| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Reviews are available about how climate change and climate variability are likely to affect health, drawing on empirical studies that document past and present risks, and predictive models that conjecture future risks.1–5 However, most of the published work focuses on directly acting temperature effects (e.g., excess mortality and morbidity due to heat waves, floods, and droughts), effects on the risk of disasters and malnutrition, and changes in the transmission of infectious diseases.1–5 With regard to a changing climate and infectious diseases, most studies have centered on malaria.10–15 Time-series analysis and predictions suggest that the population at risk of malaria in Africa will slightly increase due to rising temperatures, primarily through expansion of the disease into higher altitudes, and lengthened malaria transmission seasons.4 The potential impact of climate change on the global distribution of dengue has also been modeled. Under the scenario of contributing factors other than temperature remaining unchanged, the models predict that a large proportion of the human population would be at risk.16
Only a few attempts have been made to predict changes in the spatial distribution of schistosomiasis transmission due to global warming; results have been conflicting.17,18 Although an early model of global warming predicted that the area conducive for schistosomiasis transmission would expand,17 later models forecasted a decrease in the epidemic potential of schistosomiasis.18 Although the nature and extent of climate change on the transmission of schistosomiasis remain poorly understood,19 there is consensus that the most sensitive areas are around the borders of the current transmission.2 Clearly, new research is warranted to develop regional climate change models and to assess the biologic significance of model outcomes.4,15
Despite huge efforts in implementing and sustaining the national schistosomiasis control program in China, recent data suggest that the disease is re-emerging.20–22 Regional climate change in the face of profound demographic, ecologic, and socioeconomic transformations has been advanced as a contributing factor.20,21,23 Here, we present the results from a biology-based model developed with an emphasis on the effects of rising temperature on the future transmission of schistosomiasis in China.
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Biologic model and experiments. With the aim of assessing the effect of temperature on different critical stages of the life cycle of Schistosoma japonicum, we have developed a biologic model of the parasite and its intermediate host. Some of the data needed to generate this model were available in the literature, while additional laboratory and field experiments had to be carried out to fill some of the current data gaps.
Adult Oncomelania hupensis, the intermediate host snail of S. japonicum in China,24 were collected in November 2001 from the marshlands in Xinba, Jiangsu province (geographical coordinates: 119.53° E longitude, 32.28° N latitude). Snails were transferred to the laboratory, kept at 25°C and tested for S. japonicum-cercarial shedding after 2 and 4 weeks. All snails were non-infected, and hence they were used for assessing the effect of temperature on hibernation of O. hupensis as described elsewhere.24 In brief, groups of 30 snails were placed in Petri dishes and transferred to a colder environment (temperature: 13°C). After acclimatization for 2 days, these snails were subjected to constant temperature reductions at a rate of either 0.5 or 1°C per day. Each day, snails with a closed operculum and/or lack of movement were tested by pinching the operculum and the foot-head portion. Snails that showed no reaction were transferred to de-chlorinated water at a temperature of 13°C and kept for 2 hours. When activity resumed, snails were considered in hibernation state before observation by pinching. The remaining snails were subjected to further temperature decreases until hibernation took place. The temperature when this occurred was recorded for each snail.
The duration of a single O. hupensis generation was investigated under quasi-natural conditions. In April 2001, 200 adult O. hupensis were placed in a container containing mud and covered with a fine-meshed net in open breeding sites in Wuxi, China. As soon as eggs were discovered on the mud, the adult snails were removed and the development of eggs into young adult snails (F1) followed. Shortly before reaching maturity, the snails were removed, sexed, and couples transferred into separate containers kept in the same open breeding sites. Containers were checked daily for the presence of eggs. The duration from the first discovery of eggs from mother snails until eggs were laid by the F1 generation was considered a full development period. The ambient temperature was measured hourly throughout the study 1.5 m above ground level by a thermometer (Model ZJ1-2B, Shanghai).
In the third experiment, 750 infection-free O. hupensis were collectively exposed to 
15,000 S. japonicum miracidia (Wuxi isolate) for 4 hours at a temperature of 25°C. These snails were divided into five equally sized groups, placed in culture boxes, and raised at temperatures of 18, 21, 24, 27, or 30°C, respectively. An additional group of 150 snails was left uninfected and kept at a temperature of 24°C. The culture boxes were checked daily and dead snails were removed and counted. The initial cercarial shedding test was done after 30 days (snails kept at 30°C) or 70 days (18°C). Snails that shed cercariae were removed and counted. The remaining snails were tested again 5 days later. This procedure was repeated until no cercariae were released from snails after three consecutive tests. Snails that failed to shed cercariae were dissected to assess their infection status.
