Journal of Conservative Dentistry

ORIGINAL ARTICLE
Year
: 2011  |  Volume : 14  |  Issue : 4  |  Page : 395--400

Caries-risk assessment with a chairside optical spectroscopic sensor by monitoring bacterial-mediated acidogenic-profile of saliva in children


Annie Shrestha1, MA Mohamed- Tahir2, Jayshree Hegde3, Amir Azarpazhooh1, Anil Kishen1,  
1 Laboratory for Phototherapeutics in Oral Diseases, Discipline of Endodontics, University of Toronto, 124 Edward Street, Toronto ON M5G 1G6, Canada
2 Department of Preventive Dentistry, National University of Singapore, Singapore
3 Department of Conservative Dentistry and Endodontics, Oxford Dental College, Bangalore, India

Correspondence Address:
Anil Kishen
Laboratory for Phototherapeutics in Oral Diseases, Discipline of Endodontics, University of Toronto, 124 Edward Street, Toronto ON M5G 1G6
Canada

Abstract

Objective : This study aimed to evaluate the ability of an optical spectroscopic sensor (OSS) to monitor bacterial-mediated acidogenic-profile of saliva resulting from bacteria-sucrose interaction. Materials and Methods : Stage-1, characterization experiments were conducted to standardize the OSS. Stage-2 clinical experiments were carried out on stimulated saliva samples from 70 children of age-group 1-12 years. The bacterial-mediated acidogenic-profile of saliva mixed with sucrose was monitored using the OSS for 180 minutes. Results : The clinical patients were categorized based on the dmfs score as caries-active, caries-inactive and caries-free. The bacterial-mediated acidogenic-profile measured in terms of t1/2 monitored using the OSS was significantly different between the caries-free and caries-active (P<0.05); and caries-free and caries-inactive groups (P<0.005). Conclusions : The significant correlation of the acidogenic-profile determined using the OSS and the caries-status highlighted the OSS as a sensitive and rapid chairside tool for the quantification of the acidogenic-profile of saliva that could help in monitoring the caries-risk in children.



How to cite this article:
Shrestha A, Mohamed- Tahir M A, Hegde J, Azarpazhooh A, Kishen A. Caries-risk assessment with a chairside optical spectroscopic sensor by monitoring bacterial-mediated acidogenic-profile of saliva in children.J Conserv Dent 2011;14:395-400


How to cite this URL:
Shrestha A, Mohamed- Tahir M A, Hegde J, Azarpazhooh A, Kishen A. Caries-risk assessment with a chairside optical spectroscopic sensor by monitoring bacterial-mediated acidogenic-profile of saliva in children. J Conserv Dent [serial online] 2011 [cited 2019 Jul 19 ];14:395-400
Available from: http://www.jcd.org.in/text.asp?2011/14/4/395/87210


Full Text

 Introduction



Dental caries are considered a result of the interplay of three principal factors: host factors (teeth and saliva), microbial factors, and dietary factors, leading to demineralization of the inorganic portion and destruction of the organic substance of the tooth. [1] Due to its multifactorial nature, various parameters are used as caries-risk predictors. [2] One of the main etiological factors for dental caries is microbiota, and high and low numbers of odontopathogens have been reported in high and low caries-risk groups respectively. [3],[4] Early recognition of the caries-risk groups can be used to prevent the progression of the disease and to provide suitable treatment regimens to control the disease process which will provide a significant impact on lowering the cost of any public health caries prevention strategies directed to those most in need. Despite many attempts to introduce alternative and more reliable caries-risk assessment methods, past or present caries experience still remains the single best indicator of caries-risk. [5]

Significant correlations between mutans streptococci (MS) (Streptococcus mutans and Streptococcus sobrinus) and Lactobacilli (LB) counts with caries initiation and progression were reported. [2],[6] Further, colonization of these microorganisms in plaque and their levels in stimulated saliva has also been validated. [2],[6],[7] Studies have also shown variation in pH following glucose rinse in caries-free and caries-active subjects, with larger and faster rate of acid production in caries-active subjects. [6],[8] Dentocult SM and LB have been shown to provide a microbiological assessment of MS and LB respectively in the saliva and significant correlation with the conventional-selective-culture-based method was established. [9],[10],[11] However, these methods require a long incubation period before enumeration, do not give reliable quantitative information that can be used to compare between individuals, are expensive to use routinely, require laboratory support and have limited shelf life. The sensitivity of these chairside tests to isolate salivary MS has been shown to be comparatively low (71%). [12] The overall low predictive power of MS as well as LB in differentiating caries-risk groups emphasizes the need for some other parameters to define the risk status.

