Bacterial causes of AIDS-associated diarrhea in Thailand.

The incidence of bacterial diarrhea in AIDS patients has increased steadily and has led to enormous medical and public health problems. In this study, the clinical data together with 350 rectal swab samples each from AIDS patients with diarrhea (APD) and non-AIDS patients with diarrhea (NAPD), were collected and examined for bacterial enteropathogens at the Bamrasnaradura Infectious Diseases Hospital (BIDH), Nonthaburi, Thailand from May to December 1996. Patients were matched by age and sex. The majority of these patients were male (79%, 554/700), aged between 15 and 34 years (70.9%). The study found that the isolation rates of bacterial enteropathogens causing diarrhea in APD (18%, 62/350) were considerably lower than those in NAPD (43%, 152/350) (p<0.05). The infection rate with Salmonella group B (19.7%, 12/61) in APD was found to be significantly higher than that in NAPD (14.3%, 2/14) (p<0.05). Vibrio parahaemolyticus (53.3%, 81/152), Plesiomonas shigelloides (27%, 41/152), Aeromonas spp (19.1%, 29/152) and V. cholerae O1 (15.1%, 23/152), were more frequently detected in NAPD than in APD (p<0.05). Only nine Escherichia coli strains were isolated from APD, of which six were enteroinvasive E. coli, two enterotoxigenic E. coli and one enterohemorrhagic E. coli (non O157) possessing both vtl and vt2. No V. cholerae strains were detected in APD. The least effective antibiotics were ampicillin, tetracycline and cotrimoxazole. Antibiotic resistant patterns of the isolated organisms were similar from both groups. The results from this study might be useful in Thailand in the diagnosis and management of clinical cases of bacterial diarrhea, especially APD.

Rapid detection of polioviruses in environmental water samples by one-step duplex RT-PCR.

This study describes the rapid detection of polioviruses in environmental waters by a simple reverse transcriptase-polymerase chain reaction (RT-PCR) using two primer pairs for differentiation of poliovirus from non-polio enteroviruses in a single reaction by a one-step method, combining RT and PCR in a single tube. The detection by agarose gel electrophoresis yielded 2 bands of 153-bp and 293-bp for poliovirus tested without the need for further hybridization. The detection sensitivity of this one-step duplex RT-PCR, as measured with RNA extracted by heat treatment from supernatant of infected cell extracts, was 10(-1) 50% tissue culture effective doses (TCID50). This assay was used to evaluate the ability of sample concentration by membrane filter-based adsorption and elution, and purification by a simple RNA isolation based on guanidine isothiocyanate-phenol-chloroform extraction; the system yielded a detection limit of 5 x 10(-1) TCID50 seeded in 5 liters of tap water. This protocol was applied to the poliovirus detection in environmental water collected from 2 communities in Bangkok, Thailand during February and May 1998. Of 100 samples tested, 2 water samples collected from the same open sewage pipeline at one location were positive for polioviruses and one sample collected from another sewage pipeline was positive for non-polio enterovirus while a further 97 water samples were negative for both polioviruses and non-polio enteroviruses. With poliovirus detection by cell culture technique, none of the 100 samples tested was positive for poliovirus type 1, 2 or 3. RT-PCR was more sensitive, rapid, simple and cost-effective than the cell culture technique since the two water samples which were positive for polioviruses by RT-PCR failed to be detected by cell culture. Sequence data of 293-bp amplicons from positive samples were compared with those of reference poliovirus strains in the Genbank and the EMBL databases and identity to the sequence of type 1 strain Sabin was found to be 99%.

Detection of poliovirus, hepatitis A virus and rotavirus from sewage and water samples.

A modified adsorption-elution technique for concentration of enteric viruses from sewage and water samples was developed. The viruses in water were concentrated by negatively charged membrane filtration, eluted with 2.9% tryptose phosphate broth containing 6% glycine pH 9.0, and reconcentrated using centrifugation by a speedVac concentrator. The presence of poliovirus, hepatitis A virus (HAV) RNA, and rotavirus antigen was determined by cell culture isolation, nested polymerase chain reaction (nested PCR), and enzyme-linked immunosorbent assay (ELISA), respectively. A total of 100 sewage and water samples were collected from various sources in congested communities in Bangkok, concentrated and examined for those enteric viruses. Of 20 surface water samples from canals which located near sewage drains, 15% were positive for HAV RNA by nested PCR. Of 48 domestic sewage samples from man-holes of underground sewers, 8% were positive for rotavirus antigen by ELISA. Even though the samples were concentrated 256-2,000 fold, poliovirus was not found by isolation in cell culture.

