- Email: [email protected]
Incidental findings of thickening luminal gastrointestinal organs on computed tomography: an absolute indication for endoscopy
THE AMERICAN JOURNAL OF GASTROENTEROLOGY © 2003 by Am. Coll. of Gastroenterology Published by Elsevier Inc.
Vol. 98, No. 8, 2003 ISSN 0002-9270/03/$30.00 doi:10.1016/S0002-9270(03)00495-7
Incidental Findings of Thickening Luminal Gastrointestinal Organs on Computed Tomography: An Absolute Indication for Endoscopy Qiang Cai, M.D., Ph.D., Deborah A. Baumgarten, M.D., M.P.H., John P. Affronti, M.S., M.D., and J. Patrick Waring, M.D. Division of Digestive Diseases and Department of Radiology, Emory University School of Medicine, Atlanta, Georgia
OBJECTIVE: Incidental findings of thickened luminal GI organs (LGIO) on CT are not uncommon. However, the significance of these findings is unclear. Because of the lack of scientific data, there are no clinical guidelines for the evaluation of these radiologic abnormalities. Our objective was to determine whether endoscopic evaluation of these findings revealed significant abnormalities. METHODS: This study evaluated all incidental findings of thickened LGIO in a large medical center from October, 1997 to March, 1999 that were followed by endoscopic examinations. RESULTS: Ninety-six percent of patients with incidental findings of thickening of the sigmoid colon or rectum, 81% of patients with thickening of the distal esophagus, and 13% of patients with thickening of the cecum had significant abnormalities on further endoscopic work up. CONCLUSIONS: Although positive pathologic findings are less common in thickening of the cecum than in other LGIO, all of these incidental findings on CT warrant further endoscopic examination. (Am J Gastroenterol 2003;98: 1734 –1737. © 2003 by Am. Coll. of Gastroenterology)
INTRODUCTION Initial clinical use of CT of the luminal GI tract focused on its ability to recognize a thickened bowel wall (1). CT has now become an established diagnostic procedure in a variety of common GI disorders and is an important tool in clinical practice. The first findings of pathologic change in the GI tract recognized on CT were those in which visible wall thickening could be identified. Initial reports claimed excellent sensitivity in identifying an abnormally thickened intestinal wall but commented on the inability to distinguish inflammatory from neoplastic processes (2).
Part of this manuscript was presented October, 2000 at the American College of Gastroenterology annual conference, New York, NY, and May, 2001 at Digestive Diseases Week, Atlanta, GA.
The indication for CT depends on the clinical condition of the patient. If the history suggests a mucosal process, barium studies or endoscopic examination is an appropriate first choice. Endoscopic ultrasound (EUS) is a relatively new technology in clinical gastroenterology (3). EUS plays an important role in many clinical areas, such as in the evaluation of mediastinal lymph nodes or masses in patients with suspected lung cancer and in the evaluation of pancreatic malignancy (3). In recent years, EUS has been used more and more often for mucosal diseases in the GI system, especially to define the depth of invasion of a mucosal lesion, such as an esophageal cancer (4). However, when the extraluminal extension of disease might be more critical than the changes along the mucosal surface, CT still assumes a frontline role (1). Often in clinical practice, thickening of the luminal GI organs (LGIO), such as the esophagus, stomach, and small and large intestine, are incidentally reported on abdominal, pelvic, or chest CT scan. These patients do not have any known GI diseases, and they have been sent to CT for non–GI-related diseases, or they have a known GI disease, but the CT finding of a thickened LGIO is not directly related to their disease. An incidental finding of thickened LGIO is a common clinical occurrence; however, the significance of such findings is unclear. Primary care physicians often refer patients with incidental findings of thickened LGIO on CT to gastroenterologists for further evaluation. Because of the sparse scientific data, there are no clinical guidelines for the evaluation of incidental findings of thickened LGIO on CT. In the present study, the significance of a thickened esophagus, cecum, sigmoid colon, or rectum was investigated to determine whether these findings on CT warrant further investigation.
