HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chopra, S. Articles by Ghiatas, A. A. Search for Related Content PubMed PubMed Citation Articles by Chopra, S. Articles by Ghiatas, A. A.

Mesenteric, Omental, and Retroperitoneal Edema in Cirrhosis: Frequency and Spectrum of CT Findings¹

¹ From the Department of Radiology, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Dr, San Antonio, TX 78284-7800. Received August 7, 1998; revision requested September 3; revision received November 13; accepted December 9. Address reprint requests to S.C.

	Abstract

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 
PURPOSE: To determine the frequency and CT imaging spectrum of mesenteric, omental, and retroperitoneal edema in patients with cirrhosis.

MATERIALS AND METHODS: Eighty patients were identified with liver cirrhosis and no other cause of edema. Five radiologists jointly reviewed the abdominal CT scans of these patients to assess, by majority decision, the presence, severity, distribution, and configuration of mesenteric edema and the presence of omental and retroperitoneal edema. Subcutaneous edema, ascites, pleural effusion, splenomegaly, varices, portal venous thrombosis, and serum albumin levels were also documented.

RESULTS: Mesenteric edema was present in 69 (86%) patients. Mesenteric edema occurred alone in 26 (38%) and with omental or retroperitoneal edema in 40 (58%) of the 69 patients with edema. No patient had omental or retroperitoneal edema alone. Mesenteric edema was mild in 51 (74%) and moderate to severe in 18 (26%), patchy in 47 (68%) and diffuse in 22 (32%), purely infiltrative in 60 (87%) and infiltrative with superimposed masslike nodules in nine (13%) patients. These parameters had significant associations among themselves and with ascites, pleural effusions, subcutaneous edema, and low mean serum albumin levels but not with splenomegaly or varices.

CONCLUSION: Mesenteric, omental, and retroperitoneal edema occur commonly in patients with cirrhosis. The appearance of mesenteric edema varies from a mild infiltrative haze to a severe masslike sheath that engulfs the mesenteric vessels.

Index terms: Abdomen, abnormalities, 76.372, 76.75, 76.77, 76.794 • Abdomen, CT, 76.12111, 76.12115, 76.12119 • Liver, cirrhosis, 761.794 • Mesentery, 792.79 • Omentum, 79.79 • Retroperitoneal space, fluid, 79.79

	Introduction

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 
With the increasing routine use of computed tomographic (CT) scanning, mesenteric edema has been seen to occur frequently in cirrhosis. The CT findings of increased attenuation of mesenteric fat in patients with cirrhosis have been documented in the literature (1). However, to our knowledge, there is no mention in the literature of the various patterns of mesenteric edema in patients with cirrhosis as seen at CT. In our clinical practice, we have seen mesenteric and related edema in a variety of patterns of distribution and appearance in patients with cirrhosis. In some patients, it has been florid enough to induce false interpretations of matted lymphadenopathy or omental tumor. Prompted by these findings, we undertook a retrospective study to determine the frequency and spectrum of CT findings of mesenteric, omental, and retroperitoneal edema in patients with cirrhosis.

	MATERIALS AND METHODS

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 
From the pathology database of a large referral center (University of Texas Health Science Center at San Antonio), we identified all patients who had biopsy-proved liver cirrhosis between July 1994 and January 1997. We identified all patients from this group who had a CT scan of the abdomen. We excluded from this group patients who had any type of hepatic or extrahepatic malignancy identified at the time of CT or within 6 months by means of follow-up imaging and biopsy, by means of histopathologic examination of the cirrhotic liver resected at hepatic transplantation or other surgery, or at autopsy. We also excluded patients with any other clinically evident potential cause of mesenteric edema, like congestive cardiac failure, lymphoma, nephrotic syndrome, or intraabdominal inflammation.

