In Vitro Testing of Antimicrobial Agents

Guidance Documents For In Vitro Testing of Antimicrobial Agents

Whichever method of dilution or disk-diffusion testing a laboratory selects (and laboratories usually select a combination of methods to accommodate their needs), a standardized procedure is of the utmost importance in producing accurate and reproducible results. Various standards organizations in a variety of venues throughout the world have set and published these parameters. In the United States, the Clinical and Laboratory Standards Institute (CLSI) fulfills this role. Professional organizations (e.g., the American Society for Microbiology, Infectious Disease Society of America), private industry, and governmental organizations have worked together within the Subcommittee on Antimicrobial Susceptibility Testing under the Area Committee on Microbiology of the CLSI to establish the necessary standards. Testing conditions, quality control parameters, and interpretative guidelines—including clinical breakpoints for specific antibiotic-organism combinations—are set and published for use by clinical laboratories in the CLSI document M100 ( ) entitled Performance Standards for Antimicrobial Susceptibility Testing . Breakpoints established by the CLSI may differ for a variety of reasons from those set by the European Committee on Antimicrobial Susceptibility Testing (EUCAST) or other standards organizations. One such reason involves different dosing and administration schedules used for the same drugs in different countries. In the United States, challenges also arise when break-points differ between the CLSI and the U.S. Food and Drug Administration (FDA). The breakpoints and interpretations of the FDA regulate the manufacturers of the automated testing systems while the breakpoints and interpretations of the CLSI guide clinical microbiologists and are updated regularly based on the most recent and relevant literature. The CLSI has also published a series of documents that describe standardized procedures for susceptibility testing of fastidious and facultatively anaerobic bacteria ( ), anaerobic bacteria ( ), mycobacteria and aerobic actinomycetes ( ), filamentous fungi ( ), and yeasts ( ). The Subcommittee on Antimicrobial Susceptibility Testing revises the text of most documents on a 3-year cycle and reviews and revises on a yearly basis tables detailing MIC breakpoints or zone-diameter breakpoints. Standardized procedures published by the CLSI represent the standard of laboratory practice in the United States. Laboratories should ensure that the most current tables and texts are on hand and that revisions in the tables and texts are incorporated into the methods and policies of the laboratory when they become available.

Testing Media

It has been demonstrated that the testing components of susceptibility testing may have an effect on the results. With respect to testing media, the CLSI currently recommends Mueller-Hinton agar or broth because they demonstrate good batch-to-batch reproducibility; are low in sulfonamide, trimethoprim, and tetracycline inhibitors; and support the growth of most nonfastidious bacteria ( ). Stringent quality control must be adhered to in commercial or “in-house” preparation of media, as many physical properties of the media directly influence testing. Variations in ion concentrations and pH will also affect testing. For example, excess calcium and magnesium will cause Pseudomonas aeruginosa isolates to appear more resistant when tested against aminoglycosides, but decreased levels may demonstrate false susceptibility. The level of calcium will also affect daptomycin susceptibility results for those species against which the drug is tested, and zinc ion levels influence carbapenem results. As another example, the pH of the media will also directly affect testing; it must be between 7.2 and 7.4 for accurate results. A more acidic pH will cause some drugs—such as aminoglycosides, quinolones, and macrolides—to lose potency and drugs such as penicillin to have increased activity. A higher pH can have the opposite effect.

For some fastidious species, supplements must be added to the Mueller-Hinton media, or specialized media containing required growth factors are necessary. For example, streptococci do not grow well in or on unsupplemented Mueller-Hinton media. This problem can be overcome by adding defibrinated sheep blood to a final concentration of 5% for disk diffusion or lysed horse blood to the medium at a final concentration of 2.5% to 5.0% for broth dilution. Haemophilus influenzae requires Haemophilus test medium when performing broth dilution, and Neisseria gonorrhoeae testing requires GC agar base medium with growth supplement (CLSI M45, 2016). For susceptibility testing of anaerobes, Brucella agar supplemented with laked (i.e., defibrinated and hemolyzed by freezing) sheep blood, hemin, and vitamin K ₁ (Wadsworth method) is necessary to allow testing using the agar dilution method ( ). An exception is seen with the B. fragilis group, which can be tested using the microbroth dilution method with Brucella broth with hemin, vitamin K ₁ , and 5% lysed horse blood.

