Reduced contrast volume in coronary CT angiography: effects on coronary attenuation

Abstract

Abstract Background Coronary CT angiography (CCTA) requires intravenous iodinated contrast to achieve diagnostic intraluminal opacification. Because reducing the contrast volume administered per examination is an exposure-reduction strategy of practical interest, institutions have explored lower-volume protocols; whether attenuation can be preserved above the diagnostic threshold under real-world, non-randomized implementation is not well documented outside research settings. Methods This retrospective, single-centre, non-concurrent before-and-after protocol-implementation study was conducted at Radiology Center (January 2023–December 2024). From 200 CCTA examinations initially identified from radiology reports as potentially eligible, 106 were excluded (44 body weight >120 kg; 12 beta-blocker administered; 27 coronary calcification/calcium scoring without significant stenosis; 23 motion artefact), leaving 94 CT-angiographically normal examinations: 51 under the standard (pre-reduction) protocol and 43 under the reduced-volume (post-reduction) protocol. All examinations used a third-generation dual-source CT scanner (Siemens SOMATOM Force) with automated tube-voltage selection (CARE kV) and tube-current modulation (CARE Dose4D), retrospective ECG gating, and a 32-cm scan length. Omnipaque 350 mgI/mL was administered by power injector with bolus tracking (ascending aorta ROI, 90 HU trigger, 5 s post-trigger delay). Intraluminal attenuation was measured by a single reader in the left main trunk (LMT) and proximal right coronary artery (RCA); no inter- or intra-observer reproducibility assessment was performed. Distributions were assessed with the Shapiro–Wilk test; normally distributed variables were compared with the independent-samples (Welch) t-test and non-normally distributed variables with the Mann–Whitney U test. A multivariable linear model examined coronary attenuation as a function of protocol epoch, kVp, body weight, heart rate, and iodine delivery rate. Results Contrast volume fell from 63.4 ± 7.3 mL to 49.0 ± 8.0 mL (median 60 [IQR 60–65] vs 45 [45–52.5] mL), a 22.8% reduction (Mann–Whitney p < 0.001), with a corresponding 22.8% reduction in iodine mass (22.20 ± 2.56 vs 17.13 ± 2.79 gI). Iodine delivery rate was essentially unchanged (1.69 ± 0.14 vs 1.69 ± 0.14 gI/s; p = 0.91). Mean LMT attenuation decreased from 520.4 ± 93.6 HU to 451.9 ± 90.7 HU (− 13.2%; t-test p < 0.001) and mean RCA attenuation from 500.2 ± 107.1 HU to 418.1 ± 89.3 HU (− 16.4%; Mann–Whitney p < 0.001). At the per-patient level, 41/43 post-reduction examinations (95.3%; exact 95% CI 84.2–99.4%) reached ≥ 300 HU at the LMT and 39/43 (90.7%; 95% CI 77.9–97.4%) at the RCA; 4/43 examinations, therefore, fell below the SCCT diagnostic threshold at the RCA and 2/43 at the LMT. In the multivariable model, protocol epoch remained independently associated with lower attenuation after adjustment for kVp, weight, heart rate, and iodine delivery rate (LMT: β = − 75.7 HU, 95% CI − 111.6 to − 39.8, p < 0.001; RCA: β = − 87.8 HU, 95% CI − 128.0 to − 47.7, p < 0.001). Mean DLP was 1158.2 ± 637.4 mGy·cm (pre) versus 1072.1 ± 530.6 mGy·cm (post), a non-significant 7.4% difference (Mann–Whitney p = 0.71); the reduction in DLP standard deviation (16.8%) corresponds to an estimated 30.7% reduction in variance, which was not statistically significant (Levene’s test p = 0.24). No renal function or image-quality metric was measured. Conclusion A reduced-volume contrast protocol lowered administered contrast volume and total iodine mass by 22.8% while iodine delivery rate was maintained. Mean coronary attenuation remained above 300 HU in both groups, but a per-patient minority of post-reduction examinations fell below this threshold, particularly at the RCA. Radiation-dose differences were a co-temporal secondary observation and are not attributable to the contrast protocol. Renal outcomes were not assessed, and the findings should not be interpreted as evidence of nephroprotection or as a formal non-inferiority result.