ANTIMICROBIAL AGENTS AND CHEMOTHERAPY, Apr. 1975, p.

396-399
Copyright 0 1975 American Society for Microbiology

Vol. 7, No. 4
Printed in U.S.A.

Antibiotic Resistance Patterns of Clinical Isolates of

Serratia

marcescens

ROBERT C. COOKSEY, EDWARD R. BANNISTER,

AND

W. EDMUND FARRAR, JR.*

Departments of Medicine* and Laboratory Medicine, Medical University of South Carolina, Charleston,
South Carolina 29401

Antimicrobial susceptibility patterns of 102 clinical isolates of Serratia

marcescens from three medical centers were studied by using disk sensitivity and
agar dilution methods. The least resistance was demonstrated against gentami-

cin, nalidixic acid, chloramphenicol, and sulfisoxazole, all of which inhibited

more than 80% of the strains. Cephalothin was completely ineffective, and more
than 90% of strains were resistant to ampicillin and tetracycline. As demonstrated by the agar dilution method, the minimal inhibitory concentration of
nalidixic acid, gentamicin, tobramycin, and chloramphenicol for most strains fell
within therapeutically attainable concentrations. The prevalence of resistance to
ampicillin, cephalothin, and tetracycline was nearly the same at all three
medical centers, whereas there appeared to be patterns characteristic for each
center with regard to the other drugs used. Eleven of the isolates
produced pigment and exhibited patterns similar but not identical to those of the
nonpigmented strains, all 11 being resistant to between three and six drugs. Half
of the strains were resistant to five or more antibiotics, indicating that some
Serratia exhibit resistance to an unusually broad range of therapeutic agents.

The problems associated with hospital infections caused by Serratia marcescens have become increasingly evident (3, 5, 15). The spectrum of infections includes meningitis, pulmonary infections, septicemia, endocarditis, and a
variety of localized infections. The ability of
this opportunistic pathogen to acquire resistance to a broad range of antibiotics has made
effective therapy more difficult. Several recent
investigations have dealt with the problem of
antibiotic resistance in Serratia (3, 4, 6, 10-13).
The overall findings of these studies indicate
that the polymyxins and cephalosporins are the
least effective agents against Serratia in vitro,
and although gentamicin, kanamycin, chloramphenicol, and nalidixic acid are the most effective, resistance to kanamycin and chloramphenicol, as well as the p-lactam antibiotics,
has increased during the last decade. Serratia
strains isolated from different hospitals often
appear to have characteristic resistance patterns (13). Other studies have provided conflicting results concerning resistance patterns of
pigmented versus nonpigmented strains (2, 11).
It is evident that patterns of drug resistance
in Serratia may vary over even short periods of
time and may be characteristic for different
hospitals. For this reason and because of the
rapid development of new antibiotics, periodic

evaluation of antibiotic resistance in Serratia is

needed. This study was undertaken to obtain
up-to-date information regarding the resistance
of S. marcescens to antimicrobial agents currently available.
MATERIALS AND METHODS
A total of 102 strains of S. marcescens isolated
between June and December 1973 was examined.
Sixty-two strains were isolated at the Diagnostic
Bacteriology Laboratory of the Medical University of
South Carolina, 22 were provided by J. Shulman of
Emory University, and 18 were made available by D.
Gr6schel of the M. D. Anderson Hospital and Tumor
Institute in Houston, Tex. Each was referred to the
South Carolina State Board of Health for species
confirmation or subjected to biochemical tests including triple sugar-iron, arabinose, indole, motility, and
deoxyribonuclease. Sources of isolation included
urine, blood, throat, sputum, stool, exudate, and
wounds. Strains were inoculated onto Trypticase soy
agar slants and incubated overnight to detect pigment
production and then stored at 4 C until needed.
Disk sensitivity testing was performed by the
single-disk technique described by Bauer et al. (1) on
Mueller-Hinton agar with the following antibiotics:
streptomycin (10,ug), sulfisoxazole (0.25 mg), tetracycline (30 ag), cephalothin (30 ag), kanamycin (301Ag),
ampicillin (10 ug), nalidixic acid (30 Ag), gentamicin
(10 ag), carbenicillin (50 gg), colistin (10 Mg), tobramycin (10 Mg), and chloramphenicol (30 Mg). The
minimum inhibitory concentration (MIC) was deter-

