Jonas F Ludvigsson1,2,3,4, Simon Jarrick2, Joseph A Murray5, Louise Emilsson6,7,8. 1. Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm. 2. Department of Paediatrics, Örebro University Hospital, Örebro. 3. Division of Epidemiology and Public Health, School of Medicine, University of Nottingham, City Hospital, Nottingham, UK. 4. Department of Medicine, Columbia University College of Physicians and Surgeons, New York, NY. 5. Department of Immunology, Division of Gastroenterology and Hepatology, Mayo Clinic, Rochester, MN. 6. Department of Health Management and Health Economy, Institute of Health and Society, University of Oslo, Oslo, Norway. 7. Department of Epidemiology, Harvard T.H. Chan School of Public Health, Boston, MA. 8. Primary Care Research Unit, Vårdcentralen Värmlands Nysäter, Värmland County, Sweden.
Abstract
BACKGROUND AND AIMS: A recent study found a 10-fold increased risk of celiac disease (CD) in individuals with Henoch-Schonlein purpura (HSP), but the confidence interval (CI) was wide. METHODS: The retrospective cohort study of all patients with CD in Sweden, diagnosed through small intestinal biopsy from 1969 to 2008 (n=29,077). Each individual with CD was matched to up to 5 controls (n=144,433). Data on study participants were linked to diagnostic codes for HSP in the National Patient Registry. Through Cox regression we estimated hazard ratios for CD and later HSP. Through logistic regression we calculated odds ratios for HSP preceding CD. RESULTS: During follow-up 19 individuals with CD and 99 controls developed HSP. This corresponded to a hazard ratio of 0.96 (95% CI, 0.59-1.56). Looking backward, we found no increased risk of earlier HSP in patients with CD (odds ratio=1.02; 95% CI, 0.60-1.72). CONCLUSIONS: In this study of more than 29,000 patients with CD, we found no increased risk of HSP before or after CD.
BACKGROUND AND AIMS: A recent study found a 10-fold increased risk of celiac disease (CD) in individuals with Henoch-Schonlein purpura (HSP), but the confidence interval (CI) was wide. METHODS: The retrospective cohort study of all patients with CD in Sweden, diagnosed through small intestinal biopsy from 1969 to 2008 (n=29,077). Each individual with CD was matched to up to 5 controls (n=144,433). Data on study participants were linked to diagnostic codes for HSP in the National Patient Registry. Through Cox regression we estimated hazard ratios for CD and later HSP. Through logistic regression we calculated odds ratios for HSP preceding CD. RESULTS: During follow-up 19 individuals with CD and 99 controls developed HSP. This corresponded to a hazard ratio of 0.96 (95% CI, 0.59-1.56). Looking backward, we found no increased risk of earlier HSP in patients with CD (odds ratio=1.02; 95% CI, 0.60-1.72). CONCLUSIONS: In this study of more than 29,000 patients with CD, we found no increased risk of HSP before or after CD.