Statistical analysis and predictive modeling. For statistical analysis and predictive modeling we used SAS version 8.0 (SAS Institute Inc., NC). A probit analysis was used to establish the relationship between temperature (T) and the hibernation rate of O. hupensis (h), as expressed by Equation 1.
 |
where probit (h) is the probability of h, logT is the natural logarithm of T, a is the intercept, and b the regression coefficient.
The temperature at which 50% of snails were in hibernation (ET 50) was considered as the lowest temperature (Th) for snail development, as shown in Equation 2.
 |
We assumed that the temperature at which the development of S. japonicum within O. hupensis is arrested (T0) can be estimated from a regression model established between the development rate (d) and the temperature during the prepatent period, which was obtained from the exposure experiment. According to Equation 3, d was estimated as 1 divided by the average prepatent period (N), which is the time from snail exposure to S. japonicum cercariae until parasites are shed by the snail, as follows:
 |
The accumulated degree-days (ADD) was calculated as the difference between the mean daily temperature (Tmean) and T0 summed over the prepatent period of S. japonicum within O. hupensis (ADDS.j.), and the development period for O. hupensis (ADD O.h.). Mean values of ADD were calculated according to Equations 4 and 5, based on the experiments carried out under quasi-natural conditions, with units expressed as degree-days.
 |
 |
For each observing station ,i the ADDS.j. and ADDo.h. were calculated using the respective temperature data and the predicted temperature increases (Tp) for China by 2030 (+1.7°C) and 2050 (+2.2°C)25 according to Equations 6 and 7:
 |
 |
The potential transmission index (PTI) was calculated for each of the 193 observing stations i in 2030 and 2050, according to Equation 8. Only those PTIi values above 1 were considered to be of relevance (i.e., where S. japonicum transmission potentially can occur).
 |
The PTI for S. japonicum and O. hupensis was calculated from Equations 9 and 10, respectively.
 |
 |
Historical data suggest that the distribution of O. hupensis in China is restricted by the mean January temperature because snails have not been recorded in areas where the temperature falls below 0°C, the threshold termed the "freezing line."23,26 Using a time-series analytical approach for the period 1951–2000 and the predicted temperature increases for the years 2030 and 2050, we mapped the geographic distribution of the freezing line for these two time points. We used an autoregressive integrated moving average (ARIMA) model27,28 for each observing station. Because our recent time-series analysis of the monthly mean temperature in Jiangsu province showed strong seasonality with periodicity of 12 months,29 we developed a seasonal ARIMA(p,d,q)(P,D,Q)12 model. The model with the lowest Akaike’s information criterion (AIC)30 was used for prediction at each station.
Risk mapping.
For mapping purposes, we used a kriging approach of the spatial analyst model, employing ArcGIS software version 8.3 (ESRI, Redlands, CA). The ordinary kriging model used is as follows: Z(s)= µ+ 
(s), where (s) is the vector of locations and Z(s) is the vector of values at the respective locations. This model is based on a constant mean, µ, for the data (no trend), and random errors, (s) assuming spatial dependence.31
We adhered to the following six-step procedure. First, we selected the most suitable order to carry out the ordinary kriging analysis and developed the prediction map based on PTIi for each observing station. Second, we generated the current distribution of PTIi, by taking into account the distribution of O. hupensis and S. japonicum based on the temperature requirements of the intermediate host snail and the parasite to develop within the snail, and the freezing line for the year 2000. Third, these data were fed into a geographic information system (GIS) and used to delineate the potential schistosomiasis transmission area. Fourth, we produced a map with the prediction error (i.e., uncertainty of the prediction). Fifth, a validation of the predicted schistosomiasis transmission risk in 2000 was performed. We used an agreement test estimated by the Kappa coefficient, employing 239 village-level data extracted from the national sampling data in 2004 and 500 points sampled at random from non-endemic areas.32,33 Sixth, we produced predictive maps for the transmission of S. japonicum in the years 2030 and 2050, which combines the predicted PTIi and the predicted freezing lines for the respective years.
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
depicts schematically a biologic model emphasizing components of the S. japonicum life cycle that are influenced by temperature. Based on hypothesized future temperatures the aggregation of specific effects on the intermediate host snail, as well as the parasite itself, was then used for predictive modeling (i.e., mapping of the spatial distribution of the potential schistosomiasis transmission areas) in China in the years 2030 and 2050.
|
2 = 11.6, P = 0.394). The temperature at which half of the snails tested were found in a state of hibernation (ET50) was 5.8°C (95% confidence interval [CI]: 5.5–6.2°C). Subsequently, this value was used for estimating ADDO.h., a measure of the thermal energy needed for the development of O. hupensis.