MATERIALS AND METHODS The report logbooks for abdominal CTs performed from October, 1997 through March, 1999 in the Emory University School of Medicine, Department of Radiology were checked for patients with thickening of the distal esophagus, stomach, small intestine, cecum, sigmoid, or rectum. The
AJG – August, 2003
Thickening Luminal Gastrointestinal Organs
1735
Table 1. General Patient Information
Total patients with thickening LGIO No. patients who had endoscopy Gender (M/F) Mean age (range) (M/F) (yr) No. patients with positive findings
Distal Esophagus
Cecum
Sigmoid Colon or Rectum
48 27 (56%) 9/18 47 (35–39)/52 (27–86) 22 (81%)
18 16 (88%) 6/10 69 (58–76)/57 (40–82) 2 (13%)
51 24 (47%) 18/6 55 (36–72)/54 (42–69) 23 (96%)
M ⫽ male; F ⫽ female.
final reports were verified with the electronic record (Powerchart) for the health care system. During this period, a total of 12,021 abdominal CTs were performed. The two major criteria for enrolling patients for this study were 1) patients had no known GI disease, such as esophagitis, diverticulitis, colitis, or colon cancer before the CT scan, and 2) patients had a known GI disease, but the CT finding of a thickened LGIO was not directly related to the disease. For example, a patient with cirrhosis had CT to rule out hepatoma, and CT showed a thickened cecum. In other words, only cases with incidental findings of thickened LGIO were included in this study. Subsequent endoscopic examinations for those cases were reviewed.
RESULTS During the study period, 117 cases were identified as having incidental findings of thickened distal esophagus, cecum, sigmoid colon, or rectum based on the above criteria. Among the 117 cases, 48 patients had thickened distal esophagus, 18 patients had thickened cecum, and 51 patients had thickened sigmoid colon or rectum. Most of the patients were middle-aged (Table 1). Of the 117 patients, 67 had documented further endoscopic examination at our university hospital, including 27 patients with thickened distal esophagus, 16 patients with thickened cecum, and 24 patients with thickened sigmoid colon or rectum (Table 1). CT indications for these 67 patients are listed in Table 2. Of these 67 patients, 3 had no oral or intravenous (IV) contrast; 2 of the 67 patients had oral contrast only. The rest of the patients had both oral and IV contrast. For patients who had IV contrast, 150 ml of Omnipaque (Amersham Health, Princeton, NJ) was given (administration rate was not documented in the report). Detailed information about oral contrast administration was not documented in the report. All 67 patients had helical CT, and images were obtained at 5–7-mm increments. Among the 67 patients who underwent further evaluation by endoscopic examination, 22 of 27 patients with a thickened distal esophagus (81%) and 23 of 24 patients with a thickened sigmoid colon or rectum (96%) had abnormal findings on further evaluation. Esophagitis (accounting for 30% of the thickened distal esophagus) and colitis (accounting for 75% of the thickened sigmoid colon or rectum) were the most common findings (Table 3). Of the patients with a thickened cecum, 14 of 16 (88%) had normal findings on further evaluation (Table 3).
DISCUSSION Incidental findings of thickened LGIO on CT are not uncommon in clinical practice and might be a clue to significant pathology. To our knowledge, the present study is the first systematic investigation of incidental findings of thickened LGIO on CT. Our findings indicate that thickened LGIO on CT warrants further evaluation by endoscopic examination. CT diagnosis of GI diseases has become more common in clinical practice (5–9). CT is an accurate technique in the evaluation of intestinal intussusception, GI masses, GI bezoars, and ischemia, as well as other disorders (5–9). Thickened LGIO is commonly reported by radiologists but is a very nonspecific CT finding. It could be a real finding related to pathology or an artifact caused by poor luminal distension of the LGIO or thick tenacious material adherent to the wall of the LGIO. In some situations, radiologists cannot make a specific diagnosis based on the CT image, especially when the finding is incidental and patients do not
Table 2. Indications for CT LGIO
No. patients
Indications (n)
Distal esophagus
27
Cecum
16
Sigmoid colon or rectum
24
Chronic abdominal pain (7) Chronic abdominal pain and weight loss (5) Fever, R/O pneumonia (5) F/U lung cancer (4) Hemoptysis (3) Severe chest pain (3) Cirrhosis R/O hepatoma (5) Flank pain, hematuria (4) Chronic hepatitis, R/O cirrhosis (4) F/U of pancreatitis (2) Grossly hematuria (1) Cirrhosis, R/O hepatoma. (5) S/P BMT, fever (5) F/U ovarial cancer (3) F/U renal cancer (2) F/U chronic pancreatitis (2) Hematuria (2) Fever of unknown origin (2) Chronic hepatitis, R/O cirrhosis (2) Flank pain, H/O kidney stone (1)
BMT ⫽ bone marrow transplantation; R/O ⫽ rule out; F/U ⫽ follow up; S/P ⫽ status post; H/O ⫽ history of.
