Laboratory Problems in the Detection of Resistance
Susceptibility testing remains one of the principal functions of the clinical microbiology laboratory. If the results of in vitro susceptibility testing did not correlate with in vivo clinical effectiveness, clinicians would have no use for this information. In fact, today the clinician considers the results of susceptibility testing as being of utmost importance in directing proper antimicrobial therapy, particularly in the current era of increasing resistance. Indeed, detection of resistance has become more important as well as more difficult due to the development and spread of bacterial resistance to antimicrobial agents. Unfortunately, many of these resistant phenotypes have proven difficult to detect by routine susceptibility test methods.
For example, proficiency-testing surveys have demonstrated that decreased susceptibility to beta-lactam agents in pneumococci and decreased susceptibility to vancomycin in staphylococci are difficult to detect. In the past, resistant bacteria were easy to detect in the laboratory because the concentration of drug needed to inhibit their growth was quite high in comparison to that needed to inhibit susceptible strains. This means that today the clinical microbiology laboratory must use a variety of susceptibility test methods, each tailored specifically to a particular pathogen or group of pathogens. Previously, clinical microbiology laboratories attempted to use one method for susceptibility testing, choosing from the broth microdilution, disk diffusion, antibiotic gradient (E-test), and automated instrument methods. However, clinical microbiologists have now recognized that a single method, whether conventional or automatic, is not capable of testing all antimicrobial agents against all microorganisms and detecting all resistance mechanisms. Moreover, any susceptibility testing method used must be accurate, reliable and provide clinically relevant results.
A number of important examples of these varied methods are as follows. Glycopeptide-intermediate Staphylococcus aureus can be detected by a screening method using vancomycin-containing brain-heart infusion agar followed by confirmation by broth microdilution testing. Similar screening methods have been developed to detect vancomycin resistance in enterococci and such active surveillance has been shown to reduce the incidence of vancomycin-resistant enterococcal bacteremia. The susceptibility testing of Streptococcus pneumoniae has similarly evolved in the past two decades from no susceptibility testing at all to a screening process that only looked for penicillin resistance to susceptibility testing of multiple antimicrobial agents that utilizes specific penicillin breakpoints for pneumococcal isolates from cerebral spinal fluid. Pneumococcal resistance in vitro to macrolides and fluoroquinolones has been associated with clinical failures whereas in vitro penicillin resistance is important for meningitis, but not for other pneumococcal infections. Other techniques have evolved for screening for extended-spectrum beta-lactamases, which continue to become more prevalent and treatment of such isolates that may appear susceptible to some extended spectrum cephalosporins has been associated with high failure rates. Most clinical microbiology laboratories now use the erythromycin-induction test (D test) in order to detect the presence of in vitro inducible macrolide-lincosamide-streptogramin B resistance in clindamycin-susceptible, erythromycin-resistant methicillin-resistant Staphylococcus aureus (MRSA). Each of these methods has been shown to be accurate and reliable and is included in the CLSI (formerly NCCLS) M100 Performance Standards for Antimicrobial Susceptibility Testing. This approach using multiple susceptibility testing methods continues to evolve. It is important to appreciate that this shift in the manner in which clinical microbiology laboratories approach the clinical application and interpretation of susceptibility testing has been driven, in part, by the requirement that in vitro susceptibility test results correlate with in vivo clinical effectiveness.
