Transactions of the Royal Society of Tropical Medicine and Hygiene (2006) 100, 753759

available at www.sciencedirect.com

journal homepage: www.elsevierhealth.com/journals/trst

Estimation of the prevalence of lymphatic lariasis

by a pool screen PCR assay using blood spots
collected on lter paper
Taniawati Supali a,, Is S. Ismid a, Heri Wibowo a, Yenny Djuardi a,
Esther Majawati a,1, Praba Ginanjar a,2, Peter Fischer b,3

a
Department of Parasitology, Faculty of Medicine, University of Indonesia, Salemba 6, Jakarta 10430, Indonesia
b
Bernhard Nocht Institute for Tropical Medicine, Hamburg, Germany

Received 15 June 2005 ; received in revised form 7 October 2005; accepted 20 October 2005
Available online 25 January 2006

KEYWORDS Summary The prevalence of lymphatic lariasis was estimated by PCR-based pool screening of
night blood collected from 865 individuals living in ten areas endemic for Wuchereria bancrofti,
Lymphatic lariasis;
Brugia malayi or B. timori in Indonesia. A total of 232 microlaraemics were identied by
Blood ltration;
ltration of 1 ml of blood. The microlaria (mf) prevalence ranged from 6% to 54%, and the
Pool screen;
mf density in microlaraemics ranged from 1 mf/ml to 6028 mf/ml. PCR assays both for W.
Blood spots;
bancrofti or Brugia spp. detected a single mf present on a 30 l dried lter paper blood spot.
PCR;
One hundred and seventy-eight pools of ve blood spots in each pool (pool-5) were tested by PCR
Indonesia
and 101 (57%) pools were positive. When pool size was increased to 10 spots per pool (pool-10),
65 (70%) of 93 pools were positive. Pearsons correlation and linear regression showed a strong
correlation between ltration and pool screen PCR results for pool-10 (r = 0.835) and pool-5
(r = 0.695). Based on the determination coefcient (R), the results of pool-10 PCR (R = 0.697)
gave a better prediction compared with pool-5 PCR (R = 0.483). This study suggests that pool
screen PCR may be a useful tool for monitoring the Global Program to Eliminate Lymphatic
Filariasis.
2005 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights
reserved.

Corresponding author.
E-mail address: taniawati@yahoo.com (T. Supali).
1 Present address: Department of Parasitology, Faculty of Medicine, Krida Wacana Christian University, Jakarta, Indonesia.
2 Present address: Department of Epidemiology and Infectious Diseases, Faculty of Public Health, University of Diponegoro,

Semarang, Indonesia.
3 Present address: Infectious Diseases Division, Department of Internal Medicine, Washington University School of Medicine,

St Louis, MO 63110, USA.

