01Core principlesThe concepts and mechanisms needed to understand the subject.
Therapeutic drug monitoring links a measured concentration to an individual treatment decision. It is most useful when exposure varies substantially between patients, the margin between benefit and harm is narrow, or clinical response alone cannot answer the question promptly. The laboratory value is one observation within a pharmacokinetic story. Absorption, distribution, metabolism and elimination determine how it changes after each dose, while receptor sensitivity and the underlying condition determine what the patient experiences. A result can be analytically accurate yet clinically misleading if it was taken at the wrong time or interpreted against the wrong target.
Before requesting a concentration, write the decision it could change. Possible questions include whether an interaction has increased exposure, whether deteriorating kidney function explains new symptoms, whether missed doses contributed to loss of control, or whether a new maintenance regimen has reached an appropriate level. These questions require different urgency and sampling strategies. A routine trough collected after a stable regimen is not interchangeable with an urgent sample during suspected poisoning. Likewise, a number below a reference interval does not automatically justify escalation if the condition is controlled and the person has adverse effects. The result should sharpen clinical judgment rather than replace it.
Key points
- A drug concentration is useful only when it answers a clinical question. Establish the medicine, indication, formulation, actual doses, sampling time and organ-function context before changing treatment.
- Distinguish a trough, post-distribution sample, peak and timed random concentration. They describe different points in the concentration–time profile and cannot be compared indiscriminately with the same target.
- Lithium monitoring normally uses a sample 12 hours after the last dose; stable levels are generally checked every 3–6 months, more often with higher risk, alongside renal, thyroid and calcium surveillance.
- Routine digoxin concentrations are not recommended. When indicated, sample at least 6 hours after a dose, ideally 8–12 hours; symptoms, kidney function and electrolytes remain essential to interpretation.
- Phenytoin total concentrations commonly use a 10–20 mg/L reference range, but low albumin or altered binding can make the free concentration more informative. Do not escalate solely to normalise a total result.
- Suspected toxicity is a clinical problem even when a concentration falls within a quoted range. Withhold or treat urgently as appropriate rather than waiting for a perfectly timed routine monitoring sample.
- After an abnormal result, record the action, the next dose decision, repeat-sample timing and the clinician responsible. A laboratory flag without an implemented plan is unfinished monitoring.
02Mechanisms and patternsImportant relationships and how to distinguish them.
Consider a concentration when there is suspected toxicity, loss of control, a relevant interaction, organ-function change or uncertainty about exposure. Routine testing without a question can generate unnecessary changes in a stable patient. Ask what would happen if the result were low, expected or high; if none would change management, reconsider the test.
New coarse tremor, ataxia, vomiting or confusion during lithium treatment needs urgent assessment. Digoxin toxicity may involve gastrointestinal, visual or rhythm symptoms, while phenytoin excess can cause nystagmus, diplopia or impaired coordination. A result reported within range does not override a coherent clinical toxicity picture, particularly with organ impairment or electrolyte disturbance. When lithium toxicity is suspected with symptoms, withhold further lithium and obtain urgent concentration and U&Es with specialist advice.
Confirm the exact brand or formulation when clinically relevant, prescribed schedule, actual administrations and missed doses. Ask whether the patient took the usual morning dose before attending phlebotomy. A dose entered on the chart is not proof of administration, and a recently corrected adherence problem may make a previous low concentration unrepresentative of current exposure.
Low albumin can increase the unbound proportion of a highly protein-bound medicine such as phenytoin. Kidney decline, acute illness, pregnancy and interacting drugs can also change the relationship between dose, total concentration and effect. Record the relevant physiological change before assuming that an unexpected result reflects an error in the laboratory.
03Interpreting evidenceInformation, measurements and their limitations.
Consider the information, its meaning and its limitations before deciding what follows.
- 01
A timed sample with dose metadata - Why
- Ensure the concentration can be interpreted against the intended target.
- Interpretation and limitations
- Document the exact last-dose time, sample time, dose, formulation and duration on the current regimen. For lithium, the usual comparison point is 12 hours after dosing. For digoxin, avoid the early distribution period by sampling no sooner than 6 hours, preferably 8–12 hours, when the clinical situation allows.
- 02
Kidney function and relevant electrolytes - Why
- Assess clearance and factors that increase toxicity independently of concentration.
- Interpretation and limitations
- A stable maintenance dose can become excessive after acute kidney injury. Potassium and magnesium disturbances can increase susceptibility to digoxin-related arrhythmia. Interpret the latest concentration together with the trajectory of organ function rather than treating a recent normal result as reassurance for a newly unwell patient.
- 03
Albumin and free drug concentration when appropriate - Why
- Clarify altered protein binding that may distort the total result.
- Interpretation and limitations
- For phenytoin, a low total concentration with hypoalbuminaemia may coexist with adequate or excessive unbound exposure. A measured free concentration can be more useful than chasing a total reference interval. If a correction equation is considered, recognise its assumptions and limitations, especially in severe illness or renal dysfunction.
- 04
Condition-specific response and safety assessment - Why
- Link measured exposure to clinical benefit and harm.
- Interpretation and limitations
- Seizure frequency, mood stability, heart rate, ECG findings and adverse effects supply information a concentration cannot. Some patients are controlled at levels below a usual population interval. Treatment should preserve meaningful benefit while avoiding harm; the laboratory midpoint is not automatically the best target for every indication or person.
04Applied reasoningWorked examples connecting principles to decisions.
