Therapeutic drug monitoring — branching viva

Opening — the vancomycin level on the ward

Examiner: A six-year-old on intravenous vancomycin for a complicated MRSA bacteraemia has a trough back. What is your monitoring target, and when should the level have been drawn? [1]

Candidate (model): For serious MRSA infection my target is the 24-hour area under the concentration–time curve over the MIC — an AUC₂₄/MIC of 400 or more — as set by the 2020 ASHP, IDSA, PIDS and SIDP consensus guideline. AUC-guided monitoring, by two timed levels or Bayesian forecasting from a single level, is preferred over a trough-only strategy because it achieves target exposure with less nephrotoxicity. A true trough is drawn within 30 to 60 minutes before the next dose, at steady state — about the fourth dose in an older child — and not during the infusion or from the infusing line. [1]

Branch 1 — the trough is unexpectedly low

Examiner: The trough is lower than you expected. Walk me through your differential before you change the dose. [1] [2]

Candidate (model): First I exclude a sampling error — the wrong time, the infusing line, or a pre-steady-state level — and I confirm the dose and interval were given as written. If the sample is valid, I consider augmented renal clearance, common in the critically ill child, where inflammation, fever and vasopressors raise glomerular filtration and clear vancomycin faster than standard dosing assumes. A population pharmacokinetic analysis in infants, children and adolescents with augmented renal clearance confirms standard regimens often miss the AUC target. My response is to raise the dose, shorten the interval, or move to Bayesian AUC dosing, then recheck at steady state. [1] [2]

Branch 2 — the PICU child who is underexposed

Examiner: So how would you explain augmented renal clearance to a parent? [2]

Candidate (model): I would say that a sick child in intensive care sometimes clears medicines faster than a well child, because the body is pushing blood through the kidneys very quickly. That means the antibiotic can disappear before it has had time to work. We check the level, and if it is too low we give a little more or give it more often, and we check again to make sure we have got it right. The principle is to match the dose to the child's body, not to a standard chart. [2]

Branch 3 — the phenytoin viva opens

Examiner: Now a different child: a nine-year-old on long-term phenytoin for epilepsy, total level 28 mg/L, albumin 22 g/L, with nystagmus and ataxia. Why is this level toxic, and why is the total level misleading? [8] [9]

Candidate (model): The level and the signs fit early phenytoin toxicity — nystagmus typically appears around 20 to 30 mg/L and ataxia around 30 mg/L. Two principles explain why. First, phenytoin follows saturable Michaelis-Menten kinetics: the enzyme that clears it approaches its maximum capacity within the therapeutic range, so a small dose increase near the top of the range causes a large level rise. Second, phenytoin is about 90 per cent bound to albumin, and only the free 10 per cent acts on the brain. With an albumin of 22 g/L the free fraction rises, so the total level understates the active concentration — I would request a free phenytoin level, target 1 to 2 mg/L, or correct the total for albumin. [8] [9]

Branch 4 — immediate management and dose adjustment

Examiner: What do you do now, and how do you restart? [8]

Candidate (model): I withhold the next dose, give supportive care, protect the airway if the child is drowsy, and recheck the level as it falls. I look for an interacting drug, such as an enzyme inhibitor that has raised the level, or an inadvertent dose increase. When I restart, I lower the maintenance dose in a small step and recheck, because near the top of the range a small dose change moves the level a lot — I would never make a large maintenance increment. I document the new target, the free-level result, and the review date, and I arrange follow-up monitoring at steady state. [8] [9]

Closing — the one-liners examiners reward

Examiner: Give me the three principles you would want every registrar to remember. [1] [8]

Candidate (model): First, monitor the narrow-therapeutic-index drugs and target the right parameter — peak for aminoglycosides, AUC for vancomycin, range for anticonvulsants. Second, sample a true trough at steady state, from a clean line, and confirm the dose and sample times before you act. Third, treat the child, not the number — a level at the top of the range in a seizing child means underdosing, and the same level in an ataxic child means toxicity. [1] [8]