In Reply

We thank Dr. Hori et al. for their interest in our article¹  and their valuable and critical comments. We share their curiosity regarding the “direct lipid effect” described and agree that other explanations for the effect of the lipid emulsion alone should be ruled out before this effect can be regarded as proven. Dr. Hori et al. raise two concerns: one regarding possible differences in Na⁺ concentration between the lipid and control groups and the other concerning the presence of residual triglycerides in the centrifuged solutions.

As Dr. Hori et al. noted correctly, it is important to keep the Na⁺ concentration constant during experiments assessing Na⁺ current magnitude. Because the information on the Na⁺ concentration of Lipovenös^® (Fresenius Kabi AG, Bad Homburg, Germany) provided by the supplier is somewhat vague (up to 5 mM), we measured it ourselves by flame photometry in the initial charge of Lipovenös^® used. The result was 2.06 ± 0.05 mM (n = 3, mean ± SD), which we referred to as “approximately 2 mM” in the article, and we therefore included 2 mM Na⁺ in the corresponding control solution. Upon receiving your comments, on request to the supplier, we learned that the Na⁺ concentration of Lipovenös^® is less than 5 mM but may vary from charge to charge. We therefore also measured the Na⁺ concentration of Lipovenös^® in the second charge we used in our study. The result was 3.18 ± 0.12 mM (n = 3, mean ± SD) and therefore approximately 1 mM higher that the 2 mM used in our control solution. Nevertheless, because the final solution used for experiments contained only 10% of either Lipovenös^® or control, the difference in the Na⁺ concentration in our experiments was still marginal: 18.2 mM under control conditions and 18.3 mM in the presence of Lipovenös^®. Using the Goldman–Hodgkin–Katz equation, we calculated the expected Na⁺ current increase (at V_Pip = −40 mV) caused by increasing the Na⁺ concentration from 18.2 to 18.3 mM to be 0.7%. We therefore conclude that the measurements of I_Na were not disturbed by these marginal differences in Na⁺ concentration and that the direct lipid effect of Lipovenös^® must have a different cause.

We thank Dr. Hori et al. for sharing their interesting results regarding the removal of triglycerides by ultracentrifugation of Lipofundin^® 20% (B. Braun Melsungen AG, Melsungen, Germany). We are very pleased about their demonstration that by ultracentrifugation it is possible to remove more than 99.5% of the triglycerides from a lipid emulsion. Yet, we were surprised by the results reported by Nadrowitz et al. in their elegant and interesting article that even 0.05% Lipofundin^® has some effect on Nav1.5-mediated currents.²  A solution containing 0.05% Lipovenös^® appears clear to the eye and therefore it is perfectly possible that our centrifuged emulsions contained approximately 10 mg/dl triglycerides as suggested. Consequently, a part of a “direct lipid effect” may still be present even in the centrifuged emulsions. It should be noted, however, that Nadrowitz et al. used Lipofundin^® and Intralipid^® in their experiments, whereas we used Lipovenös^®. Even though Lipofundin^® and Lipovenös^® contain a similar mixture of triglycerides containing long- and medium-chain fatty acids, presently it is unclear whether the medium-/long-chain fatty acid mixture of Lipofundin^® or a different component (that may or may not be present also in Lipovenös^®) is responsible for the inhibition of Nav1.5-mediated currents demonstrated by Nadrowitz et al. In this regard, it is interesting that although Lipofundin^® inhibited Nav1.5-mediated currents, Intralipid^® did not. It seems prudent to assume that this is due to differences in lipid content, however, at this point this is a speculation and warrants further exploration.

In our article, to validate the results, we compared the apparent reduction of the bupivacaine concentration as assessed by concentration–response analysis of the patch clamp experiments to the actual reduction of the bupivacaine concentration as assessed by gas chromatography–mass spectrometry. We found that both approaches yielded similar bupivacaine concentrations in the centrifuged lipid emulsions. Consequently, we do not expect that residual triglycerides have significantly affected our results.

Taken together, we do not think that the reasonable concerns raised by Hori et al. can explain the “direct lipid effect” as described in our article. Yet, we are grateful for their comment as it clearly points out that the nature of what we have called “direct lipid effect” in our article is, at present, unclear. In fact, it includes every effect that cannot be attributed to the lipid sink. Clearly, we cannot rule out that a part of this effect may be explained by limitations of our experimental approach, but most importantly, more experiments are necessary to explore the nature of this effect.