Disk diffusion method

Antibiotic sensitivity testing is usually done using either the tube dilution or the agar diffusion (sometimes called “disk sensitivity”) technique. Of these, agar diffusion is by far the more common, and the Kirby-Bauer modification of this technique is the standard procedure. The Kirby-Bauer method involves (1) isolating a bacterial colony from its original growth medium, (2) allowing the bacteria to grow in a broth medium to a predetermined visual density, (3) covering the entirety of a Mueller-Hinton agar plate with the bacterial isolate, (4) placing antibiotic sensitivity disks at intervals on the surface, (5) incubating for 18-20 hours, (6) examining for clear areas around individual disks representing bacterial growth inhibition by the antibiotic-impregnated disk, and (7) measuring the diameter of these inhibition zones. Results are reported as resistant, sensitive, or intermediate, depending on previously established values for zone size based on tube dilution studies most often furnished by the disk manufacturer. Consistent results depend on strict adherence to good technique at each step, as well as the quality of the antibiotic disks and agar used. Variation in potency of antibiotic disks from different shipments and disk or agar deterioration during storage necessitate a good quality control program. The Kirby-Bauer technique has limitations. To be accurate, it can be used only when the bacterium to be tested is aerobic, nonfastidious, and grows rapidly (produces colonies within 24 hours) on the agar growth medium. Thus, it can be used for the Enterobacteriaciae, Pseudomonas organisms, and staphylococci. The Kirby-Bauer method can also be used with H. influenzae, N. gonorrheae, and S. pneumoniae, but a different agar medium is necessary.

Most organisms, fortunately, may be classified as sensitive or resistant. Intermediate sensitivity is a controversial area. In general, many believe that intermediate zones should be considered resistant, although certain organisms, such as enterococci, may be exceptions. If a particular antibiotic with intermediate degree of inhibition is important in therapy, the sensitivity test should be repeated to rule out technical problems. It might be useful to have the sensitivity test performed by the tube dilution method, if available. Another subject of dispute is the need for sensitivity testing in bacteria that almost always are sensitive to certain antibiotics. Penicillin (PCN) sensitive organisms include group A aerobic streptococci, pneumococci, Neisseria organisms, and Corynebacterium diphtheriae. A few PCN resistant or partially resistant strains of pneumococci (about 3%; range, 0%-35%) are being reported, and PCN resistant gonococci (about 2%; range, 0%-6%) are beginning to appear in various areas. At present, many laboratories do not perform sensitivity studies routinely on these bacteria. An occasional laboratory problem is requests by physicians to include additional sensitivity disks in sensitivity panels. Only a limited number of disks can be spaced on a sensitivity agar plate, so that an additional disk either means that another antibiotic must be deleted or else an additional plate must be used at extra expense. In many cases the extra antibiotic requested has a similar sensitivity spectrum to one already on the panel. Laboratories frequently include only one representative of an antibiotic family in sensitivity panels because sensitivity differences among antibiotic family members (e.g., the various tetracyclines) are usually very minor.

Serial dilution method. The other major test procedure is serial dilution (sometimes called “tube dilution”). A standardized concentration of the bacterium to be tested is placed in tubes or microtiter plate wells containing liquid culture medium. Serial dilutions of the antibiotic to be tested are added to consecutive tubes containing the bacterial suspension and the tubes are incubated, usually for 24 hours. The bacterial suspension is cloudy, and any tube in which the bacteria are sufficiently inhibited will become clear due to absence of bacterial multiplication. The last dilution tube (i.e., the tube with the highest degree of antibiotic dilution) that is clear (i.e., still shows acceptable bacterial inhibition) is reported as the minimal inhibitory concentration (MIC). This is defined as the smallest concentration of antibiotic that will inhibit bacterial growth. Therefore, the higher the antibiotic dilution before bacterial growth overcomes the effect of the antibiotic, the smaller the amount (concentration) of antibiotic necessary to inhibit the organisms. The MIC is actually reported in terms of the antibiotic concentration per milliliter of solution that is necessary to inhibit the bacteria (e.g., an MIC of 15 means a level or concentration of 15 µg of the antibiotic/ml of solution). To determine if the organism is sensitive or resistant to the antibiotic, one compares the MIC with the achievable blood level of antibiotic, which varies according to size of dose, frequency of dose, route of administration, and so forth. If the maximal achievable blood level of the antibiotic is less than the minimal concentration of the antibiotic necessary to inhibit the organism (the MIC), the antibiotic presumably will not be effective (or, to consider it in another way, the organism is resistant). In general, this means that the lower the MIC value, the more sensitive the organism is likely to be toward the antibiotic. The same tube dilution procedure is carried out for each antibiotic to be tested with each organism. As a general rule, an adequate serum antibiotic level should be at least two to four times the MIC to compensate for some of the clinical and laboratory variables.