0035-9203/$ see front matter 2005 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.trstmh.2005.10.005
754
1. Introduction
Lymphatic lariasis caused by Wuchereria bancrofti, Brugia
malayi and B. timori is a poverty-associated major cause of
morbidity in tropical and subtropical countries, affecting an
estimated 128 million people. The disease was therefore tar-
geted by the WHO for elimination by the year 2020 (Ottesen,
2000). In Indonesia, the Global Program to Eliminate Lym-
phatic Filariasis has been launched in 2002 in ve pilot areas.
The key strategy for elimination is an annual mass adminis-
tration (MDA) using diethylcarbamazine (DEC; 6 mg/kg) com-
bined with albendazole (400 mg) for 5 years (Ottesen et al.,
1997). The success of the elimination programme will be
evaluated in randomly selected sentinel sites. The goal of
interrupting transmission may be difcult to achieve in areas
with a high prevalence of microlaraemia and/or areas with
high vector densities, and extensive monitoring of micro-
laria (mf) prevalence may need to be performed during
the entire period of MDA. After cessation of MDA, randomly
selected spot check sites must be evaluated each year to
detect a possible recrudescence of transmission. For that
purpose, large numbers of blood samples (e.g. four sites with
500 samples each) have to be examined (WHO, 2000).
A common diagnostic method used in evaluation of the
lariasis elimination programme is thick lm of night blood,
which is based on microscopic detection of mf in the periph-
eral blood. The sensitivity of the method relies on the blood
volume, the time of blood collection, the mf density and the
skill of the microscopist. Filtration of night blood through
a polycarbonate membrane is usually performed on larger
amounts of blood and has a higher sensitivity compared with
the thick blood lm. Disadvantages of this method are the
relatively high costs of the membranes, the time of blood
collection and the need for venous blood collection. More-
over, neither method is easy to perform consistently under
eld conditions.
Many studies have shown that larial DNA can be
detected in human blood by PCR (Fischer et al., 2003). This
technique was proven to be species-specic and able to
detect ultralow mf densities and sometimes cryptic infec-
tion (Fischer et al., 2000, 2002). However, most PCR assays
are still difcult to perform in many laboratories in lariasis-
endemic countries. Therefore, a simple method of blood col-
lection on lter paper as well as rapid detection of the PCR
products by DNA detection test strips has been developed
(Fischer et al., 2002; Kluber et al., 2001). A major drawback
of this powerful diagnostic approach is the high costs of PCR
for individual samples. To evaluate the effect of control pro-
grammes on the human population, large numbers of blood
samples have to be examined. For the xenomonitoring of
lariasis using vector mosquitoes and to monitor the effect
of intervention on transmission, even larger numbers of vec-
tor samples have to be screened. Therefore, a pool screen
PCR approach has been developed to monitor the transmis-
sion of larial parasites in insect vectors (Williams et al.,
2002). A computer-based algorithm was used to estimate
the actual number of positive samples from the number of
positive pools, the number of pools examined and the num-
ber of vectors per pool (Helmy et al., 2004; Katholi et al.,
1995). Thus far, this pool screen approach has mainly been
used for vector screening. The aim of the present study
was to estimate the prevalence of human lariasis using a
T. Supali et al.
PCR pool screen approach on lter paper blood spots col-
lected in areas endemic for W. bancrofti, B. malayi and B.
timori. The results showed that a simple pool screen PCR
method could be reliably performed on pools of up to 10
samples to estimate the prevalence of lariasis in the human
population.
2. Materials and methods
2.1. Ethical approval
The study was approved by the Ethical Committee of the
Medical Faculty, University of Indonesia, Jakarta (ethi-
