Short communication: Use of a recombinant antigen-based ELISA to determine prevalence of brugian filariasis among Malaysian schoolchildren near Pasir Mas, Kelantan – Thailand border


correspondence N. Rahmah, Department of Microbiology and Parasitology, School of Medical Sciences, Universiti Sains Malaysia, 16150 Kubang Kerian, Kelantan, Malaysia


Brugian filariasis infects 13 million people in Asia. The routine prevalence survey method using night thick blood smear is not sensitive enough to reflect the actual infection prevalence. In 1997–2001, only three microfilaraemic cases (of 5601 individuals screened; 0.05%) were reported in Pasir Mas, a district in Kelantan (Malaysia), which shares a border with Thailand. We therefore investigated the infection prevalence in this district by employing a sensitive and specific serological assay (Brugia-Elisa). This test is based on detection of specific IgG4 antibody against a Brugia malayi recombinant antigen. A total of 5138 children, aged 7–12 years, from 16 primary schools, were tested. Eighteen pupils in eight schools, located in five subdistricts, tested positive, giving an overall prevalence rate of 0.35%. Infection in these children is significant as they represent more recent cases. These subdistricts should be included in the national filariasis elimination programme.


Lymphatic filariasis is caused by Wuchereria bancrofti, Brugia malayi and B. timori. An estimated 120 million people are infected worldwide and 1.1 billion people who live in endemic areas are at risk of acquiring the infection. Brugia malayi accounts for about 10% of the infection and occurs in India, China, Indonesia, Philippines, Thailand, Vietnam and Malaysia (WHO 1998).

The routine thick blood examination is known to be insensitive and inconvenient. Other more sensitive techniques include Knott concentration, membrane filtration and polymerase chain reaction-based assays. However, these methods are unsuitable for mass screening.

The World Health Organization has identified lymphatic filariasis to be one of the six infectious diseases that has the potential to be eliminated as a public health problem. Thus a global programme for elimination of this disease has been initiated by WHO in 1997 (WHO 1998). A yearly dose of diethylcarbamazine (or ivermectin in certain countries) and albendazole has been found to be sufficient to suppress circulating microfilaria (mf) for a year. If approximately 85% of the population is given this treatment regimen for a consecutive period of 5–20 years, the microfilaraemia rate in a population can be brought down to less than 0.1% (one per 1000 population), a level which will no longer support active transmission of the infection.

Another important aspect of this programme is the requirement of a sensitive diagnostic tool for proper mapping and surveillance exercises. For bancroftian filariasis, two kinds of monoclonal antibody-based antigen detection tests exist, i.e. the Trop-Ag W. bancrofti Elisa and the ICT filariasis card test (More & Copeman 1990; Weil et al. 1997). Efforts to develop a similar antigen-detection test for brugian filariasis have not been successful so far. However, many studies have demonstrated that antifilarial IgG4 antibodies are significantly elevated in actively infected individuals (Ottesen et al. 1985; Kwan-Lim et al. 1990; Kurniawan et al. 1993; Rahmah et al. 1998; Haarbrink et al. 1999) and decline post-treatment (Wamae et al. 1992; Kurniawan et al. 1995; McCarthy et al. 1995; Rahmah et al. 2001). Therefore, detection of IgG4 antibodies can be used as a marker of active infection in lymphatic filariasis.

We have produced a B. malayi recombinant antigen (BmR1) that shows high sensitivity and specificity for detection of brugian filariasis. An enzyme-linked immunosorbent assay (ELISA) based on this antigen (Brugia-Elisa) was blindly evaluated using 2487 serum samples from five institutions in three countries, and showed that it is 96–100% sensitive and specific (Rahmah et al. 2001a). Brugia-Elisa has also been evaluated in a blinded field trial in Setiu, Terengganu, Malaysia, where the prevalence of B. malayi infection by thick blood smear was found to be 0.26% (three of 1134) and 2.47% (28 of 1134) by Brugia-Elisa. Thus Brugia-Elisa detected 9.3 times more infections than the traditional method (Lim et al. 2001).

