| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
MATERIALS AND METHODS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). The town is located 15 km from the urban center of Iquitos (population = 500,000; 3.7°S, 73.2°W). During August 2002, a house-to-house census and survey of the community was completed. The survey collected data on basic sociodemographic indicators and risk factors for infection with intestinal parasites, including source and storage methods of water used for consumption, human waste disposal, and the frequency of the use of shoes. All participants were asked to provide one stool sample and allow collection of 3–5 mL of blood. The study protocol was reviewed and approved by the Institutional Review Boards of the Johns Hopkins Bloomberg School of Public Health, Asociacion Benéfica PRISMA, and by the Department of Public Health of Loreto Peru. All participants provided informed consent, and all those found to have helminth infections were treated according to current recommendations by the World Health Organization.
|
Modified Baermann method. Approximately 10 grams of stool were placed in gauze and partially submerged in water (40°C) in a 50-mL centrifuge tube and left for two hours. Five drops of sediment from the conical tip were removed with a transfer pipette and examined microscopically.
Simple sedimentation. Fifteen grams of sample were homogenized in water and gauze filtered into 400-mL plastic cylinders. Water was then added to the volume of the cylinder and the mixture was allowed to settle for 30 minutes or until the supernatant was nearly clear. Two-thirds of the volume of water was decanted and five drops of the sediment from the cylinder based was removed with a transfer pipette and examined microscopically.
Agar plate culture. Agar media consisting of 1.5% agar, 0.5% meat extract, 1.0% peptone, and 0.5% NaCl was prepared and plated on 90-mm petri plates. Approximately three grams of stool was placed at the center of the plate and the plate was covered and left for three days in the laboratory at ambient temperature (30°C). Plates were examined daily for trails and larval canals with a stereoscopic microscope.
Serologic assays. All blood specimens were centrifuged, and the serum was separated and frozen at –20°C pending examination by an enzyme-linked immunosorbent assay (ELISA) for IgG to S. stercoralis. Serologic response to S. stercoralis is known to cross-react with that for Wuchereria bancrofti and Onchocerca volvulus, but neither of these filarial worms has been described in the study region. However, in a malaria survey in a neighboring village, 4 (0.3%) of 1,200 thick smears were positive for Mansonella perstans. Since we assumed that the response to M. perstans might also have cross-reactivity with that to S. stercoralis, a subset of 96 randomly selected sera were evaluated by ELISA using several filarial antigens. An extract of Brugia pahangi adult worms, chosen for the extensive cross-reactivity it displays with other filarial parasites, was used for screening. Sera with positive response to the polyclonal extract were then assessed for their responsiveness to the recombinant Brugia malayi antigen Bm14 to identify true filarial infection, as opposed to cross-reactivity between S. stercoralis and Brugia antigens.7
Possible serologic cross-reactions with hookworm species were evaluated by determining the prevalence of seropositivity by the S. stercoralis ELISA in individuals who were negative for S. stercoralis in stool, but positive for hookworm. This likely substantially overestimates the true cross-reactivity because individuals who previously were infected with S. stercoralis and subsequently eliminated it will also be included in this subgroup.
