Is There a Risk of Suburban Transmission of Malaria in Selangor, Malaysia?

Background The suburban transmission of malaria in Selangor, Malaysia’s most developed and populous state still remains a concern for public health in this region. Despite much successful control efforts directed at its reduction, sporadic cases, mostly brought in by foreigners have continued to occur. In addition, cases of simian malaria caused by Plasmodium knowlesi, some with fatal outcome have caused grave concern to health workers. The aim of this study was to investigate the possibility of local malaria transmission in suburban regions of Selangor, which are adjacent to secondary rainforests. Findings A malaria survey spanning 7 years (2006 - 2012) was conducted in Selangor. A total of 1623 laboratory confirmed malaria cases were reported from Selangor’s nine districts. While 72.6% of these cases (1178/1623) were attributed to imported malaria (cases originating from other countries), 25.5% (414/1623) were local cases and 1.9% (31/1623) were considered as relapse and unclassified cases combined. In this study, the most prevalent infection was P. vivax (1239 cases, prevalence 76.3%) followed by P. falciparum (211, 13.0%), P. knowlesi (75, 4.6%), P. malariae (71, 4.4%) and P. ovale (1, 0.06%). Mixed infections comprising of P. vivax and P. falciparum were confirmed (26, 1.6%). Entomological surveys targeting the residences of malaria patients’ showed that the most commonly trapped Anopheles species was An. maculatus. No oocysts or sporozoites were found in the An. maculatus collected. Nevertheless, the possibility of An. maculatus being the malaria vector in the investigated locations was high due to its persistent occurrence in these areas. Conclusions Malaria cases reported in this study were mostly imported cases. However the co-existence of local cases and potential Plasmodium spp. vectors should be cause for concern. The results of this survey reflect the need of maintaining closely monitored malaria control programs and continuous extensive malaria surveillance in Peninsula Malaysia.


Introduction
Malaria is considered as one of the most important infectious diseases in the world. It is caused by the protozoan parasites of the genus Plasmodium [1]. There are an estimated 219 million malaria cases reported annually. Malaria is a potentially fatal disease. It kills approximately 660,000 people every year [2]. The vector of malaria is the Anopheles mosquito. Hot and humid areas are conducive for malaria transmission, where high temperature ensures the survivability of both the malarial parasites and the vectors [3,4]. In Malaysia, malaria consistently remains as a public health issue. Moreover, the infection rate have been reported to gradually increase among the mostly jungle dwelling aborigines (Orang Asli), soldiers [5], and foreign workers hailing from malaria endemic countries [6], raising concern among healthcare workers and policy makers.
Selangor is a state in Peninsula Malaysia which is highly urbanised, yet remains susceptible to malaria infections [7]. This south-western Malaysian state has an area of 7930 sq. km. It is indicated as the most populous state in Malaysia with a population of 5.46 million of which 4.99 million live in urban areas. Selangor has high level of urbanization (91.4%) and population growth [8]. Historically, Kuala Lumpur, which is the capital of Malaysia, resides in the state of Selangor before it was gazetted as Federal Territory. Anopheles maculatus had been shown to be an important vector of malaria in Kuala Lumpur since 1906 [9]. This was attributed to the removal of jungle cover from the valleys [3,9]. However, its present vectorial status needs to be reinvestigated.
The state of Selangor is made up of 9 districts, namely, Hulu Langat, Petaling, Kuala Selangor, Hulu Selangor, Gombak, Sabak Bernam, Klang, Kuala Langat and Sepang. Malaria has continued to persist in all these districts. In Selangor, malaria remains a healthcare concern due to various factors. Among them are human encroachment into jungles [7], high rate of construction and development [3], mobile local aborigines (Orang Asli) population [10][11][12] and influx of immigrant workers from malaria endemic countries [5]. Furthermore, the presence of pockets of secondary rainforests in the vicinity of urban and suburban dwellings provides niches for vector mosquitoes and monkeys that may harbor simian malaria parasites.
Recently, numerous cases of malaria caused by different species of Plasmodium have been reported in Selangor [5][6][7][13][14][15]. In a retrospective study conducted among the local aborigines admitted to Gombak Hospital 32 patients were positive for infection of Plasmodium spp [5]. (mono-infections of Plasmodium falciparum and Plasmodium vivax, mixed infections of P. falciparum and P. vivax). Reports of the simian malaria, Plasmodium knowlesi known to cause severe malaria in humans is also on the rise in Selangor. Recently, Lee et al reported an increase in the numbers of P. knowlesi malaria cases in an urban hospital [7]. Thus, there is a need to address this issue from a knowledge-based angle.
The Geographical Information System (GIS) is a new tool that can be used to monitor and evaluate a disease and its control measures. The record of diseases can be stored and presented as maps instead of tables. The correlation between geographical factors and epidemiology provides a clearer picture of the spread of this disease [16]. This technique provides an insightful picture of the dangers of malaria. We conducted epidemiological and geographical studies in Selangor and used GIS to present the cases of malaria in the state of Selangor. Our study is timely considering the fact that the malarial transmission in Selangor needs a comprehensive study to understand its dynamics, so as to either modify the control efforts or continue the already implemented measures. Findings from this study can thus be used to support malarial eradication effort. This study attempts to understand the spread of malaria as seen in Selangor, which is a highly urbanized area. The aim of this study is to map the prevalence of malaria among the population of the state of Selangor, Malaysia, as well as to investigate the hosts and vectors of malaria in malaria endemic areas.