Additional medium parameters may influence testing. For example, agar depth in plates used in disk-diffusion testing must measure 4 mm; otherwise, the diffusion gradient of the antibiotic will be affected. False resistance may occur if the agar is too deep, and false susceptibility may occur if the agar is not deep enough. When Mueller-Hinton is supplemented with blood, the zones of inhibition may be smaller by 2 to 3 mm than those obtained with unsupplemented agar.

Inoculum

Intuitively, it is evident that the quantity of bacteria present will influence results of susceptibility testing when a finite amount of antibiotic is used in the testing process. To ensure consistency in the quantity of organisms inoculated into the test system, the CLSI has mandated that the prepared inoculum suspension must have a turbidity equivalent to a 0.5 McFarland turbidity standard, which contains approximately 1 to 4 × 10 ⁸ colony-forming units/mL. These standards can be purchased commercially or a BaSO4 suspension can be prepared in the laboratory, the instructions for which are detailed in CLSI documents ( ).

Inoculum preparations can be made by several methods by directly suspending 18- to 24-hour-old colonies from nonselective media in broth or saline until turbidity of 0.5 McFarland is achieved or by suspending several colonies of the organism to be tested into nutrient broth and incubating for several hours until growth results in turbidity equivalent to 0.5 McFarland. The direct colony suspension method is necessary for very fastidious species that require special supplements or media for growth, although the growth method is preferable for species that are difficult to suspend uniformly and when 18- to 24-hour plates are not immediately available. For most commonly isolated species, the two methods are equivalent.

Incubation Conditions

Test systems must be incubated under standardized temperature and atmospheric conditions to ensure uniform results. These conditions have been described and published by the CLSI and FDA for U.S. laboratories. General conditions for disk diffusion and broth dilution of bacteria include a temperature of 35°C ± 2°C with an atmosphere of ambient air for a period of 16 to 24 hours ( ). Many exceptions to this have been noted, however. For example, for disk diffusion, Streptococcus spp. and Haemophilus spp. require incubation in an atmosphere of 5% carbon dioxide. Although zones of inhibition of Haemophilus spp. are measured after 16 to 18 hours of incubation, those for Streptococcus spp. should not be determined until after 20 to 24 hours. Similarly, although susceptibility testing for most antibiotics against Staphylococcus spp. and Enterococcus spp. requires 16 to 18 hours, when oxacillin and vancomycin against Staphylococcus spp. and vancomycin against Enterococcus spp., respectively, are tested, a full 24 hours of incubation is necessary. Differences in testing and incubation parameters among aerobic, facultative anaerobic, anaerobic, mycobacterial, filamentous fungi, and yeast species require that the relevant CLSI documents be studied carefully prior to testing.

Test Interpretation

The inhibitory parameter that forms the basis for the majority of phenotypic susceptibility tests is the minimum inhibitory concentration (MIC) , which is the lowest concentration of antibiotic that inhibits the visible growth of an organism in an in vitro system ( Fig. 58.1 ). MIC results are dependent not only on the interaction between the antimicrobial agent and organism but also on test conditions. These conditions include pH and ion concentrations of testing media, the temperature at which the test system is incubated, the incubation atmosphere, the amount of organism used in testing, and the length of time the system incubates.

