from regions adjacent to the teeth and that they invade the blood stream in

THE CLASSIFICATION OF NONHEMOLYTIC STREPTOCOCCI RECOVERED FROM BACTEREMIA OF DENTAL ORIGIN FREDERICK W. KRAUS,* DIANA W. CASEY, AND VILMA JOHNSON Tuft...
3 downloads 0 Views 580KB Size
THE CLASSIFICATION OF NONHEMOLYTIC STREPTOCOCCI RECOVERED FROM BACTEREMIA OF DENTAL ORIGIN FREDERICK W. KRAUS,* DIANA W. CASEY, AND VILMA JOHNSON Tufts College Dental School, Boston, Mass.

E XTRACTIONS of teeth occasionally give rise to subacute bacterial endocarditis, and we may postulate that the causative organisms originate from regions adjacent to the teeth and that they invade the blood stream in the course of the removal of teeth. Accordingly, we have attempted to compare streptococci isolated from bacteremia after dental extractions with those reported in subacute bacterial endocarditis and in bacteriologic studies of the oral cavity. The organisms recovered most frequently from the blood following dental procedures and during subacute bacterial endocarditis are nonhemolytic streptococci of the alpha and gamma types.7 22 23 The normal habitat of most members of this heterogeneous group is the oral cavity and the throat. Some nonhemolytic streptococci possess a group-specific C substance and can be classified, together with homologous hemolytic streptococci, under Lancefield Groups A to 0. The majority of nonhemolytic streptococci are designated as "viridans."5 Most of these have not yet been adequately defined for identification and are tentatively identified on the basis of physiologic characteristics. Recently three species have been more fully described, and their association with disease entities have been elucidated. They are characterized serologically by their reactions with type-specific antigens. Streptococcus MG, an inhabitant of the normal mouth repeatedly associated with primary atypical pneumonia, has been shown to elaborate a typespecific capsular polysaccharide.'6 It can be identified by a quellung reaction with specific MG antiserum and by a unique pattern in biochemical tests. Streptococcus sanguis, reported by Loewe, Candel, and Eiber,'4 as the nonhemolytic streptococcus most commonly recovered from 145 cases of subacute bacterial endocarditis, has been shown to be relatively homogeneous in physiologic tests and antigenic structure.'33' One of the distinguishing characteristics of Str. sanguis is its ability to synthesize large amounts of dextran from sucrose. Loewe, Plummer, Niven, and Sherman,13 working with 680 isolates from 20 healthy individuals, found no Str. sanguis in the throats of the subjects. In contrast to these results, Hehre'0 studied 50 to 70 isolates from each of 18 healthy adults for dextran synthesis. Twenty-eight strains from 9 of the subjects produced abundant dextran and showed biochemical reactions similar to those of Str. sanguis. Hehre and Neill9 found 9 strains of Group H streptococci which were strong dextran producers. Str. sanguis strains studied This study was supported by a grant from the King Foundation. Received for publication, Nov. 17, 1952; revised by authors Dec. 30, 1952. *Present address: University of Alabama School of Dentistry, Birmingham, Ala.

613

614

KRAUS, CASEY,

AND JOHNSON

J. D. Res.

by Washburn, White, and Niven34 were checked only against streptococcus antisera of Lancefield Groups A to G, so that it is open to conjecture whether any belonged to Group H. Streptococcus salivarius, another well-defined viridans streptococcus, is a normal inhabitant of the oral cavity. Since first described by Andrewes and Horder in 1906,1 it has been reported as a frequent causative organism in subacute bacterial endocarditis. Since Sherman, Niven, and Smiley29 limited its definition to those viridans streptococci which produce large mucoid colonies on 5 per cent sucrose agar, it has rarely been reported as a pathogen.'4' 24 With serologic tests, a Type I, a Type II, and probably other unknown typespecific antigens are found among Str. salivarius. In order to compare our microorganisms with those described in the oral cavity and with those mentioned in reports on subacute bacterial endocarditis, we used all methods available for taxonomic identification. Selbie, Simon, and Robinson25 and Solowey3' approached the comparison of oral and cardiac streptococci without actually identifying the various species. They based their serologic studies on the type-specific substances contained in nonhemolytic streptococci. Selbie, Simon, and Robinson25 completed a serologic classification of 96 strains from endocarditis cases, 103 strains from the throat and teeth, and 10 from other sources. Antisera were prepared against 8 strains of streptococci from endocarditis cases, including one Str. sanguis strain, and these were tested by the precipitin technic against extracts from the 209 strains. These workers concluded that nonhemolytic streptococci isolated from cases of subacute bacterial endocarditis produced a serologic pattern similar to those isolated from the teeth and throats. Their findings were in agreement with those of Solowey3' who charted the cross reactions between antisera and extracts from 107 endocarditis strains and 98 strains from the human throat and extracted teeth. These studies suggest a similarity between the nonhemolytic streptococci in the mouth and those isolated from the blood of endocarditis patients. However, Selbie Simon, and Robinson25 stated that "the largest group in streptococci from teeth and throats have less tendency to produce s.b.e. than the streptococci of the other four groups. " METHODS

