Prevalence of Escherichia coli and Salmonella spp. in street-vended food of open markets (tianguis) and general hygienic and trading practices in Mexico City.

Street-vendors in Mexico City provide ready-to-eat food to a high proportion of the inhabitants. Nevertheless, their microbiological status, general hygienic and trading practices are not well known. During spring and summer 2000, five tianguis (open markets) were visited and 48 vendors in 48 stalls interviewed. A total of 103 taco dressings were sampled for E. coli and Salmonella spp.: 44 (43%) contained E. coli and 5 (5%) Salmonella (2 S. Enteritidis phage type 8, 1 S. Agona, 2 S. B group). Both E. coli and salmonellas were isolated from three samples. Of Salmonella-positive stalls 80% (4/5) had three or more food-vendors and 80% of vendors were males, compared with 37.3% (16/43) and 46.4% (20/43) in the Salmonella-negative stalls respectively. Food-vendors kept water in buckets (reusing it all day), lacked toilet facilities, and prepared taco dressings the day before which remained at the tianguis without protection for 7.8 h on average. Consumption of street-vended food by local and tourist populations poses a health risk.

Asymptomatic salmonellosis among children in day-care centers in Mérida, Yucatan, Mexico.

BACKGROUND: Child day-care centers (DCC) have become common in many lower and middle income countries, presenting new problems that may differ from those of DCC in more developed countries. Diarrhea is a common problem in DCC in the United States, but information on the prevalence of diarrhea or specific enteropathogens among children in DCC in tropical and developing countries is limited. METHODS: Because of preliminary data from newborns and DCC attendees in Mérida, Mexico, with high rates of Salmonella infection, we conducted a 12-month longitudinal surveillance study of enteropathogens in two Mérida DCC. Seventy-eight children ages 2 months to 4 years were evaluated with demographic and clinical data, and stools were cultured monthly. RESULTS: Salmonella sp. was the most common enteropathogen detected (46 of 683 specimens, 6.7%), with higher rates in children younger than 18 months (P < 0.02), but it was found in only 1 of 10 diarrhea episodes that coincided with sampling. Other common organisms identified included Giardia lamblia (21 of 683, 3.0%) and LT-producing enterotoxigenic Escherichia coli (16 of 683, 2.3%). Salmonella was recovered from as many as 19% of children in a single month, but the large multiplicity of serotypes recovered suggested multiple sources rather than a common source outbreak. Children with Salmonella tended to have more liquid stools during the preceding 2 weeks. Salmonella was also isolated from the stool of teachers in 1 of the 2 DCC in 10 of 94 specimens (10.6%), and again multiple serotypes were represented. CONCLUSION: These data indicate the presence of multiple sources of Salmonella infection in the DCC, posing a complex situation for infection control.

Identification and strain differentiation of Vibrio cholerae by using polyclonal antibodies against outer membrane proteins.

Cholera is caused only by O1 and O139 Vibrio cholerae strains. For diagnosis, 3 working days are needed for bacterial isolation from human feces and for biochemical characterization. Here we describe the purification of bacterial outer membrane proteins (OMP) from V. cholerae O1 Ogawa, O1 Inaba, and O139 strains, as well as the production of specific antisera and their use for fecal Vibrio antigen detection. Anti-OMP antisera showed very high reactivity and specificity by enzyme-linked immunosorbent assay (ELISA) and dot-ELISA. An inmunodiagnostic assay for V. cholerae detection was developed; this assay avoids preenrichment and costly equipment and can be used for epidemiological surveillance and clinical diagnosis of cases, considering that prompt and specific identification of bacteria is mandatory in cholera.

[Salmonella serotypes identified in Mexican health services]. / Serotipos de Salmonella identificados en los servicios de salud de México.

