The Impact of Confounders on Symptom-Endoscopic Discordances in Crohn's Disease.

Abstract

Discordances between clinical and endoscopic Crohn's disease (CD) activity indices negatively impact the utility of clinic visits and efficacy assessments in clinical trials. Bile acid diarrhea (BAD) and small intestinal bacterial overgrowth (SIBO) mimic CD symptoms. This study quantified the impact of BAD and SIBO on the relationship between clinical and endoscopic disease activity indices. CD patients with 7α-hydroxy-4-cholesten-3-one (7C4) serum measurements and/or SIBO breath tests and matched clinical and endoscopic scores were included. Clinical remission (stool frequency [SF] ≤ 1 and abdominal pain score ≤ 1) rates were compared between those with and without (1) endoscopic remission, (2) BAD (7C4 > 55 ng/mL), and (3) SIBO. Of 295 CD patients, 219 had SIBO testing and 87 had 7C4 testing. Patients with elevated 7C4 had lower proportions with clinical remission (14% vs 40%, P = .007) and SF ≤ 1 (14% vs 42%, P = .004) compared to those with normal 7C4. In patients with normal 7C4, higher rates of clinical remission (65% vs 27%, P = .01) and SF ≤ 1 (71% vs 27%, P = .003) existed in patients with endoscopic remission compared to those without endoscopic remission. Conversely, among the entire 295 patient cohorts, nearly identical clinical remission rates existed between those with and without endoscopic remission (25% vs 24%, P = .8), and the Crohn's Disease Patient-Reported Outcome-2 score was not accurate for predicting endoscopic remission (Area Under the Curve (AUC): 0.48; 95% CI, 0.42-0.55). SIBO status did not impact clinical remission rates (P = 1.0). BAD, but not SIBO, contributed to symptom scores. A relationship between endoscopic inflammation and clinical remission rates only existed in patients without 7C4 elevations.