The interpretation of experimental data and extrapolation to the clinical setting is always a challenge, especially in those circumstances for which evidence-based clinical results will never be obtained. In the case of lipid rescue, although case reports are the best (and perhaps only) clinical evidence and can suggest the mechanisms underlying the supposed therapeutic effects of lipid emulsion, in fact, care must be taken in interpreting results from case reports and linking them to experimental data.1 Although case reports do assume a strong relationship between cardiac arrest and the administration of local anesthetics, as well as that between lipid infusion and the return of spontaneous circulation, other reasons for cardiac arrest and the return of spontaneous circulation cannot be excluded. In addition, factors (e.g., high likelihood of reporting bias and dose-response gradient) that may decrease or increase the strength of evidence must also be kept in mind. Therefore, we should more closely examine results of published case reports. First, there are many case reports showing successful resuscitation in patients with ropivacaine-induced cardiac toxicity, even without administrating lipid.2 To our knowledge, no published case has shown a failure of resuscitation in ropivacaine-induced cardiac toxicity, even before the era of lipid resuscitation. These data support experimental observations, demonstrating a return of circulation in almost all cases of ropivacaine-induced cardiac toxicity, but only marginal improvement in bupivacaine-induced cardiac toxicity.3,4 Notably, these and other studies all use the term “cardiac arrest,” and our use of the term cardiac arrest is sufficiently explained in the study protocol described in the Methods section so as to avoid confusion. Second, in the case reported by Ludot et al.,5 the lipid emulsion recommended was not the same as the one administered (Medialipid 20%, a solution containing soya 10 g and medium-chain fatty acids 10 g/100 mL). In contrast to the statements of Warren and Weinberg, other studies support the impression that the desired lipid sink effect can be achieved by nonrecommended lipid emulsions.5,6 In addition, Ludot et al. state that their “experience suggests that the exact formulation of the lipid emulsion may not matter.” However, as correctly pointed out by Warren and Weinberg,7 Mazoit et al.8 showed a difference in the binding capacity of lipid emulsions depending on the lipid emulsion as well as the local anesthetic itself. However, because the free local anesthetic concentration is less for the more lipophilic bupivacaine than it is for the less-lipophilic ropivacaine, the “lipid sink” effect is consequently not as pronounced for ropivacaine as for bupivacaine. Therefore, in our view, these data strongly support our experimental findings. Third, the isolated heart model allows us to investigate direct cardiac effects without metabolic, systemic, or humeral influencing factors. As mentioned in our Methods section, the design of our study was adopted from 2 former studies, and the administration of local anesthetics and the lipid emulsion was standardized throughout the experiment. Therefore, we can rule out inconsistent perfusion and are sure that our results were not biased. Notably, the regimen for administering lipid emulsion followed the recommendation by Weinberg et al.9 To our knowledge, the recommended dose has never been evaluated or even proven in a dose-dependent manner, in either animal or human studies. Therefore, it is not surprising that the lipid infusion administered varied with regard to level of lipid emulsion, duration of the experiment, and dose (range, 0.25–3 mg · kg−1 · min−1) in various studies.10 We used the recommended dose because it makes it easier for the reader and other investigators to compare the results between studies. However, future studies should determine whether humans need only a fraction of the dose that is applied to rodents.10 Similarly, we decided not to inject a bolus of lipid, because the bolus dose applied varies greatly among species and investigations (1–7.5 mL · kg−1). Furthermore, as mentioned in our study, other studies using isolated hearts did not inject a bolus of lipid. In addition, the volume of bolus used previously would have reduced the concentration of the Krebs-Henseleit solution by up to 63% in this study. Finally, when administering large doses of lipid by bolus and continuous infusion, effects other than the so-called lipid sink may partially or completely account for the observed results. Thus, we agree with Warren and Weinberg that other alternatives may explain the lipid rescue phenomenon, as recently reported by our research group.11 Therefore, the aim of our statement in the Discussion section was to clearly emphasize that approaches to address local anesthetic toxicity other than the mere application of lipid are indicated. Moreover, even with usual basic and advanced life support, successful resuscitation is possible in ropivacaine-induced (and probably in mepivacaine-induced) cardiac toxicity.2 York A. Zausig, MD, DEAA Wolfgang Zink, MD, PhD, DEAA Bernhard M. Graf, MD, PhD, MSc Department of Anaesthesiology University of Regensburg Regensburg, Germany [email protected]
Search from 250 M+ research papersSearch from 250 M+ research papers
Jun 13, 2012
Jean-Xavier Mazoit 
Open Access