The International Caries Detection and Assessment System - ICDAS: A Systematic Review.

Ekstrand KR, Gimenez T, Ferreira FR, Mendes FM, Braga MM.

The aims of this study were: (1) to evaluate the overall reproducibility and accuracy of the International Caries Detection and Assessment System (ICDAS) for assessing coronal caries lesions, and (2) to investigate the use of systems associated with the ICDAS for activity assessment of coronal caries lesions. Specific search strategies were adopted to identify studies published up to 2016. For the first objective, we selected studies that assessed primary coronal caries lesions using the ICDAS as a reference standard. A total of 54 studies were included. Meta-analyses summarized the results concerning reproducibility and accuracy (correlation with histology, summary ROC curves [SROC], and diagnostic odds ratio [DOR]). The latter 2 were expressed at D1/D3 levels. The heterogeneity of the studies was also assessed. Reproducibility values (pooled) were > 0.65. The ICDAS mostly presented a good overall performance as most areas under SROC were > 0.75 at D1 and > 0.90 at D3; DOR ≥6. For the second objective, we selected studies investigating activity assessment associated with the ICDAS. The meta-analyses pooled results based on the same methodology and parameters as above. Longitudinal findings regarding caries progression were described to estimate the validity of these systems. On average, the systems for activity assessment of caries lesions showed moderate values concerning reproducibility and overall performance. Active caries lesions were more prone to progress than inactive ones after 2 years. In conclusion, the ICDAS presented a substantial level of reproducibility and accuracy for assessing primary coronal caries lesions. Additional systems associated with the ICDAS that classify caries lesion activity can be useful as they are moderately reproducible and accurate.

Caries Res. 2018 Mar 8;52(5):406-419

Caries status in young Colombian children expressed by the ICCMS™ visual/radiographic combined caries staging system.

Cortes A, Ekstrand KR, Gamboa LF, González L, Martignon S.

The objective of this study was to report (1) the caries experience prevalence and mean, and the caries severity and distribution patterns, expressed clinically and combined with radiographs with the conventional and ICCMS™ systems in young children from Bogotá, Colombia; (2) the contribution of including radiographs to the clinical caries scoring and (3) in which surfaces the radiograph adds to the clinical caries registration. The study concluded that participants' caries experience was high. The radiographic assessment significantly contributed to caries experience. Molar and upper incisor teeth were most prone to caries.

Acta Odontol Scand. 2017 Jan;75(1):12-20

* Caries Research (Abstracts reprinted with permission of S.Karger AG, Basel)

Abstracts of the 52^nd Annual Orca Congress - Reprinted from "Caries Research" volume 39 (4) 5*

Plaque and Caries Status of Colombian Patients with Fixed Orthodontic Appliances: Visual Examination Using the OPI and ICDAS Criteria

S. Martignon a, b , K.R. Ekstrand b , M.I. Lemos a, *, M.C. Rangel a , M.P. Lozano a , C. Higuera 
a * lemosmaria@unbosque.edu.co a Universidad El Bosque, Bogotá, Colombia; 
b University of Copenhagen, Copenhagen, Denmark

The aim of this study was to assess the plaque and caries status on patients with fixed-orthodontic appliance treated for one year at the Dental Clinic, Universidad El Bosque, Bogota. Seventy-four 12–29-year-olds agreed to participate, corresponding to 79% of patients’ total number. One examiner scored as follows: Ortho-plaque- Index (OPI) [Heintze et al: Quintessence 1999, Hong-Kong, pp 67–70] on buccal surfaces: (1) mesial and distal from bracket; (2) cervical third; (3) occlusal third; OPI figures were expressed per patient: values from 0–30% indicates good, 31–50% fair and 51– 100% poor oral hygiene. Caries ICDAS-scores were: 0 sound; 1B/1W brown/white opacity after air-drying; 2B/2W (brown/white) opacity without air-drying; 3 underlying shadow; 4 surface integrity loss; 5 distinct, and 6 extensive cavity. Filled or missing-dueto- caries surfaces were also scored. Caries on buccal surfaces was further recorded at 3 sites, corresponding to OPI. The caries scoring intra-reproducibility was substantial (kappa = 0.80). Only 20% of the patients had a good oral hygiene status, 65% fair and 15% poor. The traditional mean DMF-S (D-component scores 1 4) was 6.6 8 6.3, of which 5.00 8 4.7 were fi lled and 1.1 8 2.7 missing. When lesions scored 1–4 were added, the fi gure increased to 23.6 8 9.4, 11.3 8 7.2 being white opacities. All participants had 6 1 and 50% 6 17 caries-lesions scored 1–6; 96% had 6 1; 67% 6 7; 50% 6 11, and 15% 6 20 white opacities (1W/2W). The buccal surfaces accounted for most white opacities. Thus, 37% of the lower canines, about 25% of the upper canines, lateral and central upper incisors, second lower premolars and molars had white opacities. Close to one-third of white opacities on buccal surfaces of upper anterior teeth were located around the brackets. The results showed a high incidence of white opacities related to orthodontic appliances, which indicate an urgent need to implement preventive programmes in this dental clinic.  

