Download Model of Staphylococcus aureus Central

Laboratory Animal Science
Copyright 1999
by the American Association for Laboratory Animal Science
Vol 49, No 3
June 1999
Model of Staphylococcus aureus Central Venous
Catheter-Associated Infection in Rats
Joseph S. Ulphani and Mark E. Rupp*
Background and Purpose: Staphylococcus aureus is an important cause of intravascular catheter-associated
bacteremia. We developed a rat central venous catheter (CVC)-associated infection model to study pathogenesis and treatment.
Methods: A silastic lumen-within-lumen catheter and rodent-restraint jacket were designed. Subcutaneously tunneled catheters were inserted in the jugular vein of 20 male Sprague Dawley rats. Twelve rats (group
1) were inoculated with S. aureus via the CVC; three rats (group 2) were inoculated with S. aureus via the tail
vein, five rats (group 3) served as uninfected controls; and three rats (group 4) were inoculated with S. aureus
via the tail vein but did not undergo CVC insertion. Five to eight days after inoculation, animals were
euthanized, CVCs were aseptically removed, and quantitative culture was done. Quantitative culture also
was performed on blood, heart, liver, lungs, and kidneys.
Results: Infection, characterized by bacteremia and metastatic disease, was observed in all rats inoculated
via the CVC with as few as 100 colony-forming units (CFU) of S. aureus. Rats of group 2 were not as likely to
develop CVC-associated infection, and none of the animals of groups 3 or 4 developed infection.
Conclusions: This model of CVC-associated infection should prove suitable for studying pathogenesis and
treatment of the condition.
Each year about 20 million people admitted to U.S. hospitals
undergo intravascular catheterization. About 50,000 to 120,000
develop nosocomial bacteremia (1, 2). Central venous catheter
(CVC)-related infection is a common complication and has become a major medical problem. The mortality associated with
nosocomial bacteremia approximates 25% (3). Staphylococci
are the prevailing etiologic agents, accounting for 46.9% of episodes of primary nosocomial bacteremia (4). Although Staphylococcus aureus is less frequently encountered as an etiologic
agent of catheter-related bacteremia than is S. epidermidis, it is
a more virulent pathogen, resulting in greater morbidity and
mortality (5, 6). Unfortunately, this problem is increasing at an
alarming rate. According to the Centers for Disease Control’s
National Nosocomial Infection Surveillance System, the rate of
nosocomial bacteremia caused by S. aureus increased from
122% to 283% depending on the size of hospital studied during
the 1980s (7). The rapid increase in staphylococcal bacteremia
can largely be attributed to the extensive use of intravascular
catheters.
To study the pathogenesis and treatment of this condition,
we developed a CVC-associated infection model. One important part of this model was construction of a novel Silastic
lumen-within-lumen catheter, which was tunneled subcutaneously to simulate the experience with surgically implanted central venous catheters in humans. Equally
important was the development of a special animal-restraint
jacket, which contained the catheter and facilitated rapid
and nonstressful serial blood collection.
University of Nebraska Medical Center, Omaha, Nebraska
*Address correspondence to: Dr. Mark E. Rupp, Department of Internal
Medicine, Section of Infectious Diseases, 9854031 Nebraska Medical Center, Omaha, NE 68198-4031.
Materials and Methods
Bacteria: Staphylococcus aureus strain SA-160 used in
our investigation was isolated from a patient with CVC-associated bacteremia. A single colony was incubated in 3 ml of
trypticase-soy broth overnight at 37⬚C. The overnight subculture was centrifuged at 2,000 X g for 10 min, washed, and
resuspended in phosphate buffered saline (PBS) (0.9% NaCl
in 10 mM phosphate buffer, pH 7.4). Serial dilution was performed to obtain a concentration of 105 colony-forming units
(CFU)/ml, which was confirmed by quantitative culture.
One-tenth milliliter of the bacterial suspension was added to
0.3 ml of PBS to yield a volume that completely filled the
catheter lumen.
