Advertisement

A Quantitative Polymerase Chain Reaction Assay for Rapid Detection of 9 Pathogens Directly From Stools of Travelers With Diarrhea

  • Jenni Antikainen
    Affiliations
    Helsinki University Hospital Laboratory, Department of Bacteriology, Helsinki, Finland
    Search for articles by this author
  • Anu Kantele
    Correspondence
    Reprint requests Address requests for reprints to: Anu Kantele, MD, PhD, Division of Infectious Diseases, Department of Medicine, Helsinki University Central Hospital, PO Box 348, FIN-00029 HUS, Helsinki, Finland. fax: (358) 9-471-75900.
    Affiliations
    Department of Bacteriology and Immunology, Haartman Institute, University of Helsinki, Helsinki, Finland

    Department of Medicine, Division of Infectious Diseases, Helsinki University Central Hospital, Helsinki, Finland

    Aava Travel Clinic, Medical Centre Aava, Helsinki, Finland

    Department of Clinical Medicine, University of Helsinki, Helsinki, Finland
    Search for articles by this author
  • Sari H. Pakkanen
    Affiliations
    Department of Bacteriology and Immunology, Haartman Institute, University of Helsinki, Helsinki, Finland

    Department of Medicine, Division of Infectious Diseases, Helsinki University Central Hospital, Helsinki, Finland
    Search for articles by this author
  • Tinja Lääveri
    Affiliations
    Department of Medicine, Division of Infectious Diseases, Helsinki University Central Hospital, Helsinki, Finland
    Search for articles by this author
  • Jukka Riutta
    Affiliations
    Aava Travel Clinic, Medical Centre Aava, Helsinki, Finland
    Search for articles by this author
  • Martti Vaara
    Affiliations
    Helsinki University Hospital Laboratory, Department of Bacteriology, Helsinki, Finland
    Search for articles by this author
  • Juha Kirveskari
    Affiliations
    Helsinki University Hospital Laboratory, Department of Bacteriology, Helsinki, Finland
    Search for articles by this author

      Background & Aims

      Every year, 80 million tourists traveling to tropical and subtropical areas contract traveler's diarrhea (TD). Forty percent to 80% of cases are caused by bacteria, yet clinical diagnostic tests are available to identify only a few of the strains that cause TD. We aimed to develop a quantitative polymerase chain reaction (qPCR) assay to identify all major pathogens in stool samples.

      Methods

      We developed a low-cost, high-throughput, multiplex qPCR assay for simultaneous detection of 9 bacterial pathogens in stool samples: Salmonella, Yersinia, Campylobacter, and Vibrio cholerae, as well as Shigella or enteroinvasive Escherichia coli, enterohemorrhagic E coli, enterotoxigenic E coli (ETEC), enteroaggregative E coli (EAEC), and enteropathogenic E coli (EPEC). The assay was validated using positive (n = 245) and negative (n = 243) control strains, as well as preselected positive and negative stool samples. In addition, stool samples were collected from 96 returning travelers with TD. The findings were compared with those from routine diagnostic tests.

      Results

      The assay detected the bacterial strains with 100% sensitivity and specificity, compared with results from the reference tests. Of all stool samples collected from travelers with TD, EPEC was found in 47%, EAEC in 46%, ETEC in 22%, enterohemorrhagic E coli in 7%, Campylobacter in 6%, Shigella or enteroinvasive E coli in 2%, and Salmonella in 2%. Multiple pathogens were found in 37% of all samples.

      Conclusions

      We developed a low-cost, high-throughput qPCR assay for use in routine diagnostic analysis and research. It detects the pathogenic bacteria most commonly associated with TD in stool samples with 100% sensitivity and specificity, compared with reference methods. The assay requires 4 hours, whereas current detection methods require 1 to 7 days. At least 1 TD pathogen was identified in stool samples from 76% of returning travelers, whereas conventional methods found a pathogen in only 17%. The most commonly detected bacteria were EPEC, EAEC, and ETEC.

      Keywords

      Abbreviations used in this paper:

      EAEC (enteroaggregative Escherichia coli), EHEC (enterohemorrhagic Escherichia coli), EPEC (enteropathogenic Escherichia coli), ETEC (enterotoxigenic Escherichia coli), PCR (polymerase chain reaction), qPCR (quantitative polymerase chain reaction), TD (traveler’s diarrhea)
      Two hundred million tourists from industrialized countries travel to tropical and subtropical areas every year,

      United Nations World Tourism Organization (UNWTO). World Tourism Barometer. Available from: http://dtxtq4w60xqpw.cloudfront.net/sites/all/files/pdf/unwto_barom12_02_march_excerpt.pdf. Accessed: October 28, 2012.

      and approximately 80 million of them contract diarrhea (30%–70%, according to the Centers for Disease Control

      CDC Yellow book, 2012. Travelers' health, Chapter 2. The pre-travel consultation, self-treatable conditions. Available from: http://wwwnc.cdc.gov/travel/yellowbook/2012/chapter-2-the-pre-travel-consultation/travelers-diarrhea.htm. Accessed: October 28, 2012.

