Peter Echeverria

Trends in Antibiotic Resistance Among Diarrheal Pathogens Isolated in Thailand Over 15 Years

Antibiotic resistance trends were examined for Shigella species, nontyphoidal Salmonella species, enterotoxigenic Escherichia coli (ETEC), and Campylobacter species isolates from indigenous persons and travelers in Thailand for up to 15 years. Resistance to trimethoprim-sulfamethoxazole was found in >90% of Shigella and 40% of ETEC and nontyphoidal Salmonella isolates. Resistance to nalidixic acid was found in 97%-100% of Shigella dysenteriae 1 strains isolated between 1992 and 1995. Ciprofloxacin resistance was detected in 1% of ETEC isolates in 1994 and 1995 and in one of 349 nontyphoidal Salmonella isolates in 1995. Ciprofloxacin resistance among Campylobacter species increased from zero before 1991 to 84% in 1995 (P < .0001). Azithromycin resistance was found in 7%-15% of Campylobacter isolates in 1994 and 1995, as well as 15% of ETEC and 3% of Salmonella isolates in 1995. Enteric pathogens in Thailand have developed resistance to virtually all antibiotics routinely used in the treatment of diarrhea, as well as the newer fluoroquinolone and macrolide classes of drugs.

Astroviruses as a cause of gastroenteritis in children.

BACKGROUND Infection with astroviruses has been associated with gastroenteritis in children, and serologic surveys indicate that this infection may be frequent. The importance of astroviruses as agents of gastroenteritis has not been shown in a controlled study, however. METHODS We used monoclonal antibody-based enzyme immunoassays to detect astroviruses, enteric adenoviruses, and rotaviruses in stool samples obtained from age-matched children with and children without gastroenteritis. The samples were obtained in two studies, three years apart, among patients attending an outpatient clinic in Bangkok, Thailand. RESULTS In the first study, astroviruses were detected in 8.6 percent (96 of 1111) of the children with gastroenteritis and in 2.0 percent (19 of 947) of the children without gastroenteritis. In the second study the rates were 8.6 percent (50 of 580) and 2.1 percent (11 of 512), respectively. For both studies combined, enteric adenoviruses were detected in 2.6 percent of those with gastroenteritis and in 0.5 percent of the controls, whereas rotaviruses were detected in 19 percent of those with gastroenteritis and in 1.0 percent of the controls. The clinical findings associated with astrovirus infection were similar to those associated with rotavirus infection, except for a trend toward greater dehydration in the children infected with rotaviruses. CONCLUSIONS These two controlled studies involving a total of 3150 Thai children provide evidence that astroviruses are a common cause of viral gastroenteritis. Astroviruses were found in association with gastroenteritis more frequently than were enteric adenoviruses, and with nearly half the frequency of rotaviruses.

Epidemiology of diarrhoeal illness associated with coccidian-like organism among travellers and foreign residents in Nepal

A newly described organism called CLB (coccidian-like or cyanobacterium-like body) has been identified in cases of prolonged diarrhoea. To confirm an association of CLB with disease and identify risk factors for transmission, we conducted a case-control study of travellers and foreign residents at two outpatient clinics in Kathmandu, Nepal. Patients without diarrhoea were matched to CLB cases by clinic and date of visit. For comparison, patients with other causes of diarrhoea were also studied. Stools were examined for enteric pathogens with standard microbiological and molecular genetic techniques. CLB was identified in 108 (11%) of 964 individuals with gastrointestinal symptoms compared with only 1 (1%) of 96 symptom-free controls (p = 0.003). 7% of residents in the US Embassy community acquired the infection. The diarrhoeal illness associated with CLB lasted a median of 7 weeks (interquartile range 4-9) compared with 9 days (4-19) for individuals with other causes of diarrhoea (p < 0.0001). The prevalence of other enteric pathogens was no higher among CLB cases than among symptom-free controls. Patients with CLB infection were more likely than controls to report consumption of untreated water (odds ratio 3.98; 95% CI 1.29-13.14); organisms of the same appearance were identified in an epidemiologically implicated water sample. The significant association of CLB with prolonged diarrhoea, and the low rate of other enteropathogens in CLB cases, strongly supports the hypothesis that CLB is a new pathogen. Epidemiological and environmental data suggest that the organism is waterborne.

