Colonization of Neonates in a Nursery Ward with Enteropathogenic ...

2 downloads 0 Views 883KB Size Report
May 1, 1989 - Escherichia coli and Correlation to the Clinical. Histories ... Variable numbers of enteropathogenic Escherichia coli serotype 0111:HNT strains.
JOURNAL OF CLINICAL MICROBIOLOGY, Nov. 1989, p. 2539-2543 0095-1137/89/112539-05$02.00/0

Vol. 27, No. 11

Colonization of Neonates in a Nursery Ward with Enteropathogenic Escherichia coli and Correlation to the Clinical Histories of the Children DANIEL SENERWA,"2 0RJAN OLSVIK,l* LABIUS N. MUTANDA,3 JOSEPH M. GATHUMA,2 AND KAYE WACHSMUTH4 Department of Microbiology and Immunology, The Norwegian College of Veterinary Medicine, 0033 Oslo 1, Norway'; Department of Public Health, Pharmacology and Toxicology, Faculty of Veterinary Medicine, University of Nairobi,2 and Department of Microbiology, Kenya Medical Research Institute, Nairobi, Kenya; and Enteric Bacteriology Section, Division of Bacterial Diseases, Centers for Disease Control, Atlanta, Georgia 303334 Received 1 May 1989/Accepted 4 August 1989

Stool samples were examined from 30 preterm neonates admitted to a nursery ward; 16 neonates had diarrhea, 12 constituted an age-matched control group without diarrhea, and 2 had an unknown history regarding diarrhea. Variable numbers of enteropathogenic Escherichia coli serotype 0111:HNT strains possessing the gene coding for the enteroadherence factor (EAF) were found in stool samples from 13 of the neonates. No other microbiological enteropathogen was found. A total of 294 strains (9 or 10 from each neonate, comprising 229 E. coli and 65 Klebsielka pneumoniae strains) were characterized with respect to plasmid content and grouped into 37 plasmid profile groups. Diarrhea was found not to be correlated with any specific plasmid profile or with the presence of the EAF-positive strains but rather with the number of strains with one specific plasmid profile or with the number of EAF-positive strains (of the 9 or 10 strains) isolated from each stool sample. AU the neonates who died had diarrhea (5 died of 16 with diarrhea); all five of the neonates who died possessed strains with one specific plasmid profile group, and EAF-positive strains were isolated from four of them. Of the seven neonates from whom seven or more EAF-positive isolates were isolated, three died, compared with only one of five of those from whom only a few (1 to 3 of 10) EAF-positive strains were isolated. Both plasmid profiling and genetic probing with the EAF probe were found to be good alternatives when serotyping is not available for identification of 0111:HNT enteropathogenic E. coli strains. A traditional distinguishing feature of enteropathogenic Escherichia coli (EPEC) has been the presence of certain O and H antigens, these bacteria having been shown to belong to a relatively small group of distinct serotypes (6-8, 17, 18, 23, 28). Most EPEC strains adhere to HEp-2 cells or HeLa cells in a distinct localized adherence pattern and have been referred to as class I EPEC strains (20, 26). A highly specific and sensitive enteroadherence factor (EAF) DNA probe has been used in cases of infantile enteritis to identify strains producing the EAF adhesin and localized adherence for tissue cultures (3, 21, 27; C. A. Bopp, L. K. Giljahn, V. Hundley, E. G. Sowers, T. J. Barret, J. G. Wells, and I. K. Wachsmuth, Abstr. Annu. Meet. Am. Soc. Microbiol. 1988, C113, p. 350). Outbreaks of EPEC diarrhea in the 1940s and 1950s were associated with case fatality rates as high as 70%. Risk factors for death included young age and virulence of the bacterial strain involved. Nearly all episodes of EPEC diarrhea occur in children under 2 years of age. Recent studies from several countries in North and South America and Africa have shown EPEC to be among the most important bacterial causes of diarrhea in infants (1, 4, 10, 25, 27). Prematurely born babies are immunodeficient to a certain degree and, therefore, possibly more susceptible to EPEC diarrhea than are older babies. About 50% of the children acquired hemagglutinating antibodies to EPEC strains at the age of 1 year (24). An outbreak of EPEC serotype O111:HNT occurred in a