cal clearance no. 19/PTO2.FK/ETIK/2002). Informed con-
sent was obtained from adults, and parental consent was
obtained for children. Individuals participating in the study
were registered and their name, sex and age were also
noted.
2.2. Study area
The project was performed on two islands in Indonesia.
Alor Island, East Nusa Tenggara, is endemic for W. bancrofti
and B. timori (Supali et al., 2002) and Central Sulawesi is
endemic for B. malayi (Fischer et al., 2000).
Blood samples were randomly collected from six villages
on Alor Island; three villages (Alila, Wolwal Barat and Wolwal
Tengah) endemic for W. bancrofti and three villages (Malai-
pea, Tominuku and Welai Selatan) endemic for B. timori.
Bancroftian lariasis occurred in coastal areas, with Anophe-
les subpictus as the main vector, which breeds in brackish
water; and B. timori occurred in rice-farming areas, with A.
barbirostris as the vector. In addition, blood samples were
randomly collected from four villages (Mahoni, Rogo Utara,
Rogo Selatan and Mantikole) endemic for nocturnal periodic
B. malayi in Central Sulawesi. In these villages, the main
vector of B. malayi was A. barbirostris, which also breeds
on Sulawesi in rice-farming areas.
2.3. Sample collection
Night blood samples were drawn between 19:00 hours and
23:00 hours. Samples of 2 ml venous blood were collected
from each participant in EDTA-coated vacutainers. From
each sample, 1 ml of blood was used for membrane l-
tration and 30 l of blood was spotted onto each of two
pieces of Whatman 3MM lter paper. Each spot was assigned
either to the assay using ve blood spots in each pool
(pool-5) or to the assay using ten blood spots in each pool
(pool-10). The samples were air dried, stored at room tem-
perature and transported to the laboratory at the Depart-
ment of Parasitology, Faculty of Medicine, University of
Jakarta for DNA preparation and PCR analysis. In addi-
tion, mf-positive blood samples from jirds experimentally
infected with B. malayi and from a microlaraemic W. ban-
crofti patient were available to evaluate the sensitivity
and as positive controls during DNA preparation and PCR.
Blood samples of non-infected jirds were used as negative
controls.
Estimate of lymphatic lariasis prevalence by pool screen PCR
2.4. Parasitological examination
One millilitre of venous night blood was dehaemoglobinised
using distilled water and ltered through a 5 m polycar-
bonate lter membrane (Millipore, Eschborn, Germany).
The lter membrane was transferred to a slide, xed with
methanol, stained with a 1:14 solution of Giemsa for 15 min
and microscopically examined for mf. The number of mf on
the lter membrane was counted using 100-fold magnica-
tion.
2.5. PCR amplication and detection of PCR
products
For each village, one set of lter paper samples was ran-
domly divided into pools of ve samples per pool, whilst the
other set was randomly divided in 10 samples per pool. Para-
sitological assessment of mf was independent from the pool
screen PCR and was performed by different technicians.
Preparation of DNA samples from lter paper, PCR ampli-
cation and subsequent detection of PCR products using DNA
detection test strips were performed as described previously
(Helmy et al., 2004; Kluber et al., 2001). Briey, each of the
30 l blood spots from a pool were cut out using a pair of
sterile scissors and ve or ten spots were placed together in
a microfuge tube. One millilitre of sterile deionised water
was added, incubated for 10 min at room temperature and
vortexed every 2 min. The sample was centrifuged for 2 min
at maximum speed. To bind PCR inhibitors, 10 volumes of a
5% re-suspended chelex-100 resin mixture were added, vor-
texed, incubated for 20 min at 56 C, vortexed briey and
boiled for 8 min. Afterwards, the sample was kept at 4 C
or 20 C for long-term storage. Before use in the PCR, the
chelex-100 was centrifuged for 2 min at maximum speed. As
positive controls, B. malayi or W. bancrofti mf-containing
blood samples were diluted to 1 mf/30 l using non-infected
jird blood and 30 l was spotted on lter paper. The mf
