DIAG. MICROBIOL.INFECT. DIS. 1990;13:273-276

273

Microscopic Examination of Stools and a Latex Slide Agglutination Test for the Rapid Identification of Bacterial Enteric Infections in Khmer Children Eva Nordlander, Surasith Phuphaisan, Ladaporn Bodhidatta, James Arthur, and Peter Echeverria

Eighty-five stools collected from 50 children with diarrhea at an evacuation site on the Thai-Kampuchean border were (1) examined microscopically for fecal leukocytes, (2) tested after 24 hr enrichment in brain~heart infusion broth by a latex slide agglutination test for detection of Salmonella and Shigella, and (3) examined with microbiological techniques to identify bacterial, viral, and parasitic pathogens. If the 65 specimens in which one or no pathogens are considered, 6 or more fecal leukocytes/hpf were found on microscopic examination of stools in both children infected with Shigella spp., the one child infected with Salmonella spp., and three of eight children infected with Campylobacter spp. Less than or equal to 5 leukocytes/hpf were found in 70% (7/10) of children infected with rotavirus, 100% (2/2) infected with Cryptosporidium, 100%

(2/2) infected with Giardia, 89% (8/9) infected with enterotoxigenic Escherichia coli, and 77% (24/31) with diarrhea in whom no etiologic agent was identified. The Salmonella slide latex test had a sensitivity of 50%, a specificity of 92%, and a positive predictive value of 12%. The Shigella slide latex test had a sensitivity of 0%, a specificity of 95%, and a positive predictive value of 0%. Forty-five percent of the latex slide agglutination tests from enrichment cultures were nonspecific. Microscopic examination of diarrheal stools for fecal leukocytes, though nonspecific, appears to be the best way to differentiate Shigella spp. from rotavirus and parasitic infections. Examining stools for fecal leukocytes was less helpful in differentiating Shigella from other bacterial infections.

Salmonella spp. and Shigella spp. are frequently isolated from children with diarrhea in both developing (Stoll et al., 1982; Taylor et al., 1988; Echeverria et al., 1989) and developed countries, especially in day care centers (Pickering et al., 1981). A rapid m e t h o d of diagnosing these infections would be of value in designing strategies for treating and preventing fur-

ther infections, especially in closed populations. Shigella spp., usually S. flexneri (group B) in developing countries (Stoll et al., 1982; Taylor et al., 1983; Echeverria et al., 1989) and S. sonnei (group D) in developed countries (Rosenberg et al., 1976), are wellrecognized causes of pediatric gastroenteritis. Nontyphoidal Salmonella spp. also appears to be a cause of diarrhea, primarily in very y o u n g children (Blaser et al., 1982; Echeverria et al., 1989). Standard bacteriological m e t h o d s take 48-72 hr to perform after receipt of the specimen in the laboratory. Alternative approaches to identify bacterial infections include nonspecific tests, such as examining stools for blood (Stoll et al., 1982; Radetsky., 1986) and examining stools microscopically for fecal leukocytes (Pickering et al., 1977), and specific tests, such as examining specimens with DNA probes

From the Catholic Relief Services, Surin (E.N., S.P.); and the Department of Bacteriology,Immunology, and Molecular

Genetics, Armed Forces Research Institute of Medical Sciences (L.B., J.A., P.E.), Bangkok, Thailand Address reprint requests to: Dr. Peter Echeverria, AFRIMS, Rajvithi Road, Bangkok 10400, Thailand, or A.P.O. San Francisco, CA 96346-5000. Received November 27, 1989; revised and accepted December 19, 1989. © 1990 Elsevier Science Publishing Co., Inc. 655 Avenue of the Americas, New York, NY 10010 0732-8893/90/$3.50