*

Corresponding author.

hospital in Nairobi, Kenya (27). This work describes the apparent intestinal colonization of the preterm neonates in a nursery ward by Klebsiella pneumoniae and E. coli, including EAF-positive E. coli. Patterns of colonization by bacterial strains with specific plasmid profiles and strains with the EAF genes are correlated with the clinical histories of the infants. (Parts of these results were presented at the 89th Annual Meeting of the American Society for Microbiology, New Orleans, La., 14 to 18 May 1989.) MATERIALS AND METHODS Neonates. The patients were preterm neonates, 2 to 62 days old, in a nursery ward at a hospital in Nairobi, Kenya. A stool sample was taken from each of 16 neonates with diarrhea; samples were also taken at the same time from an age-matched control group of 12 neonates without diarrhea in the same ward and from 2 neonates about whom no information regarding diarrhea was available (27). Diarrhea was defined as three or more loose stools within 24 h. Strains. Ten lactose-positive colonies of enterobacteria were isolated from each of the 30 neonates and later identified to species level according to Edwards and Ewing (7). Of 300 isolates, 229 were identified as E. coli, 65 were identified as K. pneumoniae, and 6 were lost during laboratory handling. With a natural 1-megadalton (MDa) probe, a total of 78 E. coli strains were found to possess the EAF gene located on a 65-MDa plasmid. All of these strains belonged to serogroup 0111 with nontypable H antigens and adhered to HeLa cells but did not produce cytotoxins (27). None of the 294 strains were found to possess heat-labile or heat-stable 2539

2540

J. CLIN. MICROBIOL.

SENERWA ET AL.

enterotoxin (ST) I genes, as determined with enzyme-labeled synthetic oligonucleotide probes (27). Isolation and separation of plasmid DNA. Plasmid DNA was isolated, separated, and stained as described by Kado and Liu (13). Plasmid profile groups. Plasmid profile groups were constructed by grouping strains possessing the same profile (constituted of the number and molecular mass of different plasmids) or parts of a profile constituting a core profile. Bacterial strains that carried no plasmid were regarded as constituting one plasmid profile group. Infant mouse test. All of the 65 K. pneumoniae strains were also tested for production of heat-stable enterotoxin. The strains were cultivated overnight in Trypticase soy broth in a roller drum, and the sterile-filtered broth supernatant was boiled for 10 min before being used in the infant mouse test as described by Dean et al. (5). All stool specimens and strains were tested blindly throughout the study.

RESULTS All the 294 strains investigated possessed plasmids except 7 strains isolated from three neonates. Twenty-eight different plasmid profile groups for the E. coli strains and 9 different profile groups for the K. pneumoniae strains could be defined. The molecular masses of the plasmids constituting each profile group, the number of strains belonging to each group, and the number of neonates possessing strains of each group are presented in Table 1. A total of 57 different plasmids with molecular masses ranging from 1.4 to 180 MDa were observed in the 229 E. coli strains, and 21 different plasmids were observed in the 65 K. pneumoniae strains. The most frequent plasmid profiles observed were those constituting the core profile designated no. 1, having been found in 78 strains isolated from 13 neonates. These strains also harbored the EAF-encoding gene located on a 65-MDa plasmid (27). Thirty-eight of the 65 K. pneumoniae strains belonged to one core plasmid profile. These strains were isolated from seven neonates. Eight different plasmid profiles were found in single strains of K. pneumoniae. Table 2 shows the plasmid profile groups identified in the 9 or 10 strains from each neonate. The number of plasmid profiles observed in strains from a single neonate varied from one to five. Sixteen of the 30 neonates had diarrhea, and of these, 5 died. Plasmid profile group 1 EAF-positive E. coli strains were isolated from four of the five neonates who subsequently died (Table 3). Three of seven neonates from whom a large proportion (7 to 10 of 10) of EAF-positive E. coli strains was isolated died. No fatal cases were associated with the presence of K. pneumoniae strains. Diarrhea was observed in 6 of the 13 neonates from whom EAF-positive E. coli strains were isolated, compared with 10 of 17 neonates from whom such strains were not recovered. Only one neonate of the five with 1 to 3 EAF-positive strains of a total of 10 had diarrhea, compared with four of seven with a large proportion of EAF-positive strains (7 to 10 of 10) isolated from the stool samples (Table 3). One should notice that 24 of 30 neonates had strains belonging to one predominant plasmid profile group (7 or more of 10). Six of the ten neonates with stool samples from which K. pneumoniae strains were isolated had diarrhea, as did three of the six neonates from whom a large proportion of K. pneumoniae strains (7 to 10 of 10) were isolated (Table 3). Figure 1 shows that neonates who harbored strains belong-

