Carriage rates of enterococci in the dental plaque of haemodialysis patients in Dublin

British Journal of Oral and Maxillofacial Surgery (1987) 25, 21-33 @ 1987 The British Association of Oral and Maxillofacial Surgeons

CARRIAGE RATES OF ENTEROCOCCI IN THE DENTAL PLAQUE OF

HAEMODIALYSIS PATIENTS IN DUBLIN

CYRIL J. SMYTH,* B.s~., Ph.D., M.A., DOCENT (S.U.A.S.), M. KENNETH HALPENNY,? B.D.s., F.D.S.R.C.P.S. (Glasg.), F.F.D.R.C.S. (I), SARAH J. BALLAGH,” B.A. (MOD.)

*Department of Microbiology, Moyne Institute, Trinity College, and fDepartment of Oral Surgery, Dublin Dental Hospital and School, Lincoln Place, Dublin 2,

Republic of Ireland

Summary. The incidence of carriage of enterococci in dental plaque was determined in haemodialysis patients attending three clinics in the Dublin area either as outpatients or as hospitalised patients. Their carriage rates were compared with a University group comprising normal healthy students, academic staff, technicians and ancillary personnel and with a cohort of otherwise healthy toothache patients. The carriage rates among the staffs of the dialysis units also were examined. The overall carriage rates of enterococci of the University group, the toothache patient group, the haemodialysis patients and the dialysis unit staffs did not differ significantly from each other, ranging from 5%-20%. However, the dental plaque of a mainly hospitalised group of haemodialysis patients and their attendant staff at one clinic was colonised to a statistically significant higher degree with enterococci than that of the haemodialysis patients and the staff at the outpatient clinics, both separately and as combined patient and staff groups. Age, sex, a history of recent antibiotic therapy, and elapsed time since the last dental visit did not affect isolation rates to a significant extent. The commonest enterococcus isolated from subjects was Streptococcus faecalis, followed by its variety liquefaciens. Only one subject harboured Streptococcus durans in dental plaque. Ten of the 21 subjects yielding enterococci harboured two different enterococci in their plaque. The isolation of S. faecalis var fiquefaciens alone or in combination with S. faecaks did not correlate with subject-history parameiers. The findings obtained imply that antibiotic prophylaxis specifically against enterococci may be necessary only for a small number of haemodialysis patients in whom oral carriage of enterococci has been demonstrated bacteriologically.

Introduction

Enterococci may be defined as Lancefield group D streptococci which are bile-tolerant, salt-tolerant and aesculin-hydrolysing (Facklam, 1972, 1980; Shul- man, 1979). The term enterococci includes Streptococcus faecalis and its varieties liquefaciens and zymogenes, S. faecium and S. durans. Although the transfer of these bacteria to a new genus Enterococcus has been proposed (Schleifer & Kilpper-Bslz, 1984; Collins et al., 1984), this nomenclature has not as yet been ratified and, accordingly, that still in common usage is adopted herein.

Enterococci can be isolated from the normal human mouth, but usually only represent a minor proportion of the total streptococcal flora (Hardie & Bowden, 1974). The incidence of enterococci in the oral cavity has varied from O--.50% depending on the oral source of specimens tested (saliva, root canals, plaque, tonsils, sutures) and the populations studied (normal subjects, dental clinic patients, laboratory personnel, high caries group, antibiotic-treated patients, school children, endodontic patients, naval recruits, chronic hospital patients and hospital staff) (Williams et al., 19.50; Bahn et al., 1960; Engstrdm, 1964; Gold et al., 1975; MejZxe, 1975; Sukchotiratana et al., 1975; Phillips et al., 1976; Campbell et al., 1983).

Some clinical conditions appear to predispose to oral carriage of enterococci. For

(Received 9 November 1985; accepted 20 February 1986)

21

22 BRITISH JOURNAL OF ORAL & MAXILLOFACIAL SURGERY

example, 34% of 64 naval recruits tested during an epidemic of infection with an upper respiratory tract virus yielded enterococci from saliva samples compared with 3.6% of 443 normal healthy naval recruits (x2 test, P<O.OOl) (Bahn et al., 1960). An association has also been demonstrated between the oral incidence of S. faecalis and the use of prophylactic antibiotics in oral surgical patients (Phillips et al., 1976).

In a comprehensive review of 4821 published cases of infective endocarditis in which an alleged source of infection was identified (McGowan, 1982), dental procedures were implicated in 13.2% of these episodes, extraction being the commonest precipitating factor. Enterococci were implicated as the causative organism of infective endocarditis following dentistry in only eight out of 346 cases from 1950-1975 (McGowan, cited in Campbell et al., 1983). In other reports, enterococci have been responsible for .5-X% of cases of endocarditis (Porter- field, 1950; Mandell, 1976; Parker & Ball, 1976; Moulsdale et al., 1979; 1980). Moreover, enterococcal endocarditis has been reported in six haemodialysis patients, four following dental treatment (Goodman et al., 1969; Wyler et al., 1972; Cross & Steigbigel, 1976). Infected root canals have also yielded a high incidence of enterococci (Winkler & van Amerongen, 1959; Moller, 1966; Engstriim, 1964; Mejare, 1975).

The present investigation was stimulated by the report of Campbell et al. (1983)) who showed that the carriage rate of enterococci in the dental plaque of a group of long-term haemodialysis and cardiac hospital patients, was statistically significantly higher than in groups of normal subjects comprising staff at the dialysis clinic and acute dental patients. The enterococcal isolates were also unusual in that nine out of 11 were S. fuecium, whereas the predominant enterococcus in the oral cavity in other reports has been S. faecalis and its varieties (see references above).