concentration was checked ten times for each species (20
samples) using a thick blood lm and in 16 cases at least one
mf was present. The mf-positive blood spots were pooled
with four or nine mf-negative blood spots for DNA prepara-
tion.
The PCR assays for the detection of B. malayi and B.
timori DNA targeted the HhaI repetitive sequences (Fischer
et al., 2002; Lizotte et al., 1994). The PCR assays for the
detection of W. bancrofti DNA amplied the SspI dispersed
repeat (Fischer et al., 1999; Zhong et al., 1996). The pres-
ence of amplied PCR products was checked using specic
DNA probes and DNA detection test strips (Roche, Mannheim,
Germany) as described earlier (Kluber et al., 2001). The
appearance of two clear lines on the test strips was noted
as a positive PCR result.
2.6. Data analysis
The PCR pool data were analysed by the computer pro-
gram PoolScreen version 2.0 to estimate the prevalence
of lariasis in the human population (Helmy et al., 2004;
Katholi et al., 1995). Since the data were normally dis-
tributed (Kolmogorov-Smirnov P > 0.05), a paired t-test was
used to compare the prevalences from ltration and the pool
755
screen PCR. Pearsons correlation test was used to deter-
mine the correlation between the PCR pool and membrane
ltration methods, and this was followed by constructing a
linear regression model.
3. Results
3.1. Parasitological examination
A total of 865 individuals living in ten lariasis-endemic vil-
lages on Alor and Sulawesi Islands participated in night blood
collection (Table 1). The number of samples collected in
each area varied between 33 and 83 in B. malayi villages,
between 46 and 111 in B. timori villages, and between 53
and 194 in W. bancrofti villages. The age of the partici-
pants ranged from 1 year to 75 years. One hundred and
twenty-seven individuals were less than 10 years old, 224
were 1019 years, 486 were 2059 years and 28 were >60
years. The study population consisted of 470 males (54.3%)
and 395 females (45.7%).
The prevalence of microlaraemia ranged from 6% to 41%
in B. malayi areas, from 37% to 54% in B. timori areas,
and from 6% to 21% in W. bancrofti areas (Table 1). The
total number of microlaraemics was 232; 59% of the micro-
laraemics were male and 41% were female. The highest
mf rates were found in men and women aged 20 years
and older. The mf density per ml of night blood ranged
between 1 mf/ml and 6028 mf/ml. However, in all study
villages the geometric mean mf density was usually low
(Table 1). In the B. malayi area on Sulawesi, low dosage DEC
was repeatedly distributed to a number of microlaraemics
approximately 2 years before sample collection, whereas
on Alor the mf density of B. timori or W. bancrofti was low
because 10 months prior to sample collection one round of
DEC/albendazole MDA was performed. The coverage rate
of MDA on Alor was approximately 75% (Oqueka et al.,
2005).
3.2. Pool screen PCR
Both the Brugia spp. PCR assay and the W. bancrofti PCR
assay were able to detect one mf present on a 30 l blood
spot in pools of ve (pool-5) or ten (pool-10) blood spots.
Using the B. malayi control samples, eight of ten pool-5
samples and seven of ten pool-10 samples were positive,
whilst for W. bancrofti seven of ten samples were posi-
tive for both pool-5 and pool-10. This result corresponds
well with the thick blood lm and gives a sensitivity of one
in pools of up to ten blood spots. Since sensitivity crite-
ria were fullled, eld samples from areas endemic for B.
malayi, B. timori and W. bancrofti were examined. A total
of 178 pools of pool-5 samples were tested by PCR and 101
(57%) pools showed positive results on DNA detection strips.
Among 93 pool-10 samples, 65 (70%) pools were tested pos-
itive (Table 2). In a few cases, the pool size was smaller
than ve or ten samples per pool, but the PoolScreen soft-
ware version 2 can manage varying pool sizes. Estimation of
mf rate using the PoolScreen version 2 algorithm showed a
similar range for pool-5 and pool-10. However, there was a
trend for higher estimates for pool-5 compared with pool-10
(Table 3).
756 T. Supali et al.