      ). Despite the high rate of traveler’s diarrhea (TD), in the majority of patients the causative agents remain unidentified because of poor diagnostics.
      Numerous studies have focused on a more detailed characterization of the TD pathogens, failing, however, to identify them in up to 50% of all cases.
      • Shah N.
      • DuPont H.L.
      • Ramsey D.J.
      Global etiology of travelers' diarrhea: systematic review from 1973 to the present.
      • Jiang Z.D.
      • Lowe B.
      • Verenkar M.P.
      • et al.
      Prevalence of enteric pathogens among international travelers with diarrhea acquired in Kenya (Mombasa), India (Goa), or Jamaica (Montego Bay).
      • Adachi J.A.
      • Jiang Z.D.
      • Mathewson J.J.
      • et al.
      Enteroaggregative Escherichia coli as a major etiologic agent in traveler's diarrhea in 3 regions of the world.
      The proportion remains similar even in many recent studies,
      • Pommier de Santi V.
      • Nicand E.
      • Lagathu G.
      • et al.
      Incidence, etiology, and determinants associated with acute diarrhea among French forces deployed to Chad.
      • Riddle M.S.
      • Rockabrand D.M.
      • Schlett C.
      • et al.
      A prospective study of acute diarrhea: in a cohort of United States military personnel on deployment to the Multinational Force and Observers, Sinai, Egypt.
      except for a few in which the percentage of unexplained cases has decreased to 6% to 31%.
      • Paschke C.
      • Apelt N.
      • Fleischmann E.
      • et al.
      Controlled study on enteropathogens in travellers returning from the tropics with and without diarrhoea.
      • Pandey P.
      • Bodhidatta L.
      • Lewis M.
      • et al.
      Travelers' diarrhea in Nepal: an update on the pathogens and antibiotic resistance.
      • Jiang Z.D.
      • Dupont H.L.
      • Brown E.L.
      • et al.
      Microbial etiology of travelers' diarrhea in Mexico, Guatemala, and India: importance of enterotoxigenic Bacteroides fragilis and Arcobacter species.
      • Tribble D.R.
      • Sanders J.W.
      • Pang L.W.
      • et al.
      Traveler's diarrhea in Thailand: randomized, double-blind trial comparing single-dose and 3-day azithromycin-based regimens with a 3-day levofloxacin regimen.
      In all these investigations, TD mainly is considered to be of bacterial origin. Viruses or parasites have been identified in 5% to 25% of cases.
      • Shah N.
      • DuPont H.L.
      • Ramsey D.J.
      Global etiology of travelers' diarrhea: systematic review from 1973 to the present.
      • Hill D.R.
      • Beeching N.J.
      Travelers' diarrhea.
      Most studies have reported enterotoxigenic Escherichia coli (ETEC) to be the most common pathogen, followed by Campylobacter, Salmonella, and Shigella.
      • Shah N.
      • DuPont H.L.
      • Ramsey D.J.
      Global etiology of travelers' diarrhea: systematic review from 1973 to the present.
      • Jiang Z.D.
      • Lowe B.
      • Verenkar M.P.
      • et al.
      Prevalence of enteric pathogens among international travelers with diarrhea acquired in Kenya (Mombasa), India (Goa), or Jamaica (Montego Bay).
      • Jiang Z.D.
      • Dupont H.L.
      • Brown E.L.
      • et al.
      Microbial etiology of travelers' diarrhea in Mexico, Guatemala, and India: importance of enterotoxigenic Bacteroides fragilis and Arcobacter species.
      Recently, methods based on molecular techniques have been used, and other bacteria, such as enteroaggregative E coli (EAEC), have proved substantially more common than anticipated previously.
      • Paschke C.
      • Apelt N.
      • Fleischmann E.
      • et al.
      Controlled study on enteropathogens in travellers returning from the tropics with and without diarrhoea.
      • Paredes-Paredes M.
      • Okhuysen P.C.
      • Flores J.
      • et al.
      Seasonality of diarrheagenic Escherichia coli pathotypes in the US students acquiring diarrhea in Mexico.
      The methods available for routine clinical stool analyses mostly consist of traditional cultures and immunoassays that are both slow and laborious and can identify only a few of the multitude of potential pathogens. At many laboratories this implies culturing Salmonella, Campylobacter, Shigella, and Yersinia and analyzing Clostridium difficile or enterohemorrhagic E coli (EHEC) on request. Approximately 90% of routine stool samples remain negative in traditional bacterial culture, which also makes the cost effectiveness of these analyses debatable. Newer studies using modern methodology based on polymerase chain reaction (PCR) are eagerly awaited because they may reveal a large number of causative agents, including the diarrheagenic E coli strains ETEC, EAEC, enteropathogenic E coli (EPEC), EHEC, and enteroinvasive E coli. However accurate the new PCR-based methods have proved, they have remained too laborious and time consuming to be adopted into common practice.
      • Operario D.J.
      • Houpt E.
      Defining the causes of diarrhea: novel approaches.
      We present a new multiplex quantitative PCR (qPCR) method that enables rapid detection of pathogens directly from stool samples, without cultures. Our assay uses a uniform assay design and covers 9 pathogens. The method is easy to perform, allows high-throughput analysis using a robotic platform, and provides results of a multitude of samples at low cost in a total of 4 hours. It can be applied to research as well as routine clinical work in developed countries. We report the results of validation experiments with preselected control samples and provide an account of how the method worked with immediate post-travel stool samples from travelers with diarrhea.

      Materials and Methods

       Ethics Statement

      The study protocol was approved by the ethics committee of the Helsinki University Central Hospital. The study was conducted in accordance with the principles outlined in the Declaration of Helsinki. Written informed consent was obtained from all study subjects.
      J.A. and J.K. have filed a patent application related to the new method.

       Study Design

      The study protocol is presented in Figure 1. The qPCR assay was first to be established with 1) preselected diarrheal and nondiarrheal bacterial control strains and with 2) preselected control stool samples analyzed in routine diagnostics. After this proof of concept, the assay was tested on 3) travelers with diarrhea providing stool samples immediately after their return home.
      Figure thumbnail gr1
      Figure 1The study protocol including the 2 parts of this report: the validation process and study of patients with TD.

