Introduction

Extrahepatic portal vein aneurysm (PVA) is a condition in which the extrahepatic portal vein is partially dilated into a sac-like or spindle-like shape [1]. With the development of imaging tests, the number of cases discovered incidentally is increasing. We report a case of thrombus formation in an extrahepatic PVA resulting in extrahepatic portal vein obstruction (EHO) after trauma. Ultrasonography (US) revealed interesting changes in the thrombus formation process over time. This case might help elucidate the mechanism of EHO.

Case report

The patient was a 37-year-old Japanese male with no specific medical or family history and no history of alcohol drinking, smoking, or allergies. He was brought to his local doctor as a result of a self-inflicted accident while driving a car at 40 km/h and wearing a seat belt. The airbag was deployed. The patient had mild abdominal pain. Vital signs were normal, but Focused Assessment with Sonography Trauma (FAST) showed a large extrahepatic PVA measuring 70 × 50 × 90 mm (Fig. 1). Computerized tomography (CT) showed a low-density area around the aneurysm (Fig. 2). The patient was transferred to our hospital for emergency care due to concerns about bleeding and PVA rupture.

Fig. 1
figure 1

Abdominal ultrasonography immediately after the trauma. a Right intercostal manipulation. b Longitudinal manipulation of the pericardium. An extrahepatic portal vein aneurysm of 70 × 50 × 90 mm contiguous to the main trunk of the portal vein was observed

Full size image
Fig. 2
figure 2

Computerized tomography (CT) of the abdomen immediately after the trauma. a Plain CT showed an extraportal vein aneurysm (PVA) (arrowhead), with contrast enhancement on the margins of the PVA b during the arterial phase and c in the interior during the late phase. d In coronal images, a low-density area (asterisk) was observed around the caudal side of the PVA, suggesting hemorrhage

Full size image

He was 172 cm tall and weighed 69.2 kg (body mass index, 23.2 kg/m2). Blood pressure was 131/79 mmHg, pulse was 97 beats per minute, and SpO2 was 95% while breathing room air. His mental status was normal. He complained of epigastric pain. On physical examination, his abdomen was flat and soft and he had a seatbelt indentation. Laboratory data (Table 1) showed increased white blood cell count, elevated aspartate aminotransferase (AST) and alanine aminotransferase (ALT) levels, but normal hemoglobin levels, negative CRP, and no abnormalities of the coagulation-fibrinolytic system. The radiology, vascular surgery, and gastroenterological surgery departments discussed the indications for interventional radiology and surgery. Since there was no obvious extravasation and the procedure would be very difficult, they determined that there were no indications for aggressive treatment. Therefore, the patient was admitted to our department for conservative monitoring.

Table 1 Laboratory data at the time of hospitalization
Full size table

The FIB-4 index was 0.93 and Mac-2-Binding Protein Glucosylation isomer (M2BPGi) COI was 0.46, which were not high. Liver morphology was almost normal based on US and CT. The shear wave elastography (SWE) value was 1.14 m/s. Liver stiffness based on transient elastography was 3.8 kPa. There was no liver fibrosis; liver disease was ruled out. However, since there was no previous imaging, we could not determine whether the PVA was congenital or caused by trauma. The patient’s abdominal pain resolved, vital signs remained stable, and there was no worsening of anemia or liver injury.

When abdominal US was performed again 5 days after the trauma, the interior of the PVA included a multilayered, highly echogenic structure resembling a baumkuchen and there was rapid thrombus formation (Fig. 3). Color Doppler showed loss of the blood flow signal in the center of the aneurysm, but blood flow to the liver was maintained on the dorsal side of the aneurysm. A cavernous transformation was observed on the ventral side. D-dimer levels increased from 0.5 to 3.4 μg/mL, thrombin–antithrombin complex (TAT) was as high as 3.8 ng/mL, and platelet count decreased from 241 to 160 × 103/μL, reflecting increased coagulation. Since the PVA was associated thrombus, the patient was evaluated at a tertiary liver care facility. The patient was followed conservatively because intrahepatic portal vein blood flow was maintained.

Fig. 3
figure 3

Abdominal ultrasonography with longitudinal manipulation of the pericardium at 5 days after the trauma. a The interior of the extraportal vein aneurysm (PVA) was multilayered and hyperechoic like a baumkuchen. Rapid thrombus formation was observed. b Color Doppler showed loss of the blood flow signal in the center of the PVA, but blood flow to the liver was maintained on the dorsal side of the PVA

Full size image

The patient was followed with US and contrast-enhanced CT at 1 month, 4 months, and 7 months after the trauma. At 7 months, the PVA had shrunk to 30 × 25 × 30 mm with thrombosis. Intrahepatic portal vein blood flow was maintained from a cavernous transformation that developed, but there was no blood flow within the aneurysm (Figs. 4 and 5). The spleen was enlarged and we felt that careful follow-up to monitor for the development of gastrointestinal varices was necessary.

