Braga Mariana Minatel, Mendes Fausto Medeiros, Martins Célia Regina Martins Delgado, Imparato José Carlos Pettorossi
Department of Pediatric Dentistry, School of Dentistry, University of São Paulo, Brazil.
Pediatr Dent. 2006 Sep-Oct;28(5):451-4.
The purpose of this study was to verify the influence of individual calibration on sound tooth surfaces and the number of readings on in vitro performance of a laser fluorescence (LF) device to detect occlusal caries in primary teeth.
After standard calibration (on the porcelain reference object), 72 clinically suspect sites on occlusal surfaces of 54 exfoliated or extracted (for orthodontic purposes) primary molars were assessed using the DIAGNOdent. First, after individual calibration (on a sound surface of each tooth), 3 readings were performed for each site. Subsequently, another 3 readings were taken, but without individual calibration on a sound surface of the tooth. After these assessments, sites were also evaluated with only one reading after individual calibration. Histological validation was performed as the gold standard. LF values, sensitivity, specificity, accuracy (number of correct diagnosis in both sound and diseased teeth) were calculated and compared using the McNemar change test. The area under receiver operating characteristics (ROC) curves was also compared.
Readings with standard calibration only (mean= 11.7 +/- 10.6 SD) were significantly higher statistically than assessment with 3 readings after individual calibration (10.7 +/- 10.7) and with 1 reading (10.2 +/- 8.3) after individual calibration. Nevertheless, sensitivity, specificity, accuracy, and area under ROC curve did not change significantly.
Absence of individual calibration does not affect the laser fluorescence device's performance in detecting occlusal caries in primary teeth.
本研究旨在验证个体校准对健康牙齿表面的影响以及读数次数对用于检测乳牙咬合面龋的激光荧光（LF）设备体外性能的影响。
在进行标准校准（在瓷质参考物体上）后，使用DIAGNOdent对54颗已脱落或因正畸目的而拔除的乳牙磨牙咬合面上的72个临床可疑部位进行评估。首先，在进行个体校准（在每颗牙齿的健康表面上）后，对每个部位进行3次读数。随后，再进行3次读数，但不在牙齿的健康表面上进行个体校准。在这些评估之后，在个体校准后也仅对各部位进行一次读数评估。以组织学验证作为金标准。计算LF值、灵敏度、特异性、准确性（健康牙齿和患龋牙齿中正确诊断的数量），并使用McNemar变化检验进行比较。还比较了受试者操作特征（ROC）曲线下的面积。
仅进行标准校准的读数（平均值 = 11.7 +/- 10.6标准差）在统计学上显著高于个体校准后进行3次读数（10.7 +/- 10.7）和个体校准后进行1次读数（10.2 +/- 8.3）的评估结果。然而，灵敏度、特异性、准确性和ROC曲线下的面积没有显著变化。
不进行个体校准不会影响激光荧光设备检测乳牙咬合面龋的性能。
