Antimicrobial susceptibility of Helicobacter pylori isolated from gastric biopsies in dyspeptic patients.

The objective of this study was to evaluate the prevalence of antimicrobial resistance in Helicobacter pylori isolated from the antrum and corpus of dyspeptic patients in Khon Kaen, Thailand, and to compare the antimicrobial susceptibility patterns of H. pylori isolated from the antrum and corpus in individual patients. Antimicrobial susceptibility was determined by disk diffusion, studying susceptibility to metronidazole, clarithromycin, amoxicillin, erythromycin, ciprofloxacin, and tetracycline. The H. pylori resistant rate to at least one of the six antimicrobial agents tested was 37%. The resistance rates were 30.2% for metronidazole, 9.2% for ciprofloxacin, 5% for clarithromycin, 2.4% for amoxicillin, and 1.7% for erythromycin and tetracycline. Single, double, and more than double antimicrobial resistances were found in 27.7, 6.7 and 2.5%, respectively. Antimicrobial susceptibility testing revealed 11 antibiotypes. The most common antimicrobial susceptibility pattern found was sensitivity to 6 antimicrobial agents (63%). H. pylori antimicrobial resistance in specimens isolated from the antrum and corpus were nearly equivalent, 37.3% (22/59) and 36.7% (22/60), respectively. Most of the H. pylori specimens isolated from the antrum and corpus in individual patients were identical (87.7%).

Association of ompU gene in Vibrio cholerae from patients and environment with bile resistance.

The objective of this study was to determine whether Vibrio cholerae, possessing ompU isolated from patients and the environment, conferred bile resistance and whether other virulence genes were also related to bile resistance. Fifty-two V cholerae O1 and non-O1 isolates were examined by PCR for the presence of the virulence-associated and regulatory genes, ctxA, tcpA, zot, ace, ompU, toxR, hlyA and stn/sto. V. cholerae possessing ompU resistant to equal or greater than 10% sodium deoxycholate were found in 93% of isolates but only in 9% of V. cholerae isolates not possessing ompU. The effects of other virulence genes on bile resistance could not be ascertained in this study. Thus V cholerae non-O1 with ompU and possibly other virulence genes isolated from the environment have the potential of affecting public health.

Comparison of media and antibiotic supplements for isolation of Helicobacter pylori from gastric biopsies.

Our objective was to improve the media and the antibiotic supplements in order to increase the detection rate of Helicobacter pylori from gastric biopsy specimens. For the primary isolation of H. pylori taken from gastric biopsies, we compared the efficacy of two media: Columbia blood agar (CBA, Difco); brain heart infusion agar (BHIA, Difco); and two antibiotic supplement sets--a commercial antibiotic supplement (SR147, Oxoid) and an in-house antibiotic supplement (IHS). Gastric biopsies obtained from 210 patients were diagnosed by culture, rapid urease test (RUT) and histology. The true positive criteria were defined as a culture or both urease and histology tests being positive. The H. pylori infection rate was 44.3% (93/ 210). To compare the two media, a total of 106 gastric biopsies were plated on CBA or BHIA with 7% human blood, containing the antibiotic supplement SR147 and incubated under microaerophilic conditions. Of the 106 samples, 48 (45.3%) case of H. pylori infection, compared to the true positive criteria. The isolation rate using a combination of the two media was 83% (40/48). Of the 40 samples, 36 (90%) and 35 (87.5%) were positive on CBA and BHIA, respectively. To compare the two antibiotic supplement sets, a total of 104 gastric biopsies were plated on CBA, containing the commercial antibiotic supplement SR147 (5 mg/l trimethoprim, 10 mg/l vancomycin, 5 mg/l amphotericin B and 5 mg/l cefsulodin) or containing IHS (5 mg/l trimethoprim, 10 mg/l vancomycin, 2 mg/l amphotericin B and 2,500 U/l polymyxin B). Of the 104 samples, 45 (43.2%) case of H. pylori infection were found compared to the true positive criteria. The isolation rate using a combination of the two selective supplement sets was 82% (37/45). Of the 37 samples, 35 (95%) and 34 (92%) were positive with SR147 and IHS, respectively. Our study indicates that the combination of the two media and two antibiotic supplements is useful for maximum recovery of H. pylori isolated from gastric biopsies. CBA, and the commercial antibiotic supplement SR147 provided higher detection rates for H. pylori than BHIA, and IHS but the differences were not statistically significant.