Table 1 Microlaria (mf) rate in a study population of 865 individuals from ten villages endemic for Brugia malayi, B. timori
or Wuchereria bancrofti in Indonesia

Villages No. examined/No. mf positive mf

Males Females Rate (%) Densitya

<20 years 20 years <20 years 20 years

B. malayi
Mahoni 14/3 46/15 2/1 21/4 28 4
Rogo Utara 2/1 28/12 2/1 22/8 41 5
Rogo Selatan 7/1 21/11 8/1 10/0 28 3
Mantikole 7/1 13/0 4/0 9/1 6 1
Total mf rate (%) 20 35 19 21 28 3
B. timori
Welai Selatan 29/5 27/15 29/12 26/9 37 5
Tominuku 21/8 16/11 18/8 29/17 52 18
Malaipea 20/10 9/7 10/4 7/4 54 17
Total mf rate (%) 33 63 42 48 46 10
W. bancrofti
Alila 17/0 14/2 13/1 9/0 6 1
Wolwal Barat 38/4 63/21 36/4 57/12 21 3
Wolwal Tengah 33/1 45/10 41/1 42/6 11 1
Total mf rate (%) 6 27 7 17 15 2
The mf rates in male and female children/young adults and in adults were assessed by membrane ltration of 1 ml of venous night blood.
a Total mf density was calculated as the geometric mean number of mf per ml in microlaraemic individuals.

