Diarrheal disease among military personnel during Operation Restore Hope, Somalia, 1992-1993.

The potential for widespread diarrheal disease was regarded as a substantial threat to U.S. troops participating in the early phases of Operation Restore Hope in Somalia. Outpatient surveillance of 20,859 U.S. troops deployed during the first eight weeks, however, indicated that a mean of only 0.8% (range 0.5-1.2%) of personnel sought care for diarrhea each week, and in three epidemiologic surveys, < 3% of troops reported experiencing a diarrheal illness per week. Despite these low overall attack rates, diarrhea accounted for 16% of 381 hospital admissions and 20% of 245 patients admitted with a temperature > or = 38.5 degrees C. Sixty-one specimens were obtained from inpatients and 52 were obtained from outpatients. Shigella sp. were isolated from 33%, enterotoxigenic Escherichia coli from 16%, Giardia lamblia from 4%, and rotavirus from 1% of 113 stool samples obtained from inpatient (61) and outpatient (52) troops with diarrhea. Bacterial isolates obtained in Somalia were resistant to doxycycline (78%), ampicillin (54%), and sulfamethoxazole (49%), but uniformly sensitive to ciprofloxacin. With the exception of 10 Shigella sonnei isolates that were linked epidemiologically to one eating facility, bacterial pathogens occurred sporadically and demonstrated a wide variation of serotypes and antibiotic sensitivity patterns. Additionally, three of 11 paired sera collected from persons with nausea, vomiting, and watery diarrhea demonstrated a four-fold or greater increase in titer to Norwalk virus antibody. These data indicate that large outbreaks of diarrheal disease did not occur; however, highly drug-resistant enteric bacteria, and to a lesser extent viral and parasitic pathogens, were important causes of morbidity among U.S. troops in Somalia.

Dengue fever in U.S. troops during Operation Restore Hope, Somalia, 1992-1993.

Dengue fever (DF) was considered to be a potential cause of febrile illness in U.S. troops deployed to Somalia during Operation Restore Hope in 1992-1993. A prospective study of hospitalized troops with fever and a seroepidemiologic survey of 530 troops were conducted. Among 289 febrile troops hospitalized, 129 (45%) did not have an identified cause of their fever. Dengue (DEN) virus was recovered from 41 (43%) of 96 of these patients by inoculation of admission sera into C6/36 cell cultures. Thirty-nine (41%) of the isolates were identified as DEN-2 and two (2%) as DEN-3 by an indirect immunofluorescent antibody assay. An additional 18 (49%) of 37 culture-negative cases were shown by immunoglobulin M (IgM) antibody capture enzyme-linked immunosorbent assay to have anti-DEN virus antibody. All identified DF cases recovered within 1-2 weeks; no case of dengue hemorrhagic fever or shock syndrome was observed. A seroepidemiologic survey of a unit (n = 494) with 17 culture or serologically identified DF cases and a 13% attack rate of unidentified febrile illness revealed a 7.7% prevalence of anti-DEN virus IgM antibody. Failure to use bed nets was the only identified risk factor for DEN infection (adjusted odds ratio = 2.2, 95% confidence interval = 1.4-3.0). These data indicate that DF was an important cause of febrile illness among US troops in Somalia, and demonstrate the difficulties in preventing DEN infection in troops operating in field conditions.

Etiology of acute diarrhea among United States military personnel deployed to South America and west Africa.

A study of acute diarrhea was conducted from 1985 to 1987 among U.S. military personnel participating in routine shipboard exercises in South America and West Africa and ground troops deployed to coastal Ecuador. An enteropathogen was identified in 146 (51%) of 289 acute cases of diarrhea. Enterotoxigenic Escherichia coli, found in 50 (17%) patients with diarrhea, was the most commonly identified enteropathogen. Viral enteropathogens were also found in a high percentage of acute cases of diarrhea: rotavirus was detected in 11% of the patients and Norwalk virus infection in 10%. Most enteric pathogens were acquired in equal frequencies in South America and West Africa, except for rotavirus infection which was identified more often in West Africa and enteroaggregative E. coli infection which was identified more often in South America. Bacterial enteropathogens were frequently resistant to trimethoprim/sulfamethoxazole, but no resistance to quinolone drugs was observed, indicating that quinolone drugs have become important agents for the treatment of diarrhea in South America and West Africa.

