Pharmacokinetic-pharmacodynamic modeling in anesthesia, intensive care and pain medicine

Pharmacokinetic-Pharmacodynamic Modeling of Midazolam in Pediatric Surgery

Midazolam (MDZ) is used for sedation in surgical procedures; its clinical effect is related to its receptor affinity and the dose administered. Therefore, a pharmacokinetic-pharmacodynamic (PK-PD) population model of MDZ in pediatric patients undergoing minor surgery is proposed. A descriptive, observational, prospective, and longitudinal, study that included patients of both sexes, aged 2-17 years, ASA I/II, who received MDZ in IV doses (0.05 mg/kg) before surgery. Three blood samples were randomly taken between 5-120 min; both sedation by the Bispectral Index Scale (BIS) and its adverse effects were recorded. The PK-PD relationship was determined using a nonlinear mixed-effects, bicompartmental first-order elimination model using Monolix Suite™. Concentrations and the BIS were fitted to the sigmoid Emax PK-PD population and sigmoid Emax PK/PD indirect binding models, obtaining drug concentrations at the effect site (biophase). The relationship of concentrations and BIS showed a clockwise hysteresis loop, probably indicating time-dependent protein binding. Of note, at half the dose used in pediatric patients, adequate sedation without adverse effects was demonstrated. Further PK-PD studies are needed to optimize dosing schedules and avoid overdosing or possible adverse effects.

The parameter Houlihan: A solution to high-throughput identifiability indeterminacy for brutally ill-posed problems

One way to interject knowledge into clinically impactful forecasting is to use data assimilation, a nonlinear regression that projects data onto a mechanistic physiologic model, instead of a set of functions, such as neural networks. Such regressions have an advantage of being useful with particularly sparse, non-stationary clinical data. However, physiological models are often nonlinear and can have many parameters, leading to potential problems with parameter identifiability, or the ability to find a unique set of parameters that minimize forecasting error. The identifiability problems can be minimized or eliminated by reducing the number of parameters estimated, but reducing the number of estimated parameters also reduces the flexibility of the model and hence increases forecasting error. We propose a method, the parameter Houlihan, that combines traditional machine learning techniques with data assimilation, to select the right set of model parameters to minimize forecasting error while reducing identifiability problems. The method worked well: the data assimilation-based glucose forecasts and estimates for our cohort using the Houlihan-selected parameter sets generally also minimize forecasting errors compared to other parameter selection methods such as by-hand parameter selection. Nevertheless, the forecast with the lowest forecast error does not always accurately represent physiology, but further advancements of the algorithm provide a path for improving physiologic fidelity as well. Our hope is that this methodology represents a first step toward combining machine learning with data assimilation and provides a lower-threshold entry point for using data assimilation with clinical data by helping select the right parameters to estimate.

Mechanistic machine learning: how data assimilation leverages physiologic knowledge using Bayesian inference to forecast the future, infer the present, and phenotype

We introduce data assimilation as a computational method that uses machine learning to combine data with human knowledge in the form of mechanistic models in order to forecast future states, to impute missing data from the past by smoothing, and to infer measurable and unmeasurable quantities that represent clinically and scientifically important phenotypes. We demonstrate the advantages it affords in the context of type 2 diabetes by showing how data assimilation can be used to forecast future glucose values, to impute previously missing glucose values, and to infer type 2 diabetes phenotypes. At the heart of data assimilation is the mechanistic model, here an endocrine model. Such models can vary in complexity, contain testable hypotheses about important mechanics that govern the system (eg, nutrition's effect on glucose), and, as such, constrain the model space, allowing for accurate estimation using very little data.

Comparison of three a-priori models in the prediction of serum lithium concentration

Context:

Mathematical models are valuable for optimizing drug dose and dosing regimens.

Aims:

To compare the precision and bias of three a-priori methods in the prediction of serum level of lithium in patients with bipolar disorder, and to determine their sensitivity and specificity in detecting serum lithium levels outside the therapeutic range.

Settings and Design:

Hospital-based, retrospective study.

Materials and Methods:

In a retrospective study of 31 in-patients, the serum level of lithium was calculated using three different a-priori methods. Mean Prediction Error was used as a measure of bias while Mean Absolute Error and Root Mean Squared Error were used as a measure of precision. The sensitivity and specificity of the methods was calculated.

Results:

All three models underestimated serum lithium level. Precision was best with the model described by Pepin et al., while bias of prediction was the least with the method of Abou Auda et al. The formula by Pepin et al. was able to predict serum lithium level with a mean error of 36.57%. The sensitivity and specificity of the models in identifying serum lithium levels outside the therapeutic range was 80% and 76.19% for Pepin et al., 90% and 74.19% for Zetin et al., and 90% and 66.67% for Abou-Auda et al., respectively.

Conclusion:

The study demonstrates the difference in precision and bias of three a-priori methods, with no one method being superior to the other in the prediction of serum concentration.

Designing simple PK-PD studies in children

Conducting clinical pharmacology research studies in pediatric patients is challenging because of ethical and practical constraints but necessary to ensure that drugs are used safely and effectively in this population. Developments in laboratory analytical techniques, such as improved assay sensitivity and the use of alternative sample matrices, can reduce blood loss and offer less invasive blood sampling, causing less trauma to the patient and fewer ethical concerns. Recent advances in data analysis techniques, which aim to extract the maximum amount of useful information from small sample numbers, should be considered when planning a clinical trial and incorporated into the study design. Using 'population' methodology allows a more flexible sampling strategy that enables valuable data to be collected in the course of routine clinical practice, rather than in a rigid, and potentially artificial, setting. Integration of pharmacokinetics and pharmacodynamics and the application of physiological approaches and simulation techniques to the analysis and interpretation of drug concentration and effect data offer new opportunities that have particular relevance to pharmacological research in the field of pediatric anesthetics.