Bacterial contamination of bottle milk in infants under 6 months in Children's Hospital, Bangkok, Thailand.

The bacterial contamination of bottle milk samples obtained randomly from 500 infants under 6 months of age who came to the Out-patient Department of Children's Hospital Bangkok was determined by collecting bottle milk samples prepared at home following interview of their caretakers after obtaining their consent. Bacterial contamination was found in 91.8% (459/500) of bottle milk samples. Among the positive samples, 82.8% (380/459) contained enteric bacteria, another 17.2% were unidentified bacteria. The dominant enteric bacteria isolated from bottle milk were Klebsiella spp (56.6%), Enterobacter spp (41.3%), Aeromonas spp (14.4%), E. coli (13.4 %) and Vibrio cholerae non 0-1 (1.8%). Isolated E. coli were further identified as enteropathogenic E. coli (7.8%, 4/51) and enterotoxigenic E. coli (3.9%, 2/51). About 74% of the contaminated bottle milk contained one type of bacteria, 23.7% had two types and 2.3 % had 3 or more types of bacteria. A level of bacterial contamination greater than the US government limited number (USGLN 2x10(4) CFU/ml) was found in 86.4% of total examined samples (432/500) [geometric mean (GM) of 2.9 x 10(6) CFU/ml]. About 66% (333/500) of bottle milk samples had coliforms greater than the USGLN (1 x l0(2) CFU/ml) with GM of 1.3 x 10(4) CFU/ml. Therefore, in the preparation of bottle milk, feeding practice should be emphasized in every setting of maternal-child health care and promotion of breast-feeding should be encouraged by the health personnel.

First isolation of Salmonella blockley in Thailand.

The first isolation of Salmonella blockley in Thailand was found in 2 strains of animal feed samples and 3 strains of chicken feather samples from a private poultry company in 1989. From 1987 to 1992, the number of S. blockley isolates increased and found in various sources. The major sources were the stools of diarrheal patients, mainly children. Another source of S. blockley was frozen chicken meat which increased every year studied. S. blockley isolated from human and other sources showed a high percentage resistance to streptomycin, tetracycline, kanamycin and chloramphenicol and a low percentage resistance to ampicillin and cotrimoxazole. Thus, S. blockley must now be listed as a possible cause of Salmonella food poisoning in Thailand.

Dot-ELISA for seroepidemiological study of exposure to Shigella flexneri.

A seroepidemiological study for determining serum antibodies to lipopolysaccharides (LPS) of Shigella flexneri using dot--ELISA was carried out in Krabi Province, Thailand, from January 1989 to December 1990. From 363 serum samples obtained from cord blood and from venous blood of the healthy persons aged from 6 months to over 50 years, 56% and 22%, respectively, were found to be positive for specific IgG and IgM antibodies to S. flexneri LPS. The IgG prevalence was initially detected at 3-4 years of age and then rose sharply with age. In contrast, IgM was detectable earlier, with much lower prevalence than that of IgG. The highest seroprevalence values were in the age groups 30-49 years for IgG and 15-19 years for IgM. The seroprevalence of S. flexneri infection was statistically higher among males, Buddhists, businessmen, and those with elementary education. S. flexneri infection was not associated with family income, home location, eating behaviour or water supply. These seroepidemiologic data demonstrated that most of the population in Krabi Province had been infected with S. flexneri.

ELISA for seroepidemiological study of exposure to Vibrio cholerae of population in Krabi Province, Thailand.

Seroepidemiological study of Vibrio cholerae exposure was carried out in Krabi Province during January 1989 to December 1990 using indirect ELISA to determine serum antibodies to lipopolysaccharides (LPS) of V. cholerae. Among 363 serum samples obtained from cord blood and venous blood of healthy persons, aged from 6 months to over 50 years, 65% and 64% were found positive for specific IgG and IgM against LPS of V. cholerae, respectively. The seroprevalence of V. cholerae infection increased with age from that found at 6 months, being highest in the age groups of 30-49 years for IgG and 15-29 years for IgM. The seroprevalence of V. cholerae infection was higher among female Muslims and home-makers, and increased with the family income. The seroprevalence of cholera infection was also influenced by home location, methods of food storage and water supply. These data suggested that a large number of Krabi's population had V. cholerae infection.