We identified 80 patients (45 men, 35 women; age range, 22–70 years; mean, 47 years) who fit the entry criteria. Forty of the 80 patients underwent abdominal surgery within 6 months of the CT scan. Thirty-two of these 40 patients underwent liver transplantation, and eight had other abdominal surgery. As a part of the routine surgical practice, all patients underwent thorough palpation of the mesentery and omentum to rule out masses within these structures. Of the remaining 40 patients who did not undergo surgery, 34 had negative clinical follow-up findings, and six had repeat CT findings that were negative for malignancy at least 6 months after the original CT.

The CT scans of the patients selected for the study were reviewed. All scans were obtained on a HiSpeed Advantage (GE Medical Systems, Milwaukee, Wis) or Picker 2000 (Picker International, Cleveland, Ohio) scanner. Fifty-five of the 80 patients underwent dual-phase CT scanning of the liver, which consisted of precontrast, arterial phase, and portal venous phase helical series prior to, 25 seconds after, and 70 seconds after the initiation of an intravenous power injection of 150 mL of iothalamate meglumine (Conray 60; Mallinckrodt, St Louis, Mo) or ioversol 320 (Optiray 300; Mallinckrodt) at 5 mL/sec. Scanning parameters were 250–300 mA, 120 kVp, and 7- or 8-mm collimation with a pitch of 1 from the dome of the diaphragm to the iliac crest.

Twenty-three of the 80 patients underwent helical CT scanning only in the portal venous phase 70 seconds after initiation of a power injection of the contrast media mentioned earlier at a rate of 2–5 mL/sec with use of the technical parameters described earlier. In these patients, the helical scans were followed at 120 seconds by 5-mm-thick axial sections at 10-mm intervals to the pubic symphysis. The CT scans were obtained without the intravenous administration of contrast medium owing to the presence of high creatinine levels in two patients. In all patients, the helical sections were reconstructed at 7- or 8-mm intervals to obtain contiguous sections, depending on the section thickness. All images were displayed at the window width and window level of 400 HU and 40 HU, respectively. In the patients who underwent dual-phase scanning, the portal venous phase images were used for the purpose of this study.

The CT images of each patient were read jointly by five board-certified radiologists with consensus opinion reached for the presence of mesenteric, omental, or retroperitoneal edema. Mesenteric edema was diagnosed if the adipose tissue around the mesenteric vessels or their branches had a visually perceptible increased attenuation relative to that of subcutaneous or retroperitoneal fat. Omental or retroperitoneal edema was diagnosed if there was a visually perceptible increase in the attenuation of the adipose tissue in the appropriate location. In those patients who had edema at any of these sites, the following parameters of edema were evaluated: number of sites involved by edema, severity of edema, and, only in the case of mesenteric or omental edema, the distribution and configuration of edema.

According to the number of sites involved, the edema was divided into mesenteric, omental, or retroperitoneal involvement alone or combinations of more than one of these sites. The severity of edema was graded "mild" (Figs 1, 2) if there was a minimal wispy increase in the attenuation of mesenteric, omental, or retroperitoneal adipose tissue or "moderate to severe" (Fig 3) if the increase in attenuation was more than minimal. The distribution of mesenteric or omental edema was classified as "patchy" (Figs 1, 2) or "diffuse" (Fig 3), depending on the spatial continuity of involvement. The configuration of mesenteric or omental edema was characterized as "infiltrative" if the increase in attenuation of fat was amorphous (Figs 1–3) or "masslike" (Fig 4) if the increase in the attenuation of the adipose tissue was discrete and marginated.

View larger version (138K): [in this window] [in a new window]   Figure 1. CT scan of the upper abdomen of a 42-year-old man with cirrhosis shows mild, patchy, infiltrative mesenteric edema (long straight arrow). The regions of the omentum (short straight arrow) and the retroperitoneum (curved arrow) are normal.

View larger version (141K): [in this window] [in a new window]   Figure 2. CT scan at the level of the midabdomen of a 53-year-old woman with cirrhosis shows widespread patchy mesenteric edema (long straight arrow) and florid omental edema (short straight arrow). Although there is some thickening of the retroperitoneal fasciae, the adipose tissue (curved arrow) is of normal attenuation.

