Clinical Predictors of Significant Esophagogastroduodenoscopic Findings for Asymptomatic Subjects with Positive Immunochemical Fecal Occult Blood Test and Negative Colonoscopy
Date Issued
2010
Date
2010
Author(s)
Chiang, Tsung-Hsien
Abstract
Background: Immunochemical fecal occult blood test (I-FOBT), esophagogastroduodenoscopy (EGD) and colonoscopy are common tools for detecting gastrointestinal (GI) bleeding today. The American Cancer Society recommends fecal occult blood testing for patients at average risk for colorectal cancer. And, colonoscopy is usually the initial diagnostic procedure for all subjects with positive fecal occult blood tests. However, there have been very few studies pertaining to the workup of patients who present with positive I-FOBT and negative colonoscopy. Besides, it is still controversial for EGD in asymptomatic patients with positive fecal occult blood test and no apparent source of colonic bleeding.
Objective: The aim of this study was to prospectively evaluate the diagnostic value of a positive I-FOBT for upper GI lesions after a negative colonoscopy in a large cohort of asymptomatic subjects at average risk for colorectal cancer, and to determine which subjects and clinical characteristics associated with the presence of upper GI lesions consistent with blood loss by EGD. Moreover, we tried to find the significant clinical predictors for positive-EGD in those subjects.
Design: A cross-sectional analysis of asymptomatic adults who underwent same day EGD, colonoscopy and I-FOBT examinations in our health screening program from August 2007 through July 2009 was performed in one institution. The analysis was based on data generated from personal medical history (consumption of aspirin, clopidogrel, nonsteroidal antiinflammatory drug, smoking, alcohol consumption), hemogram, endoscopy (including EGD and colonoscopy), pathological findings, H. pylori status and I-FOBT of examinees. The clinical and laboratory data were included in the univariate analysis initially to identify potential predictors for upper GI lesions and to calculate crude odds ratios (ORs). Multivariate analysis was performed with logistic regression to identify independent variables potentially associated with the presence of upper GI lesions consistent with blood loss.
Results: A total 2796 of 2871 subjects were enrolled in the study with I-FOBT and completed endoscopy. A total of 397 subjects had positive I-FOBT (14.2%), including 61.5% male subjects. The 233 of 397 subjects had negative colonoscopy. Among these 233 negative colonoscopy subjects, 46 subjects had significant upper GI lesions, including 1 angiodysplasia and 45 peptic ulcer diseases. Multivariate logistic regression analysis showed that age, H. pylori infection, low hemoglobin level, and alcohol consumption were independent factors associated with upper GI lesions. The accumulative ROC curve of area, including age, H. pylori infection, low hemoglobin, and alcohol consumption, was 0.8. The specificity and negative predictive value of positive-EGD were 75.4% and 90.4% for H. pylori infection; 90.4% and 82.8% for low hemoglobin level; 78.1% and 83.4% for alcohol consumption. The overall positive-EGD sensitivity of subjects with any one of the independent factors is 10 times than other subjects without any one of the independent factors.
Conclusion: EGD was necessary for healthy subjects with age of 50 years or older, or H. pylori infection, or low hemoglobin level, or alcohol consumption after initial positive I-FOBT and negative colonoscopy. Most of the upper GI diseases were benign and treatable peptic ulcers.
Subjects
I-FOBT
endoscopy
SDGs
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-99-P95421011-1.pdf
Size
23.32 KB
Format
Adobe PDF
Checksum
(MD5):af522527c937dbf097f615e93d8bb251