HOME
Return to Table of Contents
PREDICTORS FOR SMALL BOWEL TUMOURS IN CAPSULE ENDOSCOPY: ARE THERE ANY?
JK Law, I Bregman, W Chung, A Hassanali, P Kazemi, SE Krausz, L Sowsa, P Intaraprasong, RA Enns
Capsule endoscopy (CE) has revolutionized small bowel imaging, allowing for visualization of mucosa not easily assessed by traditional imaging techniques. Small bowel tumours are infrequent and represent only 1 to 3% of all gastrointestinal malignancies.The purpose of this study was to identify any risk factors or patient characteristics that can help predict which patients are likely to have a small bowel tumours on CE.
Methods: Patient characteristics and risk factors were prospectively collected on all patients prior to undergoing capsule endoscopy for obscure gastrointestinal bleeding (OGIB) at a single center over a five-year period (2002-07). Results were also noted and exact logistic regression was employed for univariate analysis and factor selection to identify predictors of patients with small bowel tumours. Patient age, gender, comorbidities, medication use (NSAIDs, Plavix, coumadin, selective-serotonin re-uptake inhibitors, and iron supplementation), and gastrointestinal complaints in addition to previous endoscopic procedures and transfusion requirements were analyzed.
Results: 569 patients have undergone capsule endoscopy. In 461 patients, the primary indication was for OGIB. 29 patients were found to have small bowel tumours. Of the 431 non-tumour patients, 227 (52.7%) were men and the age was 62.2 years (±17.3) compared to the 29 patients in the tumour group comprising of 13 men (44.8%) and mean age of 61.3 years (±16.2) [p=0.79 and p=0.41, respectively). There was no other significant difference in medication use, gastrointestinal symptoms, previous endoscopic procedures or transfusion requirements. There was a significant difference in patients with a diagnosis of small bowel tumour being more likely to have previously had a colonic malignancy (10.3% vs 1.6%, p<0.05). Patients with bright red blood per rectum were less likely to have a small bowel tumour (19.7% vs 6.9%, p=0.09).
CONCLUSION:There are very few clinical patient selection criteria that help identify which patients with OGIB are more likely to have a small bowel tumour versus other more benign pathology. We found that patients with a past history of colonic malignancy were significantly more likely to have a small bowel tumour while patients presenting with blood per rectum but negative endoscopic investigations were less likely to have a small bowel mitotic lesion.