Optical spectroscopic sensors (OSS) can be used to detect changes in certain parameters based on the optical properties of the material being studied. A fiber-optic bundle is used in OSS to illuminate the sample with a broad-spectrum visible light source and the absorption spectrum is acquired with a high-resolution spectrophotometer. The OSS possess several advantages: (1) They can determine multiple clinical parameters simultaneously and rapidly. [13] (2) They are highly sensitive to minor changes in the optical characteristics of the test sample. (3) They have geometric versatility, portability and ease of operation. (4) Electromagnetic interference is absent which eliminates the risk of electric shock. (5) Optical fibers are non-toxic and biocompatible. [14],[15],[16] These optical sensors are being used in different fields for continuous and reversible sensing and detection of various physical and chemical parameters. [14],[17],[18] This study aimed to evaluate the ability of an OSS to monitor bacterial-mediated acidogenic-profile of saliva resulting from bacteria-sucrose interaction. The investigation was conducted in two stages. Firstly, OSS was set up and characterized to monitor the bacterial-mediated acidogenic-profile of saliva. The findings obtained from the OSS were correlated with the pH change monitored using a pH meter, and acidogenic bacterial count determined using the conventional-selective-culture-based method. In the second stage, saliva samples were collected from children with different caries-status and were used to determine the bacterial-mediated acidogenic-profile using OSS, and pH change using a pH meter.

 Materials and Methods



Experimental setup of the optical spectroscopic sensors

According to Beer Lambert's law, when a light of a specific wavelength passes through a sample, the fraction of light radiation absorbed by the sample will be a function of the concentration of substance in the path of light. Absorption spectrometry uses this theory to make qualitative and quantitative measurement of a given sample. [19] The experimental setup configured for monitoring the bacterial activity is shown in [Figure 1]. The system consisted of a white light source from a 24V tungsten halogen lamp (Mikropack Halogen Light Source 2000, Germany), with an illuminating fiber connected to it through a SubMiniature version A connector. The acquisition fiber was connected to a high-resolution spectrophotometer (HR 4000, Ocean Optics, Dunedin, Florida, USA) comprising a 1024-pixel charged couple device (CCD) providing a resolution of 0.5 nm, over a range of 400-800 nm. The spectrophotometer was interfaced to a computer. Laboratory-grade optical fibers (Ocean Optics Inc., USA), most efficient in the range of 300-800 nm were used as the illuminating and the acquisition fibers. A cuvette holder (Ocean Optics Inc., USA) was connected to the illuminating and acquisition fiber. Samples were dispensed in semi-micro cuvettes (VITLAB, Germany) with optical path of 10 mm for the measurement. The data acquired was recorded and analyzed using customized software (OOIBase 32, Ocean Optics Inc., USA).{Figure 1}

Stage 1: Characterization of the OSS

Approximately 5 ml of stimulated saliva was collected from a known caries-free (CF) individual. The saliva sample was mixed with 1 M sucrose solution in a 3:1 ratio (3 mL saliva +1 L sucrose) and incubated for 15 min at 37°C. After incubation, 10 μL of 0.01 M of the pH indicator dye Bromophenol Blue (Bio-Rad Diagnostics, USA) prepared in deionized water was added to 1 mL of saliva+sucrose sample. Bacterial-mediated acidogenic profile was monitored hourly from 15 min to 180 min of incubation by recording the absorption spectrum using (1) the OSS and (2) UV-VIS spectrophotometer (Shimadzu, Japan). (3) A pH meter (Model 30A, Orion, USA) with a resolution of 0.01 and a relative accuracy of ΁0.02 was also used to record the pH variation of the sample (saliva+sucrose) with time. Microbiological assessment of the saliva+sucrose sample was conducted using conventional selective Mitis Salivarius agar (Biomed Diagnostics, USA) for MS. At each hourly interval, 100 μL of the incubated saliva+sucrose sample was serially diluted and plated. The colony-forming units (CFU) were counted after incubation at 37°C for 48 h and expressed as log CFU/mL saliva.

