Finger clubbing in inflammatory bowel disease: association with upper small bowel lesions and need of surgery in Crohn's disease.

OBJECTIVE Finger clubbing has been associated with inflammatory bowel disease (IBD). AIMS In a prospective single-center study, we aimed to assess the frequency of finger clubbing in a cohort of IBD patients. Whether finger clubbing is associated with clinical characteristics of IBD was also investigated. METHODS IBD patients with a detailed clinical history were enrolled. Finger clubbing was assessed by visual inspection. Data were expressed as median (range), chi-square, t-test. Multivariate logistic regression analysis was used to assess risk factors for finger clubbing, when considering demographic and clinical characteristics, smoking habits and chronic pulmonary diseases (CPD). RESULTS Finger clubbing was searched in 470 IBD patients: 267 Crohn's disease and 203 ulcerative colitis. Finger clubbing was more frequent in Crohn's disease than in ulcerative colitis: 45/267 (16.8%) vs. 15/203 (7.3%) [odds ratio (OR), 2.54 (1.37-4.70); P = 0.003]. Crohn's disease involved the ileum (59.9%), colon (4.5%), ileum-colon (25.8%) and upper gastrointestinal (GI) (9.8%). Ulcerative colitis extent included proctitis (E1) (13.4%), left-sided (E2) (43.3%) and pancolitis (E3) (43.3%). Upper GI lesions, but not other Crohn's disease localizations, were more frequent in patients with finger clubbing [9/45 (20%) vs. 17/222 (7.7%); P = 0.032]. Crohn's disease-related surgery was more frequent in patients with finger clubbing [36/45 (80%) vs. 107/222 (48.1%); P < 0.001]. In Crohn's disease, the only risk factors for finger clubbing were upper GI lesions and Crohn's disease-related surgery [OR, 2.58 (1.03-6.46), P = 0.04; OR, 4.07 (1.86-8.91), P = 0.006]. Ulcerative colitis extent was not associated with finger clubbing [E1: OR, 0.27 (0.02-3.44), P = 0.33; E2: OR, 0.93 (0.24-3.60), P = 0.92; E3:OR, 0.64 (0.22-1.86), P = 0.59]. In ulcerative colitis, but not in Crohn's disease, finger clubbing was more frequent in smokers [13/15 (86.6%) vs. 99/188 (52.6%); P = 0.01] and in patients with CPD [5/15 (33.3%) vs. 16/188 (8.5%); P = 0.002]. Smoking and CPD were the only risk factors for finger clubbing in ulcerative colitis [OR, 7.18 (1.44-35.78), P = 0.01; OR, 10.93 (2.51-47.45), P = 0.001]. CONCLUSION In the tested IBD population, finger clubbing was more frequent in Crohn's disease than in ulcerative colitis. In Crohn's disease, upper GI lesions and history of Crohn's disease-related surgery were risk factors for finger clubbing, suggesting the possible role of finger clubbing as a subclinical marker of Crohn's disease severity.