396

Downloaded from http://aac.asm.org/ on October 17, 2015 by Griffith University

Received for publication 29 November 1974

ANTIBIOTIC RESISTANCE OF S. MARCESCENS

VOL. 7, 1975

mined by the agar dilution method (7) on MuellerHinton agar, using the Steers inocula-replicating
apparatus (12) with an inoculum of approximately 5
X 104 bacteria. The MIC was defined as the lowest
concentration of drug preventing visible growth after
18 h of incubation at 37 C.
RESULTS

and were resistant to cephalothin, ampicillin,

and tetracycline. All were susceptible to chloramphenicol, kanamycin, nalidixic acid, sulfisoxazole, and gentamicin, as were the majority
of nonpigmented strains. However, eight of the
pigmented isolates were resistant to streptomycin, eight were resistant to colistin, and
five were resistant to tobramycin, indicating
that a higher percentage of pigmented strains
were resistant to these drugs than were the
nonpigmented strains.
Figure 1 shows the cumulative percentage of
strains inhibited in vitro by 11 drugs tested at
various concentrations in agar. The figures in
parentheses represent approximate serum concentrations achieved with recommended dosages except for nalidixic acid, for which approximate achievable urinary tract concentration is
listed (8, 9). Tobramycin, gentamicin, and
chloramphenicol inhibited greater than 90% of
the strains by concentrations equal to or less
than obtainable serum levels, and nalidixic acid
inhibited at least 90% of the strains at a
concentration achievable in the urine. Cephalothin and colistin failed to inhibit any of the
isolates at the achievable serum concentrations.
Levels of resistance of the 11 pigmented strains
did not appear to differ significantly from those
of the nonpigmented strains except for streptomycin, which was found to have MICs of
between 20 and 40 Ag/ml. Sixty-five percent of
the nonpigmented Serratia had MICs of streptomycin that were between 5 and 20,ug/ml. The
MICs for ampicillin, tetracycline, and cephalothin (antibiotics to which the highest percentages of strains were resistant) were higher for
those strains that were resistant to more than
five drugs, indicating that multiresistant

TABLE 1. Percentage of Serratia resistant to antibiotics as determined by disk sensitivity tests

Strains resistant (%)
Antibiotic

Cephalothin .............
Ampicillin ...............
Tetracycline .............
Colistin .................
Carbenicillin
Streptomycin ............
Tobramycin .............
Kanamycin ..............
Chloramphenicol ........
Sulfisoxazole ............
Gentamicin ..............
Nalidixic acid ...........
............

a Medical

MUSCa
(62 strains)
100
90
90
55
27
25
25
3
11
2
3
3

University of South Carolina.

Emory

(22 strains)

100
95
95
77
45
45
14
36
4
23
18
4

M. D. Anderson
(18 strains)
100
100
100
39
72
50
44
72
44
50
44

28

Total

(102 strains)

