Lipid Resuscitation of Bupivacaine Toxicity: Long-chain Triglyceride Emulsion Provides Benefits over Long- and Medium-chain Triglyceride Emulsion

• Intravenous lipid emulsion is an effective treatment for local anesthetic-induced cardiovascular collapse

• Intravenous lipid emulsions contain long-chain triglycerides and long- and medium-chain triglycerides, components which are essential to its benefits in this setting

• Long-chain triglyceride emulsion in rats with local anesthetic-induced cardiovascular collapse was superior to long- and medium-chain triglyceride emulsion and was associated with fewer recurrences of asystole after resuscitation and lower myocardial bupivacaine concentrations

FOLLOWING the serendipitous discovery of lipid rescue from local anesthetic-induced cardiac arrest by Weinberg et al.  1in 1998, a growing body of reported data has provided evidence for the efficacy of lipid rescue in both laboratory models and humans, thereby establishing intravenous lipid emulsion as an effective treatment for local anesthetic-induced cardiovascular collapse.2–3 

Intravenous lipid emulsions containing long-chain triglycerides (LCTs) and long- and medium-chain triglycerides (LCTs/MCTs) are widely used in clinical scenarios. The two formulation types of commercially available lipid emulsions most commonly used are Intralipid (a 100% LCT emulsion; Huarui Pharmaceuticals Co., Ltd., Wuxi, China) and Lipovenoes (a 50:50 wt:wt mixed LCT/MCT emulsion; Huarui Pharmaceuticals Co., Ltd.). Intralipid contains 100% soybean oil 200 g/L, phosphatide 12 g/L, and glycerol 22 g/L, whereas Lipovenoes consists of a mixture of 50% soybean oil and 50% medium chain triglycerides 200 g/L, phosphatide 12 g/L, and glycerol 25 g/L.

Most of the published work to date has focused on LCT emulsions, Intralipid4,–,9and Liposyn III.10Recently there have been several reported cases of successful resuscitation using Medialipid 20%,11–12which contains 50/50 LCTs/MCTs. Clear recommendations on lipid therapy have now been published by both the Association of Anesthetists of Great Britain and Ireland and American Society of Regional Anesthesia and Pain Medicine.2Specifically, both professional societies currently recommend administration of an initial bolus dose of 1.5 ml/kg followed by an infusion of 15 ml · kg⁻¹· h⁻¹. If cardiovascular stability is not attained or deteriorates, rebolusing at 5-min intervals and doubling the infusion rate should be considered. Nevertheless, the recommendations do not address the choice of lipid emulsions.

The exact mechanism of lipid rescue remains unclear. Currently, the most favored theoretical mechanism is the “lipid sink” theory, in which the lipophilic local anesthetics are bound by lipid, thus reducing tissue content of the toxin.1An alternative mechanism is the “lipid flux” theory, whereby lipid emulsions supply the mitochondria with sufficient substrate to enable energy production, which counters the impaired fatty acid delivery caused by local anesthetics.13Furthermore, there are reports of differences in speed of hydrolysis and oxidation,14the binding affinity of bupivacaine,15the production of ketone bodies,16and metabolic pathways of transport and utilization17between LCT and LCT/MCT emulsions. Some authors also believe that they may have different effects on bupivacaine-induced cardiotoxicity because of their different particle sizes (430 nm vs.  280 nm, respectively).18 

We hypothesize that LCT emulsions provide similar return of spontaneous circulation (ROSC), fewer recurrences of asystole after ROSC, and lower myocardial bupivacaine concentrations than did LCT/MCT emulsions. Accordingly, a double-blind prospective randomized animal study was undertaken with the primary aim of establishing the rate of mortality after ROSC following administration of LCT (Intralipid, 20%) or LCT/MCT (Lipovenoes, 20%) emulsions in an established rat model of bupivacaine cardiotoxicity, and secondarily, to determine the rate of ROSC and survival at an endpoint of 120 min. In addition, plasma and myocardial bupivacaine and triglyceride concentrations, as well as myocardial bioenergetics, were determined to account for the possible difference between them.

Two experiments were performed. In experiment A, the effects of two commercially available lipid emulsions of different compositions, on the recovery from bupivacaine toxicity in a cardiac rat model, were evaluated. In experiment B, plasma and cardiac tissue bupivacaine concentrations under the two lipid emulsion treatments at predetermined time levels were analyzed, as were triglyceride levels and myocardial bioenergetics. All studies were conducted with the approval of Wenzhou Medical College's Animal Care and Use Committee (Wenzhou, China).