       The Quantitative Polymerase Chain Reaction Assay

       Design of the quantitative polymerase chain reaction assay

      The PCR was designed to identify specific virulence genes, species-specific genes, or species-specific regions within established universal molecular clock genes (Table 1). The primers and probes were designed with Allele ID and Beacon Designer software (Palo Alto, CA) to recognize correct target genes and their global variants, including the Basic Local Alignment Search Tool search and secondary structure prediction using the National Center for Biotechnology Information (Bethesda, MD) database.
      Table 1Primers and Probes of the Nine Pathogens Assessed in This Study
      OriginGeneForward primer (5′→3′)Reverse primer (5′→3′)Probe (5′→3′)
      Multiplex 1
       EHECstx1GCGTTCTTATGTAATGACTGCTGAAGAGAAATTCTTCCTACACGAACAGAGTCTCCATGATARTCAGGCAGGACACTACTCAACCTTCC
       EHECstx2TGCATCCAGAGCAGTTCTGCCGGCGTCATCGTATACACAGGTTGTCACTGTCACAGCAGAAGCCTTACGC
       EHEC/EPECeaeCCAGGCTTCGTCACAGTTGCCAGTGAACTACCGTCAAAGTTATTACCAGATTAACCTCTGCCGTTCCATAATGTTGTAACCA
      SalmonellainvAGCTCTTCGGCACAAGTAATATCAACTCTATTTTAAATTCCGTGAAGCAAAACGCCAAACCTAAAACCAGTAAAGGCGAGCAGC
      Shigella/EIECipaHTGGTCCATCAGGCATCAGAAGGGGCAGTGCGGAGGTCATTTGTCACTCCCGACACGCCATAGAAACGCATTT
      Shigella/EIECinvETGTCTTTATAGGACATCCCTGATACTTTCTATCTACTCTTGATGCCAGAAAACTAGCACAAACAGCAAAAGAGCATAGCATCCGAGAACT
      Multiplex 2
       ETECestAAAATTGCAAAATCCGTTTAACTAATCGACTGACTAAAAGAGGGGAAAGcaaAtaTccGtgAaacaacatgac
       ETECestTCCTGAAAGCATGAATAGTAGCTTATTAATAGCACCCGGTACAAGaggAttAcaAcaCaattcacagcagt
       ETECeltCCGGCAGAGGATGGTTACAGTTGATTGATATTCCCTGAGATATATTGTGAGCAGGTTTCCCACCGGATCACCA
       EAECaggRGGAAGCAATACATATCTTAGAAATGAACTCTCGGACAACTGCAAGCATCTACTCCGTATATTATCATCAGGGCATCCTTTAGGCGT
      Campylobacter jejunirimMGAGTGAAAAAGATTTTGTTCAAGTTGAAAAGTCGCTCAGGTTATGCaagAccCacAgtTttaccaagtttt
      Campylobacter coligyrBAGTGCCTGAACCTCAATTTGTCGATAGGATTTTCTTCAAAATATTTACaacTtgGctCttCttatgtgcgt
      Y enterocolitica/pseudotuberculosisvirFGTTTGGTACAGTTTATGGCATTTCACCATGGCAATATCAACAATACTCATCTTACCCTGGATAAGCGAGCGACGTATTCTCTATGC
      Y pseudotuberculosis/pestisrumBCAGGAGCATGAGGTTCACAGTATGTCTCTGGCCCCGCACAATGAAACCAAAGCCGCCCACACCACAG
      Multiplex 3
       Oryza sativa, terminal flower geneoryCTAATCCCAGCAACCCAACCCTAATCAATGTGAGACATATGATAGAAATCcctGcaCtgGtaAgctatg
      VibrioVib16S_01GTCATCATGGCCCTTACGAGTAGCCGGACTACGACGTACTTTGTGACCTCGCGGTATCCGTGCCCTC
      V choleraectxGGGCTACAGAGATAGATATTACAGTAACTTAGCCACGGCTCTTCCCTCCAAGAACCTGCCAATCCATAACCATCTGCTGCTG
      NOTE. Lowercase letters represent normal nucleotides; capitalized letters represent locked nucleic acid modification.
      EIEC, enteroinvasive E coli.
      The qPCR was performed on the Mx3005P detection system (Stratagene, La Jolla, CA), and thermocycling was performed in the following conditions: initial denaturation and activation at 95°C for 15 minutes, 40 cycles of denaturation at 94°C for 1 minute, and annealing/extension at 60°C for 1 minute. Fluorescence was recorded at each annealing step. The 20-μL reaction contained 1× Multitect NoROX master mix (Qiagen, Venlo, the Netherlands), 1 μL of primer/probe mix (Table 1), and 0.5 μL template DNA.
      For PCR analyses, bacterial cells were collected in 100 μL of water, boiled for 15 minutes, centrifuged at 13,000 rpm for 1 minute, and the supernatant (0.5 μL) was used in PCR reactions. Bacterial DNA was purified with the NucliSENS kit (Durham, NC) using the easyMAG automatic nucleic acid purification platform (bioMérieux, Marcy l'Etoile, France), as described by the manufacturer.
      Total nucleic acids were purified directly from stool swabs with the NucliSENS kit using the easyMAG platform, as described previously.
      • Antikainen J.
      • Tarkka E.
      • Haukka K.
      • et al.
      New 16-plex PCR method for rapid detection of diarrheagenic Escherichia coli directly from stool samples.
      Briefly, stool swabs were suspended in 100 μL of Tris-EDTA buffer, purified by the general method of the easyMAG platform, and eluted to a volume of 25 μL. The eluate (0.5 μL) was used as a template in PCR.

       Specificity of the Polymerase Chain Reaction

       Preselected isolated control strains

      The specificity of the PCR was analyzed by using 245 diarrheal bacterial control strains including Salmonella, Shigella, Campylobacter, Yersinia, and Vibrio strains, as well as diarrheal E coli strains (Supplementary Table 1), and 243 control strains not belonging to diarrheal pathogens (Supplementary Tables 1 and 2).
      The strains were obtained from the Helsinki University Hospital Laboratory or the National Institute for Health and Welfare (Terveyden ja Hyvinvoinnin Laitos, THL) or received as kind gifts from M. Alexander Schmidt and Inga Benz (Westfälische Wilhelms-Universität, Münster, Germany), Isabel Scaletsky (Universidade Federal de São Paulo, Brazil), or Lin Thorstensen Brandal (The Norwegian Institute of Public Health, Norway).

       Preselected control stool samples

      Control stool samples, either negative or positive in routine diagnostics, were analyzed with the qPCR assay (Supplementary Table 3). These samples had been cultured in routine diagnostics at the HUSLAB for Salmonella, Yersinia, Shigella, Campylobacter, Vibrio cholerae, and/or EHEC with standard biochemical methods according to Clinical and Laboratory Standards Institute (Palo Alto, CA) guidelines. No clinical data were available for these control stools. However, because TD is a major indication for examining stool cultures, a significant proportion of these stools can be presumed to be from travelers.
      Diarrheal samples positive for ETEC (n = 8), EPEC (n = 11), or EAEC (n = 13) that were available from our previous study
      • Antikainen J.
      • Tarkka E.
      • Haukka K.
      • et al.
      New 16-plex PCR method for rapid detection of diarrheagenic Escherichia coli directly from stool samples.
      also were analyzed (Supplementary Table 3).

       Analytical Sensitivity of Polymerase Chain Reaction

      To analyze sensitivity for clinical use, a mixture of DNAs containing all templates purified by easyMAG for each amplicon were diluted 10-fold and analyzed by qPCR in 10 parallel reactions. The cut-off value of analytical sensitivity was determined as the lowest dilution at which 90% or more parallel reactions were amplified successfully. In addition, the amplification of each reporter was analyzed separately in 10-fold dilutions using boiled bacterial mass. Shortly, bacteria were grown on agar plates and collected to Tris-EDTA buffer, and the viable count in colony-forming units was determined. Bacteria were diluted and boiled for 15 minutes, centrifuged at 13,000 rpm for 1 minute, and the supernatant (0.5 μL) was used in PCR reactions.

       Travelers, Questionnaires, and Stool Samples

      A total of 200 travelers were enrolled during their pretravel visit to the Aava Travel Clinic, Medical Centre Aava (formerly Medicity), Helsinki, Finland, between March 1, 2009, and February 29, 2010. To obtain information on age, sex, journey, and possible diarrheal symptoms, the volunteers were asked to fill in 2 questionnaires, 1 before travel and the other immediately after travel. Stool samples were collected immediately upon returning home and, if not practicable, from their second stools. The samples were collected as 2 swabs in M40-tubes (Copan, Brescia, Italy): swab 1 for analysis with qPCR, and swab 2 for reference methods. The tubes were sealed in special mailing envelopes and delivered to the laboratory within 1 to 3 days.
      The final study group included 96 travelers, who all had returned their pretravel questionnaires, contracted diarrhea abroad, and sent their post-travel stool samples to the laboratory. The researchers defined TD according to the World Health Organization criteria

      World Health Organization (WHO). Health Topics: Diarrhoea. Available from: http://www.who.int/topics/diarrhoea/en/. Accessed: October 28, 2012.

      as the passage of 3 or more loose or liquid stools per day, or more frequently than is normal for the individual. The travelers were divided into 2 groups, those with and those without any ongoing TD-related symptoms at the time of sampling. The countries visited were grouped into 7 regions according to assumed risk for TD, as modified from the GeoSentinel Surveillance Network.
      • Freedman D.O.
      • Weld L.H.
      • Kozarsky P.E.
      • et al.
      GeoSentinel Surveillance Network
      Spectrum of disease and relation to place of exposure among ill returned travelers.