100
93
93
57
39
34
26
22
16
15
14
8

Downloaded from http://aac.asm.org/ on October 17, 2015 by Griffith University

Table 1 gives the results of disk sensitivity

testing for the 102 Serratia strains isolated. All
strains tested were resistant to cephalothin.
Ninety-three percent of the strains were resistant to ampicillin and tetracycline. The most
effective drugs were nalidixic acid (8% of strains
resistant), gentamicin (14% resistant), sulfisoxazole (15% resistant), and chloramphenicol
(16% resistant). Next in order of increasing
resistance were kanamycin (22% resistant), tobramycin (26% resistant), streptomycin (34%
resistant), carbenicillin (39% resistant), and
colistin (57% resistant). Multiple resistance was
common, as demonstrated by the observation
that 56 of the 102 isolates were resistant to at
least 5 of the drugs tested, 14 were resistant to
at least 8, and 5 were resistant to all 12.
Resistance varied among the three hospitals
(Table 1), with strains from M. D. Anderson
Hospital and Tumor Institute being more frequently resistant to every drug tested except
colistin and cephalothin. The strains from
Emory University were more resistant to colistin than the strains from either of the other two
hospitals, and the Emory strains were more
resistant to kanamycin, carbenicillin, sulfisoxazole, gentamicin, and streptomycin than were
the strains from the Medical University of
South Carolina.
Eleven of the 102 isolates produced pigment

397

398

ANTIMICROB. AGENTS CHEMOTHER.

COOKSEY, BANNISTER, AND FARRAR

_ 0-

40-

*~30.

01

0.2 0.4 0.8

6 3.2 6.3 12.5 25 50 100 200 400 800 1600 3200 6400 12,800

mg /rrd

FIG. 1. Cumulative percentage of strains inhibited in vitro by 11 drugs tested at various concentrations in
agar. Numbers in parentheses by each drug symbol, except nalidixic acid, indicate approximate concentrations
(in micrograms per milliliter) achievable in serum with recommended dosages. Abbreviations: NX, Nalidixic
acid; TB, tobramycin; GM, gentamicin; C, chloramphenicol; SM, streptomycin; AM, ampicillin; TC,
tetracycline; CB, carbenicillin; CL, colistin; KN, kanamycin; CR, cephalothin. Asterisk indicates approximate
concentration achievable in urine.

strains are often more resistant to these three

drugs than are strains resistant to fewer than
five antimicrobial agents.

DISCUSSION
Our results indicate that recently isolated
strains of S. marcescens are susceptible to
gentamicin, chloramphenicol, sulfisoxazole,
and nalidixic acid at therepeutically achievable
concentrations. These findings correspond to
previous reports concerning the efficacy-of these
antibiotics (4, 6, 10, 12). Ampicillin, cephalothin, tetracycline, and colistin are ineffective in
vitro and do not appear to be useful in the
treatment of Serratia infections. The susceptibility of these strains to kanamycin, streptomycin, tobramycin, and carbenicillin was
highly variable.
Characteristic resistance patterns were found
for each of the three hospitals with regard to
every antibiotic tested except ampicillin, tetracycline, and cephalothin. The strains from the
M. D. Anderson Hospital and Tumor Institute
were more frequently resistant than the strains
from either of the other two hospitals to every
drug tested except colistin. The high degree of
overall resistance demonstrated by these strains
may be related to selective pressure by longterm antibiotic therapy in cancer patients. The
Serratia that were isolated at the Medical
University of South Carolina were more suscep-

tible than those from Emory to every drug

tested except cephalothin, tobramycin, and
chloramphenicol. The observation that at least
90% of the 102 strains were resistant to ampicillin, tetracycline, and cephalothin may indicate
innate resistance to these drugs in this species.
A greater percentage of pigmented than nonpigmented strains were resistant to tobramycin,
streptomycin, and colistin. Pigmented strains
had higher MICs of streptomycin (20 to 40
gg/ml versus 5 to 20 jig/ml for most of the
nonpigmented strains). Previous reports have
provided conflicting results regarding antimicrobial susceptibility of nonpigmented and pigmented Serratia on the basis of differences in
cell wall lipid (2, 11). However, this premise has
been recently challenged by Winshell and Neu
(16), who found no appreciable differences in
lipid content among resistant nonpigmented,
susceptible nonpigmented, and susceptible pigmented groups of S. marcescens.
Medeiros and O'Brien (10) demonstrated
transferable resistance due to R factors in 21 of
22 strains of Serratia resistant to five or more
antibiotics. They concluded that R factors contribute additional degrees of resistance to tetracycline and ampicillin in strains already moderately resistant to these drugs. We have also
found R factors among our multiple-resistant
strains of S. marcescens, and are currently
investigating further the properties and significance of these plasmids in this species.