Malaria Cases and Treatment
Malaria transmission in Selangor is sporadic and varies from one area to another. During this research project, malaria cases were detected by the health clinics of respective districts in Selangor. Blood smears were obtained and sent to the laboratory. When malaria diagnosis was confirmed, the case was registered to the source of infection (locality of infection) accordingly. The cases were confirmed by the Klang Vector Laboratory. Thereafter, the Malaria Surveillance Unit of the Klang Vector Laboratory initiates an epidemiological investigation to determine origin of the cases (whether local or imported) and conduct mosquito collections to implicate the vector.
Treatment consisting of a single course of chloroquine was prescribed to patients with uncomplicated infections of P. falciparum, P. malariae and P. knowlesi. For patients infected with P. vivax and P.ovale, a course of chloroquine and primaquine was given. In cases of severe malaria, intravenous artesunate was used in combination with another drug such as doxycycline or chloroquine, regardless of the causative species. For vivax malaria patients, post treatment follow up of one year was performed whereas for patients infected with other species of malaria parasites (including P. knowlesi infection), a six-month-medical follow up was given.

Study Area and Entomological Mapping
The state of Selangor, Peninsula Malaysia was the target site of this study. Malaria cases from all districts were reported to the Klang Vector Laboratory. Entomological sampling was done on localities with reported malaria cases between year 2006 and 2012 whereby initial investigation revealed possibility of local malarial infection. The exact coordinates of potential vectors collected were obtained using the Garmin Geographic Positioning System (GPS) and directly plotted into GIS software system [17] as in Figure 1. Larvae obtained from the surrounding vicinity were brought and raised to adults in the laboratory. Subsequently, species identification of these adult mosquitoes was performed.

Entomological Sampling and Mosquito Identification
Sampling was done following reports of any malaria cases. Larvae were collected from water bodies adjacent (less than 1000m radius) to the vicinity where malaria cases were reported. Adult Anopheles mosquitoes were collected using bare leg landing method and double-bed net method for three consecutive days at 41 locations reported to have malaria cases between 2006 to 2012 (Table 1). Collection was carried out from 1800 to 0000 hours.
Collected mosquitoes were identified taxonomically using Anopheles mosquito identification keys of Reid and Jeffery [18,19]. The dissected mosquito was examined with a Leica ICC50 microscope (Leica Microsystems, Wetzlar, Germany) equipped with a Leica DC 300F colour video camera (Leica Microsystems, Cambridge, UK), which was connected to a computer. Photographs were analysed with Leica IM500 software (Leica Microsystems, Wetzler Germany). Several organs of each mosquito specimen were investigated: midgut to check for oocysts and salivary glands for sporozoites.

