- Email: [email protected]
Ultrasonographic Visualization of Accessory Hepatic Veins and Their Lesions in Budd–Chiari Syndrome
Ultrasound in Med. & Biol., Vol. 41, No. 8, pp. 2091–2098, 2015 Copyright Ó 2015 World Federation for Ultrasound in Medicine & Biology Printed in the USA. All rights reserved 0301-5629/$ - see front matter
http://dx.doi.org/10.1016/j.ultrasmedbio.2015.03.023
d
Original Contribution ULTRASONOGRAPHIC VISUALIZATION OF ACCESSORY HEPATIC VEINS AND THEIR LESIONS IN BUDD–CHIARI SYNDROME SHI-FENG CAI,* YONG-HAO GAI,y SHUANG MA,z BO LIANG,y GUANG-CHUAN WANG,x and QING-WEI LIU* * Department of Radiology, Provincial Hospital Affiliated to Shandong University, Jinan City, Shandong Province, China; Department of Ultrasound, Provincial Hospital Affiliated to Shandong University, Jinan City, Shandong Province, China; z Department of Ultrasound, Fifth Hospital of Jinan, Jinan City, Shandong Province, China; and x Department of Gastroenterology, Provincial Hospital Affiliated to Shandong University, Jinan City, Shandong Province, China
y
(Received 19 September 2014; revised 13 February 2015; in final form 19 March 2015)
Abstract—The aim of this study was to investigate the ultrasonographic features of accessory hepatic veins (AHVs) and their lesions in Budd–Chiari syndrome (BCS). Three hundred patients with BCS were examined by ultrasonography with multifrequency (3–6 MHz) convex transducers. Sonography was performed 1 to 2 wk before digital subtraction angiography and computed tomography angiography or magnetic resonance imaging. Using sonograms, we evaluated the number, course, diameter, orifice, lesions and hemodynamics of patent and obstructed AHVs. Ultrasonography was superior to digital subtraction angiography, computed tomography angiography and magnetic resonance imaging in revealing AHV lesions and hemodynamics. Dilated AHVs were detected in 227 patients. There were 239 caudate lobe veins in 167 patients and 168 inferior right hepatic veins in 151 patients. Both caudate lobe veins and inferior right hepatic veins were found in 91 of the 227 patients. The inlets to AHVs were located mainly on the right lateral or right anterior wall of the inferior vena cava, and the remnant, on the left lateral wall. AHV lesions comprised mainly septal obstruction and segmental stenosis. The hemodynamics of AHVs varied with the condition of inferior vena cava and AHVs. Ultrasonic examination can reveal AHVs and their lesions in patients with BCS and is helpful in choosing and planning therapeutic approaches. (E-mail: [email protected]) Ó 2015 World Federation for Ultrasound in Medicine & Biology. Key Words: Ultrasonography, Budd–Chiari syndrome, Thrombosis, Caudate lobe vein, Inferior right hepatic vein, Inferior vena cava.
other hepatic veins. Most patients with BCS have a chronic condition and have hepatic vein obstruction (Gai et al. 2014b; Zhou et al. 2014). The continuous increased pressure resulting from obstructed hepatic veins may eventually transform the anastomoses into enlarged collaterals. Blood flow from the obstructed hepatic veins can be drained by dilated AHVs through the collateral pathways. Kanamura et al. (2001) described the anatomy of caudate lobe veins, including their opening and location. Ueda et al. (1998) and Bargallo et al. (2003, 2006) visualized dilated AHVs by computed tomography and sonography in patients with BCS, respectively. Of note, Bargallo et al. (2003) suggested that visualization of the caudate vein ($3 mm in diameter) with gray-scale sonography strongly supported the diagnosis of BCS. However, descriptions of AHV lesions are rare; only Bargallo et al. (2003) mentioned that the caudate vein was obstructed in some cases, but did not describe it further. Importantly, ultrasonic examinations
INTRODUCTION Budd–Chiari syndrome (BCS) is a group of diverse conditions associated with obstruction of hepatic venous outflow in the large hepatic vein or the extrahepatic segment of the inferior vena cava (IVC). Anatomically, hepatic veins are divided into the superior and inferior groups. The superior group includes the left, middle and right hepatic veins, which are the main draining veins of the liver; the inferior group includes many hepatic veins with small diameters that are also called accessory hepatic veins (AHVs) (Williams and Bannister 1995). Normally, small anastomoses are found between adjacent AHVs and hepatic veins or between hepatic veins and Address correspondence to: Qing-wei Liu, Department of Radiology, Provincial Hospital Affiliated to Shandong University, No. 324, Jingwu Road, Jinan City 250021, Shandong Province, China. E-mail: [email protected] Conflicts of Interest: Authors declare no financial or non-financial competing interests. 2091
2092
Ultrasound in Medicine and Biology
provide an accurate way to detect AHVs and to classify obstructed AHVs in patients with BCS, which can be helpful in choosing and planning therapeutic approaches. In this study, we used ultrasonography to investigate the distribution, lesions and hemodynamic characteristics of patent and obstructed AHVs. METHODS Patients This retrospective study was based on data obtained with sonography between April 2005 and June 2014 from 300 patients with chronic BCS, including 158 male and 142 female patients (age range, 18–77 y; mean age, 45.48 6 10.82 y). In all patients, the time from the first clinical symptom to diagnosis ranged from 10 d to 24 y. However, some patients had been suffering with the disease for years before seeing a physician, and the duration of the disease before sonographic examination could not be accurately determined. The etiology was mainly idiopathic, and only four patients had previously undergone abdominal surgery. None of the BCS cases was caused by long-term oral contraceptive medication or pregnancy. Although hepatic vein thrombosis was detected in 14 patients, bone marrow aspiration and blood tests revealed that it was not related to globulism, antiphospholipid syndrome, intermittent hemoglobinuria or any myeloproliferative disease. The main clinical symptoms were right upper abdominal distention in 106 cases, abdominal wall varicosis in 59 cases and edema of both lower extremities in 118 cases. All procedures were approved by the ethics committee of Shandong University. Informed consent was obtained from all patients or their families. Ultrasonography Ultrasonography was performed using a Logiq E9 (GE Healthcare, Waukesha, WI, USA), an HDI 3500 (ATL-Philips, Bothell, WA, USA) or an Envisor HD system (Philips Healthcare, Best, The Netherlands) with multifrequency (3–6 MHz) convex transducers. All patients fasted more than 8 h before the examination and were scanned in the supine or left recumbent position. First, the liver, spleen and IVC were observed with B-mode sonography in search of hepatomegaly, splenomegaly, portal vein dilation and ascites. In the presence of hepatic vein and IVC obstruction, the number and site of outlets of draining AHVs were carefully observed, and the number and distribution of intrahepatic communicating branches were also recorded. The site we chose for measuring AHV diameter was large and was possibly the diameter at the end of inspiration. Pulsed wave Doppler sonography was used to assess blood flow direction and velocity within the draining AHVs. At sites
Volume 41, Number 8, 2015
where B-mode sonography revealed narrowing along the draining hepatic vein, color and spectral Doppler revealed aliasing and high flow velocity. We based our determination of AHVs on the fact that caudate lobe veins are the largest veins draining from the caudate (Spiegel) lobe and flow directly into the retrohepatic segment of the IVC (Bargallo et al. 2003, 2006; Kanamura et al. 2001). According to the nomenclature of Couinaud, the course of the inferior right hepatic veins lies mainly in hepatic segment VI. In transverse section, these veins appear on the posterior medial side of the posterior branch of the right portal vein, within the hepatic parenchyma, in front of the renal imprint, and enter the IVC at the level of the first portal hilum (Xing et al. 2007). Sonography was performed 1 to 2 wk before digital subtraction angiography (DSA) and computed tomography angiography (CTA) or magnetic resonance imaging (MRI). DSA was performed in all patients, whereas CTA and MRI were performed in 83 and 29 cases, respectively. When sonograms of AHVs and their lesions agreed with DSA and/or CTA (and/or MRI) images, the results were considered valid. Informed consent was obtained from each subject before DSA, CTA or MRI, and the protocol was consistent with the Declaration of Helsinki. Statistical analysis Statistical analysis was performed using SPSS Version 17.0 (SPSS, Chicago, IL, USA). Student’s t-test was used to determine the significance of statistical differences between two groups. Data are presented as the mean 6 standard deviation. Values with p , 0.05 were considered statistically significant. RESULTS Most patients with dilated AHVs have mixed BCS, whereas a small number have isolated hepatic vein-type BCS Dilated AHVs ($3 mm in diameter) were detected in 227 patients, including 32 patients with obstructed AHVs (patients with obstructed AHVs were not included in calculation of the average diameter), who accounted for 75.7% of the total number of patients (300) with BCS (Table 1). Color Doppler revealed dilation in most AHVs; the distal part of the hepatic veins, which were proximately occluded, drained through communicating branches into AHVs and subsequently into the IVC (Fig. 1). In addition to the dilated AHVs, there were a small number of tiny AHVs (,3 mm in diameter) that did not flow into collateral vessels. Long courses with acute angles between the long axis of the inferior right hepatic veins and IVC and relatively short courses of caudate lobe
Accessory hepatic vein lesions in Budd–Chiari syndrome d S.-f. CAI et al.
