A therapeutic drug monitoring panel measures selected medication levels in blood so treatment can be adjusted with more precision. It is most useful for drugs where too little may fail to control the condition and too much may cause harm. These tests are common for medications such as lithium, digoxin, phenytoin, carbamazepine, valproic acid, gentamicin, vancomycin, tacrolimus, cyclosporine, sirolimus, phenobarbital, theophylline, and some toxicology-related drugs. The result is not interpreted by number alone. Timing, last dose, kidney function, liver function, albumin level, interacting medicines, age, illness severity, and symptoms all shape the meaning. A “high” result may be expected after a dose if the sample was drawn as a peak, while the same number could be dangerous if it was supposed to be a trough. Careful interpretation prevents both undertreatment and avoidable toxicity.
- Therapeutic drug monitoring is mainly used for medicines with narrow safety margins, large person-to-person differences in clearance, or serious toxicity risk.
- A trough level is usually drawn just before the next dose; a peak level is drawn after absorption or distribution, depending on the drug.
- Toxic levels are not the same for every person because symptoms, kidney function, albumin, electrolytes, and drug interactions change risk.
- Incorrect timing is one of the most common reasons a medication level looks falsely high or falsely low.
- Urgent follow-up is more likely when a high level appears with confusion, severe vomiting, tremor, irregular heartbeat, seizures, kidney injury, or breathing problems.
Table of Contents
- What a Therapeutic Drug Monitoring Panel Measures
- Peak, Trough, and Random Levels
- Common Drugs and Typical Level Ranges
- Why Timing Changes the Result
- When Results Need Urgent Attention
- Factors That Shift Drug Levels
- How Clinicians Adjust the Dose
- How to Prepare and What to Ask
What a Therapeutic Drug Monitoring Panel Measures
A therapeutic drug monitoring panel, often shortened to TDM panel, measures the concentration of specific medications in the blood. The test helps clinicians compare the measured level with an expected treatment range and with the person’s symptoms. It is most helpful when the dose cannot be judged safely by symptoms alone.
Many drugs do not need blood level monitoring. Blood pressure medicines, for example, can often be adjusted by blood pressure readings and side effects. A pain medicine may be adjusted by pain relief, alertness, and breathing safety. TDM is different. It is used when the blood concentration itself gives useful information that cannot be seen easily from the outside.
TDM is especially common when a medicine has one or more of these features:
- A narrow therapeutic window, meaning the effective and harmful ranges are close together
- Large differences in metabolism or clearance between people
- Kidney or liver clearance that can change during illness
- Important drug interactions
- Serious consequences from low levels, such as seizure recurrence, transplant rejection, or uncontrolled infection
- Serious consequences from high levels, such as kidney injury, abnormal heart rhythm, neurologic toxicity, or coma
The panel may include one drug or several drugs, depending on the situation. For example, a person taking lithium usually needs a lithium level, along with kidney and thyroid monitoring. A person on gentamicin may need peak and trough antibiotic levels plus creatinine. A transplant patient may need tacrolimus, cyclosporine, or sirolimus trough monitoring, often alongside kidney markers. When kidney clearance is part of the interpretation, results from a kidney function blood test panel can be just as important as the drug number itself.
A TDM result answers three connected questions. Is there enough medicine present to work? Is there so much medicine present that toxicity is more likely? Does the measured level fit the dose, timing, kidney function, liver function, and clinical picture?
The third question is the one that prevents many mistakes. A drug level without timing is incomplete. “Lithium 1.1 mmol/L,” “digoxin 1.8 ng/mL,” or “vancomycin 18 mcg/mL” cannot be interpreted well unless the clinician knows when the sample was drawn, when the last dose was taken, and whether the person is stable, acutely ill, dehydrated, or taking interacting medicines.
Peak, Trough, and Random Levels
Peak, trough, and random levels describe when the blood sample was collected in relation to the dose. These timing labels are not small details. They often decide whether a result is useful or misleading.