At the beginning of this study serologic groupings were done in our own laboratory following Lancefield's precipitin technic.32 Since the majority of antisera became difficult to obtain, a large number of the strains were grouped at the Streptococcus Laboratory, Communicable Disease Center, Chamblee, Ga. We used streptococcus MG antiserum for the quellung reaction with those streptococci which exhibited the biochemical characteristics of streptococcus MG. Specific antisera for Str. salivarius and Str. sanguis could not be obtained. Among the various physiologic tests reported for the differentiation of aerobic or facultatively anaerobic streptococci, we selected those that gave consistent results when tested against known stock cultures. Hemolysis was

Volume 32 Number 5

CLASSIFICATION OF NONHEMOLYTIC STREPTOCOCCI

615

determined from poured horse blood agar plates. All tests were matched with appropriate controls. The following is a partial list of references for the procedure employed in this study: Resistance to 600 C. for 30 minutes (Shattock and Mattiek2%), hydrolysis of sodium hippurate (Hare and Colebrook,8 Coffey and Foley4), hydrolysis of arginine (Niven, Smiley, and Sherman'9), starch hydrolysis (Hodgson, Heggie, and Sutherland"), liquefaction of gelatin (Mirick, Thomas, Curnen, and Horsfall"'), splitting of esculin (Mirick, Thomas, Curnen, and Horsfall16), slime formation in 5 per cent sucrose broth (Niven, Kiziuta, and White20), mucoid colonies on Niven's agar (Niven, Smiley, and Sherman18), and growth on 10 per cent and 40 per cent bile blood agar (Belenky and Popowa2). RESULTS

In our studies of blood cultures taken immediately after extractions from

ambulatory patients of Tufts College Dental School, a high number of cases were found positive for nonhemolytic streptococci. One hundred sixty strains have been classified following Swift's chart.5 Based on Sherman's work, the entire genus Streptococcus is divided into four divisions: (1) HemolyticLancefield Groups A, B, C, E, F, G, H, K, L, 14, and 0. (2) Viridans. (3) Enterococci-Group D. (4) Lactic Group N. Of the total of 160 strains* of nonhemolytic streptococci classified, 47 (29.4 per cent) were found to belong to Lancefield Groups A to 0. Thirty-one of these were of Group K. Eight belonged to Group F, 4 to Group H, and only 1 or 2 fell into each of Groups A, C, and L. Three strains reacted with both F and K antisera and were tabulated in the group which better fitted their biochemical reactions. One strain grouping as L in our laboratory was reported by the Streptococcus Laboratory as reacting with F, K, and L antisera. No explanation for these multiple reactions was offered, but some discrepancy might be expected in testing alpha-hemolytic strains against antisera prepared with beta-hemolytic strains. In the viridans group the most common species found was Streptococcus mitis which made up 47.5 per cent of the 160 classified strains. Difficulty was encountered in fitting the other streptococci into the better-defined species. One strain was classified as Streptococcus thermophilus because it agreed with the definition of this species in all but the fermentation of maltose. Two strains were classified as Streptococcus equinus, even though one would not grow at 450 C. and one hydrolyzed sodium hippurate, Neither fermented lactose, both grew on 40 per cent bile blood agar and split esculin, and neither produced any reaction in litmus milk with 2 per cent dextrose added. Two strains were classified as streptococcus MG on the basis of biochemical tests and a positive quellung reaction with specific antiserum. Five strains were similar in many properties to Str. sanguis. They can be classified only tentatively, since no specific antisera were available for verification through serologic tests. Three additional strains were similar to Str. sanguis in hydrolyzing arginine and gelling 5 per cent sucrose broth, but *Representative strains have been preserved by freeze-drying.

J. D. Res. October, 1953

KRAUS, CASEY, AND JOHNSON

616

s

.~

o

cQ z

.

to

s~0t

bt

c

z

to r--i

I+

+-~1

+

+O

+

+

+

+

+

-%

to

z

+

+l

+lI t R z 4

zq

1+1

+1

OO

SjSaAIX

UDf)

0

NO

0

+

+1

EN tIilasa

+

+

+

+1

+1 +1

+

+1

HZO~1 asox¶dav q SOlIfuos +

+1 +

+1

+

II

+

+

0

ISOIVII31XL +1

to

asolqvw

to

asoto

0

*

+ +1

+1

0

3NINIOflI

+l +1

+

+

+1 +1 +

+I +

to to

+

I I

+

to~

+

+l

+

+1

+1 +1

+

+

+1 +1

+

+1

+i

+l

+

+ +

+

+ + +1

+

+

+1

+

+ + ++

+

+

+1

+

+

+

+

+

+

+I

II

+1

+

*

*

to

I

0

to,0d+

+l

H

to

to

11+

II0~

I

xi (9/agq0awir

to

t

)

Hlolaf asoluanls

to

01

*

+

+i

P axILOxl %v f l

to

z

+

0

z

+ +I

+1

+1 +1

+1

+1

+1

+i +

+

I+1

0~~~~0

I

s-ILflNIWN O9

'1o0

'a

09

*

-e SIS2x'IOwH c-

Suggest Documents