Prevalence, Epidemiology, and Association of Barrett’s Esophagus Among Patients With Inflammatory Bowel Disease: A National Estimate
Background:
In the past decade, the prevalence of inflammatory bowel disease (IBD), which affects 3 million people, and Barrett’s esophagus (BE), which affects 1.3% and 1.6% (3.3 million cases), respectively, has increased. IBD, a medical condition that causes inflammation, raises the risk of developing gastrointestinal dysplasia or neoplasia. Thus, this increased risk increases the risk of BE or BE-associated malignancy. Colorectal cancer risk factors include IBD. However, the evidence linking IBD to BE and BE-associated esophageal cancer is lacking. Several studies have found that people with IBD and Barrett’s esophagus have higher death rates and worse outcomes. The study sought to determine the prevalence and association of BE and BE-associated esophageal cancer with IBD patients.
Methods:
We performed a retrospective cross-sectional study using the nationwide inpatient sample (2016-2018). Adults with IBD (primary and secondary diagnosis: Crohn’s disease and Ulcerative colitis) were identified using ICD-10 codes. We evaluated the presence of BE and BE-associated esophageal cancer among IBD patients. Univariate analysis (chi-square test) and a mixed-effects, multivariable survey logistic regression analysis were performed to find out the prevalence, sociodemographic characteristics, and association of BE and BE-associated esophageal cancer among IBD patients. The adjusted odds ratio and 95% confidence interval were calculated using SAS 9.4, keeping the alpha criteria at 0.05.
Results:
We found 930,280 patients with IBD [CD: 583,765 (0.67%) and UC: 346,515 (0.39%)]. The prevalence of BE (0.77% vs 0.63% vs 0.40%, P< 0.0001), BE without dysplasia (0.76% vs 0.62% vs 0.39%, P< 0.0001), and BE with dysplasia (0.01% vs 0.01% vs 0.01%, P< 0.0001) was higher in UC and CD in comparison to those without IBD. The prevalence of BE-associated esophageal malignancy (0.08% vs 0.03% vs 0.13%, P< 0.0001) was lower in IBD patients. Patients with BE were older (67 vs 57-years old), male (0.57% vs female:0.28%), White race (0.53% vs Native American:0.32%), higher SES (0.5% vs lower SES:0.29%), hypertensives (0.51% vs 0.27%), hyperlipidemics (0.62% vs 0.30%), and alcoholics (0.63% vs 0.39%).(P< 0.0001) In regression analysis, we found higher odds of BE (CD aOR:1.67, 95%CI:1.55-1.79, UC 1.77, 1.62-1.93) and BE without dysplasia (CD 1.67, 1.55-1.80, UC 1.79, 1.64-1.96) among patients with IBD in comparison to patients without IBD. IBD was not linked to a higher risk of BE with dysplasia (CD 1.05, 0.9-2.55, UC 0.77, 0.32-1.86) or BE-associated esophageal malignancy (CD 0.24, 0.18-0.34, UC 0.49, 0.37-0.64). Among IBD patients, age group 51-75-years (2.26, 1.90-2.69), >75-years (1.99, 1.58-2.51), White (3.24, 2.34-4.49), Native American (4.47, 1.88-10.60), Hispanic (2.23, 1.47-3.39), female (1.58, 1.41-1.78), hypertension (1.29, 1.11-1.50), hyperlipidemia (1.34, 1.17-1.53), and smoking (1.39, 1.20-1.61) were associated with higher odds of diagnosing with BE.
Conclusions:
Early diagnosis of premalignant lesions in patients with IBD reduces the burden of BE. Long-term monitoring is warranted for patients with IBD. Although our study included a large cohort of US population with a large sample size and appropriate randomization based on sociodemographic diversity, the findings of the study were limited to inpatient settings without causality, long-term follow-up, sensitivity of the ICD-10 codes, severity of IBD, data on endoscopy and histopathology, etc.
