considered optimal.
Risk factors, regardless of using cut-off values or raw data, can be used to categorize patients into different groups based on cause and treatment needs and to estimate the individual risk to develop caries. For example, a patient with low salivatory pH and buffer, probably have a genetical predisposition to developing caries. In this case adding a fluoride mouthwash or switching to a toothpaste containing stannous fluoride might outweigh the genetic disadvantage. A patient with a very high count of S. mutans could have a cariogenic diet that caters the growth of sugar loving bacteria. Boosting the teeth ability to withstand low pH- values might very well help a bit, but a treatment with chlorhexidine to significantly lower the levels of bacteria followed up by a change in diet to keep the S. mutans numbers at a healthy level would most likely yield better results in the long run.
Since the S. mutans data in our study was measured in pg DNA instead of the more commonly used CFU/ml and there is no way to reliably convert these units between each other, there is no way to determine scientifically accurate cut-off values for infected and high risk infected that would be generally applicable. But in order to illustrate a table with cut-off values using the reasoning of this study, an attempt of converting pg DNA to cfu has been made using the following reasoning: S. mutans infects 40-80% of the population (Esberg et al., 2017), this would make it reasonable to assume that in our test population about 60% would be positive for S. mutans. Therefor a good cut-off value for positive/negative on that parameter was set to 0.0563 pg DNA which would include 60% of our subjects. Studies have shown that S. mutans levels above 1,000,000 cfu pose a high risk of developing dental caries (Pannu P, et al., 2013, Nanda J, et al., 2015), and that about 28% of the adult population is in that category (Pannu P, et al., 2013). The cut-off value for our high-risk S. mutans levels will therefore be set to 0.129 pg DNA which correlates to 28% in our tests. To further clarify: the cut-off values for S. mutans suggested (Fig. 2a) are made largely by assumptions in order to illustrate how cut-off values based on this study could be set, they are NOT generally applicable elsewhere.
Strengths and weaknesses
Some of the tests, including DeFS and collection of saliva was conducted by the students themselves working in pairs. This is of course a possible flaw of this study, since they were not
calibrated against each other regarding DeFS and PLI scoring, and the scoring of DeFS was done without use of radiographs these results have been omitted from this study. Also, every student could not participate during every test, which means the data is not 100 % complete.
Another flaw with this study is that the measurements for S. mutans is in pg DNA instead of the more widely used cfu. This makes it hard to compare the data with published studies regarding S. mutans and cut-off values.
One of this study’s strengths is that it conducted the same tests on the same subjects over a period of time. This gives good insight of the individual stability and natural changes of the measured parameters on an individual level.
Conducting a follow up study on this population in a five- and/or a ten-year period, measuring DeFS would give an insight as to which if any, of the salivatory risk factors measured had a correlation in predicting future caries lesions. It would also help in assessing reasonable cut- off values for the different risk factors measured in this study. It could be of clinical interest to conduct a study that would measure if large fluctuations in normally stable parameters could correlate with an increased risk of developing caries.
In conclusion, using the methods suggested in this study where we use cut-off values based on two standard deviations for risk factors following a normal distribution. And not infected, infected and an extreme value for when there could be a significant risk for disease for factors following a non-parametric distribution. We have included a table (Table. 2) with new suggested cut-off values and a graph showing the test population in regard to the suggested cut-off values for S. mutans (Fig. 2).
REFERENCES
Alaluusua S. Salivary counts of mutans streptococci and lactobacilli and past caries experience in caries prediction. Caries Res. 1993;27 Suppl 1:68-71.
Brons-Piche E, Eckert GJ, Fontana M. Predictive Validity of a Caries Risk Assessment Model at a Dental School. J Dent Educ. 2019; 83: 144-150.
Dennis DA, Gawronski TH, Sudo SZ, Harris RS, Folke LE. Variations in microbial and biochemical components of four-day plaque during a four-week controlled diet period. J Dent Res. 1975; 54:716-722.
Diagnosis and management of dental caries throughout life. NIH Consensus Statement.
2001;18:1-23.
Esberg A, Sheng N, Mårell L, Claesson R, Persson K, Borén T, et al. Streptococcus Mutans Adhesin Biotypes that Match and Predict Individual Caries Development. EBioMedicine.
2017; 24: 205-215.
Fejerskov O, Kidd E. Dental Caries the Disease and it’s Clinical Management. 2 ed. London:
Blackwell Munksgaards Ltd; 2008.
Fontana M, Zero DT. Assessing patients' caries risk. J Am Dent Assoc. 2006;137:1231-1239.
Kassebaum NJ, Bernabé E, Dahiya M, Bhandari B, Murray CJ, Marcenes W. Global burden of untreated caries: a systematic review and metaregression. J Dent Res. 2015;94:650-658.
Loesche WJ, Rowan J, Straffon LH, Loos PJ. Association of Streptococcus mutants with human dental decay. Infect Immun. 1975 ;11:1252-1260.
Martinez-Mier EA, Zandona AF. The impact of gender on caries prevalence and risk assessment. Dent Clin North Am. 2013;57:301-315.
Nanda J, Sachdev V, Sandhu M, Deep-Singh-Nanda K. Correlation between dental caries experience and mutans streptococci counts using saliva and plaque as microbial risk indicators in 3-8 year old children. A cross Sectional study. J Clin Exp Dent. 2015;7:e114-8.
Nordlund A, Johansson I, Källestål C, Ericson T, Sjöström M, Strömberg N. Improved ability of biological and previous caries multimarkers to predict caries disease as revealed by
multivariate PLS modelling. BMC Oral Health. 2009 3;9:28.
