1. Sample
This work evaluated a total of 71 hepatic CT studies, 33 from Hospital_A and 38 from Hospital_B. Fig. 2 and Fig. 3 resumed patient distribution according to age and gender, respectively.
Fig. 2: Box plot of the sample distribution according to patient's age and institution.
References: School of Health Sciences, University of Aveiro (ESSUA) - Aveiro/PT
Fig. 3: Box plot of the Sample distribution according to patients's gender
References: School of Health Sciences, University of Aveiro (ESSUA) - Aveiro/PT
With exception to gender, which had different distribution, the patient sample had no statistically significant differences between institutions, as observed in
Table 2.
Table 2: Resume table of the patients indicators comparative analysis between institutions.
References: School of Health Sciences, University of Aveiro (ESSUA) - Aveiro/PT
2. Injection Protocol Analysis
Concerning injection protocol parameters, analysed institutions presented relevant differences. Beyond the use or non-use of bolus tracking, injection flow rate and ICM dose presented statistically significant differences between Hospital_A and Hospital_B, while delays, arterial and portal venous, didn’t. Results resumed in Table 3.
Table 3: Resume table of the ICM injection protocol data with comparative analysis between institutions.
References: School of Health Sciences, University of Aveiro (ESSUA) - Aveiro/PT
3. Structures Enhancement quantitative analysis
Regarding structures enhancement, Table 4 resumed obtained results.
Table 4: Resume table of the structures enhamencement in arterial and portal venous acquisitions phases with comparative analysis between institutions.
References: School of Health Sciences, University of Aveiro (ESSUA) - Aveiro/PT
As observed in Table 5, multivariate analysis of variance presented statistically significant variation in the structures enhancement when protocol injection parameters (injection flow rate and ICM dose) were analysed.
Table 5: Resume table of the multivariate analysis of variance of structures enhancemente acoording to injection flow rate, ICM dose, and injection flow rate and ICM dose set.
References: School of Health Sciences, University of Aveiro (ESSUA) - Aveiro/PT
Univariate analysis of variance of the parameters set (flow rate*ICM dose), obtained from Tests of Between-Subjects Effects, showed that this variation had more relevance in the portal venous phase enhancement, especially in aorta artery [F(4,10) = 23.00, p < .001] and portal vein [F(4,10) = 6.33, p = .008].
Isolated, injection flow rate presented similar results (aorta artery [F(2,10) = 22.17, p < .001] and portal vein [F(2,10) = 3.97, p = .054]). Considering that, Tukey tests were performed to evaluate differences in structures enhancement between the injection flow rates, and generally, 3.5 and 4.0 mL/s presented statistically significantly increased enhancements when compared with 3.0 mL/s. Fig. 4 and Fig. 5 resumed results.
Fig. 4: Box and whisker plot of structures arterial enhancement according to used injection flow rate. [(*) - statistically significant differences between 3 mL/s and other flow rates]
References: School of Health Sciences, University of Aveiro (ESSUA) - Aveiro/PT
Fig. 5: Box and whisker plot of structures portal venous enhancement according to used injection flow rate. [(*) - statistically significant differences between 3 mL/s and other flow rates]
References: School of Health Sciences, University of Aveiro (ESSUA) - Aveiro/PT
As injection flow rate, isolated ICM dose also led to portal venous structures enhancement variation, even in this case, only aortic portal venous enhancement presented statistically significance variation [F(51,10) = 11.24, p < .001]. Also, ICM dose presented statistically significant Pearson correlations with aortic arterial enhancement and all three portal venous structures enhancement (respectively aortic arterial enhancement:[r = .512, n = 69, p < .001]; aortic portal venous enhancement: [r = .672, n = 68, p < .001]; portal vein portal venous enhancement: [r = .421, n = 68, p < .001] and hepatic parenchyma portal venous enhancement: [r = -.329, n = 68, p = .006]).
In the delay definition, the three analysed methodologies resulted in statistically significant variation in the enhancement of the structures [Wilk's Λ = 0.000; F(12, 120) = 16.14, p < .001], which Tests of Between-Subjects Effects showed that just portal vein arterial enhancement was not influenced. In general, pre-established delays methodology presented statistically significant differences in the Tukey test when compared with bolus tracking methodologies (except portal vein and hepatic parenchyma arterial enhancements). Bolus tracking methodologies just presented differences between them in aorta and portal vein portal venous delays. Fig. 6 and Fig. 7 resumed results. It was important to point out that there was a statistically significant difference in ICM dose and injection flow rate between pre-established methodology and both bolus tracking methods.
Fig. 6: Box and whisker plot of structures arterial enhancement according to delay selection method. [(*) - statistically significant differences between pre-stablished delays and bolus tracking methods]
References: School of Health Sciences, University of Aveiro (ESSUA) - Aveiro/PT
Fig. 7: Box and whisker plot of structures arterial enhancement according to delay selection method. [(*) - statistically significant differences between pre-stablished delays and Bolus tracking methods; (**) - statistically significant differences between each 3 delay selection methodologies]
References: School of Health Sciences, University of Aveiro (ESSUA) - Aveiro/PT