Healthy male Sprague-Dawley rats, 7 to 8 weeks old, were purchased from Shanghai Slac Laboratory Animal Co., Ltd. (Shanghai, China), and housed in plastic cages (4 rats per cage) at the animal center of Wenzhou Medical College in a standard 12-h reverse day/night cycle at an ambient temperature of 26°C. The rats had an ad libitum  diet of stock laboratory diet (Beijing Keao-Xieli Feedstuff Co., Ltd., Beijing, China) and tap water. They had a recovery period of 1 to 2 weeks following transportation to Wenzhou Medical College before experiments. After the animals had exhibited adequate weight gain, rats age 8–10 weeks, weighing 300–400 g, were entered into subsequent studies.

The rats were fasted for 12 h before the experiments, with free access to water. All experiments commenced at or before 9 AM. On the day of the experiment, rats were anesthetized with an intraperitoneal injection of chloral hydrate (350 mg/kg). They were then intubated via  tracheotomy and mechanically ventilated with 1% to 2% sevoflurane in 100% oxygen, using a rodent volume-controlled ventilator (HX-300, TME Technology Co., Ltd., Chengdu, China): tidal volume = 8 ml/kg, respiratory rate = 75–80 breaths/min, inspiratory:expiratory ratio = 2:3. The right internal jugular vein, the left femoral vein, and femoral artery were cannulated. Body temperature was maintained at 38° to 39°C with a heating lamp held at a safe distance. Electrocardiography, using three subcutaneous needle electrodes, and the arterial pressure were recorded continuously throughout the duration of the protocol by a MedLab data archiving and retrieval system using U/4C051 (Nanjing Medease Science and Technology Co., Ltd., Jiangsu, China). After completion of invasive procedures, all animals were then allowed to stabilize for 15 min at 0.5% sevoflurane and 100% oxygen, after which baseline heart rate, mean arterial pressure, and rate-pressure product (RPP) were recorded.

At the end of the stabilization period, sevoflurane was discontinued and immediately thereafter 20 mg/kg bupivacaine hydrochloride (Sigma-Aldrich Co., St. Louis, MO) was injected as an intravenous bolus over 20 s via  the venous cannula in the right femoral vein. The dose of 20 mg/kg was elected based on the work of Weinberg et al.  6All rats developed asystole by the end of the bupivacaine infusion, and this was taken as time zero.

Cardiopulmonary resuscitation with mechanical ventilation and manually performed external chest compressions (approximately 30% anterior-posterior chest diameter at 200 compressions/min) were instituted immediately upon the onset of cardiac arrest (time zero). At the same time, animals were entered in a lipid infusion protocol as determined by prior randomization, and received Intralipid 20% for the LCT group, or Lipovenoes 20% for the LCT/MCT group, as a 5 ml/kg bolus over 30 s followed by a continuous infusion of 1.0 ml · kg⁻¹· min⁻¹for 3 min, via  the right femoral vein and the right internal jugular vein, respectively. This lipid therapy regimen was based on the work of Di Gregorio et al.  9and Hiller et al.  19with some modifications to optimize recovery in this model of bupivacaine overdose. Total dosing of 8 ml/kg was elected in accordance with the clinical recommendation of Weinberg.20–21A bolus of 10 μg/kg epinephrine (Shanghai Harvest Pharmaceutical Co., Ltd., Shanghai, China) was then administered over the subsequent 5 s at 1-min intervals until ROSC, with a maximum cumulative dose of 100 μg/kg. Chest compressions were delivered until a native RPP more than 20% of baseline value for 1 min or longer, which was our criterion for ROSC.9All chest compressions were stopped at the 15-min point regardless of RPP, and subjects were evaluated for sustained or nonsustained recovery until the 20-min point. Ventilation with 100% oxygen was continued throughout the 20 min of resuscitation efforts. Cardiopulmonary resuscitation was not recommenced after initial ROSC in animals subsequently developing a second period of cardiac arrest. All intravenous fluids were prepared and preheated to 37°C before infusion by personnel that did not participate in resuscitation efforts. Coordinated resuscitation efforts with clear division of roles were conducted according to the recommendation of Smith22with the following modifications: (1) provider one initiated specific lipid therapy and epinephrine administration, (2) provider two performed external chest compressions, and (3) provider three managed the airway and assessed native RPP during resuscitation. The two types of lipid emulsions have the same visual appearance, and all providers were blinded regarding the group.

