Therapeutic Drug Monitoring

Paul Lewis

James W. Myers

INTRODUCTION

As antimicrobial resistance issues continue to rise and with very few novel antimicrobials in the pipeline, the emphasis on using the current agents more effectively becomes increasingly important. Understanding the pharmacokinetics, what the body does to the drug, and the pharmacodynamics, what the drug does to the body or organism, will assist in dose optimization. In conjunction with a solid understanding of the pharmacokinetic/pharmacodynamic (PK/PD) relationship is the use of therapeutic drug monitoring (TDM) (Table 54-1).

TDM has been around since the early 1970s and continues to play a role in antimicrobial management today. Several conditions support TDM. Serum drug concentration correlates with therapeutic efficacy or toxicity or both. The drug has a narrow therapeutic index, defined as less than a twofold difference between the minimum toxic concentration and the minimum effective concentration in blood. Interpatient variability in pharmacokinetics is larger than the therapeutic range. Little variation exists at steady state within an individual patient. The drug effect is difficult to assess clinically. Finally, drug assays are available to assist in dosage alteration. The most notable antibiotics utilizing TDM are the aminoglycosides and vancomycin. However, many additional antimicrobials have data to support TDM including antifungals, antiretrovirals, and antimycobacterials. This chapter evaluates current dosing strategies involving the use of TDM.

REASONS TO MONITOR DRUG LEVELS

The goal of TDM is to maximize efficacy of the antibiotic while minimizing the toxicity. Differences in patient pharmacokinetic characteristics necessitate individualized dosing. Plasma concentrations are used as a surrogate marker of therapeutic effect. This must be in conjunction with knowledge of tissue penetration. For instance, in meningitis, an antibiotic with limited penetration into the central nervous system will likely not be successful despite therapeutic plasma concentrations. Many reasons exist for monitoring of drug levels.

Plasma concentration is often a better predictor of success than dose alone.
Variation in population pharmacokinetics exists.
Many drug level assays are readily available with quick turnaround time.
Drug levels can determine if a medication is being taken correctly (compliance with antiretrovirals and antimycobacterials).
Drug-drug interactions may alter pharmacokinetics requiring dosage changes (rifampin may induce the metabolism of voriconazole).
Drug-food interactions may alter pharmacokinetics (food significantly enhances the absorption of posaconazole).

Table 54-1 Common Terms and Abbreviations

Therapeutic Drug Monitoring (TDM): The measurement and interpretation of drug concentration in the plasma to tailor dosing regimens for the safety and efficacy of drug therapy

Minimum Inhibitory Concentration (MIC): The minimum concentration of an antibiotic that will inhibit visible growth of an organism

Peak Concentration (Cmax): The maximum concentration of a drug; it is the time immediately following the end of infusion for an IV medication.

Trough Concentration (Cmin): The minimum concentration of a drug; it is the time immediately prior to the administration of an IV medication.

Area Under the Curve (AUC): The area under the plasma drug concentration vs. time (per 24 hours) curve; this is related to total drug exposure.

Elimination Rate Constant (K): For first-order kinetics, the fraction of drug removed from the body over a period of time (usually 1 hour); can be used to calculate a random concentration (C_t) at the elapsed time (t) so long as the initial concentration (C₀) is known (C_t= C₀^* e^-Kt)

Half-Life (t_1/2): The amount of time needed for a drug concentration to decrease by half; t_1/2 = 0.693/K

Volume of Distribution (Vd): A theoretical term, the apparent volume in which a drug distributes; highly lipophilic medications or medications highly distributed into tissue can have Vds greater than the total volume of the body; Vd = total dose/initial drug concentration

Timing of Levels

Peak
- Peaks are usually performed at least 1 hour after the end of an infusion.
- Should not be drawn immediately postdose due to the alpha distribution phase or tissue distribution
- Extrapolated peak can be calculated using the observed peak and the elimination rate constant (K).
  - True peak = observed Peak * e^Kt
  - t being the time in hours after the end of infusion
- Most useful for measuring concentration-dependant antibiotics (aminoglycosides)
- Primarily influenced by the dose
Trough
- Drawn just prior to the start of a dose
- If not drawn immediately prior to dose, an extrapolated trough can be drawn using the observed trough and the elimination rate constant.
  - True trough = observed trough * e^-Kt
  - t being time in hours prior to when the next dose is due
- Useful when measuring time above minimum inhibitory concentration (MIC)- dependant antibiotics or area under the curve (AUC)-dependant antibiotics when volume of distribution is stable (vancomycin) (Table 54-2)
- Primarily influenced by the dosing frequency

Random level

Drawn without regard to dosing interval

May be useful with pulse dosing (vancomycin in severe renal failure) or in antibiotics with extremely long half-lives (itraconazole)

Table 54-2 Pharmacodynamic Model Predicting Efficacy

T > MIC	Cmax/MIC	AUC/MIC
(Time Dependant)	(Concentration Dependant)	(Total Drug Exposure)
Penicillins	Aminoglycosides	Azithromycin
Cephalosporins	Fluoroquinolones	Clindamycin
Carbapenems	Daptomycin	Tetracyclines
Aztreonam	Metronidazole	Tigecycline
Erythromycin	Telithromycin	Linezolid
Clarithromycin	Echinocandins	Vancomycin
Flucytosine	Amphotericin	Triazole antifungals

Used in various nomograms to predict dosing frequency (Hartford nomogram with aminoglycosides)
Used with a second level (peak or trough) to calculate a patient-specific elimination rate constant (K)
- K =(Ln (C₂/C₁))/ΔT
- C₂ and C₁ are two serum concentrations not separated by a dose.
- ΔT is the time difference between C₂ and C₁.

