Although AVS is a well-established technique, the development of a more convenient and simple method is desired. Multi-row detector CT and magnetic resonance imaging (MRI) are commonly used in the preoperative delineation of the adrenal veins; however, there is still a degree of inadequacy in the detection rate of RAV [6–8, 16]. Although MRI is effective in avoiding the radiation exposure or complications resulting from contrast agent, CT has acquired detectability of RAV than non-contrast-enhanced MR techniques such as 4D-flow MRI (7). Because it requires a special sequence, which may be unavailable in some hospitals, and the process for obtaining images is long. Another reason to employ 4D CT is the sufficient image quality, in which submillimeter resolution is available. Also, existing study shows the difficulty in quantifying blood flow reliably through MRI acquisitions from irregular cardiac cycles (16). Additionally, the scan delay for the optimal visualization of RAV remains controversial. Ota et al. [7] reported that obtaining the late arterial phase 13 seconds after the first scan allowed the RAV to be delineated in 93% of cases. Degenhart et al. (6) performed scanning at a 90-second delay, yielding a rate of only 70–88% in the late portal phase. However, Morita et al. (17) indicated that it was possible to visualize RAV in the single venous phase and, thus, dual adrenal venous images were used, which corresponded to the period between the portal venous phase and late arterial phase. Delays of 45 and 55 seconds were applied and the RAV visualization rate increased to 98%. In the present study, the optimal time window for 4D CT was between 36 and 54 seconds in RAV and LAV. Besides, bilateral adrenal veins were detected in all patients on images and the extraction of flow dynamic features enabled us to easily evaluate enhancement patterns over time, which is more feasible than the above protocols.
A higher CT value was achieved for RAV than previously reported in each phase of less than 200 HU (17). Our results also showed that right and left tmax were 44.43 ± 6.86 and 45.39 ± 7.53 seconds, respectively. If the injection method of contrast medium is the same, even in facilities without 4D CT, RAV and LAV may be clearly visualized if the protocol is designed to take images including 44 and 45 seconds after the start of the injection. Our results could enhance the knowledge of venous physiology and pathophysiology. In addition to the complex anatomical lesions, rapid flow empties from adrenal veins to the superior veins may cause difficulty in AVS, for which understanding the venous flow dynamics is fundamentally important.
Another study demonstrated that TDC was effective in facilitating the evaluation of contrast wash-in and wash-out phases, allowing for superior visualization of blood vessels [11]. Respiration-induced motion artifacts may affect the evaluation of small regions, particularly when the effect is present in the single arterial phase. Therefore, covert variations in RAV and LAV may have been overlooked because enhancement differences were previously reported to be significant in the affected series (18). Since multiple phases were available in the present study, movement artifacts were less problematic.
A significant difference was observed in Dmax between RAV and LAV, with RAV having a lower value. This was mainly due to the short length of RAV and its diameter of < 2–3 mm, which generally does not allow for satisfactory anatomical mapping. Furthermore, the drainage of the central vein on the right side may have been duplicated and drained into the right renal vein or inferior transverse vein, which presumably led to a change in the concentration of the contrast agent [8].
In the present study, the tmax of RAV and LAV reached enhancement earlier in RAV. Since the confluence of the adrenal central vein and inferior vena cava is common, the short tmax of RAV may be related to fast outflow from the renal vein to the adrenal superior vein [20]. LAV is mainly a tributary from the left inferior transverse vein and renal capsule vein, mostly outflowing to the left renal vein, and also from the inferior transverse vein. Some cases of separation from the inferior transverse vein have been reported [21]. A previous study also identified an abnormal confluence of LAV with the periaortic left renal vein and the direct integration of LAV with the inferior vena cava [22]. Therefore, outcomes may differ if the LAV measurement position is above the confluence of the renal veins or upstream of the confluence of the inferior transverse veins. In addition, the flow of LAV may be affected by recirculation, resulting in a longer tmax.
