Research Article - Biomedical Research (2017) Volume 28, Issue 16
Accepted date: July 17, 2017
Objective: Although the action mechanism of intravenous lipid emulsion has not been fully elucidated yet, its use in liposoluble drugs intoxications. In this study, we examined the lipophilic features of causative agents and the success of the treatment ILE therapy in intoxication cases.
Methods: We reviewed 765 cases published in PubMed between 1966 and June, 2015. After applying exclusion criteria, totally 141 cases ingested single substance and received ILE therapy with 20% ILE solution were included in present study. Amount of lipid solutions given and the results were recorded. Success rate was statistically assessed according to log p values of the substances taken and the amount of lipid emulsion used.
Results: 141 patients were involved in this study; log p values were calculated for all drugs regardless of the success of ILE therapy. ILE therapy under the amount of 100 ml failed to achieve successful outcome. ALOGPS and ChemAxon log P values were higher in cases, which received ILE therapy ≤ 500 ml and showed successful results. It was found that log p value had no contribution to the treatment success in the group received ILE therapy>500 ml.
Conclusions: It was found that ILE therapy<500 ml was successful in drugs with higher lipophilicity while success rate was higher in ILE therapy>500 ml and that liposolubility had no significant contribution to treatment success.
Intravenous lipid emulsion, Lipophilicity, Poisoning.
In recent years, the lipid emulsions used for nutrition were introduced as an antidote in life-threatening drug intoxications. Weinberg et al. reported the Intravenous Lipid Emulsion (ILE) therapy given after severe bupivacaine intoxication to be effective in resuscitation in intact rats . Since then, it was shown that lipid emulsions are effective in the treatment of accidental or intentional drug intoxications in many case reports and animal studies [2-7].
Although many studies were carried out on ILE, its mechanism of action couldn’t be completely understood. However, various theories were proposed on the pharmacodynamics and pharmacokinetic aspects. First of all, a theory proposing pharmacological sink for liposoluble drugs, which is also termed as lipid sink theory, was considered [8,9]. In lipid sink theory, it is proposed that the lipid emulsion creates an expanded lipid phase, resulting in redistribution; thus, lipid emulsion leads the toxic drugs to pass into the plasma, where they then pass to the lipid phase [9,10]. In a rat study, it was found that ILE given after bupivacaine injection decreased the concentration of bupivacaine in heart, brain, lung, kidney and spleen via redistribution . In this theory, liposoluble drugs would better pass into lipid phase. Liposolubility of a drug is generally represented by logarithmic presentation (log10) of distribution between octanol and water (octanol-water partition coefficients; log p or log d (distribution-coefficient) . According to this statement, drugs with higher log p value would be more lipophilic. Second theory is the enhancement of cardiac energy support. Fatty acids are primary energy source in non-stressed, resting heart. In previous studies, it was shown that supplementation with fatty acids increased the performance in ischemic, hypo-dynamic heart [8,13]. Another action mechanism that was proposed is the direct cardiotonic effect. In a study on rats, it was demonstrated that lipid emulsions had positive inotropic and lusitropic effects, although underlying mechanism couldn't be clarified . Besides these mechanisms of action, it is also known that free fatty acid has some effects on ion channels . In cardiomyocytes, long-chain fatty acids contribute to positive inotropic effect by increasing the calcium level via the calcium channels [15,16]. In recent studies, it was reported that longchain fatty acids in lipid emulsions exerted cardioprotective effects via signaling pathway that regulates calcium homeostasis and opening of mitochondrial pores .
When considering above-mentioned mechanisms of action, it is anticipated to achieve successful outcome with ILE therapy in many types of intoxication. However, there are case reports indicating the failure of ILE therapy, including those demonstrating different outcomes in different drug classes or those demonstrating failure or success in same drug classes [2,18-20]. Such inconsistencies could be influenced by many factors such as amount of drug ingested, lipophilicity of drug and dose of ILE given. According to lipid sink theory, log p values of drugs can be helpful in predicting the clinical effectiveness of ILE therapy in lipophilic drug intoxications .
The aim of this systematic review is to investigate the relationship between the log p value of pharmacological agent leading to poisoning and the success rate of ILE therapy by reviewing case series, which ILE therapy was given for intoxication cases. By these results, we tested the theory that intoxication from more lipophilic drugs has more successful outcomes with ILE therapy; additionally, we investigated the relationship with the log p values of drugs and the success rate of ILE therapy.
