| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Original Research |
1 All authors: Department of Radiology, Ba
kent University Faculty of
Medicine, Dadalo
lu Mah., Serin Evler 39 Sok. No 6,
Yüreir, 01250 Adana, Turkey.
Received February 24, 2007;
accepted after revision May 19, 2007.
Address correspondence to Z. Koc
(koczafer{at}gmail.com).
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. From 1998 to 2006, a total of 38 portal
venous system aneurysms identified in 25 patients were retrospectively
reviewed. The data of seven patients diagnosed using color Doppler sonography
or CT before March 2004 were not consecutive, but the data recorded thereafter
comprised the analysis of 4,186 consecutive patients who underwent routine
abdominal MDCT. The patients were 14 men and 11 women (mean age 
SD,
53 17 years).
RESULTS. The prevalence of portal venous system aneurysm among 4,186 consecutive patients was 0.43%. There were no differences with respect to patient age, patient sex, and intrahepatic or extrahepatic location of aneurysm between those with and those without portal venous system aneurysm. Seven of the 25 patients with portal venous system aneurysm were symptomatic because of portal vein thrombosis, and six of them had recurrence. These patients had significantly larger aneurysms than those without symptoms. Four of those seven symptomatic patients evaluated for thrombophilia had an underlying defect. Other associated findings were splenomegaly (n = 16), portal hypertension (n = 8), cirrhosis (n = 3), psoriasis (n = 2), portal vein variation (n = 1), chronic pancreatitis (n = 1), and cutis laxa (n =1).
CONCLUSION. Portal venous system aneurysms were rare in our study group but occurred more frequently than previously thought. All thrombosed aneurysms, most with a recurrence, were symptomatic and larger in patients with symptoms of portal venous system aneurysm than in those without symptoms. There were no differences among patients with portal venous system aneurysm and those without portal venous system aneurysm with respect to patient age and patient sex. Among those with aneurysms, there was no difference between subjects with intrahepatic versus extrahepatic aneurysms. A thrombophilic defect probably played a role in development of thrombosis in the portal venous system aneurysm.
Keywords: aneurysm • color Doppler sonography • CT • MDCT • portal venous system • sonography • vein
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Most people with a portal venous system aneurysm are asymptomatic, although portal venous system aneurysms can cause symptoms. External compression and rupture are rare complications. Thrombosis of a portal venous system aneurysm, however, occurs frequently and can lead to the development of portal hypertension with clinically severe consequences [8, 9]. Knowledge of the clinical aspects and imaging characteristics of portal venous system aneurysms is helpful in the management of complications. The advent of cross-sectional imaging technology, especially MDCT, has enabled radiologists to diagnose many more venous variations and anomalies, including portal venous system aneurysms, in recent years [10–12].
Although a number of studies on portal venous system aneurysms have been reported in the clinical and radiologic literature [1–3, 5–8, 13–15], the causes, clinical manifestations, and management of that venous disorder have not yet been clarified. Because portal venous system aneurysms are rare, conducting large-scale studies is difficult. Those research topics require investigation with analyses of more entities in a large study group. The aims of this study were to present the prevalence, imaging and associated clinical findings, and possible causes of portal venous system aneurysms in patients undergoing routine abdominal CT.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
SD, 56 16
years) who had undergone routine abdominal contrast-enhanced MDCT at our
hospital were reviewed for a portal venous system aneurysm. No patient was
excluded from the study, and 26 portal venous system aneurysms were identified
in 18 patients in this group. The 12 portal venous system aneurysms diagnosed
before March 2004 in seven patients were not consecutive. In all, 14 men and
11 women (age range, 24–81 years; mean SD, 53 17
years) were diagnosed as having a portal venous system aneurysm.
Image Acquisition and Processing
Before 2004, both gray-scale and color Doppler sonography imaging
examinations of all seven patients were performed on a sonography scanner with
a 2-4–MHz convex-array transducer (Sonoline Antares, Siemens Medical
Solutions). In this group, additional CT scans were obtained in two patients
on a helical CT scanner (Somatom AR Star, Siemens Medical Solutions).
All CT scans of patients examined after March 2004 were obtained on a 4-MDCT scanner (Sensation 4, Siemens Medical Solutions). The portal phase images used to assess the abdominal venous structures were acquired 60 seconds after a 120- to 150-mL IV bolus of nonionic contrast material (300 mg/mL of iohexol [Omnipaque, Amersham]) had been administered at 3–5 mL/s. The scanning parameters were as follows: a contiguous 2.5-mm collimation and 12.5 mm per 0.5-second table speed per 360° gantry rotation with a resultant pitch value of 1.25.
