Search

Haemophilus influenzae isolates recovered from the genitourinary (GU) tract were shown to have a significantly different biotype distribution compared with respiratory tract isolates. Biotype IV strains were recovered more commonly from the GU tract, and most strains were non-serotypable. Antibiotic-susceptible strains isolated from the GU tract more frequently harbored plasmids of less than 10 megadaltons than did antibiotic-susceptible respiratory tract strains. One 2.8-megadalton plasmid resident in a GU tract isolate and one 1.8-megadalton plasmid resident in a respiratory tract isolate were shown to be related to the small ampicillin resistance plasmids previously described in H. influenzae, Haemophilus parainfluenzae, Haemophilus ducreyi, and Neisseria gonorrhoeae. This supports the suggestion that these ampicillin resistance plasmids originated by transposition or recombination of the ampicillin transposon (TnA) with cryptic endogenous Haemophilus plasmids.

Based on a limited number of biochemical properties, a system for biotyping Haemophilus influenzae (M. Kilian, Acta Pathol. Microbiol, Scand. Sect. B82:835-842, 1976) was used to analyze the relationship of biotype to source of infection and antibiotic resistance for 600 clinical strains. The distribution of biotypes from bacteremic patients was significantly different (P less than 0.001) from the distribution of biotypes from nonbacteremic patients. Although there appeared to be a correlation between biotype and source of isolation, no single biotype correlated with a specific clinical syndrome in bacteremic patients. The frequency of resistance to antibiotics (ampicillin, tetracycline, chloramphenicol, and kanamycin), which was known to be at least in part plasmid mediated, was determined. Of the 600 isolates, 43 were resistant to at least one antibiotic (30 were ampicillin resistant, 11 were tetracycline resistant, 1 was ampicillin-tetracycline resistant, and 1 was tetracycline-chloramphenicol resistant). Of these 43 resistant isolates, 42 were either biotype I or II. This distribution of biotypes among antibiotic-resustant isolates was significantly different from the overall distribution of biotypes (P is less than 0.001).

During epidemiological studies carried out in urban and rural areas of the midwestern Canadian province of Manitoba, we cultured enterotoxigenic Escherichia coli (ETEC) from 16 (1.7%) of 945 diarrheal stools and 4 (0.3%) of 1,282 normal stools. ETEC was found in not more than 2.3% of diarrheal stools obtained from any population during any season. Diarrhea associated with ETEC persisted for a mean of 9 days. Two children were dehydrated and required intravenous fluid therapy, and one adult suffered a cholera-like syndrome. Half of the children required hospitalization for management of their diarrhea. Two adults and two children who harbored ETEC were completely asymptomatic. The pattern of toxin production correlated with serotype and the serotypes encountered were (with a few exceptions) similar to those found in other areas. We conclude that ETEC is an uncommon cause of diarrhea, both in rural and urban areas of central Canada. However, the possibility that ETEC might cause severe sporadic cases or epidemics of gastroenteritis remains.

ABSTRACTPrimers were designed to amplify sequences of verocytotoxin genes and eaeAgenes of Escherichia coliO26:H11, O111:H8, and O157:H7 in a multiplex PCR assay. This assay successfully detected E. coliO26:H11 in bloody stool specimens in which other enteric pathogens were not detected by culture-based methods. Rapid assays to detect non-O157:H7 verocytotoxin-producingE. coliis important to improve methods for the etiologic diagnosis of hemorrhagic colitis.