TABLE 1. Plasmid profile groups found in 229 E. coli and 65 K. pneumoniae strains isolated from 30 neonates No. of

Plasmid profile

neonatesb

No. of strains

0 1

2 13

6 78

2

4

25

3 4 5 6 7 8 9 10 il 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 Koe Ki K2 K3 K4 K5 K6 K7 K8

3 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 7 1 1 1 1 1 1 1

21 18 9 2 1 7 7 1 1 4 1 1 1 1 1 10 1 10 9 2 1 1 2 7 1 1 38

6 6 2 1 1 1 9

Molecular mass (MDa) of plasmids

No plasmids

100d, 65, 44d, 42d 12d 7-6d 3.8d 1.9, 1.7, 1.4 100d, 68d, 62d 60d, 4.4d 4ld, 3.3, 2.9d, 1.4 130, 84, 65, 5.2, 4.4, 2.8, 1.4 160, 140, 90, 56, 6.6, 4.7 70, 67, 5.2, 4.4, 2.8, 1.4 130, 105, 83, 64 180,83,64 130,97,90,64 56 160, 92, 85, 62 140, 95, 80, 67 180, 90, 80, 5.4, 4.6, 4.1, 1.4 140, 90, 5.2, 4.8, 3.8, 1.6 140, 90 90, 80, 5.2, 4.8, 3.8, 1.4 180, 85, 60, 5.2, 4.4, 3.8, 1.4 75, 52, 5.2, 4.4, 1.4 115, 84, 67 180, 100, 7.7, 5.4, 4.1, 3.6, 1.4 95 165, 105, 78, 67, 4.3 155, 92, 85, 62 180, 155, 98, 80, 67 112, 95, 82 88, 75 85, 68, 4.4, 4.1, 3.3, 2.9, 1.4 95, 80, 60 No plasmids 180d 120, 70d, 3.3 140, 95, 85 120, 85 140, 95, 85 52 60 74, 46 100, 62, 56, 5.5, 1.4

Each plasmid profile was grouped to constitute a plasmid profile group. b Number of neonates from whom strains belonging to this plasmid profile group were isolated. C Number of strains belonging to this plasmid profile group isolated from the a

neonates. d These plasmids were not found in all the strains constituting this plasmid profile group. e The prefix K indicates plasmid profile groups from K. pneumoniae strains.

ing to a single plasmid profile group were more likely to suffer from diarrhea or die than those who had strains possessing plasmids constituting several profile groups. The mean birth weight of neonates who suffered from diarrhea was 2,171 g, compared with 1,850 g for neonates with diarrhea who later died. The neonates without diarrhea had a mean birth weight of 2,144 g. There were no statistical differences in the birth weights between these three groups of neonates. None of the neonates in this study had a birth weight below 1,000 g. DISCUSSION Chatkaemorakot et al. (3) evaluated the effectiveness of serotyping, mannose-resistant adherence to HeLa cells, and colony hybridization with the EAF probe in the identification of class I EPEC. The EAF probe was found to be more

COLONIZATION OF NEONATES WITH EPEC

VOL. 27, 1989

2541

TABLE 2. Age of neonate, duration of diarrhea, and colonization patterns determined by the plasmid profile groups in the E. coli and K. pneumoniae strains from stool samples Neonate identification no.