Enzyme-linked immunosorbent assay for S. stercoralis. Strongyloides stercoralis third-stage (L3) antigen was kindly provided by Dr. Franklin Neva (Laboratory of Parasitic Diseases, National Institute of Allergy and Infectious Diseases, National Institutes of Health, Bethesda, MD) and ELISAs were done with small variations from that previously described.8 Briefly, after performing a checkerboard the initial antigen was diluted to a concentration of 5 µg/mL with 0.06 M carbonate buffer, pH 9, and placed in polyethylene, round-bottom, 96-well plates (Immulon-2HB; Dynex, Chantilly, VA) and incubated for three hours at 37°C. Plates were washed serially with phosphate-buffered saline (PBS) with 0.05% Tween and blocked with 1% skim milk at 37°C for one hour. Sera were diluted 1:32 in PBS-Tween solution and 50 µL of test sera was incubated in wells at 37°C for one hour. Following serial washes (all washes done with PBS-Tween), peroxidase-labeled antibody to human IgG (Kirkegaard and Perry Laboratories, Gaithersburg, MD) was diluted 1:4,000 in PBS-Tween, 50 µL was added to each well, and the plates were incubated at 37°C for one hour. Following serial washes, 100 µL of the substrate solution (0.01% o-phenylenediamine plus 0.03% hydrogen peroxide in phosphate citrate buffer) was added and the plates were left in the dark for eight minutes. The reaction was stopped with 50 µL of 2N sulfuric acid and optical densities were read at 490 nm. To improve the precision of the assay, four control sera with a pre-defined target optical density were included on each ELISA plate and the plate correction-4 (PCF4) method was used. These four pools were from sera derived from children from tropical areas not endemic for nematode infections (low negative), adults from areas not endemic for helminth infections (high negative), HIV-infected adults with stools positive for Strongyloides (low positive), and HIV-negative adults from a disease-endemic area (high positive). These sera were chosen to span the optical density range likely to be encountered on each ELISA plate. The ratio of the measured optical density to the pre-defined optical density was calculated for each of the control sera as a correction factor and the mean correction factor (the PCF4 value) was multiplied by measured optical densities for all test sera as previously described.9
The cutoff value was optimized by constructing receiving operator characteristic (ROC) curves based on ELISA results from 19 patients stool positive for S. stercoralis from the Iq-uitos area (positive controls) and 37 adults from Lima (tropical negative controls), an area with frequent intestinal protozoal infections but no geohelminth infections (Gilman R, unpublished data).
Enzyme-linked immunosorbent assay for filarial species. An ELISA for filarial species using Brugia extract and recombinant Bm14 were done using the methods of Muck and others.7
Data analysis. Data were analyzed using SPSS version 12 (SPSS Inc., Chicago, IL). Pearson’s chi-square test was used for dichotomous values comparing sampled and non-sampled subpopulations, and for univariate analyses of risk factors for S. stercoralis infection. Univariate and multivariable logistic regression analyses were performed with SAS version 8 (SAS Institute, Cary, NC). Models were constructed using generalized estimating equations to account for intra-household correlation.
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Stool examination and ELISA results.
The most prevalent parasitic pathogens were Ascaris lumbricoides (41.9%), Trichuris trichura (16.9%), S. stercoralis (8.7%), Hymenolepsis nana (7.1%), Giardia intestinalis (5.2%), and hookworm (4.5%) (Table 1). A stool was classified as positive for S. stercoralis if larvae were identified by any of the three diagnostic stools tests performed. When each S. stercoralis diagnostic technique alone was compared with the gold standard of all three results taken together, the sensitivity of simple sedimentation was 79.7%, that of the modified Baermann method was 40.5%, and that of direct examination was 37.7%. The agar plate method used in a subset of samples had a sensitivity of only 60.9% because of an exuberant growth of mold on 180 (35%) of 515 specimens that impeded test interpretation.
|
). Of these, 41 (8.3%) were positive both by stool examination and ELISA, and 138 (28%) were negative by both assays. Most (311, 63.1%) had positive serologic results, but negative stool examination results; this group may include both individuals with previous strongyloidiasis and those for whom a single stool examination failed to detect an active infection. Three individuals (0.6%) were stool positive, but serologically negative; all three were positive for S. stercoralis on the agar plate test. If stool examination is taken as a gold standard for current infection, these results correspond to an ELISA sensitivity of 93% (41 of 44) and a negative predictive value of 98% (138 of 141).
|
Sera were available from 18 individuals whose stool specimen was positive for hookworm, but negative for S. stercoralis. Fourteen of these individuals (77.8%) had a positive ELISA result for S. stercoralis. The S. stercoralis seroprevalence did not differ among those with and without hookworm in their stool (P = 0.79). Hookworm infection was not significantly associated with positive S. stercoralis ELISA results in a multivariable logistic regression model adjusted for age (P = 0.38).
Risk factors for S.stercoralis infection by stool examination and ELISA.