Nontoxigenic 01 Vibrio cholerae in Peru: a report of two cases associated with diarrhea.

Two 01 Vibrio cholerae, E1 Tor strains, serogroup Ogawa were isolated from diarrheal stool material of two North American males residing or visiting in Peru. No other enteric pathogens were found. These strains did not produce cholera toxin as detected by enzyme-linked immunosorbent and Y1 adrenal cell assays, nor did they hybridize with an oligonucleotide probe for heat-labile (LT) toxin of Escherichia coli. These two cases are the first reports of 01 V. cholerae isolated from clinical specimens in South America.

Norfloxacin compared to trimethoprim/sulfamethoxazole for the treatment of travelers' diarrhea among U.S. military personnel deployed to South America and West Africa.

A randomized treatment trial of travelers' diarrhea was carried out among U.S. military personnel participating in routine exercises in several port cities in South America and West Africa. A 5-day, twice daily course of either norfloxacin (400 mg) or trimethoprim/sulfamethoxazole (TMP/SMX, 160/800 mg) was given to 142 volunteers. At the end of 5 days of treatment, diarrhea had resolved in 100% of 73 patients receiving norfloxacin and 97.1% (67/69) receiving TMP/SMX. A probable bacterial pathogen was determined in 44% of 142 subjects: 49% of the norfloxacin group and 39% of the TMP/SMX group. The most common pathogens detected were enterotoxigenic Escherichia coli in 20% of cases and rotavirus in 15%. Resistance to TMP/SMX was present in 20 (27%) bacterial isolates, while no resistance to norfloxacin was found. Eight of 10 patients in the TMP/SMX treatment group who had TMP/SMX-resistant bacterial enteropathogens improved clinically. Both norfloxacin and TMP/SMX were clinically effective in the treatment of travelers' diarrhea in this military population.

Nucleotide sequences of the genes regulating O-polysaccharide antigen chain length (rol) from Escherichia coli and Salmonella typhimurium: protein homology and functional complementation.

In this article, we report on the nucleotide sequences of the rol genes of Escherichia coli O75 and Salmonella typhimurium LT2. The rol gene in E. coli was previously shown to encode a 36-kDa protein that regulates size distribution of the O-antigen moiety of lipopolysaccharide. The E. coli and S. typhimurium rol gene sequences consist of 978 and 984 nucleotides, respectively. The homology between the nucleotide sequences of these two genes was found to be 68.9%. Both the E. coli rol and S. typhimurium rol genes are transcribed counter to the histidine operon and code for deduced polypeptides of 325 and 327 amino acids, respectively. The S. typhimurium rol gene was previously identified to encode a protein of unknown function and to share a transcription termination region with his. The homology between these deduced polypeptide sequences was observed to be 72%. A complementation test was performed in which the S. typhimurium rol gene was placed in trans with an E. coli plasmid (pRAB3) which encodes the O75 rfb gene cluster and not rol. The protein expressed from the S. typhimurium rol gene was found to regulate the distribution of the O75 O polysaccharide on the lipopolysaccharide of the host strain, E. coli S phi 874. The mechanism of Rol action may be independent of O antigen subunit structure, and its presence may be conserved in members of the family Enterobacteriaceae and other gram-negative bacilli that express O polysaccharides on their surface membrane.

Regulation by a novel protein of the bimodal distribution of lipopolysaccharide in the outer membrane of Escherichia coli.

