Clinical Efficacy of Video Capsule Endoscopy and Balloon-Assisted in Suspected Small Bowel Bleeding: A Single Center Cohort Study

Introduction: While video endoscopy capsule (VCE) and balloon assisted enteroscopy (BAE) have emerged as the gold standard for the evaluation and treatment of suspected small bowel bleeds (SSBB), little is known regarding the clinical efficacy when used in combination. We aim to assess diagnostic yields [DY] and inter-modality agreement between VCE and BAE in SSBB. Methods: Patients who underwent VCE followed by BAE from 2012 to 2022 for SSBB were prospectivelt identified. Patients with incomplete study or waiting interval between VCE and BAE > 12 months were excluded. Clinical characteristics were collected via chart review. Diagnostic yield (DY) was defined as the proportion of positive test per the Saurin classification. Waiting interval between VCE and BAE was considered short if within 30 days and long if after 30 days. Inter-modality agreement was obtained by calculating Cohen’s kappa (k) co-efficient. The agreement obtained was considered slight (k = 0.00–0.20), fair (k = 0.21–0.40), moderate (k = 0.41–0.60), substantial (k = 0.61–0.80), or almost perfect (k = 0.81–1.00). Continuous and categorical values were assessed with Fisher’s or T-test when appropriate Results: 151 patients met the inclusion criteria, of which 68 (45%) and 83 (55%) were found to have an overt and occult SSBB respectively. Patient outcomes are described in Table 1. Overall, BAE trended toward higher DY in comparison to VCE (56% vs 45%, P=0.06) with overall fair inter-modality agreement (k=0.22) In patients with shorter waiting interval (< 30 days), DY was significantly higher in BAE compared to VCE (54% vs 34%, P=0.01) with slight agreement (k=0.18). In patients with longer waiting interval ( > 30 days), DY was similar between the BAE and VCE (59% vs 62%, P=0.85) with moderate agreement (k=0.43) In patients with occult SSBB, BAE and VCE have similar DY (55% vs 55%, P=1) and fair agreement (k=0.29). In overt SSBB group, BAE has significantly higher DY compared to VCE (57% vs 32%, P=0.01) with slight agreement (k=0.15). Conclusion: Compared to VCE, DY of BAE is significantly higher in overt SSBB and shorter waiting interval. Inter-modality agreement is better in occult SSBB and longer waiting intervals. DY of BAE and inter-modality agreement appear to be inversely related with respect to time and subtypes of SSBB. A larger sample size and additional studies are needed to confirm the results and provide guidance for clinical practice. Table 1. - Clinical Efficacy of the Study Population Diagnostic Yield Inter-Modality Agreement Waiting Interval (days) VCE BAE P-Value Kappa 95% CI Strength Total (N=151) 68 (45%) 85 (56%) 0.06 0.22 0.14-0.30 Fair 1-7 (N=67) 23 (34%) 36 (53%) 0.04 0.16 0.05-0.27 Slight 1-30 (N=92) 31 (34%) 50 (54%) 0.01 0.18 0.08-0.27 Slight > 30 (N=59) 37 (62%) 35 (59%) 0.85 0.43 0.29-0.56 Moderate Subtypes Overt (N=68) 22 (32%) 39 (57%) 0.01 0.15 0.04-0.26 Slight Occult (N=83) 46 (55%) 46 (55%) 1 0.29 0.18-0.40 Fair