Development and Validation of Generalized Linear Regression Models to Predict Vessel Enhancement on Coronary CT Angiography

OBJECTIVE: We evaluated the effect of various patient characteristics and time-density curve (TDC)-factors on the test bolus-affected vessel enhancement on coronary computed tomography angiography (CCTA). We also assessed the value of generalized linear regression models (GLMs) for predicting enhancement on CCTA. MATERIALS AND METHODS: We performed univariate and multivariate regression analysis to evaluate the effect of patient characteristics and to compare contrast enhancement per gram of iodine on test bolus (ΔHUTEST) and CCTA (ΔHUCCTA). We developed GLMs to predict ΔHUCCTA. GLMs including independent variables were validated with 6-fold cross-validation using the correlation coefficient and Bland–Altman analysis. RESULTS: In multivariate analysis, only total body weight (TBW) and ΔHUTEST maintained their independent predictive value (p < 0.001). In validation analysis, the highest correlation coefficient between ΔHUCCTA and the prediction values was seen in the GLM (r = 0.75), followed by TDC (r = 0.69) and TBW (r = 0.62). The lowest Bland–Altman limit of agreement was observed with GLM-3 (mean difference, −0.0 ± 5.1 Hounsfield units/grams of iodine [HU/gI]; 95% confidence interval [CI], −10.1, 10.1), followed by ΔHUCCTA (−0.0 ± 5.9 HU/gI; 95% CI, −11.9, 11.9) and TBW (1.1 ± 6.2 HU/gI; 95% CI, −11.2, 13.4). CONCLUSION: We demonstrated that the patient's TBW and ΔHUTEST significantly affected contrast enhancement on CCTA images and that the combined use of clinical information and test bolus results is useful for predicting aortic enhancement.

Effect of Patient Characteristics on Vessel Enhancement at Lower Extremity CT Angiography

OBJECTIVE: To evaluate the effect of patient characteristics on popliteal aortic contrast enhancement at lower extremity CT angiography (LE-CTA) scanning. MATERIALS AND METHODS: Prior informed consent to participate was obtained from all 158 patients. All were examined using a routine protocol; the scanning parameters were tube voltage 100 kVp, tube current 100 mA to 770 mA (noise index 12), 0.5-second rotation, 1.25-mm detector row width, 0.516 beam pitch, and 41.2-mm table movement, and the contrast material was 85.0 mL. Cardiac output (CO) was measured with a portable electrical velocimeter within 5 minutes of starting the CT scan. To evaluate the effects of age, sex, body size, CO, and scan delay on the CT number of popliteal artery, the researchers used multivariate regression analysis. RESULTS: A significant positive correlation was seen between the CT number of the popliteal artery and the patient age (r = 0.39, p < 0.01). A significant inverse correlation was observed between the CT number of the popliteal artery and the height (r = −0.48), total body weight (r = −0.52), body mass index (r = −0.33), body surface area (BSA) (r = −0.56), lean body weight (r = −0.56), and CO (r = −0.35) (p < 0.001 for all). There was no significant correlation between the enhancement and the scan delay (r = 0.06, p = 0.47). The BSA, CO, and age had significant effects on the CT number (standardized regression: BSA −0.42, CO −0.22, age 0.15; p < 0.05, respectively). CONCLUSION: The BSA, CO, and age are significantly correlated with the CT number of the popliteal artery on LE-CTA.

Effect of Heart Rate and Body Mass Index on the Interscan and Interobserver Variability of Coronary Artery Calcium Scoring at Prospective ECG-Triggered 64-Slice CT

OBJECTIVE: To test the effects of heart rate, body mass index (BMI) and noise level on interscan and interobserver variability of coronary artery calcium (CAC) scoring on a prospective electrocardiogram (ECG)-triggered 64-slice CT. MATERIALS AND METHODS:One hundred and ten patients (76 patients with CAC) were scanned twice on prospective ECG-triggered scans. The scan parameters included 120 kV, 82 mAs, a 2.5 mm thickness, and an acquisition center at 45% of the RR interval. The interscan and interobserver variability on the CAC scores (Agatston, volume, and mass) was calculated. The factors affecting the variability were determined by plotting it against heart rate, BMI, and noise level (defined as the standard deviation: SD). RESULTS: The estimated effective dose was 1.5 +/- 0.2 mSv. The mean heart rate was 63 +/- 12 bpm (range, 44-101 bpm). The patient BMIs were 24.5 +/- 4.5 kg/m2 (range, 15.5-42.3 kg/m2). The mean and median interscan variabilities were 11% and 6%, respectively by volume, and 11% and 6%, respectively, by mass. Moreover, the mean and median of the algorithms were lower than the Agatston algorithm (16% and 9%, respectively). The mean and median interobserver variability was 10% and 4%, respectively (average of algorithms). The mean noise levels were 15 +/- 4 Hounsfield unit (HU) (range, 8-25 HU). The interscan and interobserver variability was not correlated with heart rate, BMI, or noise level. CONCLUSION: The interscan and interobserver variability of CAC on a prospective ECG-triggered 64-slice CT with high image quality and 45% of RR acquisition is not significantly affected by heart rate, BMI, or noise level. The volume or mass algorithms show reduced interscan variability compared to the Agatston scoring (p < 0.05).