While the number of haemodialysis patients in the Glasgow study was small (Campbell et al., 1983) and the reports of enterococcal endocarditis following dental treatment of such patients are few, the high incidence of carriage of enterococci in dental plaque in the Glasgow group of haemodialysis patients coupled with (i) the known antibiotic resistance patterns of enterococci (Kaye, 1982), (ii) the fact that S. faecium is more resistant to penicillin than S. faecalis (Kaye, 1982), (“‘) 111 an apparent enhanced risk of infective endocarditis in chronic haemodialysis patients (Cross & Steigbigel, 1976), and (iv) the inclusion of chronic haemodialysis patients and dialysis unit staff from only one particular hospital clinic in the Glasgow study, prompted a survey in the three haemodialysis units in Dublin to assess the carriage rates of enterococci in dental plaque by patients and staff compared with a group of acute dental outpatients and a group comprising students, academic staff, and laboratory and ancillary personnel at Dublin University.

Materials and methods

Subjects

Details of the subject groups are given in Table I. The undergraduate (24) and postgraduate (10) students, academic staff (8) and laboratory and ancillary personnel (8) of the Department of Microbiology, Trinity College were included as a ‘normal’ subject population who had a higher degree of exposure to micro- organisms, including enterococci, than individuals in society at large. The patient group attending Dublin Dental Hospital were all outpatients with a complaint of toothache.

ENTEROCOCCI IN DENTAL PLAQUE 23

The haemodialysis patients were in attendance at the dialysis units of the Meath, Jervis Street, and St. Mary’s (Phoenix Park) Hospitals, Dublin. To preserve anonymity, these hospitals have each been allocated randomly an index capital letter and are referred to hereafter solely in such terms. At Hospitals D and X, all of the subjects were outpatients, whereas at Hospital H the patients were mainly inpatients with acute problems, although a few had been on haemodialysis as outpatients for some time.

All subjects or their guardians gave informed cotisentto the dental manipulation required to examine their mouths in order to assess their oral hygiene status and to obtain dental plaque samples. All had natural, upper and lower, left second premolars and first molars, and lower central incisors. For each subject, a questionnaire was completed which recorded information on name and address, age, sex, any current antibiotic therapy or such therapy within the preceding 3 months, history of drug addiction or hepatitis, cardiac and vascular status and medical history, number of months since last known dental visit, oral hygiene status, number of months on haemodialysis, areas sampled for plaque, and an encoded reference number for ‘blind’ analysis and strain identification. Oral hygiene status was assessed according to a modification of the method of Green and Vermillion (1964): excellent = no oral plaque or stain; good = plaque covering not more than one-third of the tooth surface; fair = plaque covering one-third to two-thirds of the tooth surface; poor = plaque covering greater than two-thirds of the tooth surface; very poor = calculus deposits on teeth. Potential subjects with a history of jaundice, coronary thrombosis or a cardiovascular accident were excluded from the Microbiology Department and Dublin Dental Hospital groups. The medical histories of the dialysis unit staffs were similarly negative. None of the subjects wore dentures.

Collection of mixed dental plaque samples

The end of a sterile, disposable, precontoured plastic spatula (Johnson & Johnson, Dublin) was used to collect plaque from the lingual aspect of the lower left, second premolar and first molar of all subjects. A sterile 1 yl plastic bacteriological loop (Nunc, Copenhagen) was wetted in sterile peptone water, filled with the plaque from the tip of the spatula and shaken in 0.5 ml of peptone water (5% w/v peptone) in a bijou bottle. The sampling procedure was repeated on the buccal aspect of the upper left, second premolar and first molar using the other end of the spatula, and the same loop and bijou bottle. Each sample bottle thus contained 2 ~(1 of mixed plaque from each subject.

For 18 of the Microbiology Department or toothache patient groups, 21 haemodialysis patients and 11 dialysis unit staff, additional plaque samples were taken using the buccal aspect of the same upper teeth, but using the lingual aspect of the lower central incisors, to provide 2 ~1 of mixed plaque as used by Campbell et al. (1983). Samples were held at 4°C and usually inoculated onto isolation media within 2-4 h. A few mixed plaque samples had to be held overnight, but this procedure did not appear to affect the chance of isolation of enterococci. Moreover, control inocula onto blood agar plates as tests of viability yielded abundant growth of plaque bacteria.

Culture media and isolation of enterococci

The Selective Enterococcus (SE) agar medium of Isenberg et al. (1970), as

24 BRITISH JOURNAL- OF ORAL & MAXILLOFACIAL SURGERY

modified by Campbell et al. (1983) to contain 1.5 g bile salts instead of 10 g/l, and MacConkey agar medium No. 2 (Oxoid) were used for the primary isolation of enterococci. The mixed dental plaque samples were agitated vigorously for 30 s on a Whirlimixer (Fisons Scientific, Loughborough) and 0.25 ml aliquots spread on each of the selective media. For 70 samples chosen at random, 0.15 ml aliquots of each dispersed plaque sample were spread onto the two selective media and onto a 5% (v/v) human blood agar plate as a bacterial viability control.

Plates of the selective media were inoculated with S. faecalis strain ATCC 19433 as a control for each batch of agar plates. Agar plates were incubated aerobically at 37°C for 48-60 h. Pin-point colonies on the SE medium which were surrounded by a black zone, indicating aesculin hydrolysis, were subcultured onto SE agar, MacConkey agar and human blood agar. Identical subculturing was performed on typical enterococcal colonies on MacConkey agar.