Sixty rats were randomly allocated by the random table method before the study into LCT group and LCT/MCT group, with 30 animals in each group. The animal experimental procedures were conducted using the aforementioned bupivacaine arrest and resuscitation protocol. LCT and LCT/MCT groups received Intralipid 20% or Lipovenoes 20%, respectively. All surviving rats in both groups were observed for 120 min.

ROSC and rats surviving at 120 min were recorded. We then calculated the rate of ROSC, survival at 120 min, and mortality after resuscitation (rate of ROSC = number of rats displaying ROSC/total number of rats; survival at 120 min = number of rats survived at 120-min/total number of rats; mortality after resuscitation = cases of repeated asystole after ROSC/cases of initial ROSC). We also continuously recorded the systolic blood pressure, mean arterial pressure, heart rate, and RPP until the termination of the observation period, at which time we calculated the RPP recovery ratio (RPP recovery ratio = RPP of survival rats at specific time points/baseline RPP value). Time to first heartbeat, time to ROSC, and the cumulative dose of epinephrine were recorded.

A second group of 80 rats were assigned to 10 groups (eight animals per group) based on differing observation time using the random table method. The groups were defined as LCT⁰, LCT¹⁵, LCT³⁰, LCT⁶⁰, and LCT¹²⁰and LCT/MCT⁰, LCT/MCT¹⁵, LCT/MCT³⁰, LCT/MCT⁶⁰, and LCT/MCT¹²⁰. After asystole, LCT¹⁵–LCT¹²⁰groups and LCT/MCT¹⁵–LCT/MCT¹²⁰groups received Intralipid 20% or Lipovenoes 20%, respectively. Monitoring and evaluation of hemodynamic status and airway management were conducted continuously during cardiopulmonary resuscitation and throughout respective observation period. At the end of the observation period, all the surviving animals in LCT¹⁵–LCT¹²⁰and LCT/MCT¹⁵–LCT/MCT¹²⁰groups were killed with cervical dislocation and a 4 ml blood sample was immediately taken by needle aspiration of the cardiac apex. The same sampling method was used for all animals in LCT⁰and LCT/MCT⁰groups upon the onset of cardiac arrest (without lipid emulsions infusion). The blood plasma was separated by centrifugation (for 10 min at 3,000 revolutions per minute) and cardiac tissues were rinsed to remove any residual blood. Finally, all samples were then stored at −70°C until analysis was performed.

ROSC and rat survival to the end of their respective observation periods were recorded. For all groups, the plasma and cardiac tissue bupivacaine concentrations were analyzed similar to previous reports23from this laboratory by high-performance liquid chromatography/mass spectrometry. Cardiac tissue sample preparation for triglyceride determination was conducted using the Song24procedure. Both plasma and tissue samples were then examined by a fully automatic Architect C 8000 Analyzer (Abbott Laboratories, Chicago, IL). Adenosine triphosphate (ATP), adenosine diphosphate (ADP), and adenosine monophosphate from heart tissue were measured by high-performance liquid chromatography. Sample preparation was conducted using the Stehr25procedure. Measurements were performed on an Agilent 1100 series high-performance liquid chromatography (Agilent Technologies, Inc., Santa Clara, CA) equipped with UV-visible detection (wavelength range from 190 to 600 nm). Separations were carried out on a Hypersil ODS-2 column (200 × 4.6 mm, 5 μm; Dalian Elite Analytical Instrument Co., Ltd., Dalian, China) at 30°C. The mobile phase was a mixture of eluent A and B (pH = 6.0, 99:1, v:v) at a flow rate of 1.0 ml/min. Eluent A was phosphate buffer (12 MM Na²HPO⁴and 88 MM NaH²PO⁴, pH = 6.0) and eluent B was a mixture of 85% eluent A and 15% acetonitrile. An injection volume of 20 μl was used applying a 7725i Rheodyne manual injector (Agilent Technologies, Inc.) for both standard samples and tissue extract samples. The detection wavelength was set at 254 nm. Equally treated external standards of known concentrations were used to check retention times and to permit sample quantification based on the analysis of peak area. The adenylate energy charge was calculated according to the formula: energy charge = (ATP + 0.5 × ADP) × (ATP + ADP + adenosine monophosphate)⁻¹.