Table 1. Information on AHVs*
AHV
Number of AHVs per patient
Diameter (mm)
CLV IRHV
1.89 6 0.62 (1–4)y 1.13 6 0.39 (1–5)
7.64 6 2.74 (2–15) 9.38 6 3.05 (3–19)
Total number of AHVs/number of patients 239/167 168/151
AHV 5 accessory hepatic vein; CLV 5 caudate lobe vein; IRHV 5 inferior right hepatic vein. * Both CLVs and IRHVs were found in 91 of the 227 patients. y Mean 6 standard deviation (range).
veins nearly perpendicular to the IVC were observed in most of the patients, possibly resulting in higher blood reflux resistance of the inferior right hepatic veins in contrast to the caudate lobe veins. The inlets to all inferior right hepatic veins were located on the right lateral or right anterior wall of the IVC. Caudate lobe veins were mainly located on the anterior wall; only a small number of caudate lobe veins were located on the left lateral wall. Among the 227 patients, 197 patients had mixed BCS, in which obstructed hepatic veins and IVCs co-existed in the same patient. Only 30 cases had isolated hepatic veintype BCS, in which only the hepatic veins were obstructed. No isolated IVC type (obstruction in only the IVC) was observed in this study. These data indicated that 227 of the 300 patients had dilated AHVs, among which 197 patients had mixed BCS and 30 patients had isolated hepatic vein-type BCS. There are more lesions in inferior right hepatic veins than in caudate lobe veins There were 23 obstructed inferior right hepatic veins in 23 patients. Lesions of inferior right hepatic veins were classified mainly into septal obstruction (septum thickness #1 cm, including septal stenosis and
2093
septal occlusion) and segmental stenosis (stenosed length .1 cm). Segmental stenosis affected seven inferior right hepatic veins; the ultrasonograms revealed that the proximal lumens of the involved vessels were stenosed with a thickening wall. The stenosed length of the lumens varied from 10 to 35 mm, whereas the diameter ranged from 2 to 5 mm (Fig. 2a). Septal stenosis was detected in five inferior right hepatic veins, and septal occlusion was detected in one inferior right hepatic vein. Ultrasonography revealed that the septum was located on the inlet to the inferior right hepatic vein with or without extension to the lumen of the IVC (Fig. 2b). Thrombosis affected 10 inferior right hepatic veins, with ultrasonography revealing the hypo-echogenicity characteristic of fresh thrombus and hyper-echogenicity characteristic of calcified thrombus (Fig. 2c, d). Lesions of the caudate lobe veins were relatively rare, with only eight cases of septal stenosis and one case of septal occlusion (Fig. 3a, b). However, the inlet to the caudate lobe vein might be stenosed because of involvement of thrombus from the IVC (Fig. 3c). These data suggested that inferior right hepatic veins had more lesions than caudate lobe veins.
Hemodynamic characteristics vary depending on status of AHVs Sonograms revealed the absence of blood flow in two inferior right hepatic veins that were completely thrombosed by fresh thrombus and appeared homogeneously hypo-echoic, but a blood-filling defect was observed in the remaining veins that were partly embolized or recanalized. Reversed blood flow in the AHVs was observed in 7 patients with a cavo-hepato-atrial pathway in which the IVC was occluded between the patent AHV and hepatic vein; venous flow from the IVC
Fig. 1. (a) Sonogram in the sagittal epigastric line of a 54-y-old woman with Budd–Chiari syndrome, revealing two dilated patent caudate veins draining into the IVC. (b) Sonogram in the right intercostal antero-inferior/posterosuperior plane of a 47-y-old man with Budd–Chiari syndrome, revealing dilated patent inferior right hepatic vein with almost flat spectrum caused by the completely occluded IVC. CL 5 caudate lobe; CLV 5 caudate lobe vein; IRHV 5 inferior right hepatic vein; IVC 5 inferior vena cava. Cycle 5 3, mean velocity 5 6.92 cm/s, accelerated time 5 0.010 s.
2094
Ultrasound in Medicine and Biology
Volume 41, Number 8, 2015
Fig. 2. (a) Sonogram in the subcostal plane of a 56-y-old man with Budd–Chiari syndrome, revealing segmental stenosis of inferior right hepatic vein with aliasing blood flow (arrow). (b) Sonogram in the subcostal plane of a 72-y-old man with Budd–Chiari syndrome, revealing a septum on the inlet of the inferior right hepatic vein extending to the lumen (arrow) of the inferior vena cava with narrow aliasing blood flow. (c) Sonogram in the right intercostal antero-inferior/posterosuperior plane of a 37-y-old woman with Budd–Chiari syndrome, revealing the recanalized thrombus and blood-filling defect (arrow) of the inferior right hepatic vein. (d) Sonogram in the right intercostal antero-inferior/postero-superior plane of a 57-y-old man with Budd–Chiari syndrome, revealing the calcified thrombus and blood-filling defect (arrow) of the inferior right hepatic vein. IRHV 5 inferior right hepatic vein, IVC 5 inferior vena cava.
reversed to the AHV and intrahepatic communicating branches, onward to a hepatic vein and finally to the right atrium, as reported previously (Gai et al. 2014a). Blood from the septal occluded inferior right hepatic vein and the caudate lobe vein of the patients did not enter the IVC, but drained into the right hepatic subcapsular vein (Fig. 3b). Color Doppler studies of the remaining patients revealed that blood from obstructed hepatic veins drained through communicating branches into the AHVs and, subsequently, into the IVC (Fig. 1a, b). The spectrum of patent AHVs was similar to that of patent hepatic veins in patients with patent IVCs in this study. Using the condition of the lumen to determine whether IVC was obstructed, we classified patients with patent AHVs (excluding the 7 patients with cavohepato-atrial pathways) into the obstructed and nonobstructed groups. The peak velocity of AHVs in each group was calculated accordingly (Table 2). The results indicated that the peak velocity of AHVs in the group with obstructed IVCs was lower than that in the group with non-obstructed IVCs. In addition, more severe obstruction of IVCs led to more marked changes in spectral patterns of AHVs. When IVCs were completely occluded, the spectral patterns almost flattened (Figs. 1b and 3d).
