Performance of Different Methods for Detection of Incipient Occlusal Caries Lesions: An In Vitro Study

Abstract

Objective: The aim of this in vitro study was to evaluate the diagnostic performance of visual inspection (VI), laser fluorescence (LF pen), fluorescence camera (FC), and alternating current impedance spectroscopy technique (ACIST) for detection of incipient occlusal caries lesions. Histology was used as the gold standard to determine the interexaminer reliability of the methods, among three examiners with different levels of experiences (Examiner 1: 5 years; examiner 2: 10 years; and examiner 3: 25 years). Materials and methods: One hundred two molar teeth were selected. The occlusal surfaces of teeth were evaluated with VI, LF pen, FC, and ACIST devices independently by three examiners. Diagnostic performances of methods were evaluated with binormal receiver-operating characteristics analysis. Interexaminer agreement of detection methods was assessed using Cohen's kappa coefficient values (p < 0.001). Results: Incipient enamel lesions were determined more successfully with VI by all examiners. All detection methods were presented with statistically acceptable interexaminer agreement (p < 0.001). For D₂ and D₃ thresholds, FC for examiner 1, VI for examiner 2, and both VI and FC for examiner 3 demonstrated statistically high sensitivity and specificity (p < 0.05). Conclusions: It can be concluded that diagnostic performance of the methods was insufficient for detecting incipient occlusal caries lesions by itself. Performance can be improved by using them with VI as a traditional caries detection method.

Introduction

Even in societies aware of oral healthcare, dental caries is still a crucial health problem worldwide.¹ In developing countries, it is reported that 60–90% of school-aged children and about 100% of the adult population are affected by dental caries.²

Dental caries is a chronic disease that shows slow progression. If untreated, it can lead to pain, infection, and even tooth loss.³ Therefore, the goal of modern dentistry is to detect caries lesions in early stages.⁴ The occlusal surface is more susceptible to caries because of its anatomy, tendency to retain plaque, and maturation. This complex anatomy, widespread use of fluoride, and superficial remineralization may allow caries development on macroscopically intact enamel surfaces.⁵ For this reason, it is important to detect noncavitated lesions, especially on occlusal surfaces at early stages.

The most commonly used occlusal caries detection method is visual inspection (VI).⁶ With this method, it is important to clean and isolate tooth surfaces with good illumination for decay detection. With VI, changes in color, transparency, and hardness of the dental tissue are observed. However, as VI is a subjective method, it has lower reproducibility in detecting occlusal caries, since it involves the knowledge and clinical experience of the examiner.⁶ The International Caries Detection and Assessment System (ICDAS) has been proposed as a visual scoring system for caries detection. The system describes caries lesions in six severity stages, varying from initial visible changes in enamel to frank cavitation in dentin.^7,8 This system allows clinicians and researchers to standardize the collected data and compare results with the literature. Previous in vitro studies with ICDAS have shown good results in detecting occlusal caries.^9,10

In recent years, several diagnostic methods such as laser fluorescence (LF) (DIAGNOdent and DIAGNOdent Pen), fluorescence camera (FC, VistaProof), and alternating current impedance spectroscopy technique (ACIST, CarieScan Pro) have been introduced for the detection of occlusal caries. These methods have been developed to obtain more reliable results with VI in the detection of occlusal caries.

LF (DIAGNOdent 2995; KaVo, Biberach, Germany) and LF pen (DIAGNOdent Pen 2190; KaVo) are chair-side portable devices used with LF to detect incipient caries. The fluorescence-based devices are designed to analyze and quantify fluorescence intensity of sound dental tissues and caries lesions.¹¹ While some of the studies about these systems have reported good performance for occlusal caries detection, some researchers have shown limitless and false-positive results.^11

–15 One of the most important disadvantages is that images of the tooth surface cannot be taken with the DIAGNOdent and DIAGNOdent Pen to keep the findings for longitudinal monitoring.¹⁶

In recent years, fluorescence camera (FC, VistaProof; Dürr Dental, Bietigheim-Bissingen, Germany) has been developed and presented to the market for use in early caries detection. This device also allows the clinician to document patient databases and follow the progression and location of caries. FC demonstrated good diagnostic performance for the detection of various stages of occlusal caries lesions.^11,16 Data in the literature for FC are not quite adequate.