8* Assessment of the Oral Health Status and Treatment Needs of Institutionalized Children with HIV/AIDS in Uganda

B.T. Amaechi b , I. Mugisa a, *, R. Ndawula a , P. Lucaci a , G.G. Kugonza a , A. Mwanika

a * mugisaian@hotmail.com a Makerere University Dental School, Kampala, Uganda; 
b University of Texas Health Sciences Center at San Antonio, San Antonio, Tex., USA 

This pilot study evaluated and compared the oral health and treatment need of institutionalized children with vertically transmitted HIV with that of uninfected children. A point prevalence study was conducted in 100 children, 50 HIV/AIDS-infected on regular antiretroviral syrups and 50 uninfected, aged between 5 and 18 years living in two respective orphanages located in the same district. Children under 6 years of age were assessed using the NIDCR criteria for Early Childhood Caries, while the WHO assessment criteria were used for children aged 6 years and above. Results showed the mean DMFT and dft in infected children to be 0.17 8 0.23 and 0.24 8 0.22, respectively. Mann-Whitney U test indicated significant differences in both the DMFT (p = 0.048) and dft (p = 0.001) of the infected children when compared with those of uninfected children, with mean DMFT and dft of 0.06 8 0.07 and 0.01 8 0.04, respectively. Pearson’s _2 test indicated no significant difference in treatment need between the infected and uninfected groups, with 90% of infected and 60% of uninfected requiring treatment. 38% of the infected and 76% of the uninfected children was caries-free. It was concluded that the caries prevalence of infected children is higher than that of uninfected populations in similar living conditions. The caries prevalences for both groups, based on NIDCR and WHO criteria, were low. 

23* Analysis of QLF and ICDAS Measurements in a Clinical Trial

G.J. Eckert a ,*, B.P. Katz a , S. Ofner a , A.G. Ferreira Zandoná b , H. Eggertsson b , T. Doi b , G.K. Stookey b , J.S. Wefel c * geckert@iupui.edu

a Division of Biostatistics, 
b Department of Preventive and Community Dentistry, Indiana University, Indianapolis, Ind., 
c Dows Institute for Dental Research, University of Iowa, Iowa City, Iowa, USA

Traditionally, DMFS scores have been used for caries assessment in clinical trials, with scores calculated for each subject and averaged within a group to perform comparisons. With the development and use of new early caries detection methods comes the need to determine how data should be summarized and statistically analyzed. It is fi rst necessary to decide whether the primary outcome should include all sites or should be restricted to sites with caries at baseline, because the interpretation of results of these outcomes may be different. When following only specific sites, changes in the caries assessments are calculated for each site within a subject, but data can be analyzed by summing changes across sites, averaging changes across sites, or using individual site changes with a statistical method that accounts for within-subject correlations. Similar statistics can be calculated when following all sites regardless of baseline status. And, following directly from a traditional decayed surface count, new detection methods can be used to simply count the number of lesions, but with earlier detection of caries. Because each caries detection method may have different power to find statistical differences under different methods of statistical analysis, determination of the ‘best’ of the new detection methods may depend on the type of primary outcome required for the study. ICDAS and QLF examination baseline and 1-year results of a 2-year 2-by-2 factorial-designed clinical study of chlorhexidine gel and fluoride varnish were used to calculate various measures of change in caries. Preliminary QLF results surprisingly indicate more power to detect fl uoride varnish effects using area than _ Q, and slightly higher power for subject totals compared to individual sites and when restricting to sites with caries at baseline. 