Identification of presumptive enterococcal isolates

Isolates yielding typical colonies on SE, MacConkey and blood agars were identified as enterococcal group D streptococci on the basis of Gram stain, bile-aesculin hydrolysis, growth in 6.5% (w/v) NaCl broth (salt tolerance), and serogrouping using the Phadebact @ Strep D test (Facklam, 1972; 1980; Cowan, 1974; Pharmacia Diagnostics, 1983). The coaglutination test was performed according to the manufacturer’s instructions using bacteria grown in Todd-Hewitt broth.

Identification of the enterococcal isolates was made on the basis of a selection of physiological and biochemical tests: pyruvate fermentation (Gross et al., 1975), growth in the presence of tellurite (0.25 mg.ml-’ of Diagnostic Sensitivity Test [DST] agar (Oxoid); production of acid from arabinose, glycerol, lactose, mannitol, raffinose, salicin, sorbitol, sucrose and trehalose, using broth-based sugars with bromocresol purple as the indicator (Edwards & Ewing, 1972); arginine hydrolysis; gelatin liquefaction using nutrient gelatin deeps and Charge1 discs (Oxoid) on DST agar slopes, and the haemolysis pattern on human blood agar (Cowan, 1974; Facklam, 1980). S. f aecalis strain ATCC 19433 was used as a control organism in all tests. All incubations were at 37°C for up to 2 weeks.

Other bacteria

In the course of these studies, pin-point colonies were obtained on SE and MacConkey No. 2 agars which appeared at first sight to be enterococcal. Gram stains revealed Gram-positive, rod-to-oval shaped bacteria, often in chains. Disconcertingly, these bacteria reacted strongly with the Phadebact@ Strep D test reagent. Because of these findings, identification tests were performed by standard procedures (Buchanan & Gibbons, 1974; Cowan, 1974).

Statistical analysis

The incidences of isolation of enterococci from dental plaque between groups were compared using 2x2 contingency tables, applying Fisher’s test when the total number of observations was <50 or a x2 test (with Yates correction on tables with a total <lOO or with any cell containing a value less than 10) (Swinscow, 1983; Langley, 1979). In some instances, the standard error of difference between percentages was also calculated (Swinscow, 1983). Comparisons between dialysis unit staff and other groups were with female subpopulations of the latter (Table I).

Tabl

e I

Prof

iles

of

the

subj

ect

popu

latio

n st

udie

d

Hos

pita

l/ D

epar

tmen

t

Mic

robi

olog

y,

Trin

ity

Col

lege

D

ublin

Dub

lin

Den

tal

Hos

pita

l

Hos

pita

l D

Hos

pita

l X

Hos

pita

l H

Hos

pita

l D

Hos

pita

l X

Hos

pita

l H

Cat

egor

y N

o.

of

subj

ects

A

ge

Ran

ge

Mea

n M

edia

n R

ange

M

ean

Med

ian

Ran

ge

Mea

n M

edia

n

Staf

f an

d st

uden

ts:

Fem

ale

Mal

e To

otha

che

patie

nts:

Fe

mal

e M

ale

Dia

lysi

s ou

tpat

ient

s:

Fem

ale

Mal

e D

ialy

sis

outp

atie

nts:

Fe

mal

e M

ale

Dia

lysi

s patie

ntsc

D

ialy

sis

unit

staf

f (F

emal

e)

Dia

lysi

s un

it st

aff

(Fem

ale)

D

ialy

sis

unit

staf

f (F

emal

e)

30

2&44

25

20

21

-45

30

23

12-4

9 26

33

12

-49

26

15

1653

33

16

63

Y 17

-59

12

17-5

2

8 16

62

15

25-4

1

6 25

-37

5 26

-28

31

35

36

33

38

31

29

27

Mon

ths

sinc

e pr

evio

us

Mon

ths

on

know

n de

ntal

vi

sit”

haem

odia

lysi

s

22

1-84

9

25

I-16

8 13

26

1-22

8 13

25

l-1

80

15

32

l-240

Y

32

l-2

04

13

29

l-336

8”

31

l-2

40

17

37

l-120

6

29

1-48

11

28

l-24

10

27

6-18

10

7 13

12

18 6 12

48

36 7 7 7 Y

- : z

3-72

29

19

u

3-10

0 33

23

: el

9

3-51

18

11

r

3-11

0 23

9

; 2-

115

39

4 * 0 c m

“Ind

ivid

uals

w

ithin

gr

oups

w

ho

had

not

had

a de

ntal

vi

sit

in a

t le

ast

4 ye

ars

wer

e ex

clud

ed

from

ca

lcul

atio

ns

of m

eans

as

the

ac

cura

cy

with

w

hich

th

e pe

rson

co

uld

reca

ll da

tes

was

ca

lled

into

qu

estio

n.

‘Fiv

e w

omen

ha

d to

be

ex

clud

ed

as

defin

ed

unde

r fo

otno

te

(a).

‘Fem

ale:

mal

e=2:

6.


Results

Prevalence of enterococci in dental plaque of subjects

In all, 21 of the 209 subjects (10.0%) yielded enterococci from dental plaque. Their distribution among the subject populations is shown in Table II. Of the haemodialysis patients, 10 of 77 (13.0%) yielded enterococci. This incidence was not statistically significantly different from that of the University group or the toothache patient group or both combined. In the case of dialysis unit staffs, five out of 26 (19.2%) carried enterococci in dental plaque. Comparisons of carriage rates between this female subject group and women in the University, toothache patient and haemodialysis patient groups revealed no statistically significant differences. No differences in carriage rates were apparent between the University and toothache patient groups.

