Acta Tropica 131 (2014) 63–70

Contents lists available at ScienceDirect

Acta Tropica
journal homepage: www.elsevier.com/locate/actatropica

Comparison of the performance of two spontaneous sedimentation

techniques for the diagnosis of human intestinal parasites in the
absence of a gold standard
Alessandra Queiroga Gonçalves a,b,∗ , Rosa Abellana b , Hélio Doyle Pereira-da-Silva b ,
Ivanildes Santos a , Paula Taquita Serra a , Genimar Rebouças Julião a,1 ,
Patricia Puccinelli Orlandi a , Carlos Ascaso b,c
a
Instituto Leônidas e Maria Deane – Fiocruz Amazônia, rua Teresina 476, Adrianópolis, Manaus, Amazonas 69057-070, Brazil
b
Departament de Salut Pública, Facultat de Medicina, Universitat de Barcelona, C/ Casanova 143, Barcelona 08036, Spain
c
IDIBAPS, Institut d’Investigacions Biomèdiques August Pi i Sunyer, Barcelona, Spain

a r t i c l e i n f o a b s t r a c t

Article history: Performance evaluation of diagnostic tests is critical in the search for accurate diagnoses. A gold standard
Received 23 July 2013 test is usually absent in parasitology, thus rendering satisfactory assessment of diagnostic accuracy dif-
Received in revised form 29 October 2013 ficult. Moreover, reliability (assessed by the study of repeatability) is a rarely studied characteristic of
Accepted 29 November 2013
diagnostic tests. This study compared and evaluated the performance (repeatability, concordance and
Available online 7 December 2013
accuracy) of the spontaneous sedimentation technique (SST) and the Paratest for the diagnosis of Giar-
dia lamblia, Entamoeba histolytica complex, Blastocystis spp., Ascaris lumbricoides, hookworm, Trichuris
Keywords:
trichiura and Calodium hepaticum. Fecal samples of 143 individuals were separated into three replicates
Intestinal parasites
Spontaneous sedimentation techniques
for each test. Concordance and homogeneity of the results between replicates of each test and between
Diagnosis tests were evaluated. Proportions of positives, sensitivity and specificity were estimated using a Bayesian
Repeatability Latent Class Model. High repeatability of both tests was found for the detection of intestinal parasites,
Accuracy except for Blastocystis spp. and hookworm. Concordance between tests was generally high (concordance
Bayesian analysis correlation coefficient, 0.72–0.88), except for Blastocystis spp., hookworm and T. trichiura. The Paratest
detected more cases of Blastocystis spp. and fewer of hookworm than the SST. The tests were quite dis-
cordant in the detection of T. trichiura. A low sensitivity (39.4–49.2% for SST, 35.8–53.8% for Paratest)
and a high specificity (93.2–97.2%) were found for both tests. The Paratest presented a slightly higher
sensitivity for the diagnosis of Blastocystis spp. (53.8%), and SST did so for hookworm (49.2%). This is the
first study on repeatability and accuracy (using a Bayesian approach) of two spontaneous sedimentation
techniques. These results suggest underdiagnosis of little dense parasitic forms due to technical limita-
tions in both tests. We conclude that the combined study of repeatability, concordance and accuracy is
a key strategy for better evaluation of the performance of tests and is also useful for the identification of
technical limitations.
© 2013 Elsevier B.V. All rights reserved.

1. Introduction in regions in the developing world (Alum et al., 2010). Of par-

ticular importance worldwide are the soil-transmitted helminths
Intestinal parasites are, even today, major contributors to the Ascaris lumbricoides, hookworm and Trichuris trichiura and the
global burden of disease, affecting especially the population living protozoans Entamoeba histolytica and Giardia lamblia (Bethony
et al., 2006; Fenwick, 2012; Harhay et al., 2010). In the Ama-
zon region, Calodium hepaticum is a zoonotic helminth that has
been increasingly reported as a cause of spurious infection in
Abbreviations: SST, spontaneous sedimentation technique; Kappa IRM, kappa
index for repeatability measure; Kappa IBT, kappa index between tests; CCC, con- humans (Gonçalves et al., 2012), and Blastocystis spp. is a highly
cordance correlation coefficient; CI, confidence interval; CrI, credible interval. prevalent suspected pathogenic protozoan (Borges et al., 2009). C.
∗ Corresponding author at: Departament de Salut Pública, Facultat de Medicina, hepaticum is also the causative agent of a rarely reported liver dis-
Universitat de Barcelona, C/ Casanova 143, Barcelona 08036, Spain. ease (hepatic calodiasis) found worldwide. This helminth infects
Tel.: +34 934035269; fax: +34 934035270.
the hepatic parenchyma of various mammals (rodents being the
E-mail address: [email protected] (A.Q. Gonçalves).
1
Present address: Fundação Oswaldo Cruz – Fiocruz Rondônia, Rua da Beira 7671, principle hosts). Hepatic infection occurs following the ingestion
Lagoa, Porto Velho, Rondônia 76812-24, Brazil. of embryonated eggs present in the ground or contaminated food.

0001-706X/$ – see front matter © 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.actatropica.2013.11.026
64 A.Q. Gonçalves et al. / Acta Tropica 131 (2014) 63–70

