Abstract

Research Article

A Further Example Showing Efficiency of a Modeling Method Based on the Theory of Dynamic Systems in Pharmacokinetics

Maria Durisova*

Published: 25 January, 2017 | Volume 1 - Issue 1 | Pages: 007-012

Aims: To present a further example showing an efficiency of a modeling method based on the theory of dynamic systems in pharmacokinetics.

Study design:The goals of the current study were twofold: to present (1) a further example showing efficiency of a modeling method based on the theory of dynamic systems in pharmacokinetics, an to perform (2) a next step in tutoring the use of computational and modeling tools from the theory of dynamic systems in pharmacokinetics.

The data available in the study by Plusquellec et al. published in the October Issue of the Journal Medical Engineering & Physics were used to exemplify the method considered here. For modeling purpose an advanced mathematical modeling method was employed. Modeling was performed using the computer program named CTDB described in the study by Dedík et al. published in September 2007 issue of the Journal Diabetes Research and Clinical Practice.

Main outcome: Modeling results revealed that computational and modeling tools from the theory of dynamic systems can be successfully used in the development of a mathematical model of such a complicated process as is a multiple sites discontinuous gastrointestinal absorption.

Read Full Article HTML DOI: 10.29328/journal.hps.1001002 Cite this Article Read Full Article PDF

Keywords:

