The Truth About Therapeutic Drug Monitoring
World Small Animal Veterinary Association World Congress Proceedings, 2007
Hervé P. Lefebvre, DVM, PhD, DECVPT
UMR 181 Physiopathologie et Toxicologie Experimentales INRA, ENVT and Department of Clinical Sciences, National Veterinary School
Toulouse, France

Therapeutic drug monitoring (TDM) refers to the quantitative measurement of drugs in serum/plasma (or other biological fluids). Practically, it consists in blood collection from a patient receiving a drug, serum/plasma assay of the drug concentration and comparison with a therapeutic range (Gross, 1998). The main purpose of TDM is to assist the practitioner in drug regimen adjustment to reach an optimal drug concentration ensuring an adequate therapeutic response without drug-induced adverse effects. Drugs for which TDM has been proposed are anti-epileptic drugs, cardiovascular drugs, antimicrobials, theophylline, cyclosporine and thyroid hormones. Some reviews about TDM are available in the veterinary literature (Neff-Davis, 1988; Boothe, 1999). However, usefulness of TDM in veterinary medicine is currently highly questionable because of several major limits.

1. Potential Benefits of TDM

Without TDM, the most frequent, empirical and pragmatic approach is to initiate the treatment with the drug and to observe what happens. For example, for anti-epileptic drugs, the dog after starting therapy is then observed to determine whether or not seizures still occur. If seizures occur, the dose is increased, and the dog is again observed and the dose increased if necessary until the seizures are either controlled (pharmacodynamic effectiveness) or adverse effects appear. If adverse effects occur, the dose is decreased and another drug is added. TDM allows theoretically individualized drug therapy in a more appropriate and rapid way. For example, the new dose to be given can be estimated by the following equation:

New dose= (Old dose x target PDC)/measured PDC
With PDC, the plasma drug concentration.

The major indications of TDM are suspicion of drug toxicity, lack of therapeutic response, assessment of compliance, therapy assessment following a change in dosage regimen or clinical state of the patient, potential drug interactions and when manifestation of toxicity and disease state are similar. TDM is particularly helpful for drugs with a narrow therapeutic index, large inter- and intraindividual variability in pharmacokinetics and therapeutic response, when treatment fails on standard dosing, when the therapeutic effect is difficult to monitor, or when the drug is used to prevent the occurrence of clinical signs (e.g., anti-epileptic and anti-arrhythmic drugs).

2. Current Limits of TDM in Small Animal Medicine

TDM May Be Risky When Inappropriately Performed

Drug dosage adjustments based on the assumption that measured drug concentrations are reliable predictors of clinical response may be risky. Such adjustments may lead to therapeutic failure or toxicity. They should not be done on the basis of TDM alone. The drug concentration must be always interpreted in light of the clinical response to the treatment, the characteristics (age, body weight, breed, species...) and clinical status (e.g., chronic kidney disease, dehydration...) of the patient, the dosage regimen used, the pharmacokinetic characteristics of the drug, and other covariables (e.g., plasma potassium and ECG for digoxin).

TDM is Possible Only For Some Drugs

Assays are only available for a limited number of drugs used in human medicine. Development of such assays is relatively expensive as their validity should be demonstrated through clinical trials. Moreover, as assessing drug concentration could suggest that the drug is more toxic than that of competitors for which no assay is available, drug companies are not prone to encourage development of TDM. Clinical trial--derived pharmacokinetic information may be extremely useful to define therapeutic ranges of concentration.

Serum/Plasma Drug Concentration is Highly Variable After Dosing

Within and between-individual variability decrease the predictive power of reference values for clinical response. Absorption may also vary with the drug formulation. Effects of physiological variables (e.g., age) and concomitant disease (e.g., renal failure) on serum/plasma concentrations are not well documented in small animal medicine, which may lead to misinterpretation of the observed concentrations. TDM assesses only total drug concentration. Most drugs are bound to serum proteins to various extents, and only free drug is pharmacologically active. In certain conditions, the free drug concentration may be significantly higher than expected from the measured total concentration, and consequently toxicity may occur (Dasgupta, 2002). Noncompliance (e.g., missed doses, improper dosing interval, altered dose) is also another issue for TDM, as the dose level will be increased erroneously.

Technique and Timing of Blood Sampling Are Not Standardized

This is particularly critical for drugs with a short half-life (e.g., aminoglycosides, theophylline). The timing of blood sampling after initiation of therapy on a stable dose required that about 5 half-lives elapse before steady state concentrations can be assumed. Steady state concentrations may be reached earlier if a loading dose has been administered. However, drug with long half-lives should be monitored before steady-state is achieved to ensure that patients with concomitant disease (e.g., chronic kidney disease) are not at risk of developing toxicity. When half-life is very short compared to the dosing interval, the drug cannot accumulate with repeated administration and TDM can be performed following the first dose. Plasma/serum concentrations change throughout a dosage interval and the time of blood sample relative to the time of administration of the dose must be known for appropriate interpretation of plasma/serum level. Blood is generally collected at the end of the dosage interval, just before the next administration (trough concentration). Peak concentrations may be also measured (e.g., antibiotics). The technique of blood collection also may be a source of error. For example, serum separation tubes should be avoided because silicon gel in the tubes can bind to the drug, leading to falsely lower concentrations (Boothe et al, 1996).