Our laboratory investigations revealed a strong association between temperature and the prepatent period of S. japonicum within O. hupensis (Table 1). At the lowest temperature investigated (21°C), the prepatent period was 128.9 ± 16.1 days, whereas it was halved (62.7 ± 14.2 days) at a temperature of 30°C.
|
Table 2 summarizes the results from the O. hupensis development experiments carried out under quasi-natural conditions. Overall, 63 snail pairs were monitored and the minimum and maximum duration to complete a full generation was 200 and 385 days, respectively. The observed mean of the ADDO.h. to complete a single generation was 3846.3 degree-days.
|
Table 3 summarizes the historical mean January temperatures derived from 193 stations across China from 1960 to 2000. These data suggest that the median of the average January temperature over the mainland of China has significantly increased by 0.9°C over the past 40 years (Friedman test, 2 = 478.0, P < 0.001). The median of the predicted average January temperatures for 2030 and 2050 are –3.8°C and –3.1°C, respectively. Compared with 2000, these medians translate to increases of 0.9 and 1.6°C (P < 0.001).
|
depicts the risk map for schistosomiasis transmission and the respective prediction error distribution map in China as of 2000, taking into account the various critical temperatures that govern the transmission cycle of the disease. We found a good agreement between the predicted risk map for schistosomiasis transmission and the parasitological data from the national sampling survey in 2004; the validation agreement test showed a Kappa coefficient of 0.70 (P < 0.001). It should be noted, however that the current distribution map of schistosomiasis in China20 is influenced by past and ongoing control efforts, as shown in a recent time-series analysis of the spatio-temporal distribution of S. japonicum in Jiangsu province.34
|
. Comparison with the situation in 2000 reveals that the potential transmission areas of the disease will significantly expand; we estimate that by 2030 an additional 662,373 km2, and by 2050 an estimated 783,883 km2, might become potential risk areas for schistosomiasis transmission. Disease transmission is thus likely to occur in previously non-endemic areas, such as the southern parts of Shandong and Henan provinces. Interestingly, there is another part in north-western China where disease transmission is predicted on the basis of the PTIi. However, taking into account the predicted spatial distribution of the freezing line, it is unlikely that disease transmission can occur north of this freezing line, and hence the northwestern part of China is likely to remain schistosome-free.
|
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our attempt to assess the potential impact of climate change on the transmission of schistosomiasis in China was facilitated by the development of a biologic model, which was conceived by readily available data from the literature and supplemented by new data obtained from laboratory and field investigations. Emphasis was placed on the effect of temperature on individual components of the disease transmission cycle. Three findings warrant special mention. First, historical data suggest that the distribution of O. hupensis in China is restricted to areas where the mean January temperature is above 0°C, which roughly corresponds to 33.25° N latitude.23,26 Consequently, the mean January temperature can be used to delineate a "freezing line" north of which O. hupensis cannot survive, and hence transmission of S. japonicum would not be possible. Second, our laboratory experiments found a temperature of 5.8°C as the physiologic tolerance of O. hupensis. With regard to the development of S. japonicum within O. hupensis, a temperature of 15.4°C was found as the lowest threshold. Below this temperature, parasite development within the snail is arrested. Third, taking into account the previously mentioned temperature thresholds, we estimated the mean ADD for the development of S. japonicum in its intermediate host snail as 852.6 degree-days, and the mean ADD for the development of a generation of O. hupensis as 3846.3 degree-days. The aggregation of these findings, coupled with predicted temperature increases in China by 2030 and 2050, allowed us to draw up future risk maps for schistosomiasis transmission.
According to available temperature data for 1960 and 2000, the median January temperature, averaged across the 193 observing stations in China, increased by 0.9°C. This finding corroborates another recent time-series analysis, based on the average January temperatures from an ensemble of 652 observing stations in China, which found a temperature increase of 0.96°C between the 1960s and the 1990s.23 It has been predicted, on the basis of recent meteorological models using mean annual temperatures for the whole of China, that the mean temperature will continue to rise; indeed at an accelerated pace with predicted increases by 2030 and 2050 of 1.7 and 2.2°C, respectively.25 Our own model predictions are somewhat lower but they point in the same direction. We predict that the mean January temperature in China will increase by 0.9 in 2030 and by 1.6°C in 2050. Based on our biologic model, these temperature increases will result in an altered disease transmission, which will extend northward into currently non-endemic areas. For 2050 we predict that a surface area of 783,883 km2 might become at risk of schistosomiasis transmission, which translates to 8.1% of the total surface area of China. It is also conceivable that the transmission intensity will increase in areas already endemic for schistosomiasis. Our predictions are of considerable concern because they might explain, at least partially, the recent observations of re-emergence of schistosomiasis in areas where the criteria for transmission control, or even interruption, had been achieved.20–22
A limitation of our current modeling approach is that it emphasizes the role of temperature, but does not take into account the role of rainfall and the potential interaction between temperature and rainfall. It is difficult to say whether our model is conservative or whether these additional effects might further amplify the extent of changes predicted on the basis of temperature alone. Thus, recent improvements in modeling global trends in streamflow, precipitation, and water availability9 should become an integral part of present and future predictions of climate change and variability on infectious disease dynamics, including schistosomiasis. Flooding of the Yangtze River is implicated in snail dispersal and epidemic outbreak of schistosomiasis.37 However, floods usually originate far away, in the Tanggula mountains and only partially from local heavy rains.
