Meta-analysis: coeliac disease and hypertransaminasaemia

Abstract

There may be a positive association between coeliac disease and serum hypertransaminasaemia but evidence is conflicting. To conduct a systematic review and meta-analysis to determine the prevalence of coeliac disease in adults presenting with cryptogenic serum hypertransaminasaemia and the prevalence of hypertransaminasaemia in patients with newly diagnosed coeliac disease. MEDLINE and EMBASE were searched up to August 2010. Case series and case-control studies recruiting adults with either cryptogenic hypertransaminasaemia that applied serological tests for coeliac disease and/or distal duodenal biopsy to participants or newly diagnosed biopsy-proven coeliac disease that assessed serum transaminases were eligible. The pooled prevalence of coeliac disease in individuals presenting with abnormal serum transaminases and the pooled prevalence of hypertransaminasaemia in newly diagnosed coeliac disease were calculated with 95% confidence intervals (CI). Eleven eligible studies were identified. Pooled prevalences of positive coeliac serology and biopsy-proven coeliac disease in cryptogenic hypertransaminasaemia were 6% (95% CI 3% to 10%) and 4% (95% CI 1% to 7%) respectively. Pooled prevalence of abnormal serum transaminases in newly diagnosed coeliac disease was 27% (95% CI 13% to 44%). Exclusion of gluten led to normalisation of serum transaminase levels in 63% to 90% of patients within 1 year. Undetected coeliac disease is a potential cause for cryptogenic hypertransaminasaemia in 3% to 4% of cases. More than 20% of individuals with newly diagnosed coeliac disease may have abnormal serum transaminases and these normalise on a gluten-free diet in the majority of cases.