The haemodialysis patients were age- and sex- matched with individuals from the University and the toothache patient groups, combined as one ‘control’ group for this purpose. Each group thus contained 65 subjects with a male to female ratio of 43:22 and an age range of 16-59 years. Nine of the haemodialysis patients and six of the controls yielded enterococci from their dental plaque. The difference was not significant. A comparison of carriage rates between age-matched women among the dialysis unit staffs, haemodialysis patients, and combined University and toothache patient groups revealed no statistically significant differences. Each age-matched cohort comprised 18 subjects aged between 16 and 49 years. The carriage rates were 27.7% for the dialysis unit staff and 16.7% for each of the age-matched haemodialysis patient and combined control groups.

Inter-hospital comparisons were made for the haemodialysis patients and dialysis unit staffs. The male to female ratios and age distributions within the haemodialysis patient populations were similar. The differences in carriage rates between the Hospital H patients and each of the other two groups of haemodialysis patients were significant (Fisher’s test, P<O.Ol). The difference in carriage rates between the Hospital H patients and the outpatient groups combined was highly significant (x2, Yates Corr.=12.84, P<O.OOl; standard error of differences between percentages, P<O.OOl). The differences in enterococcal carriage rates between the staff of the dialysis unit at Hospital H and that of the Hospital D clinic or that of the Hospital D and Hospital X clinics combined were significant (Fisher’s test,

Table II Incidence of enterococci in dental plaque of different subject groups

Hospital/Department

Microbiology, TCD Dublin Dental Hospital Hospital D Hospital X Hospital H

No. of subjects yielding enterococci per group

University Toothache Haemodialysis Dialysis patients patients unit staff

(n=50) (n=56) (n=77) (n=26)

3 (6.0)” - - -

- 3 (5.4) T(6.3) 1 (6.7) 2 (9.5) 1 (16.7)

- 5 (62.5) 3 (60)

“Data in parentheses represent the percentages of subjects from each hospital or department yielding enterococci.


P<O.O5). The difference in the carriage rates between the staffs at the Hospital X and Hospital H clinics was not significant. Within each clinic, the proportions of female haemodialysis patients yielding enterococci were O/15, l/9 and l/2 for Hospitals D, X and H, respectively (c.f. dialysis staff, Tables I and II),, Ignoring the sex distributions of patients and staff, there were no differences in carriage’rates between staff and patients within each of the three clinics.

In contrast, analysing the data from the general hospital in the Glasgow study of Campbell et al. (1983) in an identical manner to those herein, the difference in carriage rates between the haemodialysis patients (606) and the dialysis unit staff (O/23) was significant (Fisher’s test, P<O:Ol). 7 : / -

a

Subject characteristics and incidence of enterococcal carriage

Differences between the incidences of carriage between sexes within an age group and between age groups within the sexes were not statistically significant. The overall rates of oral enterococcal carriage with age group were fairly even (630 years, 9.0%; 31-40 years, 11.4%; 41-63 years, 12.9%).

The incidence of enterococcal carriage as a function of the oral hygiene status of subjects was examined. None of the differences was statistically significant. Interestingly, however, more than half of the 21 subjects yielding enterococci had a good to excellent oral hygiene status. The elapsed period since last dental visit ranged from l-168 months among these individuals (median 11 m).

The question of whether current or previous antibiotic therapy might predispose to carriage of enterococci in dental plaque was also considered. There were no significant differences in the incidences of carriage of enterococci between antibiotic-treated subjects and those who had received no such treatment in the 3-month period prior to sampling of dental plaque. Among haemodialysis outpatients, of those with no immediate history of antibiotic therapy, 4/61 (6.6%) carried enterococci in dental plaque compared with l/8 (12.5%) of those who had received antibiotics; the difference was not significant. Because the overall incidence of enterococci in the antibiotic-treated subjects (4/28) was not significantly higher than in the untreated subjects (17/181), no attempt was made to assess the effect of particular antibiotic usage.

Of the 77 haemodialysis patients, only one of eight with a history of a congenital heart disorder and one of 10 with a history of coronary thrombosis and/or a cardiovascular accident yielded enterococci from dental plaque.

Influence of isolation medium and of mixed plaque sampling site on detection of enterococci

In three subjects, the enterococcal isolates were recovered from MacConkey No. 2 agar only; these isolates did, however.,. grow on the SE agar on subculture. The number of colony forming units of aesculin-hydrolysing bacteria on the SE agar medium could only be determined in eight, instances, the range being 1-18.7x lo3 c.f.u./ml of mixed plaque sample; the viable counts were too high to determine from the primary plates in the other instances.

On occasion when aesculin-hydrolysing colonies obtained on SE agar and small, intensely red colonies obtained on MacConkey No. 2 agar were subcultured onto SE agar, MacConkey agar, human blood agar and nutrient agar, they failed to grow on the latter two media and grew somewhat more slowly on the former two media than


typical enterococci. These were shown to be. isolates of Gram-positive, rod-to-oval shaped bacteria often in chains. Biochemical investigation of such isolates identified them as Lactobacillus plantarum. Moreover, before the Gram morphology of these isolates was realised, some had been found to coaglutinate with the Phadebact@ Strep D test reagent, a fact that was reconfirmed sub- sequently on several occasions.

Of the 50 randomly chosen individuals who were screened initially using two different mixed plaque samples, enterococci were isolated from seven. In three of these instances, the mixed plaque samples taken from the areas used by Campbell et al. (1983) failed to yield enterococci, whereas in the other four cases enterococcal isolates were obtained from both mixed plaque samples. On this basis, upper and lower, second premolar and first molar plaque samples only were obtained from the majority of the subjects.