1736
Cai et al.
AJG – Vol. 98, No. 8, 2003
Table 3. Positive Findings on Endoscopic Examination
LGIO
No. patients who had endoscopy
Distal esophagus
27
Cecum
16
Sigmoid colon or rectum
24
Findings
No. patients
Esophagitis Hiatal hernia Esophageal varices Normal Polyp Cancer Normal Colitis Cancer Polyp Normal
8 (30%) 7 (26%) 7 (26%) 5 (18%) 1 (6%) 1 (6%) 14 (88%) 18 (75%) 3 (13%) 2 (8%) 1 (4%)
have related GI diseases or symptoms. The observation of thickened LGIO on CT is not very difficult. The finding is based mainly on the thickness of the wall of the LGIO. Usually the wall of LGIO should be 2–3 mm (1, 10), except the rectum, which can be up to 8 mm. If the wall is thicker than 5 mm (greater than 8 mm in the rectum), the report of a thickened LGIO is made. One interesting finding in the present study is that most of the patients with thickened distal esophagus and thickened sigmoid colon or rectum had positive pathologic findings on further endoscopic workup. On the other hand, most patients with a thickened cecum had normal findings on further endoscopic workup. The reason for this difference is not absolutely clear, but at least two reasons are possible. First, if an LGIO is not distended well at the time when CT is performed, the LGIO might seem thicker than it should be. The cecum is a relatively mobile luminal organ compared with the esophagus, sigmoid colon, and rectum. The latter are relatively fixed organs and do not have much room to collapse. In other words, the esophagus, sigmoid colon, and rectum are less prone to collapse and therefore less likely to be affected by this artifact. Second, there are diseases, such as different varieties of colitis, which primarily affect the sigmoid colon and rectum. Diseases located only in the cecum are less common. Although most patients with a thickened cecum on CT had normal findings on further endoscopic examinations, there were still more than 10% who had significant abnormalities on further workup. The most common positive finding in patients with a thickened distal esophagus was esophagitis, followed by hiatal hernia and esophageal varices. There were no esophageal cancers identified in this study. We did not further subcategorize the esophagitis based on the etiology. Most of these patients had nonspecific esophagitis, probably due to gastroesophageal reflux. The seven patients with esophageal varices had no known liver cirrhosis before the examination. Another interesting finding of this study was that most of the patients with a thickened cecum had further documented evaluations. Of 18 patients with a thickened cecum, 16 (88%) had colonoscopy. Only two did not have further evaluation. In contrast, 27 of 48 patients (56%) with a
thickened distal esophagus and 24 of 51 patients (47%) with a thickened sigmoid colon or rectum had further evaluation. This might reflect the concern of the primary care physician about a thickened cecum. Because colon cancer is a relatively common disease in the United States, a thickened cecum might be perceived as more worrisome than a thickened distal esophagus or a thickened sigmoid colon and rectum, both to the patient and their physician. Approximately 50% of the patients with a thickened distal esophagus or thickened sigmoid colon or rectum did not have documented further endoscopic examination at this medical center. We are not sure whether those patients had further workup in another hospital, because we did not follow-up with these patients. Based on the endoscopic examination of our patients, the most significant findings were seven patients with esophageal varices and four patients with colon cancer. All seven patients with esophageal varices had further workup of their liver disease. Five had hepatitis C cirrhosis, and two had alcoholic cirrhosis. The four patients with colon cancer had successful surgical therapy. If there had been no further endoscopic examination for these patients, we might have missed this significant disease. It would be very interesting to know the outcomes of patients who did not have further endoscopic examination at our hospital. However, it is very difficult to