       Identification of Bacteria From Travelers’ Samples

      Swab 1 from each traveler was examined by the new qPCR method along with positive and negative control samples. If swab 1 proved positive for Salmonella, Shigella, Yersinia, Campylobacter, and EHEC, swab 2 was analyzed by bacterial culture, and strains were identified with conventional diagnostic methods. From samples with unsuccessful isolation of these bacterial strains, the corresponding genes were amplified separately and sequenced in the Sequence Core Facility at the Haartman Institute (Helsinki, Finland). Sequences were identified by the Basic Local Alignment Search Tool (available at: http://blast.ncbi.nlm.nih.gov/Blast.cgi), and sequences with 99% or greater identity were considered correct.

       Statistical Methods

      Differences between the traveler groups were analyzed by chi-square tests using SPSS 19.0.0.1 (SPSS, Inc, Chicago, IL).

      Results

       Design and Validation of the Quantitative Polymerase Chain Reaction Method

      A qPCR method was designed to identify specific virulence genes, species-specific genes, or species-specific regions within established universal molecular clock genes (Table 1) of Salmonella, Shigella/enteroinvasive E coli, Campylobacter, Yersinia enterocolitica/pseudotuberculosis, V cholerae, and diarrheagenic E coli. With each sample, an internal positive control (Oryza sativa, terminal flower gene) was tested, and no significant inhibition was detected.
      The qPCR assay was optimized and validated using the reference strains including 245 stool pathogens (Supplementary Table 1) and 243 strains belonging to other major genera (Supplementary Tables 1 and 2). All positive control strains were identified correctly, and no false-positive amplification was obtained. Thus, the assay attained 100% analytical specificity.
      The specificity also was analyzed directly from fecal samples obtained from routine diagnostics. When the routine samples positive for Salmonella, Campylobacter, Yersinia, Shigella, or EHEC (Supplementary Table 3) were analyzed with qPCR, all provided correct amplification, yet 2 additional strains, Campylobacter and EHEC, were identified by qPCR from Salmonella-positive samples and were verified by sequencing. The samples that were shown to be positive in previous studies for EAEC, EPEC, and ETEC
      • Antikainen J.
      • Tarkka E.
      • Haukka K.
      • et al.
      New 16-plex PCR method for rapid detection of diarrheagenic Escherichia coli directly from stool samples.
      proved correctly analyzed with the current qPCR method, too (Supplementary Table 3). Culture-negative samples (Supplementary Table 3) also were analyzed, and all but one proved negative for Salmonella, Campylobacter, Yersinia, Shigella, V cholerae, and/or EHEC. From the sample negative in culture, EHEC was amplified by qPCR and verified by sequencing. Thus, the clinical specificity of the assay was 100%.
      The analytical sensitivity of the PCR was defined by 10-fold dilutions of the template DNA mixture analyzed by PCR (Supplementary Table 4). The sensitivity with 40 amplification cycles was 0.1 ng/mL for EHEC and Salmonella, and 1 ng/mL for the others. For the V cholerae toxin gene, the sensitivity was 10 ng/mL. In addition, the sensitivity was measured from DNA obtained with boiling the bacterial strain. In that assay, the limit of detection was 0.5 to 50 colony-forming units per reaction (Supplementary Table 4). These results represent the lowest concentration required for correct identification (>90% positive).

       Analysis of the Travelers’ Samples

      The qPCR method was used in the analysis of the fecal samples of 96 travelers (see later for results). All samples positive for Campylobacter, Yersinia, Shigella, Salmonella, or EHEC were confirmed by colony isolation and/or by sequencing the PCR product: the results of these analyses proved consistent with each other. Because no culture method exists for other diarrheagenic E coli, they could not be verified similarly.

       Description of the travelers and travels

      The final study population comprised a total of 96 travelers whose demographics, clinical data, and travel-related data are presented in Table 2. The TD lasted 1 to 3 days in 76% of the subjects. The median time from the onset of TD symptoms (or the latest episode, if several bouts of diarrhea) to the time of stool sampling was 7 days (standard deviation, 7.7 d; range, 0–38 d).
      Table 2Details on the 96 Travelers With TD and Their Travels
      All subjects, n

      (N = 96) (% of all)
      Traveler demographics
       Female sex58 (60)
       Male sex38 (40)
       Mean age, y (range)37 (1–72)
      Destination
      Adapted from Geosentinel.17
       Sub-Saharan Africa40 (42)
       Southeast Asia27 (28)
       South and Central America and the Caribbean11 (12)
       South Asia10 (10)
       North Africa and Middle East4 (4)
       East Asia2 (2)
       Europe, North America or Australia2 (2)
      Length of journey
      Information was missing from 9 travelers.
       Mean days (range)16 (4–100)
       <15 days41 (47)
       <30 days75 (86)
      Total96 (100)
      a Adapted from Geosentinel.
      • Freedman D.O.
      • Weld L.H.
      • Kozarsky P.E.
      • et al.
      GeoSentinel Surveillance Network
      Spectrum of disease and relation to place of exposure among ill returned travelers.
      b Information was missing from 9 travelers.

       Results of quantitative polymerase chain reaction analyses for bacterial pathogens in the travelers

      A pathogen was detected in 76% of the fecal samples (Table 3); 37% had 2 or more of the potential pathogens. EPEC (47%) and EAEC (46%) were the most common findings, followed by ETEC (22%). Even after the actual TD had subsided, some of the travelers had minor symptoms, such as loose stools or abdominal discomfort, for several days. The travelers were grouped according to any ongoing symptoms related to TD at the time of sampling (Table 3). In the group with ongoing symptoms, the most common finding was EPEC (49%), followed by EAEC (38%) and ETEC (32%). In the group with symptoms that had subsided, EAEC proved to be the most common finding (53%), followed by EPEC (45%) and ETEC (12%).
      Table 3Number of Patients With TD With the Various Pathogens in Their Stool Samples
      All travelers

      (N = 96) (% of all)
      Gastrointestinal symptoms ongoing
      Defined as any ongoing symptoms related to TD (eg, diarrhea, loose stools, abdominal cramps, nausea).
      at the time of sampling

      (n = 47) (% of these)
      Gastrointestinal symptoms subsided
      Defined as any ongoing symptoms related to TD (eg, diarrhea, loose stools, abdominal cramps, nausea).
      at the time of sampling

