Utility of upper endoscopy and colonoscopy in evaluating gastrointestinal luminal wall thickening found on computed tomography.
Computed Tomography (CT) is an increasingly used modality in the initial work-up of patients who present with abdominal pain. CT scans are known to be an effective diagnostic tool in the evaluation of abdominal pathology such as tumors, obstruction, strictures, etc.
In the initial CT scan of the abdomen and the pelvis, intraluminal pathology, if found, is always mentioned. Substantial information, both abnormal findings and normal variants can be gleaned by a radiologist from studying bowel appearance. One of the most common findings reported on abdominal CT studies is thickening of the bowel wall. Bowel wall thickening may be a normal finding, or may be indicative of pathology such as inflammatory bowel disease or an acute inflammatory process. (1-3) In addition, intraluminal thickening as a radiological finding may also show esophageal pathologies upon direct visualization if found in the upper GI tract. (4-5) Furthermore, bowel wall inflammation caused by infectious etiologies is well known to manifest as bowel wall thickening on CT scans. (6)
Bowel wall thickening is defined as increased bowel wall diameter above the upper limit of normal which is 1-2 mm for the small bowel, and 2-3 mm for the colon. However, some studies seem to suggest that a large portion (20% to 60%) of patients with colonic bowel wall thickening seen on a CT scan, are found to have a normal endoscopic exam upon follow up. (3,4,9,11)
To clarify the indication for endoscopy, we therefore decided to do a retrospective review of endoscopic reports of those patients that had been referred to the Section of Digestive Diseases for bowel wall thickening seen on abdominal CT scans.
Materials and Methods
Medical records of 169 patients referred to a tertiary level hospital for abnormal findings seen on CT that thereafter received an endoscopic evaluation were identified. These patients were seen over a period of one year and their charts reviewed. The items reviewed in the charts were age, gender, chief complaint, physical findings, ordering location, CT scan indication, CT scan abnormality, endoscopic findings and the time interval between CT scan and endoscopic procedure.
We studied a total of 169 patients. The average age of the patients in this study ranged from 18 to 89 with a mean of 55 years. Of these, 41% were male (Table 1). The time period between CT scan and endoscopy ranged from the same day to a maximum of 176 days with a mean of 32 days. In this sample, the most common chief complaint was abdominal pain (66%), followed by gastrointestinal bleeding (10%), history of non-abdominal cancer (4%), diarrhea (3%), dysphagia (2%), nausea/vomiting (2%), fever (2%), and non-abdominal complaints/ others (6%) as seen in Table 1. Some patients had multiple complaints.
Physical exam was generally normal in most cases (66%). Abdominal tenderness was seen only in 34% of the cases. As far as the indications for a CT scan, 52% were for abdominal pain, followed by bleeding (7%), weight loss (4%), C. diff colitis (3%), nausea/ vomiting (2%), diverticula (2%) and other non-abdominal complaints (10%). Almost 12% of CT scans had no indication mentioned.
In the upper gastrointestinal tract, the most common CT scan finding in a total of 45 patients was esophageal, gastric, or duodenal thickening (39). Other findings seen on CT scans were of a mass (2),filling defect (2), diverticulum (1), or a smooth intraluminal appearance (1). Out of these 45 patients, 28 (62%) were found to have subsequent pathology on endoscopic examination (Table 2). Radiographic "luminal thickening" as seen on CT scans involving the upper GI tract was found to correlate highly with pathology seen on endoscopy. Out of these 39 patients, 25 (64%) had these findings. Upon upper endoscopy, the most common pathology found was gastritis (24%) followed by gastric ulcer (20%), duodenitis (16%), esophageal ulceration (16%), esophagitis (8%), gastric polyps (8%), stricture (4%) and gastric antral vascular ectasia (4%). These results are shown in Table 3. This translates to a positive predictive value of 64% for thickening seen on a CT scan correlating to an endoscopically proven upper GI pathology (Table 2).
Out of a total of 169 patients, 124 patients had pathology in the lower gastrointestinal tract on CT scans. The most common finding on the CT scans was "abnormal luminal thickening" (109/124) varying in location from the terminal ileum to rectum. Out of these 124 patients, 45 patients had positive findings on colonoscopy (Table 4). Out of the 109 patients with thickening, only 36 patients were found to have pathology on endoscopic exam. The positive predictive value for "bowel wall thickening" was only 33%.
The most common disease process seen on colonoscopy was colitis (53%), hyperplastic polyps (19%), adenocarcinoma (7%), proctitis (6%), ileocecal valve inflammation (6%), ileitis (3%), AVM (3%) and squamous cell carcinoma (3%) (Table 5). The most common form of colitis was pseudomembranous colitis (32%).
