ANTIMICROBIAL SUSCEPTIBILITY TESTS
Disk Diffusion Testing
Antibiotic Gradient Diffusion Testing
Dilution Susceptibility Tests
Serum Bactericidal Test
Combination Antimicrobial Therapy
Antiviral Susceptibility Testing
Under ideal circumstances, the use of antimicrobials would follow a logical sequence of events in which a pathogen is isolated from a clinical specimen, in vitro antimicrobial susceptibility tests are performed, and, based on the results of the laboratory tests, the patient is administered an agent that exhibits inhibitory activity against the offending microbe. In practice, however, etiologic agents frequently are not recovered from patients with an infectious syndrome, and empiric antimicrobial therapy is usually necessary. Nevertheless, in many circumstances, the clinical microbiology laboratory does play an important role in assisting the physician in selecting antimicrobial therapy.
Antimicrobial susceptibility testing is indicated for microbes that are isolated from appropriately collected clinical specimens and that have unpredictable drug susceptibility patterns. For example, susceptibility assays would be appropriate for a Staphylococcus aureus recovered from the blood of a febrile drug user or for a Pseudomonas aeruginosa isolated from the sputum of a patient with pneumonia acquired in an ICU. On the other hand, testing is not required for pathogens that are predictably susceptible to antimicrobials of choice; for example, in vitro tests are not necessary for group A streptococci, as the bacterium remains susceptible to penicillin and erythromycin. With recent changes in the susceptibility of some common human bacterial pathogens, especially Streptococcus pneumoniae and Enterococcus species, in vitro testing that was previously not necessary is now often required. Because of technical and other difficulties, the need to perform susceptibility tests on anaerobic bacteria must be determined on a case-by-case basis. Finally, because of the requirement for special techniques, antimycobacterial susceptibility testing is usually performed at a regional laboratory, and antifungal and antiviral assays at a research facility.
In most hospitals, the microbiologist, infectious disease physician, and pharmacist contribute to the decision concerning which of the many available antimicrobials are to be used in the susceptibility assays. For example, all third-generation cephalosporins are very active in vitro against Escherichia coli and other Enterobacteriaceae, and these drugs have been shown to be effective in clinical trials; nevertheless, an institution may elect to have just one or two of these drugs on the formulary and so the laboratory will test (or report results) for the formulary agents only. Similarly, although a number of cephalosporins demonstrate in vitro activity against methicillin-resistant S. aureus, clinical observations indicate that this class of drug is not effective in treating patients infected with the microbe; accordingly, the laboratory will not determine the susceptibility of methicillin-resistant S. aureus to cephalosporins. In short, the clinician should not expect to find results for every antimicrobial on susceptibility test reports.
Disk Diffusion Testing
Introduced in the 1960s, the disk diffusion test or Kerby-Bauer method is an important technique for evaluating the ability of an antimicrobial to inhibit the growth of a bacterium. Disk susceptibility testing is based on the principle that there exists an inverse linear relationship between the concentration of drug necessary to prevent bacterial proliferation and the zone or area of growth inhibition around an antimicrobial-impregnated disk; thus, the smaller the concentration of drug required to prevent growth, the larger the zone of inhibition. In the procedure, an agar plate (e.g., Mueller-Hinton medium) is streaked with a standardized bacterial inoculum containing approximately 1 × 108 colony-forming units (CFU), and as many as 12 paper disks containing a standard quantity of antimicrobial are dropped onto the plate, which is then incubated at 35°C for 16 to 18 hours. The diameter of the area around each disk that is free from visible bacterial growth is measured to the nearest millimeter. The zones of growth inhibition are compared with reference values, and the results are reported as “susceptible,” “intermediately susceptible,” or “resistant” to each agent tested; of note, a computer program is available that can provide a “calculated minimum inhibitory concentration (MIC)” based on the zone size. The disk technique is an inexpensive, rapid, reproducible, and relatively simple means of evaluating antimicrobial susceptibility; however, standardized procedures must be followed. In addition, the disk diffusion test can be employed only for microbes that grow rapidly on artificial media; thus, this technique is not reliable for evaluating the susceptibility of anaerobic and other fastidious bacteria.