Animals: Male Sprague Dawley rats (Charles River,
Wilmington, Mass.), weighing 400 to 500 g, were studied.
The rats were acclimated for 7 days prior to surgery. They
were housed in individual polypropylene shoe-box cages on
cob bedding; cages and bedding were changed twice per
week. Water and feed were given ad libitum. Feed consisted
of standard laboratory rat diet (Harlan Inc., Indianapolis,
Ind.). Animals were maintained under the following environmental conditions: lighting, 12-h on/off cycle; temperature, 22 to 25⬚C; air changes, minimum of 10/h; humidity,
50%. The rats were obtained from the supplier specified
pathogen free and were maintained using a soiled bedding
sentinel system.
Catheter construction: A Silastic lumen-within-lumen
catheter (U.S. patent no. 5762636) was constructed, using
silicon rubber tubing (Specialty Manufacturing Inc.,
Saginaw, Mich.) of three diameters. A 1.27-cm-long, 20gauge blunt needle with a luerlock hub (Figure 1 [B]) was
slid tip first into a 17.0-cm length of 0.064-cm-i.d. and 0.119-
283
Vol 49, No 3
Laboratory Animal Science
June 1999
cm-o.d. tubing to form the inner cannula (Figure 1 [D]). The
inner cannula was slid into a 10.0-cm-long tube of 0.15 cm
i.d. and 0.195 cm o.d. (outer cannula), which encases the
subcutaneous portion and the segment from the skin to the
hub of the inner cannula (Figure 1 [E]). This portion of the
catheter is subjected to maximal stress during catheter manipulation. The outer lumen simply serves to protect the inner cannula from stress caused either by handling or by the
animal. A 1.52-cm-long heat-shrink tube (Cole-Parmer Instrument Co., Vernon Hills, Ill.) was used to secure the inner
and outer cannulas to the needle (Figure 1 [C]). An injection
adaptor (Medexinc, Hilliard, Ohio) was attached to the
luerlock of the needle (Figure 1 [A]). A pair of collars with
the same diameter as that of the outer cannula was glided
over the projecting end of the inner cannula and was used
for securing the catheter in position by affixing it to the surrounding tissue with a suture. A clamp was made by slitting
a 0.076-cm-i.d. and 0.165-cm-o.d. silicon rubber tubing and
mounting it as a marker on the inner cannula at a predetermined insertional length to prevent the collars from sliding
off of the inner cannula (Figure 1 [H]). Catheters were sterilized before use by pressurized steam autoclave.
Restraint jacket: An animal-restraint jacket (U.S. patent
no. 5839393) was constructed, using an elongated sheet of
stretchable material, elastic in only the longitudinal direction (Figure 2). Spaced from the opposing ends of the sheet,
a pair of limb holes were formed. Velcro was mounted along
the entire edge of the ends so that it could be fastened along
the thorax of the animal. To receive the catheter, an aperture
was formed midway between the ends of the sheet. Two elastic straps were attached adjacent to the aperture to hold the
catheter. A cover flap wide enough to extend over the aperture and the straps was attached along one side of the sheet.
Velcro was mounted on the edge of the other end of the flap
to fasten it to the sheet.
Placement of the CVC: All procedures and protocols incorporated in this study were conducted with the approval of
the Institutional Animal Care and Use Committee. The surgical placement of CVCs was performed aseptically. Rats
were anesthetized by intraperitoneal administration (90 to
10 mg/kg of body weight) of a ketamine-xylazine solution.
The hair on an area over the scapula, cranial region of the
thorax, and neck was shaved. The rats were restrained on an
animal operating board, and their eyes were coated with boric acid ophthalmic ointment.