Stage 2: Clinical experiments conducted using OSS on saliva samples from children

Saliva samples from 70 children of the age group 1-12 years attending the undergraduate student clinic of the National University Hospital (NUH), Singapore and from school children, Bengaluru, India were collected for the study purpose. Saliva collection for this study was approved by the Institutional Review Board of the National Healthcare Group, Domain Specific Review Board (DSRB), Singapore. Informed consent was obtained from the parents/guardians of the children below six years and additional assent was obtained from the children between 6-12 years. Caries experience was recorded using the dmfs and DMFS indices. Approximately 5 mL of stimulated saliva was collected from each child after chewing on a paraffin wax for 5 min. The tests were conducted immediately after the collection.

The saliva samples collected from the patients were mixed with 1 M sucrose solution in a 3:1 ratio (3 mL saliva +1 mL sucrose) and incubated for 15 min at 37°C. After incubation, 10 μL of 0.01 M of the pH indicator dye Bromophenol blue prepared in deionized water was added to 1 mL of saliva+sucrose samples. Bacterial-mediated acidogenic profiles were recorded using the OSS for each sample at different time intervals from 15 min to 180 min of incubation. pH values for all the samples during each time interval were recorded using the pH meter.

Statistical analysis

SPSS 17.0 (SPSS, Chicago, III, USA) was used for data management and analysis. Data entry errors were verified as they were performed, and subsequently double-checked on 10% of the sample. The value of 999 was assigned as a code for missing values. Preliminary variable distributions were generated. The main outcome variable in this study was the half-life (t 1/2 ) value of the exponential reaction and decrease in the absorption intensity as a function of time monitored using the OSS and pH measured using the pH meter. The independent variable was caries-risk as determined by the dmfs values. Patients were categorized based on their dmfs values; caries-active i.e., at least one decayed tooth surface, d≥1; caries-inactive i.e., no decayed tooth surfaces, d=0; or caries-free i.e. no decayed, missing or filled (due to caries) tooth surfaces, dmfs=0. [7] The Independent-Samples t-test and Non-Parametric 2 Independent samples test (Mann-Whitney U) were used to compare means for the two groups for each stimulus. Analysis of variance (ANOVA) with Tukey HSD Post Hoc was performed to compare the means of t1/2 among different risk groups. The level of statistical significance was set at 5%.

 Results



Stage 1: Characterization of the OSS

[Figure 2] shows the typical absorption intensity response of saliva after mixing with sucrose solution, monitored for 3 h. A conspicuous decrease in the absorption intensity at 595 nm as a function of time was observed, which could be due to the increased acid formation by the salivary acidogenic bacteria. The UV-VIS spectrophotometer reading reconfirmed the sensor results as it also showed decrease in absorption intensity at 595 nm with time. The pH dropped approximately two units below the initial value in 3 h [Figure 2]. The decline in absorption intensity observed in the OSS showed a strong positive correlation (r=0.825) with the decrease in pH with time. [Figure 3] shows the increase in MS count after 120 min of incubation, determined using the culture-based technique. The decline in the absorption intensity and increase in the bacterial count showed negative correlation (r=-0.80).{Figure 2}{Figure 3}

Stage 2: Clinical experiments conducted using OSS on saliva samples from children

[Table 1] shows the patients' details of caries-scores and age. All the three variables, changes in pH monitored using the pH meter, intensity variation and the half-life (t 1/2 ) of the exponential decay curve measured using the OSS determined from different samples followed a normal curve distribution. There were 40 children in the caries-active group, 11 in the caries-inactive and 19 in caries-free group. [Table 2] shows the average values and standard deviations for the three variables and the significance levels. The pH reduction and intensity variation from the time of reaction onset to 180 min was less in case of the caries-free group than the other two groups. However, this difference was not statistically significant (P>0.05). The t 1/2 of the exponential reaction, which corresponds to the bacterial-mediated acidogenic reaction, showed significant difference between three caries-risk groups. The t 1/2 values were further analyzed using Tukey Post Hoc tests and were highly significant between caries-free and caries-inactive groups. The caries-active group also showed significant difference with the caries-free group, same as above. However, the t 1/2 values did not show a significant difference between the caries-active and caries-inactive groups, same as above. A conspicuous decrease in the absorption intensities with incubation time was observed in all the samples from different caries-status groups [Figure 4]. We identified a statistically significant negative linear relationship between t1/2 and dmfs (t= -3.415, P=0.001) [Figure 5]. This means that the slope of the regression line is not zero and, hence, that t 1/2 is useful as a predictor of dmfs. Our prediction model of "dmfs=7.602 - 0.030 x t 1/2" means that for every 100 units' increase in t 1/2 , the dmfs decreases by 3 units (95% CI ranges from -4.8 to -1.3 units).{Table 1}{Table 2}{Figure 4}{Figure 5}