In the official report of the district of Pasir Mas in the state of Kelantan, one case in 2428 samples (0.04%) from the subdistrict of Gual Periok was detected in 1997; and in 2000, two cases were detected in 803 samples (0.25%) from the subdistricts of Gual Periok and Rantau Panjang (Pasir Mas Health Office 1997–2001). However, no cases were reported in 1998 (1385 samples), 1999 (985 samples) and 2001 (1100 samples). These figures probably do not reflect the true prevalence of the infection as it is based on the insensitive thick blood smear examination. In low endemic areas of Peninsular Malaysia, the mf densities in many infected individuals are too low to be detected by this traditional method. In addition, thick blood smears do not allow for the detection of individuals harbouring non-fertile worms, pre-patent infections and single sex infections.

We aimed to determine more accurately the prevalence of B. malayi infections in areas of Pasir Mas which are located near the Thai border. As south Thailand is known to be endemic for brugian filariasis, we expected the study area also to be endemic with higher prevalence rates than previously reported. The results of this study would provide useful data to assist the recently initiated national filariasis elimination programme.

Materials and methods

Study population

Primary schoolchildren, aged 7–12 years, were selected for this study. Other than the convenience of gathering subjects for sampling, they were also expected to be a more reliable indicator of recent infection than a sample of the adult population. Schools located in the subdistricts (mukim) of Pasir Mas that are near the Thailand border were randomly selected. Sixteen national schools (Sekolah Kebangsaan, SK) from six subdistricts were studied, i.e. SK Gelang Mas (436 pupils), SK Meranti (334), SK Bakong (238), SK Kedai Tanjung (325), SK Gual To′Deh (291), SK Gual Periok (166), SK Kubang Kual (310), SK Gual Sitok (384), SK Rantau Panjang (777), SK Seri Rantau Panjang I (710), SK Seri Rantau Panjang II (453), SK Lubok Setol (156), SK Rahmat (136), SK Baroh Pial (213), SK Banggol Petai (66) and SK Bukit Perah (143). These schools are located in the subdistricts of Bunut Susu, Kubang Gadong, Alor Pasir, Gual Periok, Rantau Panjang and Kangkong. Although the subdistrict of Kangkong is not located close to the border, it was included in this study as it is located next to the mukim of Gual Periok where two of three cases were reported by the Pasir Mas Health Office in the last 5 years. Figure 1 shows the location of the schools and subdistricts involved in this study.

Figure 1.

Location of schools and subdistricts of Pasir Mas involved in the study.

In compliance with the University Ethics Committee, prior permission was obtained from the State Director of Education, Education Officer of Pasir Mas, heads of schools, parents and the pupils themselves. The pupils' middle fingers were wiped with 70% ethanol and pricked with a lancet (Becton Dickinson, USA). The blood sample was collected in a labelled microtainer with a gel serum separator (Becton Dickinson). The finger was then cleansed, and the small wound secured with a plaster. The blood samples were transported to the laboratory in cold condition. They were then immediately centrifuged at 5000 g for 15 min at 4 °C and kept at −20 °C.

Preparation of recombinant antigen

BmR1, recombinant antigen was prepared from a recombinant bacteria (BmR17D/PROEX/TOP10) with a B. malayi gene insert. The gene, which probably encodes for a secretory protein, is expressed by infective larva, mf and adult worm of B. malayi; however, the most significant expression was observed in the adult male. A colony of the bacteria was inoculated into Luria Bertani (LB) broth containing 100 μg/ml ampicillin. The flask was incubated at 37 °C overnight in a shaker-incubator. Twenty-five millilitres of the bacterial culture was subsequently transferred to a 500-ml Terrific broth containing 100 μg/ml ampicillin and incubated in a 37 °C shaker-incubator until the optical density (OD) at 600 nm reached 0.5. Next IPTG at 0.1 mm was added and the flask incubated in a 30 °C shaker for 4 h. The culture was then centrifuged at 10 000 g for 10 min at 4 °C. The resulting bacterial pellet was then resuspended in a lysis buffer containing a cocktail of protease inhibitors which inhibits a broad spectrum of serine, cysteine and metalloproteases (‘Complete’, Roche Boehringer, Mannheim, Germany). The mixture was sonicated (Labsonic 2000, B, Braun) at 100–150 V for 10 min. Dnase1 at 5 μg/ml was added to reduce the viscosity of the preparation. This was followed by centrifugation at 12 000 g, 4 °C for 30 min. The resulting supernatant was passed through a 0.22-μm membrane filter; followed by affinity purification through Ni-NTA Superflow resin (Qiagen) using a protein purification system (AKTAprime, Amersham Bioscience). The eluted protein fractions were pooled and the purity assessed by sodium dodecyl sulphate-polyacrylamide gel electrophoresis. Next, the concentration was determined by the Bio-Rad protein assay, and the antigen aliquoted and frozen at −70 °C.


