Pseudomonas aeruginosa

P. aeruginosa is widely distributed in nature and in the hospital environment with a predilection for moist areas. Its inherent resistance to many antimicrobials and its ability to produce many enzymes contribute to its pathogenic potential as both a primary and a secondary cause of infection. It is easily grown and identified in the microbiology laboratory. However, susceptibility testing remains a problem. Currently, the best approach to treatment is an aminoglycoside and an antipseudomonal beta-lactam antimicrobial. Typing can differentiate strains, but should be reserved for specific epidemiologic problems.


INTRODUCTION
P. aeruginosa was first named by Schroeter in 1872. At this time, its association with human infections was not appreciated. However, physicians had noted the occasional appearance of blue or blue-green stains on surgical dressings for many years. In 1882, Gessard isolated P. aeruginosa in pure culture from a wound and determined that the blue-green color was due to a pigment produced by this organism. 1 Shortly thereafter, Charrin demonstrated animal pathogenicity. 2 Despite subsequent isolation of P. aeruginosa from infections in many anatomic sites, its presence was long considered insignificant. Its inherent resistance to antimicrobial agents allows this organism to become dominant both as a commensal and in infectious processes when more susceptible organisms are suppressed. Hence, because of the introduction of broad-spectrum antimicrobial agents, P. aeruginosa has become a major cause of nosocomial infections.

BASIC PRINCIPLES Habitat
P. aeruginosa is a ubiquitous inhabitant of water, vegetation, and moist soil and has a worldwide distribution. In humans, this organism can be isolated from the stool of about 5% to 10% of healthy individuals 3 and is also sporadically found in moist areas of the human -skin (axilla, groin) as well as in the nose and throat. 35 P. aeruginosa is introduced into the hospital environment via patients who are already colonized with this organism. In addition, fruits, plants and vegetables are important sources which introduce this bacterium into the hospital environment. 6 Once introduced, simple nutritional requirements allow the organism to thrive in almost any moist setting containing even trace amounts of organic material. Table  1 is a list of known reservoirs of P. aeruginosa in the hospital. 7 " 12 In the hospital setting, spread of P. aeruginosa among patients occurs when personnel utilize poor handwashing technique and transport the organism from patient to patient or from reservoir to patient. 12 Airborne spread is rare because P. aeruginosa is very sensitive to drying. Fortunately, spread of Pseudomonas most often results in colonization, which is usually transient, rather than infectious. Morphology P. aeruginosa is a slender, nonsporulating, usually noncapsulated gram-negative bacillus belonging to trie family Pseudomonadaceae. These bacilli may vary considerably in size and proportion, but are usually 1.5/uto3Mlong and 0.5JI in breadth. They are arranged singly, in pairs, or in short chains. Flagellar staining will reveal a single polar flagellum by which the organism is actively motile. Motility can be demonstrated by dark-field microscopy, but not by migration through agar because the presence of molecular oxygen is necessary for motility with this organism.
Cultural Characteristics P. aeruginosa grows readily on most bacteriological media at a temperature of 30° to 37°C. Because this organism is an obligate aerobe, oxygen is required although there is some growth under anaerobic conditions in the presence of nitrates. High salt concentrations are well tolerated, but desiccation or heat is not. Colonial forms are variable, with four types of colonies seen. On a 24-hour sheep blood agar plate, the most frequen t form is a low convex to flat colony, 1 mm to 5mm in diameter, with a rough surface and irregular edges. A second type of colony is smaller, raised, and coliform-like with a smooth surface and regular edges. Another form seen is the rough colony which is raised and umbonate or frankly rugose. Finally, there are mucoid forms, commonly seen in strains recovered from patients with cystic fibrosis, which are large, convex, almost transparent colonies which are markedly mucoid and tend to coalesce early in the growth cycle; later they become flatter and contoured. Any combination of these forms may be seen in the same culture.
Colonies are often /3-hemolytic on a 24-hour plate and usually show diffuse ^-hemolysis by 48 hours. Most produce a water-and chloroform-soluble phenazine pigment called pyocyanin (from the Greek, blue pus). This pigment diffuses into the surrounding medium and imparts a blue or blue-green color to the medium but not to the colony. Presence of this pigment confirms the identity of P. aeruginosa; but the absence of pyocyanin does not exclude the identity. There are several reasons for lack of pigment: other brown to black phenazine pigments may be formed and mask the blue-green pigment; many cultures do not form pigments except on special media; some produce only small amounts which are not sufficient to color the media; and some do not produce pigment at all.
Another pigment commonly produced by most P. aeruginosa and by some other Pseudomonas species is fluorescein. Fluorescein (pyoverdin) is a water-soluble, pale yellow-green pteridine which may impart a faintly yellowish tinge to cultures, but is usually difficult to detect. It is not apparent on sheep blood agar but can usually be detected by intense greenish fluorescence under ultraviolet light on medium with magnesium and phosphate such as Mueller-Hinton agar.
Another characteristic of P. aeruginosa colonies is a grapelike odor. This is due to the production of aminoacetophenone from tryptophan.

