In patients at risk for sepsis after cardiac surgery, the efficacy of intravenous immunoglobulin (Ig) treatment was compared with a historical control population, equivalent in patient characteristics and disease severity. Using APACHE II scores, especially in the high-risk group (IgG), we could discriminate between low-risk patients (score < 19; mortality 1 percent) and the small groups at risk (score 19 to 23) and high risk (score > or = 24) with a significantly higher mortality (14 percent and 76 percent, respectively) [corrected]. Subsequently, among 1,341 consecutive patients we prospectively identified and treated (IgG n = 41 IgGMA: n = 25) these at-risk groups. In contrast to controls (risk: n = 21; high-risk; n = 21), we found a marked fall in APACHE II scores, especially in the high-risk group (IgG, n = 26: p < 0.05; IgGMA, n = 13: p = 0.08) [corrected]. In this group, Ig therapy produced higher (p < 0.05) response rates (score decrease 7 within 4 days: IgG: 54 percent, IgGMA: 62 percent; controls: 19 percent) and reduced mortality (IgG: 46 percent, IgGMA: 46 percent; controls: 76 percent), statistically significant (p < 0.05) for Ig treatment overall. Thus, early Ig treatment improves disease severity and may improve prognosis in prospectively score-identified high-risk postcardiac surgical patients.