Therapeutic Drug Monitoring in Veterinary Practice: Indications and Interpretation

By Dr. Zubair Khalid, DVM, MS, PhD · 2026-08-11

Key Takeaways

• Therapeutic Drug Monitoring (TDM) in veterinary practice is indicated for drugs with narrow therapeutic windows, significant pharmacokinetic variability, or concentration-dependent toxicity, aiming to optimize efficacy and minimize adverse effects.
• Sampling strategy is critical; peak concentrations are typically assessed 1-2 hours post-dose for oral medications, while trough concentrations are measured immediately before the next dose for drugs where minimum concentration correlates with toxicity.
• Interpretation requires comparing measured drug concentrations to established species-specific therapeutic ranges, considering factors like renal or hepatic impairment, hypothermia, and concurrent medications that alter drug clearance.
• Subtherapeutic drug levels necessitate investigation into compliance, absorption, or accelerated clearance, while toxic levels mandate dose reduction and evaluation of renal function and potential drug interactions.
• TDM supports antimicrobial stewardship by ensuring drug concentrations exceed the Minimum Inhibitory Concentration (MIC) for pathogens, thereby enhancing efficacy and limiting exposure that drives resistance.
• Emerging technologies like continuous biosensors hold promise for real-time monitoring, but standard laboratory assays remain the current clinical reference standard for dose adjustment decisions.

Therapeutic drug monitoring (TDM) is the clinical practice of measuring plasma drug concentrations and adjusting dosage to maintain levels within a defined therapeutic window. In veterinary medicine, TDM is applied when the relationship between drug exposure and clinical outcome is sufficiently predictable, when pharmacokinetic variability among patients is substantial, and when the cost of toxicity or therapeutic failure justifies the laboratory expense. This article provides a decision framework for the practicing veterinarian: which drugs warrant monitoring, how to time sample collection, how to interpret results in the context of the individual patient, and when monitoring is unlikely to change management. The focus is on drugs with an established evidence base in veterinary patients, with reference to human data where veterinary evidence is sparse.

The reader is assumed to be a qualified clinician familiar with pharmacokinetic principles. The article does not review general pharmacology theory. It addresses the practical questions that arise in daily practice: when is a trough level actionable, what does a subtherapeutic peak mean in an aminoglycoside regimen, and how should concurrent disease alter interpretation of a digoxin or anticonvulsant level. Species differences are noted where they affect sampling strategy or target ranges.

At a Glance

Why Therapeutic Drug Monitoring Works

TDM rests on three premises. First, for a given drug, a plasma concentration range exists above which toxicity becomes probable and below which therapeutic failure is likely. Second, the same dose produces widely different plasma concentrations across a patient population because of variation in absorption, distribution, metabolism, and excretion. Third, the measured concentration, instead of the administered dose, correlates with both efficacy and adverse effects.

The strength of these premises varies by drug. For aminoglycosides, concentration-dependent killing and post-antibiotic effect make peak concentration the primary efficacy driver, while trough concentration predicts nephrotoxicity. For digoxin, a narrow therapeutic index and substantial interindividual variability in clearance make routine monitoring cost-effective. For antiepileptic drugs, the correlation between serum concentration and seizure control is well established in dogs, although the optimal range for each drug remains debated.

Pharmacokinetic variability is the central justification for TDM. Renal function, hepatic enzyme activity, age, body condition, concurrent medications, and disease state all alter drug clearance. In critically ill patients, these factors change rapidly. Hypothermia, for example, impairs clearance for the majority of drugs, leading to drug and metabolite accumulation in plasma, the most striking effect is reduced clearance, and dosages should be decreased considerably during therapeutic hypothermia according to a systematic review of preclinical and clinical studies by van den Broek and colleagues. A patient with sepsis, hypothermia, or acute kidney injury cannot be dosed by formula alone.

The Pharmacokinetic Basis of Sampling Strategy

A single measured concentration is interpretable only if the sampling time is known and standardized. For most drugs, a 2-hour post-dose sample approximates the peak serum concentration, as documented for tuberculosis drugs by Alsultan and Peloquin. Adding a 6-hour sample allows the clinician to distinguish delayed absorption from malabsorption, a distinction that changes management: delayed absorption may respond to formulation changes or administration with food, whereas malabsorption requires investigation of gastrointestinal function or drug interactions.