Identification of enterococcal isolates and their relationships to subject groupings

Thirty-one isolates were obtained from the 21 subjects yielding enterococci. All except one of the isolates (viz. from subject 15006) were S. faecalis or its variety liquefaciens (Table III). The remaining isolate was S. durans (S. faecium var. durans). Twelve of the subjects harboured S. faecalis var. liquefaciens. Ten of the 21 individuals harboured more than one species, variety or biotype in their mouths. Seven had S. faecalis with S. faecalis var. liquefaciens, one had S. durans with S. faecalis, and two had two S. faecalis biotypes which differed with respect to at least five biochemical and physiological markers in an extended study (Smyth et al., unpublished data). The remaining 11 subjects harboured single strains of S. faecalis or S. faecalis var. liquefaciens. Five of the 10 haemodialysis patients yielded two types of enterococci, as did three of the six subjects in the combined University/ Dublin Dental Hospital group and two of the five dialysis unit staff who had enterococci in their dental plaque. Thus, there was no association between multiple types of enterococci and sublect group.

There were no statistically significant differences in the frequencies of isolation of S. faecalis var. liquefaciens from enterococci-positive haemodialysis patients (five out of 10) compared with enterococci-positive persons in the combined University/ Dental Hospital group (five out of six) or dialysis unit staffs combined (two out of five). Dividing the haemodialysis patients into two groups on the basis of duration of dialysis therapy for <12 or >12 months, there was no apparent relationship between the type of enterococcal isolate and the length of time on dialysis. Dividing the 21 subjects yielding enterococci into two groups on the basis of elapsed time since last dental visit, viz. during the previous 12 months and in excess of 12 months, 5/12 (41.7%) and 7/9 (77.8%), respectively, harboured S. faecalis var. liquefaciens. The difference was not, however, statistically significant.

Two of the four subjects with a history of antibiotic therapy in the preceding months harboured S. faecalis var. liquefaciens. There was no association between the sex of the subject and the occurrence of S. faecalis var liquefaciens in dental plaque. Virtually equal proportions of the men and women harboured more than one species, variety or biotype of enterococcus. Dividing the subjects harbouring enterococci on the basis of age into a group <30 years and a group >30 years, 6112 and 619, respectively, had S. faecalis var. liquefaciens in their dental plaque; within the same age groups, 4112 and 319, respectively, harboured the S. faecalis and S. faecalis var. liquefaciens combination. None of the differences was significant.

Tabl

e II

I Pr

ofile

s of

sub

ject

s yi

eldi

ng

ente

roco

ccal

is

olat

es

from

de

ntal

pl

aque

in

rel

atio

n to

the

id

entif

icat

ion

of t

hese

st

rept

ococ

ci

Subj

ect

Age

R

ef.

no.8

1101

8 41

1102

5 21

11

027

24

1201

9 23

1203

8 27

1205

4 49

13

002

38

1300

8 32

1304

3 59

1400

2 27

15

006

23

1501

0 20

16

005

26

1700

1 16

1700

2 62

17

003

35

1700

4 38

17

009

39

1800

3 26

1800

4 26

1800

5 27

Sex

M

F M

M

F F M

M

M

F F M

F F M

M

F M

F F F

Ant

ibio

tic

Ora

l th

erap

y pr

evio

us

hygi

ene

3 m

onth

s st

atus

-

- - - - - Eryt

hrom

ycin

- - - - G

enta

mic

in

- - Flox

acill

in

Peni

cilli

n C

loxa

cilli

n - - - - - - A

mox

ycill

in

Poor

Goo

d G

ood

Ver

y po

or

Goo

d

Goo

d G

ood

Goo

d

Poor

Fair

Goo

d

Poor

G

ood

Fair

Fair

Goo

d Fa

ir Fa

ir G

ood

Exce

llent

Exce

llent

Mon

ths

sinc

e C

ardi

o-

last

de

ntal

va

scul

ar

visi

t st

atus

’

15

-

72

- 5

- 84

-

12

-

1 -

24

-

60

CT,

C

VA

168

-

12

- 1

-

48

CV

A

7 -

6 -

120

CH

D

3 -

12 5

- 6

-

18

-

12

-

Mon

ths

on

haem

odia

lysi

s

- - 8 81

18

26

11 4 84

115 2 5

Ente

roco

ccal

is

olat

es’

S. f

aeca

lis

S. f

aeca

lis

var.

lique

faci

ens

6. f

aeca

lis

var.

lique

faci

ens

S. f

aeca

lis

S. f

aeca

lis

S. f

aeca

lis

var.,

liq

uefa

cien

s S.

fae

calis

S.

fae

calis

va

r. liq

uefa

cien

s S.

fae

calis

va

r. liq

uefa

cien

s S.

fae

calis

S.

fbe

calis

var

. liq

uefa

cief

ls

S. f

aeca

lis

S. f

aeca

lis.’

S. f

aeca

lis

S. f

aeca

lis

var.

lique

faci

ens

S. f

aeca

lis

S.

dura

ns

S. f

aeca

li.?