Age (days)

Presence of diarrhea (no. of days)a

17 22 5 15 2 8 26 13 21 7 6 1 3 19 18 29 23 16 10 30 28 24 12 14 20 25 9 il 4 27

14 20 23 24 21 19 28 35 5 21 17 10 22 5 9 3 NIe 9 14 3 62 14 15 5 9 5 29 12 22 7

2 (Dc) 5 (D) 3(D) 8(D) 3(D) 1 1 5 1 15 10 10 9 Yes Yes Yes NI NI No No No No No No No No No No No No

Plasmid profile groups (no. of indicated strains in group)b E.

coli

K. pneumoniae

Id (l) 1 (10) 1 (9) 1(3), 2(7)

4(10) 6(2),7(1), 8(7) 0 (5), 22 (2), 23 (1), 24 (1) 2(10) 20(10) 26 (7)

1 (5), 3 (2), 19 (1) 5 (5) 1 (9), 0 (1) 1 (10)

1(10) 1 (1), 5 (4), 12 (4), 17 (1) 1(2),4(8) 1 (1), 2 (7), 25 (2) 1 (1), 3 (9) 1 (7), 2 (1), 16 (1), 17 (1) 18 (10) 3(10) 21(9) 9 (7), 10 (1), 11 (1) 13 (1), 14 (1), 15 (1)

Ki K8 KO K3 Ki Ki

(3) (9) (1), Ki (9) (6), K4 (2), K5 (1), K6 (1)

K7 Ki Ki Ki

(1) (1), K2 (6) (10) (10)

(1) (4)

Number of days with diagnosed diarrhea. Yes indicates presence of diarrhea but no information on duration; No indicates absence of diarrhea. b These groups are presented in Table 1. C D, Neonate died. d Strains of E. coli plasmid profile group 1 (in boldface) possessed the EAF genes and were of the classical EPEC serotype 0111:HNT. e NI, No information.

a

specific and sensitive than either serotyping or the cell adherence test. Gomes et al. have shown that the EAF gene is present mainly in classical serotypes and infrequently in other serotypes and conclude that EAF determination alone will provide a reasonable indicator of the importance of EPEC strains associated with diarrheal diseases (lia). The presence of the EAF gene and the specific plasmid profile group correlated in this work well with the 01l1:HNT serotype, making these methods possible alternatives when serotyping is not available. The EAF-positive serotype found in the present investigation, E. coli serotype Ol11:HNT (27), belongs to a classic EPEC serogroup (17), several other serotypes in this serogroup having been incriminated as causes of infantile diarrhea in both developed and developing countries (14, 22, 25; Bopp et al., Abstr. Annu. Meet. Am. Soc. Microbiol. 1988). The 229 E. coli isolates in the present study were grouped into 28 plasmid profile groups. EAF-negative strains representing the most commonly observed plasmid profile group did not belong to any of the known EPEC serotypes and were negative for HeLa cell adherence and for production of cytotoxins (27). Plasmid profiling has been shown to be a good epidemiological tool in investigating epidemics or outbreaks of bacterial diseases (19). The plasmid content of the EAF-positive E. coli serotype 0111:HNT strains in this study might provide useful information in characterizing different clones

of these strains in different parts of the world. The EAF genes were located on a 65-MDa plasmid (27). This is in agreement with the observations of other workers who have shown that localized adhesion to HEp-2 cells is encoded by plasmids with molecular masses between 55 and 70 MDa (2, 16, 20, 21). Plasmid profiles of the E. coli strains showed genetic diversity among the strains in the samples; only 6 of 28 profile groups were found in more than one neonate, while 142 of the 229 strains belonged to only four plasmid profile groups (no. 1, 2, 3, and 4). Therefore, 24 of the 30 neonates had strains belonging to predominantly one (7 or more of 10) plasmid profile group. Neonates with E. coli strains belonging to one plasmid profile group had a higher frequency of fatality than did neonates infected with E. coli strains belonging to two to four different plasmid profile groups. Organisms inducing

disease are often found by microbiologists to be dominant or present in large numbers in the infected host in comparison to the levels regarded as normal in noninfected cases. as Comparison of the frequency of pathogenic bacteria such the EPEC strains present with the levels of other E. coli strains suspected to be a part of the normal flora could, therefore, be of importance. The results from this study show that the mortality rate was linked not only to the presence of EAF-positive strains but also to the degree of colonization one could estimate from the number of EAFpositive strains among the 10 strains examined from each

