Background
Under Soviet rule, malaria was reported having been nearly eradicated in Tajikistan by the end of the 1960s; only in the region bordering Afghanistan, in the southern part of the country, did a low level of
Plasmodium vivax transmission persist [
1]. However, since 1992 the situation appears to have deteriorated considerably. The World Health Organization (WHO) estimates annual incidences to be much higher than what is officially reported. Estimates are in the range of 300–400,000 cases of
P. vivax and 30–50,000 cases of
Plasmodium falciparum[
2]. This contrasts with official governmental statistics indicating a peak of 30,000 cases in 1997, followed by a decline [
3]. According to the same governmental data, more than 80% of the cases were registered in the southern province of Khatlon, where there is a substantial influx of Tajik refugees returning after the civil war from northern Afghanistan. In 1997, many Afghans also entered Tajikistan from the south, fleeing the Taliban regime and, thus, possibly importing malaria parasites. With the deterioration of basic infrastructure and social services due to the civil war, all malaria prevention measures were suspended and drainage systems are no longer adequately maintained. These developments provide a better breeding ground for mosquitoes in ditch-water reservoirs and other stagnant pools [
2].
Since the early 1990s, changes in spatial malaria distribution patterns have been observed. An increase in average temperatures and changes in land use are seen as driving factors here [
1]. In areas where the climate is marginally suitable for malaria transmission – such as areas at high altitude in Tajikistan – climatic variations may play a key role in the incidence of malaria [
4].
The current standard national malaria treatment of
P. vivax malaria consists in administrating a three-day course of chloroquine followed by a 14-day course of primaquine [
5]. Primaquine is the only available antimalarial drug able to effectively eliminate
P. vivax by killing the latent liver stages of the parasite. This treatment also plays an essential role in controlling the disease by preventing further transmission. However, there is a potential lack of compliance with the treatment due to haemolytic side effects of primaquine. People with an inherited G6PD deficiency are particularly at risk of severe haemolysis when using primaquine. Therefore, India and Pakistan have adhered to a five-day course instead of the recommended 14-day course as their standard malaria treatment. This makes it possible to limit incidences of severe haemolysis, especially in countries where the necessary facilities for detecting G6PD deficiency are often lacking. However, this strategy has the disadvantage of failing to effectively prevent relapses [
6]. The general frequency of G6PD varies among areas and ethnic groups [
7] and its geographic distribution has led several authors to suggest that G6PD deficiency is a polymorphism that builds resistance to
P. falciparum malaria [
8]. In Tajikistan, a G6PD deficiency frequency of 1.6% in the South and 0% in the North of the country has been investigated by Russian scientists in the 1980ies [
9]. Before mutation analysis was possible three different variants of G6PD gene mutations were detected: Dushanbe I and II, which have less than 10% of residual enzyme activity and Dushanbe III with normal enzyme activity [
10]. More recent G6PD deficiency studies in Central Asia have revealed quite a range in the frequency of the trait among the different ethnic groups: 2.1% among Irani to 2.9% among Afghan Tajiks over 15.8% among Pakistani Pathans up to 36.4% among Azerbaijani [
11‐
13].
So far, much remains unknown about the current malaria situation in Tajikistan. The present study aims (i) to establish a malaria risk map with special consideration of seasonal and environmental distribution patterns; (ii) to assess G6PD deficiency prevalence and (iii) to critically review the national strategy for combating P. falciparum malaria. Jointly, these aims should make it possible to provide relevant information to successfully control the disease in Tajikistan.
Discussion
Spatial and seasonal distribution of malaria
Low temperatures limit malaria transmission both spatially and temporally. Due to low temperatures, locations at high altitudes are spared the risk of malaria. The threshold for malaria transmission was estimated to be at altitudes higher than 2,500 m above sea level. Malaria epidemics at relatively high altitudes appear to be possible in Central Asia, as one study reported an outbreak in neighbouring Afghanistan at an altitude of around 2,400 m [
17]. This excludes the entire high-mountain area in the eastern part of the country – mainly the Pamirs – from malaria transmission.