3.3. Comparison between ltration and pool
screen PCR results
The prevalences of lariasis obtained by PoolScreen version
2 analysis on PCR results of pool-5 and pool-10 were
lower than the prevalence results revealed by membrane
ltration. The data were checked using the Kolmogorov-
Smirnov test and it was found that the prevalence results
by either ltration, pool-5 and pool-10 PCR were normally
distributed (P = 0.440, P = 0.304 and P = 0.767, respectively).
Statistical analysis by paired t-test showed that there was
a signicant difference in the ltration result compared
with pool screen PCR results for pool-5 (P = 0.025) as well as
for pool-10 (P = 0.003). However, no signicant difference
in prevalence results of pool-5 compared with pool-10
(P = 0.077) was found. To identify the association between

Table 2 Total number of samples examined and results of the PCR screening of blood spots in pools of ve (pool-5) and ten
(pool-10) lter paper blood spots

Village No. of samples examined pool-5 pool-10

No. of pools No. of positive pools No. of pools No. of positive pools

B. malayi 216 45 34 24 17
Mahoni 83 17 15 9 6
Rogo Utara 54 11 10 6 5
Rogo Selatan 46 10 7 5 4
Mantikole 33 7 2 4 2
B. timori 241 50 35 26 23
Welai Selatan 111 23 14 12 11
Tominuku 84 17 15 9 8
Malaipea 46 10 6 5 4
W. bancrofti 408 83 32 43 25
Alila 53 11 1 6 2
Wolwal Barat 194 39 19 20 15
Wolwal Tengah 161 33 12 17 8

Total 865 178 101 93 65

Each blood spot contained approximately 30 l of night blood collected in ten villages endemic either for Brugia malayi, B. timori or
Wuchereria bancrofti.
Estimate of lymphatic lariasis prevalence by pool screen PCR 757

the prevalence of microlaraemia. The sensitivity to detect

Table 3 Comparison of microlaria rate as determined by
mf is largely dependent on the amount of blood examined.
ltration of 1 ml of night blood or by the PCR screening of
In the present study, membrane ltration of 1 ml of night
lter paper blood spots containing 30 l of night blood per
blood was used as a reference method and compared with
sample in pools of ve (pool-5) or ten (pool-10) samples
the PCR screening for mf in pools of ve and ten lter paper
Village Microlaria rate (%) blood spots of 30 l each. It is not surprising that the sensi-
tivity of mf detection by ltration is higher compared with
Filtration pool-5 pool-10 pool screen PCR. However, the strong correlation of ltra-
B. malayi 28 25 12 tion results with pool screen PCR results suggests that pool
Mahoni 28 35 10 screen PCR is suitable for assessing the prevalence of micro-
Rogo Utara 41 38 16 laraemia.
Rogo Selatan 28 21 15 PCR-based screening of pools of blood samples for infec-
Mantikole 6 7 7 tious agents is a common strategy, especially in transfusion
medicine (Roth and Seifried, 2001; Stolz et al., 2003; Yang et
B. timori 46 21 19 al., 2001). For example, several millions of blood donor sam-
Welai Selatan 37 17 22 ples have been tested for hepatitis C virus and HIV-1 using
Tominuku 52 35 20 minipool PCR of up to 98 samples per pool (Roth et al., 2002).
Malaipea 54 17 15 In such large-scale screening projects, dilution of viral load
W. bancrofti 15 9 8 during the pooling of samples was prevented using cen-
Alila 6 2 4 trifugation for sample concentration, and standardisation
Wolwal Barat 21 13 13 was achieved using commercial kits for DNA extraction and
Wolwal Tengah 11 9 6 PCR. Our results indicate that the pool screening approach
may also be suitable for prevalence estimation of micro-
Total 27 15 11 laraemia for monitoring lariasis interventions. However,
Infection rate based on pool screening was estimated using the more operational research would be needed to achieve a
PoolScreen program. similar capacity and sensitivity of the method as developed
for viral agents. Therefore, our project has to be considered
as a pilot study to evaluate the potential of the strategy.
ltration and pool screen PCR results, Pearsons correlation
A number of PCR assays for the detection of DNA of lym-
and linear regression were used for further analysis. The
phatic lariae have been available for more than a decade
results showed that the correlation coefcient of pool-10
(Lizotte et al., 1994; Zhong et al., 1996). However, the role
was stronger (r = 0.835) compared with pool-5 (r = 0.695)
of PCR assays on human blood for monitoring the Global Pro-
(Table 4). Both pool sizes for pool screen PCR gave a positive
gram to Eliminate Lymphatic Filariasis is not well dened.
correlation, which proved the hypothesis that the mf rate
Major drawbacks of PCR assays are the requirement for well
determined by pool-5 or pool-10 PCR is an indicator for mf
trained personnel and the relatively high costs. PCR-based
prevalence as determined by ltration.
methods are nowadays also widely used in developing coun-
The determination coefcient (R), which is used to pre-
tries and a pool screening approach would reduce costs
dict the accuracy of the pool screen method in replacing
enormously. It is estimated that the cost of 10 polycarbon-
the ltration method, showed that the determination coef-
ate lters for blood ltration (approximately US$9) is at
cient in pool-10 (R = 0.697) was higher compared with pool-5
least three times higher compared with one PCR reaction
(R = 0.483). The results indicate that pool-10 gave 70% accu-
for a pool of 10 blood spots. A PCR thermocyler is usually
racy in predicting mf rate compared with ltration, whereas
no more expensive than a good research microscope. Prepa-
pool-5 gave an accuracy of 48%. The linear regression model
ration of DNA from 100 pools of lter paper blood spots,
of pool-5 was Y = 9.870 + 0.964X, whilst the linear regres-
PCR and detection of PCR products can easily be performed
sion model of pool-10 was Y = 3.476 + 2.492X (Y = ltration
on a single day, whereas staining and microscopic examina-
prevalence, X = pool screen PCR prevalence).
tion of 1000 individual samples usually takes at least three
times as long. A further advantage of PCR-based detection
4. Discussion of larial parasites is the easy storage of genetic material for
later studies, e.g. for molecular epidemiology if drug resis-
This study showed that a pool screen PCR approach using tance develops. Multiplex assays can be used to detect W.
blood spots collected on lter paper can be used to estimate bancrofti and B. malayi in a single reaction. Real-time PCR