At sites where proximal parts of the AHVs were obviously stenosed, color and spectral Doppler revealed aliasing and high-velocity regardless of whether the IVC was obstructed. The velocity of the inferior right hepatic veins ranged from 90 to 225 cm/s, and that of the caudate lobe veins, from 65 to 112 cm/s (Figs. 2a, b and 4). These data suggested that hemodynamic characteristics varied depending on the status of AHVs. Ultrasonography is superior to DSA, CTA and MRI in revealing AHV lesions and hemodynamics As mentioned earlier, 32 of the 227 patients had obstructed AHVs; the remaining 195 patients had patent AHVs and clear blood flow as observed by ultrasonography, which was consistent with the DSA results. Compared with other imaging techniques, ultrasonography was superior in accurately detecting lesions in AHVs and specifying lesion type (Table 3). Lesions that were not identified by DSA included two completely thrombosed inferior right hepatic veins, a septal-occluded inferior hepatic vein and a septal-occluded caudate lobe vein. Seven patients with AHV lesions identified by CTA included two cases of segmental stenosis of inferior right hepatic veins, three cases of thrombus in inferior right hepatic veins (two cases with calcified thrombus),
Accessory hepatic vein lesions in Budd–Chiari syndrome d S.-f. CAI et al.
2095
Fig. 3. (a) Transverse sonogram through the epigastric region of a 29-y-old woman with Budd–Chiari syndrome (BCS), revealing septal stenosis at the inlet of the caudate lobe vein. (b) Transverse sonogram through the right epigastric region of a 46-y-old man with BCS, revealing septal occlusion at the inlet of the caudate lobe vein (arrow) and blood draining into the right hepatic subcapsular vein (red). (c) Transverse sonogram through the epigastric region of a 40-y-old woman with BCS, revealing the stenosis inlet of the caudate lobe vein caused by involvement of the thrombus from the inferior vena cava. (d) Transverse sonogram through the epigastric region of a 39-y-old woman with BCS, revealing patent caudate lobe vein with flat septum caused by obstruction of the inferior vena cava. CLV 5 caudate lobe vein. IVC 5 inferior vena cava. Cycle 5 3, mean velocity 5 9.39 cm/s, accelerated time 5 0.029 s.
one case of septal stenosis of inferior right hepatic veins and one case of septal stenosis of caudate lobe veins; in the two patients with septal stenosis, CTA revealed no septum, but locally clear blood flow. However, veins with calcified thrombus and segmental stenosis were clearly visualized. Three patients with AHV lesions examined by MRI included two cases with a thrombosed inferior right hepatic veins (one fresh thrombus and one calcified thrombus) and one case with septal occlusion of caudate lobe vein. Fresh thrombus was correctly diagnosed by iso-intensity on T1-weighted imaging and hyper-intensity on T2-weighted imaging, but patients with calcified thrombus were not successfully diagnosed. For patients with septal occlusions, MRI revealed arc-like low signals from the septum (Fig. 5a, b). The results of ultrasonography, DSA, CTA and MRI are summarized in Table 3. These data suggested that ultrasonography was superior to DSA, CTA and MRI in demonstrating AHV lesions and hemodynamics. Treatment strategies based on AHV status Among the 227 patients, 195 had patent AHVs. Only one patient with a patent AHV diameter ,0.5 cm underwent surgery to relieve the hepatic vein obstruction;
the others patients with patent AHVs were not treated. Among the 32 patients with AHV lesions, all patients with septal stenosis or occlusion of the inferior right hepatic veins and caudate lobe veins underwent percutaneous transluminal angioplasty, two patients with segmental stenosis of the inferior right hepatic veins underwent percutaneous transluminal angioplasty and five patients with segmental stenosis of the inferior right hepatic veins underwent stent placement. All lumens were patent after surgery (Fig. 6). The other seven cases
Table 2. Comparison of peak velocity of AHVs between group with obstructed IVCs and group with unobstructed IVCs* Peak velocity (cm/s) IVC condition
IRHV
CLV
Obstructed group (n 5 154) Unobstructed group (n 5 30) p Valuez
12.54 6 4.26y 28.36 6 9.18 0.000
13.39 6 4.95 30.25 6 9.58 0.000
AHV 5 accessory hepatic vein; CLV 5 caudate lobe vein; IRHV 5 inferior right hepatic vein; IVC 5 inferior vena cava. * The velocity of AHVs was measured at the end of inspiration. y Mean 6 standard deviation. z p , 0.05 was considered statistically significant.
2096
Ultrasound in Medicine and Biology
Fig. 4. Sonogram in the subcostal plane of a 47-y-old woman with Budd–Chiari syndrome, revealing the aliasing and highflow velocity of 112 cm/s at the stenosis inlet of the inferior right hepatic vein. IRHV 5 inferior right hepatic vein. Peak velocity 5 112 cm/s, pressure gradient 5 5.00 mm Hg.
with thrombosed inferior right hepatic veins underwent surgery to open hepatic veins because of poor drainage. Four patients who had membranous or short segmental occlusion of the IVC with cavo-hepato-atrial pathways underwent angioplasty for re-canalization of the IVC, after which blood flow returned to normal. Three patients with a small amount of thrombus were left untreated and followed up 1–5 y without surgery; their ultrasonograms and clinical manifestations did not change significantly. Among the 227 patients, 177 had various degrees of obstruction in the IVC and underwent intervention therapy or surgery. These data suggested that ultrasonography was helpful in choosing and planning therapeutic approaches. DISCUSSION In this study, ultrasonography clearly revealed the number, course, diameter, presence of lesions and variations in hemodynamics of AHVs. First, we not Table 3. The diagnosis of ultrasonography, DSA, CTA and MRI Number of patients AHV status
Diagnosis
US
DSA
CTA
MRI
Patent*
Detected Undetected Detected Undetected Correct Incorrect
195 0 32 0 73 0 300
195 0 28 4 73 0 300
60 0 7 2 14 0 83
19 0 2 1 7 0 29
Afflicted No AHVs Total
AHV 5 accessory hepatic vein; CTA 5 computed tomography angiography; DSA 5 digital subtraction angiography; MRI 5 magnetic resonance imaging; US 5 ultrasonography. * The 7 patients with cavo-hepato-atrial pathways were included.
Volume 41, Number 8, 2015
only distinguished AHVs according to their anatomic locations, but also confirmed the locations of the inlets to AHVs, which could guide interventional therapy for both AHVs and the IVC, particularly to avoid secondary stenosis of the inlets to AHVs caused by stent oppression. Second, we found AHV lesions and evaluated their hemodynamics. When hepatic veins were obstructed and the blood from these veins was drained by AHVs, dilation of AHVs and collaterals was visualized by ultrasonography. If the IVC was obstructed at the same time, the dilation was even more obvious. Meanwhile, obstruction of IVCs led to increased hypertension and decreased blood flow velocity of AHVs, which might accelerate hepatic congestion. The results of this study indicated that the flow velocity of AHVs in the obstructed group was significantly lower than that in the non-obstructed group of IVCs. Therefore, it was suggested that the decreased flow velocity of AHVs might be used as a criterion to predict increased congestion of the liver. On the other hand, our study revealed that the spectrum of patent AHVs in normal patients was similar to that of patent hepatic veins in patients with patent IVCs, namely, phasic flow in response to both cardiac and respiratory cycles. However, in patients with partial IVC obstructions, continuous flow (also called a pseudo-portal Doppler signal) might appear with more severe obstruction of the IVC corresponding to more marked changes in spectral patterns of AHVs. When the IVC was completely occluded, the Doppler signal became almost flat. However, the Doppler signal of AHVs returned to its phasic pattern after the IVC obstruction was recanalized. Consequently, we speculate that the variation in velocity and spectrum patterns of AHVs can also be used as a criterion in evaluating the effect of recanalization of IVCs. The cavo-hepato-atrial pathway is an unusual blooddraining pathway in BCS, with the pathologic consequence that blood pressure below the obstructed portion of the IVC exceeds that of hepatic veins. Therefore, blood from the IVC reverses to the AHVs, through the enlarged intrahepatic collaterals and draining hepatic veins (orifice above occlusion), and finally to the right atrium. This blood-draining pathway relieves the symptoms caused by portal hypertension and IVC hypertension (Akaki et al. 1995; Kamba et al. 1995; Takayasu et al. 1985). Hepatopetal flow in AHVs can be clearly observed on ultrasonography, on which basis a definite diagnosis can be made (Gai et al. 2014a, 2014b). The results of our study indicated that the AHVs most involved by thrombus and segmental stenosis were the inferior right hepatic veins. The incidence of obstruction, especially obstruction caused by thrombus, was higher for the inferior right hepatic veins than for the caudate lobe veins. The greater susceptibility of inferior right hepatic veins to thrombus might be caused by
Accessory hepatic vein lesions in Budd–Chiari syndrome d S.-f. CAI et al.