S. f

aeca

lis”

S. f

aeca

lis

var.

lique

faci

ens

S. f

aeca

lis

S. J

aeca

lis

var.

lique

faci

ens

S. f

aeca

lis

var.

lique

faci

ens

S. f

aeca

lis

var.

lique

faci

ens

S. ,

faec

alis

S.

fae

cali

s S.

fae

calis

S.

fae

calis

* S.

fae

calis

S.

fae

calis

va

r. liq

uqfa

cien

s S.

fae

calis

“Cod

es

begi

nnin

g w

ith

Il=U

nive

rsity

, 12

=Dub

lin

Den

tal

Hos

pita

l, 13

/14=

Hae

mod

ialy

sis

patie

nts

and

Dia

lysi

s un

it st

aff,

Hos

pita

l D

, re

spec

tivel

y;

15/1

6=H

aem

odia

lysis

pa

tient

s an

d D

ialy

sis

unit

staf

f, re

spec

tivel

y,

Hos

pita

l X

; 17

/18=

Hae

mod

ialy

sis

patie

nts

and

Dia

lysi

s un

it st

aff,

Hos

pita

l H

, re

spec

tivel

y.

bCT=

Cor

onar

y th

rom

bosi

s;

CV

A=c

ardi

ovas

cula

r ac

cide

nt;

CH

D=c

onge

nita

i he

art

diso

rder

. “S

. fa

ecal

is”

indi

cate

s is

olat

es

of

this

sp

ecie

s w

ith

a di

ffer

ent

phys

iolo

gica

l an

d bi

oche

mic

al

phen

otyp

e fro

m

othe

r S.

Ja

ecal

is

isol

ates

lis

ted.


Discussion

In contrast to the study of Campbell et al. (1983) which pointed to a higher carriage rate of enterococci in the mouths of chronic dialysis patients compared to the mouths of dialysis unit staff and control toothache patients, the present study did not indicate a statistically significant higher incidence of enterococci in a group of dialysis patients five times larger than in the Glasgow investigation. However, a small group of mainly hospitalised dialysis patients attending the clinic of one Dublin hospital (H) carried enterococci in dental plaque at a significantly higher incidence than a cohort of haemodialysis patients attending another two clinics as outpatients. Indeed, the majority of the enterococcal isolates in the study of Campbell et al. (1983) were from hospitalised patients (Macfarlane, personal communication). In contrast to the Glasgow report, the dialysis unit staff attending to the hospitalised dialysis patients were also significantly colonised compared to dialysis unit staffs in the outpatient units, who resembled the staff of the dialysis unit in Glasgow in terms of carriage rates.

Unlike the Glasgow cohort of haemodialysis patients who had received a variety of antibiotic regimens, only 13% of the haemodialysis patients in the Dublin cohort had a history of antibiotic therapy in the 3-month period prior to sampling. However, these haemodialysis patients were not exposed to antibiotics any more than the normal individual (14.2% of the University and toothache dental patients combined). Antibiotic administration alone could not account for the marked incidence of enterococci among the hospitalised haemodialysis patient group. However, given that 803 (61.5%) of staff and patients at the Hospital H clinic carried enterococci in dental plaque compared with 7/90 (7.8%1 of staff and patients at the outpatient clinics of two other Dublin hospitals (x , Yates Corr. =24.0 P<O.OOl), antibiotics in the environment or some other environmental factor(s) may have contributed to acquisition of enterococci in the mouth by the oral-faecal route or from saliva by oral communication between staff and patients or between patients.

Haemodialysis outpatients with a history of antibiotic treatment did not have a higher incidence of oral carriage of enterococci compared to outpatients with no history of antibiotic chemotherapy in the preceding 3 months or to the control University and toothache groups. This is in line with the findings of Sukchotiratana et al. (1975) who reported an absence of S. faecalis in the oral flora of normal volunteers given penicillin or clindamycin and lincomycin, although relatively resistant strains of oral bacteria became predominant within a few days. In contrast, 11 out of 47 patients, who had received similar antibiotic regimens prophylactically just prior to and following total extraction or extraction of third molars, had S. faecalis isolated from sutures removed from their gums 5-7 days after dental treatment, although none of their extracted teeth yielded enterococci (Phillips et al., 1976).

Another aspect of the hospitalised haemodialysis patients was that their general health appeared to be poorer than that of the majority of the haemodialysis subjects attending outpatient clinics. Indeed, reduced resistance, as occurs after upper respiratory tract infections, appears to predispose to carriage of enterococci (Bahn et al., 1960). Thus, their decreased health status and the inevitable close contact between staff and patients may have permitted spread and colonisation with enterococci. Whether food, water, cleaning staff, kitchen staff and other workers possibly contributed could only have been ascertained by an epidemio- logical investigation.


The culture procedure used in the present study was modified from that of Campbell et al. (1983) in two respects, viz. the inclusion of MacConkey No. 2 agar in addition to SE agar as a selective medium and the use of mixed plaque samples from a slightly different oral area. Both of those modifications contributed to isolation efficiency. The use of the Campbell et al. (1983) modification of the Selective Enterococcus agar of Isenberg et al. (1970) had one drawback in terms of selectivity, namely that the reduced concentration of bile salts allowed the ready growth of oral lactobacilli on this nutrient-rich medium. L. planturum will grow in media containing 4% bile salts (Buchanan & Gibbons, 1974). The original medium of Isenberg et al. (1970) contains 1% bile salts compared with 0.15% in the Glasgow modification. L. plunturum also tolerates up to 4% NaCl and 250 mg.l-’ of sodium azide.

Antibodies to the glucosyl-ribitol teichoic acid from L. plantarum cross-react with the membrane-associated glycerol teichoic acid of group D streptococci because of the common immunodominant a-n-glucosyl substituents (Knox & Wicken, 1973), which confer the group D specificity of the grouping antigen. This accounts for the reactivity of the Phadebact@ Strep D test reagent with L. plunturum isolates in the present study.