Dynamic system; Mathematical model; Transfer function

References

  1. Dammann HG, Müller P, Kather H, Simon B. The new histamine H2-receptor antagonist ranitidine. Duration of action. Res Exp Med (Berl). 1981; 178: 151-154. Ref.: https://goo.gl/47Zy9Y
  2. Dawson J, Richards DA, Stables R, Dixon GT, Cockel R. Ranitidine --pharmacology and clinical use. J Clin Hosp Pharm Ther. 1983; 8: 1-13. Ref.: https://goo.gl/bkzNwF
  3. Eckardt VF, Cordes I, Janish HD, Wiemann H. Daytime acid secretion after a single dose of ranitidine and cimetidine--a double blind crossover study. Eur J Clin Pharmacol. 1984; 26: 177-182. Ref.: https://goo.gl/9HoAnV
  4. Brittain RT, Jack D. Histamine H2-antagonists--past, present and future. J Clin Gastroenterol. 1983, 5: 71-79. Ref.: https://goo.gl/EXuI0O
  5. Gaginella TS, Bauman JH. Ranitidine hydrochloride. Drug Intell Clin Pharm 1983; 17: 873-885. Ref.: https://goo.gl/ptOQfC
  6. Merki HS, Witzel L, Walt RP, Neumann J, Scheurle E, et al. Comparison of ranitidine 300 mg twice daily, 300 mg at night and placebo on 24-hour intragastric acidity of duodenal ulcer patients. Aliment. Pharmacol Ther. 1987; 1: 217-223. Ref.: https://goo.gl/bEZhOc
  7. Plusquellec Y, Efthymiopoulos C, Duthil P, Houin G. A pharmacokinetic model for multiple sites discontinuous gastrointestinal absorption. Med Eng Phys. 1999; 21: 525-532. Ref.: https://goo.gl/BkULJt
  8. Reid SR, Bayliff CD. The comparative efficacy of cimetidine and ranitidine in controlling gastric pH in critically ill patients. Canad Anaesth. 1986; 33: 287-293. Ref.: https://goo.gl/Wcm42c
  9. Higuchi K, Watanabe T, Tominaga K, Shiba M, Nakagawa K, et al. Effects of ranitidine on quality of gastric ulcer healing compared with famotidine: a randomized, controlled, multicenter trial. Int J Clin Pharmacol Res. 2005; 25: 187-194. Ref.: https://goo.gl/i6D1S1
  10. Chrenova J, Durisova M, Mirciou C, Dedik L. Effect of gastric emptying and entero-hepatic circulation on bioequivalence assessment of. Methods Find Exp Clin Pharmacol. 2010; 32: 413-419. Ref.: https://goo.gl/OanMbq
  11. van Rossum JM, de Bie JE, van Lingen G, Teeuwen HW. Pharmacokinetics from a dynamical systems point of view. J Pharmacokinet Biopharm. 1989; 17: 365-392. Ref.: https://goo.gl/VAVsFe
  12. Dedík L, Ďurišová M. CXT-MAIN: a software package for the determination of the analytical form of the pharmacokinetic system weighting function. Comput Methods Programs Biomed. 51: 183-192. Ref.: https://goo.gl/iAvtNQ
  13. Dedík L, Ďurišová M. Frequency response method in pharmacokinetics. J Pharmacokin Biopharm. 1994; 22: 237-307. Ref.: https://goo.gl/flpLxF
  14. Ďurišová M, Dedík L. Modeling in frequency domain used for assessment of in vivo dissolution profile. Pharm Res. 1997; 14: 860-864. Ref.: https://goo.gl/p6uWjQ
  15. Ďurišová M, Dedík L. A system-approach method for the adjustment of time-varying continuous drug infusion in individual patients: A simulation study. J Pharmacokinet Pharmacodyn. 2002; 29: 427-444. Ref.: https://goo.gl/1XtK6n
  16. Ďurišová M, Dedík L. New mathematical methods in pharmacokinetic modeling. Basic Clin Pharmacol Toxicol. 2005; 96: 335-342. Ref.: https://goo.gl/UNFNoG
  17. Yates JW. Structural identifiability of physiologically based pharmacokinetic models. J Pharmacokinet Pharmacodyn. 2006; 33: 421-439. Ref.: https://goo.gl/qf3HvZ
  18. Ďurišová M, Dedík L, Kristová V, Vojtko R. Mathematical model indicates nonlinearity of noradrenaline effect on rat renal artery. Physiol Res. 2008; 57: 785-788. Ref.: https://goo.gl/UW4Ees
  19. Ďurišová M. Physiologically based structure of mean residence time. Scientific World Journal. 2012. Ref.: https://goo.gl/DjpTgU
  20. Ďurišová M. A physiological view of mean residence times. Gen Physiol Biophys. 2014; 33: 75-80. Ref.: https://goo.gl/SFZu6S
  21. Ďurišová M. Mathematical model of pharmacokinetic behavior of orally administered prednisolone in healthy volunteers. J Pharmaceut & Pharmacol. 2014; 2: 1-5.
  22. Ďurišová M. Further worked out examples that illustrated the successful use of an advanced mathematical modeling method based on the theory of dynamic systems in pharmacokinetics. Int J Res Sci Res. 2015; 6: 4873-4879. Ref.: https://goo.gl/eB8xLo
  23. Ďurišová M. Mathematical model of pharmacokinetic behavior of warfarin. Adv Pharm Clin Trials. 2016; 1: 1-7. Ref.: https://goo.gl/1Zd7zi
  24. Ďurišová M. Computational analysis of pharmacokinetic behavior of ampicillin. J Appl Bioanal. 2016; 2: 84-89. Ref.: https://goo.gl/9uhDJ6
  25. Levitt DG. PK Quest: a general physiologically based pharmacokinetic model. Introduction and application to propranolol. BMC Clin Pharmacol. 2002; 15: 2-5. Ref.: https://goo.gl/2UCayb
  26. Howell BA, Yang Y, Kumar R, Woodhead JL, Harill AH, et al. In vitro to in vivo extrapolation and species response comparisons for drug-indused liver injury (DILI) using DILsymTM: a mechanistic model of DILI. J Pharmacokinet Pharmacodyn. 2012; 39: 527-541. Ref.: https://goo.gl/LveVQu
  27. Dedík L, Ďurišová M, Penesová A, Miklovičová D, Tvrdoňová M. Estimation of influence of gastric emptying on shape of glucose concentration-time profile measured in oral glucose tolerance test. Diabetes Research and Clinical Practice. 2007; 77: 377-384. Ref.: https://goo.gl/0koO9F
  28. Ďurišová M. Physiologically based structure of mean residence time. The Sci World Journal. 2012. Ref.: https://goo.gl/xPqy08
  29. Ďurišová M. A physiological view of mean residence times. Gen Physiol Biophys. 2014; 33: 75-80. Ref.: https://goo.gl/44tImL
  30. Weiss M, Pang KS. Dynamics of drug distribution. I. Role of the second and third curve moments. J Pharmacokinet Biopharm. 1992; 20: 253-278. Ref.: https://goo.gl/wiYtGU
  31. Verotta D. Concepts, properties, and applications of linear systems to describe distribution, identify input, and control endogenous substances and drugs in biological systems. Crit Rev Biomed Eng. 1996; 24: 73-139. Ref.: https://goo.gl/6lGcmI
  32. Xiao H, Song H, Yang Q, Cai H, Qi R,et al. A prodrug strategy to deliver cisplatin (IV) and paclitaxel in nanomicelles to improve efficacy and tolerance. Biomaterials. 2012; 33: 6507-6519. Ref.: https://goo.gl/UthCGZ
  33. Beckermann B, Kaliaguine V. The diagonal of the Padé table and the approximation of the Weyl function of second-order difference operators. Constr Approx. 1997; 13: 481-510. Ref.: https://goo.gl/AgA5a2
  34. Levy EC. Complex curve fitting. IRE Trans Automat Contr. 1959; 4: 37-43. Ref.: https://goo.gl/5g3G6d
  35. Akaike H. A new look at the statistical model identification. IEEE Trans Automat Contr. 1974; 19: 716-723. Ref.: https://goo.gl/KM0Sbt
  36. Lampariello F, Sciandrone M. Use of the minimum-norm search direction in a nonmonotone version of the Gauss-Newton method. J Optim Ther Appl. 2003; 119: 65-82. Ref.: https://goo.gl/JwG01Y
  37. Boyle P. Options: A Monte Carlo approach. J Fin Econom. 1977; 4: 323-338. Ref.: https://goo.gl/g8gMCC
  38. Varvel JR, Donoho DL, Shafer SL. Measuring of the predictive performance of computer controlled infusion pumps. J Pharmacokin Biopharm. 1992; 20: 63-84. Ref.: https://goo.gl/vsRmMu
  39. Siegel RA. Pharmacokinetic transfer functions and generalized clearances. J Pharmacokin Biopharm. 1986; 14: 511-521. Ref.: https://goo.gl/aRrc7I
  40. Segre G. The sojourn time and its prospective use in pharmacology.J Pharmacokin Bio- pharm. 1988; 16: 657-666. https://goo.gl/7yZs2V
  41. Yates JW. Structural identifiability of physiologically based pharmacokinetic models. J Pharmacokinet Pharmacodyn. 2006; 33: 421-439. Ref.: https://goo.gl/VPEqJ0
  42. Rescigno A. Compartmental analysis and its manifold applications to pharmacokinetics. AAPS J. 2010; 12: 61-72. Ref.: https://goo.gl/qJ6h5r
  43. Gillespie WR, Veng-Pedersen P, Berg MJ, Schkottelius DD. Linear systems approach to the analysis of an induced drug removal process. Phenobarbital removal by oral activated charcoal. J Pharmacokin Biopharm. 1986; 14: 19-28. Ref.: https://goo.gl/vL520R

Figures:

Figure 1

Figure 1

Similar Articles

Recently Viewed

Read More

Most Viewed

Read More

Help ?