View larger version (187K): [in this window] [in a new window]   Figure 3. CT scan at the level of the midabdomen of a 48-year-old man with cirrhosis shows widespread severe, diffuse, infiltrative mesenteric edema (long straight arrow), omental edema (short straight arrow), and retroperitoneal edema (curved arrow).

View larger version (128K): [in this window] [in a new window]   Figure 4a. CT scans obtained in a 50-year-old woman with cirrhosis at (a) the level of the left renal hilus and (b) approximately 2 cm below a show severe, diffuse, well-marginated, masslike mesenteric edema (straight arrow). Note how the edema forms a sheath around the superior mesenteric vessels and could be mistaken for matted mesenteric lymphadenopathy. Splenomegaly (curved arrow) is also noted.

View larger version (120K): [in this window] [in a new window]   Figure 4b. CT scans obtained in a 50-year-old woman with cirrhosis at (a) the level of the left renal hilus and (b) approximately 2 cm below a show severe, diffuse, well-marginated, masslike mesenteric edema (straight arrow). Note how the edema forms a sheath around the superior mesenteric vessels and could be mistaken for matted mesenteric lymphadenopathy. Splenomegaly (curved arrow) is also noted.

The frequency of edema at various sites was calculated. The frequency of various site combinations, grades of severity, and types of distribution and configuration of edema were calculated and mutual associations among them were determined. The associations of the parameters of edema with the presence of subcutaneous edema, ascites, pleural effusions, splenomegaly, and varices were determined.

The statistical significance of all of these associations was tested by using the Fisher exact test. Mean serum albumin levels were calculated for patients with various site combinations and grades of severity and types of distribution and configuration of edema. Mean serum albumin levels for patients with or those without ascites, pleural effusions, subcutaneous edema, splenomegaly, or varices were also calculated. The differences among the mean serum albumin levels were tested for statistical significance by using the Student t test. A P value of .05 or less was considered to indicate a statistically significant difference. Because ascites, pleural effusions, and subcutaneous edema may be pathogenically related, any independent relationship between the parameters of edema and these findings was sought by using logistic regression with backward elimination. All statistical analysis was performed by a qualified statistician by using a statistical software package (SAS; SAS Institute, Cary, NC).

	RESULTS

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 
The small-bowel mesentery and the retroperitoneal spaces were visualized in all patients. The omentum was visualized in 77 (96%) patients. The remaining three patients had a paucity of intraabdominal adipose tissue too severe for reliable identification and assessment of the omentum.

Frequency and Sites of Edema
Edema in at least one site was seen in 69 (86%) of the 80 patients. Mesenteric edema was seen in all 69, and retroperitoneal edema was seen in 13 (16%) of the 80 patients. Omental edema was seen in 40 (52%) of the 77 patients with adequate visualization of the omentum. The combinations of the sites of edema are summarized in Table 1. Isolated mesenteric edema without omental or retroperitoneal edema was seen in 26 (38%) of the 69 patients with edema (Fig 1). All patients with omental (Fig 2) or retroperitoneal edema also had mesenteric edema. All patients with retroperitoneal edema also had combined mesenteric and omental edema (Fig 3), except two patients in whom the omentum could not be evaluated.

Association with Other Findings
The frequency of other CT findings related to cirrhosis is shown in Table 3. The association of the parameters of severity of mesenteric edema with these findings is shown with the P values in Table 4. The presence of ascites had a statistically significant association with the presence of edema at additional sites, moderate to severe edema in the mesentery, and diffuse distribution of mesenteric edema. No significant association was seen between the presence of ascites and masslike configuration of mesenteric edema. Furthermore, of 39 patients with ascites, only one had no mesenteric edema, while of 41 patients without ascites, 10 had no mesenteric edema (P < .001). Pleural effusion and subcutaneous edema showed statistically significant association with additional site involvement and diffuse distribution of mesenteric edema.