In spurious infection, non-embryonated eggs are ingested (from Paulo, Brazil) is a commercial kit for spontaneous sedimentation of
the ground, contaminated food or liver of mammals) and directly preserved stool, developed with the aim of expanding new meth-
exit with the stools without causing liver disease (Fuehrer et al., ods based on the simplification of laboratory procedures thereby
2011; Gonçalves et al., 2012). Transmission of intestinal para- improving biosecurity (Brandelli et al., 2011). The kit provides a
sites depends on the availability of clean water, socio-economic stool container that has a cap equipped with a filter of 266 ␮m.
conditions, education, personal and public hygiene practices, tem- This characteristic synthesizes the manipulation and examination
perature, humidity and the survival of the environmental stages of of stool samples by performing the steps of conservation, filtration
the parasites (Alum et al., 2010). and concentration in the container itself. The amount of feces is
Evaluation of the performance of diagnostics tests is critical in standardized (2 g), whereas variable quantities can be used (1–5 g)
the search for accurate diagnostic techniques to provide adequate in the SST (Brandelli et al., 2011; De Carli, 2007a; Hoffman et al.,
patient care, assess drug efficacy, monitor the effectiveness of con- 1934). The Paratest is faster (15 or 30 min of sedimentation) than
trol programs and obtain better understanding of the epidemiology the SST and its compact structure allows the performance of the test
of intestinal parasites (Harhay et al., 2011; Tarafder et al., 2010). in remote places. Despite the widespread use of the spontaneous
In order to evaluate diagnostic tests it is important to take into sedimentation techniques, no study has evaluated their repeatabil-
account that few, if any, gold standard tests (i.e. a diagnostic test ity and accuracy taking into account the absence of a gold standard
with 100% accuracy against which the sensitivity and specificity of in the latter case.
other tests are estimated) are available in parasitology and, in par- The aim of this study was to evaluate and compare the
ticular, do not exist for the detection of intestinal parasite infection performance (repeatability, concordance and accuracy) of two
(Basso et al., 2013; Tarafder et al., 2010). Nevertheless, most studies spontaneous sedimentation techniques (SST and Paratest) in
estimating the sensitivity and specificity of tests for the diagno- the detection of infection by several pathogenic (or suspected
sis of intestinal parasites consider the results of one of two tests pathogenic) intestinal parasites (G. lamblia, E. histolytica complex,
compared (usually the traditional test) or the combination of the Blastocystis spp., A. lumbricoides, hookworm, T. trichiura and C.
results of several diagnostic tests as the gold standard (Brandelli hepaticum), using a Bayesian approach for the estimation of the
et al., 2011; Carvalho et al., 2012; Devera et al., 2008a; Dogruman- proportion of positives, sensitivity and specificity.
Al et al., 2010; Glinz et al., 2010; Inês et al., 2011; Knopp et al., 2011;
Levecke et al., 2011; Steinmann et al., 2012). These practices have
2. Materials and methods
led to biased estimations of accuracy. The use of statistical models
that consider the assumption of absence of a gold standard test can
2.1. Study area and population
overcome this problem, generating more reliable information as to
the accuracy of diagnostic tests. Up to now, only three articles in
This study was carried out in 2009 with the collection of stool
the area of human intestinal parasites have presented estimations
samples from children and adults from the agricultural community
of the sensitivity and specificity of diagnostic tests using the con-
of Rio Pardo of the municipality of Presidente Figueiredo, located
cept of absence of a gold standard test (Booth et al., 2003; Tarafder
∼160 km to the north of the city of Manaus (∼1◦ 48 S; 60◦ 19 W),
et al., 2010; Traub et al., 2009).
Amazonas State, Brazil.
Another important aspect during the evaluation of the perfor-
mance of diagnostic tests is the repeatability of the results (Sanchez
et al., 2002). Repeatability refers to the extent of agreement among 2.2. Field and laboratory procedures
repeat assessments of the same sample using the same technique
in the same laboratory by the same operator (Braun-Munzinger Participants were asked to submit one fresh stool sample. The
and Southgate, 1992; White and van den Broek, 2004). Although collection of samples was conducted in the households with two
the repeatability of a test refers to its reliability (White and van daily visits of the staff of the project, once in the morning and
den Broek, 2004), this important characteristic has been little eval- another in the afternoon. The samples were initially processed
uated in studies of the performance of diagnostic tests (Charlier 1–3 h after collection in a local laboratory unit located in the
et al., 2005; Sanchez et al., 2002; Thomas et al., 1981). community, as follows: firstly, thorough homogenization of each
Among the diagnostic tests based on optical microscopy those specimen was performed by stirring with a wooden spatula for
based on spontaneous sedimentation are among the least expen- at least 1 min. After homogenization, each sample was sepa-
sive and easiest to perform and enable the simultaneous detection rated into three equal replicates of feces for each test. For the
of helminth and protozoan intestinal infections (Brandelli et al., Paratest the replicates were deposited into three different plas-
2011; Camacho et al., 2013; Carvalho et al., 2012; Ribeiro and Furst, tic stool containers provided by the Paratest kit using a device
2012; Tello et al., 2012). For these reasons, in some economically that enables the collection of 1 g of feces. Each replicate was com-
underdeveloped settings their use is preferred over the tests based posed of 2 g of feces diluted in 7 ml of the preservative (5% buffered
on centrifuge-sedimentation or centrifuge-flotation. Nevertheless, formalin pH 7.0) contained in each container. For the SST the repli-
techniques based on centrifugation have demonstrated to be bet- cates were deposited into three different containers with sodium
ter in relation to those based only on spontaneous sedimentation acetate–acetic acid–formaldehyde (SAF). For each replicate of the
(Carvalho et al., 2012; Gomes et al., 2004), although exceptions have SST, 2 g of feces (measured with the device provided by the Paratest
been reported (Devera et al., 2008a; Tello et al., 2012). kit) were diluted in 7 ml of SAF. In the case of a diarrheal sample,
The spontaneous sedimentation technique (SST) (also known as three measurements of the device provided by the Paratest kit were
the Lutz technique or the Hoffman, Pons and Janer technique) is applied for the two tests. Samples that could not achieve the total
a traditional test widely used for clinical diagnosis and epidemio- of six replicates due to the lack of a sufficient quantity of feces were
logical surveys in Brazil and, also, Venezuela (Brandelli et al., 2011; not included in the study.
Carvalho et al., 2012; de Souza et al., 2007; Devera et al., 2008a,b; The two sedimentation techniques were processed and exam-
Pinheiro et al., 2011; Santos et al., 2013; Velásquez et al., 2005). In ined in the Leonidas e Maria Deane Institute (Fiocruz, Manaus)
this test, stool samples (previously preserved or diluted in water) by an experienced laboratory technician. The delay in time from
are filtered through a gauze strip into a conical cup and subse- stool sample processing to microscopic reading ranged from 3 to 17
quently submitted to sedimentation in tap water for 1 or 2 h (De days. The Paratest was carried out according to the manufacturer’s
Carli, 2007a). On the other hand, the Paratest (DK Diagnostics, São instructions. In brief, each container with the diluted feces was
 A.Q. Gonçalves et al. / Acta Tropica 131 (2014) 63–70 65