 Discussion



The development of an accurate practical caries prediction model would provide basic knowledge in relation to targeting based on need; economic efficiency and cost containment; appropriate care; and improved effectiveness of preventive therapies. The OSS utilized the variation in the absorption spectroscopy produced by the change in pH due to the interaction of acidogenic bacteria with sucrose. [15] This drop in pH will consequently lead to changes in the absorption property of the photosensitive pH indicator. Bromophenol blue was utilized as a photosensitive indicator in the sensor, which has a pKa value of 4 and is effective to monitor pH changes between 4 and 7.5. [20],[21] Gradual decreases in the intensity due to decreasing pH were observed with time in all the samples. It should be noted that the utilization of sucrose by the acidogenic bacteria is a chemical reaction where sucrose and bacteria are the reactants. More the number of bacteria, faster would be the utilization of sucrose. This reaction rate will be reflected as shorter half-life of exponential decay in the caries-active group as compared to the caries-inactive and caries-free groups.

In the present study, the caries-active, caries-inactive and caries-free groups showed different rates of sucrose utilization. Caries-inactive individuals who received restorative treatment few months before and properly maintained oral hygiene thereafter might have their bacterial counts closer to those of caries-free individuals. Whereas, caries-inactive individuals with recent history of restoration might still possess higher bacterial counts thus resembling caries-active individuals. Therefore, the treatment history plays an important role while comparing between caries-inactive individuals. A study by Hayes et al., reported that the acidogenic-ratio (Acidogenic: Non-acidogenic bacterial count) was 44% less in the caries-free group even though no significant difference existed in the total organisms between different caries-risk groups. [22] A highly significant relation was also observed between the acidogenic-ratio and the caries score of the caries-active group.

The extent and duration of pH drop after glucose rinse was found to be greater in cases of caries-active individuals than in caries-free or caries-inactive individuals. [8] This could be due to the higher proportion of acidogenic bacteria in caries-active individuals and has been related to the current caries-status of the individual. [6],[23] Positive numerical association of MS in stimulated saliva or dental plaque with dental caries served as an indicator of caries-risk prediction. [2],[4] In contrast, studies have also shown that caries occurred in individuals with no detectable counts of MS or LB and no caries was found in individuals with high MS and LB count. [12],[24] Previous studies have suggested that only 6% of the variation in caries prevalence can be explained by the variation of salivary MS. [24],[25] This variability in the direct association of the two main etiological odontopathogens with caries-status has led to the instability in the risk prediction. Besides, the association of MS with caries-status was found mainly applicable in the population level, while in the individual level the association was rather weak. [25]

The specificity and sensitivity values calculated using the t 1/2 values and caries scores for the OSS were 63% and 58% respectively. Specificity and sensitivity values are based on the comparison of a new tool with reference to a standard method, and must be applied to samples in a group in which the diagnosis is known. [26] However, the sensitivity and specificity values are not of great assistance to the clinician who seeks a definite answer about the given diagnostic test, whether it indicates presence or absence of disease. [27] For a chairside diagnostic tool it is more important for a clinician to determine the likelihood that a disease is present or absent, based on a given test result and is given by the predictive value. [26] In the present study, the positive predictive value of the OSS was found to be higher (68%) than the negative predictive value (53%).

In conclusion, the OSS used in this study was able to monitor the bacterial-mediated acidogenic-profile of saliva-sucrose interaction in real time. The OSS was able to show significant difference in the salivary acidogenic-profiles between patients of different caries-status. Longitudinal monitoring of salivary acidogenic-profile using OSS could help in monitoring the caries-risk in children. However, further studies with OSS need to be conducted in order to introduce this as a potential caries-risk assessment tool for a preventive approach.