      (n = 49) (% of these)
      P value with chi-square testDays (median) between onset of symptoms and sampling, patients with ongoing symptoms
      Defined as any ongoing symptoms related to TD (eg, diarrhea, loose stools, abdominal cramps, nausea).
      (information available/missing)
      Days (median) between onset of symptoms and sampling, patients with subsided symptoms
      Defined as any ongoing symptoms related to TD (eg, diarrhea, loose stools, abdominal cramps, nausea).
      (information available/missing)
      No bacterial pathogens23 (24)13 (28)10 (20).4053 (13/0)13 (5/5)
      Bacterial pathogen73 (76)34 (72)39 (80).4054 (30/4)13 (23/16)
      Single pathogen37 (39)14 (30)23 (47).0843 (12/2)13 (17/6)
      ≥2 pathogens36 (37)20 (43)16 (33).3175 (18/2)13 (6/10)
      ≥3 pathogens15 (16)8 (17)7 (14).7129 (8/0)12 (2/5)
      EPEC45 (47)23 (49)22 (45).6924 (21/2)13 (10/12)
      EAEC44 (46)18 (38)26 (53).1474 (16/2)13 (16/10)
      ETEC21 (22)15 (32)6 (12).020
      Difference was statistically significant.
      4 (14/1)8 (3/3)
      EHEC7 (7)3 (6)4 (8).7376 (2/1)9 (1/3)
      Shigella/EIEC2 (2)2 (4)0 (0).1445 (2/0)N/A (0/0)
      Campylobacter6 (6)1 (2)5 (10).10221 (1/0)16 (1/4)
      Salmonella2 (2)1 (2)1 (2).9762 (1/0)24 (1/3)
      Yersinia0 (0)0 (0)0 (0)N/AN/AN/A
      V cholerae0 (0)0 (0)0 (0)N/AN/AN/A
      Total96 (100)47 (49% of all)49 (51% of all)3 (43/4)13 (28/21)
      NOTE. The figures are provided separately for all travelers and for the groups with or without gastrointestinal symptoms at the time of sampling. The medians for days since onset of symptoms are provided for both groups.
      EIEC, enteroinvasive E coli; N/A, not applicable.
      a Defined as any ongoing symptoms related to TD (eg, diarrhea, loose stools, abdominal cramps, nausea).
      b Difference was statistically significant.

      Discussion

      Although the conventional diagnostics only manage to detect a few of the diarrheal pathogens, and actually fail to identify the most common ones, new PCR-based methods have in recent studies offered a significantly better coverage of the variety of pathogens.
      • Meraz I.M.
      • Jiang Z.D.
      • Ericsson C.D.
      • et al.
      Enterotoxigenic Escherichia coli and diffusely adherent E. coli as likely causes of a proportion of pathogen-negative travelers' diarrhea–a PCR-based study.
      They have, however, mostly remained laborious, slow, and expensive, and therefore not applicable to routine diagnostics. The new qPCR assay described in this article proved to have several advantages over the classic PCR methods, as follows: (1) broad coverage: 9 important causative agents of diarrhea were analyzed simultaneously; (2) rapidity: assays were performed in 4 hours (instead of 1–7 days as required by current routine methods); (3) high throughput: 48 samples were analyzed in 4 hours; and (4) easy performance: assay was performed directly from stools without prior culture and no specific microbiological competence was needed. All these properties are required for a method to be applicable in various clinical settings. Methods of this kind are expected to revolutionize both the diagnostics and research of TD. The present study shows that our method can be applied easily to both purposes.

       Methodologic Considerations

      Several recent articles have described single PCRs, multiplex PCRs with agarose gel electrophoresis, or SYBR Green-based melting curve detection (Life Technologies, Carlsbad, CA) for the analysis of stool pathogens.
      • Antikainen J.
      • Tarkka E.
      • Haukka K.
      • et al.
      New 16-plex PCR method for rapid detection of diarrheagenic Escherichia coli directly from stool samples.
      • Guion C.E.
      • Ochoa T.J.
      • Walker C.M.
      • et al.
      Detection of diarrheagenic Escherichia coli by use of melting-curve analysis and real-time multiplex PCR.
      • Barletta F.
      • Ochoa T.J.
      • Ecker L.
      • et al.
      Validation of five-colony pool analysis using multiplex real-time PCR for detection of diarrheagenic Escherichia coli.
      More advanced methods may use bead-based systems, and even microarrays.
      • Taniuchi M.
      • Verweij J.J.
      • Noor Z.
      • et al.
      High throughput multiplex PCR and probe-based detection with Luminex beads for seven intestinal parasites.
      • You Y.
      • Fu C.
      • Zeng X.
      • et al.
      A novel DNA microarray for rapid diagnosis of enteropathogenic bacteria in stool specimens of patients with diarrhea.
      • Jin D.
      • Qi H.
      • Chen S.
      • et al.
      Simultaneous detection of six human diarrheal pathogens by using DNA microarray combined with tyramide signal amplification.
      For routine clinical diagnostic purposes, a high-throughput method with minimal manual intervention and highly automated interpretation of results is needed. Ideally, the method of choice would be a qPCR analysis. This objective is hampered by the fact that bacterial pathogens and their virulence factors, such as ETEC toxins, often are genetically highly variable,
      • Taxt A.
      • Aasland R.
      • Sommerfelt H.
      • et al.
      Heat-stable enterotoxin of enterotoxigenic Escherichia coli as a vaccine target.
      • Jobling M.G.
      • Holmes R.K.
      Type II heat-labile enterotoxins from 50 diverse Escherichia coli isolates belong almost exclusively to the LT-IIc family and may be prophage encoded.
      yet, in a good clinical assay, they all should be detected. In addition, the virulence genes may contain repetitive elements or have a highly variable guadinine-cytosine content, which will lead to major problems when designing a high-performance assay with a uniform chemical reaction mixture composition and thermal cycling conditions. Therefore, methods covering only a part of the diarrheal pathogens have been published to date. Such solutions, however, will not suffice to replace bacterial culture in primary screening because most requirements related to minimal hands-on time and avoidance of unnecessary plating of negative samples are not fulfilled.
      Among the study patients, we were able to identify up to 4 different pathogens from one sample, showing that multiple pathogens can be detected in parallel. Nevertheless, similar to culture methods, a typical PCR reaction inherently favors the most abundant target. A false-negative result most likely occurs in the case of a low copy pathogen in a stool sample together with 1 or 2 highly abundant pathogens. This possibility can be controlled only by other methods and/or re-sampling. To minimize the risk of false-negative results, the multiplex composition was designed so that the most frequent pathogens would be kept in different multiplex reactions.
      The assay sensitivity and specificity proved extremely high, 100%, when compared with independent reference methods traditionally considered the golden standard. Combined with the numerous advantages listed earlier, these data suggest that our method could replace stool culture as a primary means of screening causative agents in traveler’s diarrhea. In any case, the high proportion of patients with diarrheagenic E coli (75%) suggests that at least these pathogens should be analyzed by PCR.