Placebo-controlled trial of co-trimoxazole for cyclospora infections among travellers and foreign residents in Nepal

Cyclospora is a coccidian (previously referred to as cyanobacterium-like bodies) that has been implicated in cases of prolonged diarrhoea. The average duration of symptoms is more than three weeks, and no specific treatment has been shown to shorten the illness. A case report suggested that co-trimoxazole may be effective. Expatriate persons with gastrointestinal complaints and cyclospora detected on examination of faeces were recruited from two clinics in Kathmandu, Nepal, between May and August, 1994. Participants were assigned in a randomised, double-blinded manner to receive either cotrimoxazole (160 mg trimethoprim, 800 mg sulphamethoxazole) or placebo tablets twice daily for 7 days. Of 40 patients included in the study, 21 received cotrimoxazole and 19 placebo. There were no significant differences between these two groups in age, sex, time in Nepal, duration or severity of illness, or presence of other enteric pathogens. After 3 days, 71% of patients receiving co-trimoxazole still had cyclospora detected, compared with 100% of patients receiving placebo (p = 0.016). After 7 days, cyclospora was detected in 1 (6%) of 16 patients treated with co-trimoxazole who submitted stool specimens compared with 15 (88%) of 17 patients receiving placebo (p < 0.0001). Eradication of the organism was correlated with clinical improvement. There was no evidence of relapse of infection among treated patients followed for an additional 7 days. Treatment with co-trimoxazole for 7 days was effective in curing cyclospora infection among an expatriate population in Nepal.

Ciprofloxacin and loperamide in the treatment of bacillary dysentery.