Follow-up imaging studies were performed 1–38 months after the
initial examination (mean interval, 8.7 7.2 months) in 17 of the 25
patients as follows: color Doppler sonography, eight patients; color Doppler
sonography and CT, two patients; CT, four patients; and CT, color Doppler
sonography, and MRI, three patients.
Image Interpretation and Data Collection
The medical records, gray-scale and color Doppler sonography images, and CT
images of the seven patients diagnosed before 2004 were retrospectively
reviewed and the imaging findings were recorded. In the second part (after
March 2004) of the study, all CT images of each patient were analyzed by the
same author, who had 6 years of experience in abdominal radiology at the time
of the review; the reviewer measured the size of the portal venous system
aneurysms and interpreted the associated imaging findings. For image
interpretation, the reviewer used the source data on a workstation (Volume
Wizard, Siemens Medical Solutions). Postprocessing techniques (maximum
intensity projection [MIP], multiplanar reconstruction, or volume rendering)
were used for detailed analysis of portal venous system aneurysms or suspected
portal venous system aneurysms. The thickness of the MIP images used in the
study ranged from 5 to 50 mm, but most ranged from 10 to 25 mm. We preferred
to adjust the thickness of the MIP according to the vein examined so that
thick MIP was used for the intrahepatic segments of the portal vein, the
branches of which are distributed in a wide volume.
The intrahepatic and extrahepatic segments of the portal vein, portal confluence, splenic vein, and superior and inferior mesenteric veins were measured at their maximum diameter. Measurements of the size of the portal vein and of any identified portal venous system aneurysm were performed with short-axis diameter at 2x magnification using a distance-measuring tool. Any focal saccular or fusiform dilatation of the portal venous system that was 21 mm or larger in diameter, including the bifurcation of the main portal vein, was accepted as diagnostic for an extrahepatic portal venous system aneurysm [1, 4, 5]. For intrahepatic portal vein branches, any vein larger than 9 mm in diameter that was significantly larger than the remaining segments of the same vein was considered an aneurysm [4, 7]. Except for measurement methods, the diagnostic criteria of portal venous system aneurysms of each imaging technique in this study are the same as those used in previous studies [1, 4, 5, 7].
When an aneurysm was diagnosed, the vein was measured at its maximum diameter. If a vein that was dilated like an aneurysm contained a segment of normal diameter between two aneurysmal sides, each side was considered a separate aneurysm. Associated portal vein variants and anomalies were also recorded, and the intrahepatic branching pattern and anomalies of the portal vein were classified according to previous reports [11, 12, 16, 17]. In addition to the data specified earlier, the radiologists noted any additional CT findings that may have been associated with a portal venous system aneurysm. Recurrent portal vein thrombosis was diagnosed if thrombosis of a new segment of the portal venous system that had not been present before was detected, flow in a recanalized segment of the vein was not present, or the patient was experiencing symptoms similar to those experienced in the first episode of portal vein thrombosis together with documented previous portal vein thrombosis. An arterial portographic study was performed in one patient, and portography during the direct transhepatic puncture of the portal vein for catheter-directed thrombolysis was performed in two patients.
In all patients with a portal venous system aneurysm, the reason for examination, clinical complaints, clinical and laboratory findings, results of follow-up imaging, treatment, and outcome were recorded if those data existed or were pertinent.
Statistical Analysis
Statistical analysis was performed with SPSS software (version 11, SSPS).
The chi-square test was used to evaluate the difference between frequencies of
portal venous system aneurysm in men and that in women. The independent-sample
Student's t test was used to assess for a relationship between
patient age and frequency of portal venous system aneurysms; significance of
difference in diameter between symptomatic and asymptomatic patients; and
significance of difference in diameter between intrahepatic and extrahepatic
portal venous system aneurysms. The analysis-of-variance test was used to
assess for significance of differences in diameter of the aneurysms in the
three most frequent locations. A p value of < 0.05 was considered
to indicate a statistically significant difference.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
|
|
|
|
|
|
Tests to identify thrombophilia were performed in four of the seven
patients with portal vein thrombosis, and all of those individuals had a
thrombophilia defect (Table 1).
The symptomatic patients had significantly larger aneurysms than did the
asymptomatic patients (mean SD, 33.6 9.9 and 23.1
3.3 mm, respectively; p < 0.0001). Three of the symptomatic
patients had direct hyperbilirubinemia, and one of these patients had biliary
colic because of compression of the biliary system by a large aneurysm
(Fig. 4A).
|
SD, 28.49.0
mm).
|
The average diameters of the extrahepatic (30 mm) and intrahepatic (25.5 mm) portal venous system aneurysms were not significantly different (p = 0.137). Among the extrahepatic aneurysms, the splenic vein aneurysms were the largest (mean, 31.8 mm), followed in order of descending size by aneurysms in the confluence (mean, 31.1 mm) and the body (mean, 29 mm) of the main portal vein. The mean aneurysmal diameters with respect to location were not different from each other (p = 0.925).
|
The additional imaging findings about the aneurysms are presented in Table 3. A patient with six portal venous system aneurysms also had an associated splenic artery aneurysm. Calcifications of the portal venous system aneurysms were observed in only three patients with portal vein thrombosis (Figs. 2A, 2B, 3A, 3B, and 4A).
