Comparing the reliability and accuracy of clinical measurements using plaster model and the digital model system...

This study aims to clarify whether 3Shape™ digital model system could be applied in orthodontic diagnostic analysis with certainty, especially under different crowding condition. Reliability, accuracy and efficiency of 3Shape™ digital model system were assessed by comparing them with traditional plaster cast.

29 plaster casts with permanent dentition were transformed into digital models by 3Shape™ D800 scanner. All 29 models were categorized into mild-crowding (arch length discrepancy <3 mm), moderate-crowding (arch length discrepancy >3 mm and <8 mm), and severe-crowding group (arch length discrepancy >8 mm). Fourteen linear measurements were made manually using a digital caliper on plaster casts and virtually using the 3Shape™ Ortho Analyzer software by two examiners. Intra-class Correlation Coefficient (ICC) was used to evaluate intra-examiner reliability, inter-examiner reliability and reliability between two model systems. Paired t test was used to evaluate accuracy between two model systems. Kruskal-Wallis test followed by Mann-Whitney U test was used to evaluate the measurement differences between 3 groups in two model systems.

Both intra-examiner and inter-examiner reliability were generally excellent for all measurements made on 3Shape™ digital model and plaster cast (ICC: 0.752-0.993). Reliability between different model systems was also excellent (ICC: 0.897-0.998). Half of the accuracy test showed statistically significant differences (p < 0.05) when digital models were compared with plaster casts. Furthermore, while assessing measurement differences between 3 groups in two model systems, the mandibular required space showed significant difference (p = 0.012) between mild crowding group (0.27 + 0.01 mm) and severe crowding group (0.20 + 0.09 mm). However, the differences were less than 0.5 mm and would not affect clinical decision.

Using 3Shape™ digital model system instead of plaster casts for orthodontic diagnostic measurements is clinically acceptable.