Celiac.com 06/03/2010 - Clinical presentation of celiac disease can vary considerably from patient to patient. Most patients with celiac disease present atypical symptoms. Moreover, most patients who present abdominal symptoms in primary care do not have celiac disease, and so diagnostic tests for celiac disease are not necessary and should be avoided.
A team of researchers recently conducted a systematic review of diagnostic testing for celiac disease among patients with abdominal symptoms.
The team included Daniëlle A. W. M. van der Windt, PhD; Petra Jellema, PhD; Chris J. Mulder, MD, PhD; C. M. Frank Kneepkens, MD, PhD; and Henriëtte E. van der Horst, MD, PhD. Their article appears in the Journal of the American Medical Association.
The goal of the research was to review and summarize evidence on the performance of diagnostic tests for spotting celiac disease in adults who present abdominal symptoms in primary care or similar settings.
To obtain initial data, the team search MEDLINE (from January 1966 through December 2009, and EMBASE from January 1947 through December 2009. They also conducted a physical search of references for additional relevant studies.
The team chose cohort or nested case-control diagnostic studies which included adults presenting non-acute abdominal symptoms, which featured celiac disease prevalence of 15% or less, and in which the tests included gastrointestinal symptoms or serum antibody screens.
Two independent reviewers conducted studies tool and data extraction. They then calculated sensitivities and specificities for each study and computed pooled estimates using bivariate analysis where there was clinical and statistical homogeneity.
In all, the team included sixteen studies encompassing 6085 cases in their review.
Specificity, sensitivity, and confidence intervals for predicting celiac disease varied with abdominal symptoms. For patients presenting with classic diarrhea, for example, predictive sensitivity ranged from 0.27 to 0.86, while specificity ranged from 0.21 to 0.86.
Pool estimates for 8 studies on IgA antiendomysial antibodies were 0.90, with a 95% confidence interval [CI] (0.80-0.95) for sensitivity and 0.99, with a 95% CI (0.98-1.00) for specificity, with a positive likelihood ratio [LR] of 171 and negative LR of 0.11.
Pool estimates for IgA antitissue transglutaminase antibodies (7 studies) were 0.89, with a 95% CI (0.82-0.94) and 0.98 at 95% CI (0.95-0.99), respectively, with a positive LR of 37.7 and negative LR of 0.11.
IgA and IgG antigliadin antibodies showed variable results, especially for sensitivity, which ranged from 0.46-0.87 for IgA, and from 0.25-0.93 for IgG.
One recent study using deamidated gliadin peptides showed good specificity (0.94), but the target population offered limited supporting evidence.
For adults who present abdominal symptoms in primary care or other unscreened settings, IgA antitissue transglutaminase antibodies and IgA antiendomysial antibodies offer high sensitivity and specificity for diagnosing celiac disease.
SOURCE: JAMA. 2010;303(17):1738-1746. doi:10.1001/jama.2010.549