Mean serum albumin level findings were available in 43 patients. The association of the parameters of severity of edema with mean serum albumin levels is also shown in Table 4. Mean serum albumin levels were significantly lower for mesenteric edema with additional site involvement as compared with isolated mesenteric edema, and they were significantly lower for moderate to severe mesenteric edema as compared with mild mesenteric edema. The mean serum albumin levels were also lower for diffuse as compared with patchy distribution, and they were lower for masslike as compared with infiltrative edema. However, these differences did not reach low enough P values to be deemed statistically significant. Mean serum albumin levels were significantly lower in patients with ascites as compared with those without ascites (P < .005), pleural effusion (P < .02), and subcutaneous edema (P = .001). No association was observed between mean serum albumin levels and splenomegaly or varices.

On review of the medical records of the patients, 40 (50%) of the 80 patients were found to have undergone a laparotomy within 6 months (range, 10–170 days) after the original CT scan. Thirty-two of these were for liver transplantation, and eight were for other indications. The CT scans of these patients included in the study showed edema in 33 patients. Fifteen of these had isolated mesenteric edema; 13 had mesenteric and omental edema; and five had mesenteric, omental, and retroperitoneal edema. During laparotomy, a thorough palpation of the peritoneal reflections had been performed and failed to show any evidence of masses or inflammation within the mesentery or the omentum.

	DISCUSSION

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 
Edema in patients with cirrhosis has a multifactorial pathogenesis (3). The most obvious pathogenic mechanism is portal hypertension causing increased hydrostatic pressure within the mesenteric veins and causing fluid to seep out into the mesentery. The other manifestations of portal hypertension are varices, splenomegaly, and ascites. In addition, liver dysfunction produces a state of water overload due to hypoalbuminemia and decreased aldosterone catabolism. This also plays a role in producing mesenteric edema, ascites, pleural effusions, and subcutaneous edema. The CT appearance of the normal mesentery and the diffuse mesenteric edema due to different causes was described by Silverman et al (4,5). Mindelzun et al (6,7) also described the appearance of diffuse mesenteric edema and named its appearance at CT "misty mesentery." Tyrrel et al (1) quantified the CT attenuation levels of the mesenteric, retroperitoneal, and subcutaneous fat in patients with cirrhosis and showed them all to be statistically significantly higher than in control subjects without cirrhosis. The visualization of mesenteric edema in cirrhosis at CT has been well established by these studies. However, in practice, when edema of the mesentery or the retroperitoneum is identified on the abdominal CT scan of a patient with cirrhosis, the question arises of an additional abdominal condition such as inflammation, hemorrhage, or neoplastic infiltration as the cause of the observed edema. Although the clinical features may suggest the underlying cause, the correct analysis of the CT findings depends on the knowledge of the spectrum of the various appearances of edema in cirrhosis. To our knowledge, there are no publications that have described the spectrum of CT imaging appearances of mesenteric and related edema in cirrhosis in detail. We undertook this study with the aim of doing so.

In our study, mesenteric edema was seen at CT in 86% of patients with cirrhosis. It occurred more commonly in combination with omental or retroperitoneal edema (62%) than it did alone (38%). Neither omental nor retroperitoneal edema occurred in the absence of mesenteric edema, and retroperitoneal edema did not occur in the absence of omental edema. There was thus a progression from mesenteric edema alone through a combination of mesenteric and omental edema to a combination of mesenteric, omental, and retroperitoneal edema.

Although mesenteric edema was more often mild, focal, and infiltrative, with increasing severity of edema it tended to become more diffuse and develop a masslike appearance. The increasing severity, diffuse distribution, masslike appearance, and recruitment of omental and retroperitoneal sites are the parameters of severity of mesenteric edema and in general correlate with other findings of severe ascites, subcutaneous edema, pleural effusions, and low mean serum albumin levels. Masslike appearance of edema was seen in the omentum in only one patient. To our knowledge, the spectrum of sites of edema with the mesentery as the primary site, the various parameters of severity, and the masslike appearance of edema in some patients have not been published before.

