|
|
||||||||
Gastrointestinal Imaging |
1 From the Departments of Radiology and Nuclear Medicine (T.R., A.H., P.M., N.G.) and Internal Medicine (S.H.), Hospital Barmherzige Brueder, Kajetanerplatz 1, 5010 Salzburg, Austria. From the 1998 RSNA scientific assembly. Received January 21, 1998; revision requested April 2; final revision received November 11; accepted January 8, 1999. Address reprint requests to T.R.
| Abstract |
|---|
|
|
|---|
MATERIALS AND METHODS: In a partly retrospective study, abdominal sonograms were obtained in 11 patients with adult celiac disease. Fifty matched control subjects also were examined. After the patients had fasted overnight, they were examined with 24-MHz abdominal and high-frequency 510-MHz linear-array US transducers.
RESULTS: The authors found several US pathologic signs in patients with untreated disease, including abnormal fluid-filled small intestine in all 11 patients, flaccid and moderately dilated small-bowel loops (2.53.5 cm) in eight, slight diffuse thickening of the small-bowel wall (35 mm) in seven, increased peristalsis of the small intestine in eight, enlarged mesenteric lymph nodes (anteroposterior diameter 510 mm) in nine, dilated caliber of the superior mesenteric artery or portal vein in seven, free fluid in the abdominal cavity in five, and increased echogenicity of the liver in six. None of these signs was present in the control group.
CONCLUSION: There are several US signs associated with adult celiac disease. None of the signs identified is specific, but a combination of signs is characteristic and indicates suspicion of this disease in a high percentage of cases. US could help to avoid diagnostic delay, especially in patients who have atypical clinical presentations.
Index terms: Abdomen, US, 74.12989, 79.12989 Celiac disease, 74.7611 Intestines, abnormalities, 74.7611 Intestines, US, 74.12989
| Introduction |
|---|
|
|
|---|
Adult celiac disease occurs largely in white individuals. Prevalence ratios in Europe range from a high of 1:300 in western Ireland to between 1:1,000 and 1:2,000 in other regions. The true prevalence is certainly higher, because a substantial number of individuals have no or mild symptoms (3).
The symptoms vary tremendously from patient to patient and depend largely on the length and severity of damage to the small bowel (4,5). The classic clinical presentation includes diarrhea, steatorrhea, flatulence, weight loss, and fatigue. Laboratory abnormalities include iron deficiency or folate deficiency anemia; low serum levels of cholesterol, calcium, and albumin; elevated levels of alkaline phosphatase and liver enzymes; and a prolonged prothrombin time. In the past 10 years, the assessment of antigliadin and antiendomysial antibodies have gained in importance in confirming the clinical suspicion of adult celiac disease and in screening the families of patients with the disease (6).
The small-bowel enema examination has been widely advocated and used as a noninvasive diagnostic examination in patients suspected of having adult celiac disease (2,7). Other imaging modalities, such as computed tomography (CT) or ultrasonography (US), have not been established in confirming the diagnosis of adult celiac disease, but they have been described as useful in demonstrating complications such as malignant transformation or huge lymphadenopathy (2,8,9). CT and US, however, have a role in other more common diseases of the small intestine such as Crohn disease, small-bowel obstruction, or tumor (1012).
The essential step in the diagnosis is biopsy of the mucosa of the proximal small intestine, with results that show a lack or reduction of villi and crypt hyperplasia with increased mitoses (1). There also is an increased number of lymphocytes, eosinophils, and mast cells in the lamina propria and epithelium (1). Although the histologic appearance of the mucosa is characteristic, it is not specific. The definite diagnosis of adult celiac disease is therefore based on two criteria: the characteristic small-intestine biopsy specimen and clinical and histologic improvement after institution of a gluten-free diet (6). A gluten-free diet for the rest of life is the treatment of choice.