A peak level estimates the higher concentration reached after a dose. For some drugs, the peak relates to effectiveness. Aminoglycoside antibiotics such as gentamicin are classic examples, although many hospitals now use extended-interval dosing methods that may rely on different timing strategies. A peak drawn too early can be falsely high because the drug has not finished distributing into tissues. A peak drawn too late can be falsely low because the drug has already started to clear.
A trough level is drawn near the lowest concentration, usually just before the next dose. Troughs are used when sustained exposure, accumulation, or toxicity risk matters. Lithium, tacrolimus, cyclosporine, sirolimus, vancomycin in some settings, and many antiseizure medicines are often interpreted from trough or near-trough samples. A true trough gives a cleaner view of how much drug remains at the end of the dosing interval.
A random level is drawn without strict peak or trough timing. Random levels can still be useful in overdose, suspected toxicity, missed-dose questions, emergency care, or drugs with long half-lives. Random levels are also used with some dosing software when exact dose and sample times are entered. Without those times, a random level may be hard to act on.
| Timing term | When it is usually drawn | Why it is used | Common examples |
|---|---|---|---|
| Peak | After a dose, at a drug-specific time | Checks whether high enough exposure is reached for effect | Gentamicin, tobramycin, amikacin in selected regimens |
| Trough | Just before the next scheduled dose | Checks accumulation, end-of-interval exposure, and toxicity risk | Lithium, tacrolimus, cyclosporine, sirolimus, some vancomycin monitoring |
| Random | At a documented time not tied to peak or trough rules | Helps in overdose, toxicity, missed doses, or model-based dosing | Acetaminophen, salicylate, methotrexate, aminoglycosides in some protocols |
Some tests also use AUC, short for area under the concentration-time curve. AUC estimates total exposure over time rather than a single high or low point. Vancomycin monitoring has increasingly moved toward AUC-guided interpretation for serious MRSA infections because a single trough may not predict both effectiveness and kidney safety as well as overall exposure. A related article on vancomycin trough and AUC monitoring covers that pattern in more detail.
The safest way to read a TDM report is to treat timing as part of the result. A number should travel with the dose, route, schedule, last dose time, draw time, and reason for testing.
Common Drugs and Typical Level Ranges
Therapeutic and toxic ranges vary by laboratory, method, drug formulation, clinical indication, age, pregnancy status, kidney function, albumin level, and local protocol. The ranges below are common adult reference examples, not universal dosing instructions. Some people need lower or higher targets under specialist care.
| Drug or drug group | Common monitoring approach | Typical therapeutic range example | Toxicity concerns |
|---|---|---|---|
| Lithium | Usually 12-hour trough | Often about 0.6–1.0 mmol/L for many maintenance situations; lower targets may be used in older adults | Tremor, vomiting, diarrhea, confusion, ataxia, kidney injury, seizures at higher levels |
| Digoxin | Level at least 6–8 hours after dose, often trough-like | Often 0.5–0.9 ng/mL for heart failure; higher historical ranges may appear on some reports | Nausea, vision changes, confusion, slow or irregular heartbeat, higher risk with abnormal potassium |
| Phenytoin | Total level, sometimes free level | Total often 10–20 mcg/mL; free often 1–2 mcg/mL | Nystagmus, unsteady gait, slurred speech, confusion; free level matters when albumin is low |
| Carbamazepine | Usually trough | Often 4–12 mcg/mL | Dizziness, double vision, low sodium, liver or blood count problems, severe toxicity at high levels |
| Valproic acid | Usually trough; free level in selected cases | Total often 50–100 or 50–125 mcg/mL depending on indication and lab | Tremor, sedation, high ammonia, liver injury, low platelets, pancreatitis |
| Phenobarbital | Usually trough or random with long half-life | Often 15–40 mcg/mL | Sedation, poor coordination, respiratory depression at high levels |
| Theophylline | Timed level based on formulation | Often 10–20 mcg/mL, though lower targets may be preferred | Nausea, tremor, fast heartbeat, seizures, dangerous arrhythmias |
| Gentamicin and other aminoglycosides | Peak/trough or extended-interval protocol | Depends strongly on infection type and dosing method | Kidney injury, hearing or balance toxicity; trough accumulation is important |
| Vancomycin | AUC-guided monitoring or trough-based protocols | AUC target often 400–600 mg·h/L for serious MRSA infections when MIC assumptions fit | Kidney injury, especially with excessive exposure or other nephrotoxic drugs |
| Tacrolimus, cyclosporine, sirolimus | Usually trough | Target depends on organ, time after transplant, rejection risk, and regimen | Kidney injury, neurologic symptoms, high blood pressure, infection risk, metabolic effects |
| Methotrexate, acetaminophen, salicylate | Timed or serial levels | Interpreted by time since dose and toxicity nomograms or clearance curves | Liver injury, metabolic acidosis, kidney injury, marrow suppression, neurologic symptoms depending on drug |
Several patterns deserve special caution.