Power analysis was based on results of preliminary data comparing mortality after ROSC between groupings. In our preliminary study, 16 rats received lipid emulsions immediately after bupivacaine-induced asystole (n = 8 for mIntralipid treatment; n = 8 for Lipovenoes treatment). One of seven (14.3%) animals displaying initial ROSC developed a second cardiac arrest after intralipid administration, whereas 3 of 6 (50%) survivors suffered a repeated asystole after Lipovenoes administration. Sample size calculations showed that a chi-square test with a type I error (two-sided) of 0.05 will have 88% power to detect a 35.7% difference in mortality after ROSC between the two groups, when the sample size in each group is 29. To account for potential attrition, we enrolled 30 rats per group.

Considering a survival of 75% (6 of 8) at 120-min endpoint for intralipid treatment and 37.5% (3 of 8) for Lipovenoes treatment shown in the preliminary experiment, eight rats were enrolled in each group so as to avoid a small number of survivors, which may statistically lead to inaccurate results obtained from determination of myocardial bupivacaine and triglyceride levels.

Data were tested for normal distribution using the Shapiro–Wilk test and the Kolmogorov–Smirnov test. Continuous variables were presented using means and SDs or medians and interquartile values, and in the case of categorical variables, frequencies were used. In experiment A, values for baseline parameters, times to ROSC, and times to first heartbeat were compared by Student t  test, and the cumulative epinephrine dose was analyzed using the Wilcoxon and Mann–Whitney U test. Dichotomous outcomes were analyzed with Fisher exact test. Continuous hemodynamic variables in animals surviving at 120 min were compared across time by two-way repeated measures ANOVA, with Bonferroni correction posttesting when significance was achieved (P < 0.05). Time-to-event data were studied with Kaplan–Meier analysis and compared using the log-rank test. In experiment B, baseline parameters were analyzed by one-way ANOVA. The comparison of bupivacaine, triglyceride, and adenine-nucleotides between groups with the same observation time was evaluated by Student t  test. As primary outcome variables, bupivacaine among the same emulsion-treated groups was compared using an initial Kruskal–Wallis H-test. If significance was achieved for differences, subanalysis according to the Bonferroni correction was applied. All statistical analysis was performed with SPSS for Windows (version 14.0; SPSS, Chicago, IL). Statistical significance was considered as P < 0.05.

Body weight, baseline values of mean arterial blood pressure, heart rate, or RPP before intervention did not differ significantly between LCT and LCT/MCT groups (table 1).

The LCT and LCT/MCT groups displayed similar ROSC (24 and 23 animals respectively, 80.0%vs.  76.7%, P = 0.754). More importantly, 8 of 23 animals displaying initial ROSC in LCT/MCT group subsequently developed a second period of intractable cardiac arrest compared with only 2 of 24 in LCT group (8.3%vs.  34.8%, P = 0.027). Almost all of repeated cardiac arrest occurred in the period from 10 to 45 min. Therefore, 22 animals in LCT group survived to protocol termination, compared with 15 animals in LCT/MCT group (73.3%vs.  50.0%, P = 0.063). There were no significant differences between LCT and LCT/MCT groups in respect to time to ROSC (P = 0.264), time to first heartbeat (P = 0.086), and cumulative dose of epinephrine (P = 0.529) (table 2). Death distribution and Kaplan–Meier survival curves of rats that exhibited ROSC in LCT and LCT/MCT groups are presented in figure 1. The median survival time of animals displaying initial ROSC was 120 (interquartile, 0) min and 120 (interquartile, 91) min in LCT and LCT/MCT groups, respectively (chi-square test = 4.643, P = 0.031).

Hemodynamic parameters heart rate (fig. 2), systolic blood pressure (fig. 3), and RPP (fig. 4) in animals surviving to protocol termination are presented graphically. LCT emulsions produced superior conditions in terms of heart rate (F = 22.69, P < 0.001), systolic blood pressure (F = 4.35, P = 0.044), and RPP (F = 17.87, P < 0.001) than did LCT/MCT emulsions. The graphical disparity reported primarily occurred at the midpoint of the observation period, and the curves of both groups flatten out gradually in the later periods. LCT emulsions resulted in a more complete recovery of RPP at 15 (P = 0.004), 30 (P = 0.001), and 60 min (P < 0.001) than did LCT/MCT emulsions. RPP recovery ratio for the 90–120 min did not differ between the groups (P > 0.05) (fig. 5).

In experiment B, there were 6, 5, 6, and 6 surviving rats in LCT¹⁵, LCT³⁰, LCT⁶⁰, and LCT¹²⁰groups, respectively, and 4, 4, 4, and 5 in LCT/MCT¹⁵, LCT/MCT³⁰, LCT/MCT⁶⁰, and LCT/MCT¹²⁰groups, respectively. Baseline body weight and hemodynamic parameters (heart rate, mean arterial pressure, and RPP) were similar among the 10 groups (table 1).

