J Vet Diagn Invest 5:363-367 (1993)

Streptococcus suis infection in swine: a retrospective

study of 256 cases. Part I. Epidemiologic factors
and antibiotic susceptibility patterns
Rachel Y. Reams, Lawrence T. Glickman, Daniel D. Harrington,
Terry L. Bowersock, H. Leon Thacker

Abstract. A retrospective study of 256 cases of naturally acquired Streptococcus suis infections in swine
submitted to the Indiana Animal Disease Diagnostic Laboratory from 1985 to 1989 was performed to determine
the epidemiologic factors and antibiotic susceptibility patterns associated with S. suis serotypes 1-8 and 1/2.
A standardized computer form was used to record the history, signalment, and clinical signs obtained from the
records of selected cases and the microscopic lesions identified after review of the histopathology slides for each
case. A computer statistics package (SAP) was used to evaluate the data. Although the number of recovered
S. suis isolates increased in the fall and winter months, most serotypes were readily isolated throughout the
year; only serotypes 1, 4, 7, and 1/2 increased in frequency of isolation in the fall, winter, and spring months.
The majority (6 1.1%) of infected pigs in this study were < 12 weeks of age. More than 75% of pigs infected
with serotypes 1, 6, 7, and 1/2 were < 12 weeks of age. There was extensive overlap in the age distributions for
pigs with each serotype, and statistically significant differences for most serotypes were not observed. Fifty
percent of pigs infected with S. suis serotypes 1 and 1/2 were 3-10 weeks of age, 50% of pigs infected with
serotype 2 were 6-14 weeks of age, and 50% of pigs infected with serotypes 3, 4, 5, 7, and 8 were 2-16 weeks
of age. Isolates of S. suis were not uniformly susceptible to penicillin, and a large percentage of isolates were
resistant to many antibiotics in common usage. The results of this study indicated that the various serotypes
of S. suis could not be readily separated based on antibiograms, epidemiologic factors (herd size, breed, etc.),
or geographic location.

Streptococcus suis is a significant pathogen of human
beings and swine that was first isolated during disease
outbreaks in swine in the early 1950s in Europe.5,13 In
the last 20 years, reports of clinical disease associated
with S. suis infection have markedly increased. The
organism has been recovered from swine, other animal
species, and human beings in various countries
throughout the world. 7,10,12,15,19 To date, at least 28 dif-
ferent serotypes of S. suis have been identified,8 al-
though many laboratories routinely screen only for se-
rotypes 1-8 and 1/2. (Isolates of S. suis serotype 1/2
contain capsular antigens common to both serotype 1
and serotype 2 and therefore will react with antibody
to both serotype 1 and serotype 2 in coagglutination
and other tests).
Streptococcus suis serotypes 1 and 2 are the most
completely characterized, and in the USA they are pri-
marily associated with meningitis in neonatal pigs and
suppurative bronchopneumonia in pigs 3-12 weeks of
age.4,20 Serotypes 3-8 and 1/2 of S. suis have not been
From the Indiana Animal Disease Diagnostic Laboratory (Reams,
Harrington, Thacker) and the Department of Veterinary Pathobiol-
ogy (Reams, Glickman, Harrington, Bowersock), Purdue University,
West Lafayette, IN 47907.
Received for publication August 21, 1992.
363
as extensively studied, and there are few published
reports concerning the epidemiologic factors, clinical
signs, and gross or microscopic lesions associated with
these serotypes in naturally infected swine. The goals
of this study were to describe the epidemiology and
pathology of S. suis serotypes 1-8 and 1/2.