We report on the cloning and characterization of the rfb gene cluster of the O75 lipopolysaccharide from a urinary tract isolate of Escherichia coli. Deletion cloning defined the minimum region of DNA that expressed the O75 antigen in E. coli host strains to be on a 12.4-kb insert. However, the E. coli strain expressing this region did not produce a polymerized O chain as detected by sodium dodecyl sulfate-polyacrylamide gel electrophoresis and silver staining. A slightly larger DNA clone of 13.4 kb produced a polymerized O chain in E. coli S phi 874 but was found to be abnormal in its distribution over the surface membrane. Normal wild-type E. coli, as with Salmonella spp., has a bimodal distribution of the lipopolysaccharide on the surface which is seen as an abundance of long and short O chains attached to the lipid A-core structure. We found in a region adjacent to the cloned rfb region, and on the opposite side from where the putative polymerase (rfc) is encoded, a novel protein of 35.5 kDa expressed from a 1.75-kb DNA fragment. This protein was shown to complement in trans the E. coli strains carrying plasmids that expressed abnormal, unregulated lipopolysaccharides. The expression of these complemented strains was bimodal in distribution. Mutation of the gene encoding this protein destroyed its ability to regulate O-chain distribution. We propose to call this regulator gene rol, for regulator of O length.

Antimicrobial susceptibility of Neisseria gonorrhoeae isolates from a military population in San Diego.

Neisseria gonorrhoeae isolates were studied to determine their patterns of antimicrobial susceptibility and possible chemotherapeutic implications. Of 370 consecutive isolates, 32 (8.7%) were penicillinase-producing N. gonorrhoeae (PPNG). The remaining 338 were subjected to disk-diffusion tests, and those apparently resistant to penicillin, tetracycline, or spectinomycin were tested by an agar-dilution method. The dilution test showed that 5.4% (20/370) were penicillin-resistant, non-PPNG strains, of which 100%, 90%, and 45% were also resistant to tetracycline, cefoxitin, and erythromycin, respectively. No resistance to spectinomycin or ceftriaxone was demonstrated, although there was an association between minimum inhibitor concentrations (MICs) of penicillin of greater than or equal to 1.0 microgram/ml and increased MICs of ceftriaxone. The overall incidence of penicillin resistant isolates, including PPNG, was 14.1% (52/370). Of the 20 penicillin-resistant, non-PPNG strains, all were also resistant to tetracycline, and another 21 exhibited tetracycline resistance but were sensitive to penicillin. The in-vitro data suggested that: (1) neither penicillin, tetracycline, nor cefoxitin were acceptable drugs for routine treatment of gonorrhea in our population during the study period; (2) spectinomycin and ceftriaxone continue to demonstrate adequate in-vitro activity against N. gonorrhoeae despite increasing in-vitro resistance to penicillin; and (3) non-plasmid-mediated resistance to penicillin may predict future resistance to ceftriaxone.

An unusual case of penicillinase-producing Neisseria gonorrhoeae resistant to spectinomycin in California.

We report a case of gonorrhea due to a penicillinase-producing strain of Neisseria gonorrhoeae resistant to spectinomycin in a 26-year-old man who had not been out of the United States for a year-and-a-half. His sexual contact also had no recent travel out of the United States. The genital and oropharyngeal infections were successfully treated with cefoxitin (1 g im) plus probenecid (1 g orally) and trimethoprim-sulfamethoxazole (80 mg of trimethoprim and 400 mg of sulfamethoxazole). The patient took nine of the latter tablets daily for five days. The organism was a serovar IB-3, proline-requiring auxotype. The patient's isolate contained both 2.6-megadalton and 4.4-megadalton plasmids. Measurement of minimal inhibitory concentrations (MICs) of antibiotics for the isolate confirmed the penicillin resistance and showed an MIC of spectinomycin of greater than 256 micrograms/ml. The epidemiologic investigation suggested that the source of the infection was a male contact with unusual clinical features, including bloody urethral discharge and a possible incubation period of 28 days.