The concentrations of bupivacaine in cardiac tissue and plasma are presented graphically in figure 6. Data indicate a lower bupivacaine level in the LCT¹⁵group (8.1 ± 2.2 μg/g; 14.1 ± 2.2 μg/ml; cardiac tissue content, plasma concentration, respectively) than the LCT/MCT¹⁵group (13.8 ± 1.7 μg/g; 29.7 ± 5.5 μg/ml), and in the LCT⁶⁰group (2.0 ± 0.8 μg/g; 3.8 ± 1.8 μg/ml) than the LCT/MCT⁶⁰group (5.9 ± 0.7 μg/g; 16.1 ± 1.8 μg/ml). Mean ± SD; P < 0.05 for all shown results. LCT emulsion-treated groups showed a monotonously decreased plasma bupivacaine, whereas LCT/MCT emulsion-treated groups exhibit an overall decline with an unexpected plasma concentration rebound that was 16.1 ± 1.8 μg/ml in the LCT/MCT⁶⁰group and 8.3 ± 2.7 μg/ml in the LCT/MCT³⁰group, P = 0.003.

There was no difference in the triglyceride content of cardiac tissue between groups with the same observation time (P > 0.05) (fig. 6B). No difference in plasma triglyceride levels was found between LCT⁰and LCT/MCT⁰, as well as between LCT¹⁵and LCT/MCT¹⁵groups (P > 0.05) (fig. 6D). There was a steady decline from the 15-min point (LCT/MCT¹⁵) to the 120-min point (LCT/MCT¹²⁰). It was lower in the LCT/MCT¹²⁰group as compared to the LCT¹²⁰group (P < 0.001).

No major differences in the myocardial bioenergetics between groups with the same observation time was observed, including ATP, ADP, adenosine monophosphate, and energy charge (fig. 7).

This rodent model demonstrated that both LCT and LCT/MCT emulsions reversed the cardiac arrest induced by the intravenous injection of 20 mg/kg bupivacaine. They produced the relatively similar rates of ROSC (P = 0.754). There was, however, a significant difference in mortality after resuscitation was observed between the LCT and LCT/MCT groups (8.3%vs.  34.8%, P = 0.027). With respect to survival at 120 min, it approached significance (P = 0.063); we had powered the study to find a 35.7% difference (β = .88), but we only found a 23.3% difference. Thus, the null hypothesis regarding survival may not have been rejected (potential Type II error). In addition, the RPP recovery ratio during the period from 15 to 60 min were also much higher in LCT group. The concentrations of bupivacaine in cardiac tissue and arterial plasma were significantly lower in LCT¹⁵and LCT⁶⁰groups than those in LCT/MCT¹⁵and LCT/MCT⁶⁰groups, respectively.

Epinephrine is a first-line drug for treating cardiac arrest because its α-agonist activity increases diastolic pressure and coronary perfusion pressure.26Recent work by Harvey et al.  5demonstrated that epinephrine was necessary for return of spontaneous circulation in lipid-based resuscitation from local anesthetic-induced cardiac arrest. This finding is consistent with our previous work in the isolated rat heart model of bupivacaine toxicity, wherein lipid emulsion combined with epinephrine produced superior coronary flow and RPP value than resuscitation with lipid emulsion alone.27 

The results in experiment B indicate that both treatments resulted in a sharp decline in total myocardial bupivacaine concentration during the initial 30 min after asystole, a time period during which cardiac tissue bupivacaine content decreased by 90.4% after LCT emulsion treatment and by 93.0% after LCT/MCT emulsion treatment. Overall, the similar measurements at 0 min and 30 min suggested they may exert their curative effects in a 30-min time period. Thereafter, it was observed that the plasma bupivacaine concentration dramatically increased in the LCT/MCT⁶⁰group relative to the LCT/MCT³⁰group, whereas it exhibited an extended decline after LCT emulsion treatment, which indicated a longer action time when LCT emulsion was compared to LCT/MCT emulsion. The half-life may have accounted for this observation: the estimated half-life was 33 min for LCTs and 17 min for MCTs,28for which clearing of the plasma LCT/MCT was completed earlier than LCT.14More studies suggest that a substantial amount of exogenous fatty acids reach both hepatic and nonhepatic tissues as intact triacylglycerol without preceding lipolysis in the plasma compartment.29On the basis of the “lipid sink” hypothesis and the profound liver-targeting property for lipid emulsions,30–31it can be hypothesized that the redistribution and metabolism of lipid emulsion upon hemodynamic recovery may have allowed subsequent hepatic release of bupivacaine that was incorporated within the core of LCT/MCT lipid emulsion. This hypothesis may explain the distinct effects in hemodynamic parameters during the period from 15 to 60 min as well as the deaths after resuscitation concentrated in 10–45 min in LCT/MCT group in experiment A. However, we did not measure the hepatic bupivacaine content, and this is a possibility for future research.