Long tailed macaque distribution
Since the long-tailed macaque monkey (Macaca fascicularis) is a known natural host of P. knowlesi [20] and is native to Peninsula Malaysia, its distribution was also taken into account in this study. Data on the macaque distribution was obtained from the Department of Wildlife and National Parks (DWNP/ PERHILITAN).

Data analysis
Raw data of malaria patients from the Klang Vector Laboratory were archived using Microsoft Excel 2010. Data analysis was done using Sigma Plot for Windows Version 12.0 Build 12.2.0.45 and Microsoft Excel 2010 (Version 14.0.6129.500 -32-bit). Based on the Garmin GPS Device, the coordinates of sampling points within study sites that were captured were then digitally plotted and mapped using ArcGIS software [17]. All photographs were taken with an Olympus camera (OM-D E-M5).

Malaria cases
A total of 1623 laboratory confirmed malaria cases were registered in the 7-year-period between 2006 and 2012. The malaria cases (both local and imported cases) were found positive by microscopic examination for Plasmodium spp. and have been reported from all the nine districts of Selangor ( Figure 2). Records of confirmed positive malaria cases for the year 2006 to 2012 of each district were analysed, and percentages of malaria cases in Selangor subsequently determined, and were represented with maps accordingly.
The age-specific infection prevalence and species distribution of all reported malaria cases are shown in Figure 4. From this study, the incidence of malaria ranged between the age of 1 and 55 year-old. The highest prevalence was found in the age group of 25-29 year-old (peak age: 25 year-old).
Based on the distribution of imported and local malaria cases in Selangor ( Figure 5), imported cases showed a sharp increase in 2007 and thereafter dropped to a level that was higher than that of year 2006. This level was generally maintained throughout the study. However, the number of local malaria cases increased. In 2012, the number of local and imported cases had almost equalised (114 local cases as compared to 139 imported cases). This is in contrast with the 5 local cases and 83 imported cases reported in 2006.

Malaria and mosquito correlation
To determine the correlation between the causative agent of malaria (Plasmodium) and the vector (Anopheles mosquito), we tried to find out the potential vectors of malaria in the districts of Selangor with reported human cases ( Figure 1; Table 1 In this investigation, the Anopheles mosquitoes were found in a total of 41 localities. The established malarial vector of Peninsula Malaysia, An. maculatus was relatively frequently collected in places with reported malaria cases. It was found in 18 out of the 41 locations with Anopheles mosquitoes. Based on examinations, none were found positive for malarial parasites. Moreover, at least 2 areas had Macaca fascicularis present coinciding with that of reported malaria cases and Anopheles mosquito prevalence (Figure 1). Both human malaria cases were that of P. knowlesi infections.

Malaria Epidemiology and Strategy in Selangor
This study is among the few which aims to investigate the correlation between malaria cases, macaque distribution and Anopheles mosquitoes in Selangor, Peninsula Malaysia. The National Strategic Plan (NSP) for Malaria Elimination 2010-2020 has been initiated by the Ministry of Health Malaysia. One of the aims of this national plan is to reduce locally acquired malaria incidence rate in Peninsula Malaysia by less than 1 per 1000 population by 2015 and in East Malaysia (Sabah and Sarawak) by 2017 and eventually, to ensure that Malaysia is free from local malaria transmission (except P.knowlesi transmission) by 2020. However, Anopheles mosquitoes would still be present in Malaysia, and possibilities of outbreaks occurring in such locations still exists if malaria cases were imported from other countries. Protection in malarious areas are provided through the use of residual spraying, impregnated bed net and anti-larvicidal control methods.
One of the main obstacles faced by NSP is the mobility of infected workers. A locality that has the source of infection (human cases) is termed vulnerable areas whereas an area that has malaria vectors is categorized as receptive areas. Receptive and vulnerable areas indicate areas that have both local malaria cases being reported as well as vector mosquito being found within that malarious area simultaneously. In the event where two or more human malaria cases occur in a certain area, an "outbreak" is said to have occurred in that area. In 2012, only two out of nine districts in Selangor were classified in this category, namely Hulu Langat, and Gombak. In order to prevent outbreaks in 'receptive and vulnerable areas' as well as in other areas, time-tested malaria prevention and control measures of the Malaria Elimination Programme must be implemented continuously.

