Abstract: Objective To evaluate and compare the radiation dose and the number of images with valid data obtained during the plain scan and combined 3D reconstruction in CT-dacryocystography,and explore the necessity of 3D reconstruction for nasolacrimal duct obstruction diagnosis. Methods A retrospective analysis was performed in 96 patients with nasolacrimal duct obstruction requiring CT-dacryocystography,of which 49 patients underwent plain nasolacrimal CT scan (plain scan group),and the other 47 patients underwent plain nasolacrimal CT scan + 3D reconstruction (reconstruction group).The dose of radiation exposed during the examination,the number of total CT images obtained,and the number of CT images with valid data of lacrimal tract were compared and analyzed between the two groups. Results The reconstruction group was exposed to more radiation (504.7±405.8mGy×cm vs 363.8±159.9mGy×cm,P=0.026) than the plain scan group,and obtained more CT images (289±74 vs 230±57,P=0.000) and CT images with valid data (132±36 vs 112±29,P=0.003),but the proportion of CT images with valid data in the reconstruction group was lower (48.70±0.71% vs 45.67±0.51%,P=0.000). Conclusion The combined three-dimensional reconstruction technique in CT-dacryocystography increases radiation exposure risk compared to plain scan.It obtains more CT images of lacrimal ducts,but the efficiency is relatively low.Reducing radiation exposure that is not related to the imaging of the lacrimal duct should be considered in both of the two methods.