Antimotility drugs are not recommended in the treatment of dysentery. In guinea pigs, the motility of the small intestine is an important defense mechanism. Eight of 10 starved guinea pigs that were experimentally infected with Shigella and given opium died, whereas 10 similarly infected guinea pigs not given opium survived [1]. In a study of diphenoxylate-atropine (Lomotil, Searle and Company, Chicago, Illinois) used alone or in combination with oxolinic acid to treat experimental shigellosis in 25 volunteers [2], the authors concluded that drugs that reduced intestinal motility should not be used to treat invasive diarrhea. The report acknowledged that the results were inconclusive and suggested that larger trials should be done in naturally occurring illness to document the frequency and severity of prolongation of fever and pathogen carriage when patients are treated with Lomotil or paregoric [2]. Loperamide, a synthetic antidiarrheal agent, has not been studied in large numbers of patients with dysentery. Although it does not contain atropine (as does Lomotil), its primary action is to slow intestinal motility. In studies designed to look at nondysenteric diarrhea, loperamide has recently been used, either alone or in combination with antibiotics (cotrimazole or ciprofloxacin), to treat more than 25 adults with Shigella infections [3-7]. Although loperamide did not deliver a definable therapeutic advantage, the drug appeared to be safe in treating patients infected with Shigella. Because loperamide is frequently used in the treatment of diarrhea, which is often due to Shigella, a double-blind, placebo-controlled, randomized clinical trial of loperamide in the treatment of bacillary dysentery (primarily due to Shigella) was done. Methods Study Design Between 28 November 1990 and 29 February 1992, we studied patients with dysentery at Bamrasnaradura Hospital, Nonthaburi, Thailand. Dysentery was defined as three or more loose stools containing blood or mucus associated with at least one of the following symptoms: abdominal cramps, nausea, vomiting, or temperature greater than 38 C. Patients entered in the study had symptoms for fewer than 60 hours. Patients were excluded if they had received antimotility drugs or antibiotics in the previous 7 days, were pregnant, were immunosuppressed, or could not be followed for the next 10 days. Adults admitted to the Bamrasnaradura Hospital who met the study criteria and gave consent were treated with 500 mg of ciprofloxacin (Cipro, Miles Pharmaceuticals, West Haven, Connecticut) twice daily for 3 days. Patients were sequentially assigned a study number and an associated vial containing 16 caplets of loperamide (Imodium, McNeil Consumer Products Company, Fort Washington, Pennsylvania) or identical placebo. Numbered vials were randomized before the beginning of the study using a computer-generated random number table. Nurses, physicians, investigators, and patients were blinded to treatment regimen throughout the study. Patients received 4 mg (two caplets) of loperamide as an initial dose under the observation of a nurse. Patients were instructed to take one caplet after every loose stool (as many as eight caplets per day). Age, duration of diarrhea in hours, number of previous unformed stools, previous medication, and history of symptoms were recorded. Patients were examined by a physician, and stool specimens were collected at study entry. Patients were interviewed daily for 3 days (or until diarrheal symptoms had resolved) and again 10 days later. Symptoms, the time of the last unformed stool, and the number of unformed stools occurring during the previous 24 hours were recorded. Stools (or rectal swabs) were collected at the time of each interview. Laboratory Studies Stools were examined microscopically for fecal leukocytes and erythrocytes after staining with methylene blue [8] and were cultured promptly on MacConkey and Hektoen agar before and after inoculation in Selenite F broth (Difco, Detroit, Michigan). Bacterial, viral, and intestinal protozoa were identified as previously described [9, 10]. Up to 10 lactose-fermenting and up to 10 nonlactose-fermenting Escherichia coli, as identified on a MacConkey plate, were saved on Dorset egg yolk media slants. Escherichia coli isolates were tested within 1 month of isolation for heat-labile and heat-stable toxin production in the Y-1 adrenal cell [11] and suckling mouse [12] assays. Escherichia coli isolates were also fixed on Whatman 541 filters (Millipore Corporation, Bedford, Massachusetts) as described by Maas [13] and were tested for genes coding for Shiga-like toxin I and II as well as those with the 17-kilobase EcoRI fragment of pWR100 [14]. Isolates that hybridized with the 17-kilobase EcoRI probe were speciated and tested using the Sereny test [15]. Shigella, enteroinvasive E. coli, and enterotoxigenic E. coli as well as Aeromonas, Plesiomonas, and Vibrio species were tested for susceptibility to ampicillin, chloramphenicol, ciprofloxacin, nalidixic acid, trimethoprim-sulfamethoxazole, and tetracycline by disc diffusion [16] using CM471 agar (Oxoid Ltd., Basingstoke, United Kingdom). Enteroinvasive and enterotoxigenic E. coli were serotyped as described by Orskov and Orskov [17]. Intestinal parasites were identified by direct microscopy of saline suspension of stools after formalin-ether concentration, polyvinyl alcohol fixation of stools, and trichrome staining [18]. Cryptosporidium was identified microscopically with a modified dimethyl-sulfoxide method [19]. Rotavirus was identified using a monoclonal enzyme-linked immunosorbent assay (Pathfinder, Kallsted Laboratories, Austin, Texas) [20]. Data Analysis All investigators, nurses, and laboratory personnel were blinded to treatment group until all end points were determined at the end of the study. Data were entered from data collection forms and laboratory records into a computer database. Patients not followed until the resolution of diarrhea and those found not to have met the enrollment criteria were excluded from the analysis before unblinding. Infections with enteroinvasive E. coli or Shigella were evaluated together because of the similar pathogenicity of these organisms [21]. Statistical comparisons were made using Epi Info software [22]. A P value of less than 0.05 was considered statistically significant. The sample size necessary to show a difference of 24 hours in duration of diarrhea (80% power, P < 0.05) was 66 patients with Shigella or enteroinvasive E. coli. We planned to conduct this study over a 1-year period or until 66 patients with Shigella or enteroinvasive E. coli infections were entered. Data analysis included a chi-square with Yates correction (or two-tailed Fisher exact test if an expected cell was less than five) for the detection of proportional differences between the two treatment groups; the Student t-test for the comparison of mean ages and weights; and the Mann-Whitney-Wilcoxon test for the evaluation of non-normally distributed outcome variables (duration of diarrhea and number of diarrheal stools). Confidence intervals (CIs) of 95% are given where appropriate. Results From November 1990 to February 1992, patients with dysentery were screened by a nurse each morning, Monday through Friday. Ninety-two adults met the enrollment criteria. Two patients refused to submit additional specimens, one vomited his initial medication, and one could not be located after discharge from the hospital. Our final sample included 88 analyzable patients with dysentery. Forty-two patients received loperamide and 46 received placebo. Characteristics of the evaluable patients are shown in Table 1. Patients in the loperamide and placebo groups did not differ significantly in age, weight, sex, fecal leukocytes or erythrocytes, proportion who had fever at enrollment, duration of diarrhea before enrollment, or number of diarrheal stools during the 24 hours immediately preceding enrollment. Shigella species were isolated from 44 (50%) and enteroinvasive E. coli from 3 (3%) study participants (Table 2). The patients in whom Shigella or enteroinvasive E. coli were isolated had a lower mean weight (53 compared with 58 kg) and had a longer duration of diarrhea before enrollment (26 compared with 8 hours) than did the patients in whom these organisms were not cultured. Severity of illness was greater in the Shigella-enteroinvasive E. coli group, which had a higher proportion of volunteers with fecal leukocytes (87% compared with 46%), fecal erythrocytes (94% compared with 34%), and fever at enrollment (47% compared with 22%) than did other patients. The median number of stools passed in the 24 hours before enrollment was similar in the two groups (10 and 8 stools, respectively). Table 1. Patient Characteristics at Study Entry* Table 2. Enteric Pathogens Identified in 88 Patients with Dysentery Receiving Ciprofloxacin and Loperamide or Placebo* After treatment, the number of diarrheal stools was fewer for patients with dysentery who received loperamide (median, 4.5 and 7.0 stools, respectively; P = 0.030). This difference was caused by patients infected with Shigella or enteroinvasive E. coli (P = 0.016) (Table 3). In addition, the duration of diarrhea was less in patients infected with Shigella or enteroinvasive E. coli who received loperamide (P = 0.028) (Table 3). During the observation period after the start of treatment, the percentage of patients infected with Shigella or enteroinvasive E. coli receiving loperamide compared with placebo did not differ statistically in the following parameters: abdominal cramps (21% compared with 27%, P = 0.282), nausea (16% compared with 18%, P = 0.903), vomiting (16% compared with 19%, P = 0.770), or constipation (5% compared with 0%, P = 0.447). An oral temperature greater than 38 C was not found in any participant for more than 1 day after enrollment. Shigella or enteroinvasive E. coli were isolated 1 day after begin