Finally, our predictions of the potential impact of regional climate change on the transmission of schistosomiasis in China must be juxtaposed to profound demographic, ecologic, and socioeconomic changes that are taking place in this part of the world.38 Recently, it has been emphasized that the interplay of climate change and health should not be viewed in isolation, but rather be taken into account alongside other ecologic changes and socioeconomic development.4 This claim is particularly pertinent in the case of China where huge ecologic transformations are underway, most notably the implementation of water-resource development projects, such as the Three Gorges dam36 and the South-to-North water transfer project.23,39,40 The implementation and maintenance of water-resource development projects have a history of facilitating the transmission of schistosomiasis,41 and hence it is likely that such projects in China, in the face of a warming climate, will have a huge impact on schistosomiasis. Rigorous monitoring and surveillance thus is of pivotal importance if the goal of schistosomiasis elimination from China should not be compromised.
Received April 21, 2007. Accepted for publication July 2, 2007.
Acknowledgments: We thank our colleagues from the National Institute of Parasitic Diseases and the Jiangsu Institute of Parasitic Diseases for help with the field work, laboratory investigations, and statistical analyses.
This project was supported by the Chinese National Science Foundation (project nos. 300070684 and 30590373), the UNICEF/UNDP/ World Bank/WHO/Special Programme for Research and Training in Tropical Diseases (TDR; grant no. 970990), and the Ministry of Science and Technology (2003DIA6N009). At the time of manuscript preparation, G. J. Yang was a recipient of a TDR grant (A10775). J. Utzinger acknowledges financial support from the Swiss National Science Foundation (project no. PPOOB-102883).
* Address correspondence to Xiao-Nong Zhou, National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention, 207 Rui Jin Er Road, Shanghai 200025, People’s Republic of China. E-mail: ipdzhouxn{at}sh163.net 
Authors’ addresses: Xiao-Nong Zhou, Kun Yang, and Xian-Hong Wang, National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention, Shanghai 200025, Peoples Republic ’ of China. Guo-Jing Yang, Kun Yang, Qing-Biao Hong, and Le-Ping Sun, Department of Schistosomiasis Control, Jiangsu Institute of Parasitic Diseases, Wuxi 214064, People’s Republic of China. Guo-Jing Yang, School for Environmental Research, Charles Darwin University, Ellengowan Drive, 0909, Darwin, Australia. John B. Malone, Department of Pathobiological Sciences, School of Veterinary Medicine, Skip Bertman Drive, Louisiana State University, Baton Rouge, LA 70803. Thomas K. Kristensen, DBL–Institute for Health Research and Development, University of Copenhagen, Jaegersborg Allé 1D, DK–2920 Charlottenlund, Denmark. N. Robert Bergquist, Ingerod 407, 454 94 Brastad, Sweden. Jürg Utzinger, Department of Public Health and Epidemiology, Swiss Tropical Institute, PO Box, CH-4002 Basel, Switzerland.
Reprint requests: Xiao-Nong Zhou, National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention, 207 Rui Jin Er Road, Shanghai 200025, People’s Republic of China. Telephone: +86 21 6473–8058, Fax: +86 21 6433–2670, E-mail: ipdzhouxn{at}sh163.net.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  L.-D. Wang, H.-G. Chen, J.-G. Guo, X.-J. Zeng, X.-L. Hong, J.-J. Xiong, X.-H. Wu, X.-H. Wang, L.-Y. Wang, G. Xia, et al. A Strategy to Control Transmission of Schistosoma japonicum in China N. Engl. J. Med., January 8, 2009; 360(2): 121 - 128. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. H. King Where Snails No Longer Fear to Tread Am J Trop Med Hyg, February 1, 2008; 78(2): 185 - 185. [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