Another method used for occlusal caries detection is ACIST (CarieScan Pro; CarieScan Ltd., Dundee, Scotland). This method measures the mineral density with low-voltage current applied directly through the tooth.¹⁷ The system has been designed to determine precavitated lesions as represented by ICDAS code 1 and 2. The literature contains few studies about the CarieScan Pro and its performance. However, in one pilot study, the device showed good accuracy for detection of precavitated lesions.¹⁸

The aim of this in vitro study was to investigate the diagnostic performance of VI, LF pen (DIAGNOdent Pen), FC (VistaProof), and ACIST (CarieScan Pro) for incipient occlusal caries lesions using histological validation as gold standard and determine the interexaminer reliability of the methods with three examiners.

Materials and Methods

Sample preparation

The research protocol was approved by the Local Ethics Committee in Ankara University, Faculty of Dentistry, Ankara, Turkey (361906000/74). One hundred two unstained, visually sound and noncavitated, freshly extracted permanent human molar teeth were selected upon VI under 2.5 × magnification by a senior researcher not involved in the study. All teeth had been extracted by dental practitioners in the Oral and Maxillofacial Surgery Department at Gazi University. Before extraction, the patients were informed about the use of their teeth for research purposes, and their consent was obtained.

After extraction, teeth were stored in 100% humidity distilled water, as storage methods can alter fluorescence measurements. Teeth with cavitated caries lesions, fissure sealant and restoration, hypoplasia or hypomineralization were excluded. Debris, calculus, and discolorations were removed using a scaler and a rubber cup. Then, the occlusal surfaces were photographed with a camera (Canon EOS 550D, Japan) at 10 × magnification and distance of 30 cm. One spot on each tooth was selected by a senior researcher, who did not take part in the examinations (test site). All assessments were performed independently by three examiners with different degrees of clinical experience (Examiner 1: 5 years; examiner 2: 10 years; examiner 3: 25 years). To reduce prejudice, all examiners were trained and calibrated using ICDAS-II with photographic examples, for the VI. All examiners were trained and calibrated by manufacturers' instructions for other detection methods.

Visual inspection—ICDAS

VI was performed in adherence to ICDAS-II criteria. The test sites were examined in the same room under the same bluish white spectrum lamp (KaVo Primus 1058; KaVo, Biberach/Riss, Germany) and air-drying for 5 sec, with observation distance of 30 cm, without probing. VI was coded as: (0) sound tooth surface; (1) first visual changes in enamel, opacity, and discoloration not or barely seen on a wet surface; (2) distinct visual changes in moist enamel and lesion that are still visible when the tooth is dry; (3) localized enamel breakdown due to caries with no visible dentin or underlying shadows; (4) underlying dark shadows in dentin with or without localized enamel breakdown; (5) distinct cavity with visible dentin and visual evidence of demineralization; and (6) extensive distinct cavity with visible dentin.

Laser fluorescence pen—DIAGNOdent Pen

For LF measurement, LF pen (DIAGNOdent Pen 2190; KaVo) was used according to the manufacturer's instructions, using the cylindrical tip designed for occlusal surfaces after calibration with the ceramic standard. After positioning the tip of the device perpendicularly to the tooth, it was rotated along its long axis. The tool then provides a numerical value ranging from 0 to 99. The measurement was repeated three times and the highest reading was recorded as the tooth's LF score.

Fluorescence camera—VistaProof

FC measurements were performed with VistaProof (VistaProof; Dürr Dental) in a black box to simulate the oral environment. Occlusal surfaces were dried for 5 sec with mild air before examination. The FC was placed parallel to the occlusal surface, and images of the surface were taken. The images were analyzed with FC-specific software (DBSWINN, Version 5.3; Dürr Dental), which translates the green and red rate of fluorescence to numbers that represent lesion severity. The numerical values observed in tooth images were recorded as the tooth's FC score.

Alternating current impedance spectroscopy—CarieScan Pro

Before using the CarieScan Pro (CarieScan Pro; CarieScan Ltd.), the device was calibrated manually according to the manufacturer's instructions. Each tooth was rinsed in tap water for 5 sec and gently dried with an air syringe, leaving the root wet. Then, teeth were held by the roots with the ungloved hand (left), and the device was held by a glove-covered hand (right) to complete the electric circuit. The sensor tip was applied into the fissures without pressure or movement during evaluation. Measurement was repeated five times and the mean score was recorded as the tooth's CarieScan Pro score.