The dominant role of temperature in transmission is obvious, given the fact that rainfall is scarce if not completely absent during summer in many parts of Tajikistan. The amount of rainfall does thus not appear to influence malaria distribution significantly (Figure
2). In Pakistan, where similar seasonal patterns in
P. vivax and
P. falciparum transmission have been observed [
18], the malaria transmission season generally follows the monsoon rains [
19]. Despite the absence of rainfall in summer, when temperatures are appropriate for malaria vectors to breed, there are plenty of open water bodies due to the high level of water discharge from snow and ice melting in the mountains. Especially irrigation systems in the plains and valleys, but also natural rivers and possibly private water collectors, provide appropriate aquatic habitats for malaria vectors. A study conducted in South Punjab (Pakistan), where rainfall is also low and irrigation is also practised on a large scale, showed most breeding sites were linked to the irrigation system [
20].
The main cause of higher malaria incidences in southern Tajikistan compared to northern regions of the country, despite similar suitable environmental conditions, is probably the massive influx of migrants in the mid 1990s from malaria-endemic areas in neighbouring Afghanistan. Apparently the different degrees of poverty in these two parts of the country only play a minor role. Degrees of poverty are known to be less severe in the northern than in the southern province of Tajikistan. Today internal migration is seen as more important for malaria transmission than cross-border movements. However, further investigation is needed to understand the relation between internal and external migration, other socio-economic factors, and malaria transmission patterns in Central Asia.
Validation of the model
Even though the model used was relatively simple, it appears to be highly plausible with respect to better understanding of the mechanisms of malaria transmission in Tajikistan. While the relative importance of the different factors influencing malaria distribution – such as the degree of immunity in the population, the quality of health services, and preventive measures – is not fully known, the model includes the most important environmental factors that directly influence the biology of the malaria parasites and vectors: temperature and potential aquatic breeding sites.
Malaria risk zones
As a result of the malaria risk map, three main categories of malaria risk in Tajikistan were identified. They can be characterized as follows:
-
Zones at high relative risk of malaria combined with high current incidence rates. This was mainly true for the province of Khatlon in southern Tajikistan. In this zone malaria control and prevention measures should be taken at all stages of the transmission cycle.
-
Zones at relatively high risk of malaria but with low current incidence rates. This mainly applies to the northern part of the country. Here the incidence of malaria should be monitored carefully, since some areas located in the centre appear to be at relatively high risk as well.
-
Zones at moderate or low risk combined with low current incidence rates. This applies mainly to locations at intermediate altitudes where temperature appears to limit the level of malaria transmission. Accordingly, a possible rise in temperature, which has been prognosticated for the long run due to climate change [
21], could lead to an extension and intensification of malaria transmission in these areas. Thus, rising average temperatures would allow malaria to spread to so far malaria-free areas, which then would become especially prone to epidemics due to the lack of immunity among the population [
4].
Prevalence of G6PD deficiency
Reliability of the test
The qualitative screening test applied for G6PD deficiency using the dye reduction method worked well. The simplified incubation method apparently did not influence the results, since the decolourization of the test was obvious and the results could be easily determined. The specificity of the test was estimated to be high, as only four participants theoretically could have been deficient while showing a negative result for G6PD deficiency. According to the greater activity of G6PD in young red blood cells, a high amount of reticulocytes in blood can give a false negative result. One study participant was suffering from anaemia, two had liver problems, and one was suspected of having malaria.
