Evaluation of the confounding factors disturbing thequality of CT images of pulmonary angiography

Since pulmonary angiography is the only non-invasive method for diagnosing pulmonary embolism, the use of radiation reduction methods without affecting the quality of images seems necessary. One of the simplest reasons for unnecessarily increasing of patient exposure is repetitive imaging due to poor image quality. Therefore, it is necessary to identify and eliminate the factors that reduce the image quality, which leads to duplication of imaging or reduced accuracy.  In this study, the reasons for decreasing the image quality of pulmonary computed tomography (CT) angiography were investigated.

This cross-sectional study was performed over a period of 6 months (May to November 2018) in a university-affiliated hospital. Data of 75 patients suspected of pulmonary embolism who were referred for CT angiography were collected. All CT scans were performed by a 16-slice CT scanner (Lights Speed, GE Healthcare) with fixed imaging parameters including 100 kVp and 100 mA. In order to evaluate the adequacy of pulmonary artery enhancement, the Hounsfield unit (HU) of the main pulmonary artery was measured in the aortic arch, upper, and lower segmental parts. Also, 8 factors affecting the image quality were examined.

The mean HU of the main pulmonary arteries was 312 in the range of 124 to 677. The mean HU in patients with and without pulmonary embolism was 358 and 302, respectively. Most CT images (61%) had poor quality for the diagnosis of pulmonary embolism. Only 2% of the images were acceptable compared to standard images and protocols. Late imaging was the most common quality confounding factor in this study.

A low-quality image reduces the accuracy of the physicianchr('39')s diagnosis that leads to misdiagnosis. Besides, sometimes poor image quality leads to repetitive imaging, which results in increased patient exposure and therefore increased radiation hazards. Therefore, modifiable confounding factors must be identified and corrected which one of the most important ones is wrong imaging timing protocols after contrast injection.