Assay Methods Are Not Validated in Dogs and Cats

The validity of drug assays depends on the assay specificity for parent drug, pro-drug and metabolites which may be different between humans and animal species. Accuracy, precision and sensitivity should be also determined. Only unbound free drug interacts with receptors to produce pharmacodynamic effects but free drug monitoring is not performed routinely. Drug assay interferences have not been investigated in small animals. Moreover, results should be available within a clinically meaningful time interval especially when drug toxicity is suspected. In many laboratories, the assays are performed in batches at fixed times and the result for a given patient consequently may not be available rapidly. Moreover, the cost of repeated measurements of drug concentration in a given patient over the treatment period is expensive.

Calculation for Dosage Regimen Adjustment

Dosage regimen adjustment is one of the major consequences of TDM. A clinical pharmacologist should be consulted when precise dosage adjustment is required (especially concerning the decision to modify the dose versus the dosing interval) because generalist practitioners are not familiar with these approaches. The target plasma drug concentration should be known to establish the new dose. Unfortunately, target plasma drug concentrations have never been rationally established for many drugs in small animal species.

Reliability of the Therapeutic Range is Questionable

TDM required knowledge in pharmacokinetics, pharmacodynamics, pharmacogenomics, pathophysiology and medical management of patients in order to achieve safe and effective therapy. Therapeutic ranges are based on probabilities. Plasma/serum concentration above or below the reference values are associated with an increased probability of adverse effects or unsatisfactory clinical response, respectively. The drug reference interval, like reference intervals for plasma biochemical variables, depends on the similarity between the reference population in which they were established and the subject being tested. In dogs and cats, to our knowledge, no prospective study has rationally established reference values in a well-defined population. Moreover other factors (e.g., age, breed, preanalytical and analytical conditions) may affect drug concentrations. The therapeutic ranges proposed in the veterinary literature are generally extrapolated from human patients. Because of potential major differences in pharmacokinetics and pharmacokinetics/pharmacodynamics relationships between humans and small animals, use of such therapeutic ranges is inappropriate. Therapeutic ranges for serum bromide and phenobarbital concentrations have been determined in epileptic dogs but retrospectively (Trepanier et al, 1998).


Measuring a concentration in isolation is unlikely to be a useful and relevant procedure. There is a current need for individualization of drug dosage regimen but there are still too many issues to encourage generalist practitioners to perform TDM. Further investigations should be performed to validate drug assays in the canine and feline species, and to define rationally therapeutic ranges in specific veterinary populations. Population pharmacokinetics is also probably a promising tool in this field. Usefulness of TDM in human medicine remains controversial. In 2007, the Cochrane Database of Systematic Reviews mentioned for example that "No evidence was found to indicate that the routine measurement of serum drug concentrations to inform drug dose adjustments is superior to drug dose adjustments made on clinical grounds alone in newly-diagnosed epilepsy patients treated with a single drug....this review does not exclude the possibility that TDM might be useful in patients with newly-diagnosed epilepsy, nor does it exclude the possible usefulness of monitoring in special situations or in selected patients".


1.  Boothe DM, Simpson G, Foster T. Effects of serum separation tubes on serum benzodiazepine and Phenobarbital concentrations in clinically normal and epileptic dogs. Am J Vet Res 1996;57:1299-1303.

2.  Boothe DA. Therapeutic drug monitoring. In: Small Animal Clinical Diagnosis by Laboratory Methods, 3rd ed., Willard MD, Tvedten H, Turnwald GH, WB Saunders, Philadelphia, 1999, pp. 348-355.

3.  Dasgupta A. Clinical utility of free drug monitoring. Clin Chem Lab Med 2002;40:986-993.

4.  Gross AS. Best practice in therapeutic drug monitoring. Br J Clin Pharmacol 1998;46:95-99.

5.  Neff-Davis CA. Therapeutic drug monitoring in veterinary medicine. Vet Clin N Am: Small Anim Pract 1988;18:1287-1307.

6.  Trepanier LA, Van Schoick A, Schwark WS, Carillo J. Therapeutic serum drug concentrations in epileptic dogs treated with potassium bromide alone or in combination with other anticonvulsants: 122 cases (1992-1996). J Am Vet med Assoc 1998;213:1449-1453.

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Hervé P. Lefebvre, DVM, PhD, DECVPT
National Veterinary School of Toulouse

MAIN : Pharmacology : Therapeutic Drug Monitoring
Powered By VIN