VANCOMYCIN

Vancomycin is the workhorse for many gram-positive infections, including those resistant to beta-lactams. A glycopeptide, vancomycin was first isolated in 1953 from Amycolatopsis orientalis and finally reached market in 1958. Due to the inability of organisms to develop resistance of the original compound, this new drug was said to vanquish infection and was coined vancomycin. Originally, vancomycin contained many impurities which earned it the term “Mississippi Med.” Concern for nephrotoxicity and ototoxicity limited the use in its early use. As methicillin-resistant Staphylococcus aureus continues to rise, vancomycin has become the mainstay for empiric coverage of hospital-acquired and health care-associated infections. Concern for nephrotoxicity and ototoxicity still exist, though some would argue the toxicities were due to the impurities seen with the earlier preparations. The Infectious Diseases Society of America (IDSA) published guidelines for the therapeutic monitoring of vancomycin in 2009 and serve as the basis for the following recommendations (Table 54-3).

For systemic infections, vancomycin should be given IV.
- To avoid red-man syndrome (a histamine-related adverse reaction), vancomycin should not be administered faster than 1 g/hr.
Vancomycin displays a two-compartment model:
- A central compartment (serum) with high perfusion
- A peripheral compartment (muscle and fat) with less perfusion

Drug is only eliminated from the serum compartment.

Drug must leave the tissue and enter back into the serum to be eliminated.

Table 54-3 Recommended Vancomycin Trough Concentration

<10 mg/L	No therapeutically accepted indication
10-15 mg/L	Complicated urinary tract infections (including pyelonephritis) cellulitis
	Complicated skin and soft tissue infections
15-20 mg/L	Bacteremia/endocarditis
	Line infections
	Pneumonia
	Meningitis/CNS infections
	Osteomyelitis
	Sepsis
	Bone and joint infections
	Intra-abdominal infections
	Febrile neutropenia
	A known pathogen with an MIC = 1 mg/L

Due to the two-compartment model, vancomycin also distributes in two phases.
- The alpha phase lasts 30 minutes to an hour and mostly involves rapid distribution into the tissue.
- During the beta phase, the drug reenters the plasma and is eliminated by the kidneys at a logarithmic rate.
Volume of distribution is roughly 0.65 L/kg (range 0.4 to 1 L/kg).
Based on Matzke equation, the elimination rate constant (K) can be calculated.
- K = 0.00083*CrCl + 0.0044
Vancomycin displays AUC:MIC-dependant killing.
For systemic infections, vancomycin is given as intermitted intravenous infusion.
- Continuous infusion is not likely to significantly impact efficacy.
Dosing should target an AUC:MIC of 400 or greater for greatest chance at clinical success.
Serum trough concentrations are the most accurate and practical measures for efficacy.
Troughs should be considered when therapy is likely to exceed 72 hours.
Troughs should be drawn prior to the fourth dose to ensure steady state has been reached.
- Troughs drawn prior to steady state are not recommended.
Regardless of indication, serum troughs should never drop below 10 mg/L to avoid the development of resistant organisms.
Complicated infections such as bacteremia, endocarditis, osteomyelitis, meningitis, and hospital-acquired pneumonia should target serum troughs of 15 to 20 mg/L.
Any organism with an MIC known to be 1 mg/L, a serum trough level of 15 to 20 mg/L, is needed to attain the target AUC:MIC, regardless of indication.
For patients with normal renal function, doses 15 to 20 mg/kg of actual body weight given every 8 to 12 hours are necessary to achieve the target serum trough.
Loading doses of 25 to 30 mg/kg may be considered for complicated infections.
As renal function declines, dosage adjustments should be made (see Table 54-4 below).

Peak concentrations are rarely needed.

A peak may be needed to determine volume of distribution, if needed.
A peak may be drawn to calculate an elimination coefficient for difficult to dose patients (obesity, amputees, significantly underweight).

Table 54-4 Vancomycin Dosing Nomogram

Dose Recommendation		Interval Recommendations
Weight^a (kg)	Load^b (mg)	Maintenance (mg)	Creatinine Clearance^c (mL/min)	Dosing Interval^d (h)
<40 kg	25-30 mg/kg	500
40-59	1,250	750	>80	12
60-74	1,500	1,000	40-79	24
75-89	1,750	1,250	20-39	48
90-110	2,000	1,500	<20	Pulse dosing or consult pharmacokinetics service if available
111-125	2,000	1,750
>126	2,000	2,000
^aBased on actual body weight. ^bConsider loading dose in seriously ill patients. ^cBased on Cockcroft-Gault equation. ^dMay consider a more aggressive interval (every 8 h) for seriously ill patients. Modified from McCluggage L, Lee K, Potter T, et al. Implementation and evaluation of vancomycin nomogram guidelines in a computerized prescriber-order-entry system. Am J Health Syst Phar 2010;67(1):70-75.