Inflammatory bowel disease (IBD) activity during pregnancy is variable and factors influencing the course unknown. We studied the relationship between maternal-fetal HLA disparity and IBD course.Women with IBD and childbirth were recruited and medical records were reviewed for five time periods. Twenty milliliters of blood was obtained from mother and child for genotyping. Each mother/child pair was assigned an HLA disparity status (+/-) for HLA A, B, C, DRB1, and DQ loci. Odds ratios were calculated comparing HLA disparity in women whose IBD improved versus those whose disease worsened or remained active.Fifty pregnancies in 38 women were studied. Forty-two of 50 pregnancies (84%) were disparate at the DRB1 locus; 34 (68%) were mismatched at the DQ locus. There was no difference in average disease score or overall activity score based on DRB1 or DQ disparity (p > 0.05 for all comparisons). There were 31 pregnancies disparate at both DRB1 and DQ loci; a significant difference was found in average disease scores and overall activity scores between women mismatched at both loci versus only one or neither locus (OR 8.4 [1.5-14, p = 0.01). Logistic regression identified prepartum disease activity and disparity at both DRB1 and DQ as significant predictors of overall disease activity during pregnancy.Improvement of IBD symptoms during pregnancy is associated with disparity in HLA class II antigens between mother and fetus. This suggests that the maternal immune response to paternal HLA antigens plays a role in pregnancy-induced remission. What is accepted and what this research adds are as follows: . The course of IBD during pregnancy is variable. . The factors involved with disease course are unknown. . The data presented here provides a scientific mechanism for disease course during pregnancy. . This is a novel work and it corroborates what has been seen in other autoimmune conditions.