Although the MIC is a very useful and reproducible indicator of the interaction of an antibiotic with a specific organism, the information it provides to a physician in the context of the infected patient by itself is limited. The MIC is a snapshot assessment of the interaction of a constant concentration of drug with the organism at a specific point in time under specified conditions in vitro. The MIC must be correlated with the complex and changing environment of the physiology of the patient to allow the physician to predict whether therapy with a particular antibiotic will be successful. This is accomplished through the use of breakpoints , which allow assessment of the efficacy of killing or inhibition of growth of the organism by the antimicrobial. Several types of breakpoints—including microbiological, pharmacokinetic/pharmacodynamic, and clinical—have been identified. Clinical breakpoints, which are intended to separate strains of organisms into categories that predict whether therapy in a patient will be successful, are breakpoints relevant to this short review and are the only type of breakpoint discussed here. Determining clinical breakpoints is complex, difficult, and subjective. The process can involve numerous clinical trials, and clinical breakpoints may vary with the determining standards organization. Breakpoints also change over time, as they are reassessed and revised according to the most recent guiding literature. Breakpoints also vary based on the clinical site of infection; for example, the breakpoint set for meningitis due to S. pneumoniae is different from that set for pneumonia caused by the same organism due to the fact that the concentration of antibiotic achieved varies between sites of clinical infection. Breakpoints are not perfectly clinically predictive for various reasons, such as differences in patient characteristics and immunity, as well as differences in organism virulence and type and site of infection. Excellent discussions of breakpoints and their determination, maintenance, and relevance can be obtained by referring to several review articles ( ; ).

Interpretive categories have been established that allow simplified and easily understood reporting of susceptibility results by laboratories. Many laboratories report these categories in addition to reporting an MIC or, in some cases, in lieu of reporting an MIC. Although highly knowledgeable infectious disease physicians may be able to make informed management decisions with the MIC, this is unlikely to be the case for physicians who do not specialize in this area. Treating physicians depend on guidance from the laboratory regarding interpretation of the actual breakpoints. The basic interpretative categories include susceptible , intermediate , and resistant . There is also a nonsusceptible category and, most recently, a susceptible dose dependent ( SDD ) category for antibacterial AST ( ). Susceptible implies that therapy with the recommended dosage of a particular antibiotic is likely to be effective in eradicating the infection, and resistant indicates that the antibiotic in the appropriate dose has not been shown to have a high likelihood of treatment success in clinical trials.

Reporting an interpretative category of intermediate has several implications. It suggests that the isolate tested may be less inhibited by the usual dose of the antibiotic than those isolates that are categorized as susceptible. It also implies that therapy with higher doses of antibiotic may be effective and that therapy may also be effective if used in anatomic sites in which the antibiotic is concentrated. An example of this latter situation is the use of β-lactam antibiotics in treating urinary tract infection. This category also includes a buffer zone , which is intended to prevent technical factors from resulting in a major interpretive error. An additional category of SDD implies that susceptibility of an isolate is dependent on a higher dosing regimen (higher doses and/or more frequent dosing intervals) than typically used and that the dose to which the patient is exposed is higher than that used when the susceptible breakpoint was established. This category has been in use for antifungal AST but is now used for antibacterial AST because it was noted that intermediate was often misinterpreted as resistant by clinicians, thereby limiting their choices for therapy.

The nonsusceptible category is intended for cases in which intermediate and resistant isolates have not been found or categories of intermediate and resistant have not been defined. It is utilized primarily for new antibiotics. If testing of an isolate yields an MIC value that is above the cutoff concentration established for the breakpoint for a susceptible category, the laboratory can report the isolate as nonsusceptible. This does not necessarily imply that resistance to the antibiotic exists, but it means that experience with isolates that demonstrate MICs at this level is limited; thus, no definitive assessment can be made.

Although reporting of antimicrobial susceptibility testing and categorization of an isolate into one of the interpretative categories are helpful to physicians in managing patients with infection, it must be remembered that antimicrobial activity in vivo depends on many factors. In addition to drug dosage, route of administration, pharmacokinetic/pharmacodynamic characteristics of the drug, and site of infection, patient-specific factors such as immune status, hepatic and renal function, diet, and concomitant therapy with other pharmaceuticals also determine a patient’s response to therapy, thus must be considered in planning treatment. Therefore, antimicrobial susceptibility testing is an adjunct to patient management and does not in itself predict response to therapy.