Acute otitis media is an extremely common disease, most children being subject to it once or more during their first 5 years of life. Thirty to forty per cent of episodes of acute otitis media are caused by Streptococcus pneumoniae. Resistance of pneumococci to antibiotics is an increasing problem particularly in France. Since 1983 a 23-valent pneumococcal polysaccharide vaccine has been available, containing in particular four serotypes that carry penicillin resistance. Unfortunately, this vaccine is therefore not fully effective until after the age of 2 years. In spite of good serologic responses, clinical results have been variable. Some found that this diminished frequency only lasted for several months and ended as serotypes not included in the 23-valent vaccines began to appear. Since 1992, studies have been carried out using different pneumococcal vaccines conjugated to diphtheria, tetanus or meningococcal proteins. Despite the incomplete nature of the data, it can already be confirmed that these vaccines are immunogenic from the very first months of life onwards. Numerous problems nevertheless remain unsolved. How many serotypes can be included in the conjugate vaccine? For the moment it seems technically impossible to conjugate all 23 serotypes of the current vaccine. While Haemophilus influenzae type b is principally responsible for invasive infections, it is traditionally said to play only a minor role in otitis, which is usually provoked by non-capsulated Haemophilus strains beyond the vaccine's reach. In fact, in most studies it has been observed that 2--5% of cases of Haemophilus otitis are due to Haemophilus type b. The vaccine could be said, therefore, to have a minor but by no means negligible role in the prevention of Haemophilus otitis. In conclusion, pneumococcal and haemophilus vaccines have not revolutionized the preventive treatment of otitis for the moment, but great hopes are still placed on the conjugate pneumococcal vaccine.