Department of Anesthesiology, University of Illinois at Chicago, Chicago, IL
Checking for direct PDF access through Ovid
We thank Zausig et al (1) for their interest in our study on intravenous lipid emulsion (ILE) resuscitation (2). We agree that differences in laboratory findings may be secondary to species and methodologic differences, but these differences per se do not address the validity of the lipid sink or other proposed mechanisms for lipid reversal of drug toxicity. We also agree that a volume effect (for example, hemodilution) could theoretically contribute to the efficacy of ILE. However, in two other studies using a similar model, saline (volume) controls compared very unfavorably with the ILE groups in all hemodynamic and metabolic metrics (3, 4). To address this issue further, we analyzed, post facto, blood and tissue samples taken from the animals used in the current report (2).Plasma derived from blood taken at the experiments' conclusion (10 mins after asystole induced by a 20-mg/kg intravenous bupivacaine bolus) was either stored at −80°C (vasopressin-treated group) or first ultracentrifuged to separate the lipid and aqueous plasma components, which were then frozen separately (ILE group); hearts were also removed and frozen at the 10-min time point. One sample from each group was lost to analysis and sample size was therefore n = 5 for both groups and all measurements. We used high-performance liquid chromatography of hexane extracts of plasma and the separated lipid cap to determine the bupivacaine concentrations in matched aqueous and lipid components of the ILE group. The mean calculated partition coefficient of these samples was 20.98 ± 3.8 (sem). This is greater than that identified in previous in vitro mixing experiment (5) (11.9) and suggests the potential for ILE to significantly lower aqueous bupivacaine content in vivo. Comparing the two groups showed that ILE treatment dramatically reduced the aqueous plasma bupivacaine concentration below that of the vasopressin group (12.7 ± 3.08 vs. 37.8 ± 5.94 mM, respectively; mean ± sem and n = 5 for both groups; p < .001 by two-way unpaired t test using Welch's correction). This 66% reduction in bupivacaine concentration is roughly threefold greater than that predicted by hemodilution alone (23%) assuming a 30% increase in blood volume as suggested by Zausig et al (1). We then measured bupivacaine content in cardiac tissue using mass spectroscopy of hexane extracts of stored myocardium samples. We found a strong linear correlation of bupivacaine content in cardiac tissue with aqueous plasma bupivacaine concentration (Fig. 1A) and a strong first-order correlation between bupivacaine content in cardiac tissue and rate–pressure product, our primary hemodynamic metric of cardiac function and recovery.This is the first study linking ILE to reduced myocardial tissue content of an offending toxin. Furthermore, the strong correlation between aqueous plasma concentration and tissue content supports the lipid sink mechanism in ILE. This relationship and their association with improved hemodynamic recovery in ILE is graphically illustrated in Figure 1C, in which the surfaces representing data from lipid- and vasopressin-treated groups are entirely nonoverlapping.Dr. Weinberg holds a patent for “Lipid Emulsion in the Treatment of Systemic Poisoning.” The remaining authors have not disclosed any potential conflicts of interest.