POLYMICROBIAL AETIOLOGY OF TRAVELLERS' DIARRHOEA

Of 35 US Peace Corps volunteers in Thailand, 20 (57%) had a total of 30 episodes of diarrhoea during their first 6 weeks in the country. Enteric pathogens were associated with 90% of the episodes. A single pathogen was identified in 17 (57%) episodes, 2-4 pathogens were identified in 10 (33%) episodes, and there were 15 symptomless infections. Enterotoxigenic Escherichia coli (ETEC) was identified in 37% of these episodes, and various salmonella serotypes were isolated in 33%. Infections with 9 other enteric pathogens were also identified: Campylobacter jejuni (17%), Plesiomonas shigelloides (13%), Aeromonas hydrophila (10%), Blastocystis hominis (7%), Norwalk virus (7%), Vibrio parahaemolyticus (3%), non-O1 Vibrio cholerae (3%), Vibrio fluvialis (3%), and rotavirus (3%). In total, 56 enteric infections were documented in 35 volunteers.

Two-Year Study of the Protective Efficacy of the Oral Whole Cell plus Recombinant B Subunit Cholera Vaccine in Peru

The protective efficacy of an oral inactivated whole cell Vibrio cholerae plus recombinant B subunit cholera vaccine was determined against El Tor cholera among Peruvian children and adults (2-65 years old) in a randomized, double-blind manner. Study subjects received 2 doses of vaccine or placebo 2 weeks apart, followed by a booster dose 10 months later. Surveillance for cholera was performed actively, with 2 visits per week to each household, and passively, at a local hospital. Stool samples were collected during diarrhea episodes and were cultured for V. cholerae. A total of 17,799 persons received 2 doses of vaccine or placebo, and 14,997 of these persons received the booster dose. After 2 doses (first surveillance period), V. cholerae biotype O1 was isolated from 17 vaccinees and 16 placebo recipients, demonstrating vaccine efficacy (VE) of -4%. After 3 doses (second surveillance period), V. cholerae O1 was isolated from 13 vaccinees and 32 placebo recipients, demonstrating VE of 61% (95% confidence interval ¿CI, 28%-79%). In the second surveillance period, the VE for illness requiring hospitalization was 82% (95% CI, 27%-96%). VE was also higher for persons >15 years old (VE, 72%; 95% CI, 28%-89%).

Antimicrobial resistance and enterotoxin production among isolates of Escherichia coli in the Far East.