J. CLIN. MICROBIOL.

SENERWA ET AL.

2542

TABLE 3. Mortality and diarrhea rates of the neonates correlated to the colonization patterns of both EAF-positive and -negative E. coli and the K. pneumoniae strains No. of cases:

Total no. of

Strains and colonization patterns

neonates

Fatal

With diarrhea

Without diarrhea

With no available

With diarrhea Without

diarrheainformation'

EAF-positive strains (total)c 1-3 EAF positive of 10 4-6 EAF positive of 10 7-10 EAF positive of 10

13 5 1 7

4 1 0 3

6 1 1 4

5 4 0 1

2 0 0 2

EAF-negative strains (total)d

17

1

10

7

0

K. pneumoniae (total) 1-3 strains of 10 4-6 strains of 10 7-10 strains of 10

10 3 1 6

0 0 0 0

6 2 1 3

4

0 0 0 0

1 0 3

a In general, 10 strains from each patient were investigated, but because of loss of strains during laboratory handling, in some cases only 9 strains were tested. b No information concerning presence or absence of diarrhea. C EAF-positive strains of E. coli serotype 0111:HNT. d Strains of E. coli and K. pneumoniae.

sample. Only 1 to 3 of 9 to 10 EAF-positive strains were isolated from neonate fecal samples in some cases, showing the importance of investigating several colonies even when the primary cultivation plate appeared to contain a pure culture. Investigations regarding EPEC have usually involved the study of 4 to 6 E. coli strains from each sample, some including at least one lactose-negative strain (lia). Lactose-negative colonies were not examined further in this study, but the lactose-negative E. coli strains should nevertheless be considered to be of importance. Nicoletti et al. (22) found E. coli strains in neonates in Somalia which were both enteroadherent and enteroinvasive. Seven of 17 neonates with no detectable EAF-positive E. coli strains had diarrhea, as did 3 of 6 neonates from whom only K. pneumoniae strains were isolated. K. pneumoniae strains have been reported to colonize neonates (29), though no correlation to diarrheal disease has been found so far (11). K. pneumoniae strains have been associated with cases of diarrhea (9, 12), and some strains have been shown to produce heat-stable enterotoxins (15). However, the strains studied in this investigation did not possess such genes

~~~~~Died

12 _ _ 100

M M Diarrhea 0 Z No diarrhea

g

l_

z8

8

z2 1

2

3

4

Number of Different Plasmid Profile Groups FIG. 1. Relationship between the number of different plasmid profiles observed in the 9 or 10 E. coli strains isolated from a single stool sample and the clinical information concerning fatality, diarrhea, or no diarrhea. All neonates who died had diarrhea.

detectable with our synthetic probes, nor did they produce ST detectable in the infant mouse test. Known ST-producing K. pneumoniae strains have also been tested by using natural DNA probes for STIa and STIb, these probes failing, however, to hybridize with the ST genes responsible for production of this toxin in K. pneumoniae strains (Ç0. Olsvik, unpublished results). Plasmid profiling made it possible to differentiate the 65 K. pneumoniae strains into eight plasmid profile groups and 37 of these strains into one core profile. Walia et al. (30) found plasmid profiling to be superior to the antimicrobial susceptibility patterns in characterizing strains of this species in an epidemiological context. Freiman et al. (10) reported that in 40 to 60% of diarrheal cases, no pathogens are isolated. This indicates that there might be several other enteropathogenic variants among species generally regarded as nonpathogenic and that the number (e.g., of 10 examined) of different strains of a pathogen such as EAF-positive E. coli present in the intestine (as reflected by findings in the fecal sample) could play an important role in the development of diarrhea. This should be taken into consideration when establishing new sensitive and specific methods such as the use of DNA probes and when employing these in routine diagnostic laboratories. LITERATURE CITED 1. Antai, S. P., and S. O. Anozie. 1987. Incidence of infantile diarrhea due to enteropathogenic Escherichia coli. J. Appl. Microbiol. 62:227-229. 2. Baldini, J. B., J. B. Kaper, M. M. Levine, C. A. Candy, and H. W. Moon. 1983. Plasmid-mediated adhesion in enteropathogenic Escherichia coli. J. Pediatr. Gastroenterol. Nutr. 2:534-538. 3. Chatkaemorakot, A., P. Echeverria, D. N. Taylor, K. A. Bettelheim, N. R. Blacklow, O. Sethabutr, J. Seriwatana, and J. Kaper. 1987. HeLa cell-adherent Escherichia coli in children with diarrhea in Thailand. J. Infect. Dis. 156:669-672. 4. Cravioto, A., R. E. Reyes, R. Ortega, G. Fernandez, R. Hernandez, and G. Lopez. 1988. Prospective study of diarrhoeal disease in a cohort of rural Mexican children: incidence and isolated pathogens during the first two years of life. Epidemiol. Infect. 101:123-134. 5. Dean, A. G., Y. C. Ching, R. G. Williams, and L. B. Harden. 1972. Test for Escherichia coli enterotoxin using infant mice: application in a study of diarrhea in children in Honolulu. J. Infect. Dis. 125:407-411.

VOL. 27, 1989

6. Edelman, R. E., and M. M. Levine. 1983. Summary of workshop on enteropathogenic Escherichia coli. J. Infect. Dis. 147:11081118. 7. Edwards, P. R., and W. H. Ewing. 1972. Identification of Enterobacteriaceae, 3rd ed. Burgess Publishing Co., Minneapolis. 8. Ewing, H. W., H. W. Tatum, and D. R. Davis. 1957. The occurrence of Escherichia coli serotypes associated with diarrhea diseases in the U.S. Public Health Lab. 15:1251-1258. 9. Finegold, S. M. 1970. Interaction of antimicrobial therapy and intestinal microflora. Am. J. Clin. Nutr. 23:1466-1477. 10. Freiman, I., E. Hartman, H. Kassel, R. M. Robins-Browne, R. D. Schoub, H. J. Koornhof, G. Lecatsas, and O. W. Proozesky. 1977. A microbiological study of gastro-enteritis in black infants. S. Afr. Med. J. 52:261-265. 11. Goldman, D. A. 1988. The bacterial flora of neonates in intensive care-monitoring and manipulation. J. Hosp. Infect.

11(Suppl):340-351. 11a.Gomes, T. A. T., M. A. M. Vieira, I. K. Wachsmuth, P. A. Blake, and L. R. Trabulsi. 1989. Serotype-specific prevalence of Escherichia coli strains with EPEC adhesion factor genes in infants with and without diarrhea in Sao Paulo, Brazil. J. Infect.

Dis. 160:131-136. 12. Gross, R. J. 1984. Acute enteritis, p. 458-476. In G. R. Smith (ed.), Topley and Wilson's principles of bacteriology, virology and immunology, vol. 3. Bacterial diseases. Butler and Tarner Ltd., London. 13. Kado, C. I., and S. T. Liu. 1981. Rapid procedure for detection and isolation of small and large plasmids. J. Bacteriol. 145: 1365-1373. 14. Karch, H., J. Heeseman, and R. Laufs. 1987. Phage-associated cytotoxin production by, and enteroadhesiveness of, enteropathogenic Escherichia coli isolated from infants with diarrhea in West Germany. J. Infect. Dis. 155:707-715. 15. Klipstein, F. A., and R. F. Engert. 1976. Purification and properties of Klebsiella pneumoniae heat-stable enterotoxin. Infect. Immun. 13:373-381. 16. Laporta, M. Z., M. L. Silva, I. C. A. Scaletsky, and L. R. Trabulsi. 1986. Plasmids coding for drug resistance and localized adherence to HeLa cells in enteropathogenic Escherichia coli. Infect. Immun. 51:715-717. 17. Levine, M. M. 1987. Escherichia coli that cause diarrhea: enterotoxigenic, enteropathogenic, enteroinvasive, enterohemorrhagic and enteroadherent. J. Infect. Dis. 155:377-389. 18. Levine, M. M., and R. Edelman. 1984. Enteropathogenic Escherichia coli of classic serotypes associated with diarrhea: epi-

COLONIZATION OF NEONATES WITH EPEC

2543

demiology and pathogenesis. Epidemiol. Rev. 6:31-51. 19. Mayer, L. M. 1988. Use of plasmid profiles in epidemiologic surveillance of disease outbreaks and in tracing the transmission of antimicrobial resistance. Clin. Microbiol. Rev. 1:228-243. 20. Nataro, J. P., K. O. Maher, P. Mackie, and J. Kaper. 1987. Characterization of plasmids encoding the adherence factor of enteropathogenic Escherichia coli. Infect. Immun. 55:23702377. 21. Nataro, J. P., I. C. A. Scaletsky, J. B. Kaper, M. M. Levine, and L. R. Trabulsi. 1985. Plasmid-mediated factors conferring diffuse and localized adherence of enteropathogenic Escherichia coli. Infect. Immun. 48:378-383. 22. Nicoletti, M., F. Superti, C. Conti, A. Calconi, and C. Zagaglia. 1988. Virulence factors of lactose-negative Escherichia coli strains isolated from children with diarrhea in Somalia. J. Clin. Microbiol. 26:524-529. 23. Orskov, F., and I. Orskov. 1984. Serotyping of Escherichia coli, p. 43-112. In T. Bergan (ed.), Methods in microbiology, vol. 14. Academic Press, Inc. (London), Ltd., London. 24. Parker, T. M. 1984. Hospital-acquired infections, p. 192-224. In G. R. Smith (ed.), Topley and Wilson's principles of bacteriology, virology and immunology, vol. 3. Bacterial diseases. Butler and Tarner Ltd., London. 25. Paulozzi, J. L., K. E. Johnson, L. M. Kamahele, R. W. Riley, and S. D. Helgerson. 1986. Diarrhea associated with enteropathogenic Escherichia coli in an infant and toddler center, Seattle, Washington. Pediatrics 77:296-300. 26. Scaletsky, C. A. I., M. Lourdnes, M. Silvia, and L. R. Trabulsi. 1984. Distinctive patterns of adherence of enteropathogenic Escherichia coli to HeLa cells. Infect. Immun. 45:534-536. 27. Senerwa, D., 0. Olsvik, L. N. Mutanda, K. J. Lindqvist, J. M. Gathuma, K. Fossum, and K. Wachsmuth. 1989. Enteropathogenic Escherichia coli serotype 0111:HNT isolated from preterm neonates in Nairobi, Kenya. J. Clin. Microbiol. 27:13071311. 28. Toledo, M. R. F., M. C. B. Alvariza, J. Murahovschi, S. R. T. S. Ramos, and L. R. Trabulsi. 1983. Enteropathogenic Escherichia coli serotypes and endemic diarrhea in infants. Infect. Immun.

39:586-589. 29. Tullus, K., B. Berglund, B. Fryklund, I. Kuhm, and L. G. Burman. 1988. Epidemiology of fecal strains in the family Enterobacteriaceae in 22 neonatal wards and the influence of antibiotic policy. J. Clin. Microbiol. 26:1166-1170. 30. Walia, S., T. M. Madhavan, T. Reuman, R. Tewari, and D. Duckworth. 1988. Plasmid profiles and klebocin types in epidemiological studies of infections by Klebsiella pneumoniae. Eur. J. Clin. Microbiol. Infect. Dis. 7:279-284.