Histological validation

After all evaluations had been completed, teeth were embedded in acrylic blocks with the occlusal surface exposed. Each tooth was hemisectioned in a mesiodistal direction through the spot where caries was previously suspected using a 1 mm thick diamond saw under water cooling (Mecatome T201A; Presi, Grenoble, France). Histological sections were viewed under a stereomicroscope (Olympus SZ 60, Japan) at 10 × magnification. Histological validation was performed by an experienced researcher, who did not take part in any stage of the experiments and assigned scores according to Downer¹⁹: (0) no demineralization; (1) demineralization extending to the outer half of the enamel; (2) demineralization extending to the inner half of the enamel; (3) demineralization extending to the outer half of the dentin; and (4) demineralization extending to the inner half of the dentin.

Statistical analysis

Statistical analysis was performed with SPSS software for Windows, version 21.0 (IBM Corp., Armonk, NY).

The efficiency of diagnostic methods in identifying incipient enamel caries lesions was assessed as a percentage distribution. Cohen's kappa coefficient values were used to evaluate the interexaminer agreement of diagnostic methods.²⁰

Sensitivity, specificity, accuracy, and area under the receiver-operating characteristics (ROC) curve were calculated for D₁, D₂, and D₃ diagnostic thresholds. These thresholds for all detection methods were determined by the cutoff points indicated by manufacturers (Table 1).

Table 1.

Distribution of the Histological Scores and Manufacturer Cutoff Values of Detection Methods According to the Diagnostic Thresholds

	Detection methods and manufacturer cutoff points
Histological scores	VI	LF	FC	ACIST
0	0	0–13	0–0.9	0 (green)
D₁
1	1	14–20	1–1.4	1–30 (green–yellow)
D₂
2	2	21–30	1.5–1.9	31–90 (yellow)
D₃
3	3-4-5-6	>30	≥2	91–100 (red)

D₁ threshold: 0 = sound, 1, 2, 3 = decayed. D₂ threshold: 0, 1 = sound, 2, 3 = decayed. D₃ threshold: 0, 1, 2 = sound, 3 = decayed.

ACIST, alternating current impedance spectroscopy technique; FC, fluorescence camera; LF, laser fluorescence; VI, visual inspection.

At the D₁ diagnostic threshold, histological scores 1–4 were classified as caries. At the D₂ diagnostic threshold, histological scores 2–4 were classified as caries. At the D₃ diagnostic threshold, histological scores 3 and 4 were classified as caries only. Using these sensitivity and specificity values, binormal ROC analyses were performed at the D₁, D₂, and D₃ thresholds for each examiner. These calculations were obtained at the cutoff values defined by the manufacturers. The area under the ROC curve (AUC) was determined using this classification²¹: 0.60–0.75, fair; 0.75–0.90, good; 0.90–0.97, very good; and 0.97–1.00, excellent. The significance level was set at α = 0.05.

Results

Histological validation of 102 occlusal sites revealed that 5 teeth (4.90%) were free of caries (0), 27 teeth (26.47%) had caries extending over the outer half of the enamel (1), 42 teeth (41.17%) had caries extending over the inner half of the enamel (2), and 28 teeth (27.45) had caries extending throughout the dentin (3–4).

Table 2 shows the percentage of incipient occlusal caries lesions detected correctly by means of diagnostic methods. Incipient occlusal caries lesions were detected successfully with VI by all examiners. LF pen and ACIST did not perform effectively in detecting the enamel lesions recommended by manufacturers.

Table 2.

Percentage Distribution of Incipient Enamel Lesions Detected By Three Examiners with All Diagnostic Methods

	Incipient enamel lesions (n = 27)
Diagnostic methods	Examiner 1, n (%)	Examiner 2, n (%)	Examiner 3, n (%)
VI	16 (59.3)	15 (55.6)	18 (66.7)
LF	1 (3.7)	1 (3.7)	1 (3.7)
FC	10 (37.0)	4 (14.8)	12 (44.4)
ACIST	0 (0.0)	3 (11.1)	3 (11.1)

Table 3 shows the Cohen's kappa coefficient values for interexaminer agreement. VI showed good agreement among examiners 1, 2, and 3. FC showed substantial agreement between examiners 1 and 2. ACIST showed substantial agreement between examiners 2 and 3.