The frequency of association between transferable extrachromosomal D.N.A. (plasmid) mediated antibiotic resistance and enterotoxin productin is unknown. The antimicrobial susceptibility of 176 enterotoxigenic Escherichia coli from 57 children and adults in the Philippines, Korea, Taiwan, and Indonesia has been examined. 126 isolates (72%) were resistant to one or more antibiotic(s); 77 (44%) were resistant to four or more antibiotics. 43 E. coli which produced both heat-labile and heat-stable toxin, 110 isolates which produced only heat-labile toxin, and 23 which produced only heat-stable toxin were frequently resistant to multiple antibiotics. 25 of 31 resistant isolates tested, 80% transferred antibiotic resistance in bacterial mating experiments. In 35% of the matings transferring antibiotic resistance, the ability to produce enterotoxin was also conferred on the recipients. This in-vitro observation suggests that the widespread use of antibiotics could increase the distribution of enterotoxigenic E. coli, as genes coding for antibiotic resistance and enterotoxin production are frequently transferred together.

HAEMOPHILUS INFLUENZAE TYPE B RESISTANT TO AMPICILLIN AND CHLORAMPHENICOL IN AN ORPHANAGE IN THAILAND

Abstract Between July 1 and Oct. 30, 1979, three children from an orphanage in Bangkok, Thailand, died of meningitis caused by Haemophilus influenzae type B which was resistant to ampicillin (Amp) and chloramphenicol (Chlor). H. influenzae type B was isolated from the nasopharynx of 38 of 219 children at the orphanage (17%); 47% (18/38) of these isolates were resistant to ampicillin and chloramphenicol. In an attempt to eradicate this organism from the nasopharynx, 93 children were treated with rifampicin (rifampin), and 92 were treated with rifampicin (10 mg/kg/12 h) and minocycline (2 mg/kg/12 h) for three days. Among 18 children colonised with H. influenzae type B when treatment was given, 83% (10/12) treated with rifampicin, and 67% (4/6) treated with rifampicin and minocycline had negative nasopharyngeal cultures one week after therapy. The carriage rate of H. influenzae type B among all children in the orphanage, however, remained unchanged (9% before treatment and 10% after treatment). Furthermore 33% of organisms isolated before therapy, and 44% isolated after therapy were resistant to ampicillin (Amp) and chloramphenicol (Chlor). 34 of 35 resistant (Amp r Chlor r , Amp r Chlor s , or Amp s Chlor r ) isolates were more resistant to minocycline than 13 sensitive strains (minimum inhibitory concentration [MIC] 0·2 vs 0·1 μg/ml), and six resistant strains were more resistant to rifampin than 16 sensitive strains (MIC 0·4 vs 0·2 μg/ml). A 4·5 megadalton plasmid encoded for β-lactamase, and 2·2 and 1·2 megadalton plasmids were associated with chloramphenicol resistance. The 4·5 megadalton plasmid was similar in size to plasmids encoding for β-lactamase in Neisseria gonorrhoeae isolated in Thailand and the Philippines, and reported in ampicillin-resistant H. influenzae from Europe and the U.S.A. Attempts to control epidemics of ampicillin and chloramphenicol resistant H. influenzae type B with either refampicin, or refampicin and minocycline may increase the prevalence of this multi-resistant pathogen.

A simple polymerase chain reaction technique to detect and differentiate Shigella and enteroinvasive escherichia coli in human feces

A simple polymerase chain reaction (PCR) procedure using IS630-specific primers was developed as a general diagnostic probe to detect Shigella and enteroinvasive Escherichia coli (EIEC). However, IS630 and the other two previously reported molecular probes, ipaH and ial, cannot be used to differentiate among Shigella serotypes and EIEC strains that cause dysentery. The sensitivity of PCR protocol was determined to be 100-200 shigellae for each PCR reaction. An enrichment incubation would allow the detection of shigellae in stool samples with low bacterial concentration; i.e., < 10(4) CFU/gram. Serotype-specific primers derived from the rfc genes of differentiate among Shigella serotypes in the laboratory, such as S. sonnei, S. flexneri, and S. dysenteriae 1. It was demonstrated further that the multiplex PCR system containing rfc-specific primers can efficiently identify the most prominent Shigella serotypes in raw stool samples of acute diarrheal patients.