obtain this information, because some patients did not come back to our hospital for continuation of medical care. There were 45 patients with diffuse small bowel thickening and 34 patients with gastric wall or antral thickening reported during this study period. However, none of these patients met the criteria of an incidental finding of thickening of LGIO. Almost all patients with small intestine thickening on CT were cirrhotic patients with low serum albumin, and none of them had further endoscopic examination. All patients with gastric wall or antral thickening were sent for CT for abdominal pain or abdominal pain and weight loss. Because those patients did not meet the criteria for incidental findings of thickening LGIO, they were not included in this study. However, some of these patients had documented endoscopic examination, and most were given a diagnosis of gastritis. Limitations of this study include patient selection and verification bias. Because patient selection was based on reports in the logbook in the radiology department, there might be omission of cases having incidental thickening of LGIO. We did not examine all of the final reports for the 12,021 patients who had a CT during this study period. Furthermore, patients who did not have endoscopic workup in our hospital but who may have had endoscopic examinations in other hospitals were also excluded from the study. Another limitation of this retrospective study is the inability to precisely quantify patients’ symptoms. All information about the patients was collected by reviewing patient charts and CT reports.
AJG – August, 2003
In conclusion, of those who underwent further workup in our hospital, most of the patients with a thickened distal esophagus and a thickened sigmoid colon or rectum on CT had an abnormality. Therefore, a thickened distal esophagus, or a thickened sigmoid colon or rectum on CT should be investigated. Of those who underwent further workup, most of the patients with an isolated thickening of the cecum on CT had normal examinations. Although the proportion of positive endoscopic findings in patients with a thickened cecum is relatively lower than that in patients with a thickened distal esophagus or thickened sigmoid colon or rectum, the finding of a thickened cecum on CT still needs further investigation because of the more than 10% possibility of localized cecal mass or polyp.
Thickening Luminal Gastrointestinal Organs
2. 3. 4. 5. 6. 7.
Reprint requests and correspondence: Qiang Cai, M.D., Emory University School of Medicine, Division of Digestive Diseases, 201 Whitehead Building, 615 Michael Street, Atlanta, GA 30322. Received Sep. 30, 2002; accepted Jan. 7, 2003.
REFERENCES 1. Megibow AJ. The gastrointestinal tract. In: Lanzieri CF, Sartors DJ, Zerhouni EA, et al., eds. Computed tomography and
8. 9. 10.
1737
magnetic resonance image of the whole body, Third edition. St. Louis: Mosby, 1994:855–94. Coscina WF, Arger PH, Levine MS, et al. Gastrointestinal tract focal mass lesions: Role of CT and barium evaluations. Radiology 1986;159:581–7. Dancygier H, Lightdale CJ. Endosonography in gastroenterology: Principles, techniques, findings. New York: Thieme, 1999. Tio TL. Diagnosis and staging of esophageal carcinoma by endoscopic ultrasonography. Endoscopy 1998;30(suppl 1): A33–40. Shiraga N. Two-dimensional and three-dimensional CT diagnosis of alimentary tract. Nippon Igaku Hoshasen Gakkai Zasshi 2001;61:672–82. Barbiera F, Cusma S, Di Giacomo D, et al. Adult intestinal intussusception: Surgery-CT correlations. Radiol Med (Torino) 2001;102:37–42. Rousso I, Hadar H, Levy A, et al. Primary CT diagnosis of gastrointestinal masses. Comput Radiol 1985;9:259 –63. Ripolles T, Garcia-Aguayo J, Martinez MJ, et al. Gastrointetinal bezoars: Sonographic and CT characteristics. AJR Am J Roentgenol 2002;177:65–9. Zaleman M, Sy M, Donckier V, et al. Helical CT signs in the diagnosis of intestinal ischemia in small-bowel obstruction. AJR Am J Roentgenol 2000;175:1601–7. Hammerman AM, Mirowitz SA, Susman N. The gastric airfluid sign: Aid in CT assessment of gastric wall thickening. Gastrointest Radiol 1989;14:109 –12.
Vol. 98, No. 8, 2003 ISSN 0002-9270/03/$30.00 doi:10.1016/S0002-9270(03)00495-7
Incidental Findings of Thickening Luminal Gastrointestinal Organs on Computed Tomography: An Absolute Indication for Endoscopy Qiang Cai, M.D., Ph.D., Deborah A. Baumgarten, M.D., M.P.H., John P. Affronti, M.S., M.D., and J. Patrick Waring, M.D. Division of Digestive Diseases and Department of Radiology, Emory University School of Medicine, Atlanta, Georgia
OBJECTIVE: Incidental findings of thickened luminal GI organs (LGIO) on CT are not uncommon. However, the significance of these findings is unclear. Because of the lack of scientific data, there are no clinical guidelines for the evaluation of these radiologic abnormalities. Our objective was to determine whether endoscopic evaluation of these findings revealed significant abnormalities. METHODS: This study evaluated all incidental findings of thickened LGIO in a large medical center from October, 1997 to March, 1999 that were followed by endoscopic examinations. RESULTS: Ninety-six percent of patients with incidental findings of thickening of the sigmoid colon or rectum, 81% of patients with thickening of the distal esophagus, and 13% of patients with thickening of the cecum had significant abnormalities on further endoscopic work up. CONCLUSIONS: Although positive pathologic findings are less common in thickening of the cecum than in other LGIO, all of these incidental findings on CT warrant further endoscopic examination. (Am J Gastroenterol 2003;98: 1734 –1737. © 2003 by Am. Coll. of Gastroenterology)
INTRODUCTION Initial clinical use of CT of the luminal GI tract focused on its ability to recognize a thickened bowel wall (1). CT has now become an established diagnostic procedure in a variety of common GI disorders and is an important tool in clinical practice. The first findings of pathologic change in the GI tract recognized on CT were those in which visible wall thickening could be identified. Initial reports claimed excellent sensitivity in identifying an abnormally thickened intestinal wall but commented on the inability to distinguish inflammatory from neoplastic processes (2).
Part of this manuscript was presented October, 2000 at the American College of Gastroenterology annual conference, New York, NY, and May, 2001 at Digestive Diseases Week, Atlanta, GA.
The indication for CT depends on the clinical condition of the patient. If the history suggests a mucosal process, barium studies or endoscopic examination is an appropriate first choice. Endoscopic ultrasound (EUS) is a relatively new technology in clinical gastroenterology (3). EUS plays an important role in many clinical areas, such as in the evaluation of mediastinal lymph nodes or masses in patients with suspected lung cancer and in the evaluation of pancreatic malignancy (3). In recent years, EUS has been used more and more often for mucosal diseases in the GI system, especially to define the depth of invasion of a mucosal lesion, such as an esophageal cancer (4). However, when the extraluminal extension of disease might be more critical than the changes along the mucosal surface, CT still assumes a frontline role (1). Often in clinical practice, thickening of the luminal GI organs (LGIO), such as the esophagus, stomach, and small and large intestine, are incidentally reported on abdominal, pelvic, or chest CT scan. These patients do not have any known GI diseases, and they have been sent to CT for non–GI-related diseases, or they have a known GI disease, but the CT finding of a thickened LGIO is not directly related to their disease. An incidental finding of thickened LGIO is a common clinical occurrence; however, the significance of such findings is unclear. Primary care physicians often refer patients with incidental findings of thickened LGIO on CT to gastroenterologists for further evaluation. Because of the sparse scientific data, there are no clinical guidelines for the evaluation of incidental findings of thickened LGIO on CT. In the present study, the significance of a thickened esophagus, cecum, sigmoid colon, or rectum was investigated to determine whether these findings on CT warrant further investigation.