O'Sullivan DM, Thibodeau EA. Caries experience and mutans streptococci as indicators of caries incidence. Pediatr Dent. 1996;18:371-374.
Oziegbe EO, Esan TA. Prevalence and clinical consequences of untreated dental caries using PUFA index in suburban Nigerian school children. Eur Arch Paediatr Dent. 2013;14:227-231.
Pannu P, Gambhir R, Sujlana A. Correlation between the salivary Streptococcus mutans levels and dental caries experience in adult population of Chandigarh, India. Eur J Dent.
2013;7:191-5.
PM och anvisningar Kariologi vt 2008. Umeå Universitet; 2008
Righolt AJ, Jevdjevic M, Marcenes W, Listl S. Global-, Regional-, and Country-Level Economic Impacts of Dental Diseases in 2015. J Dent Res. 2018 05;97:501-507.
Schwendicke F, Dörfer CE, Schlattmann P, Foster Page L, Thomson WM, Paris S.
Socioeconomic inequality and caries: a systematic review and meta-analysis. J Dent Res.
2015 ;94:10-18.
Selwitz RH, Ismail AI, Pitts NB. Dental caries. Lancet. 2007 6;369(9555):51-59.
Srilatha A, Doshi D, Kulkarni S, Reddy MP, Reddy BS, Satyanarayana D. Determination and Comparison of Dermatoglyphic Patterns and Salivary Streptococcus mutans Counts and Its Correlation with Dental Caries among 3- to 6-year-old Children. Oral Health Prev Dent.
2018;16(3):291-7.
Stephenson FH. Calculations for Molecular Biology and Biotechnology A Guide to Mathematics in the Laboratory. 2 ed. London: Academic press; 2010.
Thibodeau EA, O'Sullivan DM. Salivary mutans streptococci and dental caries patterns in pre-school children. Community Dent Oral Epidemiol. 1996;24:164-168.
Zero D, Fontana M, Lennon AM. Clinical applications and outcomes of using indicators of risk in caries management. J Dent Educ. 2001;65:1126-1132.
FIGURES
0,0 1,0 2,0 3,0 4,0 5,0
Saliva sekretion ml/min
Saliva secretion first measurment
Saliva secretion Mean
0,0 1,0 2,0 3,0 4,0 5,0 6,0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70
ml/min
Subject
Saliva secretion
First measurement Second measurement Third measurement
Figure 1: A, Showing the results for saliva secretion from the first saliva measurements with the mean in red and std segments marked in black. The second and third measurement followed the same pattern as observed in the first. B, A plot showing the different measurements for each individual.
0 50000 100000 150000 200000 250000 300000 350000 400000 450000 500000
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70
CFU
Subject
A. Actinomycetemcomitans
First measurement Second measurement Third measurement Median
0 0,5 1 1,5 2 2,5 3 3,5 4 4,5
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70
pg DNA
Subject
S. mutans
First measurement Second measurement Third measurement Infected cutoff High risk cutoff
Figure 2: A, A plot showing results from all runs in relation to cut‐off values for S. mutans. B, A plot showing the results from the a.a. measurements of all three runs
TABLES
Table 1: Cut-off values used in the student clinic
Low Normal High
Saliva secretion < 0.7 ml/min > 1 ml/min
Salivatory pH-level < pH 6.8 pH 7.0 – 7.5 > pH 7.5 Salivatory buffert
capacity
< pH 5.5 > pH 5.5
S.mutans > 250,000 CFU/ml
Table 2: A table showing suggested cut-off values based on this study
Low Normal High
Saliva secretion < 0.2 ml/min 0.2 - 4 ml/min > 4 ml/min Salivatory pH-level < pH 6.9 pH 6.9 – 8.0 > pH 8.0 Salivatory buffert
capacity
< pH 2.9 pH 2.9 - 8.7 > pH 8.7
S.mutans Not infected Infected Mutans millionaire
SUPPLEMENTALS
Supplemental 1: A bigger version of Figure 1B, to more easily be able to interpret the individual results between the different measurements.
0,0 1,0 2,0 3,0 4,0 5,0 6,0
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70
ml/min
Subject
Saliva secretion
First measurement Second measurement Third measurement
1 2 3 4 5 6 7 8 9
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70
pH
Subject
Saliva pH
First measurement second measurement Third measurement
0 1 2 3 4 5 6 7 8 9 10
0 5 10 15 20 25 30 35 40 45 50 55 60 65 70
pH
Subject
Saliva buffer capacity
First measurement Second measurement Third measurement
Supplemental 2: Plots showing results from all three runs from Saliva pH and Saliva buffer capacity to illustrate the individual stability between the different tests.
Supplemental 3: Graphs showing results from each of the different measurements of the buffer capacity of the saliva. Mean marked by a red line and each std marked with a black line.
2 3 4 5 6 7 8 9
pH
Saliva buffer capacity ‐ first measurement
pH value Mean
2 3 4 5 6 7 8 9
pH
Saliva buffer capacity ‐ third measurement
pH value mean
2 3 4 5 6 7 8 9
pH
Saliva buffer capacity ‐ second measurement
pH value Mean
Supplemental 4: Graphs showing results from each individual measurement of the pH‐value of saliva. Median is marked with a red line and each std is represented by a black line.
6,7 6,9 7,1 7,3 7,5 7,7 7,9 8,1 8,3
pH
pH in saliva first measurement
pH value Mean
6,7 6,9 7,1 7,3 7,5 7,7 7,9 8,1 8,3
pH
pH in saliva second measurement
pH value Mean
6,7 6,9 7,1 7,3 7,5 7,7 7,9 8,1 8,3
pH
pH in saliva third measurement
pH value Mean