Materials and methods
Identification of S. suis isolates. Specimens submitted from
diseased swine were cultured on 5% ovine blood agar plates
at 37 C in 5% CO2 for 24-48 hr. Colonies of catalase-negative
gram-positive cocci exhibiting or ' hemolysis were subin-
oculated onto 5% ovine blood agar plates and incubated for
24 hr in 5% CO2 at 37 C. Resulting bacterial isolates were
then serotyped using the coagglutination test. Reference strains
of capsular serotypes 1-8 and 1/2a (reference strain numbers
A228, D930, 4961, 6467, 4638, 2524, 8074, 14636, and
2851, respectively) were used as described previously to pre-
pare the antibodyb used for the coagglutination test.1 The
coagglutination test used was similar to that described pre-
viously,1 but the inactivation step was decreased from an
overnight incubation to a 30-min incubation. Briefly, suspect
colonies and colonies of reference strains were suspended in
0.02 M phosphate-buffered saline with 0.3% formalin for 30
min. The resulting bacterial suspensions were mixed 1:1 with
each of the 8 antisera on a ringed glass plate and rocked
manually for 1-2 min. Agglutination was detected with a
364 Reams et al.
Figure 1. Relative frequency of 256 Streptococcus suis isolates
recovered from porcine submissions to the Indiana Animal Disease
Diagnostic Laboratory, 1985-1989.
bright lamp against a black background. Suspect isolates were
identified by rapid (5-15 sec) agglutination in the presence
of antibody. Isolates exhibiting agglutination with antibody
prepared against reference strains of capsular serotype 1 and
2 or with the reference strain of 1/2 were designated 1/2.
Antimicrobial susceptibility. During the 4-yr study period,
antibiotic susceptibilities were determined by tube and mi-
crobroth dilution, using automated systems.c,d,e Samples were
prepared and processed following the manufacturers guide-
lines using commercially prepared reagents for each system.
Antibiotic susceptibilities were reported as susceptible, in-
termediately susceptible, moderately susceptible, or resistant
using minimum inhibitory concentration guidelines16,17 for
each category. The following bacterial strains were used for
quality control of the antibiograms: Escherichia coli, 25922;
Pseudomonas aeruginosa, 27853; Streptococcus faecalis,
29212; Staphylococcus aureus, 29213.
Case selection. The preliminary study group composed of
955 accessions contained all porcine accessions submitted to
the Indiana Animal Disease Diagnostic Laboratory from No-
vember 1, 1985 to December 31, 1989 in which S. suis was
isolated and identified as described above. The final study
Table 1. Percentage of 236 Streptococcus suis isolates from swine submissions to the Indiana Animal Disease Diagnostic Laboratory,
1985-1989, that were susceptible to commonly used antibiotics.
group was limited to 256 (26.8%) accessions that met the
following criteria: 1) tissues had been collected for micro-
scopic examination, 2) all laboratory results and histopa-
thology slides were individually labelled, 3) only 1 pig in the
accession was infected with S. suis, and 4) only 1 serotype
of S. suis was isolated from the infected pig.
Data collection and entry. A computer data form was used
to enter information obtained from each case record con-
cerning age and breed of the infected pigs, other epidemio-
logic factors, antibiograms, and gross and microscopic le-
sions. Data collected in this manner was then transferred to
SAS f for descriptive statistics and other analyses.
This paper is limited to discussion of the distribution of
serotypes of S. suis in naturally infected swine and the as-
sociated epidemiologic factors and antibiotic susceptibilities.
The clinical signs and gross and microscopic lesions observed
in these swine and the frequency of isolation of other patho-
gens will be reported separately.
21
Statistical analysis. Duncans multiple range test ( =
0.05) was used to evaluate differences in pig age between the
serotypes after deletion of values greater than the mean
2.75 SDS, square-root transformation of the age, and confir-
mation of the homogeneity of the variances for each serotype
(Bartletts test).22 Principal analysis (Pearsons correlation co-
efficient, under the null hypothesis) was used to determine
the association of various epidemiologic factors with sero-
type. The distribution of the isolates in the preliminary study
group was compared with the final study group by the chi-
square test.
Results
Distribution of isolates. The relative frequency of
isolates of each serotype included in the study is shown
in Fig. 1. This distribution did not differ significantly
2
(x = 5.4, df = 9, P = 0.80) from that of the total 955
isolates in the preliminary group (data not shown).
The county of origin was known for 237 (92.6%) of
256 submissions. Streptococcus suis was isolated from
swine submissions from 57 (62.0%) of 92 counties in
Indiana and from submissions from Illinois (7) and
Missouri (1). Indiana isolates were most often (69.9%)
 Streptococcus suis infection in swine. Part I. 365

Figure 2. Monthly distribution of 256 isolates of S. suis (sero-

types 1-8 and 1/2) submitted to the Indiana Animal Disease Di-
agnostic Laboratory, 1985-1989.

