Three large multi-center randomized double-blind placebo controlled clinical trials in the U.S. have observed and published variable results following IVIG prophylaxis for late-onset infection in the high risk neonate.Staphylococcus epidermidis was the major cause of infection in these studies. We hypothesized that these differences might in part be due to variable S. epidermidis antibody activity contained in the IVIG lots infused. We were able to obtain bottles of each of the twelve original lots of IVIG used in these trials. All IVIG lots were diluted to 5 grams% with normal saline. Each was then. evaluated to determine the amount of antibody present against a clinical (ATCC) strain of S. epidermidis (Haywood, serotype 2) using a modification of an in-vitro assay of neutrophil mediated opsonophagocytosis previously reported (Fischer et al, Journal of Infectious Diseases 1994;169:324-9). Antibody levels are the geometric mean of two assays and represent the largest reciprocal dilution which promoted ≥ 90% bacterial killing. No bacterial killing was noted in negative controls which were without IVIG, complement and both. The total IgG, measured by radial immunodiffusion, was similar for all the samples tested. The S. epidermidis opsonic activity of the IVIG lots ranged from undetectable to 20-1. Two of the IVIG lots (one with opsonic activity at 20-1 and the other at 5-1) and normal saline were then evaluated for in-vivo activity against the same strain of S. epidermidis by determining the per cent survival of lethally infected suckling rats (Fischer et al, ibid.) treated with each lot. Suckling rat survival was significantly increased (p<0.02) in animals receiving IVIG with high titers (68%, 52 of 76) versus low titers (47%, 36 of 76) ofS. epidermidis activity. Survival of pups treated with normal saline(41%, 29 of 70) were not significantly different (p>0.5) than those treated with IVIG containing low levels of S. epidermidis opsonic activity. We conclude that the IVIG lots used in the clinical trials evaluating IVIG prophylaxis in the high risk neonate had significantly variable S. epidermidis antibody activity. Further analysis will be necessary to correlate the clinical and antibody variability. If so, the development of screened or selected products with high titers of activity against neonatal specific pathogens would be warranted.