The shaved area was cleansed with povidone-iodine solution and was draped. With the rat positioned on its left side,
a 0.5-cm midline incision was made over the scapular region. The rat was turned on its back, and a small longitudinal incision was placed over the spot where the proximal
part of the jugular, acromiodeltoid, and cephalic veins join to
form the right external jugular vein. A trocar was passed
subcutaneously from the ventral to the dorsal incision down
the side of the neck. The catheter was advanced through the
trocar to the ventral incision. The trocar was then slowly
pulled out, leaving 4 to 5 cm of the catheter in a subcutaneous position. The right external jugular vein was exposed by
carefully removing the connective tissue surrounding the
vein. Two 3.0 silk sutures were placed around the vein ap-
284
Figure 1. Silastic lumen-within-lumen catheter: A, injection
adapter (heparin lock); B, needle; C, heat-shrink tube; D, inner
cannula; E, outer cannula; F, proximal collar; G, distal collar; H,
clamp.
Figure 2. Animal-restraint jacket: A, catheter; B, Velcro closure;
C, limb holes; D, elastic straps; E, injection adapter (heparin lock).
proximately 1.5 cm apart. The vein was ligated by tying the
cephalad suture, and a small incision was made between the
two sutures. A catheter filled with heparin solution (20 U of
heparin/ml of 0.9% NaCl) was slowly inserted into the vein
and was advanced into the sinus venosus. The lower suture
was tied around the vein over the proximal collar, which prevented occlusion of the catheter. A stress loop was formed,
the catheter was secured in place by tying a suture around
the distal collar, and incisions were closed. After surgery,
each animal was given buprenorphine hydrochloride (0.05
mg/kg, intravenously).
Central venous catheter-associated infection: Fortyeight h after catheter placement, the 20 rats were assigned to
three groups. Rats of group 1 (n = 12) were inoculated via the
catheter with 100 ␮l of an S. aureus suspension, ranging from
102 to 104 CFU. One rat received an inoculum of 102 CFU, three
rats received an inoculum of 103 CFU, and eight rats received
an inoculum of 104 CFU. The bacterial suspension was allowed
to dwell within the catheter lumen for 15 min. The contents of
the catheter were then flushed into the blood with 0.5 ml normal saline, and the catheter was filled with heparin solution.
Central Venous Catheter-Associated Infection Model
Rats of group 2 (n = 3) were inoculated via the tail vein and received an inoculum of 104 CFU. Group-3 (n = 5) rats (control)
did not receive a bacterial inoculum. To assess whether the
catheter had a potentiating role in bacteremia, an additional
group (group 4, n = 3) of rats was studied. This group did not
undergo any surgical procedure and was inoculated via the tail
vein with an inoculum of 104 CFU of S. aureus. Five rats of
group 1 were euthanized on day 5 by CO2 inhalation. The remaining seven rats of group 1 and rats of groups 2, 3, and 4
were sacrificed on day 8 after inoculation. The intravascular
portion of the catheters along with the surrounding vein were
removed aseptically; bacteria adhering to the catheter/tissue
were removed by vortex washing in 5 ml of PBS; and a 0.1-ml
aliquot of the bacterial suspension was taken for quantitative
plate count. The heart, lungs, liver, and kidneys of each rat
were removed, weighed, and homogenized with sterile disposable tissue grinders (Sage Products, Inc., Crystal Lake, Ill.) in 5
ml of PBS, then a 0.1-ml aliquot of the tissue suspension was
taken for quantitative plate count. One milliliter of peripheral
blood, obtained by puncture of the caudal vena cava at the time
of euthanasia, also was collected for quantitative culture.
Pulse-field gel electrophoresis: To ensure that staphylococci recovered from the rats was identical to the inoculum
strain, the isolates were characterized by use of analysis of
restriction fragment length polymorphisms with pulse-field
gel electrophoresis (PFGE). Genomic DNA was prepared for
PFGE by use of described methods (8). The following electrophoresis parameters were used: 6 V/cm; initial pulse time, 1
sec; final pulse time, 30 sec for 22 h at 14⬚C.
Statistical analysis: Bacterial counts (CFU/ml or CFU/g)
were log10 normalized, then were analyzed by use of the
Kruskal-Wallis test with the Dunn’s test for multiple comparisons. Statistical analysis was performed, using Graphpad
Prism version 2.0 (San Diego, Calif.).