37* Clinical Calibration by Five Examiners Using ICDAS on Occlusal Surfaces, Buccal Pits and Lingual Grooves

H. Eggertsson a, *, A. Papas b , S.G. Ciancio c , A. Ferreira-Zandoná a , M. Singh b , O. Shilby c , J. Fabiano c , G.J. Eckert d , D.T. Zero

a * heggerts@iupui.edu a Indiana University School of Dentistry, 
b Tufts University, 
c University at Buffalo, d Indiana University School of Medicine, Indianapolis, Ind., USA

The ICDAS criteria have shown promise as a new tool for visual detection of caries, as compared to the traditional way of assessing caries at the level of cavitation. For widespread use, the criteria must work in various populations and by diverse examiners. This study reports the results of a calibration exercise on occlusal surfaces, as well as buccal pits and lingual grooves, after two days of training. Over a 2-day period, 77 subjects, age range 10–55, were examined. Seventeen subjects were re-examined 24 h after their initial examination. The subjects’ teeth were cleaned by brushing, and then examined by a senior examiner and four other examiners. Scoring for lesion severity was done using a seven-category scale (0–6) according to the ICDAS criteria, while lesion activity was scored on a three-step confidence scale. Intra- and inter-examiner repeatability was assessed using weighted kappa (WK) statistics for occlusal surfaces and buccal pits/lingual grooves (BP/LG). Intra-examiner repeatability for occlusal surfaces ranged from 0.63 to 0.90 (WK). On BP/LG the range was 0.55 to 0.85. For activity the WK ranged from 0.15 to 0.65 on occlusal surfaces. The inter-examiner repeatability was slightly lower, ranging from 0.65 to 0.75 (WK) for occlusal surfaces, and 0.63 to 0.73 for BP/LG. The inter-examiner activity scoring was less repeatable, ranging from 0.21 to 0.40 for occlusal surfaces and 0.37 to 1.00 for BP/LG. In this 4-day calibration exercise, the intra and inter-examiner agreement for ICDAS severity scores was substantial. Further training is required to achieve acceptable agreement on lesion activity. Supported by Curozone USA Inc.  

Abstracts of the 53^rd Annual Orca Congress - Reprinted from "Caries Research" volume 40 (4)

92* Tooth-brushing Behaviour of Deprived Schoolchildren Assessed with Videotaped Sessions at School and a Questionnaire

S. Martignon*, M.C. González, A. Guzmán, V. Sáenz, M. Martínez, A. Mora, I. Quintero * martignonstefania@unbosque.edu.co

Universidad El Bosque, Bogotá, Colombia

The aim of this study was to investigate the tooth-brushing behaviour of schoolchildren by analysis of tooth-brushing (videotaped at school) and a questionnaire. 150 5–8 year-old schoolchildren (78%) from an ongoing preventive-intervention programme in two low-SES schools in Bogotá participated. Children had a mean number of 22.73 (SD 12.23) surfaces with carious lesions ac cording to the ICDAS Index [Pitts: Community Dental Health 2004; 21: 193–198]. With the Cariogram caries risk assessment [Bratthall et al.: Eur J Oral Sci 1996; 104: 486–491], 32% of subjects were scored as high and 42% as very high. The median length of tooth-brushing was 115 s (25th percentile 83 s; 75th percentile 178 s) of which most of the time (median 90 s; 25th percentile 62 s; 75th percentile 143 s) the toothbrush was in the child’s mouth. Almost all children brushed their anterior (94.6%) and posterior teeth (93.3%), of which 92% the lower and 86.6% the upper. There was no difference between right and left sides. The majority brushed the occlusal (lower 81%; upper teeth 67%) and the buccal surfaces (anterior 91%; posterior lower 74%; and posterior upper teeth 71%), as opposed to the lingual/palatal surfaces (12; 12; and 15%, respectively). Most children used the mirror (78.7%), spat out (86.7%) and rinsed their mouth (72%). Finally, the majority (85%) was confi dent that the tooth-brushing session was effective. The questionnaire disclosed the following: none of the children brush their teeth at school; at home only 1/5 (20.5%) of the children are supervised by an adult; and not everybody brushes their teeth twice a day (78%). The overall positive fi ndings of the tooth-brushing, put into context with the high caries status and the poor tooth-brushing habits at home, lead to a strong recommendation to incorporate daily supervised school-based tooth-brushing sessions. Partially supported by Colgate-Palmolive, Colombia. 

94* Caries Status in Colombian Patients with Cleft Lip-Palate: Visual Examination using the ICDAS Criteria

M.C. González*, S. Martignon, A.M. Gaona, L.F. Gamboa * gonzalezmaria@unbosque.edu.co Universidad El Bosque, Bogotá, Colombia

