← Back to portfolio
Educational pharmacokinetic tool

Vancomycin AUCmodel explorer

Explore how patient information, a published population model, and measured vancomycin levels change estimated exposure over time.

Published equationsBayesian updatingCalculations stay in your browser

Educational use only. This prototype has software verification but no independent clinical validation. Do not use it to make patient-care decisions.

1

Patient

Enter a current clinical snapshot

2

Regimen

Describe the planned dosing schedule

3

Levels

Add measured concentrations if available

4

Exposure

Review estimated AUC and concentration

Patient information

This controls model eligibility and auto-selection. It does not diagnose the patient.

Calculated values
CrCl
82.6mL/min
IBW
65.9kg
ABW
67.6kg
BMI
24.2kg/m²

Pharmacokinetic model

Model selection

General adult intermittent infusion selects Thomson 2009.

Fit for this patient profile

Within published population ranges

No range warning

No encoded age, weight, renal-range, obesity, or dialysis warning was triggered.

Renal estimate used
82.6 mL/min
CG CrCl, TBW, SCr floor 0.679 mg/dL
Weight used
70.0 kg
Total body weight

The model uses

  • CG CrCl, TBW, SCr floor 0.679 mg/dL
  • 2-compartment intermittent infusion structure
  • Published parameter variability and residual-error model

The model cannot see

  • Changing renal function or other time-varying covariates
  • Missed, delayed, interrupted, or changed administrations
  • Pre-existing vancomycin, renal replacement therapy, and assay metadata

Current patient summary: 55 years, 70.0 kg, SCr 1.00 mg/dL. Eligibility flags describe published ranges; they are not a clinical suitability determination.

Current regimen and target

Regimen

Fixed at 1 mg/L by broth microdilution for the 400-600 mg·h/L consensus target. A reported MIC above 1 must not be handled by proportionally increasing exposure.

Exposure target

Measured vancomycin levels

Add measured concentrations and their timing to update the population estimate. This explorer assumes the regimen above was unchanged and administered on schedule.

The update balances what the published model expects with how closely different parameter values reproduce the measured levels. It does not account for missed or changed doses, pre-existing vancomycin, changing renal function, or assay error.
Measured levels

No measured levels added

Add a concentration and its timing to update the population estimate

Pharmacokinetic estimates

Population estimate
CL (L/hr)
3.756
V1 (L)
47.25
V2 (L)
51.24
Fixed population value
Q (L/hr)
2.280
Fixed population value
Terminal β (hr⁻¹)
0.0238
Terminal half-life (hr)
29.1

Exposure over time

First 24 hours

AUC₀–₂₄

311mg·h/L

Begins with no modeled vancomycin present

Second 24 hours

AUC₂₄–₄₈

421mg·h/L

Second modeled day on the unchanged regimen

At steady state

AUC₂₄,ss

532mg·h/L

Steady-state average from daily dose ÷ CL

Early exposure and steady-state exposure answer different questions and should not be used interchangeably.

Preparing interactive chart…

Regimens within the selected target

Dose (mg)IntervalEnd-infusion Cmax,ssPre-dose Cmin,ssAUC₂₄,ssRationale
1000Nearest target match
q12h33.7 mg/L14.6 mg/L533 mg·h/LAUC₂₄ 532 mg·h/L within target 400-600
2000
q24h49.7 mg/L10.3 mg/L533 mg·h/LAUC₂₄ 532 mg·h/L within target 400-600
1750
q24h43.5 mg/L9.0 mg/L466 mg·h/LAUC₂₄ 466 mg·h/L within target 400-600
750
q8h32.7 mg/L18.9 mg/L599 mg·h/LAUC₂₄ 599 mg·h/L within target 400-600

Documentation

Clinical text export

Paste-ready calculation summary

Plain text for an EHR note or pharmacist handoff. It includes the entered clinical inputs, model provenance, fit warnings, levels, PK estimates, and modeled exposure.

Patient name, date of birth, medical-record number, and encounter identifiers are not collected.

Review before adding to the medical record. This export is a calculation summary and is not a medication order.

About this explorer

Validation status

Published equation translation

Verified locally

Source equations reproduced with reference cases

Structural PK mathematics

Verified locally

Infusion, superposition, steady-state, and mass-balance invariants

Bayesian implementation

Verified locally

Synthetic recovery, residual error, and optimizer-boundary checks

External retrospective validation

Not performed

Requires timestamped deidentified patient data

Prospective clinical validation

Not performed

Requires protocol, governance, and independent review

Clinical decision support status

Not authorized

Educational demonstration only

“Verified locally” means the implementation passed the repository’s deterministic software tests. It does not establish predictive performance in an external population.

Methods

Bayesian MAP

Deterministic log-grid MAP estimation uses each model’s published CL/volume variability and residual error. A local log-scale Laplace approximation makes parameter uncertainty visible without presenting it as exact.

PK Engine

One-, two-, and three-compartment IV models cover intermittent and continuous infusion. Dialysis clearance is switched on only during entered Westra HDF/HD sessions.

Population Models

Auto-selection considers clinical population and administration structure. Every model remains labeled Match, Caution, Not applicable, or Source data required; override cannot bypass structural incompatibility.

References

Daniel Tran, PharmD

UC San Diego — Skaggs School of Pharmacy

Source code MIT. Content © 2026 Daniel Tran (CC BY-NC-SA 4.0).