Trough sampling, immediately before the next dose, is appropriate for drugs whose toxicity correlates with minimum concentration. This applies to aminoglycosides, cyclosporine in some protocols, and digoxin when toxicity is suspected. The choice of peak versus trough sampling must be made before therapy begins, and the sampling time must be recorded on the laboratory submission form. A level drawn at an undocumented time is clinically useless.

Sample handling is a frequent source of error. Prompt centrifugation and freezing are required for many drugs. Isoniazid and ethionamide are not stable in human serum at room temperature, while rifampicin remains stable for more than 6 hours under the same conditions. Veterinary laboratories should be consulted for stability data specific to each drug, and samples should be shipped frozen when transport time exceeds the drug's documented stability window.

When Pharmacokinetic Variability Justifies Monitoring

The decision to monitor a drug should be made prospectively, not after toxicity or therapeutic failure has occurred. Three clinical scenarios justify TDM. The first is a narrow therapeutic window, where the difference between effective and toxic concentrations is small. Colistin exemplifies this: its pharmacokinetics are highly variable and its therapeutic window is narrow, so therapeutic drug monitoring is warranted, as documented by Grégoire and colleagues. The second scenario is a drug with unpredictable pharmacokinetics in a specific patient population, such as critically ill animals with fluctuating organ function. The third is a drug with concentration-dependent toxicity that is severe or irreversible, such as aminoglycoside nephrotoxicity or digoxin cardiotoxicity.

Monitoring is not indicated for drugs with wide therapeutic windows, predictable pharmacokinetics, and reversible toxicity. It is also not indicated when the laboratory turnaround time exceeds the clinical decision window, or when the cost of monitoring exceeds the expected benefit. For antimicrobial stewardship, TDM serves a dual purpose: ensuring that concentrations exceed the minimum inhibitory concentration for the pathogen, and avoiding excessive exposure that may select for resistance. Professional guidance on judicious antimicrobial use from the American Veterinary Medical Association supports this framework, emphasizing that dose optimization is part of responsible stewardship.

Species and Disease State Considerations

Veterinary TDM is complicated by the absence of validated therapeutic ranges for many drugs in many species. Canine ranges for digoxin, phenobarbital, and cyclosporine are reasonably established, but feline, equine, and exotic animal data are often extrapolated from other species or from small studies. Extrapolation is hazardous when protein binding, metabolic pathways, or elimination routes differ. Cats, for example, have reduced glucuronidation capacity, which alters clearance of drugs metabolized through that pathway.

Renal impairment requires particular attention. For drugs cleared renally, the maintenance dose must be reduced in proportion to the decline in glomerular filtration rate, and TDM should be used to confirm that steady-state concentrations remain within the target range. Colistin dosing must be adapted to renal function because colistin methanesulfonate is partly eliminated by the kidney, even though renal clearance of the active moiety is very low. Hepatic impairment similarly reduces clearance of drugs metabolized by the liver, but the relationship between liver enzyme activity and drug clearance is less predictable, making TDM more valuable instead of less.

The Role of Emerging Monitoring Technology

Standard TDM relies on batch laboratory assays with turnaround times of hours to days. This limits the frequency of monitoring and prevents real-time dose adjustment. Electrochemical aptamer-based sensors represent a potential advance: they can measure drug concentrations directly in blood, rapidly and without calibration, and may enable monitoring as frequent and convenient as blood glucose measurement in diabetic patients. A vancomycin aptamer sensor has demonstrated measurement of plasma vancomycin in finger-prick-scale whole blood samples with high precision, and microneedle-based patches have shown continuous real-time measurement of circulating drug pharmacokinetics through interstitial fluid in animal models.

