Lack of the uniqueness of the human dentition for bitemark investigations in orthodontically treated patients and twins

The uniqueness of the human dentition (UHD) became one of the most polemic fields in forensic dentistry in the last decade. Criticism arose in the scientific community1, especially in the interface of the UHD and bitemark analysis2–4. The scientific literature demonstrates a dissonant opinion concerning the existence or not of the UHD on the level of bitemarks5–7. The present study aimed to investigate the uniqueness of the human dentition on the level of bitemarks in human skin, systematically modifying the amount of dental material analyzed in the anterior dentition. The sample consisted of 445 dental casts digitalized with an automated motion device (XCAD 3D® (XCADCAM Technology®, São Paulo, SP, Brazil) and implemented in Geomagic Studio® (3D Systems®, Rock Hill, SC, USA) software package. The dental casts derived from randomly selected patients (n=22), orthodontically treated patients (n=59), twins (n=344), and orthodontically treated twins (n=20). Besides, casts (n=20) were collected from subjects at two different moments, to establish threshold values indicating whether or not two dentitions were equal. Within each group pair wise comparisons were performed with the Geomagic Studio®software, using automated superimposition. The comparisons were performed on 4 levels, systematically reducing the amount of dental material analyzed (level 1: 3mm of the incisal edges; level 2: 2mm of the incisal edges; level 3: 1mm of the incisal edges; level 4: a slice of 1mm not including the incisal edges of the anterior dentition). The morphological differences between dentitions were quantified separately for the maxilla and the mandible, for each groups, and for each level, and expressed in four values: the maximum positive and negative deviations, the average and the standard deviation. The four quantified values were log transformed in a single mean Euclidean distance and compared with the threshold of the respective level, using ANOVA test with a significance interval of 5%. A ROC analysis was performed to test the predictive value of the threshold group to classify unique and not unique dentitions. In all the study levels, the mean Euclidean distances of the maxilla and the mandible of orthodontically treated patients, twins and orthodontically treated twins remained below the threshold (p<0.05). The opposite was observed with the randomly selected subjects. The ROC curve resulted optimistic for the study levels 1, 2 and 3 reaching sensitivity and specificity rates between 70-100%, while it decreased to 55-80% in study level 4. Apart of that, the Area Under the Curve remained >71% in all the study levels, suggesting a proper threshold for the classification of unique and not unique dentition. The lack of the uniqueness of the human dentition on bitemark level observed in the present study was proven with advanced scientific standards. However, this finding must be carefully interpreted. In specific, the bitemark practice must not be disregarded, but applied after careful case selectionstatus: publishe