       Clinical Considerations of the Methodology

      Rapid, accurate TD diagnostics would prove highly advantageous in many clinical situations. Although the great majority of TD cases are treated in primary health care, even at hospital emergency departments diarrhea is the most common symptom with which febrile returning travelers present.
      • Siikamäki H.M.
      • Kivelä P.S.
      • Sipilä P.N.
      • et al.
      Fever in travelers returning from malaria-endemic areas: don't look for malaria only.
      Instead of 1 to 7 days required for fecal cultures, the present assay provides results in 4 hours, often helping the physician to refrain from prescribing unnecessary antimicrobials. In many cases, a number of other etiologic investigations and even potential hospitalization could be avoided.
      In Europe, no general guidelines exist, but antimicrobials are not usually recommended for patients with short-lasting diarrhea without bloody stools or fever.
      • Hill D.R.
      • Beeching N.J.
      Travelers' diarrhea.
      Notably, the finding of diarrheagenic E coli per se should not encourage treatment with antimicrobials but instead support the decision to refrain from prescribing antibiotics to patients with mild or moderate symptoms. This is especially evident with EHEC infections, in which antimicrobials are considered harmful. On the other hand, rapid detection of Campylobacter could help physicians to choose macrolides, especially for patients returning from Southeast Asia, where the majority of Campylobacter strains are fluoroquinolone-resistant.
      • Pandey P.
      • Bodhidatta L.
      • Lewis M.
      • et al.
      Travelers' diarrhea in Nepal: an update on the pathogens and antibiotic resistance.
      • Tribble D.R.
      • Sanders J.W.
      • Pang L.W.
      • et al.
      Traveler's diarrhea in Thailand: randomized, double-blind trial comparing single-dose and 3-day azithromycin-based regimens with a 3-day levofloxacin regimen.
      • Hakanen A.
      • Jousimies-Somer H.
      • Siitonen A.
      • et al.
      Fluoroquinolone resistance in Campylobacter jejuni isolates in travelers returning to Finland: association of ciprofloxacin resistance to travel destination.
      Rapid information about Salmonellosis in persons handling food would be epidemiologically valuable: in many countries they must be prevented from handling food until they no longer secrete Salmonellae. Our test also would prove valuable when showing negative results: a negative multiplex PCR may suggest a nonbacterial etiology for the diarrhea. It should discourage the use of antimicrobials and instead lead the physician to consider viral, parasitic, systemic, or noninfectious etiology. For example, diarrhea can be a symptom of more severe conditions, such as malaria or sepsis. On the other hand, finding a pathogen in the stools does not rule out other simultaneous infections.

       Considerations of the Etiology of Traveler's Diarrhea in Travelers

      Even if our study covered all the major bacterial pathogens, Aeromonas species, Plesiomonas, toxigenic Bacteroides fragilis, Arcobacter, and diffusely adherent E coli were not included. Their pathogenic role and incidence is not fully understood as yet, and, based on some previous studies, their relative proportion is expected to be low.
      • Shah N.
      • DuPont H.L.
      • Ramsey D.J.
      Global etiology of travelers' diarrhea: systematic review from 1973 to the present.
      • Jiang Z.D.
      • Lowe B.
      • Verenkar M.P.
      • et al.
      Prevalence of enteric pathogens among international travelers with diarrhea acquired in Kenya (Mombasa), India (Goa), or Jamaica (Montego Bay).
      • Jiang Z.D.
      • Dupont H.L.
      • Brown E.L.
      • et al.
      Microbial etiology of travelers' diarrhea in Mexico, Guatemala, and India: importance of enterotoxigenic Bacteroides fragilis and Arcobacter species.
      • Vila J.
      • Ruiz J.
      • Gallardo F.
      • et al.
      Aeromonas spp. and traveler's diarrhea: clinical features and antimicrobial resistance.
      Notably, the conventional diagnostic methods would have identified a bacterial pathogen in only 17% of our subjects, whereas the new method showed a pathogen in 76%. The latter figure is consistent with estimates of TD being of bacterial origin in 40% to 80% of cases.
      • Shah N.
      • DuPont H.L.
      • Ramsey D.J.
      Global etiology of travelers' diarrhea: systematic review from 1973 to the present.
      • Hill D.R.
      • Beeching N.J.
      Travelers' diarrhea.
      Importantly, 37% of our travelers had multiple pathogens. Paschke et al
      • Paschke C.
      • Apelt N.
      • Fleischmann E.
      • et al.
      Controlled study on enteropathogens in travellers returning from the tropics with and without diarrhoea.
      have reported multiple pathogens in 61% of travelers seeking medical care because of TD. These data call for future studies to confirm the actual significance of each TD pathogen as an agent causing the symptoms.
      A comparison between those with and those without ongoing symptoms showed that except for ETEC, the findings did not differ significantly between these 2 groups, but instead the secretion of the pathogens lasts beyond the symptomatic phase. Moreover, diarrheal pathogens also have been reported in asymptomatic individuals.
      • Paschke C.
      • Apelt N.
      • Fleischmann E.
      • et al.
      Controlled study on enteropathogens in travellers returning from the tropics with and without diarrhoea.
      • Pandey P.
      • Bodhidatta L.
      • Lewis M.
      • et al.
      Travelers' diarrhea in Nepal: an update on the pathogens and antibiotic resistance.
      • Keskimäki M.
      • Mattila L.
      • Peltola H.
      • et al.
      Prevalence of diarrheagenic Escherichia coli in Finns with or without diarrhea during a round-the-world trip.
      Accordingly, the significance of the stool analysis results always should be evaluated with respect to the clinical picture. Although our assay covers 9 TD pathogens, some other, less-common agents still remain unexplored. In addition, E coli strains carrying more than one virulence factor may be interpreted as 2 separate pathogens. In outbreak situations the causative organisms always should be investigated further to guarantee identification of EHEC, or any new or modified bacteria harboring additional virulence factors.
      EAEC and EPEC were the most common findings in our travelers. This stands in contrast to numerous previous studies showing ETEC as the most frequent cause of TD.
      • Shah N.
      • DuPont H.L.
      • Ramsey D.J.
      Global etiology of travelers' diarrhea: systematic review from 1973 to the present.
      • Hill D.R.
      • Beeching N.J.
      Travelers' diarrhea.
      However, in the majority of studies conducted, EPEC and/or EAEC were not determined. The clinical significance of these 2 pathogens still needs to be studied further. The high prevalence of EAEC among our travelers is consistent with findings in recent studies using modern methods.
      • Paschke C.
      • Apelt N.
      • Fleischmann E.
      • et al.
      Controlled study on enteropathogens in travellers returning from the tropics with and without diarrhoea.
      • Paredes-Paredes M.
      • Okhuysen P.C.
      • Flores J.
      • et al.
      Seasonality of diarrheagenic Escherichia coli pathotypes in the US students acquiring diarrhea in Mexico.