E. Nordlander et al.

274

(Echeverria et al., 1985; Seriwatana et al., 1988). Because standard bacteriological methods are expensive and among the most difficult to perform accurately, we evaluated a commercial latex agglutination slide test to detect Salmonella spp. and Shigella spp. infections in Khmer children with diarrhea at an evacuation site in Thailand. Stool specimens were collected from Khmer children less than 2 years of age with diarrhea seen for treatment at a field hospital at site B, an evacuation site for Khmers on the Thai-Kampuchean border. Electrical power at this hospital is provided by generators that only operate several hours each day, so it is not possible to establish a diagnostic bacteriology laboratory. Approximately 66,000 people, 25% of w h o m are less than 5 years of age, are restricted to a 3-km 2 area at site B. Water is contaminated with fecal coliforms, people live under crowded insanitary conditions, and diarrheal disease is consequently endemic. Stool specimens collected from Khmer children with diarrhea (defined as three or more loose stools in the previous 24 hrs) were examined microscopically for fecal leukocytes at the field hospital (Pickering et al., 1977). Stools, transported in Cary Blair media at 4°C, were examined for enteric pathogens, as described previously (Taylor et al., 1988), in a laboratory that had been established in the town of Surin, approximately 65 kilometers away. Stools were cultured on selective media and inoculated into brain/heart infusion (BHI) broth (Difco, Detroit, MI). After incubation at 37°C for 24 hr, 50 ~1 were tested in the slide agglutination latex test for Salmonella spp. and Shigella spp. (Bactigen Salmonella-Shigella latex agglutination slide test, Wampole Laboratories, Div. Carter-Wallace Inc. Cranbury, NJ) according to the manufacturer's instructions. Briefly, this kit consists of reagent latex 1, latex particles coated with goat antibodies shown reactive with more than 80 Salmonella spp. serotypes (according to the manufacturer, additional Salmonella spp. serotypes are detected due to the common antigenic determinants); reagent latex 2, latex particles coated with rabbit antibodies against Shigella flexneri (serotypes 1 to 6), and S. sonnei (forms I and II); and reagent latex 3, latex particles coated with rabbit antisera to Shigella dysenteriae (serotypes 1-7), and S. boydii (serotypes 1-7). One drop (50 ul) of each of the three latex reagents was mixed with 50 ul of 24-hr BHI culture of a stool on a slide provided with the kit. The slide was rotated for 10 min by hand and then observed for the presence of agglutination. A reaction that was grainy or _ was read as negative. Positive controls for each latex reagent, provided by the manufacturer, were included in each assay. A specimen that reacted with more than one latex reagent was considered to exhibit a nonspecific reaction.

TABLE 1.

Microscopic Stool Examination and Results from 65 Specimens in Which One or No Infecting Pathogens Was Identified Fecal Leukocytes (WBC/hpf) a

Pathogen (no. cases) ETEC (9)b

Campylobacter spp. (8) Shingella spp. (2) Salmonella spp. (1) Cryptosporiclium (2) Giardia (2) Rotavirus (10) No pathogen identified (31)

0-5

6-25

>25

8 5 0 0

0 3 1 1

1 0 1 0

2 2

0 0

0 0

7

2

1

24

5

2

aWBC, white blood cell. bETEC, enterotoxigenic Escherichiacoil

Rotavirus was identified in 27%, enterotoxigenic

E. coli (ETEC) in 23%, Campylobacter spp. in 16%, Giardia in 8%, Cryptospodirium in 7%, Shigella spp. in 6%, Klebsiella oxytoca in 4%, Aeromonas spp. in 3%, Plesiomonas shigelloides in 2%, Strongyloides stercoralis in 2%, Salmonella spp. in 2%, and Edwardsiella tarda in 1% of 85 episodes of diarrhea in 50 children with diarrhea (June and July 1989). Multiple pathogens were identified in 20% of episodes. More than five fecal leukocytes were found on microscopic examination of stools of both children infected with Shigella only, one child infected with Salmonella only, and three of eight children infected with Campylobacter only. Less than five leukocytes were found in 70% of children infected with rotavirus, 100% of children infected with Giardia, 89% of children infected with ETEC, and 77% of children with diarrhea in w h o m no etiologic agents were identified (Table 1). The Salmonella latex slide agglutination test had a sensitivity of 50%, a specificity of 92%, a positive predictive value of 12%, and a negative predictive value of 99%. The Shigella latex agglutination test had a sensitivity of 0%, a specificity of 95%, a positive predictive value of 0%, and a negative predictive value of 94% (Table 2). Forty-five percent (38/85) of specimens tested with the latex agglutination test were nonspecific (agglutinated in two or more latex reagents). Identification of Salmonella and Shigella infections is usually performed by culturing fecal specimens on selective media before and after incubation in selective enrichment broth, identifying isolates by their characteristic morphology on agar media, test-

Notes

275

TABLE 2.