Patients suspected of having adult celiac disease can immediately undergo jejunal mucosal biopsy without any prior imaging investigation to confirm the diagnosis. However, in a substantial number of patients, the clinical symptoms and laboratory findings are atypical or subdued; therefore, adult celiac disease is not infrequently overlooked (35). Patients with an atypical clinical presentation (for instance, anemia or weight loss only) are often referred for routine US of the abdomen as the first imaging examination. There may exist distinct US signs or a combination of signs in patients with adult celiac disease that could arouse initial suspicion and assist in the diagnosis of the disease. Subsequent mucosal biopsy would then confirm the diagnosis without delay. This study was therefore performed to investigate the various US features that we observed in patients whose ultimate diagnosis was adult celiac disease to determine whether distinct US signs exist in patients with this disease.
| MATERIALS AND METHODS |
|---|
|
|
|---|
From 1987 to 1993, the diagnosis of adult celiac disease was established in six patients. In all of them, we noted a variety of US findings, which were evaluated retrospectively in this study. From 1994 to 1996, there were five patients with adult celiac disease, and in each case, we proposed this diagnosis on the basis of US findings alone before it was confirmed by using biopsy. The US pathologic signs in these five patients were evaluated prospectively.
Abdominal US examinations performed prospectively in 50 adult control subjects (29 women, 21 men) also were included in this study. These individuals were from an abdominal US screening program that has been offered to almost all of our inpatientseven those who do not have clinical abdominal symptomssince 1988 as common practice. Informed consent was obtained from each subject, and this practice has been approved by the ethical committee at our hospital. The only selection criteria were normal weight and no sign of abdominal disease. The control subjects fasted overnight for the general screening investigation but were not specially prepared. The body mass index of the control subjects was 1825 (mean index, 21) and thus somewhat higher than that of the underweight patients with adult celiac disease (body mass index, 1323; mean index, 18). US was performed in these subjects by using the model HDI 3000 unit with 24-MHz curved-array and 510-MHz linear-array transducers (Advanced Technology Laboratories) and the model AU4 unit with 3.5-MHz curved-array and 7.510.0-MHz linear-array transducers (Esaote).
The US signs that were investigated in the 11 patients with adult celiac disease and in the 50 control subjects are (a) fluid within the small bowel, (b) diameter of the small bowel, (c) thickness of the small-bowel wall, (d) peristalsis of the small intestine, (e) diameter of the mesenteric lymph nodes, (f) diameter of the superior mesenteric artery or portal vein, (g) free fluid in the abdominal cavity, and (h) echogenicity of the liver.
The following data with regard to normal findings refer to the results in the control group in our study and to data from the literature. In a healthy individual (ie, control subject) who has fasted overnight, US will demonstrate little or no fluid within some small-bowel loops. The echogenicity of the fluid in such subjects usually is echo free or echo poor. A marked amount of fluid in numerous small-bowel loops is considered to be abnormal.
The diameter of the small bowel (ie, the outer diameter including the bowel wall) in a healthy individual (ie, control subject) with an empty stomach can be up to 2 cm. Small-bowel loops with diameters between 2.5 and 3.5 cm are considered to be moderately dilated.
The wall thickness of the small intestine in a noncontracted condition is 12 mm in healthy individuals (ie, control subjects). Normal layers of small intestine (eg, mucosa, submucosa, and muscularis propria) can hardly be differentiated in a noncontracted condition but are well distinguishable in a contracted condition. The valvulae conniventes are clearly visible in fluid-filled loops only. A wall thickness of 3-5 mm is considered to be slightly thickened.
In patients with an empty stomach, peristalsis of the small intestine is slow and relatively rare. Sometimes moderate peristalsis can be observed, but this affects only single parts of the small bowel in control subjects. In our study, if the peristalsis was permanent and marked in the entire small intestine, it was considered to be abnormal. The degree of severity of this sign is, however, difficult to objectively determine.
Normal mesenteric lymph nodes could be demonstrated with US in 35 of the 50 control subjects. These nodes had an anteroposterior diameter of 25 mm (mean diameter, 3.5 mm), transverse diameter of 414 mm (mean, 7 mm), and longitudinal diameter of 624 mm (mean, 12 mm). The anteroposterior diameter is the most reliable parameter in differentiating between normal and enlarged nodes and was used as the criterion. Lymph nodes with an anteroposterior diameter greater than 5 mm were considered to be pathologic.
The width of the superior mesenteric artery was measured 23 cm distal to the origin but proximal to any side branch. Measurements of the portal vein were obtained at its widest part in the hepatoduodenal ligament. The diameter of the superior mesenteric artery in the healthy individuals was 47 mm (mean diameter, 5.3 mm) in women and 68 mm (mean, 6.5 mm) in men; the diameter of the portal vein was 712 mm (mean, 10.9 mm) in both women and men. Vessels with diameters greater than the maximum values were considered to be dilated.
In men, free fluid in the abdominal cavity was always considered to be abnormal. However, a small amount of free fluid was found in six of the 25 women of child-bearing age. This fluid is usually visible only around the uterus and not between bowel loops.