Table 3.

Cohen's Kappa Values for Interexaminer Reproducibility of All Diagnostic Methods

Interexaminer agreement	Diagnostic methods	95% confidence interval	Cohen's kappa values (standard deviation)
Examiner 1-Examiner 2	VI	0.267–0.567	0.417 (0.077)
	LF	0.297–0.734	0.516 (0.111)
	FC	0.472–0.756	0.614 (0.072)
	ACIST	0.329–0.632	0.480 (0.077)
Examiner 1-Examiner 3	VI	0.315–0.592	0.453 (0.071)
	LF	0.300–0.773	0.537 (0.121)
	FC	0.341–0.700	0.520 (0.092)
	ACIST	0.299–0.602	0.450 (0.077)
Examiner 2-Examiner 3	VI	0.102–0.360	0.231 (0.066)
	LF	0.311–0.730	0.520 (0.107)
	FC	0.136–0.559	0.347 (0.108)
	ACIST	0.472–0.756	0.614 (0.072)

Table 4 shows sensitivity, specificity, accuracy, and AUC for VI, LF pen, FC, and ACIST according to the cutoffs determined by the manufacturers at D₁, D₂, and D₃ thresholds. At the D₁ threshold, any diagnostic method used by any examiner showed statistically significant results (p > 0.05). FC exhibited high sensitivity for both D₂ and D₃ thresholds for examiner 1. VI showed good specificity for D₂ and D₃ thresholds for examiner 2. For examiner 3, the AUC for both VI and FC were statistically significant at D₂ and D₃ thresholds. At the D₂ threshold, while VI showed moderate sensitivity and specificity, FC showed high sensitivity and moderate specificity. At the D₃ threshold, both VI and FC exhibited high sensitivity.

Table 4.

Sensitivity, Specificity, Accuracy, and AUC Values for VI, LF pen, FC, and ACIST for All Examiners at All Thresholds

Diagnostic methods	AUC	p	Sensitivity (%)	Specificity (%)	Accuracy (%)
D1
VI
Examiner 1	0.357	>0.05	—	—	—
Examiner 2	0.558	>0.05	—	—	—
Examiner 3	^a	—	—	—	—
LF pen
Examiner 1	^a	—	—	—	—
Examiner 2	0.366	>0.05	—	—	—
Examiner 3	^a	—	—	—	—
FC
Examiner 1	^a	—	—	—	—
Examiner 2	^a	—	—	—	—
Examiner 3	0.751	>0.05	—	—	—
ACIST
Examiner 1	0.720	>0.05	—	—	—
Examiner 2	^a	—	—	—	—
Examiner 3	^a	—	—	—	—
D2
VI
Examiner 1	0.617	>0.05	—	—	—
Examiner 2	0.660	<0.05	35.71	84.34	50.98
Examiner 3	0.676	<0.05	68.57	62.50	66.67
LF pen
Examiner 1	0.492	>0.05	—	—	—
Examiner 2	0.506	>0.05	—	—	—
Examiner 3	0.641	>0.05	—	—	—
FC
Examiner 1	0.732	<0.05	81.43	46.88	70.59
Examiner 2	0.685	>0.05	—	—	—
Examiner 3	0.802	<0.05	84.29	53.13	74.51
ACIST
Examiner 1	0.601	>0.05	—	—	—
Examiner 2	0.591	>0.05	—	—	—
Examiner 3	0.615	>0.05	—	—	—
D3
VI
Examiner 1	0.619	>0.05	—	—	—
Examiner 2	0.640	<0.05	39.29	74.32	64.71
Examiner 3	0.748	<0.05	89.29	52.70	62.75
LF pen
Examiner 1	0.688	>0.05	—	—	—
Examiner 2	0.447	>0.05	—	—	—
Examiner 3	0.677	>0.05	—	—	—
FC
Examiner 1	0.679	<0.05	85.71	32.43	47.06
Examiner 2	0.603	>0.05	—	—	—
Examiner 3	0.714	<0.05	85.71	32.43	47.06
ACIST
Examiner 1	0.622	>0.05	—	—	—
Examiner 2	0.559	>0.05	—	—	—
Examiner 3	0.554	>0.05	—	—	—

Calculation cannot be performed because there are not enough data in the defined range for ROC analysis.

