In vitro data suggested that laser fluorescence and visual inspection have similar accuracy [11]. There is limited evidence-based data in the literature regarding in vivo diagnostic accuracy of non cavitated carious lesions detection methods. In a previous systematic review[3] based on in vivo data, only 5 studies were eligible for meta-analysis and there was only one study assessing diagnostic accuracy of laser fluroscence (DIAGNOpen) for detection of non -cavitated lesions in smooth facial surfaces with sensitivity and specificity of (0.32–0.78) and (0.64–0.85) respectively. There were no in vivo diagnostic accuracy studies for detection of non-cavitated lesions in smooth facial surfaces using DIAGNOdent. The authors also reported that sensitivity and specificity of DIAGNOdent in previous in vivo studies for occlusal surface in permanent dentition was (0.48–1) and (0.54–1) respectively[3], there was a huge heterogeneity among studies with wide confidence intervals for the sensitivity and specificity. Moreover, comparison between in vitro and in vivo data regarding the accuracy of laser fluorescence for detection of incipient carious lesions showed that in vitro data exhibited better accuracy. Therefore, in vivo assessment of non-cavitated carious lesions using laser fluorescence needed further studies.
Visual examination is the most common used method for detection of initial carious lesions due to its convenience, reliability[12] and cost effectiveness[3]. It is well-established in the current literature that direct visual examination has good validity for detection of white spot lesions, where diagnosis is based on clinical appearance and texture of enamel rather than by-products of bacteria detected by laser fluorescence [13–16]. Authenticating visual examination method for in vivo assessment in trials had many drawbacks compared to histological assessment, which is universally accepted as the reference gold standard to assess accuracy of carious detection methods. Nevertheless, for histological assessment to be done, teeth under investigation should be extracted causing ethical issues[7]. Therefore, to assess in vivo diagnostic accuracy without histological examination, inter-examiner agreement should be used as a surrogate method to enhance accuracy[17]. The DIAGNOdent detects carious lesions through quantifying bacterial by-products (porphyrins) produced by cariogenic bacteria and detecting their biological fluorescence[18]. Furthermore, previous systematic reviews[11, 19] and other in vitro studies[14–16, 20, 21] showed that laser fluorescence detection methods such as DIAGNOdent and DIAGNOdent Pen had high in vitro sensitivity and specificity for detecting initial white spot lesions.
Although DIAGNOdent could be considered as a reliable tool for caries diagnosis, some requirements are needed to minimize false positive readings in clinical practice. For instance, DIAGNOdent overestimates readings for any minor alterations in properties of the dental substrate. Confounders such as stains, calculus, plaque and degree of mineralization can affect the accuracy of readings[22–24]. Hence, standardized measurement protocol is required to enhance reading accuracy of DIAGNOdent. Teeth surfaces should be cleaned before measurements, dryness should be optimum to standardize moisture content of teeth, as it affects light scattering due to difference in refractive index between air and water[22–25]. Light reflection of the dental unit lamp could affect DIAGNOdent results, therefore it is very crucia; to use the same lighting conditions such as intensity, angulation, and distance during measurements [24].
In the current study, DIAGNOdent had overall accuracy of 84.45% with sensitivity and specificity of 87.58% and 96.87% respectively, when score 0 represented sound tooth surface, while scores 1 and 2 were considered as clinically non-cavitated carious lesions, which indicated excellent association between ICDAS-II and DIAGNOdent methods. Moreover, when only ICDAS score 1 was considered representing first visual change in enamel, DIAGNOdent had accuracy of 74.15% with sensitivity and specificity of 83.53% and 90.62% respectively, which also indicated excellent association between ICDAS-II and DIAGNOdent methods. In the present study, when only ICDAS score 2 was considered representing distinct visual change in enamel, DIAGNOdent had accuracy of 100% with sensitivity and specificity of 100% and 100% respectively, which indicated perfect association between ICDAS-II and DIAGNOdent methods. The findings presented herein were supported by two recent systematic reviews, where Thanh et al. [7] found that fluorescence based diagnostic methods had sensitivity and specificity of 80%, while Foros et al. [3] reported that the sensitivity ranged from 0.32–0.78 and the specificity ranged from 0.64–0.85.
The current findings noticed that DIAGNOdent had less accuracy regarding early visual changes in enamel when compared to ICDAS II, while for the more extended distinct visual change in enamel, DIAGNOdent demonstrated perfect accuracy. These observations were in line with previous studies [13, 26] validating fluorescence devices for evaluation of white spot lesions, showing less accuracy with early lesions and better accuracy with more extended lesions. This may be attributed to the low level of bacterial by-products in early incipient enamel lesions [27].
In the present investigation, there was a strongly significant positive correlation (r = 0.892) between ICDAS-II scores and DIAGNOdent readings, which denotes that as the scores of ICDAS- II increased, the DIAGNOdent readings increased, and vice versa. This was also in agreement with another trial reporting a good correlation (r = 0.71) between ICDAS II and DIAGNOpen readings [28].
Furthermore, the present findings found almost perfect inter-examiner agreement for DIAGNOdent and ICDAS II (Kappa = 0.83844 and 0.94311) respectively, which was supported by similar results from other trials [10, 22, 28]. This emphasizes the importance of training and calibration before undergoing diagnostic accuracy studies. The examiners’ calibration is very crucial to ensure high reproducibility during assessment of non cavitated carious lesions [22, 29].
The ideal diagnostic tool for detection and monitoring progress of non -cavitated white spot lesions should have high sensitivity and specificity. For a diagnostic test to be beneficial, the summation of sensitivity and specificity should be at least 1.5 [30]. Additionally, the diagnostic tool for incipient caries should be user friendly, easily applied in clinical situations by less experienced operators, non-invasive, cost-effective and should also provide repeatability and reproducibility among different examiners [13].
Combining DIAGNOdent and direct visual examination for diagnosis of facial smooth surfaces might be considered beneficial, through merging the merits of assessing enamel’s mineral quantity together with texture and appearance. This could help clinicians to enhance their diagnostic efficiency and decision-making during management of early enamel lesions, thus halting lesions progression using non-invasive preventive protocols [31]. Besides, it has been reported that treatment decisions based on combining ICDAS and DIAGNOdent had better accuracy than decisions based on ICDAS only [32].