2097
Fig. 5. For the patient whose sonogram is provided in Figure 3b: (a) Magnetic resonance image revealing the hypointense septum at the inlet of the caudate lobe vein. (b) Digital subtraction angiography failed to reveal the septum.
their higher resistance to regurgitation, in contrast to the caudate lobe veins, as mentioned earlier. Ultrasonography can be used to visualize thrombi with or without calcification, as well as blood-filling defects caused by thrombi. Sonograms of segmental stenosis revealed narrow lumens with a thickening wall at the proximal part of AHVs, but the lumen is sometimes too narrow to be visualized on gray-scale sonography. The diagnosis relies only on linear aliasing blood on color Doppler with accelerated velocity. Caudate lobe vein lesions comprised mainly stenosis of the inlets caused by septum and IVC-involved thrombus. The septum that caused relative stenosis of the inlet might have arisen from the wall of the IVC, which limited the inlet dilation compared with the obviously expanded distal lumen of the caudate lobe veins. Accelerated blood flow from stenosis inlets was also detected, but was usually lower than that from stenosis inlets to the inferior right hepatic veins. In cases of stenosis of the caudate lobe veins caused by IVC-involved thrombus, it was clearly observed that the thrombus arose
Fig. 6. Sonogram in the right intercostal antero-inferior/ postero-superior plane of a 42-y-old man, revealing the patent stent in the inferior right hepatic vein (IRHV) that was affected by segmental stenosis before surgery. IVC 5 inferior vena cava.
from the IVC and extended to the lumen of the caudate lobe veins. Because of its complex etiopathogenesis and multiple types, there are many treatment strategies for BCS. Therapeutic strategies are based on imaging techniques, which can reveal blood draining pathways of AHVs and collateral circulations of obstructed hepatic veins before surgery. The present study found that ultrasonography, DSA, CTA and MRI had similar sensitivity in diagnosing patients without AHVs. However, for revealing lesions, ultrasonography was superior to the other imaging techniques, especially in depicting the septum of AHVs, even when septa were #3 mm thick. Completely thrombosed AHVs cannot be identified on DSA, but can be clearly visualized by ultrasonography. The sensitivity of CTA and MRI in detecting lesions in AHVs requires further study because of the small number of obstructions examined in this study. It is generally accepted that if the hepatic veins are incompletely obstructed (i.e., at least one of the three main hepatic veins remains patent), only the obstructed IVC requires surgery. If the three main hepatic veins are completely obstructed, treatment must be directed to the hepatic veins. Complex treatment of obstructed hepatic veins can be avoided when large draining AHVs and rich collaterals are observed (Shan et al. 2005). Even when the three main hepatic veins at the second hilum and the AHVs (draining vessels) at the third portal hilum are stenosed or obstructed, treatment can be limited to the inadequate draining vessels at the third portal hilum if rich intrahepatic communicating branches connecting the two groups of vessels can be visualized. Treatment centered on AHVs at the third portal hilum is usually easier and more tolerable than that on the hepatic veins at the second portal hilum, because the outlets of the AHVs to the right atrium are longer than those of the main hepatic veins. Therefore, detailed assessment of every AHV at the third portal hilum is mandatory before surgery. In our study, both the lesions and hemodynamics of the AHVs were clearly visualized on ultrasonography,
2098
Ultrasound in Medicine and Biology
and the operations were performed successfully after examination. In conclusion, ultrasonographic examination not only provides a convenient and accurate way to visualize AHVs and their lesions in patients with BCS, but also helps in choosing and planning therapeutic approaches. Acknowledgments—This work was supported by the Shandong Provincial Science and Technology Development Project Foundation of China (Grants 2013 GSF11827 and 2012 GSF11820).
REFERENCES Akaki S, Kanazawa S, Gochi A, Nakamura K, Yasui K, Togami I, Hiraki Y, Hamazaki K. Asymptomatic membranous obstruction of the inferior vena cava due to large intrahepatic collaterals. Cardiovasc Intervent Radiol 1995;18:403–405. Bargall o X, Gilabert R, Nicolau C, Garcıa-Pagan JC, Ayuso JR, Bru C. Sonography of Budd–Chiari syndrome. AJR Am J Roentgenol 2006; 187:W33–W41. Bargall o X, Gilabert R, Nicolau C, Garcıa-Pagan JC, Bosch J, Bru C. Sonography of the caudate vein: Value in diagnosing Budd–Chiari syndrome. AJR Am J Roentgenol 2003;181:1641–1645. Gai Y, Cai S, Fan H, Liu Q. Diagnosis of the cavo-hepato-atrial pathway in Budd–Chiari syndrome by ultrasonography. Exp Ther Med 2014a;8:793–796.
Volume 41, Number 8, 2015 Gai Y, Cai S, Guo W, Zhang C, Liang B, Jia T, Zhang G. Sonographic classification of draining pathways of obstructed hepatic veins in Budd–Chiari syndrome. J Clin Ultrasound 2014b;42:134–142. Kamba M, Ochi S, Ochi H, Maruyama S, Sato H, Suto Y. Asymptomatic membranous obstruction of the inferior vena cava forming intrahepatic collateral pathways. J Gastroenterol 1995;30:783–785. Kanamura T, Murakami G, Hirai I, Hata F, Sato TJ, Kumon M, Nakajima Y. High dorsal drainage routes of Spiegel’s lobe. J Hepatobiliary Pancreat Surg 2001;8:549–556. Shan H, Zhu K, Xiao X, Ouyang Q, Meng X, Li Z, Huang M, Jiang Z, Guan S. Budd–Chiari syndrome with occlusion of hepatic vein: Multi-slice spiral CT diagnosis and its clinical significance in the treatment. Zhonghua Yi Xue Za Zhi 2005;85:303–307. Takayasu K, Moriyama N, Muramatsu Y, Goto H, Shima Y, Yamada T, Makuuchi M, Yamasaki S, Hasegawa H, Hojo K. Intrahepatic venous collaterals forming via the inferior right hepatic vein in 3 patients with obstruction of the inferior vena cava. Radiology 1985; 154:323–328. Ueda K, Matsui O, Kadoya M, Yoshikawa J, Gabata T, Kawamori Y, Takashima T. CTAP in Budd–Chiari syndrome: Evaluation of intrahepatic portal flow. Abdom Imaging 1998;23:304–308. Williams PL, Bannister LH, (eds). Gray’s anatomy. 38th ed. London: Churchill Livingstone; 1995. Xing X, Li H, Liu W. Clinical studies on inferior right hepatic veins. Hepatobiliary Pancreat Dis Int 2007;6:579–584. Zhou P, Ren J, Han X, Wu G, Zhang W, Ding P, Bi Y. Initial imaging analysis of Budd–Chiari syndrome in Henan province of China: Most cases have combined inferior vena cava and hepatic veins involvement. PLoS One 2014;9:e85135.
