caffsim

## Caffeine Concentration Predictor

- `Caffeine Concentration Predictor` <https://asan.shinyapps.io/caff>
- `Caffeine Concentration Predictor` is open to everyone. We are happy to take your input. Please fork the repo, modify the codes and submit a pull request. <https://github.com/shanmdphd/caff>

### Reference

This work is solely dependent on the interesting paper published in Eur J Pediatr in 2015. 

- "Prediction of plasma caffeine concentrations in young adolescents following ingestion of caffeinated energy drinks: a Monte Carlo simulation." Eur J Pediatr. 2015 Dec;174(12):1671-8. doi: 10.1007/s00431-015-2581-x <https://www.ncbi.nlm.nih.gov/pubmed/26113286>
- "Clinical pharmacokinetics and pharmacodynamics: concepts and applications, 4th edition" Lippincott Williams & Wilkins. 2011. ISBN 978-0-7817-5009-7

The pharmacokinetic parameters from the paper were derived and used in the app as follows:

$$ 
\begin{split}
\begin{bmatrix}
     \eta_1 \newline
     \eta_2 \newline
     \eta_3
\end{bmatrix}
& \sim MVN \bigg(
\begin{bmatrix}
     0 \newline
     0 \newline
     0
\end{bmatrix}
, 
\begin{bmatrix}
     0.1599 & 6.095 \cdot 10^{-2} & 9.650 \cdot 10^{-2} \newline
     6.095 \cdot 10^{-2} & 4.746 \cdot 10^{-2} & 1.359 \cdot 10^{-2} \newline
     9.650 \cdot 10^{-2} & 1.359 \cdot 10^{-2} & 1.004
\end{bmatrix}
\bigg) \newline
\newline
CL\ (mg/L) & = 0.09792 \cdot W \cdot e^{\eta1} \newline
V\ (L) & = 0.7219 \cdot W \cdot e^{\eta2} \newline
k_a\ (1/hr) & = 4.268 \cdot e^{\eta3} \newline
\newline
k\ (1/hr) & = \frac{CL}{V} \newline
t_{1/2}\ (hr) & = \frac{0.693}{k} \newline
t_{max}\ (hr) & = \frac{ln(k_a) - ln(k)}{k_a - k} \newline
C_{max}\ (mg/L) & = \frac{Dose}{V} \cdot \frac{k_a}{k_a - k} \cdot (e^{-k \cdot  t_{max}} - e^{-k_a \cdot t_{max}}) \newline
AUC\ (mg \cdot hr / L)  & = \frac{Dose}{CL} \newline
\newline
C_{av,ss} & = \frac{Dose}{CL \cdot \tau} \newline 
AI & = \frac{1}{1-e^{-k_e \cdot \tau}} \newline
\end{split}
$$
(Abbreviation: $AI$, accumulation index; $AUC$, area under the plasma drug concentration-time curve; $CL$, total clearance of drug from plasma; $C_{av,ss}$, average drug concentration in plasma during a dosing interval at steady state on administering a fixed dose at equal dosing intervals; $C_{max}$, highest drug concentration observed in plasma; $MVN$, multivariate normal distribution; $V$, Volume of distribution (apparent) based on drug concentration in plasma; $W$, body weight (kg); $\eta$, interindividual random variability parameter; $k$, elimination rate constant;  $k_a$, absorption rate constant; $\tau$, dosing interval; $t_{1/2}$, elimination half-life)

### R Packages
- H. Wickham. ggplot2: Elegant Graphics for Data Analysis. Springer-Verlag New York, 2009.
- Winston Chang, Joe Cheng, JJ Allaire, Yihui Xie and Jonathan McPherson (2016). shiny: Web Application Framework for R. R package version 0.14.2. https://CRAN.R-project.org/package=shiny
- JJ Allaire, Jeffrey Horner, Vicent Marti and Natacha Porte (2015). markdown: 'Markdown' Rendering for R. R package version 0.7.7. https://CRAN.R-project.org/package=markdown
- Hadley Wickham and Romain Francois (2016). dplyr: A Grammar of Data Manipulation. R package version 0.5.0. https://CRAN.R-project.org/package=dplyr

### Caffeine contents

<div align=center><img src=http://graphs.net/wp-content/uploads/2013/01/Caffeine-Content-in-Energy-Drinks.jpg width = 450 /></div>