OBJECTIVES: Preliminary studies have shown the possibility of detecting colonic polyps and tumors using the PillCam Colon capsule endoscope (CCE) (Given Imaging, Yoqneam, Israel). The aim of this study was to evaluate the ability of CCE to detect clinically relevant colonic findings as compared with colonoscopy, and further, to test the assumption that CCE used in the frame of colorectal cancer (CRC) screening could accurately discriminate patients deserving a complete colonoscopy. METHODS: A total of 128 patients (67 men, 55+/-14 years) with an indication of colonoscopy were investigated by CCE, followed by colonoscopy under general anesthesia on the next day. Bowel cleansing was carried out according to a previously published protocol (3 l polyethylene glycol+2 doses of sodium phosphate solution). All CCE recordings were read by the same physician and all colonoscopies, performed by the same operator, were blinded to each other's results. The primary outcome of the study was the decision made by the CCE reader to indicate a colonoscopy, compared with the final result of the colonoscopy. Secondary outcomes were the agreement between CCE and colonoscopy for making a diagnosis of colorectal disease, as well as detection rate, number, and size of polyps. RESULTS: Two patients were excluded: one did not swallow the capsule and the other was diagnosed with a jejunal stenosis by the CCE. The CCE found at least one clinically relevant colonic finding in 71 patients (56.3%), and the colonoscopy results confirmed this finding in 56 patients, showing a sensitivity of CCE of 87.5% (confidence interval (CI) 79.4-95.6%) and a specificity of 75.8% (CI 65.4-86.2%). With a prevalence of 50.8% of clinically relevant findings at colonoscopy, the positive predictive value of CCE was 78.9%, and the negative predictive value (NPV) was 85.4%. The agreement between CCE and colonoscopy was significant for the detection of any colonic lesion (P<0.0001), polyps (P<0.001), as well as for the detection of the number (P<0.0001) and size (P=0.0083) of polyps. Tolerance of the procedure was excellent, and no complication was reported. CONCLUSIONS: In the setting of this study, CCE seemed to be effective in detecting clinically significant colonic findings in patients with an indication of colonoscopy. The high NPV and excellent tolerance of CCE suggest that it could be evaluated in large CRC-screening programs and further studies in screening conditions should also evaluate its cost-efficacy ratio.
OBJECTIVES: Preliminary studies have shown the possibility of detecting colonic polyps and tumors using the PillCam Colon capsule endoscope (CCE) (Given Imaging, Yoqneam, Israel). The aim of this study was to evaluate the ability of CCE to detect clinically relevant colonic findings as compared with colonoscopy, and further, to test the assumption that CCE used in the frame of colorectal cancer (CRC) screening could accurately discriminate patients deserving a complete colonoscopy. METHODS: A total of 128 patients (67 men, 55+/-14 years) with an indication of colonoscopy were investigated by CCE, followed by colonoscopy under general anesthesia on the next day. Bowel cleansing was carried out according to a previously published protocol (3 l polyethylene glycol+2 doses of sodium phosphate solution). All CCE recordings were read by the same physician and all colonoscopies, performed by the same operator, were blinded to each other's results. The primary outcome of the study was the decision made by the CCE reader to indicate a colonoscopy, compared with the final result of the colonoscopy. Secondary outcomes were the agreement between CCE and colonoscopy for making a diagnosis of colorectal disease, as well as detection rate, number, and size of polyps. RESULTS: Two patients were excluded: one did not swallow the capsule and the other was diagnosed with a jejunal stenosis by the CCE. The CCE found at least one clinically relevant colonic finding in 71 patients (56.3%), and the colonoscopy results confirmed this finding in 56 patients, showing a sensitivity of CCE of 87.5% (confidence interval (CI) 79.4-95.6%) and a specificity of 75.8% (CI 65.4-86.2%). With a prevalence of 50.8% of clinically relevant findings at colonoscopy, the positive predictive value of CCE was 78.9%, and the negative predictive value (NPV) was 85.4%. The agreement between CCE and colonoscopy was significant for the detection of any colonic lesion (P<0.0001), polyps (P<0.001), as well as for the detection of the number (P<0.0001) and size (P=0.0083) of polyps. Tolerance of the procedure was excellent, and no complication was reported. CONCLUSIONS: In the setting of this study, CCE seemed to be effective in detecting clinically significant colonic findings in patients with an indication of colonoscopy. The high NPV and excellent tolerance of CCE suggest that it could be evaluated in large CRC-screening programs and further studies in screening conditions should also evaluate its cost-efficacy ratio.