Lithium has a narrow safety margin and depends heavily on kidney clearance and body water balance. Dehydration, low sodium intake, kidney disease, thiazide diuretics, ACE inhibitors, angiotensin receptor blockers, and nonsteroidal anti-inflammatory drugs can raise lithium levels. When lithium and kidney markers are reviewed together, a focused discussion of lithium level and kidney function may help clarify why a previously stable dose can become unsafe.
Digoxin toxicity can occur even when the level does not look dramatically high, especially in older adults, kidney impairment, or abnormal potassium and magnesium states. The combination of digoxin level, symptoms, ECG findings, kidney function, and potassium is often more useful than the drug level alone. The relationship between digoxin level and potassium is especially important because electrolyte abnormalities can make the heart more sensitive to digoxin.
Phenytoin and valproic acid are highly protein-bound drugs. A total level can look “normal” while the active free level is high if albumin is low, kidney failure is present, or another drug displaces protein binding. Free phenytoin or free valproate can be more informative in these situations. For seizure medication patterns involving carbamazepine, phenytoin, and valproic acid, interpretation also depends on seizure control, side effects, adherence, liver function, and interacting medicines.
Why Timing Changes the Result
The body handles a drug in phases. After a dose, the drug is absorbed, distributed into blood and tissues, metabolized, and cleared. The blood level rises, reaches a high point, then falls. A single blood sample is a snapshot of that curve.
This is why the same drug concentration may mean different things at different times. A gentamicin level shortly after infusion may be expected to be much higher than a trough. A lithium sample drawn 4 hours after a dose may look higher than a properly timed 12-hour level. A digoxin sample drawn too soon after a dose can be misleading because distribution is not complete. A vancomycin trough drawn hours before the next dose is not a true trough.
The most useful medication level reports include:
- The exact dose and formulation
- The route, such as oral, IV, or extended-release oral
- The dosing schedule
- The last dose date and time
- The blood draw date and time
- Recent missed doses or extra doses
- Recent kidney or liver changes
- Current symptoms
- Other medicines and supplements
Steady state also matters. Steady state means the amount of drug going in and the amount being cleared have reached a relatively stable pattern. Many drug levels are easiest to interpret after several consistent doses. Drawing too early after starting or changing a dose may show a transition rather than the eventual level. However, urgent illness, suspected toxicity, overdose, kidney injury, or high-risk therapy may require earlier testing.
Half-life gives a rough sense of how quickly a drug level changes. A short half-life drug may change within hours. A long half-life drug may take days or weeks to settle after a dose change. Phenobarbital and amiodarone-like long half-life drugs behave very differently from aminoglycosides. Lithium may rise quickly when kidney clearance falls, even if the dose has not changed.
