Assessment of vertical root fractures in dental volumetric tomography images using different kilovoltages (kVp)

The aim of this study was to assess the detectibiliy of vertical root fractures (VRF) on dental volumetric tomography (DVT) by using different kVp levels. Root canals of 48 extracted single rooted premolar teeth was enlarged with the rotating ProTaper (Dentsply/Maillefer, Tulsa, OK) system. Teeth were divided into four groups of 12 following the enlargement: group 1 (control group, no VRF + no filling), group 2 (VRF + no filling), group 3 (no VRF + filling) and group 4 (VRF + filling). Group 2 and 4 were consisted of 24 teeth with incomplete VRF. Incomplete fractures were created using the Instron device. Gutta-percha filling was applied to the 24 teeth included in group 3 and 4. All of the teeth were placed on a dry human mandible and DVT images were obtained with different kVp levels (60-70-80 kVp) using the Kodak 9000 3D (Kodak, Rochester, NY) device. Four different observers evaluated the presence and absence of VRF were scored from 576 images using a 5 level scale. Detectibility of VRF with different kVp levels was assessed with calculating the area under the curve (AUC) in ROC analysis (Az). Independent samples t-test and two-way ANOVA was used for the comparison of different kVp levels. Interobserver agreement was evaluated with the kappa test. Vertical root fracture detectibility score was the highest at 60 kVp (AZ =0.98), followed by 70 (AZ =0.96) and 80 kVp (AZ = 0.73) in teeth without filling. No significant difference was found between 60 and 70 kVp (p>0.05) regarding VRF detectibility in teeth without filling. However there was a statistically significant difference between 60-70 kVp and 70-80 kVp (p<0.05). Vertical root fracture detectibility score was the highest at 70 kVp (AZ =0.79), followed by 60 (AZ =0.75) and 80 kVp (AZ =0.63) in teeth with filling. No significant difference was found between 60-70 kVp, 70-80 kVp and 60-80 kVp regarding VRF detectibility in teeth with filling (p>0.05). In counclusion different kVp levels have an effect on imcomplete VRF detectibility on DVT images. Lower kVp levels performed better regarding VRF detectibility. Given the lower radiation doses and high diagnostic capacity, 60 kVp is recommended in evaulation of suspicious VRF's in teeth with or without canal filling.