These technologies are not yet clinical tools in veterinary practice. Their relevance to the practitioner is conceptual: they illustrate that the limitation of TDM is often logistical instead of pharmacological. When continuous monitoring becomes available, the indications for TDM will expand to include drugs currently monitored only in severe disease, and closed-loop feedback control of drug delivery may become feasible. For now, the clinician must work within the constraints of standard laboratory assays and interpret each measured concentration in the context of sampling time, patient status, and the drug's documented pharmacokinetic behavior.

The Monitoring Sequence: From Indication to Dose Adjustment

The decision to monitor a drug level should follow a defined sequence. First, confirm that the drug has a demonstrated concentration-effect relationship and a narrow therapeutic index. Second, identify the specific clinical question: Is the drug being absorbed? Is the patient clearing it abnormally fast or slow? Is the current dose producing a level associated with efficacy or toxicity? Third, select the appropriate sampling time based on the drug's pharmacokinetic profile and the formulation used. Fourth, interpret the result against the established target range for that species and indication. Fifth, adjust the dose using first-order kinetics, then decide whether repeat monitoring is needed.

The sequence fails when the clinician skips the first step. Monitoring a drug without a validated target range produces numbers that cannot guide therapy. For drugs such as digoxin, phenobarbital, and cyclosporine, target ranges are supported by clinical outcome data in veterinary patients. For others, the range is extrapolated from human medicine or from in vitro susceptibility data, and interpretation must carry that caveat.

Target Ranges, Sampling Times, and Interpretation

The table below summarizes drugs for which therapeutic drug monitoring has a defined role in veterinary practice. Sampling times assume steady state has been reached unless otherwise stated. Always confirm current reference intervals and species-specific data in the MSD Veterinary Manual or a current veterinary pharmacology text before acting on a result.

The table is a starting point, not a substitute for species-specific reference data. For example, zonisamide has a more favourable therapeutic index than most other antiepileptic drugs, which reduces the urgency of monitoring in some patients, but monitoring remains useful when seizures persist despite an adequate dose or when adverse effects appear.

Troubleshooting the Subtherapeutic Level

A level below the target range triggers a structured investigation. The first question is whether the sample was drawn correctly. A trough sample drawn too early after dosing, or a sample drawn during the absorption phase, will underrepresent the true steady-state concentration. Confirm the dosing history, the time of the last dose, and the sampling time before changing therapy.

The second question is compliance. Owners may miss doses, give the drug at irregular intervals, or administer it with food that alters absorption. Ask specifically about recent changes in feeding routine, formulation brand, or administration technique.

The third question is bioavailability. Phenobarbital and cyclosporine are absorbed variably, and gastrointestinal disease can reduce uptake. If the level is low despite confirmed dosing and correct sampling, consider a malabsorption workup or a change to a parenteral or more bioavailable formulation.

The fourth question is clearance. Hepatic enzyme induction, hyperthyroidism, or concurrent drugs can accelerate metabolism. Phenobarbital induces its own metabolism over weeks, so a level drawn early in therapy may fall as steady state matures. In this situation, repeat the level after the induction phase completes instead of escalating the dose prematurely.

Troubleshooting the Toxic Level

A level above the target range demands immediate dose reduction, not necessarily drug withdrawal. The magnitude of the elevation guides the response. A digoxin level of 3.0 ng/mL with mild anorexia may respond to a 25% dose reduction and recheck in 5 to 7 days. A level of 5.0 ng/mL with ventricular arrhythmias warrants hospitalization, electrocardiographic monitoring, and temporary discontinuation.

Renal function is the most common reason for unexpected toxicity with renally cleared drugs. Digoxin and colistin both require dose adjustment when creatinine clearance falls. Colistin dosing must be adapted to renal function because the prodrug is partly eliminated by the kidney, and the pharmacokinetics are highly variable, which is why monitoring is warranted. For any renally cleared drug, measure creatinine and estimate glomerular filtration rate before adjusting the dose.

Drug interactions deserve equal attention. Ketoconazole raises cyclosporine levels. Cimetidine raises phenobarbital levels. Any newly added drug should prompt a check for metabolic or excretory interactions before the level is rechecked.