Additionally, L. plunturum hydrolyses aesculin. Thus, counting aesculin- hydrolysing colonies on SE agar may not give a viable count solely of enterococci in mixed plaque samples (Campbell et al., 1983). Indeed, Williams et al. (1950) reported that if enterococci were present in saliva, there was likely to be a higher count of lactobacilli than if enterococci were absent.

As in previous reports on the incidence of enterococci in the oral cavity (Williams et al., 1950; Winkler & van Amerongen, 1959; Gold et al., 1975), the order of frequency of occurrence of the various species of enterococci was S. fueculis, S. fuecalis var. liquefaciens and S. durans. In contrast, S. faecium was the dominant species in the Glasgow study (82% of enterococcal isolates; Campbell et al., 1983). On the contrary, S. fuecium is the most frequently isolated enterococcus from human faeces followed by S. fueculis (Unsworth, 1979). The high incidence of S. faecium in the Glasgow study may have been related to hospitalisation and the particular antibiotic regimens in use in these subjects, but unfortunately it has not proved possible to investigate this aspect retrospectively (Macfarlane, personal communication). However, faecal origin of the S. faecium isolates is likely as it was not uncommon to isolate Escherichiu coli and other enterobacteria from the oral flora of such patients in Glasgow (Macfarlane, personal communication). The possibility of wrong identification seems unlikely as the identification protocol used in the Glasgow study was essentially identical to that used in the present study.

Patients on haemodialysis are regarded as a high risk group with respect to bacterial endocarditis because of the presence of arterio-venous shunts (Goodman et al., 1969; Cross & Steigbigel, 1976; McGowan, 1982). The results obtained herein imply that antibiotic prophylaxis specifically against enterococci in association with oral surgery or dental manipulation may only be necessary for a small number of haemodialysis patients. If possible, the plaque content of enterococci should be assessed prior to endodontic therapy or extraction. Penicillin in conjunction with an aminoglycoside is generally recommended for prophylaxis against enterococci (Barry & Thornsberry, 1980; Tullman & Redding, 1982), gentamicin being usually preferred because ribosomal resistance to gentamicin is extremely rare in enterococci and synergy between the two antibiotics can be demonstrated with virtually all strains of enterococci (Foster, 1983; Ispahani, 1983).


Acknowledgements

The authors thank students, staff, technicians and ancillary staff of the Department of Microbiology, staff and patients of the Dublin Dental Hospital, haemodialysis patients and dialysis unit staffs at the Jervis Street, Meath and St. Mary’s (Phoenix Park) Hospitals for their voluntary cooperation during this study; Dr T. W. Macfarlane, Glasgow Dental Hospital, for providing additional information about the study of enterococcal carriage in haemodialysis patients in Glasgow; Dr Diarmuid Shanley for his enthusiasm; the preparation room staff of the Moyne Institute, in particular Henry Kearns, for the exhausting supplies of media; and Mairead Daly for excellent secretarial service.

References

Bahn, A. N., Shklair, I. L. Mazzarella, M. & Calandra, J. C. (1960). Incidence of oral group D streptococci. Journal of Dental Research, 39, 686.

Barry, A. L. & Thornsberry, C. (1980). Approximate minimal inhibitory concentrations of more susceptible strains among common bacterial pathogens. In E. H. Lennette, A. Balows, W. J. Hausler, Jr. & J. P. Traunt (editors), Manual of Clinical Microbiology, 3rd Ed., p. 498. Washington, D. C.: American Society’ for Microbiology.

-.

Buchanan. R. E. & Gibbons. N. E. (1974). Beraev’s Manual of Determinative Bacteriolonv, 8th Ed.. pp. 576-593. Baltimore: The W’illiams and Wilkins Company.

YI

Campbell, J., McGowan, D. A. & Macfarlane, T. W. (1983). The prevalence of enterococci in the dental plaque of chronic hospital patients. British Journal of Oral Surgery, 21, 171.

Collins, M. D., Jones, D., Farrow, J. A. E., Kilpper-Balz, R. & Schleifer, K. H. (1984). Enterococ- cus avium nom. rev., comb. nov.; E. casseliflavus nom. rev., comb. nov.; E. durans nom. rev., comb. nov.; E. gallinarum nom. rev., comb. nov.; and E. malodoratus sp. nov. International Journal of Systematic Bacteriology, 34, 220.

Cowan, S. T. (1974). Cowan and Steel’s Manual for the Identification of Medical Bacteria, 2nd Ed., pp. 51-55. Cambridge: Cambridge University Press.

Cross, A. S. & Steigbigel, R. T. (1976). Infective endocarditis and access site infections in patients on hemodialysis. Medicine (Baltimore), 55, 453.

Edwards, P. R. & Ewing, W. H. (1972). Identification of Enterobacteriaceae, 3rd Ed. Minneapolis: Burgess Publishing Company.

Engstrom, B. (1964). The significance of enterococci in root canal treatment. Odontologisk Revy, 15, 87.

Facklam, R. R. (1972). Recognition of group D streptococcal species of human origin by biochemical and physiological tests. Applied Microbiology,- 23, 1131. _

Facklam. R. R. (1980). Streotococci and aerococci. In E. H. Lennette. A. Balows, W. J. Hausler, Jr. & J. P. Truant (editors): Manual of Clinical Microbiology, 3rd Ed., p. 88. Washington, D.C.: American Society for Microbiology.

Foster, T. J. (1983). Plasmid-determined resistance to antimicrobial drugs and toxic metal ions in bacteria. Microbiological Reviews, 47. 361.

Green, J. C. & Vermillion, J. R. (1964). The simplified oral hygiene index. Journal of the American Dental Association, 68, 7.

