In vivo validity of proximal caries detection in primary teeth, with histological validation.

Abstract

Detection and diagnosis of proximal caries in primary molars are challenging. The aim of this in vivo study was to assess the validity and reproducibility of four methods of proximal caries detection in primary molar teeth. Eighty-two children (5-10years) were recruited. Initially, 1030 proximal surfaces were examined using meticulous visual examination (ICDAS) (VE1), bitewing radiographs (RE), and a laser fluorescence pen device (LF1). Temporary tooth separation (TTS) was achieved for 447 surfaces, and these were re-examined visually (VE2) and using the LF pen (LF2). Three hundred and fifty-six teeth (542 surfaces) were subsequently extracted and provided histological validation. At D1 (enamel and dentine caries) diagnostic threshold, the sensitivity of VE1, RE, VE2, LF1, and LF2 examination was 0.52, 0.14, 0.75, 0.58, and 0.60 and the specificity values were 0.89, 0.97, 0.88, 0.85, and 0.77, respectively. At D3 (dentine caries) threshold, the sensitivity values were 0.42, 0.71, 0.49, 0.63, and 0.65, respectively, whereas specificity was 0.93 for VE1 and VE2, and 0.98, 0.87, and 0.88 for RE, LF1, and LF2 examinations, respectively. ROC analysis showed radiographic examination to be superior at D3 . Meticulous caries diagnosis (ICDAS) should be supported by radiographs for detection of dentinal proximal caries in primary molars.