Linking climate to incidence of zoonotic cutaneous leishmaniasis (L. major) in pre-Saharan North Africa

Abstract

Shifts in surface climate may have changed the dynamic of zoonotic cutaneous leishmaniasis (ZCL) in the pre-Saharan zones of North Africa. Caused by Leishmania major, this form multiplies in the body of rodents serving as reservoirs of the disease. The parasite is then transmitted to human hosts by the bite of a Phlebotomine sand fly (Diptera: Psychodidae) that was previously fed by biting an infected reservoir. We examine the seasonal and interannual dynamics of the incidence of this ZCL as a function of surface climate indicators in two regions covering a large area of the semi-arid Pre-Saharan North Africa. Results suggest that in this area, changes in climate may have initiated a trophic cascade that resulted in an increase in ZCL incidence. We find the correlation between the rainy season precipitation and the same year Normalized Difference Vegetation Index (NDVI) to be strong for both regions while the number of cases of ZCL incidence lags the precipitation and NDVI by 2 years. The zoonotic cutaneous leishmaniasis seasonal dynamic appears to be controlled by minimum temperatures and presents a 2-month lag between the reported infection date and the presumed date when the infection actually occurred. The decadal increase in the number of ZCL occurrence in the region suggests that changes in climate increased minimum temperatures sufficiently and created conditions suitable for endemicity that did not previously exist. We also find that temperatures above a critical range suppress ZCL incidence by limiting the vector's reproductive activity.

Figure 1

(a) Study area, showing the two provinces of Saida and Errachidia; (b) Annual mean temperature and precipitation anomalies for Errachidia; (c) Annual mean temperature and precipitation anomalies for Saida.

Figure 2

Normalized time series of ZCL-cases for the two provinces. The maximum number of cases for Saida is 267 (population 7,500 inhabitants) and for Errachidia is 1596 (population 600,000 inhabitants).

Figure 3

Changes in rainy season precipitation and growing season NDVI from P1 to P2 (see text for details). Changes in NDVI values were scaled by 1,000 for plotting purposes. Prefix ER is for Errachidia and S for Saida.

Figure 4

Relationship between precipitation, NDVI, and ZCL-incidence in the province of Saida. Rainy season (September through April) precipitation, growing season (February through May) vegetation density and annual total ZCL-cases. Note the correspondence of pattern dynamics between temporal changes of precipitation and NDVI values. The ZCL-cases lag precipitation and vegetation by two years.

Figure 5

Proposed Trophic Cascade in Leishmaniasis complex.

Figure 6

Monthly composite relative humidity (bar, right axis). Maximum (red) and minimum (blue) temperatures (left axis) for P2 at Saida. Dark shading represents the vector’s period of hibernation. Active and Reproductive periods are indicated by horizontal bars at the top of the graph. The dark solid line represents the monthly average reported ZCL-cases during P2 and the dashed line shows the presumed ZCL-occurrence with a 2-month lag (right axis).

Figure 7

Same as Figure 6, except for Errachidia and the monthly average ZCL-cases are divided by 10 for plotting purposes.

Figure 8

Average July-August maximum temperature anomalies from the mean value over P2 (34.76 °C) and 1-year lag ZCL-cases for Saida.