Antibiotic Gradient Diffusion Testing
A relatively new method for assessing susceptibility, the E test uses a plastic strip that is coated with a continuous antimicrobial gradient on one side and has an MIC scale printed on the other. The strip is placed on an agar plate inoculated with the bacterium of interest; following incubation, the intersection of the strip with the edge of the elliptic zone of inhibition is identified, and the MIC is read from the scale. In comparative studies, data from the E test have correlated very well with the results of broth or agar dilution susceptibility assays. The E test has greatly facilitated the ability of microbiology laboratories to determine MICs for fastidious microbes, especially S. pneumoniae. On the other hand, the costs of the E test strips are substantial, and the expense of the assay has limited its use.
Although they are not susceptibility tests, assays that detect the presence of b-lactamase represent an essential component of the microbiologic evaluation of some common bacteria, especially Haemophilus influenzae and Neisseria gonorrhoeae. b-Lactamases are enzymes capable of hydrolyzing the b-lactam ring of penicillins (penicillinases) or cephalosporins (cephalosporinases), thereby inactivating the drugs. Once a pathogen has been isolated from a clinical specimen, b-lactamase production can be detected rapidly by a number of methods, such as the chromogenic cephalosporin assay. The results of screening for b-lactamase are often available 12 to 24 hours before the results of in vitro susceptibility tests are known. Production of b-lactamase by H. influenzae indicates that the pathogen is resistant to ampicillin and amoxicillin.
Dilution Susceptibility Tests
To determine the actual concentration of an antimicrobial required to inhibit bacterial growth, dilution tests must be performed. Broth (or agar) containing twofold dilutions of a drug (usual range of concentrations, 0.125 to 128.0 µg/mL) is inoculated with a standardized bacterial inoculum (5 × 105 CFU/mL), and the cultures are incubated at 35°C for 18 to 24 hours. The lowest concentration of drug that prevents visible growth of the bacterium is the MIC, which is expressed in micrograms per milliliter (µg/mL). In general, an organism is considered to be susceptible when the achievable peak serum concentration is twofold to fourfold greater than the measured MIC. Tests that assess the MIC require strict attention to detail because a variety of technical factors (e.g., size of the bacterial inoculum, concentration of cations in the medium) can influence outcome. To ensure a uniformity in the methodology and interpretation of MIC and other susceptibility tests, most microbiology facilities use the National Committee for Clinical Laboratory Standards (NCCLS).
Broth microdilution tests represent the most prevalent susceptibility assays; agar dilution testing, in which twofold dilutions of an antimicrobial are incorporated into the growth medium (e.g., Mueller-Hinton agar), remains primarily a research tool. The wide use of the broth microdilution assays results from the commercial availability of 96-well microtiter plates containing up to 12 antimicrobials and the automated systems that inoculate and read the plates and that report MIC values for each of the drugs tested. Broth microdilution tests are also useful in evaluating the antimicrobial susceptibility of fastidious bacteria, including anaerobic microbes. Obviously, the testing of an increasing number of antimicrobials against the many microbes isolated daily in a busy clinical microbiology laboratory has been facilitated by the introduction of automated broth microdilution systems. Of note, although these microdilution systems provide MIC data, the results are often reported to clinicians in terms of “susceptible,” “moderately susceptible,” or “resistant.” Finally, antifungal susceptibility is currently performed by means of a broth macrodilution assay.
On occasion, the lowest concentration of antimicrobial required to kill a bacterium must be measured; this value is referred to as the minimum bactericidal concentration (MBC) or minimum lethal concentration (MLC). To determine the MBC, a broth dilution system is employed. Tubes containing a broth medium, a standard bacterial inoculum, and varying concentrations of an antimicrobial are incubated for 18 to 20 hours. Aliquots of broth from tubes with no visible growth (the first of which corresponds to the MIC) are subcultured onto agar medium containing no drugs. Expressed in micrograms per milliliter, the MBC is the concentration of antimicrobial that kills at least 99.9% of the original bacterial inoculum. The MBC determination is a time-consuming test, and fortunately it is not often indicated. In general, an MBC determination should be reserved for patients who have infections in which host defenses do not contribute to cure (e.g., endocarditis) and who are failing to respond to apparently appropriate antimicrobial therapy (e.g., MIC of drug against offending pathogen is low).
