Whole body physiology model to simulate respiratory depression of fentanyl and associated naloxone reversal

Abstract

Background: Opioid use in the United States and abroad is an endemic part of society with yearly increases in overdose rates and deaths. In response, the use of the safe and effective reversal agent, naloxone, is being fielded and used by emergency medical technicians at a greater rate. There is evidence that repeated dosing of a naloxone nasal spray is becoming more common. Despite this we lack repeated dosing guidelines as a function of the amount of opiate the patient has taken.

Methods: To measure repeat dosing guidelines, we construct a whole-body model of the pharmacokinetics and dynamics of an opiate, fentanyl on respiratory depression. We then construct a model of nasal deposition and administration of naloxone to investigate repeat dosing requirements for large overdose scenarios. We run a single patient through multiple goal directed resuscitation protocols and measure total naloxone administered.

Results: Here we show that naloxone is highly effective at reversing the respiratory symptoms of the patient and recommend dosing requirements as a function of the fentanyl amount administered. We show that for increasing doses of fentanyl, naloxone requirements also increase. The rescue dose displays a nonlinear response to the initial opioid dose. This nonlinear response is largely logistic with three distinct phases: onset, rapid acceleration, and a plateau period for doses above 1.2 mg.

Conclusions: This paper investigates the total naloxone dose needed to properly reverse respiratory depression associated with fentanyl overdose. We show that the current guidelines for a rescue dose may be much lower than required.

Fig. 1. The models in the physiology engine that respond to Fentanyl administration (venous bolus dose).

Feedback indicates that the two models interact with one another each time step, such as the drug-drug interaction model between Fentanyl and Naloxone. Figure created using BioIcons open-source icons. Icons used in the nasal figure are copyrighted under license CC 3.0. PKPD pill icon is licensed under CC 0, the syringe and vessels are licensed under CC 3.0, Respiration icons are provided under the CC 3.0 license, Nervous and circulation are similarly provided under CC 3.0 license.

Fig. 2. 2-hour simulation of nasal naloxone dose for 4 (orange dashed) and 2 (blue) mg doses.

The peak venous concentration and long trend dynamics follow experimental data well, with minor peak concentration differences for each dose. Error bars correspond to one standard deviation of experimental data.

Fig. 3. Pharmacokinetic profiles for various routes of administration, concentrations, and total doses for experimental studies in humans.

Error bars correspond to one standard deviation of experimental data, deviations were not reported in the Han study. a and d simulate bolus dose administrations, equivalent to experimental studies. Subfigure b simulates an infusion for 5 minutes, and c simulates an infusion over 4 minutes.

Fig. 4. CO₂ response for varying levels of Fentanyl bolus dose.

a Total pulmonary ventilation, b oxygen partial pressure for a single 0.5 mg bolus dose of fentanyl. Subfigure c displays the CO₂ response, computed for varying levels of steady-state fentanyl concentrations. Respiratory depression as a function of the fentanyl administration tracks well with the experimental data from. Error bars represent one standard deviation from the mean of the data.

Fig. 5. Simulated testing of the co2 response as a function of fentanyl steady-state concentration in the blood.

Step infusions were administered to the simulated patient, and after five minutes of steady-state plasma concentrations, subfigure a displays the fentanyl plasma step concentrations, b displays the end-tidal CO₂, c displayes the O₂ partial pressure, d displays the central nervous response, e and f highlight the respiration rate and total pulmonary ventilation over the duration of the simulated experiment. We altered the inhaled substance composition to consist of ~ 10% CO₂ for 5 minutes. The environmental inhaled substance composition was returned to normal before each fentanyl increment.

Fig. 6. The primary cause and effect of fentanyl induced hypoxia are plotted here for varying levels of initial fentanyl bolus doses.

Subfigure a displays oxygen saturation curves over different initial fentanyl doses, and b highlights respiration rate over those same fentanyl doses. The purple line indicates the initial naloxone dose administered during the rescue scenario.

Fig. 7. Simulated protocol for reversal resuscitation for an opioid overdose patient.

Green denotes the start of the rescue period. Red boxes indicate decision/administration points, with blue boxes denoting observation or translation points.

Fig. 8. We group three fentanyl doses from analyzing the total administration amounts of the simulated experiment and compute the mean (pink diamonds) and standard deviation of the total naloxone dose.

The line in the middle of the boxes indicates the median of the data. Each box plot quantifies the non-linear response of the required naloxone dose as a function of fentanyl dose. N = 6 for each sample grouping.