In addition to the above, the high lipid solubility of local anesthetics and the difference in the high binding capacity of the lipid emulsions explained their distinct efficacy. Mazoit et al. ,15using the “shake-flask” method in vitro  to measure the extent of emulsification of lipophilic local anesthetics, found that the extent of local anesthetic binding to Intralipid 20% was roughly 2.5 times more than Medialipide 20% (B-Braun, Boulogne, France). This indicated that LCT emulsions were more efficient than LCT/MCT emulsions in binding to long-acting local anesthetics. Compared with the LCT/MCT¹⁵group, the lower cardiac tissue bupivacaine content in the LCT¹⁵group may chiefly result from the better capacity of LCT lipid emulsion to bind bupivacaine and an ensuing liver-targeting transport of lipid-soluble bupivacaine. The Lipovenoes in our protocol is similar to the Medialipid used by Mazoit et al. , as both of them are lipid emulsions containing equal proportions of LCTs and MCTs (although made by different manufacturers).

Under normal aerobic conditions, fatty acids are the preferred substrate for cardiomyocyte, generating about 80–90% of ATP.32Lipid emulsions could theoretically increase intracellular fatty acid content, by impeding a local anesthetic's inhibition of acylcarnitine and ATP synthesis7,13in the cardiomyocyte, thereby improving ATP production cardiac function. Carnitine is needed to transport long-chain fatty acids into the mitochondria.33Conversely, medium-chain fatty acids can enter the mitochondria by simple diffusion independent of the carnitine enzyme system.34Thus, specificity for rapid oxidation, without deposition in fat stores,14carnitine-independent transport to the site of oxidation, and the ability to supply energy with higher ketone bodies make MCTs theoretically a preferable choice from an energy-related perspective. Unfortunately, the theoretical benefits of LCT/MCT lipid emulsions are not borne out in practice. No difference in ATP, ADP, adenosine monophosphate, and energy charge throughout the duration of the protocol was observed in experiment B, which was similar with that of Stehr's research data from an isolated rat heart model.25 

Our investigation has several limitations. First, our small animal model necessitated that each animal in experiment B had to be sacrificed at its assigned sampling time point. Thus, repeated measures for each animal were not possible. Nonetheless, we stringently controlled the animal profiles including the animal source, strain, sex, age, body weight, housing condition, and experimental environment to reduce the error as much as possible. Second, the dosing protocols of key study drugs (lipid emulsions, epinephrine) that we used may not be the optimal. However, the regimen we used was found in preliminary experiments used to optimize recovery in this model of bupivacaine intoxication. The effective reversal of cardiovascular symptoms and our highly successful resuscitation suggest validation of our model. A third limitation regards the power of our study. While we were powered to find a 35.7% difference, we, nonetheless, still present a solid basis for further experimentation and discussion, especially in light of the fact that our work supports Weinberg's6,7,9,19and confirms his studies.

In conclusion, LCT emulsion may be the choice of lipid resuscitation from bupivacaine toxicity as it provided less mortality after resuscitation, lower myocardial bupivacaine concentrations, and superior hemodynamic recovery than did LCT/MCT emulsion in an intact rat model of bupivacaine-induced cardiovascular collapse, which we would attribute to the differences of “lipid sink” effect, rather than to the “lipid flux” theory. LCT emulsion's superiorities showed in the present study may further strengthen the guidelines for the management of severe local anesthetic toxicity with lipid emulsion. Further research may be warranted regarding the comparison of LCT and LCT/MCT emulsions in another humanized animal model, although it is unlikely that LCT/MCT emulsions will show any benefit over LCT emulsions.

The authors thank Lianbo Yu, Ph.D., Biostatistical Scientist, Center for Biostatistics, Columbus, Ohio, and Hongying Shi, Ph.D., Assistant Professor, Department of Preventive Medicine, Wenzhou Medical College, Wenzhou, Zhejiang, China, for analyzing the data.