moderately shaken, the sediment exit (located in the cap) was taken The median and the 95% credible interval of these samples
off and the container was inverted and placed onto a polystyrene were used as point and interval estimation of the parameter.
tray for the spontaneous sedimentation of the fecal suspension. The model was fitted with the WinBUGS 1.4 software (Spiegel-
Thirty minutes later, two drops of the sediment from each con- halter DJ, Thomas A, Best NG, 2004. WinBUGS version 1.4.;
tainer were placed on a slide and stained with lugol for subsequent http://www.mrc-bsu.cam.ac.uk/bugs/winbugs/contents.shtml).
microscopic examination. The SST was based on Lutz (1919) and
Hoffman et al. (1934). Each replicate of conserved feces was filtered 2.4. Ethics statement
through gauze folded twice and the filtrate was received in a 125 ml
polystyrene conical cup. Tap water was added to the filtrate up to a This study was approved by the Ethics Committee in Inves-
volume of 3/4 of the cup. The suspension was allowed to stand for tigation of the Oswaldo Cruz Foundation (Protocol 384/07 of
2 h, and after this period, part of the sediment was collected with a 20/08/2007). Written Informed consent was obtained from all the
pipette, and one drop was placed on a slide and stained with lugol study participants.
for subsequent microscopic examination.
3. Results
2.3. Statistical analysis
3.1. Parasitological findings by diagnostic method
The number of positive results and their percentages for each
intestinal parasite were calculated separately for each replicate A total of 143 stool samples were subjected to the analysis of
of the tests. The number (and percentages) of discordant results three replicates by each test. The cumulative percentages of pos-
between tests were calculated for each pair of replicates compared. itive cases (considering the results of the replicates of each test)
These descriptive analyses were also performed with the combina- as well as the total percentages of positive cases observed are pre-
tion of the results of the replicates of each test and considering a sented by intestinal parasite in Fig. 1. The examination of more
positive result when at least one of the three replicates was posi- than one replicate of the same stool sample by the SST did not rep-
tive. This latter information is referred to herewith as a “combined resent an increment in the percentage of positives encountered for
result”. G. lamblia, E. histolytica complex, A. lumbricoides, T. trichiura and C.
The Cochran Q test was used to test intra-test homogeneity, hepaticum. The examination of the three replicates by the SST (in
that is, that assessed if the percentage of positive results was the comparison to only one replicate) led to an increase in the percent-
same among the three replicates of each test. A generalization of age of positives for Blastocystis spp. (18.9% to 26.6%) and hookworm
the kappa index (Broemeling, 2009) (here named “kappa index (14.7% to 21.7%).
for repeatability measure (Kappa IRM)”) was calculated for the In relation to the results obtained with the Paratest, increases
assessment of diagnostic agreement among the three replicates were also observed in the percentage of positives for hookworm
of each test. The McNemar test was used to test the homogene- (6.3% to 11.9%), being even more pronounced for Blastocystis spp.
ity (same percentage of positive results) of the results between the (23.8% to 39.2%). Using the combined result of the two tests
two tests. The kappa index (here named “kappa index between tests together, we calculated the total of the positive cases observed for
(Kappa IBT)”) was calculated to assess the diagnostic concordance each intestinal parasite. On considering this combined result, an
between the SST and the Paratest (between each pair of replicates increase was observed in the percentage of positive cases of Blas-
compared and between the combined result of each test). Kappa tocystis spp. (43.4%) and hookworm (25.9%), and, to a lesser extent,
measures were interpreted as follows: <0, poor agreement; 0–0.20, of T. trichiura (9.8%).
slight agreement; 0.21–0.40, fair agreement; 0.41–0.60, moderate
agreement; 0.61–0.80, substantial agreement; and 0.81–1.0, almost 3.2. Proportion of positives and test sensitivity and specificity
perfect agreement (Landis and Koch, 1977). The concordance cor-
relation coefficient (CCC) was used to obtain an overall measure of The highest estimated proportions of positive cases were
agreement between the SST and the Paratest taking the results of encountered for Blastocystis spp. (63.8%; CrI 46.1–83.5), hookworm
the three replicates into account. (31.8%; CrI 15.0–50.9) and A. lumbricoides (23.2%; CrI 9.7–40.1). In
A Bayesian latent class approach (Joseph et al., 1995) was used to general the sensitivity was similar in both tests and for almost all
obtain estimates for the sensitivity and specificity of the two tech- the intestinal parasites evaluated, except for Blastocystis spp. and
niques and the proportion of positives for each intestinal parasite. hookworm. In these cases, the sensitivity varied from 42.8% to 46.4%
The conditional dependence between the two tests was estimated in the SST and from 43.4% to 45.6% in the Paratest. The Paratest
using a fixed parameter (Dendukuri and Joseph, 2001). presented a slightly higher sensitivity for the diagnosis of Blasto-
The proportion of positives was assumed to follow a Beta prior cystis spp. (53.8%; CrI 43.2–65.3) in comparison to the SST (39.4%;
distribution with alpha and beta parameters equal to 1 (non- CrI 30.3–49.9) and the SST presented a slightly higher sensitivity
informative distribution). Informative prior distributions were for hookworm (49.2%; CrI 36.0–62.4) in comparison to the Paratest
used for sensitivity and specificity. Taking into account the results (35.8%; CrI 25.0–51.4). The specificity was high (between 93.2% and
obtained by Gomes et al. (2004) and Brandelli et al. (2011), we 97.2%) and similar in both tests and for all the intestinal parasites
assumed that the sensitivity of the SST and the Paratest should be evaluated (Table 1).
between 30% and 60% and the specificity, between 90% and 100%.
Thus, the parameters for the Beta prior distribution for sensitivity 3.3. Repeatability
were alpha equal to 19.35 and beta 23.65, and for specificity 71.25
and 3.75, respectively. In both tests, the concordance between replicates was almost
Three different chains were run from different starting points perfect (Kappa IRM value superior to 0.93) with homogeneity of
to assess convergence to ensure robust estimation. Model con- the results of replicates for G. lamblia, E. histolytica complex, T.
vergence was assessed using Gelman and Rubin convergence trichiura and C. hepaticum. However, one exception was A. lum-
statistics. The first 5000 iterations were discarded as burn-in bricoides, in which almost perfect concordance was observed in
and the next 1500 iterations by chain were used to obtain both tests but with heterogeneity in the SST (Cochran Q, P = 0.011).
a sample of the marginal posterior density for each param- For the detection of Blastocystis spp., the SST and the Paratest pre-
eter (proportion of positive cases, sensitivity and specificity). sented substantial intra-test concordance (Kappa IRM = 0.79 and
66 A.Q. Gonçalves et al. / Acta Tropica 131 (2014) 63–70

Fig. 1. Cumulative percentages of positive cases by test and total percentages of positives, by intestinal parasite.

Table 1
Median value and 95% credible intervals of proportion of positives, sensitivity and specificity of SST and Paratest by intestinal parasites, as estimated by Bayesian analysis.

Protozoans/ Tests Proportion of positives Sensitivity Specificity

Helminths

Median CrI 95% Median CrI 95% Median CrI 95%

SST 4.4–29.1 44.5 32.5–57.7 96.4 92.6–98.8

G. lamblia 14.7
Paratest 45.6 33.7–59.0 96.1 92.2–98.6
SST 0.9–13.0 44.4 30.9–59.0 96.9 93.9–98.9
E. histolytica complex 5.0
Paratest 43.4 29.7–57.9 97.2 94.5–99.0
SST 46.1–83.5 39.4 30.3–49.9 95.7 90.2–98.7
Blastocystis spp. 63.8
Paratest 53.8 43.2–65.3 94.7 87.6–98.4
SST 9.7–40.1 46.4 35.0–58.5 96.1 91.9–98.6
A. lumbricoides 23.2
Paratest 44.6 33.4–56.5 96.6 92.5–98.9
SST 15.0–50.9 49.2 36.0–62.4 93.2 85.7–98.1
Hookworm 31.8
Paratest 35.8 25.0–51.4 96.3 91.9–98.9
SST 2.3–21.2 42.8 28.8–57.6 96.5 92.9–98.8
T. trichiura 9.5
Paratest 44.2 30.1–59.2 95.6 91.7–98.5
C. hepaticum SST 1.9–19.1 44.2 31.4–58.3 96.6 93.3–98.8
8.3
Paratest 44.2 30.9–58.3 96.5 93.2–98.8

SST = spontaneous sedimentation technique; CrI = credible interval.

