The role of mass spectrometry in antibiotic stewardship

Highlights

• •EU guidelines claim the need for widespread availability of TDM.
• •TDM of antibiotics is recognized as an important element of antibiotic stewardship.
• •However, availability of analytical services for antibiotic TDM is limited.
• •ID-LC-MS/MS instruments are still mainly restricted to centralized facilities.
• •Hospital laboratories should implement ID-LC-MS/MS for TDM of antibiotics.

The spread of bacterial strains that have acquired a high degree of resistance against antibiotics are well recognized as a big problem for human health worldwide. Analyses from the European Centre for Disease Prevention and Control (ECDC) in 2009 estimated that infections with highly resistant hospital bacteria account for 25,000 cases of death in Europe each year. Healthcare costs and productivity losses have been estimated to be at least EUR 1.5 billion [1]. With first pan-resistant bacteria finally have arrived, a scenario of a “post-antibiotic era” is meanwhile discussed. The situation is particularly dangerous since it seems that the pharmaceutical industry is addressing chronic diseases more than acute infections due to commercial considerations, although some new antibiotic substances have been developed in the last years [2]. With a lack of new antimicrobial drugs coming to market it is essential that the currently available compounds are applied in an optimal way.

The extensive and often inappropriate use of antibiotics is a main reason for this emerging threat to mankind [3]. Inappropriately low drug exposure of bacteria in human infection sites will increase the risk to both: therapeutic failure and development of resistance.

It is well documented that the threat of multi-resistant bacteria differs between nations; this is mostly attributable to the practices of prescribing antibiotics as well as hospital hygiene management, with especially high prevalence of multi-resistant bacteria for example in the Eastern Mediterranean Region [4].

Antibiotic stewardship is an emerging concept of inter-disciplinary cooperation, aiming to achieve optimal clinical outcomes in relation to antimicrobial use, minimize toxicity, reduce the costs, and limit the selection and the development of antimicrobial resistant strains. Individualization of antibiotic therapy is becoming an important part in this program [5]. Published guidelines recommend that an infectious diseases specialist and a clinical pharmacist with infectious diseases training, with the inclusion of infection control professionals, the hospital epidemiologist, a clinical microbiologist, and an information system specialist, when possible, should be included in this program [6]. Ideally, a specialist in laboratory medicine might be also involved, who may be responsible for the therapeutic drug monitoring (TDM) of antibiotics. Indeed, TDM of antibiotics is increasingly recognized as an important element of antibiotic stewardship [7], [8], [9].

It is well recognized that critically ill patients with severe infections often have substantial pharmacokinetic abnormalities in relation to healthy volunteers – which are notably addressed in PK (pharmacokinetic) studies performed by the manufacturers. Unpredictable PK situations can result from renal failure but also augmented renal clearance, hepatic abnormalities (both, impaired metabolic activity, and increased metabolism due to cytochrome induction (drug interferences)); impaired micro-circulation due to arterio-venous shunting; hemodynamic instability; altered distribution in “third spaces” such as generalized edema, capillary leakage, altered protein binding; extensive infusion regimens; organ replacement therapy (hemodialysis, extra-corporal membrane oxygenation). These variables can be superimposed to pharmacogenetic variation.

Indeed, many studies have shown that standard dosage regimens often lead to sub-therapeutic blood concentrations in a high proportion of critically ill patients for different compounds. For example, in a study by Taccone et al. [10], predefined targets for patients with severe sepsis and septic shock were reached for ceftazidime in only 28%, for cefepime in 16%, for meropenem in 75%, and for piperacillin-tazobactam in only 44%, despite the inclusion of patients with acute renal failure (27%). Other studies found in part higher proportions, in part lower proportions of inadequate concentrations in critically ill patients [11].

Measurement of the concentration of antibiotics in the circulation aims to adjust the dosage applied of one or several antibiotics to the susceptibility of involved bacteria. Ideally, data on the susceptibility pattern of the pathogen, which is responsible of the infection of an individual patient, is available from microbiological testing of sample materials (in particular from blood culture results in septicemia) and MICs of the different antibiotic substances are reported. In this case, a case-individual lower limit for the target range may be applicable. In cases where minimal inhibitory concentrations are not available, the lower limit for the target range may be based on breakpoints for the causative strain, or on data from hospital surveillance programs. Finally, in cases where the causative pathogen is unknown, the lower limit for the target range can be based on PK/PD (non-species related) breakpoints as for example provided by the European Committee on Antimicrobial Susceptibility Testing [12]. In contrast, the upper limit of the target range (the toxic threshold) is independent of the causative strain and may be available from clinical studies also including clinical trials from the manufacturer.

So far – for antibiotic drugs such as aminoglycosides and glycopeptide antibiotics – TDM mainly aims to avoid side effects and toxicity, whereas innovative approaches mainly aim to optimize treatment efficiency. A constantly growing number of studies suggest that TDM of antibiotic substances can improve the outcome for patients with severe infections [13], [14], [15], [16], [17], [18]. For example, Scaglione et al could show that measurement of drug concentrations and determination of pathogen MIC values with subsequent dose adaption significantly improved the probability of good clinical outcome and pathogen eradication in nosocomial pneumonia.

According to the literature, there is good reason to assume that indeed many lives could be saved with a widespread availability of antibiotic TDM. Recently the Commission of the European Union has released a document “EU guidelines for the prudent use of antimicrobials in human health” [19]. It claims the need for widespread availability of TDM and reminds the IVD industry to their responsibility to improve diagnostic testing with respect to resistant pathogens and life-threatening infections. The fact that antibiotic TDM is imperatively endorsed on an official international level emphasizes the public health dimension of this topic.