For antibiotics, timing links closely to the drug’s killing pattern. Aminoglycosides usually depend on high peak exposure relative to the organism’s susceptibility, while avoiding persistent trough accumulation. Vancomycin depends more on total exposure over time. Beta-lactam antibiotics, when monitored in some intensive care settings, are often judged by how long concentrations remain above the organism’s minimum inhibitory concentration.
For overdose-related levels, the clock may be even more important. Acetaminophen interpretation depends on the time since ingestion and the measured level. Methotrexate monitoring depends on the dose protocol, time after infusion, kidney clearance, hydration, urine alkalinization in some regimens, and leucovorin rescue timing. A salicylate level may need repeating because levels can rise or remain dangerous due to delayed absorption.
When Results Need Urgent Attention
A medication level needs faster attention when the number is high, the timing is believable, and the person has symptoms or organ stress that fit toxicity. Some toxic reactions are subtle at first, so clinicians often look for patterns rather than waiting for severe symptoms.
Emergency evaluation may be needed when a high or possibly toxic drug level appears with:
- Severe confusion, agitation, unusual sleepiness, fainting, or coma
- New seizures or severe tremor
- Severe vomiting, diarrhea, dehydration, or inability to keep fluids down
- Irregular heartbeat, very slow pulse, chest pain, or fainting
- Severe dizziness, double vision, slurred speech, or trouble walking
- Shortness of breath or slow breathing
- New kidney injury, very low urine output, or rapidly rising creatinine
- Severe metabolic acidosis or abnormal blood gases
- High potassium with possible digoxin toxicity
- Suspected intentional or accidental overdose
Lithium toxicity may start with stomach upset, tremor, weakness, drowsiness, or poor coordination. More severe toxicity can cause confusion, seizures, coma, and kidney problems. The measured level is important, but chronic lithium toxicity can be dangerous at levels that overlap with or sit only slightly above some therapeutic ranges because the brain has had time to accumulate lithium.
Digoxin toxicity may cause nausea, poor appetite, confusion, yellow-green visual changes, slow heart rate, heart block, or dangerous arrhythmias. Potassium can be low, normal, or high depending on the situation, but marked hyperkalemia in acute severe digoxin poisoning is concerning.
Theophylline toxicity can cause tremor, vomiting, agitation, fast heartbeat, low potassium, seizures, and life-threatening arrhythmias. Serious symptoms can occur abruptly, and the level may need repeated testing.
Antiseizure medication toxicity can look like neurologic illness. Phenytoin commonly causes nystagmus, unsteady gait, slurred speech, and confusion as levels rise. Carbamazepine can cause dizziness, double vision, low sodium, drowsiness, abnormal heart conduction, and coma at high levels. Valproic acid toxicity may cause sedation, high ammonia, low platelets, liver injury, pancreatitis, and metabolic problems.
Aminoglycosides and vancomycin are watched closely because excessive exposure can injure the kidneys. Aminoglycosides can also damage hearing or balance. Symptoms such as ringing in the ears, vertigo, hearing changes, or worsening kidney function deserve prompt review.
Transplant drugs require special caution because low levels may increase rejection risk and high levels may cause kidney injury, neurologic symptoms, high blood pressure, infection risk, or metabolic complications. For tacrolimus, creatinine trends can change the interpretation of a trough level, which is why tacrolimus and creatinine monitoring are often reviewed together.
Factors That Shift Drug Levels
Drug levels change when absorption, distribution, metabolism, or clearance changes. A dose that worked for months can become too high during dehydration, kidney injury, infection, medication changes, or major weight change.
Kidney function is one of the largest influences. Lithium, digoxin, vancomycin, aminoglycosides, and methotrexate are strongly affected by renal clearance. A rising creatinine or falling eGFR can allow drug accumulation even when the dose is unchanged. For people with unstable kidney function, a creatinine and eGFR interpretation may help explain why a drug level changed.