Fig. 4
figure 4

Changes in abdominal ultrasonography findings over time. At a 1 month, b 4 months, and c, d and 7 months after the trauma, the PVA had shrunk with thrombus organization (white arrowhead). A cavernous transformation developed around the PVA and intrahepatic portal venous blood flow was preserved

Full size image
Fig. 5
figure 5

Changes in contrast-enhanced computerized tomography (CT) findings of the abdomen over time. a At 1 month after the trauma, CT showed no thrombus within the extraportal vein aneurysm (PVA) (black arrowhead), but b by 4 months after the trauma, the interior of the PVA became thrombosed. c By 7 months, the PVA had shrunk to 30 mm. Blood flow from the main trunk of the portal vein into the liver was disrupted, but development of collateral blood vessels around the PVA was observed. Blood flow into the liver was maintained (white arrow). The spleen was enlarged (asterisk)

Full size image

Discussion

This case report describes an extrahepatic PVA that rapidly thrombosed following trauma, resulting in size reduction of the PVA and occlusion of the extrahepatic portal vein. PVA were first described by Bazilai and Kleckner in 1956 [1]. PVAs are extremely rare, with an incidence of 0.06%. A PVA is defined as a cystic or spindle-shaped dilatation of a portion of the portal vein with a diameter of 1.9 cm in patients with liver cirrhosis and 1.5 cm or more in patients with normal livers [2]. According to Ogawa et al. intrahepatic PVAs are frequently reported in Japan, while extrahepatic PVAs are relatively rare [3]. There are two theories about the etiology of PVAs: congenital, due to the fragility of the portal vein wall and developmental abnormalities (e.g., failure of the yolk vein to retract during embryogenesis) [4, 5] and acquired, secondary to surgery or trauma, portal hypertension, or weakening of the portal vein wall as a result of pancreatitis or cholangitis [6]. This patient had no history of abdominal surgery. Liver morphology, fibrosis markers, and liver stiffness based on elastography were normal. Since there was no background liver disease, congenital PVA was inferred. However, this could not be confirmed because the patient had not undergone any imaging studies in the past. In this case, it is considered that PVA and subsequent thrombosis are induced by acute mechanical endothelial damage caused by trauma. Extrahepatic portal obstruction (EHO) is induced by a thrombus in the aneurysm, then this case classified as portal venous system thrombosis caused by trauma. Florent G et al. who have reported similarly case of PVA forming after trauma and a thrombus in the aneurysm a few months later, and pointed out that there was a latent genetic abnormality for thrombosis, and that the patient’s background should be investigated for the underlying disease of thrombosis [7]. In this case, there was no history of thrombosis, and the values of PT, APTT, fibrinogen, and D-dimer immediately after the accident were normal. Furthermore, on the 7th hospital day, the quantitation of PAI-1 and factor VIII were confirmed, and all of them were normal. Since the coagulation system was normal, further investigation of the genetic disease was not conducted, but careful follow-up was necessary in the future. Considering the rapid development of cavernous transformation after trauma, it is certain that trauma triggered the PVA.

Regarding treatment or management of PVAs, patients are generally asymptomatic and can be followed, but surgery can be considered in the presence of symptoms, rupture, perforation, thrombus formation, or tendency to enlarge [2]. In our patient, symptoms were very mild at the time of admission and there was no obvious rupture or thrombosis. Thus, the patient was placed under observation. Contrast-enhanced CT immediately after trauma showed a low attenuation layer around the extrahepatic portal aneurysm, so bleeding was suspected, but no obvious extravasation was observed, there was no decrease in hemoglobin, and the abdominal symptoms were mild and improving, so we thought that there was no bleeding in this case, and we followed up. But in past reports, it has been reported that surgery is indicated if there is bleeding [2]. The surgical procedure has not been determined, and it is a very difficult operation, and the postoperative mortality rate is high. Therefore, the bleeding is considered to be a significant factor in determining the prognosis in the future. However, US at 5 days after the trauma showed an organic thrombus in the PVA. Three factors are involved in thrombus formation: (i) changes in the vessel wall, (ii) changes in blood properties, and (iii) blood congestion or stagnation, i.e., Virchow’s triad [8]. The US study immediately after the trauma showed that blood flow in the aneurysm became turbulent, like in a whirlpool (Fig. 6). Point-like hyperechogenicity, presumed to represent microthrombi, can be seen floating in a whirlpool-like pattern in the aneurysm. In addition, moderately hyperechoic structures and comet signs were observed along the vessel wall of the PVA, suggesting changes in the vessel wall. Based on the US findings, we speculated that the trauma triggered injury to the PVA wall, which contributed to increased coagulability. Turbulent flow and congestion in the PVA might have contributed to rapid thrombus formation. In other words, this is a very rare case in which Virchow’s triad could be detected with US. There have been no such reports to date. Contrast-enhanced CT immediately after the trauma shows contrast enhancement in the margins of the PVA during the arterial phase and in the interior during the late phase, which shows spiral turbulence inside the aneurysm. However, US can provide a more detailed examination.