AUC, area under the ROC curves; ROC, receiver-operating characteristics.

Significant values are indicated in bold.

Discussion

Early detection of noncavitated caries lesions is crucial in the management of dental caries, as early-stage caries lesions can be treated preventively, with minimally invasive techniques. Unidentified noncavitated caries lesions can develop under macroscopically intact enamel surfaces; the accurate and early detection of occlusal caries lesions is necessary to arrest caries progression. Therefore, several methods have been introduced for the diagnosis of caries lesions.^22
–24 This in vitro study investigated the performance of three methods (LF pen, FC, and ACIST with histological validation as gold standard) and compared them with VI, as a traditional method for detecting incipient occlusal caries lesions.

In the literature, there are many in vivo and in vitro studies investigating the performance of various methods for caries detection.^25,26 The conditions for detecting occlusal caries lesions are more suited to in vitro than in vivo studies because of the absence of stain, biofilm, bacterial plaque, and saliva.^14,27 For in vitro occlusal caries detection studies, histological evaluation of tooth sections under microscopy is the most commonly used gold (reference) standard technique.¹⁹ In addition, laboratory studies allow histopathological validation for precise evaluation of lesion depth.^28,29 The concordance between histological validation and study results is important to evaluate the performance of diagnostic methods.¹¹

Performing VI according to ICDAS criteria is a traditional method for caries detection that provides details on the phase of caries lesion development. Previous studies showed good results for ICDAS: acceptable prediction of caries depth and correlation with histological validation values.^6,30

In this study, incipient occlusal caries lesions were detected correctly with VI by all examiners. In an in vitro study, it was reported that VI was more effective than LF methods in identifying incipient caries lesions.³¹

LF pen and FC are two methods based on the fluorescence measurement of tooth structures for the detection of early-stage caries lesions. Previous studies have shown that these devices may help to complete the diagnosis, especially in cases of hidden caries.^14,32 The results of these studies were influenced by several factors such as calculus, plaque, stains, fluorescence-based prophylactic paste, and remineralization.^33
–35 Such alterations may increase the number of false-positive results. Therefore, cleaning the teeth (including the deepest pits and fissures) of deposits before fluorescence-based evaluation is essential.

In several studies evaluating the efficacy of LF pen in the accurate detection of occlusal caries, LF pen performed better than traditional methods such as VI and radiographic examination.^13,36,37 In addition, some other studies reported that these fluorescence-based devices should be used as adjuncts to traditional methods in clinical practice.^31,38

While studies use various thresholds for describing the stages of caries, some of which increase the performance of LF pen, it is not always possible to compare study results.²² Therefore, we used the cutoff limits identified by the manufacturer to standardize evaluation in the present study.

FC is a hand-held intraoral camera, which allows teeth to be captured two-dimensionally, and provides a fluorescence caries score. It uses blue light to detect porphyrin activity and was shown to be more effective than the red light used by the LF devices (DIAGNOdent and DIAGNOdent Pen) for detecting early enamel lesions.³⁹ FC readings are influenced by the surrounding light and may change the results. For this reason, a black cap is provided with the machine; the unit's light should be switched off during clinical evaluations. A black box must be used to simulate the illumination of the oral environment for in vitro experiments.⁴⁰ As recommended by the manufacturer, a protective plastic cover and black box were used in this study to imitate intraoral conditions.⁴¹

Few published studies have used FC-based devices for the detection of occlusal caries lesions. In one study, fluorescence intensity increased with the use of blue (488 nm) light compared with red light (655 nm) (DIAGNOdent and DIAGNOdent Pen).³⁹ Red light is thought to be more effective in detecting dentin caries than incipient enamel lesions.⁴⁰ Therefore, it can be concluded that blue light may be more efficient than red light in detecting early-stage caries lesions.

In an in vitro study, results obtained with FC and histological validation scores showed weak correlation.¹¹ This finding is in agreement with the results of our study. However, the finding may reflect the particular teeth included in the study and the experience of the examiners with the devices used.