OBJECTIVE: To identify the different Salmonella strain serotypes isolated at public and private laboratories in Mexico and at the Institute for Epidemiologic Diagnosis and Referral (InDRE). MATERIAL AND METHODS: A total of 24,394 Salmonella strains collected from 1972 to 1999 in public and private health laboratories of Mexico were analyzed with the Kauffmann-White method, using antisera produced by InDRE, according to Centers for Disease Control and Prevention (CDC, Atlanta, GA) standards; 15,843 (64.9%) samples were from human sources and 8,551 (35.1%) from non-human sources. RESULTS: One hundred ninety nine different serotypes were identified. The most frequent serotype in human beings was S. Typhimurium (20.4%), followed by S. Enteritidis (18.3%). In the past few years, the frequency of S. Enteritidis has been increasing, surpassing that of S. Typhimurium since 1991. Presently S. Enteritidis is the most frequently isolated serotype. In non-human sources, S. Derby (13.8%) and S. Anatum (8.5%) are the most frequent strains. CONCLUSIONS: Salmonella serotypes most frequently isolated in Mexico are: S. Typhimurium, S. Enteritidis, S. Derby, S. Agona y S. Anatum. From the epidemiologic standpoint, it is necessary to identify circulating and emerging Salmonella serotypes in order to target pertinent preventative interventions.

[Analysis of 1185 strains of Shigella isolated in Mexico from 1982 to 1993]. / Análisis de 1185 cepas de Shigella aisladas en México de 1982 a 1993.

During the period from 1982 to 1993, 1185 Shigella strains from the National Network of Diarrhoeal Laboratories were sent to the Enteric Bacteriology Laboratory of INDRE. These strains from patients of various ages with diarrhoeal illness were serologically confirmed. The frequency was as follows: S. flexneri (61.35%), S. sonnei (26%), S. dysenteriae (6.4%) and S. boydii (6.2%). S. dysenteriae 1 is an epidemiologicaly important species because it has caused diarrhoeal outbreaks on the southern border of Mexico that later spread through Central America. It must be considered that the 20 isolates obtained in 1989 were from an intentional search focused on S. dysenteriae. Authors pretend to continue with epidemiological surveillance focused on Shigella and intensify the intentional search in order to identify possible human or environmental S. dysenteriae 1 reservoires.

[Identification of Vibrio cholerae O1 by flow cytometry]. / Identificación de Vibrio cholerae O1 por citometría de flujo.

A total of 72 peptonated water samples suspected of carrying Vibrio cholerae were assessed by laser flow cytometry (LFC) and compared with positive culture. We used a direct fluorescence technique using polyclonal (PolAb) and monoclonal antibodies (MoAb) conjugated to fluorescein. The PolAb were able to detect 33 positive samples. A clear difference among the 20 positive samples was found with only three V. cholerae O1 false negatives when MoAb were used whereas all 13 V. cholerae Non O1 samples were detected. The correlation index comparing control autofluorescence with peptonated water samples show a R = 0.69, versus 0.96 with pure V. cholerae O1 strains. Our data suggest that the LFC technique is able to recognize V. cholerae O1 from a mixture of microorganisms with high sensitivity and specificity in a few hours.

[Principal serotypes of Salmonella identified in 10703 strains in Mexico from 1982 to 1993]. / Principales serotipos de Salmonella identificados en 10703 cepas en México entre 1982 y 1993.

From 1982 to 1993, 10703 Salmonella strains from The National Network of Diarroheal Laboratories of Mexico were sent to the Enteric Bacteriology Laboratory of INDRE. The strains were confirmed by serology and 119 different Salmonella serotypes were found. The most frequent serotypes were as follows: S. typhimurium, S, enteritidis, S. agona & S. typhi. The strains were classified according to the source of isolation as follows: 6671 strains (62.33%) from clinical samples, mainly of faecal origin; 2903 (27.1%) from food for human consumption; 425 from food for animal consumption, 665 (6.21%) from environment or fomites and 39 (0.36%) from animals. The most frequent serotype in clinical samples was S. typhimurium among 96 different serotypes. The main serotype from blood cultures was S. typhi although 27 other serotypes were found. Of thirteen serotypes related to diarrhoeal outbreaks the higher frequency of S. typhimurium was observed but S. typhi caused more outbreaks. A frequency of 119/> 2000 serotypes was observed, that means less than 5% of Salmonella known serotypes. A yearly variability on serotype predominance was observed as well as changes on source of isolation. This results suggest that epidemiological surveillance of salmonellosis should be continued and improved, looking for cases, asymptomatic carriers and contaminated food for human consumption.

