Radiation Doses in Cardiovascular Computed Tomography

**Background:** Cardiovascular computed tomography (CT) has become a cornerstone in diagnosing coronary artery disease, with guidelines from the European Society of Cardiology recommending it as an initial test for symptomatic patients. However, CT is a significant source of ionizing radiation exposure, and its use has increased dramatically—approximately 20-fold in the United States from 1990 to 2010. The harmful effects of ionizing radiation are classified as stochastic (e.g., cancer, genetic damage) and deterministic (e.g., cataracts, hair loss). The linear no-threshold model assumes that even low doses carry some cancer risk. According to the BEIR VII report, for every 100,000 people exposed to 100 mSv, the lifetime attributable risk (LAR) is 510 in 100,000 (5 in 1000). For 10 mSv, LAR is about 1 in 1000. Radiation also contributes to cardiovascular disease via endothelial damage, oxidative stress, and accelerated atherosclerosis. This review aims to summarize current radiation doses in cardiovascular CT, dose definitions, optimization methods, and the need for updated conversion factors.

**Methods:** This is a narrative review of published literature on radiation doses in cardiovascular CT. The authors reviewed studies including the CORE-64 study, the PROTECTION I and VI trials, the German Cardiac CT Registry, the CRESCENT study, and data from centers in China, Saudi Arabia, Australia, France, Taiwan, and the UK. They also examined dose definitions (CTDIvol, DLP, SSDE, effective dose) and optimization techniques such as tube voltage reduction, ECG-modulated tube current, iterative reconstruction, deep learning reconstruction, prospective acquisition, and high-pitch scanning. Data on TAVI (transcatheter aortic valve implantation) protocols were also reviewed.

**Key Results:** The review reports a substantial reduction in radiation doses over time. In the PROTECTION I study (2007), the diagnostic reference level (DRL) for coronary CT angiography (CCTA) was 1200 mGy*cm. By PROTECTION VI (2017), the median DLP for CCTA was 195 mGy*cm (IQR 110–338), corresponding to an effective dose of 2.7 mSv (using k=0.014) or 5.1 mSv (using k=0.026). The median total DLP for all 4502 patients was 252 mGy*cm (IQR 154–412). The CORE-64 study reported a mean effective dose of 19 mSv (range 16–26) for CCTA, with organ doses as high as 38 mSv for breasts and 35 mSv for lungs. In the University Center in Nanjing (2007–2016), median DLP decreased from 661.2 mGy*cm to 268.2 mGy*cm. The German Cardiac CT Registry showed a decline from 397 mGy*cm (2009–2010) to 176 mGy*cm (2011–2017). The PROTECTION VI study found that low tube voltage (80 kVp) reduced DLP by 68% compared to 120 kVp, and 90–100 kVp reduced DLP by 50%. Iterative reconstruction reduced dose by 33% compared to filtered back projection. Deep learning reconstruction enabled dose reductions of 30–80% compared to iterative reconstruction. For TAVI protocols, median DLP varied widely: 675 mGy*cm (IQR 477–954) in a UK survey, 357 mGy*cm with high-pitch non-ECG-gated protocols, and as low as 201 mGy*cm in some studies. The review also highlights that the organ conversion k-factor for cardiac CT may be higher than the standard chest CT factor (0.014–0.017 mSv/mGy*cm). Studies suggest k-factors of 0.026–0.030 mSv/mGy*cm for cardiac CT, meaning effective doses may be underestimated by 30–60%.

**Clinical Implications:** The review emphasizes that radiation doses in cardiovascular CT have decreased significantly due to technological advances and protocol optimization, but doses remain a concern, especially for younger patients and women (breast exposure). The lifetime risk of cancer from a single CCTA is estimated at 0.05% for a 10 mSv dose, but benefits (e.g., 50% reduction in fatal/non-fatal myocardial infarctions over 3 years) generally outweigh risks. The authors stress the importance of adhering to the ALARA principle, using dose reduction techniques (low kVp, iterative reconstruction, prospective gating, heart rate control), and avoiding unnecessary scans. They also note that current dose estimates may be systematically underestimated if standard chest CT conversion factors are used, and recommend adopting cardiac-specific k-factors. Diagnostic reference levels (e.g., DLP 400 mGy*cm from PROTECTION VI) can help centers benchmark and optimize their protocols.