Computed tomography is an important diagnostic modality in the initial evaluation of abdominal complaints. In the abdomen, it is very useful in identifying pathological processes such as gallstones, tumors of the solid organs, and masses. The CT scan is also useful in assessing intraluminal inflammatory conditions. When confusing or overlapping CT parameters are encountered or uncertainties persist, barium enema or direct endoscopic visualization of the suspected bowel segment is often done. (4)
The aim of this study was to determine whether bowel wall thickening seen on CT scans correlates with disease pathology seen endoscopically. After data collection, the study was divided anatomically into upper GI and lower GI groups according to the endoscopic intervention done.
In the upper GI cohort, 39 individuals had luminal wall thickening seen on CT scans. In patients with luminal wall thickening in the upper GI tract, 25 had endoscopically proven abnormal findings such as gastritis, duodenitis, and ulcers. Hence, the positive predictive value of thickening seen on CT scans for upper GI pathology was 64%. In a similar study done in Korea, the predictive value of thickening seen on CT scans ranged from 73% to 83%. (7)
In the lower GI group, 109 patients had luminal wall thickening seen on CT scans. Of these, only 36 patients had actual pathology seen endoscopically. Thus, the predictive value was 33%. Luminal thickening as seen on CT scans involving the lower GI tract did not correlate highly with pathology seen on follow up colonoscopy. Out of these 36 cases, the most common pathology was colitis (53%). Three cases of adenocarcinoma (7%) and one case of squamous cell carcinoma (3%) were also seen. In a similar study done on thirty-five patients with colonic wall thickening on CT scan, eight patients were found to have neoplasia (23%).8 In our study, we only found 4 cases (10%) of neoplasia.
In a recent study done on 40 patients with a radiological finding of bowel wall thickening, positive findings in the cecum and sigmoid colon/rectum ranged from 13 to 96%.9 In our study of 109 patients, only 36 had endoluminal pathology, which is a poor positive predictive value of 33%. However, it is important to point out that neoplasia comprised of 10% of those findings.
The CT scans were read by several staff radiologists and hence, there is the possibility of variability amongst the findings seen on the scans. Another factor in causing variability in the findings is the degree of thickening seen on the CT scans. Thickening expressed as a dimension of the wall diameter was often left out in the final radiology report. Thus, it is not possible to know if a greater degree of thickening would correlate with a greater incidence of inflammatory pathology. Currently, there are specific parameters for describing bowel thickening, but subjective variability still abounds. (5) While our study included all patients referred to the Digestive Diseases section, there may have been a small number of patients that may have had thickening on their CT scans and on resolution of their symptoms, were not referred for an endoscopy or were lost for follow up. These factors should be taken into account when interpreting these results.
Further research endeavors may involve increasing the sample size and perhaps more in-depth subgroup analysis which may include the subjects' age, indications, or symptoms.
In conclusion, luminal wall thickening in the upper GI tract seen on CT scan may be a very useful predictor of inflammatory pathology (64%) and thus, direct visualization for further evaluation which may include mucosal biopsy is recommended.
Some studies have shown better correlation between colonic wall thickening and pathology which may be as high as 74%.12 In our study, colonic thickening as seen on CT scan only correlated with an endoluminal pathology 33% of the time. However, direct visualization may be imperative, as 10% of findings were malignancies (Adenocarcinoma and Squamous Cell Carcinoma). Nevertheless, the decision to perform a follow up endoscopic evaluation should be based on a sound judgment of the correlation between clinical and radiological findings.
(1.) Desai RK, Tagliabue JR, Wegryn SA, CT evaluation of wall thickening in the alimentary tract. Radiographics. 1991 Sep; 11(5):771-83; discussion 784. Review.
(2.) Frager DH, Goldman M, Beneventano TC. Computed tomography in Crohn disease. J Comput Assist Tomogr. 1983 Oct; 7(5):819-24.
(3.) Fisher JK, Abnormal colonic wall thickening on computed tomography, J Comput Assist Tomogr. 1983 Feb; 7(1):90-7.
(4.) Balthazar EJ, Charles HW, Megibow AJ. Salmonella- and Shigella-induced ileitis: CT findings in four patients. J Comput Assist Tomogr. 1996 May-Jun; 20(3):375-8.
(5.) Shin WC, Jeong MJ. Clinical significance of incidentally detected bowel wall thickening on abdominal computerized tomography scan. Korean J Gastroenterol. 2005 Jun; 45 (6):409-16. Korean.
(6.) Macari M, Balthazar E. CT of Bowel Wall Thickening: Significance and pitfalls of interpretation, Am J Roentgenol. 2001 May; 176:1105-16.
(7.) Berkovich GY, Levine MS, Miller WT Jr. CT findings in patients with Esophagitis. AJR Am J Roentgenol. 2000 Nov; 175(5): 1431-4.