The aim of this study was to assess the caries status in a cleft lip-palate Colombian population. Eighty-fi ve 2–25-year-old subjects (85% from our Faculty’s craniofacial programme) participated. The sample was distributed according to dentition: primary, mixed and permanent (n = 19, 27 and 39, respectively), with age ranges of: 2–5 (mean 5); 6–11 (mean 9); and 12–25 (mean 17). Two examiners assessed dental caries using the ICDAS-visual criteria [Pitts: Community Dental Health 2004; 21: 193–198]. Filled surfaces or surfaces missing due to caries were also scored. Inter- and intra-examiner reproducibilities for caries scoring were substantial ( _ 1 0.80). The traditional mean dmf-s/DMF-S (d/D-component cavitated lesions) was for the primary-dentition group 9.3 8 10.4, of which 5.1 8 7.8 were filled and 2.6 8 6.4 missing. Corresponding figures for the mixed-dentition group were 9.6 8 9.1, 5.1 8 6.1 and 1.7 8 4.0, respectively; and for the permanent-dentition group 9.6 8 10.6, 6.1 8 8.0 and 1.8 8 5.1, respectively. When adding non-cavitated lesions, values increased to 20.2 8 12.8, 21.9 8 10.7 and 25.0 8 10.3, respectively. All participants in the primary dentition had 6 1 caries lesion (ICDAS criteria); in the mixed-dentition and permanent dentition groups 6 5. These were followed by 9–12 lesions (42%), 9–12 (33%), and 13–16 (28%), respectively. The teeth most affected with caries lesions were: primary dentition: second lower left primary molar (1.4 8 0.95); mixed dentition: fi rst lower left permanent molar (1.18 8 0.76); and permanent dentition: second lower left permanent molar (1.0 8 0.89). Regarding the anterior superior teeth, in the primary dentition group the lateral right superior incisor was the fourth most affected tooth (0.95 8 1.35): in the mixed dentition group the left superior central incisor was the third most affected tooth (0.81 8 0.96); and in the permanent dentition group the right upper canine was the fourth most affected tooth (0.85 8 0.85). The results showed a high caries experience, with an apparently normal distribution among the teeth, with the anterior superior teeth not being the most affected, as one could expect. There is an urgent need to implement preventive programmes. 

Abstracts of the 54^th Annual Orca Congress - Reprinted from "Caries Research" volume 41 (4)

59* Development of a Universal Visual Scoring System (UniViSS)
J. Kühnisch a, * , I. Goddon b , H. Senkel b , B. Monse-Schneider c ,
T. Oehme d , R. Hickel a , R. Heinrich-Weltzien e
* jkuehn@dent.med.uni-muenchen.de
a Department of Conservative Dentistry, Ludwig Maximilian University of Munich, 
b Health Department of the Eneppe-Ruhr-District, Schwelm, 
c Center for International Migration & Development, Bonn, 
d Dental practice, Lichtenstein,
e Department of Preventive Dentistry, University of Jena, Germany

During the last decade several methods for caries detection and diagnosis have been developed and tested. Bearing in mind limitations of adjunct diagnostic tools, e.g. imperfect reliability and validity, difficulties in controlling confounding factors, etc., it seems desirable to improve visual caries detection and diagnosis. Therefore, our work aimed at systematising caries lesions as precisely as possible with one universal visual scoring system (UniViSS). The development of UniViSS was influenced by the work of Ekstrand et al. [Caries Res 1998; 32: 247–254], Nyvad et al. [Caries Res 1999; 33: 252–260] and the recently inaugurated ICDAS criteria. UniViSS was systematically adjusted on the basis of the clinical experience of all authors, dental screening of children and young adults in industrialised and developing countries as well as from in vivo/in vitro diagnostic studies. Further, more than 500 photographs of different caries stages from occlusal and smooth surfaces were used to categorise all kind of lesions. As a result, UniViSS should be understood as a three-step evaluation procedure: 1. lesion detection followed by discoloration assessment (white, white-brown, brown) 2. severity assessment (first signs, established lesion, microcavity, dentine exposure, large cavity, pulp involvement) and 3. activity assessment. The cross-tabulation of discoloration and severity enables the dentist to classify all caries lesions on occlusal and smooth surfaces in the primary and permanent dentition with one system. Further, UniViSS was intended to be compatible with the WHO and ICDAS criteria. It is assumed that UniViSS could close the gap between sound and decayed lesions which will lead to a more precise clinical assessment. However, reproducibility and validity of the system needs to be tested in future studies.

