The ABO, Lewis or P blood group phenotypes are not associated with recurrent pelvic inflammatory disease.

An assumption that ABO, Lewis, or P blood group phenotypes are associated with recurrent pelvic inflammatory disease (PID) in a similar way as with recurrent urinary tract infection has been tried to establish. Of 20 patients with PID 9 (45%) had blood type A, 6 (30%) type B, 1 (5%) type AB and 4 (20%) type O; 14 (70%) had Le(a-b+), 5 (25%) had Le(a+b-), and 1 (5%) had Le(a-b-). Of the 20 controls 10 (50%) had blood type A, 3 (15%) type B, 1 (5%) type AB and 6 (30%) type O; 12 (60%) had Le(a-b+), 4 (20%) had Le(a+b-), and 4 (20%) had Le(a-b-). The difference in the proportions of the A, B, AB, and O phenotypes as well as the proportion of combined recessive and nonsecretor phenotype Le(a+/-b-) between patients with recurrent PID and controls were not statistically significant. The distribution was consistent with that in the general population. 2 of the patient group (10%) and 6 (30%) of the controls had positive blood type P1 (Fisher's exact probability = 0.0958). The distribution of P1 between the patients and controls was opposite to that in the general population. We could not demonstrate association of ABO, Lewis or P blood group phenotypes with recurrent PID.