Antibiotics in Intensive Care: A Case for Therapeutic Drug Monitoring?

Antibiotics are powerful drugs that help to treat bacterial infections by preventing bacteria from reproducing or by destroying them. They also play a life-saving role in the treatment of critically ill patients in intensive care units, who are frequently fighting with severe infections. However, the management of such infections in the intensive care unit (ICU) is challenging, even though an early and appropriate antibiotic therapy is associated with increased outcomes [1]. When the most appropriate antimicrobial agent has been chosen, it is important to then provide the right antibiotic dose regimen for the specific patient. Data shows that an optimised exposure results in improved clinical outcomes [2,3]. Preclinical infection models also indicate that higher antibiotic exposures can suppress the emergence of antibiotic resistance [4] — a continuously growing problem due to the overuse of these drugs on a global scale.

The Role of Therapeutic Drug Monitoring

The problem for both physicians and pharmacists is: what is the right dose for the specific patient? Therapeutic Drug Monitoring (TDM) is commonly used when the pharmacokinetics, and therefore the optimal dose, of a drug for an individual patient are difficult to predict [5]. This is also true for antibiotics that possess a small therapeutic window where a dose is effective.

Therapeutic window for antibiotics - Chromsystems

Fig. 1: The therapeutic window for antibiotics

The general difficulty in determining the appropriate dose is even more difficult for critically ill patients with severe infections in ICU. The patient can undergo physiological changes which often leads to the drug’s concentration fluctuating dramatically. In sepsis, the clearance of drugs may be affected due to an increased renal blood flow — up to three times the normal rate, or the opposite, whereby the flow is reduced with a much slower metabolic rate. This can go as far as a treatment failure despite the patient being susceptible to the prescribed antibiotic. The volume of distribution, a pharmacokinetics parameter to characterise drug distribution, can also be altered by a range of factors in intensive care patients, such as physiologic disturbances with an increased severity of illness. Obesity and other influences also need to be considered, and for paediatric patients, it is again a completely different story. Therefore, applying standard dosing regimens under certain circumstances in the ICU can result in sub-therapeutic or toxic concentrations for a large number of patients.

That is where TDM can help: the concentration of the antibiotic is measured in the specific patient and the dosing can be adjusted accordingly. This prevents both over- and underdosing, regardless of whether the patient is critically ill, obese or a very young child. This ensures that the medicines are used appropriately and efficiently, which leads to better patient outcomes and lower costs for the hospital. One great example is linezolid: researchers identified a high variability of drug levels and showed a substantial proportion of ICU patients with insufficiently low levels. The authors concluded that this contributed to a high mortality rate and severity of infection in these ICU patients [6]. Another study showed that age-differentiated standard dosing regimens of linezolid resulted in suboptimal plasma levels in approximately half of the cases among children [7]. Linezolid, however, is also a concern for clinicians, especially for long-term use which can bring about serious dose-dependent adverse effects, such as thrombocytopenia and low blood levels of thrombocytes in the blood [8]. TDM of linezolid would help to prevent both under- and overdosing and help to reduce potentially harmful side effects.

Best Ways for Therapeutic Drug Monitoring

The clinical lab itself needs to answer a range of questions as to how therapeutic drug monitoring should be performed. As with any test in the clinical environment, the assay needs to be fast, robust and possess a high precision and accuracy. There are, however, also special requirements with respect to antibiotic measurements. One problem is the very short half-life of antibiotics, which is usually due to either their degradation after sampling and/or their inherent chemical instability. One example is meropenem which has an extremely low chemical stability of less than one hour [9]. The preanalytical treatment must also be uncritical for drugs with such low stabilities. Furthermore, the sample preparation should be reliable and easy, and the assay needs to provide the required specificity in clinical diagnostics. Overall, the assay and the equipment required should be also cost-effective —there is no use for an assay that can provide great results, if it is not affordable by the hospital.

Clinical labs can rely on a range of assays. Traditionally, immunoassays are used to provide the highest speed, and a minimal amount of lab resources are required due to a high degree of automation. However, the low accuracy of test results can be questioned. The antibodies used in the assay can provoke cross reactivity, leading to falsely higher values. In contrast, HPLC offers a much better accuracy and is gaining traction: it has a simple sample prep and enables the determination of several analytes in one run. HPLC instruments are also an affordable technology that is already widely used in clinical laboratories. Mass spec also has a fairly high upfront investment, but with comparably cheap run-costs once the instrument is installed.

Tab.1: Comparison of the different technologies for TDM of antibiotics

Method Multianalyte CE-IVD EUR/analysis Pros Cons
Immunoassay No Available €€€ Easy to use Direct from primary tube

Low accuracyTime-consuming calibration

HPLC Yes Available High accuracy Affordable widespread technology Manual sample prep
LC-MS/MS Yes No €€ High sensitivity/accuracy Initial costs complexity of technology

Choosing the appropriate method is one thing, the other consideration is the relatively low stability of many antibiotics in the patient. Some drugs, like meropenem, have an extremely low stability of 1 hour in the patient [9]. This raises concerns towards the validity of the results but can also represent a problem when working with quality control and calibrator materials. Chromsystems developed a way of stabilising real patient samples, as well as quality control and calibrator material (see table 2).