[Phenotypic and genotypic characterization of Vibrio cholerae O1]. / Caracterizatión fenotípica y genotípica de Vibrio cholerae O1.

We made 52180 tests for isolation and identification of toxigenic V. cholerae O1 from rectal swabs and reference strains. We isolated 17.6% V. cholerae O1 strains in 1991, 43.5% in 1992 and 38.9% in 1993. The main serovar in 1991 was Inaba, whereas in 1993 a similar percentage was serovar Ogawa. The phenotype of V. cholerae strains was determined by hemolysis test, Voges-Proskauer test, polymyxin B resistance and phages 4 and 5 resistance. All of the mexican strains were El Tor. There were 2.9-0.75% hemolytic strains from 1991 to 1993, but they were negative when the test was made in tube with human erythrocytes. The resistotypes were performed in 24526 selected strains by Kirby-Bauer method and MIC tests. All of the strains were sensitive, except more than 100 strains isolated in Veracruz that were resistant to tetracycline and doxycycline. Detection of cholera toxin was made by ELISA and on culture of Vero and CHO cells. All the V. cholerae O1 strains were toxigenic. The genotype was determined by PCR and ribotyping. The PCR amplified one 564 pb fragment on V. cholerae O1. The ribotypes of mexican strains were 5 and 6a.

[Evaluation of the ELISA method for cholera toxin determination in Vibrio cholerae cultures]. / Evaluación del método de ELISA para determinación de toxina colérica en cultivos de Vibrio cholerae.

ELISA test was evaluated in 503 cultures of Vibrio cholerae O1 y 303 Non-O1. The cultures were isolated from sewage from different states of México between june 1991 and october 1992. The sensitivity was 100% and specificity was 96%. Only 12 strains of V. cholerae Non-O1 were positive for CT toxin. When these cultures were confirmed by polymerase chain reaction (PCR) for cholera toxin, the results were negative. ELISA test is a good alternative to be used for toxin production in cultures of V. cholerae, it needs confirmation only with O1 negative and Non-O1 positive reactions.

[Cytotoxic effect of Vibrio cholerae non-O1 on Vero cells]. / Efecto citotóxico de Vibrio cholerae no-O1 en células Vero.

At the present time there is still in Mexico a diarrhoeal outbreak due to Vibrio cholerae O1. In INDRE we have isolated from the same outbreak last year (jan-apr), 70 strains of Vibrio cholerae Non-O1. These were isolated from patients with a diarrhoeal illness different from cholera. Patients were of different ages and sex, and from various geographic areas. The isolated strains were confirmed by serological agglutination test with polyclonal antisera, and they neither belong to O1 serogroup or O139. We assayed all the 70 strains in Vero cells, searching for cytotoxic effect, probably attributed to cholera toxin, or any other toxin. The strains were screened by PCR for cholera toxin gene detection, and negative results were obtained. We have found only one CT-producer strain, but it was a rough one so, we are not able to affirm that is not a V. cholerae O1 serotype. Vibrio cholerae Non-O1 strains, tested in Vero cells assay, produced cytotoxic effect within 24 h. It was found that 48/70 strains (66.6%), had cytotoxic activity, showing rounding and then lysis of cells. From our results we concluded that this cytotoxic effect, is not cholera toxin related, instead we propose it could be due to an unknown virulence factor, probably a different toxin in mexican Vibrio cholerae Non-O1 strains.