61* Relationship between ICDAS II Scores and Histological Lesion Depth on Proximal Surfaces of Primary and Permanent Teeth
S. Martignon a, * , K.R. Ekstrand b , S. Cuevas a , J.F. Reyes a , C. Torres a ,
M. Tamayo a , G. Bautista a * martignonstefania@unbosque.edu.co
a Research Unit UNICA-B, Universidad El Bosque, Bogotá,
Colombia; 
b University of Copenhagen, Denmark

This study aimed to determine the relationship between lesion severity assessed by ICDAS criteria and histological depth on sound and carious proximal surfaces. The ICDAS scores used were: 0 sound; 1B/1W brown/white opacity after air-drying; 2B/2W brown/white opacity without air-drying; 3 microcavities; 4 underlying shadow; 5 distinct cavity and 6 extensive cavity. The sample consisted of 140 permanent teeth and 108 primary teeth stored in thymol water at the University of El Bosque, Bogotá. A total of 160 carious or sound permanent proximal surfaces and 136 carious or sound primary proximal surfaces were finally assessed by the main investigator SM. The surfaces were cleaned and clinically assessed by means of head light, WHO probes and air-drying. Re-examination of all teeth was conducted after 8 days. Afterwards teeth were cut longitudinally at the center of the lesions and 220- _ m-thick sections were stereomicroscopically assessed for demineralization by KE, as follows: 0 no demineralisation; 1 enamel demineralisation limited to the outer 50% of the enamel layer; 2 demineralisation involving 50% of the enamel to one-third of the dentine; 3 demineralisation involving the middle 1/3 of the dentine; and 4 demineralisation involving the inner one-third of the dentine. ICDAS scores were distributed as follows for the permanent teeth: 0 29; 1B 13; 1W 2; 2B 32; 2W 31; 3 20; 4 7; 5 5 and 6 21; for the primary teeth: 0 19; 1B 3 lesions; 1W 4; 2B 11; 2W 37; 3 13; 4 9; 5 7 and 6 33: Intra-examiner reproducibilities (kappa) for the ICDAS scores were for the permanent and the primary teeth 0.86 and 0.92 respectively. Spearman’s correlation coefficient (r S ) was 0.87 for the permanent teeth and 0.92 for the primary teeth. Results showed that the correlation between ICDAS scores and histological changes was excellent both for primary and permanent proximal lesions; intra-reproducibility was also excellent.

62* Reproducibility of the International Caries Detection and Assessment System (ICDAS II) on Occlusal Surfaces
A. Jablonski-Momeni a, * , D.N.J. Ricketts b , V. Stachniss a , K. Pieper a
* momeni@staff.uni-marburg.de
a Philipps University of Marburg, Germany; b University of Dundee, UK

This study aimed to evaluate the reproducibility of the ICDASII visual classification system. One experienced examiner (A) trained 3 other examiners (B, C, D) who had no previous experience in the ICDAS-II classification system in a one-hour session. The codes included: 0 = sound, 1 = first visible sign of non-cavitated (NC) lesion seen only when the tooth is dried, 2 = clinically visible NC lesion seen when wet, 3 = microcavitation in enamel, 4 = NC lesion extending into dentin seen as an undermining shadow, 5 and 6 = cavitated lesion with visible dentine (small and large respectively). 100 unrestored permanent posterior teeth were selected from a group of extracted teeth and carefully cleaned. One to four sites within the pit and fissure system of each tooth were chosen for investigation by each examiner (total sites 181). Examiners B, C and D re-examined all the investigation sites after 3 weeks in order to determine the intra-examiner reproducibility. Weighted kappa values for inter- and intra-examiner reproducibility were calculated, both for all the individual ICDAS-II-codes (group 1) and on a combination of the codes (group 2: 0, 1–3, 4–5, 6). Inter-examiner reproducibility kappa values (group 1) were for examiners A-B 0.8, A-C 0.69, A-D 0.76, B-C 0.68, B-D 0.74 and C-D 0.61. Inter-examiner kappa values for the combined codes (group 2) were 0.71, 0.67, 0.7, 0.63, 0.7 and 0.57 respectively. Intra-examiner reproducibilities were: B 0.8, C 0.73 and D 0.83 (group 1) and for combined codes (group 2) were 0.8, 0.74 and 0.82. The ICDAS-II system achieved substantial to good agreement between and within examiners after a short training session.

63* Reproducibility of ICDAS II Criteria for Occlusal and Approximal Caries Detection in Primary Teeth
L. Shoaib a, * , C. Deery b , Z.J. Nugent c , D.N.J. Ricketts d
* c.deery@sheffield.ac.uk
a Edinburgh Dental Institute, University of Edinburgh, b School of
Clinical Dentistry, Sheffield, UK; c CancerCare Manitoba, Canada;
d Dundee Dental Hospital and School, University of Dundee, UK

The aim of this in vitro study was to assess the reproducibility of the detection of occlusal and approximal caries in primary teeth using the ICDAS II Criteria [Pitts: Community Dent Health 2004;21:193–198], working at the D 1 (enamel and dentine caries) and ICDAS II codes 2/3 (the first visible sign of cavitation or dentine caries) thresholds. Three trained examiners independently examined 112 extracted primary molars under dental surgery conditions, using the ICDAS II criteria. The teeth were cleaned and setup in groups of 4, in pink impression putty to mimic their anatomical positions. The condition of the teeth used ranged from clinically sound to cavitated; extensively broken down teeth were excluded. As per the ICDAS II criteria a 3: 1 syringe and blunt probe were used during the examinations and the most advanced lesion on each surface scored. At all other times the teeth were kept wet. Each examination was conducted blind and repeated by each examiner after a gap of at least 24 h. Kappa values were calculated to assess reproducibility of the examinations. The intraexaminer reproducibility ranged from 0.74 to 0.83 and from 0.72 to 0.85, at the D 1 and ICDAS II codes 2/3 diagnostic thresholds, respectively. The inter-examiner reproducibility ranged from 0.60 to 0.72 and from 0.63 to 0.80 at the D 1 and code 2/3 diagnostic thresholds respectively. These values generally represent ‘substantial agreement’ [Landis and Koch: Biometrics 1977;33:159– 174]. In conclusion the reproducibility of the ICDAS II was acceptable when applied to primary molar teeth.

64* A Training and Calibration Study Using ICDAS II in Association with Light Fluorescence
A.G. Ferreira Zandona a, * , H. Eggertsson a , E. Delgado b ,
P. Hernandez b , G.J. Eckert c , M.S. Mau a , J. Tran a , D.T. Zero a
* azandona@iupui.edu
a Indiana University School of Dentistry, Indianapolis, b School
of Dentistry Research Center, University of Puerto Rico,
c Section of Biostatistics, Indiana University School of Medicine,
Indianapolis, USA

The aim of this study was to assess inter and intra-examiner agreement using ICDAS II in association with QLF (Quantitative Light Induced Fluorescence). Two independent phases were performed consecutively. Both phases included lecture, in vitro and in vivo training, and calibration. In Phase 1, 2 examiners were trained and calibrated against an expert examiner on ICDAS. All permanent teeth surfaces of 53 subjects (aged 7–17) were examined using ICDAS. 20 subjects were re-examined for intra-examiner repeatability assessment. In Phase 2, 3 examiners were trained and calibrated against an expert examiner using modified ICDAS criteria (QLF-I criteria) which incorporated the ICDAS exam and the interpretation of the light fluorescence image obtained with QLF. First and second permanent molars of 49 subjects were examined. 21 subjects were re-examined to assess intra-examiner agreement. ICDAS criteria for each surface were recorded and a QLF image was obtained. On live QLF images, the following criteria were used: (0) sound tooth surface; (1) slight fluorescence change; (2) distinct fluorescence change; (3) visible enamel breakdown with a distinct fluorescence change; (4) poorly delineated distinct fluorescence change with or without enamel breakdown; (5) cavitation visible with distinct fluorescence change. ICDAS examination weighted kappa scores for inter-examiner agreements were 0.67 and 0.68 (occlusal surfaces); 0.68 and 0.61 (smooth surfaces) and 0.61 (both examiners) (interproximal surfaces). Repeatability was good for all surfaces and ranged from 0.60 (interproximal) to 0.76 (occlusal). For QLF-I exam inter-examiner agreement ranged from 0.66 to 0.77 (occlusal surfaces) and from 0.63 to 0.70 (smooth surfaces). Intra-examiner agreement ranged from 0.73 to 0.90 (smooth surfaces) and from 0.78 to 0.85 (occlusal surfaces). In conclusion, all examiners were able to achieve good agreement with expert examiners and had good repeatability with both methods. Supported by NIH/NIDCR RO1DE017890-01.

65* Experience of Calibration Exercises Conducted in Colombia with the ICDAS II Visual Criteria for Primary Coronal Caries
L.F. Gamboa a, b, * , F. Gil b , S. Martignon a
* gamboaluis@unbosque.edu.co
a Caries Research Unit UNICA-B, Universidad El Bosque, and
b Pontificia Universidad Javeriana Dentistry and Medical School,
Bogotá, Colombia

