Diagnostics: Antibiotic susceptibility

2019-12-12

For determining if an antibiotic treatment will be effective and appropriate, a species-level identification is not sufficient. With some bacteria and in some settings, it can be easy to determine which antibiotic is most appropriate. For example, penicillin resistance among Group A streptococci has never been reported, and as a consequence strep throats can be treated successfully without susceptibility testing. But as resistance among bacteria increases, so does the factual need for susceptibility testing in order to find the best available antibiotic. But susceptibility testing is also a central component of surveillance that can feed information about resistance levels both to clinicians to guide empiric therapy, as well as national or international surveillance systems that monitor resistance levels and trends.

Metrics

There are two main categories of measures to describe the susceptibility of bacteria to antibiotics: MICs and SIR categories. The MIC, or minimal inhibitory concentration is defined as the lowest concentration of an antibiotic that inhibits the visible growth of a bacterium under standardized conditions. This value is then related to predefined breakpoint values, either clinical breakpoints or epidemiological cutoffs.

The SIR system is a classification of the susceptibility in relation to expected clinical effect. A bacterium found in a patient sample can be tested and classified as either “S”, susceptible, “I” susceptible with increased exposure, or “R”, resistant. The terminology used by EUCAST was updated in 2019, when the former definition of “I” as “intermediate” was changed to “susceptible, increased exposure”.

Culture-based methods

The most common methods to determine the susceptibility of a bacterium to antibiotics are based on growing the bacteria in a laboratory. The gold standard method is still broth dilution, a method where the bacterium is exposed to a series of antibiotic concentrations in liquid culture media. While the original method was done in test tubes, the current standard is microdilution with smaller volumes in a handy microplate format. Once the dilution series is prepared and the bacteria are added, the plates or tubes are incubated in 35 degrees Celsius for 16-20 hours. After incubation, the lowest concentration where no visible growth is seen is the MIC.

As this method is laborious, a simpler method called disk-diffusion is used in most clinical laboratories. The test is done on a solid agar medium with paper disks containing antibiotics. The bacteria are spread evenly onto the plate surface before adding the disks. After 16-20 hours incubation, the bacteria will have grown on the plate, except around the disks that contain antibiotics that have an effect on the bacteria. The zone of inhibition is measured and related to breakpoint measures. The size of the zone will vary depending on the bacterium, but also on the antibiotic and the amount of antibiotic in the disk. This method will not give a MIC, but the SIR category.

A variant of this method is the gradient test. In this test, a plastic strip contains a gradient of the antibiotic that is transferred to the surface of the medium. A scale on the plastic strip can then be used to read the MIC after incubation.

Nucleic acid amplification tests (NAAT)

The above described methods determine the resistance of the bacterium to the antibiotic based on its phenotype, or ability to grow while being exposed to the antibiotic. But in many cases, antibiotic resistance is caused by a gene, or a mutation in a gene, that causes the resistance. Examples are mutations in the DNA gyrase gene that cause ciprofloxacin resistance and ESBL-genes that cause resistance to beta-lactam antibiotics. A wide variety of genetic tools can thus be used to detect these genes and mutations, from amplification of the gene of interest to whole genome sequencing. These types of methods are especially valuable for bacteria that are difficult to culture, such as Mycobacterium tuberculosis, or in epidemiological studies that aim to track the prevalence or spread of said genes.

The major downsides of DNA-based methods is that not all resistance is directly linked to a specific gene – and any new genes that emerge are of course not identified by these tests. This leads to the conclusion that DNA-based methods tend to err towards missing resistance if the resistance mechanism is not sought for.

Rapid diagnostics

In recent years, rapid diagnostics have also entered the field of susceptibility testing. A new protocol from EUCAST is, in some cases, able to identify susceptibility in a disk diffusion test within 4-8 hours. Many DNA based methods also give results within a few hours, and a MALDI-TOF can detect some known resistance mechanisms within hours. Newer advances include microfluidics, cellular level imaging and image processing as well as detection of cell-level responses to antibiotics. The field has picked up pace as the consequences of antibiotic resistance have become more evident also among non-specialists.

Considerations for adoption

Aside the more mundane considerations such as cost and laboratory capacity, deciding which diagnostic method(s) to implement depend on factors like throughput, perceived need for rapid diagnosis and, most importantly, what the diagnostic is needed for. For some bacteria, NAAT is the only viable option, but for most bacteria encountered in a clinical laboratory the choice is open. It is important to remember that a diagnostic test only answers the specific question that is asked by the design of the test – in the case of most NAATs, the question is “Is this genetic element present in the sample?”. In the case of culture-based methods, the question is “Is this bacterium susceptible or resistant to these antibiotics?”.

The difference between these questions has led many clinical microbiologists to advocate for culture-based diagnostic methods. As for the new rapid diagnostic tests, an additional question is: “How rapid is rapid – how will the increased speed translate into improved clinical outcomes or prescription?” As an example, the standard methods described in this and previous articles require approximately 48-72 hours from sampling to final diagnosis – for some blood cultures even longer. Even if a new test can shave off for example 12 hours from this, a patient with a severe infection will receive antibiotics before the culture results. However, if a combination of rapid detection and resistance testing could reduce time in each step, or even run species identification and susceptibility tests simultaneously, time savings may be significant. But will that be enough?