MATERIALS AND METHODS The report logbooks for abdominal CTs performed from October, 1997 through March, 1999 in the Emory University School of Medicine, Department of Radiology were checked for patients with thickening of the distal esophagus, stomach, small intestine, cecum, sigmoid, or rectum. The
AJG – August, 2003
Thickening Luminal Gastrointestinal Organs
1735
Table 1. General Patient Information
Total patients with thickening LGIO No. patients who had endoscopy Gender (M/F) Mean age (range) (M/F) (yr) No. patients with positive findings
Distal Esophagus
Cecum
Sigmoid Colon or Rectum
48 27 (56%) 9/18 47 (35–39)/52 (27–86) 22 (81%)
18 16 (88%) 6/10 69 (58–76)/57 (40–82) 2 (13%)
51 24 (47%) 18/6 55 (36–72)/54 (42–69) 23 (96%)
M ⫽ male; F ⫽ female.
final reports were verified with the electronic record (Powerchart) for the health care system. During this period, a total of 12,021 abdominal CTs were performed. The two major criteria for enrolling patients for this study were 1) patients had no known GI disease, such as esophagitis, diverticulitis, colitis, or colon cancer before the CT scan, and 2) patients had a known GI disease, but the CT finding of a thickened LGIO was not directly related to the disease. For example, a patient with cirrhosis had CT to rule out hepatoma, and CT showed a thickened cecum. In other words, only cases with incidental findings of thickened LGIO were included in this study. Subsequent endoscopic examinations for those cases were reviewed.
RESULTS During the study period, 117 cases were identified as having incidental findings of thickened distal esophagus, cecum, sigmoid colon, or rectum based on the above criteria. Among the 117 cases, 48 patients had thickened distal esophagus, 18 patients had thickened cecum, and 51 patients had thickened sigmoid colon or rectum. Most of the patients were middle-aged (Table 1). Of the 117 patients, 67 had documented further endoscopic examination at our university hospital, including 27 patients with thickened distal esophagus, 16 patients with thickened cecum, and 24 patients with thickened sigmoid colon or rectum (Table 1). CT indications for these 67 patients are listed in Table 2. Of these 67 patients, 3 had no oral or intravenous (IV) contrast; 2 of the 67 patients had oral contrast only. The rest of the patients had both oral and IV contrast. For patients who had IV contrast, 150 ml of Omnipaque (Amersham Health, Princeton, NJ) was given (administration rate was not documented in the report). Detailed information about oral contrast administration was not documented in the report. All 67 patients had helical CT, and images were obtained at 5–7-mm increments. Among the 67 patients who underwent further evaluation by endoscopic examination, 22 of 27 patients with a thickened distal esophagus (81%) and 23 of 24 patients with a thickened sigmoid colon or rectum (96%) had abnormal findings on further evaluation. Esophagitis (accounting for 30% of the thickened distal esophagus) and colitis (accounting for 75% of the thickened sigmoid colon or rectum) were the most common findings (Table 3). Of the patients with a thickened cecum, 14 of 16 (88%) had normal findings on further evaluation (Table 3).
DISCUSSION Incidental findings of thickened LGIO on CT are not uncommon in clinical practice and might be a clue to significant pathology. To our knowledge, the present study is the first systematic investigation of incidental findings of thickened LGIO on CT. Our findings indicate that thickened LGIO on CT warrants further evaluation by endoscopic examination. CT diagnosis of GI diseases has become more common in clinical practice (5–9). CT is an accurate technique in the evaluation of intestinal intussusception, GI masses, GI bezoars, and ischemia, as well as other disorders (5–9). Thickened LGIO is commonly reported by radiologists but is a very nonspecific CT finding. It could be a real finding related to pathology or an artifact caused by poor luminal distension of the LGIO or thick tenacious material adherent to the wall of the LGIO. In some situations, radiologists cannot make a specific diagnosis based on the CT image, especially when the finding is incidental and patients do not
Table 2. Indications for CT LGIO
No. patients
Indications (n)
Distal esophagus
27
Cecum
16
Sigmoid colon or rectum
24
Chronic abdominal pain (7) Chronic abdominal pain and weight loss (5) Fever, R/O pneumonia (5) F/U lung cancer (4) Hemoptysis (3) Severe chest pain (3) Cirrhosis R/O hepatoma (5) Flank pain, hematuria (4) Chronic hepatitis, R/O cirrhosis (4) F/U of pancreatitis (2) Grossly hematuria (1) Cirrhosis, R/O hepatoma. (5) S/P BMT, fever (5) F/U ovarial cancer (3) F/U renal cancer (2) F/U chronic pancreatitis (2) Hematuria (2) Fever of unknown origin (2) Chronic hepatitis, R/O cirrhosis (2) Flank pain, H/O kidney stone (1)
BMT ⫽ bone marrow transplantation; R/O ⫽ rule out; F/U ⫽ follow up; S/P ⫽ status post; H/O ⫽ history of.