http://dx.doi.org/10.1016/j.ultrasmedbio.2015.03.023
d
Original Contribution ULTRASONOGRAPHIC VISUALIZATION OF ACCESSORY HEPATIC VEINS AND THEIR LESIONS IN BUDD–CHIARI SYNDROME SHI-FENG CAI,* YONG-HAO GAI,y SHUANG MA,z BO LIANG,y GUANG-CHUAN WANG,x and QING-WEI LIU* * Department of Radiology, Provincial Hospital Affiliated to Shandong University, Jinan City, Shandong Province, China; Department of Ultrasound, Provincial Hospital Affiliated to Shandong University, Jinan City, Shandong Province, China; z Department of Ultrasound, Fifth Hospital of Jinan, Jinan City, Shandong Province, China; and x Department of Gastroenterology, Provincial Hospital Affiliated to Shandong University, Jinan City, Shandong Province, China
y
(Received 19 September 2014; revised 13 February 2015; in final form 19 March 2015)
Abstract—The aim of this study was to investigate the ultrasonographic features of accessory hepatic veins (AHVs) and their lesions in Budd–Chiari syndrome (BCS). Three hundred patients with BCS were examined by ultrasonography with multifrequency (3–6 MHz) convex transducers. Sonography was performed 1 to 2 wk before digital subtraction angiography and computed tomography angiography or magnetic resonance imaging. Using sonograms, we evaluated the number, course, diameter, orifice, lesions and hemodynamics of patent and obstructed AHVs. Ultrasonography was superior to digital subtraction angiography, computed tomography angiography and magnetic resonance imaging in revealing AHV lesions and hemodynamics. Dilated AHVs were detected in 227 patients. There were 239 caudate lobe veins in 167 patients and 168 inferior right hepatic veins in 151 patients. Both caudate lobe veins and inferior right hepatic veins were found in 91 of the 227 patients. The inlets to AHVs were located mainly on the right lateral or right anterior wall of the inferior vena cava, and the remnant, on the left lateral wall. AHV lesions comprised mainly septal obstruction and segmental stenosis. The hemodynamics of AHVs varied with the condition of inferior vena cava and AHVs. Ultrasonic examination can reveal AHVs and their lesions in patients with BCS and is helpful in choosing and planning therapeutic approaches. (E-mail: [email protected]) Ó 2015 World Federation for Ultrasound in Medicine & Biology. Key Words: Ultrasonography, Budd–Chiari syndrome, Thrombosis, Caudate lobe vein, Inferior right hepatic vein, Inferior vena cava.
other hepatic veins. Most patients with BCS have a chronic condition and have hepatic vein obstruction (Gai et al. 2014b; Zhou et al. 2014). The continuous increased pressure resulting from obstructed hepatic veins may eventually transform the anastomoses into enlarged collaterals. Blood flow from the obstructed hepatic veins can be drained by dilated AHVs through the collateral pathways. Kanamura et al. (2001) described the anatomy of caudate lobe veins, including their opening and location. Ueda et al. (1998) and Bargallo et al. (2003, 2006) visualized dilated AHVs by computed tomography and sonography in patients with BCS, respectively. Of note, Bargallo et al. (2003) suggested that visualization of the caudate vein ($3 mm in diameter) with gray-scale sonography strongly supported the diagnosis of BCS. However, descriptions of AHV lesions are rare; only Bargallo et al. (2003) mentioned that the caudate vein was obstructed in some cases, but did not describe it further. Importantly, ultrasonic examinations
INTRODUCTION Budd–Chiari syndrome (BCS) is a group of diverse conditions associated with obstruction of hepatic venous outflow in the large hepatic vein or the extrahepatic segment of the inferior vena cava (IVC). Anatomically, hepatic veins are divided into the superior and inferior groups. The superior group includes the left, middle and right hepatic veins, which are the main draining veins of the liver; the inferior group includes many hepatic veins with small diameters that are also called accessory hepatic veins (AHVs) (Williams and Bannister 1995). Normally, small anastomoses are found between adjacent AHVs and hepatic veins or between hepatic veins and Address correspondence to: Qing-wei Liu, Department of Radiology, Provincial Hospital Affiliated to Shandong University, No. 324, Jingwu Road, Jinan City 250021, Shandong Province, China. E-mail: [email protected] Conflicts of Interest: Authors declare no financial or non-financial competing interests. 2091
2092
Ultrasound in Medicine and Biology
provide an accurate way to detect AHVs and to classify obstructed AHVs in patients with BCS, which can be helpful in choosing and planning therapeutic approaches. In this study, we used ultrasonography to investigate the distribution, lesions and hemodynamic characteristics of patent and obstructed AHVs. METHODS Patients This retrospective study was based on data obtained with sonography between April 2005 and June 2014 from 300 patients with chronic BCS, including 158 male and 142 female patients (age range, 18–77 y; mean age, 45.48 6 10.82 y). In all patients, the time from the first clinical symptom to diagnosis ranged from 10 d to 24 y. However, some patients had been suffering with the disease for years before seeing a physician, and the duration of the disease before sonographic examination could not be accurately determined. The etiology was mainly idiopathic, and only four patients had previously undergone abdominal surgery. None of the BCS cases was caused by long-term oral contraceptive medication or pregnancy. Although hepatic vein thrombosis was detected in 14 patients, bone marrow aspiration and blood tests revealed that it was not related to globulism, antiphospholipid syndrome, intermittent hemoglobinuria or any myeloproliferative disease. The main clinical symptoms were right upper abdominal distention in 106 cases, abdominal wall varicosis in 59 cases and edema of both lower extremities in 118 cases. All procedures were approved by the ethics committee of Shandong University. Informed consent was obtained from all patients or their families. Ultrasonography Ultrasonography was performed using a Logiq E9 (GE Healthcare, Waukesha, WI, USA), an HDI 3500 (ATL-Philips, Bothell, WA, USA) or an Envisor HD system (Philips Healthcare, Best, The Netherlands) with multifrequency (3–6 MHz) convex transducers. All patients fasted more than 8 h before the examination and were scanned in the supine or left recumbent position. First, the liver, spleen and IVC were observed with B-mode sonography in search of hepatomegaly, splenomegaly, portal vein dilation and ascites. In the presence of hepatic vein and IVC obstruction, the number and site of outlets of draining AHVs were carefully observed, and the number and distribution of intrahepatic communicating branches were also recorded. The site we chose for measuring AHV diameter was large and was possibly the diameter at the end of inspiration. Pulsed wave Doppler sonography was used to assess blood flow direction and velocity within the draining AHVs. At sites
Volume 41, Number 8, 2015
where B-mode sonography revealed narrowing along the draining hepatic vein, color and spectral Doppler revealed aliasing and high flow velocity. We based our determination of AHVs on the fact that caudate lobe veins are the largest veins draining from the caudate (Spiegel) lobe and flow directly into the retrohepatic segment of the IVC (Bargallo et al. 2003, 2006; Kanamura et al. 2001). According to the nomenclature of Couinaud, the course of the inferior right hepatic veins lies mainly in hepatic segment VI. In transverse section, these veins appear on the posterior medial side of the posterior branch of the right portal vein, within the hepatic parenchyma, in front of the renal imprint, and enter the IVC at the level of the first portal hilum (Xing et al. 2007). Sonography was performed 1 to 2 wk before digital subtraction angiography (DSA) and computed tomography angiography (CTA) or magnetic resonance imaging (MRI). DSA was performed in all patients, whereas CTA and MRI were performed in 83 and 29 cases, respectively. When sonograms of AHVs and their lesions agreed with DSA and/or CTA (and/or MRI) images, the results were considered valid. Informed consent was obtained from each subject before DSA, CTA or MRI, and the protocol was consistent with the Declaration of Helsinki. Statistical analysis Statistical analysis was performed using SPSS Version 17.0 (SPSS, Chicago, IL, USA). Student’s t-test was used to determine the significance of statistical differences between two groups. Data are presented as the mean 6 standard deviation. Values with p , 0.05 were considered statistically significant. RESULTS Most patients with dilated AHVs have mixed BCS, whereas a small number have isolated hepatic vein-type BCS Dilated AHVs ($3 mm in diameter) were detected in 227 patients, including 32 patients with obstructed AHVs (patients with obstructed AHVs were not included in calculation of the average diameter), who accounted for 75.7% of the total number of patients (300) with BCS (Table 1). Color Doppler revealed dilation in most AHVs; the distal part of the hepatic veins, which were proximately occluded, drained through communicating branches into AHVs and subsequently into the IVC (Fig. 1). In addition to the dilated AHVs, there were a small number of tiny AHVs (,3 mm in diameter) that did not flow into collateral vessels. Long courses with acute angles between the long axis of the inferior right hepatic veins and IVC and relatively short courses of caudate lobe
Accessory hepatic vein lesions in Budd–Chiari syndrome d S.-f. CAI et al.