Fluid status matters as well. Dehydration can raise lithium and worsen kidney clearance. Critical illness can do the opposite for some antibiotics by increasing volume of distribution or increasing renal clearance early in sepsis, leading to unexpectedly low antibiotic levels. Edema, burns, pregnancy, extracorporeal membrane oxygenation, and dialysis can also change drug distribution and clearance.
Liver function affects many antiseizure medicines, theophylline, and immunosuppressants. Carbamazepine, phenytoin, valproic acid, tacrolimus, cyclosporine, and sirolimus can all be influenced by hepatic metabolism and drug interactions. When liver injury is part of the concern, a hepatic function panel can add context.
Albumin changes the meaning of protein-bound drugs. Phenytoin and valproic acid bind strongly to albumin. If albumin is low, the free active portion may be higher than the total level suggests. Kidney failure and competing drugs can also increase the free fraction. In those cases, clinicians may order free drug levels rather than relying only on total levels.
Drug interactions are a frequent cause of unexpected results. Some drugs inhibit metabolism, raising levels. Others induce metabolism, lowering levels. Grapefruit products can raise levels of some medicines that use intestinal CYP3A pathways. Rifampin, some seizure medicines, macrolide antibiotics, azole antifungals, amiodarone, verapamil, diltiazem, and many HIV or transplant-related drugs can cause important interactions depending on the medication being monitored.
Electrolytes can change toxicity risk even when the drug number is not extreme. Potassium and magnesium are especially important for digoxin and some arrhythmia risks. Sodium and hydration are important for lithium. For broader context, an electrolyte panel can help connect symptoms, kidney function, and medication safety.
Adherence also matters. A low level may mean the dose is too low, the drug is not being absorbed, the person missed doses, the sample was drawn late, or another medicine is speeding clearance. A high level may mean too much dose, reduced clearance, an interaction, a sample drawn too soon, or accidental extra dosing. The lab result points toward a question; it does not always answer it by itself.
How Clinicians Adjust the Dose
Clinicians adjust therapy by combining the level with the reason for treatment. A person taking an antiseizure medicine who is seizure-free but mildly above a reference range may be handled differently from someone with side effects. A patient with a serious infection and low antibiotic exposure may need faster dose adjustment than someone being monitored during stable long-term therapy.
Dose adjustment usually follows a few steps.
- Confirm the timing. The clinician checks whether the sample was really a trough, peak, timed post-dose level, or random level. If timing is wrong and the person is stable, repeating the test may be safer than changing the dose.
- Review the clinical status. Symptoms, vital signs, kidney function, liver function, albumin, electrolytes, ECG findings, infection severity, seizure control, transplant status, and toxicity signs all shape the decision.
- Check recent dose history. Missed doses, extra doses, formulation changes, IV-to-oral changes, hospital medication holds, vomiting after doses, or pharmacy substitutions can explain unexpected levels.
- Look for interactions. New antibiotics, antifungals, heart medicines, psychiatric medicines, seizure medicines, diuretics, anti-inflammatory drugs, and supplements can change levels.
- Decide whether to hold, reduce, increase, or continue. Toxic levels may require holding the drug, giving supportive care, using an antidote in selected poisonings, increasing monitoring, or using dialysis for certain severe toxicities. Low levels may lead to a dose increase, loading dose, adherence review, or change in route.
- Plan the next level. The next test is timed based on the drug’s half-life, severity of illness, kidney or liver changes, and how quickly safety decisions are needed.
Dose changes are not always proportional. Phenytoin is a classic example: small dose increases can produce unexpectedly large concentration increases because its metabolism can become saturated. For phenytoin, clinicians often adjust cautiously and consider free levels in people with low albumin or kidney disease.
For antibiotics, modern dosing may use pharmacokinetic software. Bayesian dosing programs can estimate exposure using one or more levels plus patient factors. This approach is increasingly used for vancomycin AUC monitoring and some complex antimicrobial dosing. Still, software depends on accurate dose times, blood draw times, kidney data, and the correct model for the patient.