Enzyme-linked immunosorbent assay for serodiagnosis of Helicobacter pylori in dyspeptic patients and volunteer blood donors.

Helicobacter pylori, an important etiological agent in the development of gastritis, peptic ulcer and gastric carcinoma, can be detected by the enzyme-linked immunosorbent assay (ELISA). Our objectives were: (1) to evaluate the efficacy of a commercial ELISA kit (Pyloriset EIA-G III) in sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy for diagnosis of H. pylori infection in Thai dyspeptic patients in Khon Kaen Thailand; and (2) to examine the seroprevalence of H. pylori among blood donors at Srinagarind Hospital's Blood Bank, Khon Kaen University, by the commercial ELISA. Gastric biopsies obtained from 137 dyspeptic patients were diagnosed by culture, rapid urease test (RUT) and histology. Serum samples from the same dyspeptic patients and 100 healthy blood donors were assayed using the commercial ELISA. H. pylori infection in dyspeptic patients was considered positive when the culture or both RUT and histology were positive. Using a cut-off value at a titer of 20 U/ml (as recommended by the manufacturer), we found the commercial ELISA kit had a sensitivity of 93.3%, specificity of 75.3%, PPV of 74.7%, NPV of 93.5% and accuracy of 83.2%. The overall H. pylori seroprevalence in the healthy blood donors was 57%. Of the 100 healthy blood donors, 39 (60.9%) of the males and 18 (50.0%) of the females were seropositive.

Enteropathogenic bacteria and enterotoxin-producing Staphylococcus aureus isolated from ready-to-eat foods in Khon Kaen, Thailand.

The objective of this study was to investigate the microbiological quality of ready-to-eat food in the Municipality of Khon Kaen, Thailand. Four categories of 186 food samples were collected: (1) high heat food; (2) low heat food; (3) no heat food; and, 4) on-site prepared fruit juices and beverages. Of the food samples, 145 (78%) failed to meet acceptable microbiological standards, including fruit juice and beverages (100%), no heat food (91.7%), low heat food (81.7%) and high heat food (57.9%). The most frequent bacterial indexes indicating unacceptability were the most probable number (MPN) of coliforms (78%), the bacterial colony count (58%), and the MPN of E. coli (46%). Pathogenic bacteria were found in 6.5% of food samples. Salmonella, Vibrio cholerae non O1 and Aeromonas hydrophila were found in 4.3, 1.6 and 0.5% of the total food samples, respectively. The serovars of Salmonella found in food were S. Derby, S. Give, S. Krefield, S. Paratyphi B, S. Verchow, S. Lexington and S. Senftenberg. Staphylococcus aureus concentrations of >10(2) CFU/g and >10(5) CFU/g were found in 10.8% and 1.1% of the food samples. Enterotoxin types AB and A of S. aureus were found in 2.7% of the food samples. These results indicate that more than half of the ready-to-eat foods tested in Khon Kaen municipality did not meet microbiological national standards and many kinds of enteropathogenic bacteria were found, suggesting food stalls may be a source of foodborne disease.

Sensitivity and specificity of an in-house rapid urease test for detecting Helicobacter pylori infection on gastric biopsy.

