Excellent review of lipid rescue therapy

February 25, 2015, 6:47 pm


Intravenous Lipid Emulsion in the Emergency Department: A Systematic Review of Recent Literature. Cao D et al. J Emerg Med 2014 Dec 19 [Epub ahead of print]


This excellent comprehensive review of lipid rescue therapy (LRT) is vitiated only by the unavoidable fact that available clinical evidence  is so inconclusive. As the authors point out, published literature consists mostly of case reports and small case series. The vast majority of these reported cases have good outcomes and reflect positive effects from ILE, but the evidence is marred by multiple confounding variables (such as concurrent treatment) that are impossible to correct for, and by publication bias. There have been no prospective randomized controlled trials.Intravenous lipid emulsion is now considered first-line therapy for local anesthetic systemic toxicity (LAST,) but generally reserved as a last-ditch effort for patients severely ill from other drugs who are not responding to more established interventions.

The authors conducted an extensive literature review to identify publications describing human overdose cases treated with LRT in which outcomes were reported. They found 94 articles and 40 abstracts, which are cited in a very complete bibliography current to 2014.

This entire review is well worth reading. I will touch on some points I found particularly interesting.

Table 1 lists the different drugs involved in overdose cases where LRT was used, along with whether the observed result was “positive effect” or “no apparent effect.” (For some reason they do not report possible negative effects, although one article cited in their reference list described 2 patients who developed asystole shortly after receiving LRT.) The list includes:

  • local anesthetics (primarily bupivacaine)
  • anti-depressants (including amitriptyline, citalopram, bupropion, and venlafaxine)
  • anti-psychotics (including quetiapine and olanzapine)
  • cardiovascular medications (primarily calcium-channel-blockers, beta-blockers and anti-arrhythmics)

There are cases cited of overdose with various other agents, including aconite, lamotrigine, baclofen, and chloroquine. Most of the drugs involved were lipophilic, but positive effects were observed also with water-soluble drugs, such as metoprolol, atenolol, labetalol, and amphetamine.

The authors note that at this time the recommendations for LRT dosing in overdose cases is empiric, based on the experience with treating LAST. They do not discuss a point we touched on in our recent podcast with Dr. Guy Weinberg. With LAST, a relatively large dose of toxic agent is rapidly taken up systemically, and quickly eliminated if the patient survives. With an oral overdose, systemic absorption can continue for hours or even days. Optimal LRT treatment in these cases might require a completely different dosing schedule. Unfortunately, that question has not been studied.

Related posts:

TPR Podcast #7: Interview with Guy Weinberg about lipid rescue therapy

Acute respiratory distress syndrome following intralipid emulsion therapy 

Complications associated with lipid emulsion therapy

Lipid rescue therapy can interfere with critical lab values

Case report: cocaine cardiotoxicity treated with intravenous lipid infusion

Lipid emulsion therapy for poisonings: a review

Lipid emulsion overdose


  1. Guy Weinberg Says:

    I would like to offer a few thoughts on the generally excellent review of Cao et al and Leon’s comments. They have done a service by providing an exhaustive analysis of the available clinical literature on ILE – a daunting undertaking for any team. I’m slightly disappointed that it took almost a year (Dec 2014) since the end of their literature search (Feb 2014) for the paper to get to print. There have been several interesting (and arguably) important reports in the intervening period. I prefer to blame the publisher – but it would have been nice to see their analysis of the relevant literature from that period.

    The authors mention the oft-cited issue of ‘positive reporting bias’. I see several factors that make this less problematic than assumed. First, is that publishing bias per se is more about prejudice in favor of significant results than positive results. In fact, as a favorable clinical response is more established (as for LAST) then one expects 1) increasing negative bias in reporting as these results become more noteworthy; I don’t believe we’re seeing this and 2) a higher standard for ‘significance’ to justify publication…that is, cases become publishable only when they are remarkable in some way. I think we do see this now. The main effect of 2) is increasing ‘file cabinet’ phenomenon: a save isn’t reportable unless it’s really amazing so that cases illustrating good clinical response go unreported in part because they don’t reach that standard of ‘amazing’. The authors state in their conclusion that ‘..substantial publication bias toward positive results precludes lipid emulsion therapy as a first line agent for indications other than local anesthetic systemic toxicity.’ I don’t’ necessarily disagree (at this stage) with the authors’ conclusion but neither do I agree with their rationale. Positive reporting bias doesn’t indicate a lack of efficacy. After all, if there weren’t a predominance of positive case reports (here, 85%) we wouldn’t be having this discussion. Case reports alone don’t tell the entire story. Solid scientific study is needed to test the null hypothesis.

    Another key reason cases aren’t reported is publishing (author) fatigue – no one bothers to write up positive cases because it’s too much effort or they figure publishers (and readers) aren’t interested. I’m painfully aware of this phenomenon as I’m very (very) often met by people at meetings who tell me of their most recent successful lipid save(s). None of these gets written up. Look for example to the link Leon provides (at Chris Nickson’s blog, lifeinthefastlane.com) and see the quote about TCA at the bottom by Sarah Silver “I should’ve written up case reports. Have had successful case of reversal of cardiac arrest with intralipid therapy: overdose of amitryptyline (roughly 90 tablets of 50 mg preparation). “ Publishing fatigue.

    Another factor that both Leon and the authors cite as a weakness in such reports is the confounding effects of multiple treatments. Granted it’s impossible in such scenarios to ascribe resuscitation to any one therapy … unless it is obvious that nothing has worked and a dramatic effect is then temporally related to lipid therapy. While this isn’t always the case, in those where it is, the treating physicians will likely believe the association of lipid with resuscitation is causal. See for example the well-known case of bupropion/lamotrigine overdose reported by Sirianni et al; there are many other such examples that one could argue have weight that carries them beyond the confounders.

    The authors believe that using a treatment very close to the end of resuscitation places it ‘most proximal to the desired outcome’. I believe this actually makes the test for efficacy much more stringent since these patients are literally closer to death and have more profound biochemical, metabolic and circulatory derangements – the chances of survival much worse than at the beginning of resuscitation. If something works then, it’s not likely because all the other treatments suddenly kicked in – anyone who’s done their share of CPR knows this.

    More later …. To avoid reader fatigue.

  2. Leon Says:


    Those are all great points. i actually think the fact that the authors’ most recent reference was published only a year ago shows that the editorial process was remarkably streamlined. As Joe Lex has said:

    If you want to know how we practiced medicine 5 years ago, read a textbook

    If you want to know how we practiced medicine 2 years ago, read a journal

    If you want to know how we practice medicine now, go to a conference

    If you want to know how we will practice medicine in the future, listen in the hallways and use FOAM