Documentation and Repeat Monitoring

Every monitoring event should produce a record that includes the drug, dose, route, formulation, sampling time relative to the last dose, the measured level, the laboratory reference interval, and the clinical decision that followed. This record supports future dose adjustments and provides a defense if a regulatory question arises. The FDA Center for Veterinary Medicine maintains labeling and extralabel use information that may apply when monitoring is performed for approved or extralabel indications.

Repeat monitoring is indicated when the dose changes, when renal or hepatic function changes, when a interacting drug is added or removed, or when the clinical response is inadequate despite an apparently therapeutic level. Routine monitoring at fixed intervals is not necessary for every patient on every drug. A stable epileptic dog on phenobarbital with controlled seizures and no adverse effects may need a level only once or twice yearly, whereas a dog with breakthrough seizures needs a level at the time of the event.

Species and Production System Adjustments

The correct approach changes with the patient. In food animals, tissue residues and withdrawal periods constrain the use of monitored drugs, and regulatory requirements vary by jurisdiction. The World Organization for Animal Health terrestrial standards address residue control and prudent use in production systems. In these patients, the decision to monitor must account for the withdrawal implications of the drug and the feasibility of sampling under field conditions.

In exotic species, reference intervals are rarely established, and monitoring is often limited to drugs with well-characterized pharmacokinetics in related domestic species. Extrapolation carries risk, and the clinician should state the basis for the chosen target range in the record.

In critically ill patients, clearance can change rapidly. Hypothermia, for example, impairs drug clearance and can cause accumulation of drugs and metabolites, increasing the risk of toxicity. A systematic review of hypothermia effects on pharmacokinetics found that impaired clearance is the most striking alteration, and dosages should be decreased considerably in hypothermic patients. Monitoring in this setting should be more frequent, and the target range may need to be adjusted downward to avoid toxicity despite a "normal" level.

Recognized Complications and Early Detection

The principal complications of TDM in veterinary patients fall into three categories: sampling errors, assay interference, and misinterpretation of results in the face of altered physiology. Each has a characteriztic presentation and a discriminating check.

Sampling errors produce results that do not reflect the true steady-state concentration. A sample drawn from the same catheter through which the drug was infused will show a spuriously high concentration, while a sample drawn before the true trough time will overestimate the nadir. The discriminating check is to review the sampling record against the administration record before acting on any result. If the timing cannot be confirmed, repeat the sample.

Assay interference is less common but more insidious. Immunoassays used for drugs such as digoxin and cyclosporine can cross-react with metabolites or endogenous substances, producing results that diverge from chromatographic methods. When a result does not match the clinical picture, ask the laboratory whether the assay is specific for the parent drug and whether known cross-reactants have been reported for the species in question.

Altered physiology is the third failure mode. Hypothermia, for example, impairs drug clearance and can cause drug and metabolite accumulation in plasma, with the most striking effect being reduced clearance of the majority of drugs van den Broek et al., effects of hypothermia on pharmacokinetics and pharmacodynamics. A patient that becomes hypothermic between the initial dose and the monitoring sample will show higher concentrations than predicted, and the clinician may wrongly reduce the dose. The check is to measure body temperature at the time of sampling and to account for it in interpretation.

Common Errors and Corrective Action

The most frequent error is sampling at the wrong time. Less experienced clinicians often draw a single sample at an arbitrary interval after dosing and compare it to a published target range without confirming that the target applies to that sampling time. The corrective action is to fix the sampling window before the dose is given and to record the exact time on the laboratory form.

A second error is interpreting a single concentration in isolation. A subtherapeutic peak may reflect rapid clearance, poor absorption, or a dose that is simply too low. The clinician who immediately increases the dose without checking a follow-up sample may overshoot into toxicity. The corrective action is to repeat the measurement after a dose adjustment and to compare the trend instead of the single value.

A third error is failure to account for protein binding. For highly bound drugs, total concentration may be misleading when albumin is low. The corrective action is to measure albumin concurrently and to interpret the result with the binding fraction in mind, or to request free drug measurement where the laboratory offers it.