Measuring the MBC might reveal tolerance, the phenomenon in which normally microbicidal drugs, such as penicillins and cephalosporins, inhibit the growth of a bacterium but do not kill it. Tolerance has been defined on the basis of the ratio of MBC to MIC, and it is said to be present when MBC/MIC is greater than 16 or 32. Usually associated with S. aureus, tolerance should be suspected in the patient with endocarditis who fails to respond to ostensibly adequate antimicrobial therapy.
Serum Bactericidal Test
The serum bactericidal test is an assay that quantifies the killing ability of serum from an antimicrobial-treated patient. In this assay, serum is obtained immediately before (trough) or soon after (peak) the administration of an antimicrobial. Twofold dilutions of the serum are made with culture medium (Mueller-Hinton broth with pooled human serum), and the mixtures are inoculated with the patient’s bacterial isolate. The serum bactericidal titer is defined as the highest dilution at which more than 99% of the bacterial inoculum is killed. The serum bactericidal test has been used to monitor antimicrobial therapy in patients with infectious endocarditis. The test has also been used in granulocytopenic or immunosuppressed patients who have gram-negative bacteremia and in patients whose therapy has been changed from the parenteral to the oral route, including children with acute osteomyelitis and IV drug users with endocarditis. In general, a peak titer of 1:8 or 1:16 or more is considered desirable. Because its precise role remains uncertain, the test should be reserved for selected adults with bacterial endocarditis or other life-threatening infections that fail to respond to therapy.
Combination Antimicrobial Therapy
Patients who are critically ill with gram-negative pneumonia or other serious infections are often administered a number of antimicrobials, each of which possesses in vitro activity against the offending pathogen. A combination of drugs can produce a bactericidal effect that is equal to, greater than, or less than the sum of the activities of the individual agents; these effects are referred to as indifference, synergism, or antagonism, respectively.
Therapy with predictably synergistic antimicrobial combinations represents an attractive method of treating normal or immunosuppressed patients with serious infections caused by pathogens that are difficult to eradicate. In general, the combination of an aminoglycoside with a b-lactam drug will produce synergy against most aerobic gram-negative bacilli; thus, for the patient with a severe pneumonia caused by P. aeruginosa, Serratia marcescens, or Enterobacter species, the combination of an aminoglycoside with a third-generation cephalosporin, a ureidopenicillin (e.g., ticarcillin), or a monobactam (e.g., aztreonam) has appeal. However, therapy with synergistic combinations has proved to offer a significant advantage over treatment with a single drug or with nonsynergistic pairings in few infections; of note, enterococcal endocarditis is one disease in which the administration of a synergistic combination (penicillin plus gentamicin) is essential for cure. It must be emphasized that predicting which drug combinations will exert synergistic killing in an individual patient can be difficult, and some drug combinations can be antagonistic. In vitro assays (the two-dimensional broth dilution method and the time-kill curve method) have been developed to quantify the bactericidal activity of combinations of antimicrobials. These tests are technically difficult and labor-intensive, and they should be considered only in selected circumstances.
Antiviral Susceptibility Testing
During the past two decades, a number of antiviral medications have been introduced into clinical practice; obviously, the AIDS pandemic has accelerated the widespread use of agents with activity against herpes simplex virus, varicella-zoster virus, cytomegalovirus, and, of course, HIV. Not surprisingly, treatment failures have occurred, and interest in the in vitro susceptibility testing of important viral pathogens has developed. Although not generally available, these assays have the potential to detect resistance and permit the selection of more effective therapy. In the most common clinical circumstance, a change in susceptibility to the antiviral agent is inferred when the expected outcome is not achieved. For example, a rising “viral load” of HIV-1 as determined by a quantitative molecular assay (polymerase chain reaction) suggests the evolution of resistance to the antiretroviral agents the patient is receiving; conversely, a declining viral load indicates susceptibility.
Antiviral susceptibility tests are characterized as genotypic or phenotypic assays. Genotypic assays allow the detection, by polymerase chain reaction amplification, of viral genes that are known to confer resistance; these remain primarily research tools. Phenotypic assays are somewhat similar to the common bacterial susceptibility tests noted earlier. Cell cultures containing various concentrations of the antiviral drug are infected with a standard inoculum of the virus. Following incubation, the activity of the drug is assessed by quantifying phenomena such as decrease in virus yield, inhibition of cytopathic effect, or reduction in the elaboration of viral products. Results are expressed as the drug concentration that produces a 50% inhibition (IC50). Phenotypic assays are commonly employed to assess the susceptibility of herpes simplex virus, varicella-zoster virus, HIV-1, and others. Many factors can influence the outcome of these tests, and the results do not invariably correlate with the clinical response to an agent.
Although the results of susceptibility assays are valuable in guiding antimicrobial therapy, they do not guarantee a response to treatment, and they cannot predict outcome. Host factors and the nature of the infection play critical roles in determining whether a patient will respond to antiinfective agents. For example, the patient with an abscess or an obstructed biliary tract may fail to respond to antimicrobial therapy, regardless of the results of in vitro susceptibility tests. By extension, surgical intervention to drain an abscess or to relieve an obstruction may be more important than potent antimicrobials in improving the likelihood of a favorable outcome. To strengthen the chances of a response, adequate dosages of the antimicrobial must be administered, and to achieve this goal, the clinician may need to review the pharmacokinetics of the agents being considered for use and to monitor serum drug levels during therapy. (A.L.E.)
Brook I. Inoculum effect. Rev Infect Dis 1989;11:361.
For some antimicrobials, especially b-lactam drugs, an increase in the size of the bacterial inoculum produces a substantial elevation in the MIC. This article provides a detailed review of the problem and its clinical relevance.
Gutmann L, et al. Synergism and antagonism in double b-lactam antibiotic combinations. Am J Med 1986;80 (Suppl 5C):21.
In this extensive review, the authors report that most combinations of the b-lactam antimicrobials, such as the ureidopenicillins (e.g., piperacillin), third-generation cephalosporins (e.g., ceftazidime), and monobactams (e.g., aztreonam), produce an indifferent effect, and they caution that antagonism can occur when b-lactam drugs are combined.
Jorgensen JH. Laboratory issues in the detection and reporting of antimicrobial resistance. Infect Dis Clin North Am 1997;11:785.
A contemporay review of the methodology and limitations of in vitro testing in the detection of antimicrobial resistance.
Washington JA. In vitro testing of antimicrobial drugs. Infect Dis Clin North Am 1989;3:375.
A thorough review of the indications, methodology, and limitations of in vitro testing. The importance of technical and biologic variables in the performance of bactericidal tests and in the detection of tolerance is noted.
Weinstein MP, et al. Multicenter collaborative evaluation of a standardized serum bactericidal test as a prognostic indicator in infective endocarditis. Am J Med 1985;78:262.
Employing a standard method to evaluate the serum bactericidal activity in patients with endocarditis, the investigators found that peak titers of 1:64 or more and trough titers of 1:32 or more were associated with a bacteriologic cure in all patients.
Weinstein MP, et al. Multicenter collaborative evaluation of a standardized serum bactericidal test as a predictor of therapeutic efficacy in acute and chronic osteomyelitis. Am J Med 1987;83:218.
Based on data from 48 patients with infections of bone, the authors conclude that for patients with acute osteomyelitis, serum bactericidal titers of 1:2 or greater should be maintained at all times; for patients with chronic osteomyelitis, titers of 1:4 or greater should be achieved.
Wolfson JS, Swartz MW. Serum bactericidal activity as a monitor of antibiotic therapy. N Engl J Med 1985;312:968.
A thorough review of this controversial assay that emphasizes the paucity of clinical data concerning efficacy.