BACKGROUND: Recent advances in gadolinium-enhanced magnetic resonance imaging (G-MRI) have been developed to enhance the resolution of the intestinal mucosa and facilitate the differentiation of ulcerative colitis (UC) from Crohn's disease (CD). The objective of this study is to apply this technology in Pediatrics. METHODS: A G-MRI was performed on 58 consecutive children with suspected IBD between 1999 and 2002 using intravenous gadolinium, fat suppression, and respiration-suspended sequences to enhance the resolution of the intestinal wall. The sensitivity and specificity in diagnosing either UC or CD was determined by comparing the G-MRI to the established histologic diagnosis. RESULTS: G-MRI confirmed the diagnosis of either CD (21) or UC (7) with a sensitivity and specificity of 96% and 92%, respectively. Among the 21 patients with CD, 14 showed proximal small bowel involvement by G-MRI. In total, 17 patients were diagnosed with indeterminate colitis (IC) based on histologic criteria alone, and among these patients, G-MRI had a significantly lower non-classification rate (P < 0.02). In comparison, endoscopy was less sensitive (57%), but more specific (100%) than either histology or G-MRI in diagnosing IBD. G-MRI also showed a strong concordance with computed tomography in diagnosing CD (P = 0.001). CONCLUSION: G-MRI is a both a sensitive and specific radiologic tool in diagnosing pediatric IBD. In patients with CD, G-MRI may be useful in identifying proximal small bowel involvement. Longitudinal follow-up studies are needed in those patients diagnosed with IC to determine the predictive value of G-MRI testing.
BACKGROUND: Recent advances in gadolinium-enhanced magnetic resonance imaging (G-MRI) have been developed to enhance the resolution of the intestinal mucosa and facilitate the differentiation of ulcerative colitis (UC) from Crohn's disease (CD). The objective of this study is to apply this technology in Pediatrics. METHODS: A G-MRI was performed on 58 consecutive children with suspected IBD between 1999 and 2002 using intravenous gadolinium, fat suppression, and respiration-suspended sequences to enhance the resolution of the intestinal wall. The sensitivity and specificity in diagnosing either UC or CD was determined by comparing the G-MRI to the established histologic diagnosis. RESULTS: G-MRI confirmed the diagnosis of either CD (21) or UC (7) with a sensitivity and specificity of 96% and 92%, respectively. Among the 21 patients with CD, 14 showed proximal small bowel involvement by G-MRI. In total, 17 patients were diagnosed with indeterminate colitis (IC) based on histologic criteria alone, and among these patients, G-MRI had a significantly lower non-classification rate (P < 0.02). In comparison, endoscopy was less sensitive (57%), but more specific (100%) than either histology or G-MRI in diagnosing IBD. G-MRI also showed a strong concordance with computed tomography in diagnosing CD (P = 0.001). CONCLUSION: G-MRI is a both a sensitive and specific radiologic tool in diagnosing pediatric IBD. In patients with CD, G-MRI may be useful in identifying proximal small bowel involvement. Longitudinal follow-up studies are needed in those patients diagnosed with IC to determine the predictive value of G-MRI testing.