The aim of this study was to evaluate hypothetical screening strategies in a Swedish celiac disease (CD) mass screening.
Of 10,041 Swedish sixth graders born in 1993 invited to a population-based CD mass screening, 7208 participated. Anti-tissue transglutaminase (tTG) immunoglobulin (Ig) A were analyzed in all children and total serum IgA (s-IgA) in 7161 children. Additional analyses of tTG-IgG, endomysial antibodies (EMA) IgA and IgG, and human leukocyte antigen (HLA) alleles were performed according to a standardized protocol. Children with elevated levels of serological markers were recommended to undergo a small intestinal biopsy to verify diagnosis, and 153 children with CD were thus identified. Sensitivity, specificity, positive predictive values (PPVs) and negative predictive values (NPVs) were calculated and receiver operating characteristic curves were plotted.
By lowering the cutoff for tTG-IgA, 17 additional cases of CD were identified at the cost of 32 biopsies. All children with tTG-IgA >50 U/mL (10 times the recommended upper limit of normal) had gluten enteropathy. Area under the receiver operating characteristic curve for tTG-IgA was 0.988. All cases carried HLA-DQ2 or HLA-DQ8, as did 53% of the controls. For different hypothetical screening strategies, sensitivity, specificity, PPV, and NPV ranged between 87.6% and 100%, 99.5% and 99.9%, 79.7% and 89.7%, and 99.7% and 100%, respectively. Efforts to increase sensitivity by lowering tTG-IgA cutoff would result in increased number of small intestinal biopsies and lower PPV. Sequential testing for both EMA and HLA-DQ genotyping would reduce the number of negative small intestinal biopsies.
tTG-IgA is a robust marker when used in CD mass screening and its performance can be enhanced by sequential testing for EMA or HLA-DQ genotyping.
*Department of Public Health and Clinical Medicine, Epidemiology and Global Health
†Department of Clinical Sciences, Pediatrics, Umeå University, Umeå
‡Department of Clinical Sciences, Pediatrics, Skånes University Hospital, Lund University, Lund
§Pediatric Clinic, Norrköping Hospital, Norrköping, Sweden.
Address correspondence and reprint requests to Olof Sandström, Department of Clinical Sciences, Pediatrics, Umeå University, Umeå 90187, Sweden (e-mail: email@example.com).
Received 4 December, 2012
Accepted 31 May, 2013
The study was funded by the European Union–supported project FP6–2005-FOOD-4B- 36383-PREVENTCD, the Swedish Research Council (grants 521–2004–7093 and 521–2007- 2953), the Swedish Research Council for Environment, Agricultural Sciences and Spatial Planning (grants 222–2004–1918 and 222–2007–1394).
The authors report no conflicts of interest.