Results
All rats of group 1 developed CVC-associated infection
(Figure 3). Mean number of organisms recovered from the
catheters at the time of euthanasia was 6.1 x 106 CFU/catheter. In contrast, mean number of organisms recovered from
the catheters or rats of group 2 was 10 CFU/catheter, and no
rats of groups 3 or 4 had bacteria recovered from the catheters (P < 0.01, for group 1 versus group 2, 3, or 4). Mean
number of staphylococci recovered from the blood of rats in
group 1 was 2.7 x 104 CFU/ml, compared with 7, 0, and
0 CFU/ml for groups 2, 3, and 4, respectively (P < 0.05 for
group 1 versus group 2, 3, or 4) (Figure 3). All rats of group 1
had evidence of metastatic disease with large numbers of
staphylococci recovered from the heart, lungs, liver, and kidneys (Figure 3). In contrast, there was minimal evidence of
metastatic disease in group-2 rats (P < 0.01 for group 1 versus group 2, 3, or 4). Staphylococcus aureus was the only
bacterial species recovered from the rats and, as indicated
by PFGE (gel not shown), the isolates recovered were identical to the strain inoculated.
Discussion
The purpose of this study was to develop an effective, economical, invivo model to simulate human CVC-associated
Figure 3. Staphylococcus aureus recovered from catheter (CFU/
catheter), blood (CFU/ml), and various organs (CFU/g). Group 1
(n = 12): catheterized and inoculated via the catheter, five
euthanized day 5, seven euthanized day 8. Group 2 (n = 3): catheterized and inoculated via the tail vein, euthanized day 8. Group
3 (n = 5): catheterized, not inoculated (control), euthanized day 8.
Group 4 (n = 3): not catheterized, inoculated via the tail vein,
euthanized day 8. Bars represent mean value, and lines indicate
SEM.
infection. The pathogenesis of CVC-associated infection is
complex and involves interactions between the microbe, the
device, and the host. Therefore, in vivo models should be
used to examine questions regarding pathogenesis and
treatment.
To better understand the rationale for this study, a brief
description of previously used intravascular catheter-associated infection models is in order. For years, the rabbit model
of infective endocarditis, first introduced by Garrison and
Freedman (1970), modified by Durack and Beeson (1972),
and later described in the rat by Santoro and Levison (1978),
has been considered reliable to study catheter-associated infections (9–11). This model has proven useful for investigating the pathophysiology of infective endocarditis. However,
some features of the endocarditis model constitute a substantial departure from catheter-associated infection in humans (12, 13). Most importantly, to make the animal
susceptible to infection, trauma to cardiac valves is induced
by moving the catheter backward and forward, and by keeping the tip of the catheter between the valve leaflets. This is
pathophysiologically different from the condition that predisposes humans to catheter-associated infection. Also, a
larger inoculum of bacteria was used to induce infection
than that encountered in clinical practice in humans.
In 1993, Paston et al. introduced a rat model of CVC-related
sepsis (14). This model was useful in describing the dynamics of
catheter-related sepsis. Our model differs from that of Paston
et al. in several respects. The CVC in our model is subcutaneously tunneled, which more closely mimics the conditions in
human surgically implanted catheters, such as the Hickman or
Broviac types. Also, the catheters were kept patent, allowing
collection of sequential blood samples for culture. Finally, the
need for a tethering device was obviated by the use of a rodentrestraint jacket that protected the catheter yet allowed free
animal mobility and ready access to the catheter. Dennis et al.
described an intravascular catheter-associated infection model
285
Vol 49, No 3
Laboratory Animal Science
June 1999
in the rabbit (15). Our model differs from that of Dennis et al.
in that rats are used rather than rabbits. This introduces an
economic advantage as rats are considerably less expensive
than rabbits. In addition, the advantages of using the rodentrestraint jacket described herein have already been mentioned.
