Relating increasing hantavirus incidences to the changing climate: the mast connection

Abstract

Background: Nephropathia epidemica (NE), an emerging rodent-borne viral disease, has become the most important cause of infectious acute renal failure in Belgium, with sharp increases in incidence occurring for more than a decade. Bank voles are the rodent reservoir of the responsible hantavirus and are known to display cyclic population peaks. We tried to relate these peaks to the cyclic NE outbreaks observed since 1993. Our hypothesis was that the ecological causal connection was the staple food source for voles, being seeds of deciduous broad-leaf trees, commonly called "mast". We also examined whether past temperature and precipitation preceding "mast years" were statistically linked to these NE outbreaks.

Results: Since 1993, each NE peak is immediately preceded by a mast year, resulting in significantly higher NE case numbers during these peaks (Spearman R = -0.82; P = 0.034). NE peaks are significantly related to warmer autumns the year before (R = 0.51; P < 0.001), hotter summers two years before (R = 0.32; P < 0.001), but also to colder (R = -0.25; P < 0.01) and more moist summers (R = 0.39; P < 0.001) three years before. Summer correlations were even more pronounced, when only July was singled out as the most representative summer month.

Conclusion: NE peaks in year 0 are induced by abundant mast formation in year-1, facilitating bank vole survival during winter, thus putting the local human population at risk from the spring onwards of year 0. This bank vole survival is further promoted by higher autumn temperatures in year-1, whereas mast formation itself is primed by higher summer temperatures in year-2. Both summer and autumn temperatures have been rising to significantly higher levels during recent years, explaining the virtually continuous epidemic state since 2005 of a zoonosis, considered rare until recently. Moreover, in 2007 a NE peak and an abundant mast formation occurred for the first time within the same year, thus forecasting yet another record NE incidence for 2008. We therefore predict that with the anticipated climate changes due to global warming, NE might become a highly endemic disease in Belgium and surrounding countries.

Figure 1

Map of different biotopes and NE distribution in Europe and Western Russia. Temperate broad-leaf forests and boreal forests are the preferred habitats of the Puumala virus rodent reservoir Myodes glareolus, and hence often concur with NE endemic zones, as indicated. However, some regions such as the U.K. have so far no known NE endemicity, despite having broad-leaf forests as an adequate biotope and a documented presence of M. glareolus. The vegetation map was derived from Olson, D. M, E. Dinerstein, E.D. Wikramanayake, N.D. Burgess, G.V.N. Powell, E.C. Underwood, J.A. D'amico, I. Itoua, H.E. Strand, J.C. Morrison, C.J. Loucks, T.F. Allnutt, T.H. Ricketts, Y. Kura, J.F. Lamoreux, W.W.Wettengel, P. Hedao, & K.R. Kassem.

Figure 2

Yearly numbers of NE cases in Belgium 1985-half 2008. Serologically confirmed acute NE in Belgium, 1985-half 2008. For the current year 2008, marked with asterisk, only data of the first six months were available. Numbers above each column are the cases/year, the number between brackets above the 2008 column is only the half-yearly number. Mast years are indicated with black full arrows.

Figure 3

Climate Graph Royal Meteorological Institute, Brussels, (Belgium) 1985–2007. Monthly variations of mean temperature (red lines, in °C) and mean precipitation (black lines, in mm) of the current 1996–2007 study period, compared to the previous decade 1985–1995.