The human body is comprised of body water, protein, minerals, and body fat. The weight of human tissues such as muscles and somatic cells can be calculated from their relationship to the different components.
To our knowledge, this is the first clinical study to assess the effect of items included in BIA on IDCE on CE-CT scans of the aorta and hepatic parenchyma. In BIA, the body water- and the body components comprising the weight of the human body are further subdivided into their constituents and calculated.
The TBW exhibited the greatest effect on IDCE of the abdominal aorta scanned during the HAP, it also had the greatest effect on hepatic parenchymal enhancement observed during the PVP. The SMI most strongly affected enhancement of the hepatic parenchyma during the PVP. The other components addressed in BIA showed a moderate to high correlation with the TBW and SMI.
Earlier studies suggested that among assessed body parameters, the LBW and the BSA exhibited the strongest correlation with aortic and hepatic enhancement 14,17. The most commonly used parameter for determining the iodine dose has been the TWB. However, in obese patients, the iodine dose may be excessive because a large proportion of their TBW is comprised of poorly perfused adipose tissue in which iodine distributes poorly 18–21. In small patients, it may also be excessive.
Ours diverge from earlier findings possibly because the difference between the LBW or the BSA and the TBW was relatively small with respect to IDCE reported earlier 14. Also, while the LBW and the BSA are factors that influence IDCE more strongly than the TBW in overweight and underweight patients, the weight distribution in our subjects was very narrow and there were only a few high-weight patients.
We found that during the PVP, the SMI exhibited the greatest effect on IDCE of the hepatic parenchyma. The SMI was calculated by dividing the limb skeletal muscle mass (kg) by the square of the height (m2). The SMI is an index for the skeletal muscle mass and it is used as a criterion for the diagnosis of sarcopenia, defined as an age-related loss in the skeletal muscle mass that results in a reduction in muscular strength and physical function 22. The blood vessels in muscles contain a larger blood volume than does fat; this is thought to exert a strong effect on contrast enhancement 23–25. The SMI is highly correlated with liver fibrosis and a correlation between liver fibrosis and contrast enhancement of the hepatic parenchyma has been reported 26. Consequently, IDCE on CE-DCT scans may be strongly affected by the SMI.
We found that many body parameters included in BIA affect IDCE on CE-DCT images. Most exhibited a more than moderate correlation with IDCE of vessels and the hepatic parenchyma.
The performance of machine-learning algorithms applied to medical data is high and machine learning is useful for predicting clinical outcomes 27,28. Therefore, we think that the integration of parameters included in BIA into machine learning may yield more accurate predictions.
Our study has some limitations. Ours was a retrospective, single-center study and the study population was small. As the median and the range of the TBW of Japanese- is lower than of Western populations, our findings cannot be extrapolated to non-Japanese individuals. In this study we did not compare contrast enhancement with the BSA index, we did not confirm the relationship between contrast enhancement and the image quality, and we did not address the detectability of liver tumors. These issues are currently under investigation.
In conclusion, BIA showed that among the components comprising the weight of the human body, the TBW had the strongest effect on aortic and hepatic enhancement on CE-DCT scans. With respect to hepatic parenchymal contrast enhancement during the PVP; the SMI exhibited the greatest effect.