We developed an in-house rapid urease test (iRUT) and evaluated the efficacy and the agreement of the iRUT and the cRUT compared with culture and histology for the detection of H. pylori infection. Five iRUT media were tested with H. pylori isolates and other bacteria. The most suitable iRUT medium was further evaluated for detection of H. pylori infection. Gastric biopsies from 120 patients were diagnosed by culture, iRUT, cRUT and histology. The results of the iRUT and cRUT were read at 30 minutes, 1 hour and up to 24 hours. A true positive result was either the culture or both the RUT (cRUT or iRUT) and the histological examination being positive. The sensitivity and specificity of the iRUT result at 30 minutes, 1 hour and up to 24 hours were 77.1% and 100%, 77.6% and 100%, and 94.1% and 94.2%, respectively. Values for the same parameters of cRUT were 87.5% and 100%, 89.8% and 100%, and 100% and 94.2%, respectively. The agreement between the iRUT and cRUT was very good (kappa values > or = 0.82). Our results indicate that the iRUT is a-sensitive, specific and cost effective test. It can be appropriately applied for detecting H. pylori infection in gastric biopsy specimens.

Epidemiology of antimicrobial resistance in Salmonella isolated from pork, chicken meat and humans in Thailand.

Forty samples each of pork and chicken meat were collected from local retail markets in Khon Kaen, northeast Thailand, for Salmonella isolation and identification during 2003. Fifty-four isolates of Salmonella obtained from diarrheal patients admitted at a hospital located in the same town were serotyped. All isolates were also tested for antimicrobial sensitivity against amoxicillin (Amx), chloramphenicol (Chl), norfloxacin (Nor), ciprofloxacin (Cip), gentamicin (Gm), sulfamethoxazole and Trimethoprim (Sxt), Tetracycline (Tet), Streptomycin (Str) and sulfamethoxazole (Sulfa). The results showed that 26 (65%) pork samples and 30 (75%) chicken meat samples were contaminated with Salmonella. The most prevalent serovar in pork was S. Rissen (61.5%), followed by S. Stanley and S. Lexington (11.5%). In chicken meat, the most prevalent serovar was S. Anatum (33.3%), followed by S. Rissen (16.7%). Among isolates from human patients, S. Rissen (20.4%) and S. Stanley (18.5%) were the most frequently identified serovars. All the isolates were resistant to Str and Sulfa. None were resistant to Nor and Cip. Resistance to Amx, Chl, Gm, Sxt, and Tet in pork was 15.4, 15.4, 3.9, 15.4 and 88.5%, respectively. The resistance to those antimicrobial agents in chicken meat was 30.0, 26.7, 6.7, 20 and 100%, respectively, and in human patients was 27.8, 20.4, 5.6, 31.5 and 92.6%, respectively. Statistical analysis found no difference in the rate of resistance to Amx, Chl, Gm, and Sxt among the different sources of the Salmonella isolates (p > 0.05 for each antimicrobial agent). Our results indicate that antimicrobial resistant Salmonella strains were widely spread among pork and chicken meat, and in humans.

Analysis of three phenotyping methods and pulsed-field gel electrophoresis for differentiation of methicillin resistant Staphylococcus aureus isolated from two hospitals in Thailand.

Methicillin resistant Staphylococcus aureus (MRSA) is an important hospital and community-acquired pathogen. Rapid and reliable epidemiologic typing is necessary for controlling the spread of MRSA outbreak. The objective of this study was to compare the phenotyping with the genotyping method to differentiate MRSA isolates obtained from the two hospitals in Thailand (central and northeastern). Seventy-four MRSA isolates were randomly collected and confirmed by the presence of mecA gene. Antibiogram, phage typing and enterotoxin production were used for the phenotyping analysis. Pulsed-field gel electrophoresis (PFGE) with Smal digestion of chromosomal DNA was used for the genotyping analysis. We found 17 distinct profiles by the 3 phenotypic typing methods and 18 PFGE types designated as 5 major types (A-E) and 13 subtypes. The most frequent PFGE types and their related subtypes found in both hospitals were A and C, comprising 54 and 27%, respectively. The antibiogram could differentiate 6 different types. All isolates were resistant to the majority of antimicrobial agents tested, but were susceptible to vancomycin and fosfomycin. Ten (13.5%) MRSA isolates produced enterotoxin A. Nontypable phage and phage type 77 were found predominantly in MRSA isolated from the northeast and central hospital, respectively. A significant correlation was found between the phenotyping and the genotyping methods and there was a good correlation between antibiogram and PFGE. Antibiogram typing alone can be used as a useful epidemiological marker for practical purposes. PFGE types A and C were the common endemic MRSA clones in both hospitals in Thailand.