Portal hypertension or low serum albumin level has been described as the major pathogenic mechanism in the production of intraabdominal edema and ascites by various authors at different times (3,8). However, in our study, we found no significant association between the parameters of severity of edema and the manifestations of portal hypertension (ie, splenomegaly and varices). On the other hand, in general there was a significant association between the parameters of severity of mesenteric edema and the manifestations of water overload (ie, subcutaneous edema, pleural effusion, and ascites) and low mean serum albumin levels. In addition, subcutaneous edema, which is unaffected by portal hypertension, showed an independent association with mesenteric edema. This demonstrates that the volume-expanded state plays a much greater role than portal hypertension and is the predominant mechanism in the pathogenesis of mesenteric edema in cirrhosis.

Our observations have some points of clinical relevance. When abdominal CT scans of patients with cirrhosis are interpreted, mesenteric edema should be expected to be seen commonly. However, the presence of omental or retroperitoneal edema should not be ascribed to cirrhosis alone. Other conditions such as inflammation, hemorrhage, or malignant infiltration should be considered. Similarly, the presence of severe ascites in a patient with cirrhosis without mesenteric edema is so uncommon that in this situation additional causes of ascites, such as hepatocellular carcinoma, portal or hepatic venous occlusion, or spontaneous bacterial peritonitis, should be considered and sought out. A nodular pattern of edema may occur in the mesentery and rarely in the omentum. Its presence should not be mistaken for other causes of nodular abnormalities such as benign or malignant lymphadenopathy (9) or tumor nodules in these areas. In such cases, careful correlation with the severity of cirrhosis and other findings is necessary.

This study has some limitations. First, we selected our patients from a cohort who had biopsy-proved cirrhosis. In clinical practice, patients with cirrhosis do not undergo a random liver biopsy unless there is a specific clinical question to be answered. Therefore, our population probably does not represent the usual spectrum of patients with cirrhosis. On the other hand, we strongly thought that in our study the diagnosis of cirrhosis must be beyond any doubt so that patients with other forms of chronic liver disease were not included and patients with cirrhosis but without florid morphologic manifestations were not excluded.

Second, we used a subjective technique to grade the severity of edema. We opted not to measure the attenuation values of the areas of edema because in clinical practice the visually discernible increase in the attenuation of fat is used for diagnosing and grading edema. To increase the interobserver reliability, we simplified the categories of severity. Mild edema was defined as a minimal wispy increase in the attenuation of adipose tissue. Any increase in attenuation that was more than mild was classified as moderate to severe. In our opinion, this mimics the clinical environment more closely than measurements of CT attenuation values and is therefore more meaningful.

Third, this study is limited by the lack of an absolute standard, in some patients, of proof of edema and exclusion of concomitant disease that could produce some of the assessed findings. We performed our study on the assumption that the increased attenuation detected in the adipose tissue was caused by edema produced by the cirrhotic process. However, we recognize that tumoral infiltration or edema of other causes could have caused the attenuation change. To minimize this potential error, we excluded all patients with overt evidence of malignancy and required a laparotomy with negative findings or a stable repeat CT examination at 6 months as entry criteria into our study. Thirty-three patients with edema underwent subsequent laparotomy with no evidence of any other pathologic process. In addition, we excluded all patients with other types of systemic disease that could cause mesenteric, omental, or retroperitoneal edema. Therefore, the CT findings read as edema were probably due to the presence of edema in the mesentery, omentum, and retroperitoneum.

In conclusion, with this study we have described the spectrum of mesenteric, omental, and retroperitoneal edema in patients with cirrhosis, their interrelationship, and associations with other findings at abdominal CT. An awareness of these findings should allow more accurate interpretations of the abdominal CT findings of patients with cirrhosis.

	Acknowledgments

 
The authors thank John Schoolfield, MS, for statistical analysis and Baltazar Farias for photography.

