TEXT

Top Abstract Text References

The genus Providencia belongs to the family Enterobacteriaceae. Four species of this genus have been isolated from humans: Providencia alcalifaciens, P. stuartii, P. rettgeri, and P. rustigianii (6). Before the genus Providencia was classified into four species, the bacteria were collectively called the Providencia group of organisms (13) and were found to be causative agents of diarrhea in some studies (5, 12); the clinical spectrum suggested that they caused invasive diarrhea (5, 12). A recent study in the United Kingdom reported that the proportion of diarrheal British travelers returning from developing countries from whom P. alcalifaciens was isolated was significantly higher than that of diarrheal patients who had not traveled overseas and had P. alcalifaciens (10). Although this was not a case-control study, it suggested that P. alcalifaciens could cause travelers' diarrhea. Our studies in laboratory models confirmed the diarrheagenic potential of P. alcalifaciens (1, 2, 11). However, case-control studies showing an epidemiological association of P. alcalifaciens with diarrhea are lacking. Therefore, we carried out a study in which we compared the rate of isolation of P. alcalifaciens from the stool specimens of children with diarrhea with that in matched children without diarrhea. The results are reported in this article.

Children younger than 5 years of age with acute diarrhea (defined as three or more loose stools a day) seen at the Clinical Research and Service Centre of the International Centre for Diarrhoeal Disease Research, Bangladesh, in Dhaka, Bangladesh, were studied from February 1993 to July 1994. These children were the routine 4% of diarrheal surveillance patients (every 25th patient) seen at the Clinical Research and Service Centre (14). The patients were first seen in a triage area, and those requiring further care were admitted to the hospital ward. A physician performed a physical examination and assessed the patients' dehydration status as none, mild, moderate, or severe according to clinical signs (15). A health assistant administered a questionnaire regarding demographic background, medical history, and previous treatment to the patient or to an adult guardian. A stool specimen collected immediately after admission was used for etiological studies.

Controls were healthy children from the same neighborhood as the case patients. After leaving the household of a patient in the community, a health worker tossed a pencil onto the street and then walked in the direction the sharpened end of the pencil pointed. After passing the first 10 houses, the health worker located the first house with an age-matched control and collected the same information from the control that had been collected from the patients. Control children had not taken antibiotics during the previous 2 weeks. Stool samples from controls were collected within 2 weeks of samples from patients and were transported in refrigerated boxes to the laboratory within a few hours.

Stool specimens were cultured for P. alcalifaciens on modified proteeae isolation medium (10). Characteristic colonies causing browning of the medium were further tested for P. alcalifaciens as described previously (1). These colonies were confirmed to be P. alcalifaciens by their reaction on an API-20E strip (API System, Montalieu, Vercieu, France). The isolates were studied for invasiveness and actin condensation in the HEp-2 cell monolayer as described previously (2). Briefly, for invasiveness, 10⁷ CFU of bacteria were inoculated onto the monolayer and allowed to interact with it for 2 h. After the monolayer was washed to remove nonadherent bacteria, it was incubated for another hour in the presence of gentamicin. The gentamicin-resistant bacteria, which are an indicator of invasiveness, were enumerated after the monolayer was lysed with 1% Triton X-100. To detect actin condensation in the HEp-2 cell monolayer, the monolayer was incubated with bacteria for 3 h; after the nonadherent bacteria were removed by washing, the monolayer was stained with fluorescein-labeled phalloidin and viewed under a fluorescent microscope.

Stool specimens were also examined for other enteric pathogens, including rotavirus, Giardia lamblia, Entamoeba histolytica, Cryptosporidium sp., and recognized bacterial pathogens as described previously (3). The bacterial pathogens sought were Vibrio cholerae O1 and O139, Salmonella spp., Shigella spp., Campylobacter spp., Aeromonas spp., Plesiomonas sp., Clostridium difficile, and diarrheagenic Escherichia coli (enterotoxigenic, enteropathogenic [EPEC], enteroaggregative, enteroinvasive, enterohemorrhagic, and diffuse adherent [DAEC]).

During the period of study, 814 children with diarrhea and the same number of matched controls were tested for enteric pathogens, including P. alcalifaciens. P. alcalifaciens was cultured from 17 patients with diarrhea and four control children without diarrhea (P = 0.004 by ² test), indicating a significant association of the organism with diarrhea. Data for the isolation of P. alcalifaciens in relation to that of other enteric pathogens are presented in Table 1; in terms of rank order, P. alcalifaciens was next to Shigella spp. However, whereas P. alcalifaciens was the only pathogen that was isolated from five patients, other patients had copathogens (Table 2). Two of the four control children positive for P. alcalifaciens had copathogens; one had Plesiomonas shigelloides, and the other had DAEC. Two of the P. alcalifaciens isolates from the control children and 9 of 17 isolates from the patients were available for studies on invasion and actin condensation in the HEp-2 cell monolayer (2). All isolates from control children and patients invaded the cell monolayer (gentamicin-resistant progeny, >10³ CFU per ml), with associated actin condensation (2).

The invasive nature of the isolates is consistent with our previous findings: P. alcalifaciens invaded HEp-2 cells, with actin condensation, produced inflammatory changes associated with the invasion of tissues in adult rabbit ileal loops, and produced invasive diarrhea in a removable intestinal tie adult rabbit diarrhea model (1). Electron microscopy of the ileal mucosa of these animals showed two modes of entry into intestinal epithelial cells, one directly by endocytosis associated with polymerization of cytoskeletal components and the other by disruption of tight junctions, with entry into and proliferation in intercellular spaces (11). Further studies with additional isolates from diarrheal stool samples in Bangladesh confirmed the invasiveness of P. alcalifaciens and showed that both invasion and actin condensation were inhibited by cytochalasin D, an antimetabolite that inhibits microfilament formation (2). Invasiveness and actin condensation were also shown to be present in P. alcalifaciens isolates from diarrheal patients in Brazil (9). Consistent with the invasive nature of P. alcalifaciens, four of the five patients infected only with P. alcalifaciens had manifestations of invasive diarrhea (Table 3).

This first case-control study of children with diarrhea to investigate the etiologic role of P. alcalifaciens suggested that the organism is associated with diarrhea. However, in the present study, P. alcalifaciens was associated with diarrhea in only a small proportion (2.1%) of the children. Many recognized enteric pathogens are normally isolated from some apparently healthy subjects in case-control studies of diarrhea in developing countries (4, 8), as in the present study. (P. alcalifaciens was isolated from four apparently healthy children.) The majority of the patients from whom P. alcalifaciens was isolated also had copathogens. The isolation of multiple enteric pathogens from stool samples is common in many developing countries, including Bangladesh (7). This problem reflects the high degree of fecal contamination in the environments of developing countries.