Table 1. Information on AHVs*
AHV
Number of AHVs per patient
Diameter (mm)
CLV IRHV
1.89 6 0.62 (1–4)y 1.13 6 0.39 (1–5)
7.64 6 2.74 (2–15) 9.38 6 3.05 (3–19)
Total number of AHVs/number of patients 239/167 168/151
AHV 5 accessory hepatic vein; CLV 5 caudate lobe vein; IRHV 5 inferior right hepatic vein. * Both CLVs and IRHVs were found in 91 of the 227 patients. y Mean 6 standard deviation (range).
veins nearly perpendicular to the IVC were observed in most of the patients, possibly resulting in higher blood reflux resistance of the inferior right hepatic veins in contrast to the caudate lobe veins. The inlets to all inferior right hepatic veins were located on the right lateral or right anterior wall of the IVC. Caudate lobe veins were mainly located on the anterior wall; only a small number of caudate lobe veins were located on the left lateral wall. Among the 227 patients, 197 patients had mixed BCS, in which obstructed hepatic veins and IVCs co-existed in the same patient. Only 30 cases had isolated hepatic veintype BCS, in which only the hepatic veins were obstructed. No isolated IVC type (obstruction in only the IVC) was observed in this study. These data indicated that 227 of the 300 patients had dilated AHVs, among which 197 patients had mixed BCS and 30 patients had isolated hepatic vein-type BCS. There are more lesions in inferior right hepatic veins than in caudate lobe veins There were 23 obstructed inferior right hepatic veins in 23 patients. Lesions of inferior right hepatic veins were classified mainly into septal obstruction (septum thickness #1 cm, including septal stenosis and
2093
septal occlusion) and segmental stenosis (stenosed length .1 cm). Segmental stenosis affected seven inferior right hepatic veins; the ultrasonograms revealed that the proximal lumens of the involved vessels were stenosed with a thickening wall. The stenosed length of the lumens varied from 10 to 35 mm, whereas the diameter ranged from 2 to 5 mm (Fig. 2a). Septal stenosis was detected in five inferior right hepatic veins, and septal occlusion was detected in one inferior right hepatic vein. Ultrasonography revealed that the septum was located on the inlet to the inferior right hepatic vein with or without extension to the lumen of the IVC (Fig. 2b). Thrombosis affected 10 inferior right hepatic veins, with ultrasonography revealing the hypo-echogenicity characteristic of fresh thrombus and hyper-echogenicity characteristic of calcified thrombus (Fig. 2c, d). Lesions of the caudate lobe veins were relatively rare, with only eight cases of septal stenosis and one case of septal occlusion (Fig. 3a, b). However, the inlet to the caudate lobe vein might be stenosed because of involvement of thrombus from the IVC (Fig. 3c). These data suggested that inferior right hepatic veins had more lesions than caudate lobe veins.
Hemodynamic characteristics vary depending on status of AHVs Sonograms revealed the absence of blood flow in two inferior right hepatic veins that were completely thrombosed by fresh thrombus and appeared homogeneously hypo-echoic, but a blood-filling defect was observed in the remaining veins that were partly embolized or recanalized. Reversed blood flow in the AHVs was observed in 7 patients with a cavo-hepato-atrial pathway in which the IVC was occluded between the patent AHV and hepatic vein; venous flow from the IVC
Fig. 1. (a) Sonogram in the sagittal epigastric line of a 54-y-old woman with Budd–Chiari syndrome, revealing two dilated patent caudate veins draining into the IVC. (b) Sonogram in the right intercostal antero-inferior/posterosuperior plane of a 47-y-old man with Budd–Chiari syndrome, revealing dilated patent inferior right hepatic vein with almost flat spectrum caused by the completely occluded IVC. CL 5 caudate lobe; CLV 5 caudate lobe vein; IRHV 5 inferior right hepatic vein; IVC 5 inferior vena cava. Cycle 5 3, mean velocity 5 6.92 cm/s, accelerated time 5 0.010 s.
2094
Ultrasound in Medicine and Biology
Volume 41, Number 8, 2015
Fig. 2. (a) Sonogram in the subcostal plane of a 56-y-old man with Budd–Chiari syndrome, revealing segmental stenosis of inferior right hepatic vein with aliasing blood flow (arrow). (b) Sonogram in the subcostal plane of a 72-y-old man with Budd–Chiari syndrome, revealing a septum on the inlet of the inferior right hepatic vein extending to the lumen (arrow) of the inferior vena cava with narrow aliasing blood flow. (c) Sonogram in the right intercostal antero-inferior/posterosuperior plane of a 37-y-old woman with Budd–Chiari syndrome, revealing the recanalized thrombus and blood-filling defect (arrow) of the inferior right hepatic vein. (d) Sonogram in the right intercostal antero-inferior/postero-superior plane of a 57-y-old man with Budd–Chiari syndrome, revealing the calcified thrombus and blood-filling defect (arrow) of the inferior right hepatic vein. IRHV 5 inferior right hepatic vein, IVC 5 inferior vena cava.
reversed to the AHV and intrahepatic communicating branches, onward to a hepatic vein and finally to the right atrium, as reported previously (Gai et al. 2014a). Blood from the septal occluded inferior right hepatic vein and the caudate lobe vein of the patients did not enter the IVC, but drained into the right hepatic subcapsular vein (Fig. 3b). Color Doppler studies of the remaining patients revealed that blood from obstructed hepatic veins drained through communicating branches into the AHVs and, subsequently, into the IVC (Fig. 1a, b). The spectrum of patent AHVs was similar to that of patent hepatic veins in patients with patent IVCs in this study. Using the condition of the lumen to determine whether IVC was obstructed, we classified patients with patent AHVs (excluding the 7 patients with cavohepato-atrial pathways) into the obstructed and nonobstructed groups. The peak velocity of AHVs in each group was calculated accordingly (Table 2). The results indicated that the peak velocity of AHVs in the group with obstructed IVCs was lower than that in the group with non-obstructed IVCs. In addition, more severe obstruction of IVCs led to more marked changes in spectral patterns of AHVs. When IVCs were completely occluded, the spectral patterns almost flattened (Figs. 1b and 3d).