01Worked caseReject an incorrectly timed routine lithium comparisonA stable 44-year-old takes lithium at 08:00 and 20:00. She has no toxicity symptoms and normal unchanged kidney function. A routine sample taken at 10:00 measures 0.95 mmol/L; her documented target applies to a 12-hour post-dose sample.+
- 1Calculate elapsed time from the actual 08:00 dose to the 10:00 sample: 2 hours. Identify that this is not the 12-hour comparison point used by her monitoring plan, so the target cannot be applied as if the sample were a trough.
- 2Check that the supplied stability is accurate: ask about tremor, coordination, gastrointestinal symptoms, missed doses, illness and new interacting medicines. If toxicity were suspected, the clinical response would be urgent and would not wait for a routine retimed sample.
- 3For the supplied asymptomatic stable case, avoid changing the dose solely from the 2-hour result. Arrange a correctly timed sample approximately 12 hours after an actual dose, before the next scheduled dose in this twice-daily regimen, with clear phlebotomy instructions.
- 4Give the final action as repeat valid monitoring rather than automatic dose reduction. Independently verify that an 08:00 administration followed by a sample just before 20:00 gives approximately 12 hours, document the timing and assign the result to the clinician managing her lithium.
02ToxicityAct when symptoms outweigh a routine reference rangeA patient taking a monitored medicine develops compatible new adverse effects.+
- 1Assess severity and vital signs, identify the medicine and actual exposure, and seek urgent clinical help where neurological, cardiac or systemic instability is possible. Decide whether further doses must be withheld while the immediate problem is assessed.
- 2Obtain an urgent concentration and relevant safety tests, recording dose and sample times even when the sample cannot meet the routine schedule. For suspected lithium toxicity with symptoms, withhold lithium, obtain an urgent lithium concentration and U&Es, and seek specialist advice; do not wait for a routine 12-hour sample.
- 3Interpret symptoms, ECG or other physiological findings together with the concentration. Consider active metabolites, altered binding, electrolyte sensitivity and a changing renal state; an apparently normal laboratory interval must not dismiss a convincing toxicity presentation.
- 4Document treatment, observation and the eventual restart or alternative plan. A previously tolerated regimen may no longer be appropriate after an interaction or organ decline, so resumption requires a deliberate assessment of what changed.
03AdjustmentReview an unexpectedly low concentrationA monitoring result is below the expected interval or individual target.+
- 1Check whether the sample timing, units and recent regimen match the target. Confirm actual adherence and supply, because a missed dose, wrong formulation or mistimed sample can create a low result without proving that the prescribed maintenance dose is inadequate.
- 2Assess clinical control and adverse effects, then review organ function, protein binding and interactions. In phenytoin treatment, a low total concentration with reduced albumin should prompt consideration of free exposure before a dose increase.
- 3If adjustment is indicated, follow the medicine-specific titration method and allow an appropriate interval for reassessment. Phenytoin has nonlinear kinetics, so a proportional increase intended to double a concentration can produce an unexpectedly large rise.
- 4Record the new regimen, the expected clinical response and the repeat test with exact timing instructions. Verify that the patient has the correct product and understands which old doses or strengths should no longer be used.
05Checking understandingVerify the reasoning, revisit uncertainties and apply feedback.
- For stable lithium treatment, check concentrations every 3–6 months and kidney, thyroid and calcium measures every 6 months; increase frequency with age-related vulnerability, interacting drugs, pregnancy or renal decline. Concentration monitoring does not screen for every long-term adverse effect.
- For digoxin, request levels when the clinical question justifies them, such as suspected toxicity, an interaction or deteriorating renal function. After a dose change in normal renal function, SPS describes checking at about 7 days; impaired clearance may require longer to reach the new steady state.
- For phenytoin or carbamazepine, relate any concentration to seizure control, adherence, interactions and specific safety monitoring. Routine concentration testing is not a replacement for recognising rash, blood dyscrasia, hepatic problems or other clinically important adverse effects.
- Close the loop on every result: verify receipt, interpret it, communicate the action and confirm the next dose or appointment. If the responsible clinician is unavailable, a defined cover arrangement prevents a high-risk result sitting unread.
06Special situationsVariants, exceptions and circumstances that change the usual approach.
Steady state is a timing assumption
After a maintenance change, a concentration may still be moving towards its new pattern. The relevant interval depends on elimination and can lengthen with organ impairment. A sample taken too soon may underestimate the eventual accumulation.
Total concentration and active exposure differ
Laboratories often measure bound plus unbound drug, while the unbound component is available to act. Reduced binding can therefore make a lower total result misleading. Clinical findings and, where available, a free concentration help resolve the discrepancy.
Population ranges are guides
A quoted interval describes a useful reference, not a guaranteed boundary between effectiveness and toxicity. An individual’s previous effective concentration can be valuable, provided the indication, timing and physiological circumstances are comparable.
A calculation can be pharmacologically wrong
Correct arithmetic cannot rescue an inappropriate model. Assuming a linear dose–concentration relationship for phenytoin may produce unsafe advice even when every multiplication is accurate; understand the kinetics before using a proportional formula.
07Common pitfallsFrequent interpretation and management errors.
- 01
Changing lithium solely because a post-dose sample differs from a trough target can create a problem that began with sample timing rather than dose.
- 02
Escalating phenytoin automatically for a low total concentration in hypoalbuminaemia can increase already adequate or excessive free exposure.
- 03
Ordering routine digoxin levels at every visit without a clinical question can distract from kidney function, electrolytes and symptoms that more directly guide care.
- 04
Assuming that a laboratory result within range excludes toxicity can delay treatment when physiology, interactions or altered sensitivity change the clinical effect.