1736
Cai et al.
AJG – Vol. 98, No. 8, 2003
Table 3. Positive Findings on Endoscopic Examination
LGIO
No. patients who had endoscopy
Distal esophagus
27
Cecum
16
Sigmoid colon or rectum
24
Findings
No. patients
Esophagitis Hiatal hernia Esophageal varices Normal Polyp Cancer Normal Colitis Cancer Polyp Normal
8 (30%) 7 (26%) 7 (26%) 5 (18%) 1 (6%) 1 (6%) 14 (88%) 18 (75%) 3 (13%) 2 (8%) 1 (4%)
have related GI diseases or symptoms. The observation of thickened LGIO on CT is not very difficult. The finding is based mainly on the thickness of the wall of the LGIO. Usually the wall of LGIO should be 2–3 mm (1, 10), except the rectum, which can be up to 8 mm. If the wall is thicker than 5 mm (greater than 8 mm in the rectum), the report of a thickened LGIO is made. One interesting finding in the present study is that most of the patients with thickened distal esophagus and thickened sigmoid colon or rectum had positive pathologic findings on further endoscopic workup. On the other hand, most patients with a thickened cecum had normal findings on further endoscopic workup. The reason for this difference is not absolutely clear, but at least two reasons are possible. First, if an LGIO is not distended well at the time when CT is performed, the LGIO might seem thicker than it should be. The cecum is a relatively mobile luminal organ compared with the esophagus, sigmoid colon, and rectum. The latter are relatively fixed organs and do not have much room to collapse. In other words, the esophagus, sigmoid colon, and rectum are less prone to collapse and therefore less likely to be affected by this artifact. Second, there are diseases, such as different varieties of colitis, which primarily affect the sigmoid colon and rectum. Diseases located only in the cecum are less common. Although most patients with a thickened cecum on CT had normal findings on further endoscopic examinations, there were still more than 10% who had significant abnormalities on further workup. The most common positive finding in patients with a thickened distal esophagus was esophagitis, followed by hiatal hernia and esophageal varices. There were no esophageal cancers identified in this study. We did not further subcategorize the esophagitis based on the etiology. Most of these patients had nonspecific esophagitis, probably due to gastroesophageal reflux. The seven patients with esophageal varices had no known liver cirrhosis before the examination. Another interesting finding of this study was that most of the patients with a thickened cecum had further documented evaluations. Of 18 patients with a thickened cecum, 16 (88%) had colonoscopy. Only two did not have further evaluation. In contrast, 27 of 48 patients (56%) with a
thickened distal esophagus and 24 of 51 patients (47%) with a thickened sigmoid colon or rectum had further evaluation. This might reflect the concern of the primary care physician about a thickened cecum. Because colon cancer is a relatively common disease in the United States, a thickened cecum might be perceived as more worrisome than a thickened distal esophagus or a thickened sigmoid colon and rectum, both to the patient and their physician. Approximately 50% of the patients with a thickened distal esophagus or thickened sigmoid colon or rectum did not have documented further endoscopic examination at this medical center. We are not sure whether those patients had further workup in another hospital, because we did not follow-up with these patients. Based on the endoscopic examination of our patients, the most significant findings were seven patients with esophageal varices and four patients with colon cancer. All seven patients with esophageal varices had further workup of their liver disease. Five had hepatitis C cirrhosis, and two had alcoholic cirrhosis. The four patients with colon cancer had successful surgical therapy. If there had been no further endoscopic examination for these patients, we might have missed this significant disease. It would be very interesting to know the outcomes of patients who did not have further endoscopic examination at our hospital. However, it is very difficult to