CarieScan Pro is a new device based on ACIST that is used to characterize early and noncavitated stages of caries lesion development.⁴² There are limited published data about the in vitro performance of this device.^3,17,42

The in vitro studies performed with ACIST must be performed with caution. The storage media, temperature, dryness of the tooth, and electrical conductivity may have affected the method's performance.^3,43 To prevent these limitations, we gently dried the tooth, held the tooth by the roots with ungloved and moist fingers to facilitate electrical conductance and changed the sensor after every 20 measurements, as recommended by the manufacturer to avoid reducing sensitivity. However, the measurements may have been affected by the efficiency of the sensor or overdrying of the extracted teeth.

The results obtained with ACIST in the present study were not so encouraging, as diagnostic performance was significantly lower than observed with the fluorescence system. For a device designed to evaluate the initial stages of caries lesions, the lack of performance was of concern. Teo et al. also reported poor performance for ACIST.⁴⁴

One of the most important obligations of the ideal diagnostic method for occlusal caries detection is good interexaminer agreement, because measurements and results must be reproducible. Cohen's kappa values are used to evaluate the intra- and interexaminer reliability of diagnostic methods.²⁰ In our study, all methods had different kappa values. For VI, moderate agreement was found between examiners 1 and 2 and between examiners 1 and 3. An in vitro study by Rodrigues et al., using ICDAS-II criteria at occlusal surfaces confirmed the interexaminer reliability reported in this study.¹¹ These different stages of agreement by examiners can be correlated with the clinical experiences of examiners and the heterogeneous distribution of caries stages in the teeth involved in the study.

ROC analysis is a statistical method that allows researchers and clinicians to evaluate the sensitivity and specificity for all thresholds used to characterize the diagnostic utility of a certain method.⁴⁵ In this in vitro study, the cutoff points determined by the manufacturer's instructions were used. For all examiners, with the D₁ threshold, no method yielded statistically significant values. This may reflect the fact that, with histological validation, the numbers of sound teeth were inadequate. The D1 threshold is used to identify the effectiveness of methods in characterizing intact surfaces as well as those with caries.

The D₂ threshold is used to determine the sensitivity, specificity, and accuracy of methods used to detect caries in enamel. At the D₂ threshold, the ICDAS-II had higher specificity and lower sensitivity for examiner 2 and moderate sensitivity and specificity for examiner 3; FC had high sensitivity and low specificity for examiners 1 and 3. These findings suggest that these methods are sufficient for detecting changes in enamel surfaces. Betrisey et al. compared FC and LF for detecting incipient occlusal caries lesions; FC showed higher sensitivity values as observed in the present study.⁴⁰ This result could be explained by the high effectiveness of blue light, compared with the red light used for LF pen, in characterizing early-stage enamel lesions.³⁹

At the D₃ threshold, ICDAS-II for examiner 2 and FC for examiners 1 and 3 presented high sensitivity and low specificity. Based on these findings, both methods are successful for detecting lesions in dentin. In two other in vitro studies, researchers reported similar results for the detection of dentin caries.^11,32 In contrast, researchers in an in vitro study reported higher specificity than sensitivity.¹⁰ These differences could be explained by the examiners' clinical experience and the distribution of caries stages in the teeth investigated in this study.

Within the limitation of this in vitro study design, it can be concluded that VI and FC were more effective in detecting enamel and dentin caries lesions. It is important to point out that, although all methods evaluated in this in vitro study have advantages, they also have limitations. These diagnostic methods can be considered as adjunct devices that can be used with traditional methods such as VI. To obtain results that will be useful for clinical practice, further in vitro and in vivo studies are needed to investigate the possible use of FC and ACIST during caries excavation and for detecting caries lesions.

Footnotes

Author Disclosure Statement

No competing financial interests exist.

References

de Paula

, Campos

, Diniz

, Hebling

, Rodrigues

. In situ and in vitro comparison of laser fluorescence with visual inspection in detecting occlusal caries lesions. Lasers Med Sci, 2011; 26:1–5.

Petersen

, Bourgeois

, Ogawa

, Estupinan-Day

, Ndiaye

. The global burden of oral diseases and risks to oral health. Bull World Health Organ, 2005; 83:661–669.