The ELISA procedure was as described previously (Rahmah et al. 2001a). Each well of a microtitre plate (Nunc Immuno Maxisorp, Denmark) was coated with 20 μg/ml protein in NaHCO3 buffer (pH 9.6) and kept at 4 °C overnight in a humid container. This was followed by a washing step with 0.05% Tween-20 in phosphate-buffered saline (PBST), each well being blocked with 1% blocking solution (Roche Boehringer Mannheim) for 0.5 h at 37 °C. Next serum samples were added at 1:50 dilution (in duplicate wells) and incubated at 37 °C for 2 h. After washing off the excess sera, mouse monoclonal antihuman IgG4-HRP (CLB, Netherlands) was added at 1:4500 dilution (in PBS) and incubated for 0.5 h. Following a washing step, 2,2′-Azino-di-3-ethylbenzthiazoline sulphonate (ABTS) substrate (Roche Boehringer Mannheim, Germany) was added and the results were read at 410 nm with an ELISA spectrophotometer (Dynatech, USA). A PBS control, positive and negative controls were included in each plate and the OD readings were blanked with the PBS control. OD readings of ≥0.300 (determined previously) were considered as being positive.


Table 1 shows the summary of results: 5138 pupils, comprising 2439 boys (47.5%) and 2699 girls (52.5%) were screened. Brugia malayi infection was detected in 18 children, giving an overall prevalence of 0.35% (18 of 5138). They comprised 13 boys and five girls; two were 12 years old, three were 11 years old and the others were each 10, 8 and 7 years of age.

Table 1.  Prevalence of Brugia malayi infection among Malaysian school children near Pasir Mas-Thailand border, as determined by Brugia-Elisa
SubdistrictSchoolNo. of pupilsNo. of pupils
positive by
Brugia-Elisa (%)
95% confidence
intervals (CI)
1. Bunut SusuSK Gelang Mas436 1 (0.23%)−0.0022–0.0068
2. Kubang GadongSK Meranti334 0 
SK Bakong238 0 
3. Alor PasirSK Kedai Tanjung325 1 (0.3%)−0.0029–0.0091
4. Gual PeriokSK Gual To'Deh291 2 (0.69%)−0.0026–0.0164
SK Gual Periok166 0 
SK Kubang Kual310 8 (2.6%)0.0082–0.0434
5. Rantau PanjangSK Gual Sitok384 0 
SK Rantau Panjang777 0 
SK Seri Rantau Panjang (1)710 0 
SK Seri Rantau Panjang (2)453 1 (0.22%)−0.0021–0.0065
SK Lubok Setol156 3 (1.92%)−0.0023–0.0408
SK Rahmat136 0 
6. KangkongSK Baroh Pial213 1 (0.47%)−0.0045–0.0139
SK Banggol Petai66 0 
SK Bukit Perah143 1 (0.70%)−0.0067–0.0207
Total 513818 (0.35%)0.0019–0.0051
   (2439 boys,
2699 girls)
(13 boys, five girls)

Prevalence rates according to schools were as follows: SK Kubang Kual recorded the highest rate, i.e. 2.60% (eight of 310), followed by SK Lubok Setol 1.92% (three of 156), SK Bukit Perah 0.70% (one of 143), SK Gual To'Deh 0.69% (two of 291), SK Baroh Pial 0.47% (one of 213), Sk Kedai Tanjung 0.30% (one of 325), SK Gelang Mas 0.23% (one of 436) and SK Seri Rantau Panjang II 0.22% (one of 453). The prevalence rates according to subdistricts were as follows: Gual Periok (1.30%, 10 of 767), Kangkong (0.47%, two of 422), Alor Pasir (0.30%, one of 325), Bunut Susu (0.23%, one of 436), Rantau Panjang (0.15%, four of 2616), Kubang Gadong (0%, none of 572).


