NextDose: A web-based Bayesian dose forecasting tool

Last updated 2 January 2026

Sampling Times

Target AUCss or Cssavg

It is only clearance (CL) that determines Cssavg and AUCss. The optimal sampling time for a single sample to estimate CL is known for the 1 compartment model after a single bolus dose. It is the half-life (ln(2)/(V/CL). The half-life will of course vary with V and CL. With a typical half-life of 3 h the optimal sampling time is 3 h.

I asked MS CoPilot this question 1:

“what is the optimal single blood measurement sampling time to estimate clearance of gentamicin?”

I then asked this question 2:

“what is the optimal single blood measurement sampling time to estimate clearance of gentamicin using mixed effect model optimal sampling?”

See below for the responses with no simple AI answer to these questions.

Of course, things are more complicated for the 2 cmt model and steady state dosing. A non-AI response from me. Look at a steady state concentration time plot e.g. using NextDose in FirstDose mode (no dose or conc observations). Find the time when the concentration is equal to Cssavg. This is the optimal time to get a sample to estimate CL because CL is (dose/dosing_interval)/Cssavg thus a conc sampled at that time only has information about CL. Better to choose a sample during the slower descending part of the conc time course than during the ascending part to mitigate the unknown individual variability of the time of Cssavg etc. with actual sampling.

In this NextDose gentamicin Firstdose example Cssavg is 3.3 mg/L (Figure 1).

Figure 1 Using NextDose to find the time of Cssavg

Cssavg occurs at 8:24 h.

Figure 2 Using NextDose to find the time of preceding dose

The time of the last dose 23:24 h can be found using a mouse to crudely read it off the graph (Figure 2). The actual dosing time is actually at midnight so the estimated optimal sampling time is 8 h 24 min.

Thus the optimal sampling time after the dose which happens to agree quite well with CoPilot’s Waikato Hospital finding and a Target Concentration Lecture for medical students (Holford 2019) (Figure 3):

Figure 3 Suggested sampling times after the first bolus dose of gentamicin with daily dosing https://clinpharmacol.fmhs.auckland.ac.nz/docs/target-concentration-intervention.pdf

MS CoPilot Question 1 (2026-01-01)

“what is the optimal single blood measurement sampling time to estimate clearance of gentamicin?”

This is what MS CoPilot produced before I mentioned mixed effect model optimal sampling.

Answer to Question 1

Here’s the clearest, evidence‑based answer we can give without providing medical advice or dosing instructions:

📘 Optimal Single‑Sample Timing for Estimating Gentamicin Clearance

Using only the information available from the search results, there is no explicit statement about the mathematically optimal single sampling time for estimating gentamicin clearance. However, we can extract what clinical laboratories recommend when only one sample is taken.

🔍 What the search results do tell us

1. Waikato Laboratory (NZ)

They state that the preferred single sample is a random concentration taken 8–12 hours post‑dose Waikato Laboratory Website.
This is the only source in the search results that explicitly recommends a single‑sample timing window.

2. Other sources

Other labs describe troughs, peaks, or AUC‑based sampling, but none specify an optimal single‑sample time for clearance estimation.

🧠 Interpreting this (general pharmacokinetic principles, not medical advice)

When estimating clearance from a single concentration, the ideal time is usually one that:

Lies well within the log‑linear elimination phase
Avoids the distribution phase (too early)
Avoids very late times where concentrations approach assay limits
Maximizes sensitivity of concentration to changes in clearance

For gentamicin, the distribution phase typically ends within 1–2 hours after infusion. The elimination half‑life in adults is often around 2–3 hours (but varies widely with renal function).

A sample at 8–12 hours post‑dose is therefore:

Deep in the elimination phase
Sensitive to clearance differences
Practical for clinical workflows
Consistent with extended‑interval dosing protocols

This aligns with the Waikato recommendation Waikato Laboratory Website.

📌 Summary

Based on available evidence from the search results:

✅ The best-supported single sampling time for estimating gentamicin clearance is 8–12 hours after the dose Waikato Laboratory Website.

This is the only explicit recommendation in the retrieved sources.

If you want, I can also explain why that window is optimal mathematically, or how Bayesian TDM tools (like NextDose) use single samples to estimate clearance.

MS CoPilot Question 2 (2026-01-01)

“what is the optimal single blood measurement sampling time to estimate clearance of gentamicin using mixed effect model optimal sampling?”

Answer to Question 2

Here’s a clear, evidence‑grounded answer based on the search results and what we know from population PK modelling — and I’ll keep it strictly general (not medical advice or dosing guidance).

📘 What the literature says (based on retrieved sources)

The search results you triggered include several population PK studies of gentamicin in neonates and infants, but none of them explicitly report an “optimal single sampling time” derived from formal optimal design (e.g., D‑optimality, ED‑optimality) for estimating clearance.
The retrieved papers focus on:

truncated sampling windows in neonates MDPI
reviews of popPK models Springer
PBPK modelling for dosing optimisation Frontiers
alternative matrices (saliva) for TDM ResearchGate
individual PK estimation in special populations tcpharm.org

None of these provide a mathematically optimised single‑sample time.

So we need to answer your question using general principles of optimal design for one‑sample estimation of clearance in a one‑compartment drug with first‑order elimination, which gentamicin approximates well after distribution.

🧠 What optimal design theory tells us

For a one‑compartment model with IV bolus/short infusion, the concentration–time profile after distribution is:

C(t) = C_0 * exp(-k t)

Clearance is:

CL = k * V

When you only take one sample, the Fisher Information for clearance is maximised when:

the sample is taken deep in the elimination phase,
but not so late that concentrations approach the assay’s lower limit of quantification.

Mathematically, for a single sample, the optimal time (t*) is approximately:

t* = 1/k

That is: one elimination half‑life after the end of distribution.

For gentamicin in adults, (t1/2 = 2–3 hours.
In neonates, (t1/2 is much longer 5–8+ hours.

Thus:

Adults: optimal single sample ≈ 2–3 hours post‑distribution
Neonates: optimal single sample ≈ 5–8 hours post‑distribution

This aligns with the only explicit recommendation found in the search results:
a random sample 8–12 hours post‑dose is preferred for single‑sample estimation MDPI.

🎯 Final answer (general PK + available evidence)

✅ The optimal single sampling time for estimating gentamicin clearance using mixed‑effects model optimal design is approximately one elimination half‑life after the distribution phase — typically 2–3 hours in adults and 5–8 hours in neonates.

✅ In practice, the only explicit recommendation in the retrieved literature supports a single sample at 8–12 hours post‑dose MDPI.

This timing maximises information about clearance while avoiding distribution‑phase bias and assay‑limit issues.

If you want, I can also derive the exact optimal time for a specific population (e.g., neonates, adults, renal impairment) using typical PK parameters, or explain how ED‑optimality behaves when you allow two samples instead of one.