At sites where proximal parts of the AHVs were obviously stenosed, color and spectral Doppler revealed aliasing and high-velocity regardless of whether the IVC was obstructed. The velocity of the inferior right hepatic veins ranged from 90 to 225 cm/s, and that of the caudate lobe veins, from 65 to 112 cm/s (Figs. 2a, b and 4). These data suggested that hemodynamic characteristics varied depending on the status of AHVs. Ultrasonography is superior to DSA, CTA and MRI in revealing AHV lesions and hemodynamics As mentioned earlier, 32 of the 227 patients had obstructed AHVs; the remaining 195 patients had patent AHVs and clear blood flow as observed by ultrasonography, which was consistent with the DSA results. Compared with other imaging techniques, ultrasonography was superior in accurately detecting lesions in AHVs and specifying lesion type (Table 3). Lesions that were not identified by DSA included two completely thrombosed inferior right hepatic veins, a septal-occluded inferior hepatic vein and a septal-occluded caudate lobe vein. Seven patients with AHV lesions identified by CTA included two cases of segmental stenosis of inferior right hepatic veins, three cases of thrombus in inferior right hepatic veins (two cases with calcified thrombus),
Accessory hepatic vein lesions in Budd–Chiari syndrome d S.-f. CAI et al.
2095
Fig. 3. (a) Transverse sonogram through the epigastric region of a 29-y-old woman with Budd–Chiari syndrome (BCS), revealing septal stenosis at the inlet of the caudate lobe vein. (b) Transverse sonogram through the right epigastric region of a 46-y-old man with BCS, revealing septal occlusion at the inlet of the caudate lobe vein (arrow) and blood draining into the right hepatic subcapsular vein (red). (c) Transverse sonogram through the epigastric region of a 40-y-old woman with BCS, revealing the stenosis inlet of the caudate lobe vein caused by involvement of the thrombus from the inferior vena cava. (d) Transverse sonogram through the epigastric region of a 39-y-old woman with BCS, revealing patent caudate lobe vein with flat septum caused by obstruction of the inferior vena cava. CLV 5 caudate lobe vein. IVC 5 inferior vena cava. Cycle 5 3, mean velocity 5 9.39 cm/s, accelerated time 5 0.029 s.
one case of septal stenosis of inferior right hepatic veins and one case of septal stenosis of caudate lobe veins; in the two patients with septal stenosis, CTA revealed no septum, but locally clear blood flow. However, veins with calcified thrombus and segmental stenosis were clearly visualized. Three patients with AHV lesions examined by MRI included two cases with a thrombosed inferior right hepatic veins (one fresh thrombus and one calcified thrombus) and one case with septal occlusion of caudate lobe vein. Fresh thrombus was correctly diagnosed by iso-intensity on T1-weighted imaging and hyper-intensity on T2-weighted imaging, but patients with calcified thrombus were not successfully diagnosed. For patients with septal occlusions, MRI revealed arc-like low signals from the septum (Fig. 5a, b). The results of ultrasonography, DSA, CTA and MRI are summarized in Table 3. These data suggested that ultrasonography was superior to DSA, CTA and MRI in demonstrating AHV lesions and hemodynamics. Treatment strategies based on AHV status Among the 227 patients, 195 had patent AHVs. Only one patient with a patent AHV diameter ,0.5 cm underwent surgery to relieve the hepatic vein obstruction;
the others patients with patent AHVs were not treated. Among the 32 patients with AHV lesions, all patients with septal stenosis or occlusion of the inferior right hepatic veins and caudate lobe veins underwent percutaneous transluminal angioplasty, two patients with segmental stenosis of the inferior right hepatic veins underwent percutaneous transluminal angioplasty and five patients with segmental stenosis of the inferior right hepatic veins underwent stent placement. All lumens were patent after surgery (Fig. 6). The other seven cases
Table 2. Comparison of peak velocity of AHVs between group with obstructed IVCs and group with unobstructed IVCs* Peak velocity (cm/s) IVC condition
IRHV
CLV
Obstructed group (n 5 154) Unobstructed group (n 5 30) p Valuez
12.54 6 4.26y 28.36 6 9.18 0.000
13.39 6 4.95 30.25 6 9.58 0.000
AHV 5 accessory hepatic vein; CLV 5 caudate lobe vein; IRHV 5 inferior right hepatic vein; IVC 5 inferior vena cava. * The velocity of AHVs was measured at the end of inspiration. y Mean 6 standard deviation. z p , 0.05 was considered statistically significant.
2096
Ultrasound in Medicine and Biology
Fig. 4. Sonogram in the subcostal plane of a 47-y-old woman with Budd–Chiari syndrome, revealing the aliasing and highflow velocity of 112 cm/s at the stenosis inlet of the inferior right hepatic vein. IRHV 5 inferior right hepatic vein. Peak velocity 5 112 cm/s, pressure gradient 5 5.00 mm Hg.
with thrombosed inferior right hepatic veins underwent surgery to open hepatic veins because of poor drainage. Four patients who had membranous or short segmental occlusion of the IVC with cavo-hepato-atrial pathways underwent angioplasty for re-canalization of the IVC, after which blood flow returned to normal. Three patients with a small amount of thrombus were left untreated and followed up 1–5 y without surgery; their ultrasonograms and clinical manifestations did not change significantly. Among the 227 patients, 177 had various degrees of obstruction in the IVC and underwent intervention therapy or surgery. These data suggested that ultrasonography was helpful in choosing and planning therapeutic approaches. DISCUSSION In this study, ultrasonography clearly revealed the number, course, diameter, presence of lesions and variations in hemodynamics of AHVs. First, we not Table 3. The diagnosis of ultrasonography, DSA, CTA and MRI Number of patients AHV status
Diagnosis
US
DSA
CTA
MRI
Patent*
Detected Undetected Detected Undetected Correct Incorrect
195 0 32 0 73 0 300
195 0 28 4 73 0 300
60 0 7 2 14 0 83
19 0 2 1 7 0 29
Afflicted No AHVs Total
AHV 5 accessory hepatic vein; CTA 5 computed tomography angiography; DSA 5 digital subtraction angiography; MRI 5 magnetic resonance imaging; US 5 ultrasonography. * The 7 patients with cavo-hepato-atrial pathways were included.
Volume 41, Number 8, 2015
only distinguished AHVs according to their anatomic locations, but also confirmed the locations of the inlets to AHVs, which could guide interventional therapy for both AHVs and the IVC, particularly to avoid secondary stenosis of the inlets to AHVs caused by stent oppression. Second, we found AHV lesions and evaluated their hemodynamics. When hepatic veins were obstructed and the blood from these veins was drained by AHVs, dilation of AHVs and collaterals was visualized by ultrasonography. If the IVC was obstructed at the same time, the dilation was even more obvious. Meanwhile, obstruction of IVCs led to increased hypertension and decreased blood flow velocity of AHVs, which might accelerate hepatic congestion. The results of this study indicated that the flow velocity of AHVs in the obstructed group was significantly lower than that in the non-obstructed group of IVCs. Therefore, it was suggested that the decreased flow velocity of AHVs might be used as a criterion to predict increased congestion of the liver. On the other hand, our study revealed that the spectrum of patent AHVs in normal patients was similar to that of patent hepatic veins in patients with patent IVCs, namely, phasic flow in response to both cardiac and respiratory cycles. However, in patients with partial IVC obstructions, continuous flow (also called a pseudo-portal Doppler signal) might appear with more severe obstruction of the IVC corresponding to more marked changes in spectral patterns of AHVs. When the IVC was completely occluded, the Doppler signal became almost flat. However, the Doppler signal of AHVs returned to its phasic pattern after the IVC obstruction was recanalized. Consequently, we speculate that the variation in velocity and spectrum patterns of AHVs can also be used as a criterion in evaluating the effect of recanalization of IVCs. The cavo-hepato-atrial pathway is an unusual blooddraining pathway in BCS, with the pathologic consequence that blood pressure below the obstructed portion of the IVC exceeds that of hepatic veins. Therefore, blood from the IVC reverses to the AHVs, through the enlarged intrahepatic collaterals and draining hepatic veins (orifice above occlusion), and finally to the right atrium. This blood-draining pathway relieves the symptoms caused by portal hypertension and IVC hypertension (Akaki et al. 1995; Kamba et al. 1995; Takayasu et al. 1985). Hepatopetal flow in AHVs can be clearly observed on ultrasonography, on which basis a definite diagnosis can be made (Gai et al. 2014a, 2014b). The results of our study indicated that the AHVs most involved by thrombus and segmental stenosis were the inferior right hepatic veins. The incidence of obstruction, especially obstruction caused by thrombus, was higher for the inferior right hepatic veins than for the caudate lobe veins. The greater susceptibility of inferior right hepatic veins to thrombus might be caused by
Accessory hepatic vein lesions in Budd–Chiari syndrome d S.-f. CAI et al.