	Footnotes

 
Author contributions: Guarantors of integrity of entire study, C.C.E., S.C.; study concepts, G.D.D.; study design, G.D.D., S.C., C.C.E., K.N.C., A.A.G.; definition of intellectual content, G.D.D.; literature research, S.C.; data acquisition, S.C., K.N.C., A.A.G., C.C.E.; data analysis, S.C.; manuscript preparation, S.C., G.D.D.; manuscript editing, S.C.; manuscript review, K.N.C., S.C., G.D.D.

	References

TOP Abstract Introduction MATERIALS AND METHODS RESULTS DISCUSSION References

 

1. Tyrrel RT, Montemayor KA, Bernardino ME. CT density of mesenteric, retroperitoneal and subcutaneous fat in cirrhotic patients: comparison with control subjects. AJR 1990; 155:73-75.[Abstract/Free Full Text]
2. Federle MP. Computed tomography of the spleen. In: Moss AA, Gamsu G, Genant HK, eds. Computed tomography of the body. Philadelphia, Pa: Saunders, 1983; 877-906.
3. Gines P, Fernandez-Esparrach G, Arroyo V, Rodes J. Pathogenesis of ascites in cirrhosis. Semin Liver Dis 1997; 17:175-189.[Medline]
4. Silverman PM, Kelvin FM, Korobkin M, Dunnick NR. Computed tomography of normal mesentery. AJR 1984; 143:953-957.[Abstract/Free Full Text]
5. Silverman PM, Baker ME, Cooper C, Kelvin FM. CT appearance of diffuse mesenteric edema. J Comput Assist Tomogr 1986; 10:67-70.[Medline]
6. Mindelzun RE. The abdominal mesenteries: anatomy and diseases. Contemp Diagn Radiol 1995; 18:1-6.
7. Mindelzun RE, Jeffery RB, Lane MJ, Silverman PM. The misty mesentery on CT: differential diagnosis. AJR 1996; 167:61-65.[Free Full Text]
8. Diner WC. Small intestinal edema in cirrhosis: its disappearance with diuresis. Radiology 1968; 91:792-794.[Medline]
9. Dodd GD, III, Baron RL, Oliver JH, Federle MP, Baumgartel PB. Enlarged abdominal lymph nodes in end-stage cirrhosis: CT-histopathologic correlation in 507 patients. Radiology 1997; 203:127-130.[Abstract/Free Full Text]

This article has been cited by other articles:

				E. L. Ormsby, C. Duffield, F. Ostovar-Sirjani, J. P. McGahan, and C. Troppmann Colonoscopy Findings in End-Stage Liver Disease Patients with Incidental CT Colonic Wall Thickening Am. J. Roentgenol., November 1, 2007; 189(5): 1112 - 1117. [Abstract] [Full Text] [PDF]

				E. Yoo, J. H. Kim, M.-J. Kim, J.-S. Yu, J.-J. Chung, H.-S. Yoo, and K. W. Kim Greater and Lesser Omenta: Normal Anatomy and Pathologic Processes RadioGraphics, May 1, 2007; 27(3): 707 - 720. [Abstract] [Full Text] [PDF]

				B. K. Seo, H. K. Ha, A. Y. Kim, T. K. Kim, M. J. Kim, J. H. Byun, P. N. Kim, M.-G. Lee, S.-K. Yang, E.-S. Yu, et al. Segmental Misty Mesentery: Analysis of CT Features and Primary Causes Radiology, January 1, 2003; 226(1): 86 - 94. [Abstract] [Full Text] [PDF]

				G Ege, H Akman, and G Cakiroglu Mesenteric panniculitis associated with abdominal tuberculous lymphadenitis: a case report and review of the literature Br. J. Radiol., April 1, 2002; 75(892): 378 - 380. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Chopra, S. Articles by Ghiatas, A. A. Search for Related Content PubMed PubMed Citation Articles by Chopra, S. Articles by Ghiatas, A. A.