Clinical Review
The American Journal of Gastroenterology (2006) 101, 2410–2422; doi:10.1111/j.1572-0241.2006.00840.x
Diagnostic Precision of Anti-Saccharomyces cerevisiae Antibodies and Perinuclear Antineutrophil Cytoplasmic Antibodies in Inflammatory Bowel Disease
George E Reese MRCS1, Vasilis A Constantinides MBBS1, Constantinos Simillis MBBS1, Ara W Darzi MD1, Timothy R Orchard MD2, Victor W Fazio MD3 and Paris P Tekkis MD1,3
- 1Department of Biosurgery and Surgical Technology, Imperial College London, St Mary's Hospital, London, United Kingdom
- 2Department of Gastroenterology, St Mary's Hospital, London, United Kingdom
- 3Department of Colorectal Surgery, The Cleveland Clinic Foundation, Cleveland, Ohio
Correspondence: Paris P Tekkis, MD, FRCS, Imperial College London, Department of Biosurgery and Surgical Technology, St. Mary's Hospital, 10th Floor QEQM Building, Praed Street, London W2 1NY, United Kingdom.
Received 23 March 2006; Accepted 30 May 2006.
Abstract
AIMS:
The aim of this study was to assess the diagnostic precision of antiSaccharomyces cerevisiae (ASCA) and perinuclear antineutrophil cytoplasmic antibodies (pANCA) in inflammatory bowel disease (IBD) and evaluate their discriminative ability between ulcerative colitis (UC) and Crohn's disease (CD).
METHODS:
Meta-analysis of studies reporting on ASCA and pANCA in IBD was performed. Sensitivity, specificity, and likelihood ratios (LR+, LR–) were calculated for different test combinations for CD, UC, and for IBD compared with controls. Meta-regression was used to analyze the effect of age, DNAse, colonic CD, and assay type.
RESULTS:
Sixty studies comprising 3,841 UC and 4,019 CD patients were included. The ASCA+ with pANCA– test offered the best sensitivity for CD (54.6%) with 92.8% specificity and an area under the ROC (receiver operating characteristic) curve (AUC) of 0.85 (LR+ = 6.5, LR– = 0.5). Sensitivity and specificity of pANCA+ tests for UC were 55.3% and 88.5%, respectively (AUC of 0.82; LR+ = 4.5, LR– = 0.5). Sensitivity and specificity were improved to 70.3% and 93.4% in a pediatric subgroup when combined with an ASCA– test. Meta-regression analysis showed decreased diagnostic precision of ASCA for isolated colonic CD (RDOR = 0.3).
CONCLUSIONS:
ASCA and pANCA testing are specific but not sensitive for CD and UC. It may be particularly useful for differentiating between CD and UC in the pediatric population.
