We studied 47 adults (21 men, 26 women), with biopsy-proven celiac sprue and anti-endomysin antibody (EmA) positivity while untreated, to evaluate the usefulness of both serologic markers of celiac sprue [i.e., immunoglobulin A (IgA)-EmA and total Ig-anti-gliadin (AGA) antibodies] and of a detailed dietary inquiry in predicting the mucosal pattern after gluten withdrawal. A second biopsy was repeated 8-30 months after beginning a gluten-free diet, along with EmA and AGA determinations and the dietary inquiry. Both EmA and AGA were appraised by indirect immunofluorescence on monkey esophagus and rat kidney, respectively. Intestinal biopsy was graded according to Cooke's criteria. After gluten withdrawal, intestinal mucosa reverted to normal in only nine patients. Both EmA and AGA showed high positive but low negative predictive values on intestinal histologic outcome. The positive predictive value of admission of dietary lapses was 100%, whereas the negative predictive value was 39.1%. Neither serologic markers nor dietary inquiries are to be regarded as reliable predictors of intestinal outcome after a gluten-free diet. Biopsy remains the best means of ascertaining mucosal recovery.