 A.Q. Gonçalves et al. / Acta Tropica 131 (2014) 63–70 67

Table 2 test results. However, the interpretation of these results together

Concordance (repeatability) and homogeneity between replicates of SST and of
indicates that the tests were completely discordant in the detection
Paratest by intestinal parasite.
of positive cases, that is, when a test was positive for T. trichiura eggs
Protozoans/ Tests Kappa IRM Cochran Q the other was negative, and vice versa.
Helminths
Parasitic forms present a heterogeneous distribution in stool
Median (CI 95%) P value samples (De Carli, 2007b). Homogenization of a stool prior to sam-
SST 1.00 0.97–1.00 0.368 ple processing has been suggested as a procedure to overcome
G. lamblia this problem and improve the diagnostic accuracy (Krauth et al.,
Paratest 0.98 0.95–1.00 0.717

E. histolytica complex
SST 0.98 0.95–1.00 0.050 2012). Up to now, the usefulness of homogenization by stirring has
Paratest 0.99 0.96–1.00 0.607 only been demonstrated for Schistosoma japonicum and Schistosoma
SST 0.79 0.70–0.86 0.355
Blastocystis spp.
Paratest 0.76 0.68–0.84 0.017
mansoni in reducing intra-sample variation of egg location and of
SST 0.96 0.92–0.98 0.011 fecal egg counts, respectively (Krauth et al., 2012; Ye et al., 1998).
A. lumbricoides
Paratest 0.94 0.89–0.97 0.459 For A. lumbricoides, T. trichiura and hookworm this procedure was
SST 0.76 0.66–0.84 0.003 not helpful (Krauth et al., 2012; Ye et al., 1997). In our study, the
Hookworm
Paratest 0.88 0.81–0.93 0.305
homogenization was performed for each stool sample before sep-
SST 0.95 0.96–0.98 0.368
T. trichiura
Paratest 0.93 0.87–0.97 0.273 aration into replicates. The result of high repeatability of both tests
SST 0.99 0.96–1.00 0.223 for the diagnosis of almost all the intestinal parasites studied means
C. hepaticum
Paratest 0.97 0.94–0.99 0.472 that the homogenization contributed to obtain replicates with sim-
SST = spontaneous sedimentation technique; Kappa IRM = kappa index for repeat- ilar qualitative content of parasitic forms. On the other hand, the
ability measure; CI = confidence interval. low repeatability of both tests for Blastocystis spp. and hookworm
detection suggests that this result could have been influenced by
the heterogeneous distribution of their forms in the stools, which,
0.76, respectively). Moreover, the results of replicates were het-
in these cases, could persist even after the homogenization of stools.
erogeneous only in the Paratest (Cochran Q, P = 0.017). Regarding
We believe that technical characteristics of the two techniques
hookworm, the two tests did not present the same results. The
may also explain the low repeatability of the tests. During the pro-
Paratest presented almost perfect concordance (Kappa IRM = 0.88)
cess of the two tests, sediment is generated by the action of gravity,
and the SST lower concordance (Kappa IRM = 0.76) since the results
being characterized by a heterogeneous distribution of parasitic
of the replicates were heterogeneous in the SST (Cochran Q,
forms, which is dependent on the settling velocity of these forms
P = 0.003) (Table 2).
in the liquid media (Sengupta et al., 2011). The homogenization of
the entire sediment before the collection of aliquots for microscopic
3.4. Concordance between tests evaluation is not carried out in either of these tests. However, in the
SST, aliquot collection is influenced by the handling of the tech-
For each intestinal parasite, the results of the kappa IBT were nician since the aliquots may be removed with the pipette from
concordant with the general measure given by the CCC. The con- any part of the sediment. Since handling is a known external factor
cordance between tests for the diagnosis of G. lamblia and C. of variation in diagnostic tests (Sanchez et al., 2002), we believe
hepaticum was almost perfect (CCC of 0.88 and 0.81, respectively) that some variability in our results conferred by this factor may
and was substantial for E. histolytica complex and A. lumbricoides be expected. This problem is minimized in the Paratest because
(CCC of 0.72 and 0.78, respectively). For the diagnosis of Blasto- the sediment generated is not directly manipulated by the techni-
cystis spp., hookworm and T. trichiura the concordance was fair or cian. In this test, the two drops analyzed by microscopy are almost
slight (CCC of 0.38, 0.17 and 0.10, respectively) and the results of always the first of the most posterior end of the sediment that are
the tests were heterogeneous for Blastocystis spp. and hookworm eliminated through the exit located in the cap of the container.
(McNemar test, P < 0.001 and P = 0.009, respectively) (Table 3). The intestinal parasites that did not have reproducible results
in the study of repeatability are characterized by presenting par-
4. Discussion asitic forms of small size (Blastocystis spp., the diameter of which
may vary between 2 and 200 ␮m) (Tan, 2008) and low specific den-
In the present study we analyzed the repeatability and esti- sity (hookworm, with a specific density of approximately 1.055
mated the accuracy of two spontaneous sedimentation techniques in zinc sulphate solution) (Sawitz et al., 1939). Size and den-
applied in the diagnosis of intestinal parasite infections using a sity are important characteristics for the determination of the
Bayesian approach for the first time. dynamics of sedimentation of parasitic forms in liquid media.
The study of repeatability showed that the SST and the Parat- According to some authors, the sedimentation of particles in water
est presented high repeatability for the detection of nearly all the is expected to follow Stokes’ law which implies that the settling
intestinal parasites evaluated. Low repeatability was observed in velocity depends on particle size, difference in density between
the two tests only for the diagnosis of Blastocystis spp. and hook- particles and water, and water viscosity (Medema et al., 1998;
worm, and this may be explained by the different percentages of Sengupta et al., 2011). Shuval (1978) reported that the settling
positive cases between replicates or because the positive cases did velocity of hookworm (0.39 m h−1 ) in clean water is slower than
not correspond to the same individuals in the replicates. that of Ascaris (0.65 m h−1 ) and Trichuris (1.53 m h−1 ) and as such
The concordance between tests was high (substantial or almost the eggs of hookworm are one of the last in settle, suggesting
perfect concordance) except for the diagnosis of Blastocystis spp., that their eggs probably occupy the most anterior part of the
hookworm and T. trichiura. The low concordance encountered for sediment. Information about the settling velocity of Blastocys-
Blastocystis spp. is explained by the fact that the Paratest detected tis spp. is not available in the literature. However, we believe
more positive cases than SST, thereby making the Paratest the best that its forms are also little dense (and probable settle slowly)
test for the detection of this protozoan. On the other hand, the lack since microscopic examination shows that any capillary action
of concordance obtained in the diagnosis of hookworm was due to produced in the slide easily produces the displacement or flota-
the detection of less positive cases by the Paratest, thereby making tion of these forms (Gonçalves, personal communication). The
the SST better in this case. Regarding T. trichiura, the result of low low repeatability of the SST for the diagnosis of hookworm and
concordance was in contrast with the finding of homogeneity of the Blastocystis spp. could be explained by the previously mentioned
68 A.Q. Gonçalves et al. / Acta Tropica 131 (2014) 63–70

Table 3
Concordance and homogeneity between SST and Paratest by pair of replicates and by combined result of the tests for each intestinal parasite.

