We examined venograms during BRTO for morphometric assessment of LIPV in patients with portal hypertension. All venograms of LIPV showed a narrowing just above the common trunk. The diameter of LIPV and narrowed LIPV were significantly larger in males than in females, whereas no significant differences were observed in the narrowing rate or the distance to narrowed LIPV from the left renal vein.
LIPV is rarely discussed in classical anatomy textbooks16. However, in recent years, with the development of endoscopic and endovascular treatments for esophageal and gastric varices, LIPV, which serves as an important portosystemic shunt in patients with portal hypertension, has been attracting increasing attention12,17,18,19. Loukas et al. classified and described variations in the origins and distributions of LIPV in cadavers19. Among 300 cadaveric specimens, 37% of LIPV drained into the IVC below the diaphragm, 25% into the left suprarenal vein, 15% into the left renal vein, 14% into the left hepatic vein, and 1% into both the IVC and left suprarenal vein. On the other hand, Araki et al. reviewed venograms obtained during adrenal vein sampling and reported that LIPV merged with the left adrenal vein in 92.8% of cases, and the mean distance from the confluence to the renal vein was 16.4 mm, with a standard deviation of 4.7 mm12. They also noted that 87.2% of patients without portal hypertension had narrowed LIPV, which was presumably formed by the valves. Saad et al. reviewed the BRTO technique and reported a web-like narrowing at the junction of LIPV with the common trunk, which made catheter insertion difficult11.
The present study is the first to describe morphometric features of LIPV and its narrowing. Since LIPV serves as a major portosystemic shunt, it was assumed that the diameter of LIPV correlates with the severity of portal hypertension. However, only total bilirubin level and Child–Pugh score correlated with the diameter of LIPV in the present study. Other factors should be considered to determine the association between the diameter of the LIPV and portal hypertension, including portal blood flow, portal pressure, and the degree of development of collateral pathways.
The results revealed the presence of a narrowing just above the common trunk of LIPV. There were no significant differences in the diameter of LIPV or narrowing rate between on venograms and on axial CT images. On the other hand, the diameter of narrowed LIPV was significantly larger on CT images (6.1 ± 2.3 mm) than on venograms (5.1 ± 2.3 mm) (p = 0.03). However, considering that the 95% confidence interval was − 1.65 to 0.098 mm, the difference of the diameter of narrowed LIPV between on venograms and on CT images was not considered clinically significant.
The mean distance to narrowed LIPV from the left renal vein was 20.0 mm with a standard deviation of 7.4 mm. This was longer than that previously reported by Araki et al.12, and may be attributed to differences in patient characteristics. The study by Araki et al. examined patients without portal hypertension, whereas we investigated and obtained measurements from patients with portal hypertension. LIPV may be dilated and tortuous in patients with portal hypertension, and, thus, the distance to narrowed LIPV from the left renal vein may be longer.
Based on the results obtained, In BRTO, careful catheter manipulation is required to pass through a narrowing of LIPV that is located at approximately 2 cm from the left renal vein. Once a catheter passes, the narrowing makes a very effective choke point11. In BRTO, it is essential for successful treatment that GRS is completely balloon occluded before injecting sclerosant into gastric varices. A balloon placed just above the narrowing can be stable and adequately occlude GRS10,11.
The present study has some limitations. First, the sample size was small. Second, venograms of LIPV during BRTO were obtained following a manual contrast injection in order to prevent rupture of the vein. Differences in the volume of contrast material and the injection speed may affect the depiction of LIPV and its narrowing. Moreover, we speculated that the narrowing of LIPV was formed by venous valves (downward-flow valves), which was consistent with the reason for the difficulty of catheter passage over the narrowing. However, the presence of venous valves has not yet to be confirmed histologically; therefore, further investigations on cadavers are needed.