      Conclusions

      We present a novel, low-cost, and rapid high-throughput multiplex qPCR with high specificity and sensitivity, and applicability to routine diagnostics as well as research. Direct analysis of stool samples revealed a pathogen in 76% of travelers, with TD, EPEC, EAEC, and ETEC being the most common findings; 37% had multiple pathogens. Clinical routine diagnostics would have detected a pathogen in only 17% of cases, leaving the 3 most common findings unidentified. This method has indeed become officially available at our laboratory routine diagnostics. It enables heath care personnel to make rapid and informed decisions.

      Acknowledgments

      J.A. and A.K. contributed equally.

      Supplementary Material

      Supplementary Table 1Strains Used as Preselected Positive (n = 245) and Negative (n = 243) Controls in the qPCR Assay
      SpeciesTotal number of strainsStrains positive with qPCR
      Positive control strains245245
      Campylobacter4343
      Salmonella2929
      Yersinia2727
      Shigella1616
      V cholera44
       EHEC4747
       EAEC1818
       ETEC1919
       EPEC3131
       Enteroinvasive E coli1111
      Negative control strains
      The list of negative control strains tested is shown in Supplementary Table 2.
      2430
      Total488
      NOTE. The figures indicate numbers of strains belonging to the given species. The species initially were identified with reference methods including traditional cultures or biochemical methods.
      a The list of negative control strains tested is shown in Supplementary Table 2.
      Supplementary Table 2Strains Used as Preselected Negative (n = 243) Controls in the qPCR Assay
      Negative control strainNumberNegative control strainNumber
      Achromobacterium species1Klebsiella pneumoniae5
      Acinetobacter baumannii3Lactobacillus species4
      Acinetobacter species2Lactococcus species1
      Actinomyces naeslundii1Leuconoctoc species1
      Actinomyces odontolyticus1Listeria monocytogenes3
      Actinomyces species3Micrococcus species3
      Aerococcus urinae2Parvimonas micra1
      Aerococcus viridans2Moraxella catarrhalis1
      Aeromonas caviae2Moraxella nonliquefaciens1
      Aeromonas hydrophila2Moraxella species1
      Aeromonas sobria1Morganella morganii3
      Aeromonas species3Neisseria meningitidis B2
      Aggregatibacter actinomycetemcomitans2Neisseria meningitidis C3
      Alcaligenes faecalis1Neisseria species1
      Arcanobacterium haemolyticus1Nocardia species2
      Bacillus cereus3Pasteurella multocida1
      Bacillus species3Peptoniphilus asaccharolyticus1
      Bacteroid fragilis group4Peptostreptococcus species1
      Brevundimonas diminuta1Prevotella bivia1
      Chryseomanas luteola2Prevotella buccae1
      Citrobacter amalonaticus2Prevotella loesheii1
      Citrobacter freundii3Prevotella oralis1
      Clostridium clostridioforme1Prevotella species1
      Clostridium difficile1Propionibacterium acnes3
      Clostridium perfringens2Propionibacterium species3
      Clostridium septicum2Proteus mirabilis3
      Clostridium sordellii4Proteus vulgaris3
      Clostridium tertium2Pseudomonas aeruginosa5
      Corynebacterium jeikeium3Pseudomonas fluorescens3
      Corynebacterium species2Pseudomonas putida1
      Delftia acidovorans1Pseudomonas stutzeri2
      Eggerthella lenta2Rothia mucilaginosus1
      Eikenella corrodens1Rothia species3
      Enterobacter aerogenes3Serratia fonticola1
      Enterobacter cloacae2Serratia marcescens3
      Enterobacter species1Staphylococcus aureus3
      Enterococcus avium3Staphylococcus epidermidis3
      Enterococcus casseliflavus3Staphylococcus haemolyticus species1
      Enterococcus durans1Staphylococcus hominis1
      Enterococcus faecalis3Stenotrophomonas maltophilia3
      Enterococcus faecium3Streptococcus agalactiae (B)3
      Enterococcus gallinarum3Streptococcus anginosus group3
      Escherichia coli6Streptococcus bovis (D)3
      Eubacterium species2Streptococcus constellatus3
      Fusobacterium necrophorum1Streptococcus dysgalactiae species equisimilis5
      Fusobacterium nucleatum3Streptococcus intermedius4
      Gardnerella vaginalis3Streptococcus mitis5
      Gemella haemolysans1Streptococcus mutans3
      Gemella species1Streptococcus pneumoniae3
      Granulicatella adiacens1Streptococcus pyogenes (A)3
      Hafnia alvei3Streptococcus salivarius6
      Haemophilus influenzae3Streptococcus sanquinis2
      Klebsiella oxytoca4Veillonella2
      NOTE. The figures indicate the number of strains belonging to the given species. The species initially were identified with reference methods including traditional cultures or various biochemical methods.
      Supplementary Table 3PCR Results in Control Stool Samples
      SpeciesStool samples positive with reference methodsStool samples positive with qPCR
      Campylobacter5556
      Disconcordant results were verified by sequencing.
      Salmonella5252
      Yersinia88
      Shigella2121
      EHEC1415
      Disconcordant results were verified by sequencing.
      EAEC1313
      ETEC88
      EPEC1111
      Stool samples negative with routine reference methodsStool samples negative with qPCR
      Campylobacter112112
      Salmonella112112
      Yersinia112112
      Shigella112112
      EHEC4342
      One sample that was negative in culture proved positive in PCR and was verified by sequencing.
      V cholerae5050
      a Disconcordant results were verified by sequencing.
      b One sample that was negative in culture proved positive in PCR and was verified by sequencing.
      Supplementary Table 4Analytical Sensitivity of the Assay
      DNA, ng/mLBacteria/mL
      Multiplex 1
       EHEC0.11.00E+04
       EHEC/EPEC11.00E+04
      Salmonella0.11.00E+04
      Shigella/EIEC11.00E+05
      Multiplex 2
       ETEC11.00E+04
       EPEC11.00E+05
      Campylobacter
      Campylobacter coli: 1.00E+03 colony-forming units/mL.
      11.00E+04
      Yersinia11.00E+04
      Multiplex 3
       Internal controlN/AN/A
      V cholerae11.00E+03
      V cholerae toxin101.00E+05
      NOTE. The sensitivity cut-off limit was 90% of parallel reactions being positive.
      EIEC, enteroinvasive E coli; N/A, not applicable.
      a Campylobacter coli: 1.00E+03 colony-forming units/mL.

      References

      1. United Nations World Tourism Organization (UNWTO). World Tourism Barometer. Available from: http://dtxtq4w60xqpw.cloudfront.net/sites/all/files/pdf/unwto_barom12_02_march_excerpt.pdf. Accessed: October 28, 2012.

      2. CDC Yellow book, 2012. Travelers' health, Chapter 2. The pre-travel consultation, self-treatable conditions. Available from: http://wwwnc.cdc.gov/travel/yellowbook/2012/chapter-2-the-pre-travel-consultation/travelers-diarrhea.htm. Accessed: October 28, 2012.