Transmission of malaria in Selangor
In this study, we show that malaria transmission is still being reported in the State of Selangor. From the GIS data ( Figure  2), local and imported malaria cases were still being reported from all districts of this state. It is interesting to note that although no malaria was reported within Kuala Lumpur, imported malaria transmission was reported from all the districts within the State of Selangor, which surround the Federal Territory strongly suggesting that foreigners and travellers from areas infected with malaria are crucial in the introduction of malaria to non-malarious areas [6].
On the other hand, local malaria transmission was only reported from three districts (Gombak, Hulu Langat, Hulu Selangor) that adjoins sub-urban areas and the Titiwangsa mountain range. We suggest that such distribution pattern is mainly attributed to the clearing of forests for development in these districts. Similar correlation of deforestation and malaria transmission was observed in the State of Perak, Peninsula Malaysia in 1997 [11].

Factors contributing to a potential risk of malaria transmission in Selangor
(i) Presence of mosquito vector. We noted that Anopheles spp. mosquitoes were localized in areas which have undergone deforestation due to the construction of homes for human habitation in areas where secondary rainforests bordered townships [7]. A similar phenomenon has also been noted in areas of the Peruvian Amazon where malaria is on the rise due to road development and ecological changes attributed to deforestation, which in turn caused changes in the biting habits of An. darlingi [21,22]. In the locations studied, the larvae of An. maculatus in particular was found in shallow pools with clear water and plants [23].
Urbanised Kuala Lumpur has successfully controlled the prevalence of Anopheles mosquitoes by building the sub-soil pipe drainage system [24]. As a result, it is free of An. maculatus although having the right ecological conditions for the breeding of An. maculatus. Thus, it is regarded as "Maculatus Reception Areas". This is because in the event that the drainage system built during the Malaria Eradication Programme (MEP) in the 1960s breaks down, such areas may have a sudden increase in the prevalence of Anopheles mosquitoes.
Dissection performed on the Anopheles mosquitoes collected during the fieldwork showed that none of the specimens were infected by Plasmodium spp. Although An. maculatus, the main vector of Plasmodium spp. (P. falciparum, P. malariae, P. vivax) in Peninsula Malaysia [25] was not incriminated in this study, it is believed to be the main vector in the country and is considered to play a role as a potential vector of P. knowlesi due to the frequent presence of this mosquito in areas where P. knowlesi infection was reported in humans.
(ii) Presence of source of infection. Among all districts in Selangor, Petaling had the highest infection rate of malaria that was attributed to imported malaria brought in by foreign workers from malaria endemic countries. In 2012, the largest source of imported malaria was Indonesians followed by citizens of Pakistan, Myanmar, India, Bangladesh and Nepal. The two outbreaks that occurred in Hulu Langat and Gombak district in 2012 originated from the quarters of foreign workers in these districts.

Macaque distribution
The overall data on macaque distribution in Peninsula Malaysia was obtained from the Department of Wildlife and National Parks, and detailed data on malaria cases were obtained from Health Departments of the Petaling District and