Characterization of diarrheagenic Escherichia coli isolated from food in Khon Kaen, Thailand.

Four categories of 186 ready-to-eat food samples in Khon Kaen municipality, Thailand, were collected and investigated for fecal contamination by enumeration of Escherichia coli using the most probable number (MPN) method. Then, the E. coli isolates were presumptively identified as diarrheagenic E. coil by agglutinating with polyvalent O-antisera and monovalent O-antisera commonly found in diarrheagenic strains and were subsequently investigated for the presence of the recognized virulence genes for enteroaggregative (EAEC), enteroinvasive (EIEC), enteropathogenic (EPEC), enterotoxigenic (ETEC), and shiga toxin-producing E. coli (STEC or EHEC) by multiplex PCR assays. All E, coli isolates were examined for antimicrobial susceptibilities by the agar disc diffusion method, and the results were compared with those obtained from clinical samples. The percentage of each type of food with E. coli, including no heat food, low heat food, high heat food, and fruit juices and beverages, was higher than accepted standards at 60.4, 46.5, 38.6 and 20%, respectively. Of 140 E. coli isolates obtained from food samples, 11 isolates (7.9%) agglutinated with 6 monovalent O-antisera, including one isolate each of O6, O8, O114 and O159, two isolates of O1, and five isolates of O157. None of the 11 isolates harbored the virulence genes for EPEC, ETEC, EAEC, EIEC and STEC. Although O157 E. coli isolates were found, the most frequent, E. coli O157:H7, was not found in this study. The astA gene, however, was found in 1 E. coli isolate that showed weakly positive agglutination against the polyvalent antisera. Approximately 50% of the 140 E. coli isolates were resistance to at least one antimicrobial agent. The resistant strains showed high resistance to tetracycline (43%), co-trimoxazole (36%), ampicillin (26%) and chloramphenicol (23%), respectively. The resistance of E. coli was high for nearly all antimicrobial agents, particularly ampicillin (76%), tetracycline (70%), co-trimoxazole (69%) and nalidixic acid (44%). The results show that nearly half of the ready-to-eat food samples evaluated in Khon Kaen Municipality had levels of E. coli higher than acceptable standards. Of the diarrheagenic E. coli classified by serogroup, almost none of the isolates had virulence genes. These results indicate the disadvantage of relying on serogrouping alone for the recognition of diarrheagenic E. coli. E. coli isolated from food may not be an enteropathogenic strain. We also found that E. coli antimicrobial resistant strains are widespread in both food and humans.

Detection of H. pylori in dyspeptic patients and correlation with clinical outcomes.

The objectives of this study were to evaluate the methods used to diagnose Helicobacter pylon infection in gastric biopsies, and to evaluate the correlation between H. pylori infection and clinical outcomes. Gastric biopsies, obtained from 210 patients, were evaluated for H. pylori by culture, a commercial rapid urease test (RUT, Pronto Dry) and histological examination. A true positive result was either the culture or both the RUT and histological examination were positive. The results showed a H. pylori infection rate of 44.3% (93/210). The sensitivities, specificities, positive predictive values and negative predictive values were 88.2, 100, 100, and 91.4 % by the culture; 95.7, 98.3, 97.8, and 96.6% by RUT; and 96.8, 59.8, 59.8, and 65.7% by histological examination, respectively. The prevalences of H. pylori in non-ulcer dyspepsia (NUD), peptic ulcer dyspepsia (PUD) and gastric cancer (GCA) patients were 41.2, 57.9 and 70.6%, respectively. The chi-squared-test showed that GCA patients were significantly more frequent infected with H. pylori than NUD patients (p<0.05). Our study indicates that the RUT method was highly sensitive, specific and appropriate for routine clinical use.

Salmonellosis and the food chain in Khon Kaen, northeastern Thailand.