from counties with a swine population of 50,000

head. Submissions were made by 100 different refer-
ring practitioners. Accessions were submitted from a
total of 240 different farms. Two hundred twenty-five
(93.8%) of the farms had a single submission during
the 4-year period, whereas multiple submissions (2-4
per farm) were submitted by the remaining 15 farms.
Antibiograms. Antibiograms were recorded for 236
(92.2%) of the 256 isolates. The most frequently tested
antibiotics are listed in Table 1. The percentage of
isolates susceptible to various antibiotics was extreme-
ly variable among the serotypes.
Season. The seasonal isolation rate for the S. suis
isolates in the study was similar to that of the prelim-
inary group (data not shown), with the lowest recovery
rates in the months of July, August, and September
(5.9%, 5.1%, and 4.3%, respectively) and the highest
recovery rates in the fall and winter months (Fig. 2).
Table 2. Age data for pigs naturally infected with Streptococcus
suis serotypes 1-8 and 1/2.
Figure 3. Box plots of the age distribution for 226 isolates of S.
suis from swine submissions to the Indiana Animal Disease Diag-
nostic Laboratory, 1985-1989. The age range (minimum to maxi-
mum) for each serotype is represented by the capped lines, the x
denotes the mean age, and the box encloses the interquartile range.
Only serotypes 1, 4, 7, and 1/2 had similar decrease
in isolation rate in the summer months (data not
shown). Other serotypes failed to demonstrate a sea-
sonal pattern. Serotype 2 had a constant monthly re-
covery rate throughout the calendar year, ranging from
7.7% to 10.8%.
Age. The majority (61.1%) of affected pigs in this
study were < 12 weeks of age. More than 75% of all
pigs infected with serotypes 1, 6, 7, and 1/2 were < 12
weeks of age. The mean ages of the affected pigs are
listed in Table 2. The pig age distributions for each
serotype are shown in Fig. 3.
The mean age of pigs infected with serotype 6 was
significantly lower (Duncans new multiple range test,
a = 0.05) than that of pigs infected with other sero-
types, whereas the mean age of pigs infected with se-
rotype 2 was significantly greater ( = 0.05) than that
of pigs infected with serotypes 1, 1/2, 5, 6, and 7. There
was extensive overlap in mean age of pigs infected with
366 Reams et al.
serotypes 3, 8, 4, 5, 7, and 1/2, and no statistically
significant differences in mean age of pigs infected with
these serotypes were observed.
Breed. Two hundred sixteen (94.3%) of the 229 pigs
in which the breed had been noted were crossbred pigs.
There were 5 Duroc and 4 Yorkshire pigs.
Herd size, group size, number affected, and number
dead. Pigs submitted came from herds ranging in size
from 14 to 15,000 pigs. Herd size was weakly corre-
lated with the size of the clinically affected group (r2
= 0.28, P < 0.01) and with the number of ill pigs (r2
= 0.41, P < 0.01); larger herds had larger groups of
affected pigs and more total pigs affected. Serotype was
not correlated with herd or group size, with the total
number of pigs affected, or with the number of dead
pigs per affected farm.
Discussion
The frequency distribution, by serotype, of the 256
isolates of S. suis included in the final study group did
not differ significantly from that of the preliminary
group of 955 isolates. Therefore the distribution of the
isolates in the final study group was considered rep-
resentative of the isolates recovered from diseased swine
at the Indiana Animal Disease Diagnostic Laboratory.
As in other studies, S. suis serotype 2 was isolated more
frequently than any other serotype.6,11,23 Marked vari-
ability in the frequency distribution of serotypes is
common among laboratories within the same country
and among countries. This variability most likely re-
flects a real difference in geographical distribution.
However, misidentification of serotypes, cross-reactiv-
ity among the various serotypes, and variation in the
quality and source of the antibodies used for capsular
serotyping4,18 probably contribute to this variability.