Gold, 0. G., Jordan, H. V. and van Houte, J. (1975). The prevalence of enterococci in the human mouth and their pathogenicity in animal models. Archives of Oral Biology, 20, 473.

Goodman, J. S., Crews, H. D., Ginn, H. E. & Koenig, M. G. (1969). Bacterial endocarditis as a possible complication of chronic hemodialysis. New England Journal of Medicine, 280, 876.

Gross, K. C., Houghton, M. P. & Senterfit, L. B. (1975). Presumptive speciation of Streptococcus bovis and other group D streptococci from human sources by using arginine and pyruvate tests. Journal of Clinical Microbiology, 1, 54.

Hardie, J. M. & Bowden, G. H. (1974). The normal microbial flora of the mouth. In F. A. Skinner & J. G. Carr (editors), The Normal Microbial Flora of Man, Symposium Series No. 3, Society for Applied Bacteriology, p, 47. London: Academic Press.

Isenberg, H. D., Goldberg, D. .& Sampson, J. (1970). Laboratory studies with a selective enterococcus medium. Applied Microbiology, 20, 433.

Ispahani, P. (1983). Bacteraemia and endocarditis. In D. Greenwood (editor), Antimicrobial Chemotherapy, p, 214. London: Bailliere Tindall.

Kaye, D. (1982). Enterococci: biologic and epidemiologic characteristics and in vitro susceptibility. Archives of Internal Medicine, 142, 2006.

Knox. K. W. & Wicken. A. F. (1973). Immunological properties of teichoic acids. Bacteriological Reviews, 37,, 21 J.

Langley, R. (1979). Practical Statistics Explained, 2nd Ed., pp. 269-317. London: Pan Books.


Mandell, G. L. (1976). Enterococcal endocarditis. In D. Kaye (editor), Infective Endocarditis, p. 101. Baltimore: University Park Press.

McGowan, D. A. (1982). Endodontics and infective endocarditis. International Endodontic Journal, 1.5, 127.

Mejare, B. (1975). On the facultative anaerobic streptococci in infected dental root canals in man. Odontologisk Revy, 26, Supplement 33, 5.

Moller, A. J. R. (1966). Microbiological examination of root canals and periapical tissue of human teeth. Odontologisk Tidskrift, 74, No 5-6, special article.

Moulsdale, M. T., Eykyn, S. & Phillips, I. (1979). Infective endocarditis 197@78: A review of 81 patients with particular reference to streptococci. In M. T. Parker (editor), Pathogenic Streptococci, Proceedings of the VII International Symposium on Streptococci and Streptococcal Diseases, p. 229. Chertsey, England: Reedbooks Ltd.

Moulsdale, M. T., Eykyn, S. J. & Phillips, I. (1980). Infective endocarditis 1970-79: A study of culture positive cases at St. Thomas’s Hospital. Quarterly Journal of Medicine, 49, 315.

Parker, M. T. & Ball, L. C. (1976). Streptococci and aerococci associated with systemic infection in man. Journal of Medical Microbiology, 9, 275.

Pharmacia Diagnostics (1983). Phadebact@ Strep D Test: Directions for Use, pp. 1-16. Uppsala: Pharmacia Diagnostics, AB.

Phillips, I., Warren, C.: Harrison J. M., Sharples, P., Ball, L. C. & Parker, M. T. (1976). Antibiotic susceptibilities of streptococci from the mouth and blood of patients treated with penicillin or lincomycin and clindamycin. Journal of M&cQ~ Microbiology, 9, 393.

Porterfield, J. S. (1950). Classification of the streptococci of subacute bacterial endocarditis. Journal of General Microbiology, 4, 92.

Schleifer, K. H. & Kilpper-Balz, R. (1984). T ransfer of Streptococcus faecalis and Streptococcus faecium to the genus Enterococcus nom. rev. as Enterococuss faecalis comb. nov. and Enterococcus faecium comb. nov. International Journal of Systematic Bacteriology, 34, 31.

Shulman, J. A. (1979). Streptococcus faecalis and other enterococci. In G. L. Mandell, R. G. Douglas, Jr. & J. E. Bennett (editors), Principles and Practice of Infectious Diseases, p. 1605. New York: John Wiley & Sons.

Sukchotiratana, M., Linton, A. H., Fletcher, J. P. (1975). Antibiotics and the oral streptococci of man. Journal of Applied Bacteriology, 38, 277.

Swinscow, T.D.V. (1983). Statistics at Square One, 3rd Ed., pp. 28-32 & 43-53. London: British Medical Association.

Tullman, M. J. & Redding, S. W. (1982). Systemic Disease in Dental Treatment, pp. 23-29. New York: Appleton-Century-Crofts.

Unsworth, P. F. (1979). Streptococci in human faeces. In M. T. Parker (editor), Pathogenic Streptococci, Proceedings of the VII International Symposium on Streptococci and Streptococcal Diseases, p. 239. Chertsey, England: Reedbooks Ltd.

Williams, N.B., Forbes, M. A., Blau, E. & Eickenberg, C. F. (1950). A study of the simultaneous occurrence of enterococci, lactobacilli, and yeasts in saliva from human beings. Journal of Dental Research, 29, 563.

Winkler, K. C. & van Amerongen, J. (1959). Bacteriologic results from 4000 root canal cultures. Oral Surgery, Oral Medicine, Oral Pathology, 12, 857.

Wyler, D. J., Golde, D. W. & Grausz, H. (1972). Bacterial endocarditis in a patient with a saphenous vein graft A-V fistula receiving dental work. California Medicine, 117, 75.

Documents

Carriage rates of enterococci in the dental plaque of haemodialysis patients in Dublin