The echogenicity of the liver in individuals who are healthy and of normal weight is comparable to that of the normal kidney. Increased echogenicity of the liver, which is a sensitive indication of fatty degeneration, is uncommon in patients with a low body mass index and is considered to be pathologic. None of the control subjects in this study had liver tissue texture changes.
| RESULTS |
|---|
|
|
|---|
|
|
|
|
|
|
Diffuse thickening of the small-bowel wall and valvulae conniventes was noted at US in seven (64%) patients (Figs 1, 2). The thickening was slight (range, 35 mm; mean, 3.5 mm), and the layers of the wall were preserved.
Marked peristalsis of the small intestine was noted at US in eight (73%) patients. The peristalsis was permanent and not confined to a single part of the small bowel.
Numerous moderately enlarged mesenteric lymph nodes (anteroposterior diameter, 510 mm; mean, 7 mm) were evident at US in nine (82%) patients (Fig 3). The nodes were ovoid, relatively echogenic, and had a normal hilum (Fig 3). Their appearance was that of typical reactive enlarged nodes.
|
|
|
A diffuse increase in echogenicity of the liver was noted at US in six (55%) patients, although their body mass indexes were low (range, 1323; mean, 18). All patients who had steatosis had pathologic serum liver enzymes (viral infections were excluded by serologic examinations), and the patients who had almost normal or normal liver parameters had no increased liver echogenicity at US (Table 2).
|
| DISCUSSION |
|---|
|
|
|---|
To our knowledge, this is the first study in which several pathologic signs in patients with adult celiac disease detected by using US are described and in which it is proposed that a strong suspicion of adult celiac disease can be raised by using US alone in a high percentage of cases. In our hospital, US is now the first step in establishing this diagnosis. For example, the medical history of our most recent patient (patient 10), who presented with a 6-week history of diarrhea and weight loss of 4 kg, was uneventful, and all laboratory parameters were normal (Table 2). After US revealed the characteristic combination of pathologic signs (Table 1), the patient immediately underwent biopsy. Because the presentation of the patient was uncharacteristic, he might have left the hospital without a diagnosis of adult celiac disease if US findings had not supported the suspicion.
The fluid-filled small intestine in patients with adult celiac disease can be explained by impaired absorption and hypersecretion (20). US is very sensitive in the detection of echo-poor fluid within the bowel (10). This US finding has not been described in association with adult celiac disease previously, but indirect signs of this phenomenon such as segmentation, dilution, and flocculation of contrast agent at small-bowel enema studies are well-described radiographic diagnostic criteria for adult celiac disease (13,20,21). Increased fluid within the bowel, however, is not specific; it also occurs after ingestion, after resection or removal of the stomach, in diseases like enteritis, small-bowel obstruction, peritoneal carcinosis, and Wipple disease, and in other diseases with malabsorption (10).
In several radiographic studies (2022) with small-bowel enema examinations, moderate dilatation of flaccid small-bowel loops has been described as an important criterion for adult celiac disease, with a prevalence between 69% and 88%. In the present study, the radiographic method was expected to demonstrate this sign in a higher percentage of cases than was US, because we examined patients who had fasted overnight. When there is not enough fluid within the small bowel, dilatation cannot be demonstrated with US, whereas the small-bowel enema fills the intestine with 1.001.50 L of contrast agent and therefore can demonstrate dilatation of flaccid loops in a higher percentage of cases. This sign, however, also can be found in small-bowel obstruction, where the bowel loops appear to be stiff, as opposed to in adult celiac disease, where they appear to be flaccid. Dilated flaccid small-bowel loops and increased intraluminal fluid also are described with severe enteritis, jejunal diverticulosis, and scleroderma (2,23).
To our knowledge, moderate thickening of the small-bowel wall and increased peristalsis in patients who have adult celiac disease have not been described previously by using US, but they are well documented signs of adult celiac disease at small-bowel enema studies (13,14,2022). These findings can also be found in severe enteritis, which is often pronounced in the right lower region of the small intestine (ie, ileum) (24). Hyperperistalsis of the fluid-filled small intestine without wall thickening can be found after ingestion, in patients who have enteritis, and in some types of ileus (20).