Throughout the 4-year study period, the antibiotic
panels used by the laboratory for aerobic swine bac-
terial isolates were changed frequently to reflect changes
in federal regulations, field usage, etc. In the past, S.
suis was reported to be uniformly susceptible to pen-
icillin 9,20,23 However, as in a recent study,2 many of
the isolates in this study were resistant to penicillin
and other antibiotics in common usage, such as tet-
racycline. As the prevalence of infection with S. suis
and the number of herds with enzootic infections in-
creases, the continual use of various antibiotics may
lead to selection of resistant strains of S. suis.
Previous investigations 11,20 found different patterns
of seasonal fluctuations for S. suis infection. The results
of this study indicate that seasonal influence is negli-
gible, with only a mild increase in the number of iso-
lates recovered during the fall, winter, and early spring
months for some serotypes.
The age distribution for pigs infected with S. suis
serotypes 1 and 2 have been reported previously. In
general, serotype 1 affects pigs 2-10 weeks of age,14,24
and serotype 2 affects pigs 3-12 weeks of age but has
been recovered from growing pigs of all ages.3,24 Our
results agree with those of these previous reports.
The majority of samples obtained were collected from
crossbred swine because of the large number of com-
mercial feeder pig and finishing operations in Indiana.
Because of the small number of purebred pigs exam-
ined, no conclusion could be drawn regarding breed
disposition.
There was a wide range in herd size for the study
pigs. In general, larger swine herds had larger groups
of affected pigs with a greater number of clinically ill
or dead pigs per affected group. This finding may reflect
the widespread use of large confinement operations in
Indiana. Such large groups may promote exposure of
susceptible swine to healthy carrier animals infected
with S. suis.
3,19,24
Use of intensive confinement systems
has been previously associated with increased rates of
infection with S. suis in swine.3
The various serotypes of S. suis examined could not
be readily separated from one another based on fre-
quency of isolation, antibiograms, or epidemiologic
factors. Serotypes 1-8 and 1/2 were isolated from dis-
eased swine throughout the state of Indiana. No ob-
vious geographic distribution pattern of isolates was
evident, and S. suis was recovered throughout the year
for most serotypes, with only a few serotypes showing
an increased rate of isolation in the fall, winter, and
spring. Isolates of S. suis in this study were resistant
to penicillin and other commonly used antibiotics.
The only epidemiologic factor that differed signifi-
cantly among the various serotypes was age of affected
pigs. In general, most swine infected with S. suis were
< 12 weeks of age, but sporadic cases in older pigs,
including adults, were observed. In this study, S. suis
serotypes 1 and 1/2 affected pigs 3-10 weeks of age,
serotype 2 affected pigs 6-14 weeks of age, and sero-
types 3-5, 7, and 8 affected pigs 2-16 weeks of age.
Acknowledgements
R. Y. Reams thanks Mrs. Nita Glickman for her assistance
with formulation of the computer data forms used in this
study and Mrs. Rose Begthol-Hunt for the excellent figures.
Sources and manufacturers
a. S. Rosendal, Department of Veterinary Microbiology and Im-
munology, Ontario Veterinary College, University of Guelph,
Guelph, Ontario.
b. C. A. Armstrong, ADDL, Purdue University, West Lafayette,
IN.
Autobac , Organon Teknika, Durham, NC.
Sensititre , Radiometer America, Westlake, OH.
Vitek, Vitek Systems, Hazelwood, MO.
SAS , SAS Institute, Gary, NC.
 Streptococcus suis infection in swine. Part I.
References
1. Armstrong CH, Janas LY, Barkdull JA: 1986, A comparison
of coagglutination and the DMS Rapid Strep System as methods
of identifying Streptococcus suis. Proc Int Pig Vet SOC 9302.
2. Cantin M, Hare1 J, Higgins R, Gottschalk M: 1992, Antimi-
crobial resistance patterns and plasmid profiles of Streptococcus
suis isolates. J Vet Diagn Invest 4: 170-174.
3. Clifton-Hadley FA: 1984, Studies of Streptococcus suis type 2
infection in pigs. Vet Res Commun 8:217-227.
