The old combined A and C meningitis vaccine was a "polysaccharide", made of a bit of the coat that forms the bacterial wall of Neisseria meningitidis strain C. Like the old Hib meningitis polysaccharide vaccine, it doesn't work very well. So Wyeth created a "conjugated" version, which would marry the polysaccharide of the strain C bug with CRM197, a diphtheria toxin, which has, as far as the vaccine community sees it, a proven track record for stimulating the immune system and getting it to work.

By piggybacking C strain cells onto a proven entity, the idea, says Wyeth's David Hall, is that the CRM cells put the body on red alert that it is under attack. Under this heightened situation, it is more likely then to recognise the polysaccharide cells as foreign and worthy of developing antibodies and memory cells, which help the body remember how to generate specific antibodies in the future if needed.At least that's the theory. Nevertheless, numerous studies show substantial vaccine failure rates among the supposedly successful Hib conjugate vaccine. One study showed the protective effect of one conjugate Hib vaccine was only 74 per cent (Lancet, 1991; 338: 395-8), and only 35 per cent among Alaskan infants, who are at higher risk of Hib meningitis than other children.

Another study showed that 52 per cent of Hib meningitis cases occurred among vaccinated children (Pediatrics, 1995; 96: 424-7).

And has the Hib vaccine eliminated Hib meningitis? It's true there's been a drastic drop in cases. But that's consistent with previously identified cyclic variation in this disease, with certain peaks and troughs in its incidence (JAMA, 1993; 269: 227-31).

Furthermore, the incidence has also fallen among babies who haven't been vaccinated (Lancet, 1997; 349: 1197-1201). The supposed Hib vaccine success story could be nothing more than the cyclical downturn in a disease. It's too early to tell.