Detection of Shiga toxin-producing and other diarrheagenic Escherichia coli by the BioFire FilmArray® Gastrointestinal Panel in human fecal samples.

The purpose of this investigation was the evaluation of the performance of the BioFire FilmArray® Gastrointestinal (FA-GI) Panel, a multiplexed molecular stool screening assay, for the detection of diarrheagenic Escherichia coli (DEC), with emphasis on Shiga toxin-producing E. coli (STEC). A dilution series of 12 STEC reference strains was tested with the FA-GI Panel to assess the analytical sensitivity. A total of 389 patient samples were analyzed with the FA-GI Panel and confirmation of the detected DEC was attempted with an in-house culture-based polymerase chain reaction (PCR) method. All Shiga toxin genes, except the one encoding Stx2f, were detected in bacterial dilutions ranging from 10(4) to 10(2) colony-forming units (CFU)/ml. eae + stx2f + STEC was misclassified as enteropathogenic E. coli (EPEC). Different sensitivities for various gene targets present in one isolate led to differing identifications depending on the concentration. Using the in-house method as a reference, the FA-GI Panel had a sensitivity of 90.6 % [confidence interval (CI) 75.0 %-98.0 %] and a specificity of 97.2 % (CI 94.9 %-98.6 %) for STEC detection in feces. At least one DEC was reported in 35.5 % (171/389) of the patient specimens, with EPEC being the most prevalent (n = 71). Only 59.7 % of the detected DEC could be confirmed, presumably because the comparator method was not applied directly on feces. The FA-GI Panel could not detect the stx2f subtype, misclassified certain pathogens, and the high detection rate of EPEC needs further investigation. Nevertheless, we believe that this sensitive and convenient system will prove to be an invaluable tool for the rapid diagnosis of most DEC infections, but culturing of the detected microorganisms should always be attempted.

The influence of alcohol disinfection of nail samples in the laboratory.

Samples of two hundred finger- and toenails cultured after a 70% alcohol disinfection for 2 min were compared with non-disinfected samples cultured on a classical way. Alcohol treatment caused a significant decrease in isolation of yeast and moulds. Both organisms were cultivated at 25.5 and 31%, respectively, when not treated with alcohol, whereas the isolation rate dropped to 10 and 17.5%, respectively, after alcohol treatment. Isolation of dermatophytes decreased from 27.5 to 23.5% when alcohol disinfection was used. In conclusion, although alcohol disinfection of nail samples in the laboratory effectively decreases the isolation rate of unwanted contaminants on Sabouraud glucose agar without cycloheximide, it hampers the isolation of dermatophytes and pathogenic yeasts on cycloheximide-containing media.

Isolation of Campylobacter concisus from feces of children with and without diarrhea.

A prospective study compared fecal isolation rates of Campylobacter concisus for children with diarrhea and without diarrhea by a filter technique in which media were incubated for 4 days in a microaerobic atmosphere. No statistically significant difference in isolation rates was found (13.2% in patients with diarrhea and 9% in controls). Moreover, 35 of 37 children attending the same day care center harbored different C. concisus strains, as was demonstrated by arbitrary primer PCR DNA fingerprinting. These data suggest a lack of a pathogenic role for C. concisus in enteritis.

Results of screening for verocytotoxin-producing Escherichia coli in faeces in Belgium.

During an 18-month period all stools submitted to a microbiology laboratory in Belgium for culture were screened for Verocytotoxin-producing Escherichia coli (VTEC) serotype O157. In the stool samples from 3940 patients, eight (0.2%) VTEC O157 strains were isolated, seven of which were O157:H7. Additional screening for other serotypes of VTEC in 332 selected stool samples yielded four more strains (serotypes O2:K1:H6, O111:H-, O117:K1:H7 and O"C70/86":H-). The 0.3% isolation rate for all VTEC was comparable to that for Shigella spp. Eight children under 30 months and two adults suffered from uncomplicated gastroenteritis. A 5-month-old child and a 41-year-old woman presented with hemolytic uremic syndrome a few days after onset of a diarrheal episode.

Postpartum bacteremia and placental colonization with genital mycoplasmas and pregnancy outcome.

The influence of placental colonization and postpartum bacteremia with genital mycoplasmas on the course of delivery and the immediate postpartum period was evaluated in 511 women who gave birth to live infants of at least 26 weeks' gestation. Genital mycoplasmas were isolated from the placenta in 153 patients (29.9%) and from blood in four patients (0.8%). These four isolates were all Ureaplasma urealyticum. Patients with genital mycoplasmas isolated from the placenta were delivered of infants with birth weights and gestational ages similar to those of infants of patients who did not have genital mycoplasmas in the placenta (3260 gm and 39.2 weeks versus 3272 gm and 39.3 weeks). No adverse effects of maternal postpartum bacteremia with genital mycoplasmas were observed, either in the mother or in the baby. We conclude that, whereas genital mycoplasmas frequently can be isolated from the placenta, there is no evident relationship between the presence of genital mycoplasmas and pregnancy outcome. In a few instances U. urealyticum has been isolated from the blood of afebrile postpartum women. In these women the presence of this bacteria is probably related to the birth process. This bacteremia does not precede an infectious complication.

Serotypes of Ureaplasma urealyticum isolated from normal pregnant women and patients with pregnancy complications.

To compare the distribution of Ureaplasma urealyticum serotypes 1 to 10 in different patient populations, the serotypes of 240 U. urealyticum strains from 207 patients were determined by the indirect immunofluorescence test by using U. urealyticum antisera 1 to 10. Strains were obtained from the following four patient groups: group 1, 24 couples in which the women had a history of recurrent spontaneous abortion; group 2, 25 patients who had their first spontaneous abortion; group 3, 14 pregnant patients with pregnancy complications (premature delivery, intrauterine death); and group 4, 138 patients with uneventful pregnancies. The serotypes most often found in these 207 patients were as follows: serotype 3, 52.2%; serotype 6, 30.3%; serotype 10, 11.4%; serotype 1, 9.5%; serotype 4, 6.5%; serotype 8, 6.5%. Serotypes 2, 5, 7, and 9 were found in less than 1% of the patients. More than one serotype was found in 16.9% of the patients. The overall distribution of the 10 serotypes in the different groups was similar, except for that of serotype 4. Serotype 4 was isolated from 20.8% of the patients in group 1 and from 5.1% of the patients in group 4 (P less than 0.01). Results of this study indicate that U. urealyticum serotype 4 can be isolated more frequently from patients with a history of recurrent miscarriages than from normal pregnant women.