4. Erickson ED, Doster AR, Pokomy TS: 1984, Isolation of Strep-
tococcus suis from swine in Nebraska. J Am Vet Med Assoc
185:666-668.
5. Field HI, Buntain D, Done JT: 1954, Studies on piglet mor-
tality. I. Streptococcal meningitis and arthritis. Vet Rec 66:453-
455.
6. Galina L, Collins JE, Pijoan C: 1992, Porcine Streptococcus
suis in Minnesota. J Vet Diagn Invest 4:195-196.
7. Gottschalk M, Higgins R, Jacques M, et al.: 1989, Description
of 14 new capsular types of Streptococcus suis. J Clin Microbiol
27:2633-2636.
8. Gottschalk M, Higgins R, Jacques M, et al.: 1991, Character-
ization of six new capsular types (23 through 28) of Streptococcus
suis. J Clin Microbiol 29:2590-2594.
9. Hariharan H, Bryenton J, St. Onge J, et al.: 1989, Antimicrobial
drug susceptibility of Streptococcus suis type 2. Ir Vet J 42: 113-
114.
10. Higgins R, Gottschalk M: 1990, An update on Streptococcus
suis identification. J Vet Diagn Invest 2:249-252.
11. Higgins R, Gottschalk M, Mittal KR, Beaudoin M: 1990, Strep-
tococcus suis infection in swine. A sixteen month study. Can J
Vet Res 54:170-173.
12. Hommez J, Devriese LA, Henrichsen J, Castryck F: 1986,
Identification and characterization of Streptococcus suis. Vet
Microbiol 11:349-355.
13. Jansen EJ, Van Dorssen CA: 1951, Meningoencephalitis bij
varkens door streptococcen. Tijdschr Diergeneeskd 76:815-832.
367
14. Jubb KVF, Kennedy PC, Palmer N: 1985, Bones and joints-
diseases of joints. In: Pathology of domestic animals, 3rd ed.,
vol. 1, p. 114. Academic Press, New York, NY.
15. Keymer IF: 1983, Streptococcus suis type II infection in a rac-
coon dog (Nyctereutes procyonoides) family Canidae. Vet Rec
113:624.
16. National Committee for Clinical Laboratory Standards: 1985,
Methods for dilution antimicrobial susceptibility tests for bac-
teria that grow aerobically, 1st ed., Approved standard. NCCLS
Document M7-A2. National Committee for Clinical Laboratory
Standards, Villanova, PA.
17. National Committee for Clinical Laboratory Standards: 1988,
Methods for dilution antimicrobial susceptibility tests for bac-
teria that grow aerobically, 1st ed., Tentative standard. NCCLS
Document M7-T2. National Committee for Clinical Laboratory
Standards, Villanova, PA.
18. Rosendal S, Breton J, Henrichsen J, et al.: 1986, Isolation of
S. suis using a selective medium. Can J Vet Res 50:537-539.
19. Sanford SE, Ross RF: 1986, Streptococcal diseases in swine.
In: Diseases of swine, ed. Leman AD, Straw B, Glock RD, et
al., 6th ed., pp. 607-615. Iowa State University Press, Ames,
IA.
20. Sanford SE, Tilker ME: 1982, Streptococcus suis type II-asso-
ciated diseases in swine: observation of a one-year study. J Am
Vet Med Assoc 181:673-676.
21. Steel RGD, Torrie JH: 1980, Multiple comparisons. In: Prin-
ciples and procedures of statistics, 2nd ed., pp. 187-188. Mc-
Graw-Hill, New York, NY.
22. Steel RGD, Torrie JH: 1980, Some uses of chi-square. In:
Principles and procedures of statistics, 2nd ed., pp. 471-472.
McGraw-Hill, New York, NY.
23. Touil F, Higgins R, Nadeau M: 1988, Isolation of Streptococcus
suis from diseased pigs in Canada. Vet Microbiol 17:171-177.
24. van Leengoed LAMG, Vecht U, Verheyen ERM: 1987, Strep-
tococcus suis type 2 infections in pigs in The Netherlands (part
two). Vet Q 9: 111-117.