Identification
The combination of the typical colony form, pyocyanin production, and fruity odor may allow the experienced technologist to predict accurately the presence of P. aeruginosa in a culture before the results of biochemical tests are available. This may be useful as it allows the selection of appropriate antipseudomonal antimicrobics for the susceptibility test. In addition, there are other metabolic and biochemical characteristics which should be used to confirm the identity of P. aeruginosa. Table 2 lists those characteristics most useful for the identification of P. aeruginosa.™ Most laboratories use only a few of these characteristics such as oxidase, pyocyanin production, colony morphology, and a triple sugar iron slant. Additional characterization in the routine clinical laboratory increases costs without improving the accuracy of identification.

Pathogenicity
P. aeruginosa is an unusual pathogen in that it can infect a wide variety of life-forms. 14 These include warmand cold-blooded vertebrates, insects, plants, and both terrestial and aquatic animals. As might be expected from such a versatile pathogen, there are a number of factors that appear to enhance its pathogenicity. The first of these is lipopolysaccharide (LPS) which is an endotoxin similar to that of the Enterobacteriaceae. There are differences in the hydroxy acids of its lipid A moiety which may account for the fact that LPS from P. aeruginosa is a weaker endotoxin than are those from the Enterobacteriaceae. Despite its relative lack of toxicity, this LPS has an LD in mice of about 0.1 ng. Moreover, in man, antibody directed at LPS appears to be an important host defense mechanism. 15 ' 16 In addition to LPS, which is an endotoxin, most strains of P. aeruginosa elaborate a number of exotoxins. Exotoxin A is the most potent of these toxins and is 20,000 times as toxic as LPS in mice. 17 This toxin inhibits protein synthesis by the same mechanism as diphtheria toxin. It is not produced by other species of Pseudomonas or by the Enterobacteriaceae species. Another toxin produced is leukocidin which is a cell-bound protein that is lethal for human leukocytes.
Its role in human infection is speculative although it is well recognized that polymorphonuclear leukocytes are crucial in the host defenses against Pseudomonas infection." Any interference with leukocytes or with opsonizing antibodies may have dire results.
A number of enzymes are also produced by P. aeruginosa. These include proteases, elastase, collagenase, phospholipases, lipase, and lecithinase. The phospholipases are responsible for the hemolytic activity of P. aeruginosa. The pathogenesis of necrotizing vasculitis such as seen with ecthyma gangrenosum may be related to the elastase. The effects of all of these enzymes are generally localized to the infected tissue.
Finally, most strains of P. aeruginosa elaborate a viscous extracellular slime that is glycoprotein in nature and shares heat-stable " O " antigenicity with the cell wall. However, it is different than LPS. This slime differs from the capsule of pneumococcus in that it is not firmly bound to the cell surface, but is similar in that it too appears to inhibit phagocytosis unless opsonizing antibodies are present. 2 The slime is toxic, killing mice when injected; and antigenic with active or passive immunization against slime protecting the mice from lethal challenge with live bacteria.
Surprisingly, these many potential virulence factors do not appear to make P. aeruginosa as capable of causing disease as are some gram-negative bacilli such as Francisella tularensis. Healthy persons are rarely infected with P. aeruginosa. Rather, this pathogen infects patients with defective host immune defense systems. Even with these patients colonization occurs far more frequently than infection.