Table 4 Pearsons correlation and linear regression results for the association between ltration and PCR pool screening on
pools of ve (pool-5) or ten (pool-10) blood spots

Variable r R Linear regression model P-value

pool-5 PCR 0.695 0.483 Y = 9.870 + 0.964X 0.026

pool-10 PCR 0.835 0.697 Y = 3.476 + 2.492X 0.003
r, regression coefcient; R, determination coefcient; Y, prevalence based on parasitological examination; X, prevalence based on pool
screen PCR result.
758
assays require expensive equipment but offer the advantage
that the analysis of PCR products is performed automatically
and standardisation of methods is simplied. Sensitive and
highly specic real-time PCR assays for lymphatic lariae
have been developed and are a promising new monitoring
tool (Rao and co-workers, personal communication).
In many countries, lymphatic lariasis is endemic in
remote areas where transportation is limited and where it
is not possible to perform even the simplest diagnostic pro-
cedure such as thick blood lms or blood ltration. Rapid
antigen testing may be performed for W. bancrofti, but this
is not available for Brugia spp. (Weil et al., 1997). There-
fore, knowledge regarding the distribution of W. bancrofti
is a prerequisite to employing antigen detection as a diag-
nostic tool. In remote areas of Asia where it is not known
which larial species causes lymphatic lariasis, collection
and preservation of capillary blood on lter paper for pool
screening by PCR would be an alternative that can be easily
performed by local health workers. Samples can be trans-
ported without problems to a central reference laboratory
where standardisation of PCR assays and quality control can
be ensured. Proper monitoring of lariasis control requires
the selection of sentinel and spot check sites. These sites
are often selected according to their accessibility for the
external survey team. This may cause a bias, because MDA
may also be more efcient in these areas. Therefore, col-
lection of blood spots on lter paper may offer an alter-
native to include villages that are more difcult to reach
but where local staff is able to perform the night blood
collection.
The volume of blood used for parasitological and PCR-
based diagnostics of lariasis a major determinant of the
sensitivity of the assay. It is known that membrane ltration,
usually performed with 1 ml of blood, has a higher sensitiv-
ity compared with a thick blood lm, which is usually done
with 3060 l of blood (Kimura et al., 1984). The differ-
ence in sensitivity is dependent on the mf density: in areas
where most infected individuals have mf densities above
33 mf/ml, the difference is small; and in areas where most
infected individuals have lower mf densities (e.g. due to
MDA) this difference can be large. This can explain the dif-
ference between ltration results and PCR pool screening in
our study, since PCR was performed on pools of blood spots
of 30 l each.
Based on the determination coefcient (R), the results
of pool-10 PCR gave a better estimate of microlaraemia
prevalence compared with pool-5 PCR. One explanation for
this would be that for pool-5 the relationship of target
sequence and PCR inhibitors leads to a higher sensitivity.
Although as little as one mf present in pool-5 or pool-10 can
reliably be detected, it cannot be excluded that remnants
of mf or free DNA were also detected in pool-5. Taking into
account that the volume of blood per sample examined by
pool screen PCR was 1/33 of the volume used for ltration,
both pool sizes gave a good estimate of prevalence. How-
ever, for monitoring a lariasis elimination programme, the
use of pool-10 is more efcient and cheaper compared with
pool-5.
Taken together, a pool screen method for the detection
of lymphatic lariae of B. malayi, B. timori and W. bancrofti
in blood spots collected on lter paper was established. The
results suggest that pool screen PCR can be a useful tool
T. Supali et al.
for monitoring the Global Program to Eliminate Lymphatic
Filariasis.
Conicts of interest statement
The authors have no conicts of interest concerning the work
reported in this paper.
Acknowledgements
The authors would like to thank the District Health Authority
and the health workers for their encouragement and help
during the eldwork. The authors are also grateful to all
inhabitants of the villages who participated in the study.
The authors wish to thank Tom Unnasch for providing the
PoolScreen program version 2, and Edwin Michael for sta-
tistical advice. This investigation received nancial support
from the UNDP/World Bank/WHO Special Programme for
Research and Training in Tropical Diseases (TDR) and from
GlaxoSmithKline, UK.
