In 1973, Ruth Bishop and her colleagues in Melbourne were the first to identify rotavirus gastroenteritis in infants. Rotavirus has since been recognised as the most frequent cause of severe childhood gastroenteritis worldwide. In temperate climates, infection occurs predominantly in winter and spring.

The spectrum of disease ranges from severe gastroenteritis to mild or clinically inapparent infection. Reinfection is common. In Australia, the peak age of hospitalisation for rotavirus infection is 6–24 months; however, in Indigenous children, infection occurs earlier and is more severe.1 About half of the 20 000 annual hospitalisations for acute gastroenteritis in Australian children under 5 years of age are attributed to rotavirus.2

The first rotavirus vaccine licensed in the United States (RotaShield [Wyeth, Madison, NJ, USA]) was withdrawn from use in 1999, after barely 9 months, due to an increased occurrence of intussusception after vaccination (attributable risk, one case per 10 000 vaccine recipients). Two other live attenuated oral vaccines (RotaTeq [Merck, Whitehouse Station, NJ, USA], a pentavalent human–bovine reassortant vaccine; and Rotarix [GlaxoSmithKline, Rixensart, Belgium], a human monovalent vaccine) have recently been approved for use in many countries after extensive clinical trials showed them to be effective against the most common circulating strains, and found no evidence of an association with intussusception.3,4

The RotaTeq vaccine was introduced into the US vaccination schedule in February 2006. A report from a national network of sentinel laboratories, published in June 20085 and updated in October 2008,6,7 shows that the 2007–2008 US rotavirus season was significantly delayed, shortened and diminished compared with the 2000–2006 seasons. Only 4% of children aged under 3 years who were hospitalised with acute gastroenteritis had rotavirus-positive stools in 2008 compared with 56% in 2006, and only 9% of emergency room patients with acute gastroenteritis tested positive for rotavirus compared with 58% in 2006. Reductions in hospitalisation of up to 85% for rotavirus gastroenteritis and 56% for all-cause diarrhoea in young children were also reported, suggesting very large savings in hospital costs.6 Although these findings need confirmation over future seasons, this rapid vaccine impact is particularly striking given that, at the time, the estimated vaccine coverage was just 56% for one dose in infants aged 3 months and 34% for three doses by age 13 months.5

Reassuringly, preliminary data from surveillance in the US over the first 19 months of its rotavirus program, when over 9 million doses of vaccine were distributed, do not indicate any association of RotaTeq with intussusception or other serious adverse events.8

In Australia, rotavirus vaccination commenced in the Northern Territory in October 2006 and in the remaining jurisdictions in July 2007. Rotarix is used in the NT, New South Wales, Tasmania and the Australian Capital Territory, and was used in Western Australia until May 2009. RotaTeq is used in Victoria, South Australia and Queensland. This geographical split results in about half of the birth cohort receiving Rotarix (two-dose schedule) and half receiving RotaTeq (three-dose schedule). Program implementation has been faster in Australia than in the US, with an estimated coverage, by December 2008, of 87% for at least one dose of vaccine received by 4 months of age, and 84% for a full vaccine course (either two or three doses) received by 13 months of age (Mr Brynley Hull, Epidemiologist, National Centre for Immunisation Research and Surveillance, personal communication).

Australia is in a unique position to assess the impact of the two vaccines because of the geographical split. However, it is important to recognise that the prevailing strains of rotavirus vary by region and change unpredictably from year to year.9 As the composition of the two licensed vaccines differs, effectiveness against certain strains, such as those not contained in the vaccines, may vary. An early study during a rotavirus outbreak in the NT showed that Rotarix was effective against the lately emerged G9 strain.10 Although Australia has an established national strain surveillance system,9 it needs to be enhanced, as samples for strain typing have not always been representative of all geographical areas or accompanied by clinical data. National disease notification using laboratory-confirmed cases needs to be implemented, and further studies of field effectiveness should be undertaken.

Adverse events following vaccination in Australia are reported and reviewed nationally. The vaccines have not been associated with an excess of serious adverse events during the first 6 months of the program.11 Additionally, all cases of any-cause intussusception are reported by paediatricians to the Australian Paediatric Surveillance Unit, with enhanced active surveillance for intussusception being undertaken in a pilot program.

Although efficacy of both vaccines persisted in clinical trials for 2 years,12,13 the full extent and duration of vaccine effectiveness, the degree of herd immunity, and whether a shift in disease incidence to older age groups has occurred, are yet to be determined. Identifying the emergence of new strains of rotavirus, either spontaneously or under vaccine pressure, will also be essential. Successful implementation of rotavirus vaccination in developing countries — where more than 95% of the world’s annual rotavirus deaths occur — is also much anticipated, possibly using new, less costly vaccines.7

In this issue of the Journal, Lambert and colleagues14 report on rotavirus notifications and laboratory tests performed in Queensland before and after the introduction of an infant rotavirus vaccination program. This is the first published evidence of a substantial population-based decline in rotavirus disease activity in Australia since introduction of the vaccine. Furthermore, laboratory-confirmed rotavirus infection appears to have declined not only in the cohort eligible for vaccination (infants aged under 2 years), but also in older children and in adults, suggesting a herd immunity effect, as has been observed in preliminary data from the US.7

Much points to the potential for Australia’s rotavirus vaccine program to have a substantial impact on acute childhood gastroenteritis, both in terms of health care utilisation and broader social and economic effects. It is important to document this with well targeted research and surveillance.