To determine whether an IgM-positive crossmatch adversely affects the results of heart transplantation, we conducted a retrospective study of 125 orthotopic heart transplant recipients. A direct donor-recipient crossmatch was performed retrospectively on sera from all patients by the standard National Institutes of Health (NIH) method and the antihuman globulin (AHG) procedure. The patients were then divided into 3 groups as follows. Group 1 comprised 110 patients with a negative NIH and AHG crossmatch (control group). Group 2 comprised 5 patients with a positive NIH crossmatch and a negative AHG crossmatch. Group 3 comprised 10 patients with positive NIH and AHG crossmatches. All positive crossmatches in group 3 patients converted to negative after treatment of sera with dithioerythritol, indicating that the initial result was due to IgM antibodies. All patients received standard immunosuppressive treatment. An IgM-positive crossmatch did not affect the number or severity of rejection episodes among the 3 groups, nor did it have an effect on the incidence of infection. Whereas coronary artery disease was detected by angiography in 16 of 110 patients (14.6%) in group 1 and in 1 of 10 patients (10%) in group 3 (P = NS), no patient in group 2 was affected. Actuarial survival at 1 and 2 years post-transplant was significantly better for patients with an IgM-positive crossmatch (group 2) (100% survival at 2 years) than for patients with a negative crossmatch (group 1) (73% at 1 year and 71% at 2 years, P < 0.05). Based on our study, the effect of an IgM-positive crossmatch on survival is difficult to interpret because of the small sample size. An IgM-positive crossmatch, however;did not appear to have a deleterious effect on survival. It may be that the IgM antibody has an immunoregulatory role. A larger series of patients with positive crossmatches and longer follow-up will be necessary to evaluate the importance of these results.