Rat polyvinyl sponge model for the study of infections: host factors and microbial proliferation

Use of an experimental model to evaluate infection resistance of meshes in abdominal wall surgery

BACKGROUND

Staphylococcal species are the most common organisms causing prosthetic mesh infections, however, infections due to rapidly growing mycobacteria are increasing. This study evaluates the resistance of biomaterial for abdominal wall prostheses against the development of postoperative infection in a rat model.

MATERIAL AND METHODS

In 75 rats, we intramuscularly implanted three different types of prostheses: (1) low-density polypropylene monofilament mesh (PMM), (2) high-density PMM, and (3) a composite prosthesis composed of low-density PMM and a nonporous hydrophilic film. Meshes were inoculated with a suspension containing 10⁸ colony-forming units of Staphylococcus aureus, Staphylococcus epidermidis, Mycobacterium fortuitum, or Mycobacterium abscessus before wound closure. Animals were sacrificed on the eighth day postoperatively for clinical evaluation, and the implants were removed for bacteriologic analyses.

RESULTS

Prostheses infected with S aureus showed a higher bacterial viability, worse integration, and clinical outcome compared with infection by other bacteria. Composite prostheses showed a higher number of viable colonies of both M fortuitum and Staphylococcus spp., with poorer integration in host tissue. However, when the composite prosthesis was infected with M abscessus, a lower number of viable bacteria were isolated and a better integration was observed compared with infection by other bacteria.

CONCLUSIONS

Considering M abscessus, a smaller collagen-free contact surface shows better resistance to infection, however, depending on the type of bacteria, prostheses with a large surface, and covered with collagen shows reduced resistance to infection, worse integration, and worse clinical outcome.

Interaction of beta-lactam antibiotics with the bactericidal activity of leukocytes against Escherichia coli

The effect of beta-lactam antibiotics on phagocytosis and intracellular killing of four isogenic Escherichia coli strains differing in their 0- and K antigens was studied by adopting the rat polyvinyl-sponge model. The penicillins mezlocillin, ticarcillin and piperacillin rendered all four isogenic E. coli strains more susceptible to intraleukocyte killing; the cefalosporins tested exhibited inhomogenous effects; lamoxactam was marginally effective, whereas cefoxitin was completely ineffective; cefotaxime caused an increase in intracellular killing of the capsule-defective mutant only. The beta-lactam promoted increase in intracellular killing could be inhibited by alpha-methylmannoside but not by alpha-methylglucoside. Free-flow electrophoretic separation of mezlocillin-treated bacteria and guinea pig erythrocytes revealed that co-migration of E. coli and erythrocytes respectively could be inhibited by alpha-methylmannoside but not by alpha-methylglucoside. These data indicate that mezlocillin interferes with the mannose sensitive adhesins of E. coli.

Murine neutrophils collected from subcutaneously implanted sponges

Polyvinyl sponges implanted subcutaneously in mice were used as a source of murine neutrophils (PMNs). Yield of PMNs from sponges averaged 6.3 X 10(5) PMNs per mouse and viability of sponge PMNs was 96.5%. These results were comparable to the yield and viability of PMNs harvested from peripheral blood. The chemotactic activity of sponge PMNs was significantly greater than the chemotactic activity of peripheral blood PMNs (chemotactic index of 4.78 +/- 1.66 in sponge PMNs versus 2.16 +/- 1.08 in peripheral blood PMNs, P less than 0.001). The difference in chemotactic activity was not attributable to hypotonic injury of PMNs nor the presence of soluble factors in sponge fluids. Phagocytic activity of sponge PMNs was comparable to the activity of blood PMNs (mean of 31.3 +/- 16.4% phagocytic cells for sponge PMNs versus 33.0 +/- 23.9% for blood PMNs). In addition, the number of fluorescent spheres ingested by sponge PMNs was not different (mean of 2.81 +/- 0.86 in sponge PMNs versus 2.65 +/- 0.68 spheres per cell for blood PMNs). These studies indicate that subcutaneously implanted sponges can be used as a source of functioning murine PMNs.

Effects of ovarian hormones on manifestation of purulent endometritis in rat uteruses infected with Escherichia coli

To assess the influence of hormones on uterine infections, Escherichia coli was infused into uterine lumens of ovariectomized or adrenoovariectomized rats receiving exogenous administration of various doses of ovarian hormones. Large numbers of E. coli were recovered from the rat uterine lumens, irrespective of hormonal influences. The number of leukocytes in the uterine flushings, representing the magnitude of purulent inflammation, differed significantly depending upon the hormonal regimen given to each host. Purulent endometritis was induced by E. coli in ovariectomized rats receiving progesterone or corn oil (hormone vehicle). Infections were asymptomatic in rats receiving estradiol, but promethazine-treated uterine horns were susceptible to infection. When progesterone was administered along with estradiol, purulent inflammation was caused by E. coli, but the number of leukocytes in the uterine lumens was significantly less than that obtained from the rats treated with progesterone or corn oil. These effects of ovarian hormones on uterine infections were observed in adrenoovariectomized rats as well as in ovariectomized rats. It is suggested that estradiol alters the nature of endometrial epithelium and prevent manifestation of purulent endometritis; progesterone antagonizes estradiol. Adrenal hormones appear not to participate in the pathogenesis of endometritis induced by E. coli.

Synergy between acylureidopenicillins and immunoglobulin G in experimental animals

The interaction of a commercially available 7S modified immunoserumglobulin and beta-lactam antibiotics was studied in animal experiments (granuloma pouch model) as well as an ex vivo system (rat polyvinyl sponge model). Infections of the pouches were caused by gram-negative rods and gram-positive cocci, respectively. Therapy of pouches being infected with beta-lactamase-producing strains with beta-lactam antibiotics and immunoserumglobulin was as effective as beta-lactam antibiotic monotherapy of beta-lactamase-negative strains. This synergistic effect between immunoserumglobulin and beta-lactam antibiotics against beta-lactamase-producing bacteria is due to inactivation of enzymic beta-lactamase activity by specific antibodies against beta-lactamases. Immune phagocytosis was studied by adopting the in vivo and ex vivo models, respectively. Immune phagocytosis was most effectively stimulated by immunoserumglobulin whereas a pepsin-degraded product or a preparation obtained by pH4 treatment caused only minor effects. Furthermore, immunoserumglobulin stimulated phagocytosis and intracellular killing of gram-negative bacteria even in the absence of specific antibodies against these strains. Analogous effects were obtained with spermidine and albumin. These results indicate that immunoserumglobulin may stimulate phagocytosis nonspecifically, too. Thus, immunoserumglobulin may play a dual role in host defense mechanisms; in addition immunoserumglobulin acts as a beta-lactamase inhibitor, thus protecting beta-lactam antibiotics from hydrolysis.