Non-typhoidal salmonellosis is a major cause of food-borne illness in Thailand. Specific serotyping of Salmonellae, linked with certain foods, can be used to identify outbreaks, transmission, and for surveillance. We aimed to identify the chain of non-typhoidal Salmonella transmission from food to humans in five slums, two open markets, four supermarkets and an abattoir in the municipality of Khon Kaen. During three months representing the cool-dry, hot-dry, and rainy seasons of 2002, culture samples were collected from water, food, pork, and chicken. Stool cultures of food venders, and others in the same area, were performed. Serological typing was done by the WHO National Salmonella and Shigella Center in Thailand. Of the food, drinking water, and stool samples from food handlers and healthy persons, 18, 7, 11, and 5%, respectively, were positive for Salmonella. Nearly all (96-98%) of the fresh pork and chicken, both from the open markets and supermarkets, were positive for Salmonella. The major Salmonella serovars were S. Anatum, S. Rissen, S. Virchow, S. Enteritidis and S. Panama, similar throughout the food chain and to the other reports that year. To reduce the incidence of human salmonellosis, several preventative measures must be taken where animals are produced, slaughtered and processed, and at home and in eateries. Vulnerable groups, such as infants, the elderly and the immuno-compromised, should be made aware of their increased susceptibility to food-borne disease.

Evaluation of polymerase chain reaction, conventional and MRSA screen latex agglutination methods for detection of methicillin-resistant, -borderline and -susceptible Staphylococcus aureus.

Methicillin-resistant Staphylococcus aureus (MRSA), is difficult and expensive to treat, therefore early screening is essential. Several phenotypic and genotypic methods are used to detect MRSA; however, the method of choice remains problematic. We have evaluated four phenotypic methods, broth microdilution (MIC), oxacillin disk agar diffusion (ODD), oxacillin screening salt agar (OSS), and a new rapid phenotypic (MRSA screen latex agglutination, MSLA) with the genotypic gold standard of PCR mecA detection to determine the most appropriate method for routine laboratory use. We randomly collected 203 S. aureus isolates from patients and carriers at two hospitals in Thailand. Using MIC method, three sub-groups were differentiated from among these isolates, namely MRSA (106 isolates), borderline-resistant S. aureus (BRSA) (65 isolates), and methicillin-susceptible S. aureus (MSSA)(32 isolates). A total of 10 methicillin-resistant S. epidermidis (MRSE) isolates were also included. The sensitivity and specificity of MIC, ODD, OSS, and MSLA were 99 and 96, 100 and 97, 100 and 97, and 100 and 100%, respectively. Our study indicated that ODD is still appropriate for routine laboratory. MSLA had the highest sensitivity and specificity and is rapid but expensive, so is the most appropriate method for emergency cases. MIC method was better for BRSA detection and OSS method was more appropriate for screening clinical specimens and carriers.

Evaluation of different primers for detecting mecA gene by PCR in comparison with phenotypic methods for discrimination of methicillin-resistant Staphylococcus aureus.

Detection of the mecA gene by polymerase chain reaction (PCR) is the gold standard for identifying methicillin-resistant Staphylococcus aureus (MRSA). PCR assays, employing MR1-MR2 primers (primer set 1) and MR3-MR4 primers (primer set 2) to generate 154 and 533 bp fragment, respectively, are most widely used for amplification of mecA gene. The purpose of this study was to evaluate the presence of mecA gene in 100 clinical isolates of S. aureus using PCR with the two pairs of primers. The results were compared to the broth dilution MIC method, oxacillin salt screening method (OSS) and oxacillin disk agar diffusion method (ODD). Fifteen of the 100 isolates showed a discrepancy between the mecA primer sets 1 and 2. Three isolates (3%) without the mecA gene showed discrepancies with phenotypic methods. The sensitivity, specificity and positive and negative predictive values for the 154 and 533 bp products of mecA were 79, 85, 83, 81 and 94, 100, 100, 94%, respectively. The results indicated that primer set 2 was more appropriate than primer set 1 for the detection of mecA gene in MRSA. There was a good correlation among the mecA gene detection, ODD and OSS methods. The discrepancy of three isolates between PCR and phenotypic methods should be clarified for other resistant mechanisms.