Regional differences in the frequency of G6PD deficiency
The revealed regional differences in the frequency of G6PD deficiency appear to be related to the ethnic composition of the study populations rather than to malaria incidence rates or other potential determinants. Indeed, the results of this research confirmed previous results obtained in a survey about G6PD deficiency among refugees in Northern Pakistan: a high rate of G6PD deficiency of 9.1% among Uzbek refugees, and a rate of only 2.9% among Tajik refugees [
12]. The higher frequency of G6PD deficiency among Uzbeks might be explained by their residence in
P. falciparum endemic areas in the northern provinces of Afghanistan for many centuries. Information gathered from Russian explorers at the beginning of the 20th century indicates that in southern Uzbekistan and Western Kuhistan, which includes the area of the province of Khatlon, present-day Uzbeks comprised a large number of small tribes, each with specific locations and tribal affiliations. Qunqurat tribes settled in northern Afghanistan during the 14th and 15th centuries. At the end of the 19th and the beginning of the 20th centuries, they were still semi-nomadic, and can be found today in Southern Tajikistan [
22]. An effect of selection through
P. falciparum can therefore not be excluded. In Tajikistan, it is estimated that the
P. falciparum burden was not great enough to exert selective pressure.
Conclusion
Malaria transmission in Tajikistan is a mono-seasonal phenomenon, lasting mainly from April to November. Temperature appears to be the main determining environmental factor, not only for seasonal but also for spatial distribution patterns. Based on the established risk map, three different risk zones could be identified:
(i)
zones with a high relative risk of malaria combined with high current incidence rates, which can be found in districts of the southern province of Khatlon. In these zones it is recommended to take preventive measures at all stages of the malaria transmission cycle.
(ii)
zones at relatively high risk of malaria combined with low current incidence. Here it is recommended to carefully observe the malaria situation and be ready to take measures if necessary.
(iii)
zones with intermediate and low risk of malaria combined with low current incidence rates. This is true for locations at intermediate altitudes. A rise in temperature can lead to an increase in malaria incidence and would, moreover, allow malaria transmisson to spread to areas so far free from malaria. Such new areas would be particularly prone to malaria epidemics.
The 2.1% of G6PD deficiency prevalence found suggests that current national standard malaria treatment guidelines for P. vivax should be maintained. However, in order to limit incidences of severe haemolysis after primaquine treatment, family members of patients who have had haemolytic episodes should be screened preventively for G6PD deficiency. Persons who have been tested positive should be given a card indicating their deficiency status and containing a list of drugs to be avoided or dosed carefully by qualified medical staff. Specific G6PD deficiency-sensitive training for doctors in all malaria-affected districts of Tajikistan should be offered to raise awareness about this issue. In a first phase, this might focus on districts with a higher proportion of ethnic Uzbeks.
Acknowledgements
Research for this paper was supported by the Swiss National Centre of Competence in Research (NCCR) North-South: Research Partnerships for Mitigating Syndromes of Global Change, co-funded by the Swiss National Science Foundation (SNSF) and the Swiss Agency for Development and Cooperation (SDC). Additional support was provided by the Swiss Federal Institute of Technology Zurich to cover travel expenses. The study was conducted jointly by the Swiss Tropical Institute (STI, Basle), the Centre for Development and Environment (CDE, Berne), and the Republican Centre of Tropical Diseases (RTDC, Dushanbe). We greatly appreciated the assistance of the laboratory team of RTDC in the G6PD deficiency testing procedure, and we also acknowledge the non-governmental organisation MERLIN for providing us with malaria incidence data directly collected at laboratories. In addition we would like to thank the Department of State Statistics, the Ministry of Water Resources and Melioration, and the Republic State Committee on Preservation of the Environment for supplying us with critical data and information.
Authors' contributions
CER planned and led the survey on G6PD deficiency and was the main author of the article. AJM collected data, analysed the influence of major environmental factors regarding the distribution of malaria, and developed the malaria risk map. KS took part in the planning of activities, assisted with obtaining agreement from the Ministry of Health, and informed local public authorities and hospital directors about the G6PD deficiency survey. KW and DM conceived the study, assisted AJM and CER in designing and discussing their research, and contributed to the elaboration of the manuscript.