Following the 1997 outbreak of EV71 associated HFMD in Sarawak, Malaysia, the Health Department installed a sentinel surveillance programme with the expectation that we would be able to study epidemiological trends and begin to predict when to expect outbreaks with sufficient accuracy in order to implement public health interventions to reduce the burden of the disease. Although the surveillance programme is still ongoing in Sarawak, we have sought to glean some preliminary information from the data generated over the first 7 years of the programme. During this time there have been 2 outbreaks of HEV71 in Sarawak with smaller clusters of HFMD associated with CVA16 and other species A human enteroviruses occurring concomitantly with as well as independently of the two HEV71 outbreaks. A recent report on a similar surveillance programme in Yamagata Prefecture in Japan (1998–2003) suggests that in Yamagata there is frequent importation of HEV71 from surrounding countries seeding the clusters of cases seen annually in this community. The HEV71 strains in this study were isolated from small clusters of cases that tended to be seen in the summer months while in our situation we observed outbreaks of HEV71 every 3 years with cases being seen much earlier in the year, well before the northern summer. In 2003 both Sarawak and Yamagata experienced a large outbreak and in both situations, a genogroup shift from C to B was noted.
It is interesting that in Sarawak, of the genogroup C viruses, only genogroup C1 strains have been observed, while genogroup B viruses appear to be changing from outbreak to outbreak, suggesting that it is likely that genogroup B viruses are evolving within Borneo and that the outbreaks we have experienced are being seeded from within rather than from imported viruses. Since the outbreaks in Sarawak typically begin early in the year, it is also possible that genogroup B strains generated in Sarawak may seed HEV71 outbreaks in the region, which typically occur later than the Sarawak outbreaks. This temporal sequence of regional outbreaks is also true of those occurring in Singapore and in Peninsula Malaysia.
The data we have obtained through 7 years of our sentinel surveillance programme for HFMD in Sarawak have provided useful clues to understanding the epidemiology of HEV71 in the state. It is clear that the appearance of HEV71 associated HFMD in sentinel clinics signals the start of an outbreak, but the rise in the number of cases is so rapid that this approach is not a suitable early warning system. In 2003 there were only 5 weeks between the time the first HEV71 cases were seen and the peak of the outbreak. Clearly this could be explained by rapid and effective response by the public health teams, but we have no way to know.
Alternatively, the 3-year cycle of HEV71 outbreaks we have observed could, if verified in the coming years, provide public health officials with the relevant information to plan and to implement their intervention programmes to reduce the disease burden in the years when an HEV71 outbreak is expected. Although this is not expected to prevent the outbreaks entirely, effective public health measures put into place early enough can limit the spread, reduce mortality and reduce the burden on the community and the health system.
It is important to note that epidemiological curves showing HFMD alone, without distinguishing the infecting agent for each case, can stretch broadly over many months, with non-HEV71 enteroviruses continuing to be isolated after cessation of HEV71 activity. This was especially evident in 2000, when HEV71 associated fatal cases were reported in neighbouring Singapore in September and October 2000, and the media attention surrounding these events generated a high index of suspicion in Sarawak as well. No HEV71 was isolated in Sarawak after August that year, but numerous CVA16 continued to be isolated until the end of 2000. Thus even though sociological factors affect the shape of the HFMD epidemiological curves in Sarawak, epidemiological curves specifically showing genogroup B strains of HEV71 were consistently sharp and well defined in 2000 and 2003.
The mean and median age of children with HFMD was 36 months and 30 months respectively, but the mean ages did not differ between the groups infected with the different serotypes. It is thus intriguing that HEV71 has caused much larger and sharper outbreaks than either CVA16 or CVA10. This suggests that HEV71 has the capacity to spread rapidly through the susceptible population and then become quiescent in the community. In the third year after any HEV71 outbreak, the whole cohort of children under 3 years of age has not been exposed to HEV71 and all of these children are then susceptible, providing the conditions for another sweeping transmission of HEV71 through the community. The annual birth cohort in Sarawak is 48 to 49 thousand and thus in 3 years there are up to 150,000 susceptible children in the state.
According to the trends we have reported, we expect that the next outbreak of HEV71 in Sarawak will be in 2006. At the time of writing we have already begun to pick up HEV71 cases in our sentinel programme and from past experience, an outbreak in Sarawak is often followed by outbreaks in other countries in the region. We have therefore decided to put our data into the public domain in order that other public health practitioners in the Asia Pacific region may benefit from this experience and prepare for a spread of HEV71 in the region once again in the months to come.