Tab. 2: Stability of several antibiotics without or with stabilisation by the Chromsystems assay

Antibiotic Without Stabilisation With Stabilisation
Ampicillin 15 hrs 24 hrs
Cefepime 4 hrs 24 hrs
Ceftazidime 12 hrs 12 hrs
Linezolid 24 hrs 48 hrs
Meropenem 1 hr 6 hrs
Piperacillin 4 hrs 24 hrs

Stewardship Programs

While labs can establish a laboratory routine for the TDM of antibiotics, a more holistic approach is often needed that brings together all stakeholders involved: physician, lab head, hygienist and pharmacist. The main idea behind stewardship programs is that physicians treating the patient and the lab performing the analysis work more closely together with microbiologists and pharmacists. Pharmacists are the bridge between the clinician and the patient, and they must provide the most appropriate antimicrobial with the most optimal dosage regimen. The pharmacist should also work closely with the microbiologist to rapidly identify any inadequate antimicrobial therapies, whether it is inadequate for a specific patient, ineffective at a specific dose or an antibiotic with an overly broad spectrum. A growing amount of research also indicates that these programs can improve the treatment of infections, while also reducing adverse events associated with antibiotic use [10]. Consequently, in 2014, the CDC (Center for Disease Control) in the US recommended all acute care hospitals to implement such Antibiotic Stewardship Programs. A detailed review can be found on the CDC website.

Stewardship programs not only help in improving the quality of patient care, but they also increase cure rates, reduce treatment failures, and increase the frequency of a correct prescription [11,12]. These programs are also aimed at reducing the emergence of antibiotic resistant bacterial strains [13], while helping to save costs [14,15].


Appropriate drug concentrations in antibiotic therapies are critical, especially in intensive care, and helps to prevent over- and underdosing in patients that are difficult to assess. This approach also reduces the risk of new multidrug resistant strains emerging and helps to provide the most effective drug dose to the patient. Properly implemented TDM should account for the low stability of some antibiotics. Ready-to-use CE-IVD solutions are available for HPLC, which are accurate and also affordable for smaller laboratories that support intensive care physicians .

[1] Ferrer R et al. (2014) “Empiric antibiotic treatment reduces mortality in severe sepsis and septic shock from the first hour: results from a guideline-based performance improvement program.” Crit. Care Med. 42(8):1749-55. doi: 10.1097/CCM.0000000000000330.
[2] Rayner CR et al. (2003) Clinical pharmacodynamics of linezolid in seriously ill patients treated in a compassionate use programme. Clin. Pharmacokinet. 42(15):1411-23.
[3] van Lent-Evers NA et al. (1999) “Impact of goal-oriented and model-based clinical pharmacokinetic dosing of aminoglycosides on clinical outcome: a cost-effectiveness analysis.” Feb;21(1):63-73.
[4] Vanscoy B et al. (2013) “Relationship between ceftolozane-tazobactam exposure and drug resistance amplification in a hollow-fiber infection model.” Antimicrob. Agents Chemother. Sep;57(9):4134-8. doi: 10.1128/AAC.00461-13.
[5] Wong G et al. (2014). “An international, multicentre survey of β-lactam antibiotic therapeutic drug monitoring practice in intensive care units.” J Antimicrob Chemother. May;69(5):1416-23.
[6] Zoller M et al. (2014) “Variability of linezolid concentrations after standard dosing in critically ill patients: a prospective observational study.” Crit. Care. Jul 10;18(4):R148. doi: 10.1186/cc13984.
[7] Cojutti P (2015) “Pharmacokinetic/pharmacodynamic evaluation of linezolid in hospitalized paediatric patients: a step toward dose optimization by means of therapeutic drug monitoring and Monte Carlo simulation.” J. Antimicrob. Chemother. Jan;70(1):198-206.
[8] Sasaki et al. (2011) “Population pharmacokinetic and pharmacodynamic analysis of linezolid and a hematologic side effect, thrombocytopenia, in Japanese patients.” J. Antimicrob. Agents Chemother. May;55(5):1867-73.
[9] Pinder et al. (2017) “Therapeutic drug monitoring of beta-lactam antibiotics - Influence of sample stability on the analysis of piperacillin, meropenem, ceftazidime and flucloxacillin by HPLC-UV.” J Pharm Biomed Anal. Sep 5;143:86-93. doi: 10.1016/j.jpba.2017.05.037.
[10] Malani AN et al. (2013) “Clinical and economic outcomes from a community hospital's antimicrobial stewardship program.” Am. J. Infect. Control. 41(2):145-8. doi: 10.1016/j.ajic.2012.02.021.
[11] Kaki R et al. (2011) “Impact of antimicrobial stewardship in critical care: a systematic review.” J. Antimicrob. Chemother. 66(6):1223-30. doi: 10.1093/jac/dkr137.
[12] Nowak MA et al. (2012) “Clinical and economic outcomes of a prospective antimicrobial stewardship program.” Am. J. Health Syst. Pharm. 69(17):1500-8. doi: 10.2146/ajhp110603.
[13] DiazGranados CA et al. (2012) Prospective audit for antimicrobial stewardship in intensive care: impact on resistance and clinical outcomes. Am. J. Infect. Control. 40(6):526-9. doi: 10.1016/j.ajic.2011.07.011. Epub 2011 Sep 21
[14] Sick AC et al. (2013) „Sustained savings from a longitudinal cost analysis of an internet-based preapproval antimicrobial stewardship program.” Infection control and hospital epidemiology : the official journal of the Society of Hospital Epidemiologists of America. 34(6):573-580.
[15] Roberts RR et al. (2009) “Hospital and societal costs of antimicrobial-resistant infections in a Chicago teaching hospital: implications for antibiotic stewardship.” Clin. Infect. Dis. 2009 49(8):1175-84. doi: 10.1086/605630.