RESISTANCE TO ANTIMICROBIAL AGENTS
Resistance to most antimicrobial agents is a characteristic of P. aeruginosa. This resistance is mediated by the chromosome, by plasmids, or by both. Such intrinsic resistance has made the treatment of serious Pseudomonas infections difficult. During the last two decades a steadily increasing number of antipseudomonal compounds have been developed. These compounds include both aminoglycosides and beta-lactam agents. They all show excellent activity against Pseudomonas in vitro; yet they have failed to affect substantially the mortality associated with these infections. 21 ' 24 Examination of some of the factors which are responsible for this discrepancy is indicated. Obviously these patients were often more seriously ill, and many were immunosuppressed. However, some of the fault lies with susceptibility testing methods.
An example is the aminoglycosides which have a narrow toxic-therapeutic ratio. There could be no better illustration of the theoretical utility of using minimal inhibitory concentrations (MICs) to guide the treatment of infection than in the case of aminoglycoside treatment of Pseudomonas infections. However, the MICs of aminoglycosides for Pseudomonas aeruginosa vary depending on which dilution method is used. 25 The MICs are also known to be altered by variations in calcium and magnesium ion concentrations 26 ' 28 and can be falsely low unless the medium is supplemented. Although such supplementation is recommended, 28 some commercial broth dilution procedures have failed to do so. 29 ' 30 Animal models have confirmed that the current susceptibility criteria in use for aminoglycosides and Pseudomonas overestimates aminoglycoside susceptibility. 3 ' It is not surprising, then, to find a dismal cure rate in endocarditis caused by Pseudomonas aeruginosa despite the fact that in vitro susceptibility data show MICs to be very low. 33 Improved survival can be accomplished by higher doses of the aminoglycoside. 34 Another problem with MICs and Pseudomonas aeruginosa is encountered with the beta-lactam antimicrobial agents. The antimicrobial activity of these agents against Pseudomonas is impaired by large inocula as opposed to aminoglycosides whose activity is not influenced by large inocula. 35 ' 37 In addition, the beta-lactam antipseudomonal agents do not inhibit elongation of most Pseudomonas aeruginosa strains during the first few hours, although they do inhibit bacterial division during this t i m e . 3 5 8 MIC tests cannot differentiate satisfactorily between suppression with eventual lysis and suppression in which regrowth can occur. Time-kill curves obtained by continuous turbidimetric monitoring reveal differences in the responses of the bacteria which do not correlate with simple conventional MIC results. 35 ' 36 Finally, the beta-lactam antipseudomonal agents may be inactivated by induced beta-lactamase, or "trapped" by the beta-lactamase barrier phenomenon, 39 ' 40 neither of which may be detected by conventional MIC testing. Clinical trials have now shown that the newest and "more active" beta-lactam agents are similar to carbenicillin 41 in that in vivo development of resistance can rapidly occur and result in therapeutic failure. 42 " 44 Disc diffusion susceptibility testing of Pseudomonas aeruginosa is not without problems either. It is affected by the cation effect of the agar. 26 " 38 ' 45 ' 46 Moreover, an analysis of error rates for disc diffusion testing of Pseudomonas versus aminoglycosides revealed that unacceptably high rates of error (2% to 3% false-susceptible) were associated with all fixed endpoints. 47 ' 48 Because of this, some microbiologists feel that the standardized disc diffusion test should not be used for testing Pseudomonas aeruginosa against aminoglycosides.
It would be anticipated that the rapid automated susceptibility test methods are going to have difficulties with Pseudomonas aeruginosa. This appears to be the case. 49 Additional experience with these automated systems will undoubtedly result in more examples of false susceptibilities as has been the case with methicillinresistant Staphylococcus aureus.

EPIDEMIOLOGICAL TYPING SYSTEMS
Although neither the pathogenicity nor the lethality of P. aeruginosa is related to the strain type, such typing may be useful as an epidemiologic tool. A number of typing schemes have been developed for such purposes because biochemical typing and antimicrobial susceptibility patterns have not proven to be useful markers. These methods include seroagglutination, bacteriophage typing and pyocin typing.
There are several similar serologic typing systems in use. 50 The main differences are in their selection of strains and in their method of preparation of the strain used to provide the antisera. Serotyping is easy to perform and inexpensive with commercial antisera being currently available. In many epidemiologic settings, seroagglutination can rapidly and easily determine whether two isolates are identical. When larger numbers of isolates are involved, a combination of typing methods can be used to insure better discrimination between strains.
Another typing system involves bacteriophages. 52 ' 53 The pattern of lysis by a standard set of bacteriophages allows an unknown strain to be typed. Similarly, the pattern of growth inhibition of indicator strains by the pyocin type produced by an unknown strain forms the basis for the pyocin typing method. 54 No single method is completely adequate for epidemiologic purposes; a combination of two or more of these methods is often needed. SUMMARY P. aeruginosa is widely distributed in nature and in the hospital environment with a predilection for moist areas. Its inherent resistance to many antimicrobials and its ability to produce many enzymes contribute to its pathogenic potential as both a primary and a secondary cause of infection. It is easily grown and identified in the microbiology laboratory. However, susceptibility testing remains a problem. Currently, the best approach to treatment is an aminoglycoside and an antipseudomonal beta-lactam antimicrobial. Typing can differentiate strains, but should be reserved for specific epidemiologic problems.