The prevalence of mesenteric lymphadenopathy in patients with adult celiac disease was 12% in one study (25) but 0% in several others (26,27). In contrast, our observations indicate that lymph node enlargement exists in many cases of adult celiac disease. This is in accordance with one of the earlier reports (28) on adult celiac disease, in which all four patients who underwent laparotomy were found to have enlarged mesenteric nodes. Histologic evaluation of the nodes demonstrated chronic inflammatory changes and fibrosis (9,28). It can be assumed that mesenteric lymphadenopathy is present in many cases of adult celiac disease, but frequently it is not recognized because laparotomy is no longer necessary to obtain jejunal biopsy specimens (29). In the literature, there are also case reports on huge nonmalignant lymph node enlargement in adult celiac disease (8,9,29). It must be emphasized that lymph node enlargement can also be a sign of malignant lymphoma complicating adult celiac disease (8). Therefore, careful examination and follow-up with US or CT should confirm an improvement or disappearance of adenopathy after therapy.
Our observation of dilated mesenteric vessels in patients with adult celiac disease is in accordance with that in an angiographic study (30) in which this sign was described in a high percentage of cases. It is also known from the literature that adult celiac disease is associated with hemodynamic disturbances in the intestinal circulation, such as shortening of the blood transit time and increased blood flow volume (3032). It is assumed that the causes of these phenomena are raised levels of vasoactive hormones, the destruction of the capillary bed of the small-intestine mucosa, and the creation of arteriovenous shunts (3032). Increased blood flow in the mesenteric vessels has also been described in the literature as accompanying inflammatory bowel diseases such as severe Crohn disease and severe ulcerative colitis (33).
Free fluid in the abdominal cavity has not been previously described in association with adult celiac disease, but Riccabona and Rossipal (18,19) found this sign in infants with celiac disease in about 76% of cases. To our knowledge, no explanation for this phenomenon is available in the literature. One explanation could be the chronic inflammatory process in the small intestine and mesenteric lymph nodes in these patients. Free fluid is not a specific sign; it also occurs in patients with cirrhosis of the liver, peritoneal carcinosis, vascular and inflammatory processes of the abdomen, cardiac decompensation, and hypoproteinemia.
It is known from several studies (34,35) that abnormalities in liver enzymes are common with untreated adult celiac disease, but they usually respond to dietary gluten exclusion. The US images obtained in all of the patients with pathologic liver enzymes in our study showed increased hepatic echogenicity, which is a sensitive indication of fatty degeneration. The common causes of fatty degeneration of the liver include alcoholism, diabetes mellitus, hyperalimentation, toxic disturbance, therapy with glucocorticoids, protein deficiency, and intestinal malabsorption. The reasons for liver abnormalities in patients with adult celiac disease are still unknown, but it is assumed that they are a consequence of metabolic derangement from malabsorption (29).
A limitation of this study is that the patient group was small and the study was designed prospectively for only the last five patients. US technology improved during the study period, and in the first three patients, we did not use high-frequency transducers. Therefore, the prevalences of some of the described criteria might have been underestimated. The true prevalences must be determined prospectively with a larger patient population.
Each US sign described in this study is nonspecific and can occur in various other diseases. A combination of the signs, however, is characteristic and thus cause for suspicion of adult celiac disease in a high percentage of cases and enables the differential diagnosis of other more common diseases such as small-bowel obstruction or infection. Dilated fluid-filled bowel loops can also be found in small-bowel obstruction, where they appear stiff and round on transverse US sections and the colon is usually contracted. Infection leads to abnormal fluid collections of the bowel, increased peristalsis, and possibly to enlarged mesenteric lymph nodes, but there are no dilated mesenteric vessels and the diameter of the small-bowel loops is usually normal.
In conclusion, various US signs that enable the identification of adult celiac disease in a high percentage of cases are detectable in patients who have untreated adult celiac disease. US will not eliminate either of the two essential steps of diagnosisbiopsy and clinical improvement after institution of a gluten-free dietbut it could become a useful tool in shortening the diagnostic process, especially in patients with atypical clinical presentations. Because many patients who have adult celiac disease present with nonspecific symptoms, US may be the first imaging examination performed. The sonographer should, therefore, be familiar with the US appearance of this condition.
| Footnotes |
|---|
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
M. Fraquelli, A. Colli, A. Colucci, M. T. Bardella, C. Trovato, R. Pometta, M. Pagliarulo, and D. Conte Accuracy of Ultrasonography in Predicting Celiac Disease Arch Intern Med, January 26, 2004; 164(2): 169 - 174. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |