Abstract
Objectives
To assess the reliability of the BEWE index on 3D models and to compare 3D-assessed erosive tooth wear scores with clinically detected scores.
Methods
In total, 1964 members of the Northern Finland Birth Cohort 1966 participated in a standardized clinical dental examination including the Basic Erosive Wear Examination (BEWE) and dental 3D modelling at the age of 45–46 years. Of those examined, 586 were randomly selected for this study. 3D models were assessed using the same BEWE criteria as in the clinical examination. Calculated kappa values as well as the prevalence and severity of erosive wear according to the clinical examination and 3D models were compared. Re-examinations were performed to calculate intra- and inter-method and −examiner agreements.
Results
The BEWE index on 3D models was reproducible; the mean intra- and inter-examiner agreement were 0.89 and 0.87, respectively, for sextant level, and 0.64 and 1, respectively, for BEWE sum scores. Erosive tooth wear was recorded as more severe in 3D models than in the clinical examination, and inter-method agreement was 0.41 for severe erosive wear (BEWE sum > 8). The biggest inter-method differences were found in upper posterior sextants.
Conclusions
The BEWE index is reliable for recording erosive tooth wear on 3D models. 3D models seem to be especially sensitive in detecting initial erosive wear. Additionally, it seems that erosive wear may be underscored in the upper posterior sextants when assessed clinically. Due to the nature of 3D models, the assessment of erosive wear clinically and on 3D models may not be entirely comparable.
Clinical significance
3D models can serve as an additional tool to detect and document erosive wear, especially during the early stages of the condition and in assessing the progression of wear. When scoring erosive wear clinically, care must be taken especially when assessing upper posterior sextants.
1
Introduction
Erosive tooth wear results from the combined effect of chemical erosion caused by acids and the mechanical wear of acid-softened enamel by attrition and abrasion . Owing to this multifactorial etiology, erosive wear can manifest itself in different ways and early diagnosis is particularly challenging . Currently, there are a large number of different grading scales used for diagnosis; among them quite recently introduced Basic Erosive Wear Examination (BEWE) . In the four-step BEWE classification, each surface of a tooth is examined and registered separately but recorded according to sextant; the most damaged tooth surface represents the BEWE score of a sextant (from 0 to 3). In class 0, no signs of erosive wear are observed, whereas class 1 represents initial loss of surface texture. In class 2, erosive lesions are more pronounced and there is hard tissue loss less than 50% of the surface area. In class 3, hard tissue loss covers more than 50% of the surface area. The patient BEWE score is the sum score of the sextants (from 0 to 18). The BEWE has been developed by internationally recognized experts in the field and is intended to be an internationally accepted, standardized and validated index and a simple tool for clinical practice .
Studies concerning the reliability and repeatability of the BEWE index are rare, but moderate reliabilities have been reported and the BEWE has been approved for clinical and epidemiological use. The BEWE was originally developed to be suitable also for grading dental casts , but to date there are no studies assessing the reliability of BEWE on either dental casts or digital 3D models. However, BEWE has been shown to be suitable for evaluating erosive wear on photographs .
Irrespective of the grading scale or method used, erosive tooth wear appears to be challenging to detect in its initial stages . Concerning tooth wear assessment in general, conventional clinical examination has even been suggested to be the least sensitive when compared to more sophisticated methods , including recording from high-resolution 3D modelling. However, it has been shown that initial wear is hard to distinguish also on study casts . 3D models have been used clinically in orthodontics, oral surgery and implant treatments for years. More recently, computer-aided design and manufacturing, referred to as CAD-CAM-technology, has become increasingly popular in designing and manufacturing restorations, crowns and fixed prosthodontics. In general, CAD-CAM-technology and 3D models are expected to improve implementation of medical treatments and documentation of treatment outcomes . This technology might also be useful in tooth wear assessment and treatment planning and monitoring. It has even been proposed that comparing sequential 3D models is the most accurate method for measuring tooth wear progression . However, no epidemiologic studies concerning the use of 3D models in erosive tooth wear assessment have been reported.
The aim of this study was to assess whether classifying erosive tooth wear on 3D models using the BEWE index is reliable and whether erosive wear scores on 3D models are comparable with clinical findings. The hypothesis was that the BEWE index is reliable in the assessment of erosive tooth wear on 3D models, and that 3D models are suitable for detecting erosive tooth wear.
2
Materials and methods
2.1
Study population
The study population was a part of the Northern Finland Birth Cohort 1966 (NFBC 1966), initially comprising all 12,058 children whose expected time of delivery was in the year 1966 . The cohort has been evaluated regularly since birth by means of health questionnaires and clinical examinations . In connection with the 46-year follow-up survey (2012–2014), a subgroup of NFBC 1966 (a total of 3181 persons currently living in the city of Oulu or within 100 km of Oulu, including rural areas) was invited by mail to participate in clinical oral examinations. Examination protocol is more precisely described on our previous paper . Participation in the study was voluntary and 1962 subjects agreed to participate. One-third of this original subgroup was randomly selected to comprise the present study population (N = 600).
2.2
Clinical examination
Clinical dental examinations were performed between April 2012 and June 2013. Examinations were carried out supine in a dental chair with modern equipment and optimal lightning with a probe, oral mirror and WHO ball pointed gingival probe. The teeth were blow-dried using a three-in-one syringe before assessment. In the clinical examination, the BEWE index (0–3) was used to register erosive tooth wear in every sextant according to the original BEWE definitions . All tooth surfaces were examined and the highest score for each sextant was recorded. The BEWE scoring criteria are presented in Table 1 . The examiners were advised not to include surfaces with only wedge-shaped defects or flat attritional facets in the assessment, if no signs of erosion were present. Erosive tooth wear could not be recorded if there were extensive restorations or prosthetic work covering the entire tooth surface. Wisdom teeth were excluded from the assessment.
2.3
Validation in clinical examination of erosive wear
A dentist specialized in the field of cariology (VuA) instructed and calibrated seven dentists with respect to the clinical examination, including the erosive wear examination. The instruction comprised lectures on the use of the BEWE index to ensure that all examiners would have similar theoretical knowledge. Criteria for classifications were presented as a PowerPoint presentation on a screen using a PC and data projector. Calibration was repeated every third month. The criteria for the BEWE classifications were available to the examiners throughout the study in written and graphical form in the examination room. A senior dentist and researcher (MLL), familiar with the study protocol and criteria, acted as a gold standard and ensured that the examiners followed the study protocol precisely throughout the entire field study. To assess the intra-examiner agreement, the examiners re-examined one quadrant of an average of 5 patients approximately one month after the first examination. To assess the inter-examiner agreement, the gold standard examiner (MLL) re-examined one quadrant of an average of 10 patients from each examiner.
2.4
3D models
On the clinical examination day, 3D models were obtained by dental hygienist trained by the manufacturer using an iTero 3D scanner (Cadent, San Jose, CA, USA). Models were scanned and stored unrevised in the server of the cohort. The 3D models were assessed on a PC screen (screen information) using 3Shape Ortho Analyzer™ software. Erosive tooth wear was assessed using the BEWE index and the same scoring criteria as in the clinical examination ( Table 1 ). In the software, 3D models were yellow with different degrees of shading. It was possible to zoom and rotate the models while analyzing; thus it was possible to inspect models from different angles. For analyzing the 3D models, the gold standard examiner (MLL) and the instructor with regard to the clinical examination (VuA) trained three examiners (two fourth-year dental students with more than one year clinical experience (RP, HS) and one dentist (ViA)), using the same basis of instruction as in the clinical examination. The examiners were calibrated first through the coordinated analysis of 3D models until agreement was reached regarding the severity of erosive changes between themselves and with the gold standard examiner of the clinical examination. Analyses were performed over one week at the same location with the same lightning. Analyses were first timed to maximize the sample size in a limited period of time. Two fourth-year dental students analyzed all 600 3D models together, and the dentist (ViA) acted as the gold standard examiner for re-examinations and was available in the examination room throughout the study. The criteria for the BEWE classifications were available to the examiners in written and graphical form in the analyzing room. To assess intra-examiner agreement 19 (144 sextants) randomly picked 3D models were re-examined during the analysis of all 600 models. To assess inter-examiner agreement 11 (66 sextants) randomly chosen 3D models were re-examined after analyzing all 600 3D models. The subjects were randomized by choosing every twentieth person from a list in which the subjects were sorted according to their project ID. The re-examinations were performed at the same time at the same space, and the students were allowed to discuss using a few words with each other while deciding the score.
2.5
Statistics
To describe and analyze the prevalence and severity of erosive tooth wear according to the clinical examination and analyzed 3D models, the sum scores of the BEWE index were calculated and interpreted as follows (adjusted from the original scoring system by combining the two highest risk groups): BEWE sum score 0–2 = no or only mild erosive tooth wear with no treatment needed; BEWE sum score 3–8 = moderate erosive tooth wear, treatment needed; and BEWE sum score 9–18 = severe erosive tooth wear, treatment needed.
Descriptive statistics such as frequencies, distributions, means and standard deviations were calculated. The kappa coefficient and percentage agreement was calculated to analyze inter- and intra-examiner as well as inter-method agreements. Kappa values were interpreted as suggested by Landis and Koch . The inter-method difference between mean BEWE outcomes per sextants was analyzed using the Wilcoxon signed rank test and the mean BEWE sum score using a paired sample t -test. The differences of the BEWE sum score distributions between the methods were tested using the chi-square test and Fisher’s exact test. A P -value < 0.05 was considered statistically significant. All statistical analyses were performed using SPSS (version 22.0, SPSS, Inc., Chicago, IL, USA).
2
Materials and methods
2.1
Study population
The study population was a part of the Northern Finland Birth Cohort 1966 (NFBC 1966), initially comprising all 12,058 children whose expected time of delivery was in the year 1966 . The cohort has been evaluated regularly since birth by means of health questionnaires and clinical examinations . In connection with the 46-year follow-up survey (2012–2014), a subgroup of NFBC 1966 (a total of 3181 persons currently living in the city of Oulu or within 100 km of Oulu, including rural areas) was invited by mail to participate in clinical oral examinations. Examination protocol is more precisely described on our previous paper . Participation in the study was voluntary and 1962 subjects agreed to participate. One-third of this original subgroup was randomly selected to comprise the present study population (N = 600).
2.2
Clinical examination
Clinical dental examinations were performed between April 2012 and June 2013. Examinations were carried out supine in a dental chair with modern equipment and optimal lightning with a probe, oral mirror and WHO ball pointed gingival probe. The teeth were blow-dried using a three-in-one syringe before assessment. In the clinical examination, the BEWE index (0–3) was used to register erosive tooth wear in every sextant according to the original BEWE definitions . All tooth surfaces were examined and the highest score for each sextant was recorded. The BEWE scoring criteria are presented in Table 1 . The examiners were advised not to include surfaces with only wedge-shaped defects or flat attritional facets in the assessment, if no signs of erosion were present. Erosive tooth wear could not be recorded if there were extensive restorations or prosthetic work covering the entire tooth surface. Wisdom teeth were excluded from the assessment.
2.3
Validation in clinical examination of erosive wear
A dentist specialized in the field of cariology (VuA) instructed and calibrated seven dentists with respect to the clinical examination, including the erosive wear examination. The instruction comprised lectures on the use of the BEWE index to ensure that all examiners would have similar theoretical knowledge. Criteria for classifications were presented as a PowerPoint presentation on a screen using a PC and data projector. Calibration was repeated every third month. The criteria for the BEWE classifications were available to the examiners throughout the study in written and graphical form in the examination room. A senior dentist and researcher (MLL), familiar with the study protocol and criteria, acted as a gold standard and ensured that the examiners followed the study protocol precisely throughout the entire field study. To assess the intra-examiner agreement, the examiners re-examined one quadrant of an average of 5 patients approximately one month after the first examination. To assess the inter-examiner agreement, the gold standard examiner (MLL) re-examined one quadrant of an average of 10 patients from each examiner.
2.4
3D models
On the clinical examination day, 3D models were obtained by dental hygienist trained by the manufacturer using an iTero 3D scanner (Cadent, San Jose, CA, USA). Models were scanned and stored unrevised in the server of the cohort. The 3D models were assessed on a PC screen (screen information) using 3Shape Ortho Analyzer™ software. Erosive tooth wear was assessed using the BEWE index and the same scoring criteria as in the clinical examination ( Table 1 ). In the software, 3D models were yellow with different degrees of shading. It was possible to zoom and rotate the models while analyzing; thus it was possible to inspect models from different angles. For analyzing the 3D models, the gold standard examiner (MLL) and the instructor with regard to the clinical examination (VuA) trained three examiners (two fourth-year dental students with more than one year clinical experience (RP, HS) and one dentist (ViA)), using the same basis of instruction as in the clinical examination. The examiners were calibrated first through the coordinated analysis of 3D models until agreement was reached regarding the severity of erosive changes between themselves and with the gold standard examiner of the clinical examination. Analyses were performed over one week at the same location with the same lightning. Analyses were first timed to maximize the sample size in a limited period of time. Two fourth-year dental students analyzed all 600 3D models together, and the dentist (ViA) acted as the gold standard examiner for re-examinations and was available in the examination room throughout the study. The criteria for the BEWE classifications were available to the examiners in written and graphical form in the analyzing room. To assess intra-examiner agreement 19 (144 sextants) randomly picked 3D models were re-examined during the analysis of all 600 models. To assess inter-examiner agreement 11 (66 sextants) randomly chosen 3D models were re-examined after analyzing all 600 3D models. The subjects were randomized by choosing every twentieth person from a list in which the subjects were sorted according to their project ID. The re-examinations were performed at the same time at the same space, and the students were allowed to discuss using a few words with each other while deciding the score.
2.5
Statistics
To describe and analyze the prevalence and severity of erosive tooth wear according to the clinical examination and analyzed 3D models, the sum scores of the BEWE index were calculated and interpreted as follows (adjusted from the original scoring system by combining the two highest risk groups): BEWE sum score 0–2 = no or only mild erosive tooth wear with no treatment needed; BEWE sum score 3–8 = moderate erosive tooth wear, treatment needed; and BEWE sum score 9–18 = severe erosive tooth wear, treatment needed.
Descriptive statistics such as frequencies, distributions, means and standard deviations were calculated. The kappa coefficient and percentage agreement was calculated to analyze inter- and intra-examiner as well as inter-method agreements. Kappa values were interpreted as suggested by Landis and Koch . The inter-method difference between mean BEWE outcomes per sextants was analyzed using the Wilcoxon signed rank test and the mean BEWE sum score using a paired sample t -test. The differences of the BEWE sum score distributions between the methods were tested using the chi-square test and Fisher’s exact test. A P -value < 0.05 was considered statistically significant. All statistical analyses were performed using SPSS (version 22.0, SPSS, Inc., Chicago, IL, USA).
Stay updated, free dental videos. Join our Telegram channel
VIDEdental - Online dental courses