In univariate analyses, males were more likely than females to be stool positive for S. stercoralis (relative risk [RR] = 1.64, 95% confidence interval [CI] = 1.00–2.71). The prevalence of S. stercoralis in stool increased rapidly in early childhood from 2.9% positive for children less than two 2 years of age to 8.9% for children 2–5 years of age (Figure 2). In a univariate logistic regression analysis, the odds of being S. stercoralis stool positive increased with age (OR = for each one-year increase, 95% CI = 1.00–1.03). Those who occasionally or never wore shoes were almost twice as likely to have stools positive for S. stercoralis than those who frequently or always wore shoes (RR = 1.97, 95% CI = 1.20–3.25). Occupation was categorized as predominantly outdoor or indoor; participants who worked in predominantly outdoor occupations were at higher risk than those with predominantly indoor occupations (RR = 1.70, 95% CI = 0.90–3.21), but this association failed to reach statistical significance. There was no significant association between risk of S. stercoralis-positive stool and the location of human waste disposal or water source for the household. Participants with stool positive for hookworm were more than four times as likely to have S. stercoralis identified in the same stool sample as those negative for hookworm (RR = 4.54, 95% CI = 2.10–9.87). In the multivariable logistic regression model, increasing age, presence of hookworm in the stool, and occasionally or never wearing shoes were associated with an increased risk of having S. stercoralis in stool (Table 3).
|
|
), and the serologic pattern, like that for S. stercoralis in stool, showed a rapid increase in prevalence during early childhood. Although no children less than two years of age were seropositive, 39% of 2–5-year-old children and 62.5% of 6–14-year-old children were seropositive. The odds of being seropositive increased with age (OR = 1.04 for each one-year increase, 95% CI = 1.02–1.05). In univariate models, there was a trend for wearing shoes occasionally or never to be associated with increased risk (RR = 1.37, 95% CI = 0.92–2.02). Univariate analysis showed that those who worked in predominantly out-door occupations were more likely to be seropositive than those working in indoor occupations (RR = 1.89, 95% CI = 1.09–3.25). Waste disposal location and water source were not associated with altered risk of seropositivity. In the multivariable model, increasing age and occasionally or never wearing shoes remained significantly associated with an increased risk of positive serologic results for S. stercoralis (Table 3). |
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Because of the poor performance of the agar plate method, we used the composite result of three less sensitive tests to define current infection status. We believe that the large amount of stool screened in these serial exams (more than 25 grams per sample) to some extent overcomes the limited sensitivity of each test performed individually. Nevertheless, some participants classified as stool negative might have been found to be positive if multiple stool specimens had been examined. Remarkably, the simple sedimentation procedure was more sensitive than the modified Baermann method or direct examination. This simple technique is appropriate for screening of large numbers of specimens, and we recommend its use for epidemiologic studies, and for the assessment of helminth control programs.
For the S. stercoralis ELISA, we used the PCF4 correction method to improve the precision, and to decrease within-plate and plate-to-plate variability.9 In epidemiologic studies or large-scale screening programs, plate-to-plate variation in particular must be minimized to accurately classify the test results. Precision in the testing of an ELISA is often assumed in assay development and serologic studies, despite the observation that plate-to-plate coefficients of variation are generally at least 10% even under ideal conditions,10 and may reach 
15%.11,12 Although the PCF4 control sera occupy 8 wells of the 96-well plate, the improved precision and simplicity of the method make this adjustment worthwhile, especially in laboratories without automated plate washers.
Although the ELISA used in this study has been available for several years, its use has been limited by significant cross-reactivity with filarial species, which are often co-endemic with S. stercoralis. Western blots have been used to overcome this problem, but are only partially effective in doing so13 and are relatively cumbersome to perform. An ELISA with additional preincubation of the test sera with Onchocerca gutturosa has been reported to reduce cross-reactions in patients with lymphatic filariasis and infection with Necator americanus.14 An ELISA based on a recombinant antigen has recently been reported to have a sensitivity of 87.5% and a specificity of 93.5%, with no cross-reactivity with O. volvulus or L. loa; however, difficulties in the concentration of the expressed protein limit the current availability of this protein.15
In our data, we used an ELISA with good sensitivity and specificity, and our study area was not endemic for most of the filarial species known to cross-react. Nevertheless, potential cross-reactions with hookworm species and other nematodes were of concern to us. In our study, participants stool positive for hookworm were four times as likely to be positive for S. stercoralis, an observation also noted in an earlier study.16 This is not surprising because the two nematodes share similar risk factors, such as going barefoot in contaminated soil. However, in our data, there was no demonstrable cross-reactivity between hookworm infection and seropositivity for S. stercoralis.