Fig. 6
figure 6

Review of abdominal ultrasonography findings immediately after trauma. Color Doppler shows spiral and turbulent blood flow signals in the extraportal vein aneurysm (PVA) (a right intercostal manipulation, b longitudinal manipulation of the infundibulum). c Pale echogenic structures (asterisks) and comet signs (arrows) were seen along the wall of the portal vein aneurysm, suggesting vessel wall injury. d A point-like hyperechoic structures (arrowhead) floating in the bloodstream like a spiral was observed, suggesting a microthrombus

Full size image

Duvoux C et al. performed outpatient anticoagulation therapy for traumatic portal vein thrombus and disappeared [9]. In addition, a liver biopsy was performed after the portal vein thrombus disappeared, but no obvious fibrosis was observed. No other causes, such as infection, blood disease, or hereditary coagulation abnormalities, were found. In this case if there was no bleeding, portal thrombolysis by anticoagulant therapy was recommended. In this case, after confirming the portal vein thrombus, the development of collateral circulation was observed by echo, the portal blood flow was maintained, and there was no decrease in liver reserve, so anticoagulation therapy was not performed and follow-up was performed, but the size of the portal vein thrombus tended to decrease gradually.

The patient had rapid thrombus formation. Surgery was reconsidered. However, the postoperative mortality rate of PVA is extremely high and patients with PVA require a complex, multidisciplinary approach. They should be referred to a high-volume tertiary hepatobiliary center for optimal management [2], which was done in this case. The main trunk of the portal vein was deficient in blood flow due to thrombus in the aneurysm, but intrahepatic portal vein blood flow was maintained by a cavernous transformation and the aneurysm tended to shrink. Therefore, the patient was considered not to be at high risk of death, and conservative follow-up was performed. Indeed, there has been a case report of PVA regression due to thrombus organization. Machida et al. reported a case in which intrahepatic portal vein blood flow was restored by formation of a cavernous transformation; the PVA shrank over approximately 6 months [10]. Turbulent flow within the PVA was observed with US, as in our patient. The authors speculated that a thrombus subsequently formed, which was absorbed and shrank the aneurysm. Thus, US is useful, because it allows for real-time observation of the PVA walls and blood flow in the PVA.

On the other hand, this case is suggestive of the mechanism of EHO. EHO can be classified into primary cases with no apparent cause and secondary cases with an apparent cause. Secondary causes include neonatal umbilical inflammation, tumor, extrahepatic portal vein thrombosis associated with cirrhosis or idiopathic portal hypertension (IPH), cholangitis, pancreatitis, and intraperitoneal surgery [11]. In our patient, the cause of portal vein thrombosis is certain; the patient had secondary EHO. The keywords “portal vein aneurysm” and “extrahepatic portal vein obstruction” were used in a search of the Japan Medical Abstracts Society database and PubMed. One case was reported in each database. Ishimuro et al. reported a case in which a 6 cm extrahepatic PVA thrombosed and resulted in EHO, but the patient did not develop portal hypertension due to the development of collateral channels, suggesting that surgery is not necessary if there is no background liver disease. Collateral channels can be formed without rupture [12]. In our patient similarly, a hepatopetal collateral channel was seen at the time of portal vein thrombus development and the PVA shrank as the portal vein thrombus grew, showing a favorable course with regards to reduction of aneurysm size without surgery. On the other hand, Yoshimoto et al. reported a case in which a thrombus developed in a congenital extrahepatic PVA that was removed via incision and thrombectomy. Subsequently, EHO developed due to recurrent portal vein thrombosis and splenomegaly and esophageal varices developed with the growth of collateral vessels, resulting in Hassab’s operation [13]. The complication rate of esophageal varices with EHO is considered to be high [14]. Our patient presented with splenomegaly, so it is necessary to be aware of complications associated with portal hypertension such as gastroesophageal varices. This case can be classified as secondary EHO, because the changes were captured over time, but if the course of the disease had remained unknown, it might have been diagnosed as primary EHO based solely on the final image. Among the patients diagnosed with primary EHO, it is possible that some may have developed EHO as a result of a similar course. This case is highly suggestive in considering the mechanism of EHO.

Conclusion

The PVA shrank and intrahepatic blood flow was maintained in the collateral vessels, but the patient had EHO and should be followed with attention to symptoms associated with portal hypertension. This case is suggestive of the mechanism of EHO.