References
Fischer, P., Liu, X., Lizotte-Waniewski, M., Kamal, I.H., Ramzy, R.M.,
Williams, S.A., 1999. Development of a quantitative, competi-
tive polymerase chain reactionenzyme-linked immunosorbent
assay for the detection of Wuchereria bancrofti DNA. Parasitol.
Res. 85, 176183.
Fischer, P., Supali, T., Wibowo, H., Bonow, I., Williams, S.A., 2000.
Detection of DNA of nocturnally periodic Brugia malayi in night
and day blood samples by a polymerase chain reactionELISA-
based method using an internal control DNA. Am. J. Trop. Med.
Hyg. 62, 291296.
Fischer, P., Wibowo, H., Pischke, S., Ruckert, P., Liebau, E., Ismid,
I.S., Supali, T., 2002. PCR-based detection and identication of
the larial parasite Brugia timori from Alor Island, Indonesia.
Ann. Trop. Med. Parasitol. 96, 809821.
Fischer, P., Boakye, D., Hamburger, J., 2003. Polymerase chain
reaction-based detection of lymphatic lariasis. Med. Microbiol.
Immunol. (Berl.) 192, 37.
Helmy, H., Fischer, P., Farid, H.A., Bradley, M.H., Ramzy, R.M., 2004.
Test strip detection of Wuchereria bancrofti amplied DNA in
wild-caught Culex pipiens and estimation of infection rate by a
PoolScreen algorithm. Trop. Med. Int. Health 9, 158163.
Katholi, C.R., Toe, L., Merriweather, A., Unnasch, T.R., 1995. Deter-
mining the prevalence of Onchocerca volvulus infection in vector
populations by polymerase chain reaction screening of pools of
black ies. J. Infect. Dis. 172, 14141417.
Kimura, E., Penaia, L., Spears, G.F., 1984. Comparison of meth-
ods for the detection of microlariae of Wuchereria bancrofti in
Western Samoa. Southeast Asian J. Trop. Med. Public Health 15,
167174.
Kluber, S., Supali, T., Williams, S.A., Liebau, E., Fischer, P., 2001.
Rapid PCR-based detection of Brugia malayi DNA from blood
spots by DNA Detection Test Strips. Trans. R. Soc. Trop. Med.
Hyg. 95, 169170.
Lizotte, M.R., Supali, T., Partono, F., Williams, S.A., 1994. A poly-
merase chain reaction assay for the detection of Brugia malayi
in blood. Am. J. Trop. Med. Hyg. 51, 314321.
Oqueka, T., Supali, T., Ismid, I.S., Purnomo, Ruckert, P., Bradley,
M., Fischer, P., 2005. Impact of two rounds of mass drug admin-
istration using diethylcarbamazine combined with albendazole
on the prevalence of Brugia timori and of intestinal helminths
on Alor Island, Indonesia. Filaria J. 4, 5.
Ottesen, E.A., 2000. The global programme to eliminate lymphatic
lariasis. Trop. Med. Int. Health 5, 591594.
Estimate of lymphatic lariasis prevalence by pool screen PCR
Ottesen, E.A., Duke, B.O., Karam, M., Behbehani, K., 1997. Strate-
gies and tools for the control/elimination of lymphatic lariasis.
Bull. World Health Organ. 75, 491503.
Roth, W.K., Seifried, E., 2001. Yield and future issues of nucleic acid
testing. Transfus. Clin. Biol. 8, 282284.
Roth, W.K., Weber, M., Buhr, S., Drosten, C., Weichert, W., Sireis,
W., Hedges, D., Seifried, E., 2002. Yield of HCV and HIV-1 NAT
after screening of 3.6 million blood donations in central Europe.
Transfusion 42, 862868.
Stolz, M., Gilgen, M., Niederhauser, C., 2003. Hepatitis C virus-
polymerase chain reaction minipool testing: 3 years in the
largest Swiss blood transfusion service. Vox Sang. 84, 105110.
Supali, T., Wibowo, H., Ruckert, P., Fischer, K., Ismid, I.S., Purnomo,
Djuardi, Y., Fischer, P., 2002. High prevalence of Brugia timori
infection in the highland of Alor Island, Indonesia. Am. J. Trop.
Med. Hyg. 66, 560565.
Weil, G.J., Lammie, P.J., Weiss, N., 1997. The ICT Filariasis Test: a
rapid-format antigen test for diagnosis of bancroftian lariasis.
Parasitol. Today 13, 401404.
759
WHO, 2000. Preparing and implementing a national plan to elimi-
nate lymphatic lariasis. World Health Organization, Geneva.
Williams, S.A., Laney, S.J., Bierwert, L.A., Saunders, L.J., Boakye,
D.A., Fischer, P., Goodman, D., Helmy, H., Hoti, S.L., Vasuki, V.,
Lammie, P.J., Plichart, C., Ramzy, R.M., Ottesen, E.A., 2002.
Development and standardization of a rapid, PCR-based method
for the detection of Wuchereria bancrofti in mosquitoes, for
xenomonitoring the human prevalence of bancroftian lariasis.
Ann. Trop. Med. Parasitol. 96 (Suppl. 2), S41S46.
Yang, Y., Lamendola, M.H., Mendoza, M., Xu, D., Nguyen, M., Yeh,
S., Wu, Y., Ku, J., Rosenstraus, M., Sun, R., 2001. Performance
characteristics of the COBAS AmpliScreen HIV-1 test, version 1.5,
an assay designed for screening plasma mini-pools. Transfusion
41, 643651.
Zhong, M., McCarthy, J., Bierwert, L., Lizotte-Waniewski, M.,
Chanteau, S., Nutman, T.B., Ottesen, E.A., Williams, S.A., 1996.
A polymerase chain reaction assay for detection of the parasite
Wuchereria bancrofti in human blood samples. Am. J. Trop. Med.
Hyg. 54, 357363.