In an open randomized study, serum antibodies against Haemophilus influenzae type b capsular polysaccharide (PRP) and tetanus toxoid were determined in 146 Swedish infants; 75 of them received PRP conjugated to tetanus toxoid (PRP-T) concurrently with diphtheria toxoid vaccine, and 71 received PRP conjugated to an outer membrane complex of Neisseria meningitidis (PRP-OMP) concurrently with diphtheria-tetanus toxoid vaccine. Injections were given subcutaneously at ages 3, 5 and 12 months. One month after the second injection, the PRP-T recipients had a geometric mean (GM) concentration of 0.38 microgram/ml and only 69% had PRP antibodies > or = 0.15 microgram/ml (considered a protective level). In the PRP-OMP group the GM concentration was 0.44 microgram/ml and 85% had PRP antibodies > or = 0.15 microgram/ml. One month after the third injection, 99% of the infants in both groups had PRP antibodies > or = 0.15 microgram/ml, but PRP-T recipients had significantly higher GM concentration than infants vaccinated with PRP-OMP, 10.21 micrograms/ml vs. 1.90 micrograms/ml (P < 0.001). After all three injections the diphtheria-tetanus toxoid vaccine elicited higher GM concentrations of tetanus toxoid antibodies than did the PRP-T vaccine, but both vaccines induced antibodies above the proposed protective level, 0.01 IU/ml. The reason for the lower than expected immunogenicity of the two Haemophilus influenzae type b conjugate vaccines has yet not been established. For PRP-OMP the most probable explanation is the use of a lot of low immunogenicity, but the route of administration also has to be considered.(ABSTRACT TRUNCATED AT 250 WORDS)