Virtual Library

Start Your Search

C. Hui



Author of

  • +

    P1.06 - Poster Session/ Screening and Early Detection (ID 218)

    • Event: WCLC 2015
    • Type: Poster
    • Track: Screening and Early Detection
    • Presentations: 1
    • +

      P1.06-001 - Ultra-Low Dose-CT Accurately Detects Significant Lung Nodules with a Fraction of the Radiation of Conventional Low Dose-CT (ID 2621)

      09:30 - 17:00  |  Author(s): C. Hui

      • Abstract

      Background:
      Indeterminate lung nodules are a common and increasing incidental finding on CT imaging and there are widely accepted surveillance protocols. However, even when using Low Dose (LD)-CT with a total effective dose of ~1mSv, concerns exist regarding the cumulative radiation exposure of subjects under surveillance, particularly in individuals not at high risk of lung cancer. By utilizing the Model Based Iterative Reconstruction (MBIR) technique, CT images can be obtained with a radiation dose comparable to chest x-ray (0.06-0.1 mSv). At this Ultra-Low Dose (ULD), MBIR images have generally less signal to noise ratio which may prevent small nodule detection. The aim of this prospective study was to assess the efficacy of ULD-CT in detecting clinically significant lung nodules (≥4mm) as compared to LD-CT.

      Methods:
      Following approval from the local Human Research Ethics Committee, adult subjects undergoing CT surveillance for incidental lung nodules were recruited from a tertiary hospital. Once informed consent was obtained, both standard LD- and a ULD-CT chest were performed. Scans were performed on the GE750HD Discovery scanner. Demographic information including lung cancer risk factor evaluation was obtained by questionnaire. Patients who withdrew consent or whose images were degraded by gross movement or metallic artefacts were excluded. Images from the ULD-CT were reconstructed with MBIR prior to reading. Each of LD/ULD-CT image sets was read blindly, randomly and independently by two experienced thoracic radiologists. The number, size and location of nodules was reported and subsequently compared.

      Results:
      100 subjects were recruited with a mean age of 65 years (range 32-87). Around 62% were ever smokers, with 30% smoking ≥30 pack years. Around 30% had risk factors other than smoking, but only ⅓ of these (9%) did not have a significant smoking history. Only a small proportion were high risk as evidenced by only 8 meeting Lung Cancer screening criteria (NLST criteria). A total of 200 nodules ≥4mm were detected, with all seen on both LD and ULD-CTs. In addition, there were 244 nodules <4mm seen on the LD-CT, with greater than 80% sensitivity for the ULD-CT, with minor variation between lobes. There were no false positive findings. There was a 10 fold reduction in effective radiation when comparing ULD-CT (0.09mSv) imaging with the standard LD-CT (1.11mSv). Lung nodules were subjectively better seen on the ULD-CT.

      Conclusion:
      ULD-CT with the advanced MBIR allows detection of all clinically significant lung nodules while achieving a radiation dose comparable to that of plain chest radiography. Particularly in low-risk populations, the use of ULD-CT for surveillance of lung nodules has the potential to significantly reduce cumulative radiation exposure.