Three calibration exercises on the ICDAS caries criteria have been conducted in Colombia by a trained examiner (SM). The aims of this study were to identify the ICDAS codes in which most failures (differences) between the gold standard and the other examiners occurred, and to determine the code combinations that allow higher inter-examiner agreement. 69 examiners without previous ICDAS experience participated (2005a: n = 22; 2005b: n = 13; 2006: n = 34). The exercise lasted 4 days and involved: (1) state of the art lecture in modern concepts of dental caries and ICDAS; (2) exercises on natural teeth with a stereomicroscope, presentation and discussion of clinical cases on patients and first preclinical examination of 40 cases on pictures; (3) First clinical examination (approximately 110 teeth among 16 subjects) and second preclinical examination; and (4) Second clinical examination conducted after a one-day recess. 19% of examiners achieved acceptable levels of agreement (kappa 6 0.60; mean 0.61 8 0.001 SD). Remaining examiners obtained a mean kappa of 0.43 8 0.10. Most failures occurred when scoring code 1W (first white change in enamel visible after 5 s air-drying), with only 28.7% of agreements. Disagreements (71.3%) were distributed as follows: 0 (sound) 50%, and 2W (distinct white visual change in enamel when viewed wet) 17.7%. If ICDAS codes corresponding to white and brown opacities (1W, 2W, 1B, 2B) were combined, kappa 6 0.60 were achieved in 23 examiners (33%). The first and the second kappa values were significantly different (Mann-Whitney test p ! 0.01). Combining ICDAS codes will increase the number of examiners who reach the required level of agreement, but it will also reduce the capacity of the criteria to discriminate between stages of caries.

66* Epidemiology of Dental Caries Using ICDAS-II Criteria among Schoolchildren Living in a Fluorosis Endemic Area
S. Martignon, M. Tellez * , R.M. Santamaria, Y. Muñoz
* mtellez@umich.edu
Caries Research Unit UNICA-B, Universidad El Bosque, Bogotá,
Colombia

The purpose of this study was to determine the caries and fluorosis status in 110 12-year-old children from the state of Huila, Colombia, a fluorosis endemic area. One calibrated examiner (SM) conducted the clinical examinations after tooth-brushing, using a portable dental unit, a head-light and a ball-ended probe. teeth using the Thylstrup-Fejerskov Index (TFI) and dental caries was assessed with the ICDAS-II visual criteria on all surfaces. Bivariate analyses were conducted to assess the relationship between dental caries and fluorosis. Mean DFS including cavitated and non-cavitated lesions was 10 8 5.1, with a higher mean number of surfaces affected for girls (11 8 5.3) than boys (8.9 8 4.7), and the D-component being the principal contributor to DFS (80%). The prevalence of dental fluorosis reached 100% with most children falling into the moderate severity categories (score 1 0.9%; score 2 15.5%; score 3 61.8%; score 4 16.4%; score 5 3.6%; score 6 1.8%). 22.1% of the children with TFI scores 3 and 4 also had a high proportion of surfaces coded as ICDAS code 2 (distinct visual change in enamel). The association between fluorosis and dental caries did not reach statistical significance. However, more surfaces with localized enamel breakdown, underlying dark dentine shadow and distinct cavity appeared to occur when the severity of fluorosis reached a rating of TFI 3–4 as indicated by the mean DFS (TFI 1–2/DFS 3.8 8 5.7; TFI 3–4/DFS 4.6 8 4.0; TFI 6 5/DFS 3.7 8 3.9). It is concluded that the ICDAS-II criteria are useful in evaluating the relationship between dental caries and dental fluorosis in populations where early carious lesions are more prevalent than cavitated lesions.

67* Visual (ICDAS I) and Radiographic Detection of Approximal Caries in a National Oral Health Survey
H. Eggertsson a, * , H. Gudmundsdottir b, e , H. Agustsdottir c ,
I.B. Arnadottir d , S.T. Eliasson d , S.R. Saemundsson d , S.H. Jonsson e ,
W.P. Holbrook d
* heggerts@iupui.edu
a Indiana University School of Dentistry, USA; b Center for Oral
Health and c Ministry of Health and Social Security, Iceland;
d University of Iceland, e Public Health Institute of Iceland

The purpose of this study was to compare the diagnostic yield from two critical sources of information on approximal caries, visual detection and radiographic detection. The Icelandic Oral Health Survey, a national survey was conducted in 2005, where 20% of 6-, 12- and 15-year-old children in the country were examined. Examinations were conducted using portable dental units, following thorough brushing and flossing by a dental professional. The new ICDAS I visual detection criteria, which require complete air drying of the teeth, were used. Digital bitewing radiographs of the 12- and 15-year-olds were obtained using Digora storage phosphor system and appropriate software. The radiographs were then scored under conditions of reduced lighting. ICDAS and the radiographic lesion detection both allowed estimation of lesion severity on a scale from 0 to 6. Filled surfaces and missing teeth were excluded. Out of 17,155 approximal surfaces available for examination for 12-year-olds, visual examination detected lesions in 1,109 surfaces (6.9%), of which 96.5% were non-cavitated (ICDAS codes 1 or 2). The radiographic examination detected 1,781 lesions in those 17,155 surfaces (10.4%), of which 88.0% were in enamel (D1–2 level). For 15-year-olds a total of 18,032 surfaces were examined with both methods. Visually lesions were found in 2,150 surfaces (11.9%) with 96.4% of those being non-cavitated, but radiographically lesions were found in 3,949 surfaces (21.9%), and 83.6% of those were confined to enamel. The increase in caries levels detected with the addition of the radiographic examination was similar in both age groups, or roughly two-fold increase in the number of lesions. Even though the ICDAS criteria allow for very thorough examination of the approximal surfaces, more lesions were found on radiographs. The radiographs also indicated greater lesion severity. Supported by Center for Oral Health, Primary Health Care Unit, and the Public Health Institute of Iceland.