Protozoans/ Tests SST+ P+ SST+ P− SST− P+ Kappa McNemar CCC

Helminths

N % N % N % N % IBT P value (CI 95%)

G. lamblia SST1/P1 13 9.1 13 9.1 2 1.4 2 1.4 0.83 1.000 0.88

SST2/P2 12 8.4 12 8.4 1 0.7 1 0.7 0.91 1.000 (0.85,0.91)
SST3/P3 12 8.4 12 8.4 1 0.7 1 0.7 0.91 1.000
SSTc/Pc 13 9.1 14 9.8 1 0.7 2 1.4 0.88 1.000 –
E. histolytica complex SST1/P1 5 3.5 3 2.1 3 2.1 1 0.7 0.49 0.625 0.72
SST2/P2 2 1.4 2 1.4 0 0.0 0 0.0 1.00 1.000 (0.67,0.77)
SST3/P3 2 1.4 3 2.1 0 0.0 1 0.7 0.80 1.000
SSTc/Pc 5 3.5 4 2.8 2 1.4 1 0.7 0.66 1.000 –
Blastocystis spp. SST1/P1 27 18.9 34 23.8 11 7.7 18 12.6 0.40 0.265 0.38
SST2/P2 21 14.7 40 28.0 4 2.8 23 16.1 0.45 <0.001 (0.28,0.47)
SST3/P3 23 16.1 47 32.9 8 5.6 32 22.4 0.27 <0.001
SSTc/Pc 38 26.6 56 39.2 6 4.2 24 16.8 0.53 <0.001 –
A. lumbricoides SST1/P1 20 14.0 12 8.4 8 5.6 0 0.0 0.72 0.008 0.78
SST2/P2 17 11.9 15 10.5 3 2.1 1 0.7 0.86 0.625 (0.73,0.83)
SST3/P3 14 9.8 13 9.1 3 2.1 2 1.4 0.80 1.000
SSTc/Pc 20 14.0 18 12.6 2 1.4 0 0.0 0.94 0.500 –
Hookworm SST1/P1 21 14.7 9 6.3 17 11.9 5 3.5 0.20 0.017 0.17
SST2/P2 16 11.2 7 4.9 13 9.1 4 2.8 0.21 0.049 (0.09,0.24)
SST3/P3 6 4.2 4 2.8 6 4.2 4 2.8 -0.03 0.754
SSTc/Pc 31 21.7 17 11.9 20 14.0 6 4.2 0.36 0.009 –
T. trichiura SST1/P1 5 3.5 6 4.2 5 3.5 6 4.2 -0.04 1.000 0.10
SST2/P2 3 2.1 3 2.1 2 1.4 2 1.4 0.32 1.000 (0.03,0.18)
SST3/P3 2 1.4 2 1.4 2 1.4 2 1.4 -0.01 1.000
SSTc/Pc 8 5.6 10 7.0 4 2.8 6 4.2 0.41 0.754 –
C. hepaticum SST1/P1 6 4.2 7 4.9 0 0.0 1 0.7 0.92 1.000 0.81
SST2/P2 8 5.6 6 4.2 3 2.1 1 0.7 0.70 0.625 (0.77,0.85)
SST3/P3 7 4.9 5 3.5 2 1.4 0 0.0 0.83 0.500
SSTc/Pc 8 5.6 8 5.6 2 1.4 2 1.4 0.74 1.000 –

SST = spontaneous sedimentation technique; SST+ = positives for SST; P+ = positives for Paratest; SST+ P− = positives for SST and negatives for Paratest; SST− P+ = negatives for
SST and positives for Paratest; N = number of cases; % = percentage of cases; Kappa IBT = kappa index between tests; CCC = concordance correlation coefficient; CI = confidence
interval; SST1, SST2 and SST3 = results of SST for replicates 1, 2 and 3, respectively; P1, P2 and P3 = results of Paratest for replicates 1, 2 and 3, respectively; SSTc = combined
result of SST; Pc = combined result of Paratest.

handling problem. The pipette introduced into the sediment to forms of Blastocystis spp. during processing of the SST. Some authors
collect an aliquot is usually introduced into the most posterior have previously reported the detection of Blastocystis spp. in feces
part of the sediment and almost never at the more superficial part, conserved in formaldehyde and subsequently submitted to sponta-
where the majority of little dense forms are probably located. This neous sedimentation in water (Eymael et al., 2010; Velásquez et al.,
means that during the microscopic evaluation there is probably a 2005), sometimes with a lower proportion of detection in com-
lack of representativeness of the most anterior part of the sedi- parison with other tests (Eymael et al., 2010). On the other hand,
ment. In the Paratest, the low repeatability also encountered for although both tests also presented low repeatability in the diagno-
Blastocystis spp. and hookworm detection may be explained in the sis of hookworm, the SST detected more cases thereby suggesting
same way. Since the most posterior part of the sediment is always the possible presence of some technical problems in the Paratest in
the first accessed, the most anterior portion is probably underrep- relation to hookworm detection. Beyond the previously discussed
resented in the diagnosis. Furthermore, the time of sedimentation problem about the settling of parasites with a low specific density,
suggested by the kit (15–30 min) may not be sufficient for satisfac- the capacity of adherence of eggs could be another characteristic
tory sedimentation of the little dense eggs of hookworm and the producing limitations of the test. The ability of eggs to adhere to
possible little dense forms of Blastocystis spp. the plastic that composes the filter of the Paratest and also their
The dynamics of sedimentation of the parasitic forms of pro- ability to attach to flocculated particles from the fecal suspension
tozoans and helminths in water or preservative liquids of stool should be evaluated. These characteristics could contribute to ham-
remains little studied. The settling velocity is also dependent pering the passage of eggs through the filter of the test. Brandelli
on other characteristics of the parasites, such as the capacity of et al. (2011) emphasized that the filtration system of the Paratest
some parasitic forms to adhere to surfaces and suspended matter could be a major factor contributing to the false-negative results
(Gaspard et al., 1994; Sengupta et al., 2011). Some authors have observed for larvae in their study.
reported that high particle concentrations in wastewater typically We report a low sensitivity of the SST (39.4–49.2%) and the
results in flocculation of the particles (Droppo, 2001) and this may Paratest (35.8–53.8%) in the diagnosis of some intestinal para-
lead to attachment and entrapment of eggs of helminths or cysts sites. The Paratest was more sensitive (53.8%) for the diagnosis of
and oocysts of protozoans to these flocs (Medema et al., 1998), Blastocystis spp. and the SST for hookworm detection (49.2%). For
affecting the settling velocity of the parasites. parasites obtaining the same sensitivity in both tests, we also found
We observed that although both tests presented low repeatabil- a good concordance between the tests for their detection, except
ity in the diagnosis of Blastocystis spp., the Paratest detected more for T. trichiura. So far, previously published data about the accu-
positive cases. The better results of the Paratest were expected racy of the SST and the Paratest are unreliable since the authors
because it is known that forms of this protozoan are lysed in con- defined the gold standard as the combination of the results of two
tact with water (Amato Neto et al., 2003; Stenzel and Boreham, or more tests or considered the SST as the gold standard (with
1996). We believe that although the feces were previously con- 100% of accuracy). The widespread use of the SST in Brazil has
served in SAF, it was not determinative to avoid the lyses of some made it frequently and erroneously considered as the gold standard
 A.Q. Gonçalves et al. / Acta Tropica 131 (2014) 63–70 69