Mortensen

, Dannemand

, Twetman

, Keller

. Detection of non-cavitated occlusal caries with impedance spectroscopy and laser fluorescence: an in vitro study. Open Dent J, 2014; 8:28–32.

Pitts

. Are we ready to move from operative to non-operative/preventive treatment of dental caries in clinical practice?. Caries Res, 2004; 38:294–304.

Ricketts

, Kidd

, Smith

, Wilson

. Clinical and radiographic diagnosis of occlusal caries: a study in vitro. J Oral Rehabil, 1995; 22:15–20.

Souza

, Boldieri

, Diniz

, Rodrigues

, Lussi

, Cordeiro

. Traditional and novel methods for occlusal caries detection: performance on primary teeth. Lasers Med Sci, 2013; 28:287–295.

Zandona

, Zero

. Diagnostic tools for early caries detection. J Am Dent Assoc, 2006; 137:1675–1684; quiz 730.

Ismail

, Sohn

, Tellez

, et al. The International Caries Detection and Assessment System (ICDAS): an integrated system for measuring dental caries. Community Dent Oral Epidemiol, 2007; 35:170–178.

Diniz

, Rodrigues

, Hug

, Cordeiro Rde

, Lussi

. Reproducibility and accuracy of the ICDAS-II for occlusal caries detection. Community Dent Oral Epidemiol, 2009; 37:399–404.

10.

Jablonski-Momeni

, Stachniss

, Ricketts

, Heinzel-Gutenbrunner

, Pieper

. Reproducibility and accuracy of the ICDAS-II for detection of occlusal caries in vitro. Caries Res, 2008; 42:79–87.

11.

Rodrigues

, Hug

, Diniz

, Lussi

. Performance of fluorescence methods, radiographic examination and ICDAS II on occlusal surfaces in vitro. Caries Res, 2008; 42:297–304.

12.

Akarsu

, Koprulu

. In vivo comparison of the efficacy of DIAGNOdent by visual inspection and radiographic diagnostic techniques in the diagnosis of occlusal caries. J Clin Dent, 2006; 17:53–58.

13.

Neuhaus

, Rodrigues

, Hug

, Stich

, Lussi

. Performance of laser fluorescence devices, visual and radiographic examination for the detection of occlusal caries in primary molars. Clin Oral Investig, 2011; 15:635–641.

14.

Rodrigues

, Diniz

, Josgrilberg

, Cordeiro

. In vitro comparison of laser fluorescence performance with visual examination for detection of occlusal caries in permanent and primary molars. Lasers Med Sci, 2009; 24:501–506.

15.

Lussi

, Francescut

. Performance of conventional and new methods for the detection of occlusal caries in deciduous teeth. Caries Res, 2003; 37:2–7.

16.

Jablonski-Momeni

, Schipper

, Rosen

, et al. Performance of a fluorescence camera for detection of occlusal caries in vitro. Odontology, 2011; 99:55–61.

17.

Huysmans

, Longbottom

, Pitts

, Los

, Bruce

. Impedance spectroscopy of teeth with and without approximal caries lesions—an in vitro study. J Dent Res, 1996; 75:1871–1878.

18.

Pitts

, Longbottom

, Czajczynska-Waszkiewicz

, et al. Diagnostic accuracy of an optimised ac impedance device to aid caries detection and monitoring. Caries Res, 2008; 42:211–212.

19.

Downer

. Concurrent validity of an epidemiological diagnostic system for caries with the histological appearance of extracted teeth as validating criterion. Caries Res, 1975; 9:231–246.

20.

Landis

, Koch

. The measurement of observer agreement for categorical data. Biometrics, 1977; 33:159–174.

21.

Swets

. Measuring the accuracy of diagnostic systems. Science, 1988; 240:1285–1293.

22.

Aktan

, Cebe

, Ciftci

, Sirin Karaarslan

. A novel LED-based device for occlusal caries detection. Lasers Med Sci, 2012; 27:1157–1163.

23.

Diniz

, Rodrigues

, de Paula

, Cordeiro Rde

. In vivo evaluation of laser fluorescence performance using different cut-off limits for occlusal caries detection. Lasers Med Sci, 2009; 24:295–300.

24.

Lussi

, Hellwig

. Performance of a new laser fluorescence device for the detection of occlusal caries in vitro. J Dent, 2006; 34:467–471.