A hurdle in the implementation of TDM of antibiotics for laboratories and “customers” is the lack of defined therapeutic ranges for the different antibiotics. For all classes of antimicrobial agents, which may be used for therapeutic drug monitoring in critically ill patients, potential PK-PD targets remain poorly defined with a wide range of potential target concentrations [20]. In an international, multicentre survey of β-lactam antibiotic therapeutic drug monitoring practice in intensive care units performed recently, significant variation in the PK/PD targets (100% fT > MIC up to 100% fT > 4 × MIC) and in dose adjustment strategies used by each of the sites were observed [21]. Guidelines, how to perform TDM for different antibiotic substances similar to the “Consensus Guidelines for Therapeutic Drug Monitoring in Neuropsychopharmacology: Update 2017” would be therefore of great value [22]. In these guidelines detailed recommendations are given for indication, target concentrations, and preanalytics of almost all compounds used in psychiatry.

On the side of the clinical “customers” of antibiotic TDM, a potential other hurdle which limits acceptance and compliance is the need to potentially deviate from manufacturers’ recommendation and drug licensing in order to actually achieve concentration of antibiotics in the desired concentration range, as a partial off-label use. However, in most cases therapeutic ranges based on actual minimal inhibitory concentrations can be achieved, if the causative strain is known. If applicable, another appropriate antibiotic substance might be selected where the therapeutic range can be reached more likely. Results outside the therapeutic range can then be a rationale for individualized dosing in actual cases.

The crucial problem of a widespread use of TDM of antibiotic substances, however, is the limited availability of analytical services for antibiotic TDM. Except the abovementioned assays for aminoglycosides and glycopeptide antibiotics, no antibiotic serum testing is available for standard clinical laboratory analyzer platforms yet. This represents a very substantial shortcoming in laboratory medicine at present. Different technologies could be considered to measure concentrations of antibiotic substances. The advantage of immunoassays would be the easy implementation and daily performance and the possible widespread use of this technology; however, immunoassays are currently only available for the abovementioned antibiotic substances. Instrument costs of HPLC are lower than for isotope dilution (ID)-LC-MS/MS, however, quantification tests by HPLC are not always specific due to co-eluting compounds. This is a crucial problem for tests that are predominantly used for samples from critical ill patients often treated with a wide range of co-medication. ID-LC-MS/MS instruments are still more complex to handle compared to HPLC systems but are highly specific also in extensive poly-medication and organ failure. Variable matrix effects are compensated by the principle of isotope dilution internal standardization, characterizing methods as of higher-order metrological level. Indeed, the only two reference methods for antibiotics (vancomycin [23] and gentamicin [24]) that are listed at the Joint Committee for Traceability in Laboratory medicine (JCTLM) are LC-MS/MS method. Several compounds potentially along with their metabolites can be monitored simultaneously using LC-MS/MS. Since concentrations of antibiotics are in general rather high, most antibiotics are no particular challenge from an analytical perspective. Besides triple quadrupole technology, alternative MS technologies are applicable for TDM of antibiotics, such as Quadrupole-time-of-flight (Q-TOF) and orbitrap technology.

Although LC-MS/MS is widely used now in many countries, this method is often available only in centralized facilities. Here, a high level of technical expertise with highly trained staff is available – which is required to run diagnostic LC-MS/MS tests, since the practicability of systems is still very poor compared to standard laboratory analyzer systems. For the analytes which are currently mainly addressed with LC-MS/MS (e.g., vitamin D, steroids, drug of abuse confirmation testing, pain management), turnaround times (TAT) are often not critical and shipment over extended distances is feasible. For antibiotic TDM, in contrast, TAT is critical – requiring de-centralized testing in hospital laboratories.

For the healthcare system it seems to be important to enable a large number of hospital laboratories to implement LC-MS/MS as the platform of choice for TDM of antibiotics (at least as long as no other technology with a high specificity are available for these substances). Widespread implementation of LC-MS/MS in hospitals also requires continued training and networking in this field, since LC-MS/MS analyses are currently not within the content of technicians training plans. A prerequisite for a widespread use of LC-MS/MS technologies seems to be user-friendly solutions provided by the analytical or the diagnostic industry as soon as possible. This includes software surfaces, simplified handling, improved robustness, improved service support, kitted solutions, and ultimately fully automated MS-based auto-analyzer systems. Such solutions could enable many laboratories to guarantee mostly short TATs.

Furthermore, it would be important to implement external quality management systems including an unbroken chain of traceability as rapidly as possible. This includes manufacturing and specification of reference material preparations, development of reference measurements procedures, and accreditation of calibration laboratories for antibiotic TDM in national reference institutes worldwide.

Antibiotic monitoring in de-centralized settings is a challenge for the community of clinical MS experts; but this emerging field of application also can potentially contribute substantially to shape the arena of clinical mass spectrometry. The quantification of immunosuppressant drugs was an important trigger to implement MS in many laboratories mainly in university hospitals. Due to limited availability and performance of commercial immunometric tests this group of compounds became one of the nuclei of clinical mass spectrometry. However, the burden to manage patients with life-threatening infections is far bigger for healthcare systems compared to the globally rather low number of transplantations (which will rather not see a further substantial growth).

Widespread implementation of resources for antibiotic TDM are a substantial challenge for laboratory medicine, healthcare systems, and industry. However, a wealth of data now strongly suggests that a significant improvement of patients’ outcomes can be expected from these efforts with a substantial decrease of years of life lost through infections worldwide – and mass spectrometric methods are likely to play a major role in this context.

Conflict of interest

None to be declared.

Appendix A. Supplementary data

The following are the Supplementary data to this article: