Single or double headed capsules for the investigation of suspected small bowel bleeding: Are two heads better than one

Background Capsule endoscopy is now the accepted first line investigation for suspected small bowel (SB) bleeding. Recent evidence suggests the diagnostic yield for SB pathology may be higher for tailored double headed (DH) SB capsules. Whether other forms of bidirectional capsules offer a similar advantage is less clear. Aim To compare the efficacy of single headed versus bidirectional capsules in detecting pathology in patients with suspected small bowel bleeding. Methods A single centre prospective comparison study was conducted over an 8 month period in a tertiary care hospital. Patients referred with overt or suspected SB bleeding were assigned to either SB3 Medtronic SB capsule (SH) during the initial four months or PillCam Colon 2 Medtronic capsule (DH) during the subsequent four months. Studies were analysed by trained Capsule Endoscopists and approved by our institutions capsule review board. Findings were compared between SH and DH capsules using a chi2 or t-test as appropriate. A p value of <0.05 was considered significant. Results 201 subjects were included, mean age 61.8 years, 90 (45%) male. Majority referred with occult bleeding, 153 (76%). DH and SH capsule used in 100 and 101 cases, respectively. 90% (n=181) capsules were complete and overall diagnostic yield was 57% (n=114). Diagnostic yield was similar between both groups - DH 53% (n=53), SH 60% (n=61). Positive finding in overt bleeding; SH 85% (n=22) versus DH 50% (n=11), p<0.02. SH capsules more frequently detected SB inflammation, 27 (27%) versus 9 (9%), p<0.002. More patients had another diagnosis in the DH (19) than the SH (9), p<0.04, the majority were type 1a vascular lesions, “red spots” or diminutive colonic polyps. Conclusion Single head and double head capsules perform similary in terms of diagnostic yield overall. This supports the continued use of standard small bowel capsules for investigation of the small bowel.