        • Shah N.
        • DuPont H.L.
        • Ramsey D.J.
        Global etiology of travelers' diarrhea: systematic review from 1973 to the present.
        Am J Trop Med Hyg. 2009; 80: 609-614
        • Jiang Z.D.
        • Lowe B.
        • Verenkar M.P.
        • et al.
        Prevalence of enteric pathogens among international travelers with diarrhea acquired in Kenya (Mombasa), India (Goa), or Jamaica (Montego Bay).
        J Infect Dis. 2002; 185: 497-502
        • Adachi J.A.
        • Jiang Z.D.
        • Mathewson J.J.
        • et al.
        Enteroaggregative Escherichia coli as a major etiologic agent in traveler's diarrhea in 3 regions of the world.
        Clin Infect Dis. 2001; 32: 1706-1709
        • Pommier de Santi V.
        • Nicand E.
        • Lagathu G.
        • et al.
        Incidence, etiology, and determinants associated with acute diarrhea among French forces deployed to Chad.
        J Travel Med. 2011; 18: 115-120
        • Riddle M.S.
        • Rockabrand D.M.
        • Schlett C.
        • et al.
        A prospective study of acute diarrhea: in a cohort of United States military personnel on deployment to the Multinational Force and Observers, Sinai, Egypt.
        Am J Trop Med Hyg. 2011; 84: 59-64
        • Paschke C.
        • Apelt N.
        • Fleischmann E.
        • et al.
        Controlled study on enteropathogens in travellers returning from the tropics with and without diarrhoea.
        Clin Microbiol Infect. 2011; 17: 1194-1200
        • Pandey P.
        • Bodhidatta L.
        • Lewis M.
        • et al.
        Travelers' diarrhea in Nepal: an update on the pathogens and antibiotic resistance.
        J Travel Med. 2011; 18: 102-108
        • Jiang Z.D.
        • Dupont H.L.
        • Brown E.L.
        • et al.
        Microbial etiology of travelers' diarrhea in Mexico, Guatemala, and India: importance of enterotoxigenic Bacteroides fragilis and Arcobacter species.
        J Clin Microbiol. 2010; 48: 1417-1419
        • Tribble D.R.
        • Sanders J.W.
        • Pang L.W.
        • et al.
        Traveler's diarrhea in Thailand: randomized, double-blind trial comparing single-dose and 3-day azithromycin-based regimens with a 3-day levofloxacin regimen.
        Clin Infect Dis. 2007; 44: 338-346
        • Hill D.R.
        • Beeching N.J.
        Travelers' diarrhea.
        Curr Opin Infect Dis. 2010; 23: 481-487
        • Paredes-Paredes M.
        • Okhuysen P.C.
        • Flores J.
        • et al.
        Seasonality of diarrheagenic Escherichia coli pathotypes in the US students acquiring diarrhea in Mexico.
        J Travel Med. 2011; 18: 121-125
        • Operario D.J.
        • Houpt E.
        Defining the causes of diarrhea: novel approaches.
        Curr Opin Infect Dis. 2011; 24: 464-471
        • Antikainen J.
        • Tarkka E.
        • Haukka K.
        • et al.
        New 16-plex PCR method for rapid detection of diarrheagenic Escherichia coli directly from stool samples.
        Eur J Clin Microbiol Infect Dis. 2009; 28: 899-908
      3. World Health Organization (WHO). Health Topics: Diarrhoea. Available from: http://www.who.int/topics/diarrhoea/en/. Accessed: October 28, 2012.

        • Freedman D.O.
        • Weld L.H.
        • Kozarsky P.E.
        • et al.
        • GeoSentinel Surveillance Network
        Spectrum of disease and relation to place of exposure among ill returned travelers.
        N Engl J Med. 2006; 54: 119-130
        • Meraz I.M.
        • Jiang Z.D.
        • Ericsson C.D.
        • et al.
        Enterotoxigenic Escherichia coli and diffusely adherent E. coli as likely causes of a proportion of pathogen-negative travelers' diarrhea–a PCR-based study.
        J Travel Med. 2008; 15: 412-418
        • Guion C.E.
        • Ochoa T.J.
        • Walker C.M.
        • et al.
        Detection of diarrheagenic Escherichia coli by use of melting-curve analysis and real-time multiplex PCR.
        J Clin Microbiol. 2008; 46: 1752-1757
        • Barletta F.
        • Ochoa T.J.
        • Ecker L.
        • et al.
        Validation of five-colony pool analysis using multiplex real-time PCR for detection of diarrheagenic Escherichia coli.
        J Clin Microbiol. 2009; 47: 1915-1917
        • Taniuchi M.
        • Verweij J.J.
        • Noor Z.
        • et al.
        High throughput multiplex PCR and probe-based detection with Luminex beads for seven intestinal parasites.
        Am J Trop Med Hyg. 2011; 84: 332-337
        • You Y.
        • Fu C.
        • Zeng X.
        • et al.
        A novel DNA microarray for rapid diagnosis of enteropathogenic bacteria in stool specimens of patients with diarrhea.
        J Microbiol Methods. 2008; 75: 566-571
        • Jin D.
        • Qi H.
        • Chen S.
        • et al.
        Simultaneous detection of six human diarrheal pathogens by using DNA microarray combined with tyramide signal amplification.
        Microbiol Methods. 2008; 75: 365-368
        • Taxt A.
        • Aasland R.
        • Sommerfelt H.
        • et al.
        Heat-stable enterotoxin of enterotoxigenic Escherichia coli as a vaccine target.
        Infect Immun. 2010; 78: 1824-1831
        • Jobling M.G.
        • Holmes R.K.
        Type II heat-labile enterotoxins from 50 diverse Escherichia coli isolates belong almost exclusively to the LT-IIc family and may be prophage encoded.
        PLoS One. 2012; 7: e29898
        • Siikamäki H.M.
        • Kivelä P.S.
        • Sipilä P.N.
        • et al.
        Fever in travelers returning from malaria-endemic areas: don't look for malaria only.
        J Travel Med. 2011; 18: 239-244
        • Hakanen A.
        • Jousimies-Somer H.
        • Siitonen A.
        • et al.
        Fluoroquinolone resistance in Campylobacter jejuni isolates in travelers returning to Finland: association of ciprofloxacin resistance to travel destination.
        Emerg Infect Dis. 2003; 9: 267-270
        • Vila J.
        • Ruiz J.
        • Gallardo F.
        • et al.
        Aeromonas spp. and traveler's diarrhea: clinical features and antimicrobial resistance.
        Emerg Infect Dis. 2003; 9: 552-555
        • Keskimäki M.
        • Mattila L.
        • Peltola H.
        • et al.
        Prevalence of diarrheagenic Escherichia coli in Finns with or without diarrhea during a round-the-world trip.
        J Clin Microbiol. 2000; 38: 4425-4429