It should be emphasized that the model developed in our
laboratory simulates the human condition. We were successful in inducing CVC-associated infection with secondary bacteremia and metastatic disease with a range of inocula.
Animals sacrificed as soon as day 5 after infection had a substantial degree of infection. As expected, progression of infection and metastatic disease was observed in animals
sacrificed on day 8. Infection was present at all sites, which
was the result of seeding of bacteria into the bloodstream
from the infected CVC. The difference in the degree of infection seen in rats of groups 2 and 4 clearly indicates the importance of the indwelling vascular catheters in potentiation
of the infection. This finding is in agreement with those of
other studies (5, 16, 17).
The route of inoculation also mimics that of the human condition. Migration of organisms down the external surface of the
catheter appears to be important in the pathogenesis of infections involving short-term, nontunneled catheters (18, 19). In
contrast, in long-term, subcutaneously tunneled catheters, the
major route of inoculation appears to be via the catheter hubs
and intraluminal migration (20–22).
Our CVC-associated infection model has been fully developed and standardized to study staphylococcal disease. In
this study, we have documented reproducible induction of
CVC-associated infection with a range of inocula. The importance of the catheter and the route of inoculation was unambiguously documented in the tail vein inoculation
experiments. In other studies, this model has been used to
elucidate the importance of the polysaccharide intercellular
adhesin/hemagglutinin (PIA/HA) of S. epidermidis and the
autolysin of S. epidermidis in the pathogenesis of intravascular catheter-related infection and to study the efficacy of
antistaphylococcal antibiotics (23–25). Future efforts will be
directed at further study of the pathogenesis of staphylococcal intravascular catheter-associated infection. However,
this model can be used for a broad range of applications, including the investigation of pharmacokinetics, pharmacodynamics, efficacy, and disposition of novel antimicrobial
agents. The rodent intravascular catheter and the rodentrestraint jacket developed in our laboratory make the model
efficient. Together these devices obviate elaborate preparation, equipment, and practical skills otherwise required for
multiple blood sample collection studies.
In conclusion, use of our in vivo model, which simulates
human S. aureus CVC-associated infection, should prove
suitable in elucidating the pathogenesis and treatment of
this and other similar conditions.
Acknowledgements
We thank Firdous Ulphani for assistance with production of
the rodent-restraint jacket.
This work was supported in part by a Grant-in-Aid from the
American Heart Association, (MER) no. 96006810.
286
References
1. Maki, A. G. 1992. Infection due to infusion therapy,
p. 849–898. In J. V. Bennet and P. S. Brachman (ed.), Hospital
infections. Little, Brown & Co., Boston, Mass.
2. Raad, I. I., and G. P. Bodey. 1992. Infectious complication
of indwelling catheters. Clin. Infect. Dis. 15:197–208.
3. Wenzel, R. P. 1988. The mortality of hospital-acquired bloodstream infections: need for a new vital statistic. Int. J.
Epidemiol. 17:225–227.
4. U. S. Department of Health and Human Services. 1997. National nosocomial infections surveillance (NNIS) report, data
summary from October 1986–April 1997. Am. J. Infect. Control 25:477–487.
5. Dugdale, D. C., and P. G. Ramsey. 1990. Staphylococcus
aureus bacteremia in patients with Hickman catheters. Am.
J. Med. 89:137–141.
6. Raad, I., J. Narro, A. Khan, et al. 1992. Serious complications of vascular catheter-related Staphylococcus aureus bacteremia in cancer patients. Eur. J. Microbiol. Infect. Dis.
11:675–682.
7. Banerjee, S. N., T. G. Emori, D. H. Culver, et al. 1991. Secular trends in nosocomial primary bloodstream infections in the
United States, 1980–1989. Am. J. Med. 91(Suppl. 3B):86–89.
8. Bannerman, T. L., G. A. Hancock, F. C. Tenover, et al.
1995. Pulsed-field gel electrophoresis as a replacement for
phage typing of Staphylococcus aureus. J. Clin. Microbiol.
33:551–555.