68* Agreement on the Site of Caries Using ICDAS II Criteria and Two Laser Fluorescence Devices in Primary Teeth
C. Deery a, * , Z.J. Nugent b , D.N.J. Ricketts c , L. Shoaib d
* c.deery@sheffield.ac.uk
a School of Clinical Dentistry, University of Sheffield, UK;
b CancerCare Manitoba, Canada; c Dental Hospital and
School, University of Dundee, d Dental Institute, University of
Edinburgh, UK

The aim of this in vitro study was to compare the agreement on the site of the most advanced occlusal carious lesion in primary teeth using the ICDAS II criteria [Pitts: Community Dental Health 2004; 21: 193–198] and two laser fluorescence devices: DIAGNOdent (KaVO, Germany) (LF) and DIAGNOdent pen (KaVo, Germany) (LF pen). Three trained examiners independently examined 112 extracted primary molars under dental surgery conditions, using ICDAS II criteria and the LF devices. The teeth were cleaned and set up in groups of 4, in pink impression putty to mimic their anatomical positions. The condition of the teeth used ranged from clinically sound to cavitated. Extensively broken down teeth were excluded. As per the ICDAS II criteria a 3: 1 syringe was used during the examinations. The examiners examined the occlusal surfaces and recorded the site of the most advanced caries using the ICDAS II criteria and marked the site of the lesion on a diagram. The LF devices were scanned over the entire surface, the highest reading recorded and the site of this reading was marked on a diagram. Each examination was conducted blind and repeated after a break of at least 24 h. Lesion site was then compared for each examination. There was agreement between all three methods and all three examiners on the site of the most advanced lesion on 45% of occasions. All three examiners agreed on the site on 79%, 62%, 65% of occasions for ICDAS II, LF, LF pen (p ! 0.05), respectively. In conclusion, visual examination, using the ICDAS II criteria, achieved a higher level of agreement on the site of most advanced lesions than either of the two LF devices.

81* Caries Detection by Optimal Clinical Visual, Radiographic, Laser Fluorescence and AC-Impedance Spectroscopy Techniques
C. Longbottom a, b, * , A.F. Hall a, , A. Czajczynska-Waszkiewicz a, ,
N.B. Pitts a, b , P. Los a, b , S. Kochan b , D.J.N. Ricketts c
* c.longbottom@dundee.ac.uk
a University of Dundee Centre for Clinical Innovations, b IDMoS
Dental Systems Ltd, Dundee, c University of Dundee Dental
School, Dundee, UK

The aim of this study was to compare in vitro the diagnostic performance of several techniques: optimal clinical visual (ICDAS), simulated bitewing radiography (BW), and laser fluorescence (Dd-pen), with pre-production ac-impedance spectroscopy (ac-IS) devices, using a combined clinical/micro-CT method as validation. 23 extracted human permanent premolars and molars were set in quasi-quadrant blocks and examined at 34 occlusal and 34 free smooth surface sites, using each of the detection methods at occlusal and free smooth surface sites. The ICDAS and Ddpen methods were carried out twice by one examiner. The BW and validation methods were carried out by 3 examiners using a consensus method and the ac-IS method was completed by two examiners (3 times). The performance data are presented as success ratios (number of correct measurements/total number of measurements) for 2 classes: (a) sound; (b) enamel and dentine caries, i.e. the D 1 threshold. The ac-IS results were obtained using a TREE method of analysis of assessing electrical parameters. The number of sites in each of the classes was: occlusal surfaces 3 sound, 31 carious; free smooth surfaces 15 sound, 19 carious. Success ratios were for occlusal surfaces ICDAS 80%, BW 26%, Ddpen 48%, ac-IS 93%, and for free smooth surfaces ICDAS 76%, BW n/a, Dd-pen 67%, ac-IS 92%. The results demonstrate that, at the D 1 threshold, the ac-IS method has a superior detection performance to the other techniques. Supported by IDMoS Dental Systems Ltd, Dundee, UK.