There are few previous studies of the seroepidemiology of strongyloidiasis in the literature. Some studies used sampling methods that select for higher risk individuals, and therefore cannot give valid estimates for risk factors or age-specific rates of infection.17,18 A study conducted in a leper colony in Thailand provides some data for comparison.19 Of 177 participants concurrently screened by stool examination and serologic tests, 11% had S. stercoralis identified in stool and 45% were seropositive. No association was found between age and serologic status, but the exclusion of children less than six years of age from serologic testing limited the ability to delineate the age-specific pattern of seropositivity. The area in question was reported to be non-endemic for filariasis, but because the village was a leper colony, people from other parts of Thailand may have migrated there, and individuals with prior exposure to filariasis may have been included.19
Other studies from the Amazon region show rates of S. stercoralis in stool of 16–20%, demonstrating that our study area is part of a highly disease-endemic zone.20,21 In addition, the factors associated with higher risk of strongyloidiasis by both assays, increasing age, going without shoes, and for stool infection, concurrent hookworm, are consistent with conventional knowledge of the parasite. The rate of infection based on our serologic assay was very high; 85% of adults 18 years old and 61% of children, showed seroreactivity for S. stercoralis. Published data show that ELISA optical density values decrease after successful anthelmintic treatment,22–24 but in one study, optical density values remained above the diagnostic cutoff in 32% of the cases.22 Thus, a person with past infection that cleared spontaneously or after treatment may remain seropositive, and positive serologic results in our data likely represent a mixture of past and current infections.
Given the persistent nature of infection and the mortality associated with the hyperinfection syndrome, we recommend that residents of the Amazon region be routinely screened for S. stercoralis if a diagnosis of HIV or another immunosuppressive condition is made, or prior to the institution of immunosuppressive therapy. Since the ELISA has a negative predictive value of 98%, this assay is a highly appropriate screening tool, as long as possible co-endemicity with cross-reacting filarial species is taken into consideration.
Received June 13, 2005. Accepted for publication August 17, 2005.
Acknowledgments: We thank Hermman Silva for his interest and collaboration in the current project, Franklin Neva for the provision of L3 extract and protocol for its use, Patrick J. Lammie for advice in the interpretation of serologic testing, M. A. Single for providing input and support on early versions of the work, and S. T. Unt for a critical revision of the final manuscript.
Financial support: Margaret Kosek was supported by Fogarty International Center International Research Scientist Development Award KO1TW005717. The project received additional support from grants T35AI107646 and Tropical Medicine Research Center Grant P01AI52976-01.
* Address correspondence to Margaret Kosek, Department of International Health, Johns Hopkins School of Public Health, 615 North Wolfe Street, Room 5515, Baltimore, MD 21205. E-mail: mkosek{at}jhsph.edu 
Authors’ addresses: Pablo P. Yori, Julianna Cordova, Cesar Banda Chavez, Maribel Paredes Olortegui, Carmen Montalvan, Graciela Meza Sanchez, Bevelle Worthen, James Worthen, Fay Leung, and Carlos Vidal Oré, Asociacion Benéfica PRISMA, Morona 452, Iquitos, Maynas, Peru, Telephone and Fax: 51-6-524-2226, E-mail: iqtlab{at}jhsph.edu. Margaret Kosek and Robert H. Gilman, Department of International Health, Johns Hopkins School of Public Health, 615 North Wolfe Street, Room 5515, Baltimore, MD 21205, Telephone: 410-995-4195, Fax: 410-510-1284, E-mails: mkosek{at}jhsph.edu and gilmanbob{at}yahoo.com. Caryn Bern, Division of Parasitic Diseases, National Center for Infectious Diseases, Centers for Disease Control and Prevention, Mailstop F-22, 4770 Buford Highway NE, Atlanta, GA 30341, Telephone: 770-488-7654, Fax: 770-488-7761, E-mail: cbern{at}cdc.gov.
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  M. Boscolo, M. Gobbo, W. Mantovani, M. Degani, M. Anselmi, G. B. Monteiro, S. Marocco, A. Angheben, M. Mistretta, M. Santacatterina, et al. Evaluation of an Indirect Immunofluorescence Assay for Strongyloidiasis as a Tool for Diagnosis and Follow-Up Clin. Vaccine Immunol., February 1, 2007; 14(2): 129 - 133. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
