Introduction Experimental evaluation of endometric devices usually relies on visual, subjective detection of the apical constriction to determine the accuracy of measurements. The aim of the present study was to analyze the accuracy of measurements of Raypex 6 and EndoPilot using a novel, objective image-analysis system. Methods Onehundred and twenty teeth were randomized and allocated to three groups: After coronal flaring, either Raypex 6 or EndoPilot were used to determine the endodontic working length during instrumentation using manual files (RPM and EPM group respectively). In addition, EndoPilot was used for continuous, automatic measurement during rotating instrumentation (EPA group). If the working length had been reached according to endometric results, the files were fixed in place. Tooth and file were then embedded and prepared for analysis. Subsequently, the distance between the tip of the file and the apical constriction (DAC) or the apical foramen (DAF) was calculated using trigonometric analysis and the position of the file relative to AC and AF was analyzed. Results Both inter- and intra-examiner-reliability of the trigonometric analysis were nearly perfect (ICC = 0.999, p<0.001). DAC was not significantly different between groups (p>0.05, t-test). DAF was significantly decreased when EPA had been used compared to EPM (p<0.05, Exact-test). EPA resulted in files being positioned beyond AF significantly more often than the other two methods (p<0.01). Conclusions All methods allowed reliable detection of AC. However, EPA significantly increased the risk of overpreparation. Objective, digital assessment based on image analysis was suitable to compare the accuracy of different endometric devices.