The multivariate analysis showed that BMI, plasma renin activity, and potassium significantly affected the 4D CT image intensity of blood flow in RAV. BMI was reported to inversely correlate with peak aortic enhancement, which is consistent with present result showing the relation between BMI and lower enhancement in the RAV. This finding suggests that BMI alterations specific to the peak enhancement may allow for the personalization of RAV delineation (19). In addition, previous studies suggested that contrast medium concentration varies with plasma volume and extracellular fluid volume in parenchymal organs, and that lean body mass and fluid retention, which are associated with BMI, can affect venous enhancement [25]. Although primary aldosteronism patients are characterized by an elevated extracellular volume, the expansion and distribution of extracellular fluids beside RAV may differ from that around LAV because of the closely adjacent liver parenchyma, and the TDC of the liver was previously shown to be similar to that of RAV (19, 20). On the other hand, the contrast materials administered in the blood compartment may be diluted less in RAV than in LAV due to its small size, resulting in a lower concentration in blood flowing in RAV (21). Conversely, in small vessels and extracellular areas, the clearance of contrast agent may be slower and the concentration may be higher. These parameters need to be considered when performing a dynamic analysis of RAV to avoid over- or underestimations of enhancement and contrast concentrations in veins.
The aldosterone/ plasma renin activity ratio is widely used in screening tests for primary aldosteronism because elevated aldosterone with suppressed plasma renin activity is a characteristic finding of primary aldosteronism (22). However, few studies have reported a correlation between blood flow and plasma renin activity, and the usefulness of plasma renin activity as a reference marker to evaluate RAV has not yet been examined (23). Experimental observations showed that renovascular hypertension was mainly maintained by the renin–angiotensin–aldosterone system (RAAS). Volume retention increased blood pressure, and fluid expansion returned negative feedback to the RAAS, leading to a higher plasma renin activity level than expected with volume overload. In addition, high plasma renin activity was associated with a vasoconstrictive mechanism (24). Specifically, calcium mobilization may be activated under high plasma renin activity conditions and this vasoconstrictor effect may result in renal artery stenosis (25). In the present study, data from RAV suggested a peak enhancement change with plasma renin activity, which may have resulted from venous blood volume and vascular lumen sizes influencing blood flow (24, 25).
The classical paradigm indicates that retained excess sodium is excreted, thereby returning the extracellular fluid volume excess to a normal level and deactivating negative feedback on the RAAS [29]. While this study showed that the contrast medium concentration in the adrenal vein was not correlated with the sodium concentration, and the potassium concentration significantly affected the 4D CT image intensity of blood flow in the RAV. Some findings support a role for potassium ions in RAAS, and the signal produced by changes in electrolyte concentrations in the macula densa may be caused by sodium. Baudrand et al. [26] reported that urinary potassium excretion was higher than sodium excretion and did not stimulate renin; however, our results showed that the stimulatory effects of potassium on blood flow were stronger than those of sodium. The reason for this discrepancy remains unclear and may be due to the lack of confounding factors controlling RAAS in the present study, such as antihypertensive medication and the dietary intake of potassium and sodium; therefore, further studies are warranted.
A linear relationship between sex and the time-to-peak parameter was observed in RAV. Although previous studies demonstrated that sex strongly correlated with sustained hypertension in primary aldosteronism patients [31], the relationship between adrenal vein hemodynamics and sex remains unknown. The optimal time window for delineating bilateral adrenal veins in our protocol was the same; however, a precise time still needs to be considered for taking advantage of optimal venous enhancement.
There were some limitations that need to be addressed. The present study was retrospective, performed at a single center, and included a small sample with potential selective bias. Thus, more in-depth statistical analysis and multicenter study are necessary to eliminate bias and to establish the undisputed value of hemodynamics extracted from 4D CT for delineating adrenal veins. Furthermore, the present results were based only on imaging features without sub-analyses after dividing by the anatomical patterns, and all these interpretations may not be consistent with the anatomy of patients. Moreover, we were unable to confirm the performance of images because they were obtained with contrast agents and were influenced by the protocol and baseline characteristics of patients. Another limitation is that ROI was placed manually with possible measurement errors, which may be avoided with automated selection in the future. In addition, the non-uniform distribution of contrast agent, pulsation, continuous contrast agent diffusion, and blood turbulence may have had a considerable impact. These limitations cannot be ignored when interpreting the present results and, thus, further studies are required.
In conclusion, a lower contrast effect and earlier peak enhancement was achieved for blood flow in RAV than in LAV. BMI, plasma renin activity, and potassium positively correlated with the blood flow density in RAV, and sex was independently associated with the time at which maximum CT attenuation values of RAV were obtained.