Other factors
Detection and control of malaria remains a public health concern, especially among the Orang Asli who stay in remote areas and make forays into the jungle in search of jungle produce, and thus exposing themselves to Anopheline mosquitoes infected with malaria parasites. The most prevalent malaria parasite in this study was P. vivax, followed by P. falciparum. This was similar to a previous study conducted in Pos Piah, Perak by Norhayatiet al., 2001 [26]. Other studies among Orang Asli population in Selangor reported that the most prevalent species was P. falciparum, followed by P. vivax [5].
Human activities such as hydroelectric dam construction, paddy field activities and irrigation canals which ecologically alter the jungle environment and encourage re-invasion of the typically efficient Anopheles mosquito which may play the role as vectors [11]. When a forested area is disturbed, streams and seepage water that were originally shaded are now exposed to sunlight. These clean, sun-lit water bodies are the breeding sites of choice for An. maculatus, the predominant mosquito vector of Peninsula Malaysia [23,25,27]. The ability of some soil types to trap water, such as those of granite origin, further contributes to this problem. Moreover, the replanting activities in agricultural industry such as those of rubber tree plantations and oil palm plantations further encourage the prevalence of the Anopheles mosquitoes [18,25,27]. It is possible that the construction of the dam in Kampung Pertak, Hulu Selangor district has allowed exposure of clean water bodies to sunlight, forming the breeding sites for Anopheles mosquitoes, such as, An. maculatus.

Plasmodium knowlesi infection in Selangor
Over the 7-year-period, the first report of P. knowlesi infections (2 cases) were reported to the Klang Vector Laboratory in 2008. P. knowlesi cases reported in this study were mainly from jungle fringes. There were no cases reported from urban areas. In 2012, two districts reported P. knowlesi cases; Hulu Selangor (32 cases, 11.8%) and Gombak (5 cases, 1.8%). Both these districts had sightings of Macaca fascicularis, which were also verified by distribution data given by the DWNP. Prior to 2008, in this study there was no malaria reported caused by P. knowlesi. The discovery of a large cohort of P. knowlesi cases thought to have been P. malariae by Singh et al. [28] has led to awareness of the presence of human knowlesi malaria in the region. Laboratory personnel and clinicians have thus become more aware of this parasite and since then, P. knowlesi infections have been reported in neighbouring countries such as Thailand [29], Myanmar [30], Vietnam [31], Singapore [32], the Philippines [33], and China [34]. Besides, there were reports of P. knowlesi infection acquired by travellers visiting Peninsula Malaysia [35]. P. knowlesi infection is potentially fatal due to its rapid development (24 hour-erythrocytic cycle) and great potential of achieving high parasitemia (> 5000/μl blood) [36,37].

Limitations
Malaria parasites species in this study were detected using light microscopy alone and thus could have led to premature analysis of results as early trophozoites that appear as ring forms resembling those of P. falciparum while late trophozoites "band forms" appear similar to P. malariae [28]. Furthermore, the malaria results provided for this study did not include infections that were not reported or those that did not go for treatment. Therefore, it may not reflect the actual malaria transmission in the state.
Another potential limitation in this study was that vegetation cover was not represented in the GIS map to show human activities such as deforestation, as well as encroachment in the sampling sites in the various districts into the potential Anopheles inhabited environments. However, this correlation was partially verified by our cursory field observations and findings.
Frequent immediate spraying by health authorities subsequent to a malaria case report resulted in low number of mosquitoes caught and might have led to the negative infection findings on the captured mosquitoes. Furthermore, we were limited by our capacity to enter into the depths of the jungle and had to restrict our collection close to human habitation.

Conclusion
In line with the National Malaria Elimination Strategic Plan 2010-2020, vector control strategies consisting of residual spraying, the use of impregnated bed net and larvicidal control activities have to be done in malaria reported areas. GIS to display malaria information in a graphical representation that is easily interpreted can be extensively used in the planning of malaria control programs in all the states of Malaysia. This will be beneficial to the planning, funding and control measures that can be undertaken by the state health sector to prevent the occurrence of malaria.
Although Selangor has demonstrated a general decrease in imported malaria cases since 2007, local malaria cases, especially those caused by P.knowlesi are on the rise. Thus, there is still a possibility for malaria resurgence here. It is of serious concern that the presence of potential vectors in regions housing Plasmodium infected foreigners may lead to an increased risk of malaria transmission. Hence, it is imperative that the link between human populations and malaria vector should be controlled to break the transmission chain.