obtain this information, because some patients did not come back to our hospital for continuation of medical care. There were 45 patients with diffuse small bowel thickening and 34 patients with gastric wall or antral thickening reported during this study period. However, none of these patients met the criteria of an incidental finding of thickening of LGIO. Almost all patients with small intestine thickening on CT were cirrhotic patients with low serum albumin, and none of them had further endoscopic examination. All patients with gastric wall or antral thickening were sent for CT for abdominal pain or abdominal pain and weight loss. Because those patients did not meet the criteria for incidental findings of thickening LGIO, they were not included in this study. However, some of these patients had documented endoscopic examination, and most were given a diagnosis of gastritis. Limitations of this study include patient selection and verification bias. Because patient selection was based on reports in the logbook in the radiology department, there might be omission of cases having incidental thickening of LGIO. We did not examine all of the final reports for the 12,021 patients who had a CT during this study period. Furthermore, patients who did not have endoscopic workup in our hospital but who may have had endoscopic examinations in other hospitals were also excluded from the study. Another limitation of this retrospective study is the inability to precisely quantify patients’ symptoms. All information about the patients was collected by reviewing patient charts and CT reports.
AJG – August, 2003
In conclusion, of those who underwent further workup in our hospital, most of the patients with a thickened distal esophagus and a thickened sigmoid colon or rectum on CT had an abnormality. Therefore, a thickened distal esophagus, or a thickened sigmoid colon or rectum on CT should be investigated. Of those who underwent further workup, most of the patients with an isolated thickening of the cecum on CT had normal examinations. Although the proportion of positive endoscopic findings in patients with a thickened cecum is relatively lower than that in patients with a thickened distal esophagus or thickened sigmoid colon or rectum, the finding of a thickened cecum on CT still needs further investigation because of the more than 10% possibility of localized cecal mass or polyp.
Thickening Luminal Gastrointestinal Organs
2. 3. 4. 5. 6. 7.
Reprint requests and correspondence: Qiang Cai, M.D., Emory University School of Medicine, Division of Digestive Diseases, 201 Whitehead Building, 615 Michael Street, Atlanta, GA 30322. Received Sep. 30, 2002; accepted Jan. 7, 2003.
REFERENCES 1. Megibow AJ. The gastrointestinal tract. In: Lanzieri CF, Sartors DJ, Zerhouni EA, et al., eds. Computed tomography and
8. 9. 10.
1737
magnetic resonance image of the whole body, Third edition. St. Louis: Mosby, 1994:855–94. Coscina WF, Arger PH, Levine MS, et al. Gastrointestinal tract focal mass lesions: Role of CT and barium evaluations. Radiology 1986;159:581–7. Dancygier H, Lightdale CJ. Endosonography in gastroenterology: Principles, techniques, findings. New York: Thieme, 1999. Tio TL. Diagnosis and staging of esophageal carcinoma by endoscopic ultrasonography. Endoscopy 1998;30(suppl 1): A33–40. Shiraga N. Two-dimensional and three-dimensional CT diagnosis of alimentary tract. Nippon Igaku Hoshasen Gakkai Zasshi 2001;61:672–82. Barbiera F, Cusma S, Di Giacomo D, et al. Adult intestinal intussusception: Surgery-CT correlations. Radiol Med (Torino) 2001;102:37–42. Rousso I, Hadar H, Levy A, et al. Primary CT diagnosis of gastrointestinal masses. Comput Radiol 1985;9:259 –63. Ripolles T, Garcia-Aguayo J, Martinez MJ, et al. Gastrointetinal bezoars: Sonographic and CT characteristics. AJR Am J Roentgenol 2002;177:65–9. Zaleman M, Sy M, Donckier V, et al. Helical CT signs in the diagnosis of intestinal ischemia in small-bowel obstruction. AJR Am J Roentgenol 2000;175:1601–7. Hammerman AM, Mirowitz SA, Susman N. The gastric airfluid sign: Aid in CT assessment of gastric wall thickening. Gastrointest Radiol 1989;14:109 –12.