Results of Rapid Latex Test on 85 Stool Specimens from 50 Khmer Children Latex Test Positive Reagent

Pathogen Identified (no. cases)

Ia

2b

3c

ETEC (20)b Campylobacter spp. (14) Shingella spp. (5) Salmonella spp. (2)

3a 1a 0 1

2e 0 0 0

0 0 0 0

Cryptosporidium (6)

0

Ie

0

Giardia (7)

1

0

1

Rotavirus (23)

4a

1e

0

No pathogen identified (31)

2

1

0

aSalmonella spp. bShingellaflexneri and S. sonnei. cShigelladysenteriaeand S. boydii. aThe Salmonellalatex test was positivein two childreninfected with ETEC and rotavirus and one child infected with ETEC C. jejuni and rotavirus. eTheS. flexneriand S. sonneilatextest was positivein one child infected with ETEC, Crytosporidium, and rotavirus. ing for various biochemical reactions and, finally, testing isolates for agglutination in specific antisera (Kelly et al., 1985). Immunological assays to detect enteric pathogens that have been developed include latex agglutination tests or enzyme-linked immunosorbent assays (ELISAs) to identify bacteria: S. typhi (Robinson et al., 1983; Metzler and Nachamkin, 1988); nontyphoid Salmonella (Robinson et al., 1983; Metzler and Nachamkin, 1988); Shigella (Metzler and Nachamkin, 1988); and Vibrio cholerae (Jesudason et al., 1984); viruses; rotavirus (Sanekata et al., 1981); enteric adenovirus (Johansson et al., 1985); and intestinal parasites, Entamoeba histolytica (Root et al., 1978), and G. lamblia (Unger et al., 1984). The observed isolation rate of nontyphoidal Salmonella from Khmer children (2%) was lower than from Thai children (12%) in Bangkok (Echeverria et al., 1989). The isolation rate of Shigella (6%) was similar to the isolation rate in Thai children of the same age (Taylor et al., 1988; Echeverria et al., 1989). There are a number of explanations w h y the latex agglutination test was insensitive and nonspecific in identifying Salmonella spp. and Shigella spp. infections. First, malnourished Khmer children, 35 of w h o m had recent episodes of diarrhea, may have been infected with fewer Salmonella spp. and Shigella spp.

organisms than nonimmune children cultured after their first episode of Salmonella spp. or Shigella spp. infection. Second, these children may have had interfering coproantibodies to Salmonella spp. and Shigella spp. that react with the slide agglutination test. Finally, Aeromonas spp., Plesiomonas spp., and Klebsiella oxytoca, which were frequently isolated from stools of Khmer children, may have caused the nonspecific readings of almost half the stools tested in the latex agglutination test. In the United States, the latex agglutination slide test had a sensitivity of 86% and a specificity of 96% in detecting Salmonella isolates, and two of the four Shigella isolates were identified correctly. Only 1% of cultures of these cited specimens tested by the latex slide agglutination were nonspecific (Metzler and Nachamkin, 1988). Identifying enteric pathogens directly in stools would save considerably in the cost of reagents and technician time. Furthermore, rapid tests would be very useful in this studied situation where it is very difficult or impossible to establish a clinical microbiology laboratory. Latex particles coated with polyclonal antibodies appear, however, to be nonspecific and insensitive when used with enrichment cultures of stools in a tropical developing country. This contrasts to an ELISA test using monoclonal antibodies to detect S. typhi antigens in urine samples that had a sensitivity of 65% and a specificity of 100% in identifying patients with typhoid fever (Chaicumpa et al., 1988) and was a marked improvement over ELISAs that used polyclonal antibodies (Rockhill et al., 1980; Taylor et al., 1983). One polyclonal ELISA was positive in 34% of typhoid patients and 46% of febrile controls (Taylor et al., 1983). Latex particles coated with specific monoclonal antibodies need to be developed and evaluated as rapid diagnostic tests in Salmonella spp. and Shigella spp. infections in this type of environment.

We thank Orapan Chivaratonond, Thamma Sakulkaipeara, Vitaya Khungvalert, Chittima Pitarangsi, Prani Ratarasarn, Apichai Srijan, and Suchitra Changchawalit for help in performing clinical bacteriology both in Surin and in Bangkok; and Sajee Pinnoi, Ovath Thonglee, Nisara Wongkamhaeng, and Ubonrat Tareetaifor help in collecting specimens. We also thank Sok Vantha and other Khmer laboratory technicians for allowing us to use their laboratory, the Catholic Relief Service physicians and Khmer medics for allowing us to collect specimens from their patients, and Dr. Sakti Paul, the UNBRO Medical Coordinator for his help in obtaining passes to work at site B.

