Background and aim: Capsule endoscopy (CE) has an important role in the diagnosis of patients with obscure gastrointestinal bleeding. However, there was still controversy regarding the best candidates for CE. The present retrospective study aimed to access the best candidates for CE.
Methods: There were 91 consecutive patients referred to the present study for 94 CE examinations from May 2002 to January 2005. They were divided into two groups (41 with active bleeding, and 50 with previous bleeding).
Results: The CE findings were positive in 74.7%, suspicious in 11% of cases and negative in 14.3% of cases, respectively. The positive and suspected positive yield of CE were 75.6% and 19.5% in the active bleeding group, 74% and 4% in the previous bleeding group (P = 0.01), 75.7% and 21.6% in the overt bleeding group, 75% and 0% in the occult bleeding group (P = 0.206), 89.7% and 3.4% in patients who bled 2 weeks previously, 52.38% and 4.76% in patients who bled >2 weeks previously (P = 0.003), 75.6% and 19.5% in active bleeding group, 89.7% and 3.4% in patients who bled 2 weeks ago (P = 0.128).
Conclusions: The best candidates for the procedure are those with active bleeding, or active bleeding in the previous 2 weeks.