Lymphatic filariasis used to be regarded as a disease of adults and children were generally thought to be unaffected. However, more recent studies using sensitive antigen detection ELISAs (which employ Og4C3 monoclonal antibodies) have shown that many children in bancroftian filariasis endemic areas are infected (Lammie et al. 1994; Simonsen et al. 1996; Steel et al. 2001). Thus damage to the lymphatics begins in early childhood and may lead to disfiguring conditions later in their life (Witt & Ottesen 2001). Therefore it is essential that early infection in children be detected and this can be achieved by using a sensitive diagnostic test.

Our study was aimed at determining the prevalence of B. malayi infection in 5138 children in 16 schools, located in six subdistricts. Brugia-Elisa was the sole detection method used as it has been previously shown, in blinded laboratory and field evaluations, to be sensitive and specific for the detection of brugian filariasis (Lim et al. 2001; Rahmah et al. 2001). Our study area is also near (approximately 150 km away) the area where the previous field evaluation was performed. Furthermore, in an earlier study, 737 individuals (comprising all age groups) from six villages in the subdistrict of Gual Periok had been screened by both Brugia-Elisa and microscopy. Two individuals were found to be microfilaraemic (0.27%) and 10 individuals, including the two mf+ individuals, were found to be positive (1.36%).

Eighteen pupils (0.35%) were found to be positive and further investigations revealed that all of them resided in the same subdistrict as the location of their respective schools. The prevalence of infection among boys was 0.53% and that among girls was 0.19%; and the risk of acquiring the infection for boys is 2.88 times greater than for girls (Table 2). The positive cases came from eight schools, located in five subdistricts, i.e. Gual Periok, Kangkong, Alor Pasir, Bunut Susu and Rantau Panjang (in decreasing order of prevalence). Three schools are situated in localities where filariasis has not been detected for 10 years, i.e. SK Gual To'Deh in subdistrict Gual Periok, SK Gelang Mas in subdistrict Bunut Susu and SK Kedai Tanjung in subdistrict Alor Pasir. The latter two are localities in which filariasis has never been reported. This study showed no infected pupils in SK Meranti (334 samples) and SK Bakong (238 samples), both located in the subdistrict of Kubang Gadong.

Table 2.  Analysis of the results on the study of prevalence of B. malayi infection among Malaysian school children near Pasir Mas-Thailand, as determined by Brugia-Elisa
 PositiveNegativeTotalPrevalence (%)95% CI
  1. Relative risk (RR) of boys to girls is 2.88 (95% CI is 0.95–8.73).


Among the eight schools with positive cases, two showed prevalence of more than 1.0%, i.e. SK Kubang Kual (2.6%, eight of 310), and SK Lubok Setol (1.92%, three of 156). The majority of pupils screened came from SK Rantau Panjang (777 samples), but no positive cases were detected in this school. In a nearby school, SK Seri Rantau Panjang, one pupil was found to be infected and this child's home is located right at the border. All eight schools recorded a prevalence of more than 0.1%. This is significant as, according to WHO, an infection prevalence of 0.1% or more will support active transmission of the infection in the community.

It is interesting to note that the prevalence of brugian filariasis (by Brugia-Elisa) in the earlier study involving six villages in subdistrict Gual Periok was 1.36% (10 of 737), thus very similar to the prevalence in the schoolchildren in the same subdistrict, i.e. 1.30% (10 of 767). This shows that the data obtained from sampling of schoolchildren seemed to reflect the prevalence in the general community, thus strengthening WHO's recommendation that children must be screened prior to certifying elimination of the infection.

In conclusion, this study demonstrated that brugian filariasis is endemic in the Pasir Mas areas near the Thailand border. A similar prevalence rate may exist among children in south Thailand areas juxtaposing Pasir Mas. In Malaysia, the target year for elimination of filariasis is 2013; and the programme in the Peninsula employs subdistrict as the implementation unit for the mapping exercise. Thus this study had succeeded in identifying the ‘red’ subdistricts along the Pasir Mas-Thailand border. The detailed report of this study has been submitted to the State Director of Health to enable these subdistricts to be included in the national filariasis elimination programme.


This study was fully funded by the Felda Foundation (No. 304/PPSP/6150025/Y104). We also thank the following people for their valuable assistance: Suharni Mohamad, Mohd. Hafiz Mohamad, Suriati Abd. Ghani, Mehdi Riazi, School Heads, school teachers and the pupils.