British Journal of Radiology 75 (2002),645-647 © 2002 The British Institute of Radiology
Superficial collateral veins on abdominal CT: findings in cirrhosis and systemic venous obstruction
A M Groves, BSc, MRCP and
A K Dixon, MD, FRCR
Department of Radiology, Addenbrooke's Hospital NHS Trust and the University of Cambridge, Hills Road, Cambridge CB2 2QQ, UK
Correspondence: Dr A M Groves, Department of Radiology, Box 218, Addenbrooke's NHS Trust, Hills Road, Cambridge CB2 2QQ, UK
 |
Abstract
|
|---|
Inspection of the superficial veins of the abdominal wall has long been a routine part of the physical examination. To date, radiologists have given such veins rather scant attention, even though they are elegantly demonstrated by CT. We have performed a study of 21 patients with cirrhosis, 7 patients with caval obstruction and 28 normal control counterparts in order to determine whether superficial veins were more numerous in these two clinical conditions. The 7 patients with caval obstruction included 4 with superior and 3 with inferior vena cava obstructions. Electronic data from the CT examinations of these 28 cases and 28 controls were analysed on a viewing console. Superficial veins were significantly more numerous in patients with cirrhosis (mean maximum=5, p<0.01) and caval obstruction (mean maximum=9.1, p<0.01) than in the normal controls (mean maximum=2.1). The combination of too many superficial veins and a large superior mesenteric vein is a pointer towards cirrhosis. The presence of excessive superficial veins is yet another clue to the presence of underlying disease when analysing abdominal CT.
|
Introduction
|
|---|
Clinicians attach great importance to examination of the superficial veins of the abdomen. The most basic texts are full of phrases such as "to assess the venous pressure, and reap a harvest of highly important clinical data, it is only necessary to retract the garment..." [1]. So distinctive are the clinical appearances of dilated abdominal veins that their appearance has been likened to the head of one of the three Greek mythological Gorgons whose hair was turned to snakes, the "Caput Medusa". Some clinicians take this further and manually milk the veins to observe the direction of in-filling, to help differentiate inferior vena cava obstruction from portal vein hypertension [2]. By comparison, radiologists tend not to pursue these signs with the same vigour.
On review of the CT literature, some investigators have studied collateral vessels in superior caval obstruction [3, 4], while others have concentrated on the importance of intraperitoneal and retroperitoneal collaterals in portal hypertension [5, 6]. There has, however, been a relative lack of recognition of the importance of superficial abdominal wall collateral vessels in intraabdominal disease. Stanley et al [7] briefly mentioned abdominal wall collaterals with portal hypertension, in the form of a CT image legend. Pagani et al [8] provided an image of abdominal wall collaterals, but in a patient with inferior vena caval obstruction, as well as discussing the anatomical pathways involved. Finally, in a series of 30 patients, Ishikawa et al [9] identified two patients with CT demonstrable signs of Caput Medusa.
Modern CT machines provide exquisite anatomical information of the skin and subcutaneous fat (Figure 1
). Even in a patient with relatively little subcutaneous fat, superficial collateral vessels can be identified. In turn these may drain into larger, deeper vessels (Figure 2).
View larger version (50K):
[in this window]
[in a new window]
|
Figure 1. CT appearance of the anterior abdominal wall in a normal patient at the level of the umbilicus. No superficial or deep vessels are identified.
|
|
View larger version (69K):
[in this window]
[in a new window]
|
Figure 2. CT appearance of the anterior abdominal wall showing multiple collateral superficial veins in a patient with inferior vena cava obstructions (at least 11 on this image). Note, also, one enlarged deep collateral (arrow).
|
|
For these reasons, we performed a matched controlled study examining CT appearances of collateral veins in the fat of the anterior abdominal wall in patients with cirrhosis and patients with various forms of caval obstruction.
|
Methods, materials and pateints
|
|---|
The CT cases examined over a 6-month period were systematically reviewed and 21 patients with documented cirrhosis and 7 patients with caval obstruction (4 superior and 3 inferior vena cava obstructions) identified. For each of these patients a matched control was obtained by selecting patients who had abdominal CT immediately before or after the study cases. The clinical history of the control patients was reviewed to ensure they did not have chronic liver disease or abnormalities that would render the assessment of the anterior abdominal wall difficult. In order to avoid inter-observer variation, soft copy data of every CT examination was reviewed by consensus of the authors at a stand alone workstation. At this consensus only one of the authors knew the patient's history, in an attempt to "blind" the other observer. However, CT features evident on the images often prevented a true blind reading.
The number of abdominal wall superficial and deep collateral vessels was counted on the axial image at the level of the umbilicus. A vessel was described as superficial if it lay superficial to the subcutaneous fascial plane (Figure 2
) [10]. The maximum number of superficial collaterals at any point on the abdominal wall was also counted. The superior mesenteric vein was identified and an image selected where this was largest (before it joins the portal vein). On this image the minimum diameter was measured with electronic callipers. The diameter of the superior mesenteric vein has long been regarded as a surrogate measure of the degree of portal hypertension [11]. The presence of ascites and sacral oedema were also recorded.