A relatively recent method is called the E test, that combines aspects of Kirby-Bauer with tube dilution MICs. Somewhat simplified, a single bacteria isolate is spread over the surface of a Mueller-Hinton type of agar plate (a few other media have been used experimentally for several special requirement bacteria). A rectangular plastic strip containing a continuous gradient of a single antibiotic along one side and an MIC scale with gradation representing twofold (doubling) dilutions of the antibiotic concentration along the other side is placed on the innoculated plate. After 24 hours of incubation (some have used shorter time periods), the point where a clear zone of bacterial suppression begins after the areas of bacterial growth is read from the MIC scale. More than one antibiotic-impregnated strip can be used at the same time, although the number is consideraly limited by the need to keep the strips far enough apart to prevent interference. Results have generally been reported to correlate 90%-95% with standard MIC results, although some organisms correlate in the 80%-90% range and not all organisms can be tested by this method.

Disk method versus dilution method. There are certain advantages and disadvantages to each antibiotic sensitivity technique. The disk method is cheaper and easier than tube dilution if both are done manually. Semiautomated and fully automated equipment is available for tube dilution, but the cost of the equipment usually restricts this to larger or high-volume laboratories. The disk diffusion method can be used only for certain common, rapidly growing organisms. Tube dilution can be used with more organisms than disk diffusion, but many cannot be tested with either technique. Neither technique can be used for anaerobes, under usual conditions. The disks employed in the Kirby-Bauer disk sensitivity method contain antibiotic concentration based on serum antibiotic levels achieved with usual drug doses. In certain body tissues or body fluids such as urine, the concentration of the antibiotic may be considerably more or less than the serum concentration, and the disk sensitivity result may therefore be misleading. For example, the urine concentration of some antibiotics may be much greater than the serum concentration. Thus, the organism may not be inhibited by the disk (serum) antibiotic concentration but may actually be inhibited at the higher antibiotic concentration in urine. In this case, a disk sensitivity result showing the organism to be sensitive is correct, but a disk result showing the organism to be resistant may not be correct. The opposite could be true if infection took place in a location where the antibiotic concentration was considerably less than the level in serum. Another problem, of course, is that the disk sensitivity method uses a single concentration of antibiotic. The organism might be inhibited at a lower concentration than the disk contains. Even more important, the actual concentration of that antibiotic in the serum of any individual patient may be quite different from the disk concentration for a variety of reasons (differences in dosage, absorption, degradation, excretion rate, etc.).

The tube dilution method shares some of the drawbacks of the disk diffusion method, the principal difficulty being that neither method takes into account the actual antibiotic concentration in patient serum or in the area of infection, these concentrations being unknown. However, the tube dilution method does have the advantage in that it roughly indicates the actual antibiotic concentration necessary to inhibit the organism. As noted previously, if one can learn the theoretical antibiotic concentration from a given dosage at the expected site of infection (in micrograms per milliliter), one can compare this with the level needed to inhibit the organism (the MIC, also reported in micrograms per milliliter) and have a better estimate of probable therapeutic effectiveness. However, this does not guarantee that the actual concentration of antibiotic at the site of infection is the same as the theoretical or experimental concentration. It is also possible to test the effects of antibiotic combinations using the tube dilution method; this is not possible with the Kirby-Bauer method. Finally, it is possible to obtain the minimal bactericidal concentration (MBC) using a modification of the tube dilution method. The MBC is the smallest concentration of the antibiotic necessary to kill at least 99.9% of the bacteria. This may or may not require a higher concentration of antibiotic than the MIC and might be useful information if the patient’s theoretical or actual antibiotic blood level is higher than that required by the MIC but the patient is not responding to therapy.

Bacterial resistance: beta-lactamase. Certain antibiotics, notably penicillin and the cephalosporins, have a certain structural area containing a nitrogen atom known as a beta-lactam ring. A group of bacterial enzymes (of which penicillinase was the first to be recognized) can split the beta-lactam ring and destroy the antibacterial activity of the molecule. These bacterial enzymes are now collectively called beta-lactamase. Certain tests have been devised to demonstrate bacterial production of beta-lactamase. There is considerable variation in the technical details of these tests and some variation in accuracy. Most, however, can be done in less than 1 hour and are reasonably accurate, possibly more so when indicating a positive reaction. A positive beta-lactamase test result suggests that the organism should be considered resistant to penicillin, ampicillin, and the first- and second-generation cephalosporins until results of antibiotic sensitivity studies are available. The test is particularly important in H. influenzae type B infection; up to 20%-30% (range, 6%-38%, depending on geographical location) have been reported to produce beta-lactamase. Staphylococcus aureus or epidermidas are even more likely to produce beta-lactamase, so much so that resistance to penicillin is usually taken for granted pending sensitivity study results. N. gonorrheae may produce beta-lactamase but is usually not tested for this except under special circumstances. Other antibiotic resistance mechanisms also exist.

Methacillin-resistant S. aureus (MRSA) is not only a therapeutic problem but also presents difficulties in susceptibility testing. Standard susceptibility protocols will not demonstrate methacillin resistance in a significant minority of these organisms. There are certain changes in temperature and length of incubation, salt content of the media, and density of the bacterial inoculum that will provide the greatest rate of detection. Many laboratories do not use some or any of these recommended changes. Some antibiotics, such as certain cephalosporins, may sometimes appear to be effective against MRSA by in vitro sensitivity tests, whereas they will usually not be effective if given to the patient.