test in many studies. These practices are responsible for the high sensitivity and high specificity were encountered by both tests.
variability of the reports of the sensitivity of SST encountered in We conclude that the combined study of repeatability, concord-
the literature, with values ranging from 38.4% to 100% (Brandelli ance and accuracy (in the absence of a gold standard test) is a
et al., 2011; Devera et al., 2008a; Gomes et al., 2004). Regarding the key strategy for better evaluation of the performance of tests and
Paratest, the only data available reported 33% and 55% of sensitiv- is also useful for the identification of technical limitations, pro-
ity for the diagnosis of eggs/larvae and cysts, respectively (Brandelli viding opportunities for the generation of proposals for technical
et al., 2011). In our study, the specificity of the two tests was high, improvements. Additional studies on the dynamics of sedimenta-
as previously reported, although the 100% specificity reported by tion of diverse parasitic forms in liquid media are needed in order
some authors is unreal if considering the possibility of occurrence to improve the traditional tests currently used for the diagnosis
of false positives (Tarafder et al., 2010). of intestinal parasites. Moreover, for tests based on sedimenta-
To date, only one article has estimated the sensitivity and speci- tion, the standardization of homogenization of the total sediment
ficity of a centrifuge-sedimentation technique (Danish Bilharziasis before the collection of aliquots for microscopic evaluation should
Laboratory (DBL) technique) for the diagnosis of a human intestinal be implemented in all situations.
parasite using a Bayesian method. The DBL-technique presented a
sensitivity ranging from 65% to 78% for the diagnosis of S. japon- Conflicts of interest statement
icum from some non-human mammals, and a high specificity
(92.6–99.1%), when only one stool sample was considered (Carabin The authors declare that they have no conflicts of interest.
et al., 2005). Since our study was based on the analysis of a single
stool sample, the day-to-day variation in the output of parasitic
Acknowledgments
forms was not assessed, thus, the estimations of sensitivities could
have been higher if more samples had been analyzed.
The authors would like to thank the Secretary of Health Care
Low sensitivity of diagnostic tests for the detection of hookworm
of the Municipality of Presidente Figueredo for the logistic sup-
infection may be related to rapid degeneration of hookworm eggs
port. The scientific unit of the Fundação Oswaldo Cruz in Manaus,
over time. The sensitivity is influenced by delays in time between
Amazonas (Fiocruz Amazônia) for overall support. Dr. Gonçalo
stool production and the processing of samples in the laboratories
Ferraz for the help with study design. This research was funded
(Dacombe et al., 2007; Knopp et al., 2008; Krauth et al., 2012) and,
by the Fundação de Amparo à Pesquisa do Estado do Amazonas
between the processing of samples and microscopic reading when
(Fapeam) (grant number 265/08). Additional support was pro-
using the Kato-Katz technique (Knopp et al., 2008; Tarafder et al.,
vided by the Conselho Nacional de Desenvolvimento Científico e
2010; WHO, 1994). A decrease in sensitivity of almost 50% for the
Tecnológico (CNPq), the Coordenação de Aperfeiçoamento de Pes-
detection of hookworm was reported with the formol-ether con-
soal de Nível Superior (CAPES) and the Fundação Oswaldo Cruz
centration method when preservation with formalin was delayed
(Fiocruz)-Fapeam agreement. The funders had no role in study
by more than 3 h (Dacombe et al., 2007). In our study the time from
design, data collection and analysis, decision to publish, or prepa-
stool collection in households until their processing (preservation)
ration of the manuscript. This paper is contribution number 16 of
was of up to 3 h. However, the delay in the processing was proba-
the Research Program on Infectious Disease Ecology in the Amazon
bly higher if considering that was not possible for us ascertain the
(RP-IDEA) of the Instituto Leônidas e Maria Deane.
time of stool production by the participants. Because of the difficul-
ties of the participants to deliver the stools to the field laboratory,
References
samples were collected in the households by two daily visits of the
staff of the project. The delay in time between stool production and Alum, A., Rubino, J.R., Ijaz, M.K., 2010. The global war against intestinal parasites –
processing in the field laboratory probably contributed, to some should we use a holistic approach? Int. J. Infect. Dis. 14, e732–e738.
extent, to the low sensitivity encountered by both tests in hook- Amato Neto, V., Rodríguez Alarcón, R.S., Gakiya, E., Bezerra, R.C., Ferreira, C.S., Braz,
L.M.A., 2003. Blastocystosis: controversy and indefinedness. Rev. Soc. Bras. Med.
worm detection. This aspect could also be true for Blastocystis spp.,
Trop. 36, 515–517.
since the vacuolar form of this protozoan is fragile (Stenzel and Basso, W., Hartnack, S., Pardini, L., Maksimov, P., Koudela, B., Venturini, M.C., Schares,
Boreham, 1996). G., Sidler, X., Lewis, F.I., Deplazes, P., 2013. Assessment of diagnostic accuracy
of a commercial ELISA for the detection of Toxoplasma gondii infection in pigs
When considering the analysis of three replicates for each test,
compared with IFAT, TgSAG1-ELISA and Western blot, using a Bayesian latent
the increment obtained in the percentage of positives of Blastocystis class approach. Int. J. Parasitol. 43, 565–570.
spp. and hookworm indicates that the processing of three replicates Bethony, J., Brooker, S., Albonico, M., Geiger, S.M., Loukas, A., Diemert, D., Hotez, P.J.,
contributed to better diagnosis. This last result coincides with that 2006. Soil-transmitted helminth infections: ascariasis, trichuriasis, and hook-
worm. Lancet 367, 1521–1532.
of low repeatability of both tests for Blastocystis spp. and hook- Booth, M., Vounatsou, P., N’Goran, E.K., Tanner, M., Utzinger, J., 2003. The influence
worm detection. Thus, the great variability between the results of of sampling effort and the performance of the Kato-Katz technique in diag-
the replicates led to a great increase in the percentage of positives nosing Schistosoma mansoni and hookworm co-infections in rural Côte d’Ivoire.
Parasitology 127, 525–531.
observed when considering the cumulative results of the repli- Borges, J.D., Alarcón, R.S.R., Neto, V.A., Gakiya, E., 2009. Intestinal parasitosis in
cates. Regarding T. trichiura, on the detection of different positive Indians of the Mapuera community (Oriximiná, State of Pará, Brazil): high preva-
cases by the two tests, the combined use of the two tests provided lence of Blastocystis hominis and finding of Cryptosporidium sp and Cyclospora
cayetanensis. Rev. Soc. Bras. Med. Trop. 42, 348–350.
a better diagnosis. However, despite the general improvement of Brandelli, C.L.C., Cargnin, S.T., Willers, D.M.C., Oliveira, K.R.P., Tasca, T., 2011. Com-
the diagnosis in these cases, the low sensitivity reported for both parison between spontaneous sedimentation method and Paratest® for the
tests indicates an important underestimation of the total number diagnosis of intestinal parasitic infections. Trans. R. Soc. Trop. Med. Hyg. 105,
604–606.
of positive cases.
Braun-Munzinger, R.A., Southgate, B.A., 1992. Repeatability and reproducibility
of egg counts of Schistosoma haematobium in urine. Trop. Med. Parasitol. 43,
149–154.
Broemeling, L., 2009. Bayesian Methods for Measures of Agreement. Chapman &
5. Conclusions
Hall/CRC/Taylor and Francis, Boca Raton, FL, USA.
Camacho, M., Pessanha, T., Leles, D., Dutra, J.M.F., Silva, R., de Souza, S.M., Araujo, A.,
In this study we report an overall high repeatability for the SST 2013. Lutz’s spontaneous sedimentation technique and the paleoparasitological
and the Paratest (except for Blastocystis spp. and hookworm by analysis of sambaqui (shell mound) sediments. Mem. Inst. Oswaldo Cruz 108,
155–159.
both tests) and high concordance between the two tests (except for Carabin, H., Balolong, E., Joseph, L., McGarvey, S.T., Johansen, M.V., Fernandez, T.,
the diagnosis of Blastocystis spp., hookworm and T. trichiura). Low Willingham, A.L., Olveda, R., 2005. Estimating sensitivity and specificity of a
70 A.Q. Gonçalves et al. / Acta Tropica 131 (2014) 63–70