(8.) Nino-Murcia M, Stark P, Triadafilopoulos G. Esophageal wall thickening: a CT finding in diffuse esophageal spasm. J Comput Assist Tomogr. 1997 Mar-Apr; 1(2):318-21.
(9.) Bleibel W, Guerrero JE, Kim S, Leao L, Ghosh T, Kenney TJ Jr. The clinical significance of incidental computer tomography finding of gastrointestinal luminal wall thickening as evaluated by endoscopy. Dig Dis Sci. 2007 Mar; 52:1709-12.
(10.) Wolff JH, Rubin A, Potter JD, Lattimore W, Resnick MB, Murphy BL, Moss SF. Clinical significance of colonoscopic findings associated with colonic thickening on computed tomography: Is colonoscopy warranted when thickening is detected? J Clin Gastroenterol 2008 May/June; 42(5):472-75.
(11.) Choi D, Jin Lee S, Ah Cho Y, Lim HK. Bowel wall thickening in patients with Crohn's disease: CT patterns and correlation with inflammatory activity. Clin Radiol. 2003 Jan; 58 (1):68-74.
(12.) Cai Q, Baumgarten DA, Affronti JP, Waring JP. Incidental findings of thickening luminal gastrointestinal organs on computed tomography: an absolute indication for endoscopy. Am J Gastroenterol. 2003 Aug; 98 (8):1734-7.
Muhammad Sohaib Karim, MD
Max C. Miranda, MD
John M. Shamma'a, MD
Stephan U. Goebel, MD
Uma Sundaram, MD
Section of Digestive Diseases, Department of Medicine West Virginia University School of Medicine, Morgantown
Table 1: Patient Demographics Gender Male 70/169 (41%) Female 99/169 (59%) Chief Complaint on Admission * Abdominal Pain 112/169 Bleeding 16/169 History of Non-GI cancer 7/169 Diarrhea 5/169 Dysphagia 4/169 Nausea & Vomiting 4/169 Fever 4/169 Other Non-Abdominal complaints 10/169 None indicated 12/169 CT Indication * Abdominal Pain 88/169 N/A 20/169 Mass 13/169 Bleeding 11/169 Weight Loss 8/169 C. Diff Colitis 5/169 Nausea & Vomiting 4/169 Diverticula 4/169 Other Non-Abdominal complaints 17/169 No Indication Listed 20/169 Physical Examination Abdominal Tenderness 57/169 Benign Exam 112/169 Type of Endoscopic Evaluation EGD 45/169 (27%) Colonoscopy 124/169 (73%) * Some studies had multiple entries Table 2: Correlation of Pathology on Upper Endoscopy and CT Scan Finding (n=39) CT Scan Finding Pathology seen Pathology not on EGD seen on EGD Thickening 25/39 (64%) 14/39 (36%) Mass 1/2 (50%) 1/2 (50%) Filling Defect 1/2 (50%) 1/2 (50%) Smooth Appearance 1/1 (100%) 0/1 (0%) Diverticula 0/1 (0) 1/1 (100%) Table 3: Endoscopic findings of patients with pathology and luminal thickening on CT scan (n=25) Gastritis 6/25 Gastric Ulcer 5/25 Duodenitis 4/25 Esophageal Ulcer 4/25 Esophagitis 2/25 Gastric Polyp 2/25 Gastric antrum vascular ectasia 1/25 Peptic Stricture 1/25 Table 4: Correlation of Pathology on Colonoscopy and Abnormal CT Scan (n=124) CT Scan Finding Pathology present Pathology absent on on Colonoscopy (45/124) Colonoscopy (79/124) Thickening 36/109 (33%) 73/109 (67%) Mass 9/15 (60%) 6/15 (40%) Table 5: Colonoscopic findings in patients with pathology and luminal thickening on CT (n=36) Colitis 19/36 (53%) Clostridium Difficile Colitis 6/19 (32%) Inflammatory Bowel Disease 6/19 (32%) Ischemic Colitis 4/19 (21%) Colitis (unspecified) 3/19 (15%) Hyperplastic Polyps 7/36 (19%) Adenocarcinoma 3/36 (7%) Ileocecal Valve Inflammation 2/36 (6%) Proctitis 2/36 (6%) Ileitis 1/36 (3%) Angiodysplasia 1/36 (3%) Squamous Cell Carcinoma 1/36 (3%)
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Scientific Article|
|Author:||Karim, Muhammad Sohaib; Miranda, Max C.; Shamma'a, John M.; Goebel, Stephan U.; Sundaram, Uma|
|Publication:||West Virginia Medical Journal|
|Date:||Nov 1, 2010|
|Previous Article:||Superior mesenteric artery aneurysm: an uncommon disease with potentially serious complications.|
|Next Article:||Leptospirosis induced pulmonary alveolar hemorrhage.|