Downloaded from http://aac.asm.org/ on October 17, 2015 by Griffith University

o 50-

VOL. 7, 1975

ANTIBIOTIC RESISTANCE OF S. MARCESCENS

ACKNOWLEDGMENTS
We are grateful to Dieter Groschel and Jonas Shulman for
providing bacterial strains.

9. Kucers, A. 1972. The use of antibiotics, p. 66-280.

William Hinnemann Medical Books, Ltd., London.
10. Medeiros, A. A., and T. F. O'Brien. 1969. Contribution of
R factors to the antibiotic resistance of hospital isolates
of Serratia, p. 30-35. Antimicrob. Ag. Chemother.
1968.
11. Miller, M. A., C. Y. Chang, and J. C. Tsang. 1973.
Antibiograms and lipid contents of pigmented and
nonpigmented strains of Serratia marcescens. Antimicrob. Agents Chemother. 4:66-81.
12. Oberhofer, T. R., and R. Hajkowski. 1970. Evaluation of
non-lactose-fermenting members of the KlebsiellaEnterobacter-Serratia division. II. Antibiotic susceptibility. Am. J. Clin. Pathol. 54:726-732.
13. Schaefler, S., J. Winter, A. Catelli, J. Greene, and B.
Toharski. 1971. Specific distribution of R factors in
Serratia marcescens strains isolated from hospital
infections. Appl. Microbiol. 22:339-343.
14. Steers, E., E. L. Foltz, and B. S. Graves. 1959. An inocula
replicating apparatus for routine testing of bacterial
susceptibility to antibiotics. Antibiot. Chemother.
9:307-311.
15. Wilfert, J. N., F. F. Barrett, W. H. Ewing, M. Finland,
and E. H. Kass. 1970. Serratia marcescens: biochemical, serological, and epidemiological characteristics
and antibiotic susceptibility of strains isolated at
Boston City Hospital. Appl. Microbiol. 19:345-352.
16. Winshell, E. B., and H. C. Neu. 1974. Relation of cell wall
lipid content of Serratia marcescens to resistance to
antimicrobial agents. Antimicrob. Agents Chemother.
6:73-75.

Downloaded from http://aac.asm.org/ on October 17, 2015 by Griffith University

LITERATURE CITED
1. Bauer, A. W., W. M. M. Kirby, J. C. Sherris, and M.
Turck. 1966. Antibiotic susceptibility testing by standardized single disc method. Am. J. Clin. Pathol.
45:493-496.
2. Chang, C., R. E. Molar and J. C. Tsang. 1972. Lipid
content of antibiotic-resistant and -sensitive strains of
Serratia marcescens. Appl. Microbiol. 24:972-976.
3. Dodson, W. H. 1968. Serratia marcescens septicemia.
Arch. Intern. Med. 121:145-150.
4. Edmondson, E. B., and J. P. Sanford. 1967. The Klebsiella-Enterobacter (Aerobacter)-Serratia group. A clinical and bacteriological evaluation. Medicine 46:323.
5. Ewing, W. H., J. G. Johnson, and B. R. Davis. 1962. The
occurrence of Serratia marcescens in nosocomial infections. Center for Disease Control, Atlanta, Ga.
6. Greenup, P., and D. J. Blazevic. 1971. Antibiotic susceptibilities of Serratia marcescens and Enterobacter liquifaciens. Appl. Microbiol. 22:309-314.
7. Grove, D. C., and W. A. Randall. 1955. Methods for the
determination of bacterial susceptibility to antibiotics,
p. 188-196. In Assay methods of antibiotics. Medical
Encyclopedia, Inc., New York.
8. Jackson, G. G. 1974. Gentamicin and tobramycin, p.
42-63. In B. M. Kagan (ed.), Antimicrobial therapy.
W. B. Saunders Co., Philadelphia.

399