faecal examination method for Schistosoma japonicum infection in cats, dogs, 2011. Diagnostic accuracy of Kato-Katz and FLOTAC for assessing anthelmintic
water buffaloes, pigs, and rats in Western Samar and Sorsogon Provinces, The drug efficacy. PLoS Negl. Trop. Dis. 5, e1036.
Philippines. Int. J. Parasitol. 35, 1517–1524. Krauth, S.J., Coulibaly, J.T., Knopp, S., Traoré, M., N’Goran, E.K., Utzinger, J., 2012.
Carvalho, G.L.X.de, Moreira, L.E., Pena, J.L., Marinho, C.C., Bahia, M.T., Machado- An in-depth analysis of a piece of shit: distribution of Schistosoma mansoni and
Coelho, G.L.L., 2012. A comparative study of the TF-Test® , Kato-Katz, hookworm eggs in human stool. PLoS Negl. Trop. Dis. 6, e1969.
Hoffman-Pons-Janer, Willis and Baermann-Moraes coprologic methods for the Landis, J.R., Koch, G.G., 1977. The measurement of observer agreement for categorical
detection of human parasitosis. Mem. Inst. Oswaldo Cruz 107, 80–84. data. Biometrics 33, 159–174.
Charlier, J., Duchateau, L., Claerebout, E., Vercruysse, J., 2005. Assessment of the Levecke, B., Behnke, J.M., Ajjampur, S.S.R., Albonico, M., Ame, S.M., Charlier, J., Geiger,
repeatability of a milk Ostertagia ostertagi ELISA and effects of sample prepara- S.M., Hoa, N.T.V., Kamwa Ngassam, R.I., Kotze, A.C., McCarthy, J.S., Montresor,
tion. Prev. Vet. Med. 68, 277–288. A., Periago, M.V., Roy, S., Tchuem Tchuenté, L.-A., Thach, D.T.C., Vercruysse, J.,
Dacombe, R.J., Crampin, A.C., Floyd, S., Randall, A., Ndhlovu, R., Bickle, Q., Fine, P.E.M., 2011. A comparison of the sensitivity and fecal egg counts of the McMaster egg
2007. Time delays between patient and laboratory selectively affect accuracy of counting and Kato-Katz thick smear methods for soil-transmitted helminths.
helminth diagnosis. Trans. R. Soc. Trop. Med. Hyg. 101, 140–145. PLoS Negl. Trop. Dis. 5, e1201.
De Carli, G.A., 2007a. Exames macroscópico e microscópico da amostra fecal fresca Lutz, A., 1919. Schistosomum mansoni and Schistosomatosis observed in Brazil. Mem.
e preservada. In: De Carli, G.A. (Ed.), Parasitologia Clínica: Seleção de Métodos e Inst. Oswaldo Cruz 11, 121–155.
Técnicas de Laboratório Para o Diagnóstico Das Parasitoses Humanas. Atheneu, Medema, G.J., Schets, F.M., Teunis, P.F., Havelaar, A.H., 1998. Sedimentation of free
São Paulo, pp. 29–82. and attached Cryptosporidium oocysts and Giardia cysts in water. Appl. Environ.
De Carli, G.A., 2007b. Colheita e preservação da amostra fecal. In: De Carli, G.A. Microbiol. 64, 4460–4466.
(Ed.), Parasitologia Clínica: Seleção de Métodos e Técnicas de Laboratório Para Pinheiro, I. de O., de Castro, M.F., Mitterofhe, A., Pires, F.A.C., Abramo, C., Ribeiro, L.C.,
o Diagnóstico Das Parasitoses Humanas. Atheneu, São Paulo, pp. 3–27. Tibiriçá, S.H.C., Coimbra, E.S., 2011. Prevalence and risk factors for giardiasis and
de Souza, E.A., da Silva-Nunes, M., Malafronte, R.D.S., Muniz, P.T., Cardoso, M.A., soil-transmitted helminthiasis in three municipalities of Southeastern Minas
Ferreira, M.U., 2007. Prevalence and spatial distribution of intestinal parasitic Gerais State, Brazil: risk factors for giardiasis and soil-transmitted helminthiasis.
infections in a rural Amazonian settlement, Acre State, Brazil. Cad. Saúde Pública Parasitol. Res. 108, 1123–1130.
23, 427–434. Ribeiro, S.R., Furst, C., 2012. Parasitological stool sample exam by spontaneous sed-
Dendukuri, N., Joseph, L., 2001. Bayesian approaches to modeling the conditional imentation method using conical tubes: effectiveness, practice, and biosafety.
dependence between multiple diagnostic tests. Biometrics 57, 158–167. Rev. Soc. Bras. Med. Trop. 45, 399–401.
Devera, R., Aponte, M., Belandria, M., Blanco, Y., Requena, I., 2008a. Uso del método Sanchez, J., Dohoo, I.R., Markham, F., Leslie, K., Conboy, G., 2002. Evaluation of the
de sedimentación espontanea en el diagnóstico de parásitos intestinales. Saber repeatability of a crude adult indirect Ostertagia ostertagi ELISA and methods of
20, 163–171. expressing test results. Vet. Parasitol. 109, 75–90.
Devera, R., Blanco, Y., Hernández, H., Simoes, D., 2008b. Toxocara spp. and other Santos, R.V.D., Nunes, J., da, S., Camargo, J.A., de, S.A., Rocha, E.M.M.da, Fontes, G.,
helminths in squares and parks of Ciudad Bolívar, Bolivar State (Venezuela). Camargo, L.M.A., 2013. High occurrence of Entamoeba histolytica in the munici-
Enferm. Infecc. Microbiol. Clín. 26, 23–26. palities of Ariquemes and Monte Negro, State of Rondônia, Western Amazonia,
Dogruman-Al, F., Simsek, Z., Boorom, K., Ekici, E., Sahin, M., Tuncer, C., Kustimur, S., Brazil. Rev. Inst. Med. Trop. São Paulo 55, 193–196.
Altinbas, A., 2010. Comparison of methods for detection of Blastocystis infection Sawitz, W., Tobie, J.E., Katz, G., 1939. The specific gravity of hookworm eggs. Am. J.
in routinely submitted stool samples, and also in IBS/IBD patients in Ankara, Trop. Med. Hyg. 19, 171–179.
Turkey. PLoS ONE 5, e15484. Sengupta, M.E., Thamsborg, S.M., Andersen, T.J., Olsen, A., Dalsgaard, A., 2011. Sedi-
Droppo, I.G., 2001. Rethinking what constitutes suspended sediment. Hydrol. Pro- mentation of helminth eggs in water. Water Res. 45, 4651–4660.
cess. 15, 1551–1564. Shuval, H.I., 1978. Parasitic disease and waste-water irrigation. In: Pacey, A. (Ed.),
Eymael, D., Schuh, G.M., Tavares, R.G., 2010. Standardization of Blastocystis homi- Sanitation in Developing Countries. John Wiley & Sons, Chichester, pp. 210–223.
