“The work should be applauded as a creative modeling exercise resulting in a significant gain in insight into the mechanisms of lipid resuscitation.…”
ALTHOUGH it is clear that the intravenous infusion of lipid emulsions can successfully reverse systemic toxicity from local anesthetics and other basic drugs, exactly how this works remains controversial. Several mechanisms have been suggested including simple hemodilution, cardiotonic or metabolic effects, and the “lipid sink theory.”1
The latter proposes that an increase in the concentration of lipid in the circulation can “suck” lipid-soluble drugs out of the heart by simple partition, thereby reversing toxicity. Clearly, a rigorous mechanistic understanding of lipid resuscitation would be of considerable value in suggesting ways of optimizing the procedure for greatest efficacy and safety. But, how might the contending mechanisms be evaluated and dissected experimentally?
For obvious reasons, it is not possible to perform prospective, controlled studies of local anesthetic overdose in humans. However, the next best thing is to carry out simulations in virtual patients confined within the safety of a computer. Kuo and Apka2
used this approach to assess the validity of “lipid trapping,” as a purely pharmacokinetic explanation of lipid resuscitation. To do this, they constructed a physiologically based pharmacokinetic (PBPK) model of bupivacaine after intravenous injection. PBPK models,3
as opposed to empirical or compartmental pharmacokinetic models,4
are essentially mechanistic representations based on the anatomical arrangement of organs and tissues connected by the circulation. Building these models requires prior knowledge of system parameters (demographics, organs sizes and blood flows, enzyme and transporter abundances, levels of plasma proteins, hematocrit, and many more) as well as specific drug properties (lipid solubility and other physicochemical characteristics, concentration-dependent plasma binding, uptake by erythrocytes, enzyme and transporter kinetics, and many more). Thus, they allow essentially “bottom-up” quantitative prediction of pharmacokinetic behavior, and provide an ability to address “what if” questions, such as the impact of binding to exogenous lipid added to the circulation (based on in vitro
binding isotherms) on tissue to plasma drug concentration ratios over time. The PBPK model of bupivacaine built by Kuo and Apka2
successfully captured observed plasma drug concentrations after safe, short intravenous infusions in real humans.5
They then went on to show that a 20% long-chain triglyceride emulsion given as an intravenous bolus dose (1.5 ml/kg) followed by an infusion (0.25 ml kg−1
) over 1 h might be expected to decrease the concentration of bupivacaine in the heart by 11% within 3 min after a rapid intravenous injection of 112.5 mg of the local anesthetic, with partial rebound of tissue drug concentration after approximately 20 min. On this basis, it was concluded that the “lipid sink” theory is insufficient to guarantee an efficient reversal of systemic toxicity. Acknowledged limitations of this modeling exercise were no assessment of interindividual variability in outcome and no allowance for potential partial saturation of bupivacaine metabolism together with potential feedback effects of changes in cardiovascular dynamics on bupivacaine kinetics after higher, more toxic doses. Nevertheless, these considerations are also potentially amenable to further simulation, as is manipulation of the dose and type of lipid emulsion and its duration of administration to assess likely maximum benefit with respect to the lowering of target tissue drug concentration.
In the current issue, the PBPK study has been extended to the application of a PBPK-pharmacodynamic model.7
Cardiovascular metrics were monitored in anesthetized rats after rapid intravenous injection of a dose of bupivacaine producing transient cardiovascular toxicity followed by a random assignment to four treatments. Time to recovery was clearly in the ascending order of resuscitation with 30% intravenous lipid emulsion, 20% lipid emulsion, intravenous saline, and no treatment. To recover the observed data, the simulated heart concentrations of bupivacaine were linked to decrease in cardiac output by a Hill function. A similar function related plasma lipid concentration to its inotropic effect. The flow-promoting effect of fluid infusion was represented as an increase in cardiac output proportional to the fractional increase in venous return, and feedback control was applied to the upward departure of cardiac output from baseline. By appropriate application and combination of the PBPK model together with the pharmacodynamic functions for inotropy and volume preservation, the relative contributions of sequestration, inotropic effect, and maintenance of blood volume to the observed recoveries were teased out. Based on the criterion of time within the 95% confidence limits of the cardiovascular rate–pressure product (mean arterial pressure × heart rate), it was concluded that both the direct cardiotonic effect of the lipid emulsion and its ability to provide a “sink” for bupivacaine sequestration are involved in the mechanism of lipid rescue, with the former phenomenon being the predominant one. This was a complex experiment requiring data analysis of highly nonlinear, nonadditive relationships. The work should be applauded as a creative modeling exercise resulting in a significant gain in insight into the mechanisms of lipid resuscitation which opens the way for further investigation of the optimal clinical implementation of the procedure with respect to type of lipid and its dosage regimen. There are important implications beyond regional anesthesia with regard to the use of lipid resuscitation for the treatment of cardiotoxicity associated with the overdose of many other lipid-soluble drugs, such as antidepressants.
The author is a founder and employee of Simcyp (Certara Ltd., Sheffield, United Kingdom), a spin-out company from the University of Sheffield that provides services to a consortium of pharmaceutical companies and regulatory and academic institutions regarding the prediction of pharmacokinetics–pharmacodynamics using physiologically based modeling.
1. Weinberg GT. Lipid emulsion infusion: Resuscitation for local anesthetic and other drug overdose. ANESTHESIOLOGY. 2012;117:180–7
2. Kuo I, Akpa BS. Validity of the lipid sink as a mechanism for the reversal of local anesthetic systemic toxicity: A physiologically based pharmacokinetic model study. ANESTHESIOLOGY. 2013;118:1350–61
3. Rowland M, Peck C, Tucker G. Physiologically-based pharmacokinetics in drug development and regulatory science. Annu Rev Pharmacol Toxicol. 2011;51:45–73
4. Tucker GT. Pharmacokinetics and pharmacodynamics—Evolution of current concepts. Anaesthetic Pharmacol Rev. 1994;2:177–87
5. Tucker GT, Mather LE. Pharmacokinetics of local anaesthetic agents. Br J Anaesth. 1975;47:213–24
6. Burm AG, de Boer AG, van Kleef JW, Vermeulen NP, de Leede LG, Spierdijk J, Breimer DD. Pharmacokinetics of lidocaine and bupivacaine and stable isotope labelled analogues: A study in healthy volunteers. Biopharm Drug Dispos. 1988;9:85–95
7. Fettiplace MR, Akpa BS, Ripper R, Zider B, Lang J, Rubinstein I, Weinberg G. Rescuscitation with lipid emulsion: Dose-dependent recovery from cardiac pharmacotoxicity requires a cardiotonic effect. ANESTHESIOLOGY. 2014;120:915–25
© 2014 American Society of Anesthesiologists, Inc.