Dose-Quality Study in the Temporal Bone Computed Tomography in Routine Care

Overview

Because of the complexity of the structures studied, imaging the temporal bone requires compliance with a number of image quality criteria, in order to achieve optimum spatial resolution of the most clinically relevant anatomical landmarks. On the other hand, limiting patient exposure to ionizing radiation is a growing concern for the medical world. The challenge here is twofold: to meet a certain number of image quality criteria, in order to achieve optimum spatial resolution for the most clinically relevant anatomical landmarks, while aiming to reduce the radiation dose delivered to patients as much as possible. Despite its routine use, few studies have compared the various commercially available solutions in terms of image quality or dose delivered, in daily practice. The primary objective of our study is a qualitative comparison of 7 different types of CT scans on 17 anatomical structures, useful in clinical practice, taking into account the contrast-to-noise ratio (CNR). The secondary endpoint was a comparison of radiation levels.

Because of the complexity of the structures studied, imaging the temporal bone requires compliance with a number of image quality criteria, in order to achieve optimum spatial resolution of the most clinically relevant anatomical landmarks. The spatial resolution of conventional high-resolution CT is 0.5mm, whereas the dimensions of the smallest structure in the middle ear, the stapes, are of the order of 0.19mm). Despite their small size, pathological changes to these structures can have major consequences for hearing and balance, and may require exploratory or reconstructive surgery. On the other hand, limiting patient exposure to ionizing radiation is a growing concern for the medical world, patients and public authorities alike. In the literature, effective doses vary from 50 to 970 µSv, and most protocols expose the crystalline lens (direct exposure), varying from 10 to 40 mGy in helical acquisition, depending on the technical protocol. The International Commission on Radiological Protection has therefore proposed that the "As Low As Reasonably Achievable Radiation Exposure" (ALARA) principle should be followed, i.e. maintaining optimum spatial resolution, particularly for key anatomical landmarks in rock imaging, while reducing noise levels and radiation dose. The challenge here is twofold: to meet a certain number of image quality criteria, in order to achieve optimum spatial resolution for the most clinically relevant anatomical landmarks, while aiming to reduce the radiation dose delivered to patients as much as possible. Despite its routine use, few studies have compared the various commercially available solutions in terms of image quality or dose delivered, in daily practice. To the best of our knowledge, no Diagnostic Reference Level has yet been established in the literature. The idea would therefore be to determine objective quality criteria, based on key, clinically relevant anatomical structures, evaluated in several French centers, as part of a standard offer of care, in order to seek the right balance between the dose delivered to the patient and the quality of the image required, and thus provide clinicians with precise, reproducible tools for imaging the rock, for diagnostic or interventional purposes. The primary objective of our study is a qualitative comparison of 7 different types of CT scans on 17 anatomical structures, useful in clinical practice, taking into account the contrast-to-noise ratio (CNR). The secondary endpoint was a comparison of radiation levels. This is a retrospective data collection, carried out between 2023 and 2024, including 60 patients who underwent a CT scan of the temporal bone in a hospital center in mainland France, during their course of care. Image quality was assessed independently, in a blinded fashion, by two radiologists and two ENT physicians, on the basis of a 5-point Likert semi-quantitative visual scale. To calculate radiation exposure, the dose volume index (CTDIvol) was retrieved from the dose report for each patient, and the length-dose product (LDP) was provided automatically by the various consoles.