9. Garrison, P. K., and L. R. Freedman. 1970. Experimental
endocarditis. Staphylococcal endocarditis in rabbits resulting
from placement of a polyethylene catheter in the right side of
the heart. Yale J. Biol. Med. 42:394–410.
10. Durack, A. T., and P. B. Beeson. 1972. Experimental bacterial endocarditis. Survival of bacteria in endocardial vegetations. Br. J. Exp. Pathol. 53:50–53.
11. Santoro, J., and M. E. Levison. 1978. Rat model of experimental endocarditis. Infect. Immun. 19:915–918.
12. Barza, M. 1978. A critique of animal models in antibiotic research. Scand. J. Infect. Dis. Suppl. 14:109–117.
13. Wright, A. J., and W. R. Wilson. 1982. Experimental animal endocarditis. Mayo Clin. Proc. 57:10–14.
14. Paston, M. J., R. A. Meguid, M. Muscaritoli, et al. 1993.
Dynamics of central venous catheter-related sepsis in rats.
J. Clin. Microbiol. 31:1652–1655.
15. Dennis, M. B., Jr., D. R. Jones, and F. C. Tenover. 1989.
Chlorine dioxide sterilization of implanted right atrial catheters in rabbits. Lab. Anim. Sci. 39:51–55.
16. Raviglione, M. C., R. Battan, A. Pablos-Mendez, et al.
1989. Infections associated with Hickman catheters in patients
with acquired immunodeficiency syndrome. Am. J. Med.
96:780–786.
17. Gutschik, E. 1983. Experimental staphylococcal endocarditis: an overview. Scand. J. Infect. Dis. Suppl. 41:87–92.
18. Mermal, L. A., R. D. McCormick, S. R. Springman, et al.
1991. The pathogenesis and epidemiology of catheter-related
infection with pulmonary artery Swan-Ganz catheters: a prospective study utilizing molecular subtyping. Am. J. Med.
91(Suppl. 3B):197S–205S.
19. Maki, D., M. Ringer, and C. J. Alvarado. 1991. Prospective
randomized trial of povidone-iodine, alcohol, and chlorhexidine
for prevention of infection associated with central venous and
arterial catheters. Lancet 338:339–343.
20. Linares, J., A. Sitges-Serra, J. Garau, et al. 1985. Pathogenesis of catheter sepsis: a prospective study with quantitative and semiquantitative cultures of catheter hub and segments. J. Clin. Microbiol. 21:357–360.
21. Tenney, J. H., M. R. Moody, K. A. Newman, et al. 1986.
Adherent microorganisms on lumenal surfaces of long-term
intravenous catheters: importance of Staphylococcus
epidermidis in patients with cancer. Arch. Intern. Med.
146:1949–1954.
Central Venous Catheter-Associated Infection Model
22. Cheesbrough, J. S., R. G. Finch, and R. P. Burden. 1986.
A prospective study of the mechanisms of infection associated
with hemodialysis catheters. J. Infect. Dis. 154:579–588.
23. Rupp, M. E., J. S. Ulphani, and D. Mack. 1997. Characterization of Staphylococcus epidermidis 1457 and an isogenic
PIA-negative/HA-negative mutant in a rat CVC-associated infection model. Abstr. 388, p. 143. Abstracts of the 35th Annual
Meeting of the Infectious Diseases Society of America.
24. Rupp, M. E., and J. S. Ulphani. 1998. Efficacy of LY333328
in a rat Staphylococcus aureus central venous catheter-associated infection model. Abstr. F-111, p. 260. Abstracts of the
38th Interscience Conference on Antimicrobial Agents and
Chemotherapy.
25. Rupp, M. E., J. S. Ulphani, P. D. Fey, et al. 1998. Characterization of Staphylococcus epidermidis O-47 and an isogenic
autolysin-deficient mutant in a rat central venous catheterassociated infection model. Abstr. B-23, p. 50. Abstracts of the
38th Interscience Conference on Antimicrobial Agents and
Chemotherapy.
287