References

1Thylstrup A, Fejerskov O. "Textbook of Cariology". chapter 1. Copenhagen: Munksgaard; 1986. p. 14.
2van Houte J. Microbiological predictors of caries risk. Adv Dent Res 1993;7:7-96.
3Beighton D, Hellyer PH, Lynch EJ, Heath MR. Salivary levels of mutans streptococci, lactobacilli, yeasts, and root caries prevalence in non- institutionalized elderly dental patients. Community Dent Oral Epidemiol 1991;19:302-7.
4van Houte J. Role of microorganisms in caries etiology. J Dent Res 1994;73:672-81.
5Main P, Azarpazhooh A. Risk assessment for the prevention of dental caries within the Children's Oral Health Initiative (COHI): Evidence-based report. Prepared for Dental and Pharmacy Programs, Primary Health Care and Public Health Directorate, First Nations and Inuit Health Branch (FNIHB), Health Canada; Mar 2007. p. 47.
6Loesche WJ. Role of Streptococcus mutans in human dental decay. Microbiol Rev 1986;50:353-80.
7Alaluusua S, Kleemola-Kujala E, Gronroos L, Evalahti M. Salivary caries-related tests as predictors of future caries increment in teenagers. A three-year longitudinal study. Oral Microbiol Immunol 1990;5:77-81.
8Stephan RM. Intra-oral hydrogen ion concentrations associated with dental caries activity. J Dent Res 1944;23:257-66.
9Alaluusua S, Savolainen J, Tuompo H, Gronroos L. Slide-scoring method for estimation of Streptococcus mutans levels in saliva. Scand J Dent Res 1984;92:127-33.
10Jensen B, Bratthall D. A new method for the estimation of mutans streptococci in human saliva. J Dent Res 1989;68:468-71.
11Park JH, Tanabe Y, Tinanoff N, Turng BF, Lilli H, Minah GE. Evaluation of microbiological screening systems using dental plaque specimens from young children aged 6-36 months. Caries Res 2006;40:277-80.
12Hildebrandt GH, Bretz WA. Comparison of culture media and chairside assays for enumerating mutans streptococci. J Appl Microbiol 2006;100:1339-47.
13John MS, Radhakrishnan P, Nampoori VP, Vallabhan CP. A fiber optic evanescent wave sensor for monitoring the rate of pulsed lased deposition of metal thin films. Meas Sci Technol 1999;10:N17-201.
14John MS, Kishen A, Singh LC, Asundi A. Determination of bacterial activity by use of an evanescent-wave fiber-optic sensor. Appl Opt 2002;41:7334-8.
15Kishen A, John MS, Lim CS, Asundi A. A fiber optic biosensor to monitor mutans streptococci in human saliva. Biosens Bioelectron 2003;18:1371-8.
16Kishen A. Fiber Optic Diagnostic Sensors; in Fundamental and Applications of Biophotonics in Dentistry. In: Kishen A, Asundi A, editors. London: Imperial College Press; 2007. p. 301-7.
17Preejith PV, Lim CS, Kishen A, John MS, Asundi A. Total Protein measurement using a fiber-optic evanescent wave-based biosensor. Biotechnol lett 2003;25:105-10.
18Wolfbeis OS. Fiber-optic chemical sensors and biosensors. Anal Chem 2006;78:3859-74.
19Ball DW. The basics of spectroscopy. Washington: SPIE Press; 2001. p. 53-74.
20Winans R, Brown CA. Fading of Bromophenol Blue: A combined synthesis and spectrophotometric kinetics experiment. J Chem Edu 1975;52:526-7.
21Gupta BD, Sharma DK. Evanescent wave absorption based fiber optic pH sensor prepared by dye doped sol-gel immobilization technique. Opt Commun 1997;140:32-5.
22Hayes ML, Carter EC, Griffiths SJ. The acidogenic microbial composition of dental plaque from caries-free and caries-prone people. Arch Oral Biol 1983;28:381-6.
23Dong YM, Pearce EI, Yue L, Larsen MJ, Gao XJ, Wang JD. Plaque pH and associated parameters in relation to caries. Caries Res 1999;33:428-36.
24Granath L, Cleaton-Jones P, Fatti LP, Grossman ES. Prevalence of dental caries in 4 to 5 year-old children partly explained by presence of salivary mutans streptococci. J Clin Microbiol 1993;31:66-70.
25Sullivan A, Granath L, Widenheim J. Correlation between child caries incidence and S. mutans/lactobacilli in saliva after correction for confounding factors. Community Dent Oral Epidemiol 1989;17:240-4.
26Flecthcer RW, Flecthcer FS. Clinical Epidemiology: The Essentials. 4 th ed. Philadelphia: Lippincott Williams and Wilkins; 2005.
27Spitalnic S. Test properties I: Sensitivity, specificity, and predictive values. Hosp Physician 2004;40:27-31.