Limitations of the Evidence and Areas of Dispute

The evidence base for TDM in veterinary medicine is thinner than in human medicine. Many target ranges are extrapolated from human data or from small animal studies, and the pharmacokinetic behavior of a drug in one species does not reliably predict another. For colistin, the pharmacokinetics are highly variable and the therapeutic window is narrow, which supports monitoring, but the same review notes that experimental issues arise because colistin binds to laboratory materials, making assays technically demanding Grégoire et al., clinical pharmacokinetics and pharmacodynamics of colistin. Clinicians should expect that published ranges may not transfer across species and should interpret results with the individual patient's response as the final arbiter.

Expert opinion still differs on how aggressively to adjust doses in response to a single out-of-range value. Some authorities advocate immediate dose change, others recommend confirming the result before acting, particularly when the patient is clinically stable. The safer course is to repeat the measurement when the result is unexpected and the patient is not showing signs of toxicity or therapeutic failure.

Referral, Consultation, and Reporting

Referral or specialist consultation is warranted when the patient does not respond to dose adjustment despite concentrations within the target range, when the clinician is unfamiliar with the drug's pharmacokinetics in the species being treated, or when the laboratory result cannot be reconciled with the clinical picture. Clinical pharmacologists and veterinary teaching hospitals can provide population pharmacokinetic support and assay methods not available in commercial laboratories.

Laboratory involvement is appropriate when assay interference is suspected, when free drug measurement is needed, or when the laboratory can provide therapeutic ranges validated for the species in question. The laboratory should be told the species, the drug, the sampling time, and the clinical question.

Regulatory reporting applies to adverse drug events and to antimicrobial use in food animals. Practitioners should follow the adverse event reporting pathways of their national regulatory body, and should consult the FDA Center for Veterinary Medicine animal drug information or the equivalent agency in their jurisdiction for reporting requirements. For antimicrobial stewardship, the AVMA antimicrobial use and stewardship resources provide guidance on judicious use, and international standards for food animal treatment are set out in the WOAH terrestrial animal health code. Where TDM reveals a drug concentration that is unexpectedly high or low in a food animal, the clinician should also verify that withdrawal periods remain appropriate, consulting current label and formulary references before releasing the animal to slaughter.

Frequently Asked Questions

How Do I Prioritize TDM When Budget or Laboratory Access Is Limited?

When resources are constrained, prioritize drugs with the narrowest therapeutic windows and the most serious toxicity consequences, such as aminoglycosides, vancomycin, and digoxin. For these agents, a single trough or peak sample can prevent irreversible harm. If the reference laboratory is remote, verify sample stability for the specific drug before shipping. Isoniazid and ethionamide degrade rapidly at room temperature, whereas rifampicin remains stable for over six hours, so drug-specific handling protocols must be confirmed in advance. Where TDM is unavailable, substitute with intensified clinical monitoring, including serial renal function tests for aminoglycosides and electrocardiography for digoxin. Document the limitation and the monitoring plan in the medical record.

Can I Interpret Drug Levels Using Human Reference Ranges in Exotic or Production Species?

Human reference ranges should not be applied directly to veterinary species without supporting pharmacokinetic data. Drug disposition differs substantially across species due to variations in metabolism, protein binding, and elimination pathways. For example, zonisamide pharmacokinetics and tolerability were established in human trials, and extrapolation to dogs or cats requires species-specific validation. When no veterinary range exists, consult the MSD Veterinary Manual for species-specific pharmacology, or contact a veterinary clinical pharmacologist. In food animals, also consider that altered clearance during illness may affect tissue residues, and withdrawal intervals derived from healthy animals may not apply. Regulatory oversight of extralabel drug use is described in FDA animal drug information, and local requirements vary by jurisdiction.

What Should I Do When the Assay Result Does Not Match the Clinical Picture?

First, verify the sampling time relative to the last dose. A 2-hour post-dose sample approximates peak concentration for most drugs, and adding a 6-hour sample distinguishes delayed absorption from malabsorption. Second, check for drug interactions or formulation changes that alter bioavailability. Third, consider assay error, especially for drugs that bind to laboratory materials, such as colistin. If the level is unexpectedly low but the patient responds well, the target range may not apply to that individual, or the free fraction may differ due to hypoalbuminaemia. If the level is unexpectedly high without toxicity, protein binding alterations may explain the discrepancy. Repeat the measurement once before adjusting the dose, and document the reasoning.

How Should I Document TDM Results and Dose Adjustments in the Medical Record?

Record the indication for monitoring, the drug and formulation, the exact sampling time relative to the last dose, the assay result with units, the reference range used, and the clinical status at sampling. State the dose adjustment made and the rationale, citing the target range and any concurrent disease that influenced the decision. Note the planned timing of the next level and the clinical parameters to recheck. If the result was discussed with a specialist or the owner, record that communication. This documentation supports continuity of care and defends clinical decisions if outcomes are questioned. Consistent records also allow retrospective review of whether TDM improved therapeutic outcomes in your practice population.

How Do I Explain the Value of TDM to an Owner Who Is Reluctant to Pay for Repeated Blood Tests?

Frame TDM as a cost-control measure, not an added expense. Explain that one measured level can prevent weeks of ineffective dosing or a hospitalization for toxicity. Use a concrete example, such as an anticonvulsant where the dose is adjusted until the level falls in the therapeutic window, reducing breakthrough seizure frequency. Emphasize that the test answers a specific question: whether the current dose is right for this individual animal. Acknowledge that the first measurement may need repeating, but that subsequent monitoring is usually less frequent once stability is confirmed. Offer to combine blood sampling with other needed diagnostics to reduce handling stress and visit costs.

When Should I Refer a Case for Specialist TDM Consultation instead of Managing It Myself?

Refer when the drug has complex pharmacokinetics that you do not use regularly, when the patient has concurrent organ dysfunction that complicates interpretation, or when the patient has failed to respond despite levels within the target range. Refer also when toxicity has occurred and the cause is unclear, or when managing drugs such as colistin, where the therapeutic window is narrow and pharmacokinetics are highly variable. Specialist input is valuable for designing a sampling strategy in patients with renal or hepatic impairment, where clearance is unpredictable. The AVMA practice resources may help identify referral pathways and clinical pharmacology services in your region.

Related Clinical & Scientific Guides

• Veterinary Formulary Essentials: Navigating Drug References
• Drug Interactions with Antiepileptic Drugs in Veterinary Patients: Managing Polypharmacy
• Drug Interactions with Corticosteroids in Veterinary Patients: A Comprehensive Review

References and Further Reading

• Therapeutic drug monitoring in the treatment of tuberculosis: an update.. 2014.
• Clinical Pharmacokinetics and Pharmacodynamics of Colistin.. 2017.
• Electrochemical Aptamer-Based Sensors for Improved Therapeutic Drug Monitoring and High-Precision, Feedback-Controlled Drug Delivery.. 2019.
• Wearable microneedle-based electrochemical aptamer biosensing for precision dosing of drugs with narrow therapeutic windows.. 2022.
• Effects of hypothermia on pharmacokinetics and pharmacodynamics: a systematic review of preclinical and clinical studies.. 2010.
• Zonisamide. A review of its pharmacodynamic and pharmacokinetic properties, and therapeutic potential in epilepsy.. 1993.
• FDA Center for Veterinary Medicine: Animal Drug Information. FDA CVM.
• AVMA Antimicrobial Use and Stewardship. American Veterinary Medical Association.
• MSD Veterinary Manual, Professional Edition. MSD Veterinary Manual.

Related Articles

• Therapeutic Drug Monitoring of Aminoglycosides in Veterinary Patients: Indications and Protocols
• Therapeutic Drug Monitoring of Cyclosporine in Veterinary Dermatology: Indications and Interpretation
• Therapeutic Drug Monitoring of Digoxin in Canine Cardiac Disease: Indications and Interpretation
• Therapeutic Drug Monitoring in Epileptic Dogs: Antiepileptic Drug Levels
• Corticosteroid Injections in Dogs: Indications, Risks, and Monitoring

This article is educational professional reference material for veterinary audiences. It is not a substitute for veterinary diagnosis, individual clinical judgment, current product labeling, or applicable regulatory requirements.