For transplant medications, targets change over time. Early after transplant, trough targets may be higher. Later, targets may be lower to reduce toxicity while maintaining rejection prevention. Targets also differ by organ type, rejection history, infection risk, interacting drugs, and the rest of the immunosuppressive regimen.
For toxicology-related drugs, the result may trigger a treatment protocol rather than a routine dose adjustment. Acetaminophen levels are interpreted by time since ingestion and may guide N-acetylcysteine treatment. Salicylate levels may require serial monitoring, alkalinization, or dialysis in severe cases. Methotrexate levels after high-dose therapy guide leucovorin rescue and clearance monitoring. For overdose risk, acetaminophen level and liver enzyme interpretation depends heavily on timing.
How to Prepare and What to Ask
The most helpful preparation is accurate timing. Before the blood draw, confirm whether the test should be a trough, peak, random, or timed post-dose level. Do not skip, delay, or double a dose unless the prescribing team gives clear instructions.
Bring or record the following information:
- Medication name, dose, formulation, and route
- Time of the last dose
- Usual dosing schedule
- Any missed, late, vomited, or extra doses
- New prescriptions, over-the-counter medicines, supplements, or herbal products
- Recent dehydration, diarrhea, vomiting, fever, infection, or hospitalization
- Recent kidney, liver, albumin, or electrolyte abnormalities
- Current symptoms, even if they seem mild
For a trough level, the blood draw is usually scheduled just before the next dose. In many cases, the dose is taken after the sample is collected, but instructions differ by drug and setting. For a lithium 12-hour level, for example, blood is commonly drawn about 12 hours after the last dose. For a morning draw, this often means taking the evening dose at a consistent time the night before. The prescriber or lab should give exact instructions.
For peak levels, ask exactly when the clock starts. It may start at the beginning of an infusion, the end of an infusion, or after an oral dose, depending on the drug. Drawing at the wrong time can make the result hard to interpret.
Useful questions include:
- Was this result drawn as a true peak, trough, random, or timed level?
- Does the number match my dose and timing?
- Are my kidney function, liver function, albumin, and electrolytes affecting this result?
- Do any of my other medicines raise or lower this drug level?
- Should the dose change, or should the test be repeated with better timing?
- What symptoms should prompt urgent care?
- When should the next level be checked?
A single abnormal drug level should not be ignored, but it also should not be interpreted in isolation. The safest interpretation connects the number to the person, the timing, and the treatment purpose. That is the real strength of therapeutic drug monitoring: it turns a standard dose into a more individualized plan.
References
- The Steps to Therapeutic Drug Monitoring: A Structured Approach Illustrated With Imatinib 2020 (Review)
- Therapeutic monitoring of vancomycin for serious methicillin-resistant Staphylococcus aureus infections: A revised consensus guideline and review by the American Society of Health-System Pharmacists, the Infectious Diseases Society of America, the Pediatric Infectious Diseases Society, and the Society of Infectious Diseases Pharmacists 2020 (Guideline)
- Antimicrobial therapeutic drug monitoring in critically ill adult patients: a Position Paper# 2020 (Position Statement)
- Model-Informed Precision Dosing: Background, Requirements, Validation, Implementation, and Forward Trajectory of Individualizing Drug Therapy 2021 (Review)
- Software Tools for Model-Informed Precision Dosing: How Well Do They Satisfy the Needs? 2020 (Review)
Disclaimer
Therapeutic drug monitoring results must be interpreted by a qualified clinician who knows the medication, dose timing, medical history, symptoms, and other lab results. Do not change, stop, or restart a monitored medication based only on a blood level without medical guidance. Seek urgent care if a possible toxic level occurs with confusion, fainting, seizures, severe vomiting, breathing problems, abnormal heartbeat, or signs of overdose.