nis diagnosis using different staining techniques. Rev. Soc. Bras. Med. Trop. 43, Steinmann, P., Cringoli, G., Bruschi, F., Matthys, B., Lohourignon, L.K., Castagna, B.,
309–312. Maurelli, M.P., Morgoglione, M.E., Utzinger, J., Rinaldi, L., 2012. FLOTAC for the
Fenwick, A., 2012. The global burden of neglected tropical diseases. Public Health diagnosis of Hymenolepis spp. infection: proof-of-concept and comparing diag-
126, 233–236. nostic accuracy with other methods. Parasitol. Res. 111, 749–754.
Fuehrer, H.-P., Igel, P., Auer, H., 2011. Capillaria hepatica in man – an overview of Stenzel, D.J., Boreham, P.F., 1996. Blastocystis hominis revisited. Clin. Microbiol. Rev.
hepatic capillariosis and spurious infections. Parasitol. Res. 109, 969–979. 9, 563–584.
Gaspard, P.G., Wiart, J., Schwartzbrod, J., 1994. Experimental study of the helminth Tan, K.S.W., 2008. New insights on classification, identification, and clinical rele-
eggs adhesion (Ascaris suum): analysis of the environmental implications. Rev. vance of Blastocystis spp. Clin. Microbiol. Rev. 21, 639–665.
Sci. Eau. 7, 367–376. Tarafder, M.R., Carabin, H., Joseph, L., Balolong Jr., E., Olveda, R., McGarvey, S.T., 2010.
Glinz, D., Silué, K.D., Knopp, S., Lohourignon, L.K., Yao, K.P., Steinmann, P., Rinaldi, Estimating the sensitivity and specificity of Kato-Katz stool examination tech-
L., Cringoli, G., N’Goran, E.K., Utzinger, J., 2010. Comparing diagnostic accuracy nique for detection of hookworms, Ascaris lumbricoides and Trichuris trichiura
of Kato-Katz, Koga agar plate, ether-concentration, and FLOTAC for Schistosoma infections in humans in the absence of a gold standard. Int. J. Parasitol. 40,
mansoni and soil-transmitted helminths. PLoS Negl. Trop. Dis. 4, e754. 399–404.
Gomes, J.F., Hoshino-Shimizu, S., Dias, L.C.S., Araujo, A.J.S.A., Castilho, V.L.P., Neves, Tello, R., Terashima, A., Marcos, L.A., Machicado, J., Canales, M., Gotuzzo, E., 2012.
F.A.M.A., 2004. Evaluation of a novel kit (TF-test) for the diagnosis of intestinal Highly effective and inexpensive parasitological technique for diagnosis of
parasitic infections. J. Clin. Lab. Anal. 18, 132–138. intestinal parasites in developing countries: spontaneous sedimentation tech-
Gonçalves, A.Q., Ascaso, C., Santos, I., Serra, P.T., Julião, G.R., Orlandi, P.P., 2012. nique in tube. Int. J. Infect. Dis. 16, e414–e416.
Calodium hepaticum: household clustering transmission and the finding of a Thomas, V., Sinniah, B., Leng, Y.P., 1981. Assessment of the sensitivity, specificity, and
source of human spurious infection in a community of the Amazon region. PLoS reproducibility of the indirect immunofluorescent technique for the diagnosis
Negl. Trop. Dis. 6, e1943. of amebiasis. Am. J. Trop. Med. Hyg. 30, 57–62.
Harhay, M.O., Horton, J., Olliaro, P.L., 2010. Epidemiology and control of human Traub, R.J., Inpankaew, T., Reid, S.A., Sutthikornchai, C., Sukthana, Y., Robertson, I.D.,
gastrointestinal parasites in children. Expert Rev. Anti-Infect. Ther. 8, 219–234. Thompson, R.C.A., 2009. Transmission cycles of Giardia duodenalis in dogs and
Harhay, M.O., Horton, J., Olliaro, P.L., Utzinger, J., 2011. Diagnostics are central for a humans in Temple communities in Bangkok – a critical evaluation of its preva-
truly holistic approach against intestinal parasitic diseases. Int. J. Infect. Dis. 15, lence using three diagnostic tests in the field in the absence of a gold standard.
e76–e77. Acta Trop. 111, 125–132.
Hoffman, W., Pons, J.A., Janer, J.L., 1934. The sedimentation–concentration method Velásquez, V., Caldera, R., Wong, W., Cermeño, G., Fuentes, M., Blanco, Y., Aponte,
in schistosomiasis mansoni. P.R. J. Publ. Health Trop. Med. 9, 283–291. M., Devera, R., 2005. Blastocystosis: a high prevalence of cases found in patients
Inês, E.deJ., Souza, J.N., Santos, R.C., Souza, E.S., Santos, F.L., Silva, M.L.S., Silva, M.P., from Health Center of Soledad, Anzoategui State, Venezuela. Rev. Soc. Bras. Med.
Teixeira, M.C.A., Soares, N.M., 2011. Efficacy of parasitological methods for the Trop. 38, 356–357.
diagnosis of Strongyloides stercoralis and hookworm in faecal specimens. Acta White, S.A., van den Broek, N.R., 2004. Methods for assessing reliability and validity
Trop. 120, 206–210. for a measurement tool: a case study and critique using the WHO haemoglobin
Joseph, L., Gyorkos, T.W., Coupal, L., 1995. Bayesian estimation of disease prevalence colour scale. Stat. Med. 23, 1603–1619.
and the parameters of diagnostic tests in the absence of a gold standard. Am. J. WHO, 1994. Bench Aids for the Diagnosis of Intestinal Parasites. World Health Orga-
Epidemiol. 141, 263–272. nization, Geneva.
Knopp, S., Mgeni, A.F., Khamis, I.S., Steinmann, P., Stothard, J.R., Rollinson, D., Marti, Ye, X.P., Donnelly, C.A., Anderson, R.M., Fu, Y.L., Agnew, A., 1998. The distribution of
H., Utzinger, J., 2008. Diagnosis of soil-transmitted helminths in the era of pre- Schistosoma japonicum eggs in faeces and the effect of stirring faecal specimens.
ventive chemotherapy: effect of multiple stool sampling and use of different Ann. Trop. Med. Parasitol. 92, 181–185.
diagnostic techniques. PLoS Negl. Trop. Dis. 2, e331. Ye, X.P., Donnelly, C.A., Fu, Y.L., Wu, Z.X., 1997. The non-randomness of the distribu-
Knopp, S., Speich, B., Hattendorf, J., Rinaldi, L., Mohammed, K.A., Khamis, I.S., tion of Trichuris trichiura and Ascaris lumbricoides eggs in faeces and the effect
Mohammed, A.S., Albonico, M., Rollinson, D., Marti, H., Cringoli, G., Utzinger, J., of stirring faecal specimens. Trop. Med. Int. Health 2, 261–264.