Possible etiologic factors of portal venous system aneurysm were identified in 15 of the 25 patients (Table 4) but not in the remaining 10 patients. The results of clinical evaluation revealed portal hypertension in eight patients. In five of those patients, portal hypertension may have developed as a result of recurrent portal vein thrombosis, and in the remaining three patients, portal hypertension was caused by cirrhosis (Fig. 4A). A portal vein branching variation was directly related to the portal venous system aneurysm in one patient. Increased portal vein flow caused by a primary splenic lymphoma was identified in two patients. The associated findings of portal venous system aneurysm were psoriasis in two patients, one of whom also had recurrent portal vein thrombosis and thrombophilia defects. The other patient had a portal venous system aneurysm that was an incidental finding. Chronic pancreatitis (Fig. 4B), previous pancreatic surgery, or cutis laxa was identified in one patient each (Fig. 4C).
|
|
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Congenitally or developmentally defective segments of the portal venous system may give rise to an aneurysm caused by an increase in portal venous pressure [18]. Portal hypertension is an accepted cause of portal venous system aneurysms [4, 5]. Portal vein thrombosis is the second most common cause of portal hypertension [19], a finding seen in five of our patients. The association of portal hypertension and portal vein thrombosis has been previously reported [19, 20], and our findings suggest an association between portal venous system aneurysms and thrombosis.
In patients with a thrombophilic defect, an aneurysm can trigger thrombus formation by causing turbulent flow and stasis. On the other hand, thrombosis of the portal vein can cause portal venous system aneurysm formation. Venous thrombosis has been regarded as a multicausal disease, and as the number of risk factors increases, the probability of deep vein thrombosis also increases. This phenomenon may occur in an additive manner without interaction or in a supraadditive manner that results in much higher risk than the sum of their separate effects [21].
The pathogenesis of portal venous system aneurysms has been widely discussed but remains controversial [1, 5, 22]. A portal venous system aneurysm may be congenital or may be acquired as a result of weakening of the vascular wall [4, 5]. Congenital factors include an abnormality of the internal walls of the vessel, incomplete regression of the distal right primitive vitellin vein [4, 5, 20], or a variant branching pattern of the portal vein [17]. The latter was seen in one of our patients as total ramification of the intrahepatic portal vein branches [17]. Portal venous system aneurysm in one of our young patients with cutis laxa (patient 10) probably resulted from an inherent weakness of the vessel wall caused by the connective tissue defect (Tables 1 and 4). Arterial aneurysms related to weakness of vascular wall are often seen in patients with neurofibromatosis 1 [23] and Marfan syndrome [24], but venous aneurysm is very rare [25]. Associated neurofibromatosis 1 or Marfan syndrome was not identified in our patients with portal venous system aneurysms. No apparent etiologic factors were identified in 10 of our patients.
Chronic liver disease, portal hypertension, pancreatitis, trauma, and the effects of surgery have been reported as acquired causes of portal venous system aneurysm [1, 4–6, 15, 20]. Although the number of reported cases of portal venous system aneurysm has increased, that disorder is rarely described in patients with cirrhosis [4, 14]. In our study, portal venous system aneurysm was associated with cirrhosis in only three patients (12%) and with portal hypertension in eight patients (32%). These findings support the hypothesis that chronic liver disease and portal hypertension may be contributory, but are not essential, factors in the development of portal venous system aneurysm [1, 4]. We could not find any report revealing a possible association of psoriasis and portal venous system aneurysm with or without thrombosis in the English-language literature.
Although the reported mean diameter of a healthy portal vein varies considerably [5, 20], the reported maximum diameter of the portal vein does not exceed 15 mm in healthy subjects [2] and 19 mm in cirrhotic patients [26]. Therefore, a maximum portal vein diameter of more than 20 mm is commonly accepted as the diagnostic standard for extrahepatic portal venous system aneurysm [1, 4, 5]. There is not a universally accepted upper limit for the diameter of a portal vein that indicates an intrahepatic portal venous system aneurysm. We have used the definition of other experts that states that any vein larger than 9 mm in diameter and significantly larger than the remaining segments of the same vein is considered to be an aneurysm [7, 14].
