Cardiac variation of IJV has not yet been widely evaluated and our results indicate that it is as useful as respiratory variation of IJV in evaluating intravascular volume status. This study demonstrates for the first time that cardiac variation of IJV, measured using an automatic tracing programme, can detect not only intravascular volume loss but also response to fluid administration. Using a clinical model of blood donation by preoperative stable patients is another unique point of this study, because this situation mimics classⅠhaemorrhagic shock that was promptly treated by fluid resuscitation.
Significant changes were observed in cardiac variation of IJV during blood donation and subsequent fluid resuscitation (Table 2), while clinical parameters such as BP and HR could not detect changes in volume status, especially during fluid administration (Table 2), indicating the superiority of cardiac variation indices of IJV over regular vital signs.
Among the cardiac variation parameters evaluated, collapse indices CIa and CIc could better predict change of intravascular volume compared to absolute values of area and circumference, suggesting that interindividual anatomical differences in IJV need to be considered. Additionally, our data show a temporal trend rather than single timepoint change, and this is expected to be better than CVP monitoring described previously 19, 20, because the relative temporal changes in CIa and CIc were more sensitive than clinically used parameters such as BP and HR (Table 3).
Several previous studies have investigated the utility of IJV diameter measurement by ultrasonography for intravascular volume assessment during blood donation11, 12. The reported variation in IVJ area during inspiration and expiration was 47 ± 18% before donation and 73 ± 18% after donation11. Another study has reported on indirect CVP, defined as vertical height of IJV from the point of complete collapse to the sternal angle of Louis detected by ultrasonography plus 5 cm, to be 6.67 cm (95%CI 6.72–7.07) before donation and 5.98 cm (95%CI 6.09–6.40) after donation12. These studies indicate that respiratory variation of IJV might be a reliable indicator of intravascular volume loss.
Previously, we have shown an increase in the collapse index of IVC during dehydration using a passive leg raising test and medical anti-shock trousers15. Further, we found a correlation between collapse index of IJV and indirect CVP after dehydration using a sauna16; however, these preliminary studies had small sample sizes. Here we show that the collapse index of IJV could be used to monitor dynamic changes in volume status induced by blood loss (200 ~ 400 ml blood donation) and fluid administration (500 ml crystalloid), which are clinically more relevant than those observed in our previous studies. Further, the sample size in this study was much larger (n = 104), but further investigation is necessary to verify these observations in clinically relevant situations such as haemorrhagic trauma and septic shock.
The image tracking software used enabled the analysis of the collapse index because its variation was small over a short period of time. The advantages of the software developed by us for IJV images are objective assessment and automatic analysis. In contrast, in previous studies, respiratory variation analysis was almost always done by a physician, which may be subject to many biases in each physician’s skill set13,14. The extraction programme for blood vessel boundary identification from ultrasound data used in this study was based on snake and speckle tracking models, which is an active contour method that can trace objects in the previous image18. Importantly, the software can be manually adjusted to mark the exact boundary, if there were inter-rater disagreement in measurement. Software use allows better objective assessment than physician measurements, and the development of software with respect to analysis speed and tracking ability will help improve volume status assessment in the clinical settings. Recently, a wearable ultrasound patch monitor that can continuously measure BP change has been developed, and our results indicate a potential for continuous monitoring of cardiac variation of IVJ for intravascular volume evaluation. Future devices that can measure cardiac variation of IVJ continuously will further develop 0volume evaluation technology in the future21.
This study has several limitations. First, a comparison of cardiac variation with respiratory variation was not conducted, and many previous studies have reported on the performance of respiratory variation for evaluating intravascular volume loss and fluid challenge response22,23. Other clinically available indicators such as CVP, SVV and CRT need to be compared with cardiac variation of IVJ16, 25. Second, we cannot exclude the possibility that the patients’ sympathetic nerve activation might have affected the obtained results. Although the enrolled patients took rest before blood donation, HR decreased just after donation was started, which may be because of relief from psychological stress. Third, age and BMI affected CIa and CIc; therefore, CIa and CIc may not be able to correctly predict fluid status in high and/or low BMI, and older age patients. It is necessary to analyse BMI and age stratified data in a larger cohort.