Results were analysed statistically using McNemar's test for categorical data and the paired t-test for numerical data.
|
Results
|
|---|
Results are summarized in Tables 1 and 2.
The maximal number of superficial collaterals on a CT image was significantly greater (p<0.02) in the cirrhotic cohort than the control cohort (Table 1). The difference was not so pronounced when collaterals were counted at the level of the umbilicus. In patients with caval obstruction (Table 2), the difference between the maximum number of collaterals compared with controls was also highly significant (p<0.01). In this group the number of superficial collaterals was again greatest at levels other than at the umbilicus. There were no significant differences in the number of deep collaterals seen within the three groups of patients.
The maximal diameter of the superior mesenteric vein was significantly greater (p<0.01) in the cirrhotic cohort (mean 12.9 mm) than in the controls cohort (mean 9.58 mm). In the cirrhotic cohort, ascites (p<0.01) and sacral oedema (p<0.02) occurred much more frequently than in the control cohort. In the smaller caval obstruction cohort, ascites was seen in two patients and sacral oedema in three, with neither evident in the controls.
|
Discussion
|
|---|
Our study shows that it is relatively straightforward to examine the superficial abdominal wall collaterals on CT and that there is a link between the number of these collaterals and the presence of underlying venous abnormality. It was interesting that more collateral veins were seen in patients with systemic venous obstruction than portal venous problems. Contrary to standard advice for clinical examination, the level of the umbilicus was not the best site to examine superficial collaterals by CT. It may be that clinicians choose this site as the veins are most superficial and hence most visible at this point. We also identified other useful signs. Since the superior mesenteric vein is a direct tributary of the portal vein it is recognized to enlarge if pressure is raised [11]. We confirmed enlargement of the superior mesenteric vein in cirrhotic patients. There was no increase in superior mesenteric vein diameter in patients with systemic venous obstruction. Thus, the finding of skin collaterals and a normal sized superior mesenteric vein would support a diagnosis of caval obstruction rather than cirrhosis.
Abdominal wall collateral veins are easily demonstrable by CT and provide a useful marker of intrabdominal venous problems. Consequently, radiologists should examine them with the same tenacity as their clinical colleagues.
|
Acknowledgments
|
|---|
The authors would like to acknowledge Prof D J Lomas for advice and Ms H Franklin for help with illustrations.
Received for publication November 6, 2001.
Revision received March 25, 2002.
Accepted for publication April 9, 2002.
|
References
|
|---|
- Bailey H. Basic physical signs. In: Clain A, editor. Hamilton Bailey's demonstrations of physical signs in clinical surgery (17th edn). Aylesbury, UK: Butterworth-Heineman Ltd, 1986:5–22.
- Olgivie C, Evans CC. The digestive system. In: Olgivie C, Evans CC, editors. Chamberlain's symptoms and signs in clinical medicine (11th edn). Bristol, UK: Wright, 1987:65–130.
- Kim H, Kim HS, Chung SH. CT Diagnosis of superior vena cava syndrome: importance of collateral vessels. AJR 1993;161:539–42.[Abstract/Free Full Text]
- Bashist B, Parisi A, Frager DH, Suster B. Abdominal CT findings when the superior vena cava, brachiocephalic vein or subclavian vein is obstructed. AJR 1996;167:1457–63.[Abstract/Free Full Text]
- Marn CS, Glazer GM, Williams DM, Francis IR. CT-angiographic correlation of collateral venous pathways in isolated splenic vein occlusion: new observations. Radiology 1990;175:375–80.[Abstract/Free Full Text]
- Balthazar EJ, Megibow A, Naidich D, LeFleur RS. Computed tomographic recognition of gastric varices. AJR 1984;142:1121–25.[Abstract/Free Full Text]
- Stanley RJ, Sagel SS, Levitt RG. Computed tomography of the liver. Radiol Clin N Am 1977;15:331–48.[Medline]
- Pagani JJ, Thomas JL, Bernardino ME. Computed tomographic manifestations of abdominal and pelvic venous collaterals. Radiology 1982;142:415–9.[Abstract/Free Full Text]
- Ishikawa T, Tsukune Y, Ohyama Y, Fujikawa M, Sakuuyama K, Fujii M. Venous abnormalities in portal hypertension demonstrated by CT. AJR 1979;134:271–6.
- Johnson D, Dixon AK, Abrahams PH. The abdominal subcutaneous tissue: computed tomographic, magnetic resonance and anatomical observations. Clin Anat 1996;9:19–24.[Medline]
- Rector WG Jr, Campra J, Ralls PW, Charms M. Utility and limitations of splanchnic venous ultrasonography in diagnosis of portal hypertension. J Clin Ultrasound 1986;14:689–96.[Medline]
Top
Abstract
Introduction