25.

Nyvad

. Diagnosis versus detection of caries. Caries Res, 2004; 38:192–198.

26.

Ricketts

, Kidd

, Wilson

. A re-evaluation of electrical resistance measurements for the diagnosis of occlusal caries. Br Dent J, 1995; 178:11–17.

27.

Lussi

, Megert

, Longbottom

, Reich

, Francescut

. Clinical performance of a laser fluorescence device for detection of occlusal caries lesions. Eur J Oral Sci, 2001; 109:14–19.

28.

Bader

, Shugars

, Bonito

. A systematic review of the performance of methods for identifying carious lesions. J Public Health Dent, 2002; 62:201–213.

29.

Ekstrand

, Ricketts

, Kidd

. Reproducibility and accuracy of three methods for assessment of demineralization depth of the occlusal surface: an in vitro examination. Caries Res, 1997; 31:224–231.

30.

Kuhnisch

, Berger

, Goddon

, Senkel

, Pitts

, Heinrich-Weltzien

. Occlusal caries detection in permanent molars according to WHO basic methods, ICDAS II and laser fluorescence measurements. Community Dent Oral Epidemiol, 2008; 36:475–484.

31.

Attrill

, Ashley

. Occlusal caries detection in primary teeth: a comparison of DIAGNOdent with conventional methods. Br Dent J, 2001; 190:440–443.

32.

Ricketts

, Kidd

, Weerheijm

, de Soet

. Hidden caries: what is it? Does it exist? Does it matter?. Int Dent J, 1997; 47:259–265.

33.

Diniz

, Sciasci

, Rodrigues

, Lussi

, Cordeiro

. Influence of different professional prophylactic methods on fluorescence measurements for detection of occlusal caries. Caries Res, 2011; 45:264–268.

34.

Lussi

, Reich

. The influence of toothpastes and prophylaxis pastes on fluorescence measurements for caries detection in vitro. Eur J Oral Sci, 2005; 113:141–144.

35.

Lussi

, Longbottom

, Gygax

, Braig

. Influence of professional cleaning and drying of occlusal surfaces on laser fluorescence in vivo. Caries Res, 2005; 39:284–286.

36.

Baseren

, Gokalp

. Validity of a laser fluorescence system (DIAGNOdent) for detection of occlusal caries in third molars: an in vitro study. J Oral Rehabil, 2003; 30:1190–1194.

37.

Sridhar

, Tandon

, Rao

. A comparative evaluation of DIAGNOdent with visual and radiography for detection of occlusal caries: an in vitro study. Indian J Dent Res, 2009; 20:326–331.

38.

Huth

, Neuhaus

, Gygax

, et al. Clinical performance of a new laser fluorescence device for detection of occlusal caries lesions in permanent molars. J Dent, 2008; 36:1033–1040.

39.

Hibst

, Paulus

, Lussi

. Detection of occlusal caries by laser fluorescence: basic and clinical investigations. Med Laser Appl, 2001; 16:205–213.

40.

Betrisey

, Rizcalla

, Krejci

, Ardu

. Caries diagnosis using light fluorescence devices: VistaProof and DIAGNOdent. Odontology, 2014; 102:330–335.

41.

De Benedetto

, Morais

, Novaes

, de Almeida Rodrigues

, Braga

, Mendes

. Comparing the reliability of a new fluorescence camera with conventional laser fluorescence devices in detecting caries lesions in occlusal and smooth surfaces of primary teeth. Lasers Med Sci, 2011; 26:157–162.

42.

Longbottom

, Huysmans

, Pitts

, Los

, Bruce

. Detection of dental decay and its extent using a.c. impedance spectroscopy. Nat Med, 1996; 2:235–237.

43.

Francescut

, Zimmerli

, Lussi

. Influence of different storage methods on laser fluorescence values: a two-year study. Caries Res, 2006; 40:181–185.

44.

Teo

, Ashley

, Louca

. An in vivo and in vitro investigation of the use of ICDAS, DIAGNOdent pen and CarieScan PRO for the detection and assessment of occlusal caries in primary molar teeth. Clin Oral Investig, 2014; 18:737–744.

45.

Obuchowski

. Receiver operating characteristic curves and their use in radiology. Radiology, 2003; 229:3–8.