2097
Fig. 5. For the patient whose sonogram is provided in Figure 3b: (a) Magnetic resonance image revealing the hypointense septum at the inlet of the caudate lobe vein. (b) Digital subtraction angiography failed to reveal the septum.
their higher resistance to regurgitation, in contrast to the caudate lobe veins, as mentioned earlier. Ultrasonography can be used to visualize thrombi with or without calcification, as well as blood-filling defects caused by thrombi. Sonograms of segmental stenosis revealed narrow lumens with a thickening wall at the proximal part of AHVs, but the lumen is sometimes too narrow to be visualized on gray-scale sonography. The diagnosis relies only on linear aliasing blood on color Doppler with accelerated velocity. Caudate lobe vein lesions comprised mainly stenosis of the inlets caused by septum and IVC-involved thrombus. The septum that caused relative stenosis of the inlet might have arisen from the wall of the IVC, which limited the inlet dilation compared with the obviously expanded distal lumen of the caudate lobe veins. Accelerated blood flow from stenosis inlets was also detected, but was usually lower than that from stenosis inlets to the inferior right hepatic veins. In cases of stenosis of the caudate lobe veins caused by IVC-involved thrombus, it was clearly observed that the thrombus arose
Fig. 6. Sonogram in the right intercostal antero-inferior/ postero-superior plane of a 42-y-old man, revealing the patent stent in the inferior right hepatic vein (IRHV) that was affected by segmental stenosis before surgery. IVC 5 inferior vena cava.
from the IVC and extended to the lumen of the caudate lobe veins. Because of its complex etiopathogenesis and multiple types, there are many treatment strategies for BCS. Therapeutic strategies are based on imaging techniques, which can reveal blood draining pathways of AHVs and collateral circulations of obstructed hepatic veins before surgery. The present study found that ultrasonography, DSA, CTA and MRI had similar sensitivity in diagnosing patients without AHVs. However, for revealing lesions, ultrasonography was superior to the other imaging techniques, especially in depicting the septum of AHVs, even when septa were #3 mm thick. Completely thrombosed AHVs cannot be identified on DSA, but can be clearly visualized by ultrasonography. The sensitivity of CTA and MRI in detecting lesions in AHVs requires further study because of the small number of obstructions examined in this study. It is generally accepted that if the hepatic veins are incompletely obstructed (i.e., at least one of the three main hepatic veins remains patent), only the obstructed IVC requires surgery. If the three main hepatic veins are completely obstructed, treatment must be directed to the hepatic veins. Complex treatment of obstructed hepatic veins can be avoided when large draining AHVs and rich collaterals are observed (Shan et al. 2005). Even when the three main hepatic veins at the second hilum and the AHVs (draining vessels) at the third portal hilum are stenosed or obstructed, treatment can be limited to the inadequate draining vessels at the third portal hilum if rich intrahepatic communicating branches connecting the two groups of vessels can be visualized. Treatment centered on AHVs at the third portal hilum is usually easier and more tolerable than that on the hepatic veins at the second portal hilum, because the outlets of the AHVs to the right atrium are longer than those of the main hepatic veins. Therefore, detailed assessment of every AHV at the third portal hilum is mandatory before surgery. In our study, both the lesions and hemodynamics of the AHVs were clearly visualized on ultrasonography,
2098
Ultrasound in Medicine and Biology
and the operations were performed successfully after examination. In conclusion, ultrasonographic examination not only provides a convenient and accurate way to visualize AHVs and their lesions in patients with BCS, but also helps in choosing and planning therapeutic approaches. Acknowledgments—This work was supported by the Shandong Provincial Science and Technology Development Project Foundation of China (Grants 2013 GSF11827 and 2012 GSF11820).
REFERENCES Akaki S, Kanazawa S, Gochi A, Nakamura K, Yasui K, Togami I, Hiraki Y, Hamazaki K. Asymptomatic membranous obstruction of the inferior vena cava due to large intrahepatic collaterals. Cardiovasc Intervent Radiol 1995;18:403–405. Bargall o X, Gilabert R, Nicolau C, Garcıa-Pagan JC, Ayuso JR, Bru C. Sonography of Budd–Chiari syndrome. AJR Am J Roentgenol 2006; 187:W33–W41. Bargall o X, Gilabert R, Nicolau C, Garcıa-Pagan JC, Bosch J, Bru C. Sonography of the caudate vein: Value in diagnosing Budd–Chiari syndrome. AJR Am J Roentgenol 2003;181:1641–1645. Gai Y, Cai S, Fan H, Liu Q. Diagnosis of the cavo-hepato-atrial pathway in Budd–Chiari syndrome by ultrasonography. Exp Ther Med 2014a;8:793–796.
Volume 41, Number 8, 2015 Gai Y, Cai S, Guo W, Zhang C, Liang B, Jia T, Zhang G. Sonographic classification of draining pathways of obstructed hepatic veins in Budd–Chiari syndrome. J Clin Ultrasound 2014b;42:134–142. Kamba M, Ochi S, Ochi H, Maruyama S, Sato H, Suto Y. Asymptomatic membranous obstruction of the inferior vena cava forming intrahepatic collateral pathways. J Gastroenterol 1995;30:783–785. Kanamura T, Murakami G, Hirai I, Hata F, Sato TJ, Kumon M, Nakajima Y. High dorsal drainage routes of Spiegel’s lobe. J Hepatobiliary Pancreat Surg 2001;8:549–556. Shan H, Zhu K, Xiao X, Ouyang Q, Meng X, Li Z, Huang M, Jiang Z, Guan S. Budd–Chiari syndrome with occlusion of hepatic vein: Multi-slice spiral CT diagnosis and its clinical significance in the treatment. Zhonghua Yi Xue Za Zhi 2005;85:303–307. Takayasu K, Moriyama N, Muramatsu Y, Goto H, Shima Y, Yamada T, Makuuchi M, Yamasaki S, Hasegawa H, Hojo K. Intrahepatic venous collaterals forming via the inferior right hepatic vein in 3 patients with obstruction of the inferior vena cava. Radiology 1985; 154:323–328. Ueda K, Matsui O, Kadoya M, Yoshikawa J, Gabata T, Kawamori Y, Takashima T. CTAP in Budd–Chiari syndrome: Evaluation of intrahepatic portal flow. Abdom Imaging 1998;23:304–308. Williams PL, Bannister LH, (eds). Gray’s anatomy. 38th ed. London: Churchill Livingstone; 1995. Xing X, Li H, Liu W. Clinical studies on inferior right hepatic veins. Hepatobiliary Pancreat Dis Int 2007;6:579–584. Zhou P, Ren J, Han X, Wu G, Zhang W, Ding P, Bi Y. Initial imaging analysis of Budd–Chiari syndrome in Henan province of China: Most cases have combined inferior vena cava and hepatic veins involvement. PLoS One 2014;9:e85135.