Portal venous system aneurysms were once considered extremely rare lesions, but the number of incidentally diagnosed portal venous system aneurysms has recently increased, probably because of the advent of noninvasive imaging methods [14]. In our series, the prevalence of portal venous system aneurysm of 4.3 per 1,000 patients in patients who underwent CT was higher than the previously reported number of 0.6 per 1,000 patients, but the results of the latter study were based on sonographic findings [7]. New imaging techniques and more imaging examinations increase the chance for us to find portal venous system aneurysms. The referral pattern of our institution, which is a tertiary care center, also may be a contributing factor for the higher prevalence of portal venous system aneurysms than has previously been reported.
Our results revealed that portal venous system aneurysms were frequently (63%) extrahepatic. The main portal vein (53%) and splenic vein (21%) were more susceptible to the development of aneurysms than were other segments of the portal vein system (Table 2). Portal confluence (26%) was the most frequently affected segment of the main portal vein. These findings are similar to those of previous reports [4, 5]. It has been reported that extrahepatic aneurysms are larger than intrahepatic portal venous system aneurysms [5], but our study did not support that finding (p = 0.137). Most of the portal venous system aneurysms were fusiform in configuration, and neither the age nor the sex of the patient was a factor in portal venous system aneurysm development.
Color Doppler sonography and CT have been considered accurate and reliable methods for the diagnosis and follow-up imaging of portal venous system aneurysms and their complications [1, 4, 5, 15]. The diagnosis of portal venous system aneurysm can be made using MRI, which offers multiplanar capabilities and is noninvasive [5]. CT and MRI are helpful in the differential diagnosis of acute thrombosis of a portal venous system aneurysm [6, 26]. The use of venographic studies has been restricted to patients who require an interventional procedure [5].
The clinical aspects of portal venous system aneurysms are related to their size. Small aneurysms often produce no symptoms [14]. Patients may present with recurrent upper abdominal or epigastric pain; jaundice; or, rarely, gastrointestinal bleeding [4, 5, 14]. Portal vein thrombosis, portal hypertension, compression of the adjacent structures, and aneurysmal rupture are complications caused by portal venous system aneurysms [4]. Large portal venous system aneurysms can cause compression of the duodenum or bile ducts, which results in obstructive jaundice and direct hyperbilirubinemia, as seen in three of our patients [4].
The reported prevalence of portal vein thrombosis was 30% in the portal venous system aneurysm cases reported by Gallego et al. [4]. Patients with a thrombosed portal venous system aneurysm may be asymptomatic or may present with acute abdominal pain, nausea, vomiting, and fever [5, 27]. Recurrent thrombosis can cause portal vein occlusion that results in acute or chronic symptoms of portal hypertension [8, 20]. In our study, seven (28%) of the patients had portal vein thrombosis and were symptomatic, a rate similar to that reported in the literature [4]. Calcifications of the portal vein are very rare and are often associated with portal vein thrombosis and concomitant portal hypertension [28]. Therefore, the presence of portal vein calcifications is considered a sign of portal vein thrombosis and may influence the surgical treatment for portal hypertension and liver transplantation [28–30]. Calcifications of portal vein were identified in three of our patients with recurrent portal vein thrombosis.
Most portal venous system aneurysms require no treatment; follow-up is sufficient [5]. Anticoagulation therapy is recommended for patients with acute portal vein thrombosis because complete or partial recanalization is achieved in up to 80–90% patients [8, 19, 20]. Portal vein thrombosis may necessitate percutaneous intervention with thrombectomy or thrombolysis [19, 20]. Bypass surgery or aneurysmorrhaphy are therapeutic options that are rarely required when an aneurysm is enlarging or an interventional procedure is insufficient for recanalization of the occluded main portal vein [5, 6].
The first part of the study consists of non-consecutive patients, and this is one of the limitations of our study. However, we aimed to use that part of the study for radiologic findings and not for determining rates such as prevalence. The retrospective nature of the study is another limitation. Because portal venous system aneurysms are relatively rare, conducting prospective studies in large patient populations is difficult and our study comprises, to our knowledge, the largest patient group on this subject in the literature.
In conclusion, this study revealed that the prevalence of portal venous system aneurysms was 0.43% in patients who underwent abdominal CT examination, and this prevalence is higher than previously thought. All the thrombosed aneurysms, most with a recurrence, were associated with symptoms, and those without thrombosis had no symptoms or had symptoms related to documented diseases other than portal venous system aneurysms. There was no significant age- or sex-related difference in the frequency of portal venous system aneurysms. The diameters of the extrahepatic and intrahepatic portal venous system aneurysms were not different, and the symptomatic patients had significantly larger aneurysms than did the asymptomatic patients. A thrombophilic defect together with aneurysmal dilatation of the vein might be a trigger to portal venous system aneurysm thrombosis and the subsequent development of symptoms.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