276

E. N o r d l a n d e r et al.

REFERENCES Blaser MJ, Newman LS (1982) A review of human salmonellosis: 1 infective dose. Rev Infect Dis 4:1096-1106 Chaicumpa W, Thin-inta W, KuSmith S, Tapchaisri P, Echeverria P, Kalambaheti T, Chongsa-nguan M (1988) Detection of Salmonella typhi antigen 9 in patients' specimens with monoclonal antibody. J Clin Microbio126:18241830. Echeverria P, Seriwatana J, Taylor DN, Tirapat C, Chaicumpa W, Rowe B (1985) Identification by DNA hybridization of enterotoxigenic Escherichia coli in a longitudinal study of villages in Thailand. J Infect Dis 151: 124-130. Echeverria P, Taylor DN, Lexomboon U, Bhaibulaya M, Blacklow NR, Tamura K, Sakazaki R (1989) Case-control study of endemic diarrheal disease in Thai children under 5 years of age. J Infect Dis 159:543-548. Jesudason MV, Tangavela CP, Lalitha MK (1984) Rapid screening of fecal specimens for Vibrio cholerae by a coagglutination technique. J CLin Microbiol 19:712-713. Johansson ME, Uhnoo I, Svensson L, Pettersson CA, Werdell G (1985) Enzyme-linked immunosorbent assay for detection of enteric adenovirus 41. J Med Virol 17:1927. Kelly MT, Brenner DJ, Farmer JJ (1985) Enterobacteriacae Manual of Clinical Microbiology, 4th ed. Ed. E.H. Lennette. Washington, DC: American Society for Microbiology, pp 264-265. Metzler J, Nachamkin I (1988) Evaluation of a latex agglutination test for the detection of Salmonella and Shigella spp. by using broth enrichment. J Clin Microbio126:25012504. Pickering LK, DuPont HL, Olarte J, Conklin R, Ericson C (1977) Fecal leukocytes in enteric infections. Am J Clin Pathol 68:562-565. Pickering LK, Evans DG, DuPont HL (1981) Diarrhea caused by Shigella, rotavirus, and Giardia in day care centers: prospective study. J Pediatr 99:(1):51-56. Radetsky M (1986) Laboratory evaluation of acute diarrhea. Pediatr Infect Dis 5:230-238. Robinson BJ, Pretzman CI, Manningly JA (1983) Enzyme immunoassay in which a myeloma protein is used for

detection of salmonellae. Appl Environ Microbiol 45:18161821. Rockhill RC, Rumans LW, Lesmana M, Dennis DT (1980) Detection of Salmonella D Vi and d antigens by slide agglutination in urine from patients with typhoid fever. J Clin Microbiol 11:213-216. Root DM, Cole FX, Williamson JA (1978) The development and standardization of an ELISA method for detection of Entamoeba histolytica antigens in fecal specimens. Arch Invest Med Mex 95:203-210. Rosenberg ML, Weissman JB, Gangarosa EJ (1976) Shigellosis in the United States: ten-year review of nationwide surveillance, 1964-1973. Am J Epidemio1104:543550. Sanekata T, Yoshida Y, Okada M (1981) Detection of rotavirus in feces by latex agglutination. J Immunol Methods 41:177-180. Seriwatana J, Brown JE, Echeverria P, Taylor DN, Suthienkul O, Newland J (1988) DNA probes to identify Shiga-like toxin I and II producing enteric bacterial pathogens isolated from patients with diarrhea in Thailand. J Clin Microbiol 26:1614-1615. Sethabutr O, Hanchalay S, Echeverria P, Taylor DN, Leksornboon U (1985) A nonradioactive probe to identify Shigella and enteroinvasive Escherichia coli in stools of children with diarrhea. Lancet 2:1095-1097. Stoll BJ, Glass RI, Huq MI, Khan MU, Banu H, Holt J (1982) Epidemiologic and clinical features of patients infected with Shigella who attended a diarrheal disease hospital in Bangladesh. J Infect Dis 146:177-183. Taylor DN, Harris JR, Barrett JJ, Hargrett NT, Prentzel I, Valdivieso C, Palomino C, Levine MM, Blake P (1983) Detection of using Vi antigen as a diagnostic test for typhoid fever. J Clin Microbiol 18:872-876. Taylor DN, Echeverria P, Sethabutr O, Pitarangsi C, Lexomboon U, Blacklow NR, Rowe B, Gross R, Cross J (1988) Clinical and microbiologic features of Shigella and enteroinvasive Escherichia coli infections detected by DNA hybridization. J Clin Microbiol 26:1362-1366. Ungar BL, Yolken PRN, Nash TE, Quinn TC (1984) Enzyme-linked immunosorbent assay for detection of Giardia lamblia in fecal specimens. J Infect Dis 149:9097.

Microscopic examination of stools and a latex slide agglutination test for the rapid identification of bacterial enteric infections in Khmer children.

Eighty-five stools collected from 50 children with diarrhea at an evacuation site on the Thai-Kampuchean border were (1) examined microscopically for ...
381KB Sizes 0 Downloads 0 Views