Increased detection of schistosomiasis with Kato-Katz and SWAP-IgG-ELISA in a Northeastern Brazil low-intensity transmission area

Introduction: The laboratory diagnosis of schistosomiasis is based mainly on the detection of parasite eggs in stool samples through the Kato-Katz (KK) technique, reading one slide by test. However, a widely known limitation of parasitological methods is reduced sensitivity, particularly in low endemic areas. Methods: To increase sensitivity, we conducted further slide readings from the same stool sample using the parasitological method associated with a serological test. We used the KK method (three slides) and the IgG anti-Schistosoma mansoni– enzyme-linked immunosorbent assay (ELISA) technique to diagnose schistosomiasis in low endemic areas in the Brazilian State of Ceará. Fecal samples and sera from 250 individuals were analyzed. Results: Sixteen percent and 47.2% of samples were positive in parasitological tests and serological tests, respectively. Parasitological methods showed that 32 (80%) individuals tested positive on the first slide, 6 (15%) on the second slide, and 2 (5%) on the third. The performance of the ELISA test in the diagnosis, using the KK method as diagnostic reference, showed a negative predictive value of 100%, with specificity and positive predictive values of 62.8% and 33.9%, respectively. Conclusions: In this study, the increase from one to three slides analyzed per sample using the KK technique was shown to be a useful procedure for increasing the diagnostic sensitivity of this technique.

Schistosomiasis is still considered one of the most important chronic parasitic diseases in tropical regions, affecting approximately 200 million people worldwide despite the continued implementation of control measures 1 .In Northeast Brazil, some isolated foci of intestinal schistosomiasis remain, with low intensity of transmission and, consequently, low parasite burden.Current data from the Schistosomiasis Control Program, or Programa de controle da esquistossomose (PCE), characterize the State of Ceará, Brazil, as an area of low prevalence, but indicate that this state has the locality with the highest positivity rate for schistosomiasis (Planalto do Cajueiro in the municipality of Maranguape), where the prevalence was 8.5% in 2006 and 13.8% in 2007.The World Health Organization recommends the Kato-Katz (KK) technique 2 as the standard diagnostic method for intestinal schistosomiasis, as it is simple and low cost, and the intensity of infection can be estimated 3 .However, KK smears and other parasitological methods frequently fail to detect low-intensity infections 4 , which leads to an underestimation of the actual prevalence of disease 5,6 .
To overcome this limitation, examination of a larger quantity of fecal material is necessary 7 .By increasing the number of KK smears per sample and the number of stool samples taken, it is possible to increase sensitivity considerably 8 .It is worth noting that the control of schistosomiasis-related morbidity has become feasible due to the development of single-dose oral drugs such as oxamniquine and praziquantel, which are given to heavily infected patients (high worm burden) who are easily detected by field-applicable parasitological methods.With the transition to lower morbidity, more sensitive diagnostic methods are needed 9 .In this context, the enzyme-linked immunosorbent assay (ELISA) immunological technique has been proposed for use in the PCE 10 .It is important to conduct studies about

RESULTS
Carneiro TR et al -Improving diagnosis of schistosomiasis hookworm and Ascaris lumbricoides co-infections, as literature data show the possibility of cross-reaction with Schistosoma mansoni 11 .In this study, we used the KK method (three slides) and the IgG-ELISA technique to detect anti-S.mansoni IgG class antibodies, with the aim of increasing sensitivity in the diagnosis of schistosomiasis in the locality that, according to the PCE, has the highest prevalence of this disease in Ceará, which is an area of low endemicity.

Study area, population, and design
We conducted a cross-sectional survey in the community of Planalto do Cajueiro, in the municipality of Maranguape, Ceará (in Northeastern Brazil).Maranguape (population 98,429) is located 30km from the state capital, Fortaleza, at an altitude of about 736m.Planalto do Cajueiro is a semi-urban locality, with poor sanitation and infrastructure conditions, and has two small rivers that exhibit populations of the intermediate host Biomphalaria straminea.In the study area, the state's PCE has been conducting KK stool examinations and selective treatment of positive individuals with praziquantel for more than 30 years.
At the time of the survey (2009-2010), Planalto do Cajueiro consisted of 903 inhabitants, with sources of income based primarily on subsistence agriculture and textile/footwear industries.The study included all individuals over two years of age who reside in this town and who agreed to participate, resulting in a target population of 250 individuals.Community meetings were held to explain the objectives of the study.Each household in the community was visited.Family members present on this occasion were interviewed, using pre-tested structured questionnaires.

Stool examinations and ELISA
For KK stool examinations, labeled plastic containers were distributed for collection of fecal samples.The following day the containers were collected and sent to the local health care center for parasitological examination.Stool samples were prepared in a field laboratory on the same day.To identify S. mansoni eggs, approximately 50mg of feces were prepared using the KK method (Helm-Test ® Kit).The intensity of infection, expressed as eggs per gram (epg) of feces, was calculated by multiplying each slide count by 24.To identify other intestinal helminth eggs, the Lutz method was used 12 .
For the ELISA testing, venous blood was collected using a commercially available system (Becton Dickinson Vacutainer Systems, Franklin Lakes, USA).To detect S. mansoni IgG antibodies from serum samples, we used adult worm antigen (SWAP) of S. mansoni and performed ELISA testing according to standard procedures 13 .The cut-off was defined based on the mean value of the optical density (OD) ± 2 standard deviations (SD) in a base population (n = 35) from the City of Fortaleza with negative parasitological fecal examination.To reduce disturbance of the study population, we did not repeat fecal examinations in the case of positive KK results.OD was measured on an ELISA automatic reader (BioTeck ® /USA), using a 490-nm filter.

Statistical analysis
Data were entered using an Excel® spreadsheet and checked for entry-related errors, by comparing data entries with original data forms.Then data were transferred to software GraphPad PRISM Version 5.0 ® for analysis.
The sensitivity and specificity as well as the positive predictive value (PPV) and negative predictive value (NPV) of IgG-ELISA were calculated by taking the result of the Kato-Katz examination as the "reference standard." To reduce inter observer variation, all slides were read by a single investigator.To perform quality control, 10% of slides were randomly selected for cross-reading by a reference microscopist who was blinded to the results of the first reading.Divergence was less than 5%.

Ethical considerations
The study was approved by the Ethical Review Board of the Faculty of Medicine of the Federal University of Ceará, Brazil (no.165/09).Informed written consent was obtained from study participants or, in the case of minors, from their legal guardians.In case of S. mansoni infection, a single dose of praziquantel (60mg/kg body weight) was given.Participants with positive fecal examination for intestinal helminths were treated with albendazole (400mg/single dose).

Comparative analysis of the methods
Figure 1 shows the cumulative fraction of subjects that tested positive using the KK method, according to the number of slides examined (sensitivity).With the examination of a single slide, 32 (80%) infected individuals were detected.This increased to 40 individuals when three slides were examined, a major improvement for detection when aiming to control the disease.

Figure 2 details the results of individuals, according to ELISA and KK results
. There were no false-negative ELISA results; ELISA was positive in all 40 cases in which eggs were detected by the KK method.
Of the 78 subjects with positive ELISA and negative KK, 57 (73%) stated in the interview that they never had schistosomiasis and had never received any specific treatment.The performance of the ELISA test in the diagnosis using the KK method as a diagnostic reference showed that the negative predictive value of the ELISA test was 100%; specificity was 62.8%, and positive predictive value was 33.9%.

DISCUSSION
Since 1976, parasitological testing of feces has been used by state schistosomiasis control programs as a criterion for selective treatment with praziquantel 14 .The worldwide methodological strategy currently applied for the diagnosis of mansoni schistosomiasis, using the KK method with one sample and one slide to determine the infection rate, does not have the sensitivity required for such diagnosis and particularly in areas of low endemicity underestimates the prevalence of this parasitosis 15 .Our data confirm that in areas of low endemicity, the stool test has major value for mass diagnosis and monitoring efforts of control, at least when the number of slides is increased.
Increasing the number of KK smears per sample and/or the number of stool samples analyzed increases sensitivity considerably.Enk et al. 16 demonstrated an increase in the estimated prevalence from 13.8% (one slide) to 19% (six slides) using a single stool sample.In this study, seeking to minimize the difficulties that come with repeated collection of stool samples, we worked with one fecal sample as well and analyzed three slides.Using this strategy, we observed an increase in the sensitivity of the KK method, from 32 (12.8%) of 250 subjects when analyzing one slide, to 40 (16%) of 250 when analyzing three slides.
Consecutive stool examinations may present logistical problems and higher costs 17 , as well as the need to increase the number of practitioners in the field.Thus, the strategy of increasing the number of slides, from one to three, from a single stool sample is shown to be easily applied in routine of the PCE, since even with limited infrastructure and human resources, the strategy can be applied with no considerable financial or logistical increase in the public healthcare sector.
The number of eggs per slide of the 40 individuals who tested positive for S. mansoni, i.e., the individual parasite load, varied from 8epg to 560epg, with an average of 58epg (SD = 93epg).Thus, we verified that this region, despite being an area of low endemicity, had individuals with elimination of eggs greater than 100epg.
The introduction of serological methods in epidemiological studies, thus identifying potential infection sites, may contribute to reducing transmission in low endemic areas 18 .In Planalto do Cajueiro, we observed that 118 (47.2%) of the 250 individuals analyzed were positive using the IgG-ELISA method; similar data were obtained by da Frota et al. 8 in the community of Caititú de Cima (also in the State of Ceará), where 96 (33.4%) of 287 subjects were serum reactive.
When we compared the serology results with the parasitology results, we found that the rates of seropositivity were higher (nearly three times greater) than the rates obtained by the KK method.In the 250 individuals evaluated, 118 (47.2%) were seropositive, whereas only 40 (16%) were parasitologically positive.In this seropositivity there is certainly a portion of individuals who have been cured with treatment, since the technique does not distinguish between active and past infections 10 , and another portion that represents the non-specificity of test, which can occur due to frequent similarity between the constituent antigens of the parasite 19 .In this context, to analyze cross-reactivity, we detected 43 of 250 individuals with the presence of other parasites, using the Lutz method.Studies conducted in Brazil and Venezuela show that sera from patients infected with hookworm and A. lumbricoides exhibit significant cross-reactivity with antigens of S. mansoni 11,20 .
Nevertheless, the immunological methods for diagnosing schistosomiasis take on greater importance every day, and their inclusion in control programs in low endemic areas is crucial 21 .In previous studies in low endemic areas, in which the positivity rates of immunological and parasitological methods were compared [22][23][24] , the positivity rates of immunological methods were two to six times higher than those of parasitological methods.
The methods of serological and parasitological diagnosis are complementary 25 .When only the stool parasitology method is used to diagnose schistosomiasis, the true prevalence of the disease is underestimated, because of the low efficiency of this method in detecting cases with a small number of eggs.The possibility of