We performed a literature search in PubMed between 1966 and June, 2015. The search was conducted by using Medical Subject Headings terms (MeSH): “((((((toxicology) OR poisoning) OR rescue) OR arrest) OR toxic)) and ((((lipid emulsion) OR fat emulsion) OR intralipid) OR intravenous lipid emulsion)”. In total, 765 publications were identified. Among them, the animal studies and those on multidrug ingestion were excluded. Finally, 323 publications involving single agent ingestion were included to the review.
Two researchers reviewed congress abstracts and articles separately. The congress abstracts, which were subsequently published as an article, were considered as a single case report. The results were assessed, and 1 case of ingestion of a drug termed "Bonzai" was excluded since there was no data about the drug composition. Again, another case was excluded due to ingestion of a drug classified as unknown TCA. In addition, 15 cases were excluded since amount of lipid given was unspecified. Overall, 141 cases (75 publications) were included to the review. There was concurrent alcohol consumption in 4 of 141 cases with single agent ingestion. These cases were included to the review as ethanol is a water-soluble substance (log p=-0.4 (ALOGPS)).
The concentration of the lipid given in all patients was 20% lipid emulsion. The amount of lipid given was stratified as<100 ml, 100-500 ml and >500 ml. Success was defined as complication-free result. Cases with complication or nonsurvivors were considered as failure.
Log p value for each substance was searched from “www.drugbank.ca” website. Log p values were recorded as Experimental, ALOGPS and ChemAxon data. Analyses were performed for all three parameters.
All statistical analyses were performed using SPSS 17.0 (IBM, New York, USA) and MS Office Excel. The continuous variables were expressed as mean ± SD, whereas the categorical variables were expressed as n (%). The difference between the mean values of the continuous variables was calculated using the Mann Whitney U-test. The correlations among continuous variables were calculated using Spearman’s Rho correlation. Logistic regression analysis was performed in order to determine the independent effect of the log p values. P ≤ 0.05 was considered statistically significant.
Among 141 cases, 15 drug classes and 31 drugs were identified. Table 1 presents log p values and number patients for 31 drugs identified. In Table 2, log p values and number of patients for 15 drug classes are presented. Table 3 summarizes the amount of lipid given and success rates.
|Medication or chemical agent||n||Log P|
|Ca2+ channel blocker||Diltiazem||2||2.800||3.090||2.730|
|Alpha and beta blocker||Carvedilol||5||4.190||3.050||3.420|
|Central muscular blocker||Baclofenone||2||1.300||-0.820||-0.780|
|Na+ channel blocker||Propafenone||1||3.200||3.100||3.540|
Table 1: Log P values of medication or toxic agents of survey.
|n||Log P||Log P||Log P|
|Experimental (mean ± SD)*||ALOGPS (mean ± SD)*||ChemAxon (mean ± SD)*|
|Beta blocker||8||3.080 ± 0.741||2.723 ± 0.569||2.375 ± 0.380|
|Ca2+ channel blocker||14||3.302 ± 0.426||3.562 ± 1.465||3.168 ± 1.651|
|Na channel blocker||4||3.635 ± 0.290||3.010 ± 0.060||3.278 ± 0.175|
|Local anesthetic||41||3.244 ± 0.329||3.155 ± 0.314||4.345 ± 0.355|
|TCA||18||4.727 ± 0.305||4.901 ± 0.374||4.623 ± 0.354|
|Antipsychotic||10||2.790 ± 0.626||3.143 ± 0.344||3.011 ± 0.332|
|Antiepileptic||3||2.483 ± 0.029||1.947 ± 0.133||2.210 ± 0.485|
|Herbicide||26||-3.731 ± 1.395||0.04||-3.1|
|Central muscle relaxator||2||1.3||-0.82||-0.78|
|Alpha and beta blocker||5||4.19||3.05||3.42|
Table 2: Log P values of agent types.
|≤ 100 ml||100-500 ml||>500 ml||Success|
|Beta blocker||-||6 (75%)||2 (25%)||7 (87.5%)|
|Ca2+ channel blocker||1 (7.1%)||9 (64.3%)||4 (28.6%)||10 (71.4%)|
|Local anesthetic||12 (29.3%)||22 (53.7%)||7 (17.1%)||40 (97.7%)|
|Narcotic agents||1 (33%)||1 (33%)||1 (33%)||2 (66.7%)|
|Alpha and beta blocker||-||5 (100%)||-||5 (100%)|
|TCA||-||10 (55.6%)||8 (44.4%)||18 (100%)|
|Antipsychotic||-||7 (70%)||3 (30%)||8 (80%)|
|Antiepileptic||-||1 (33%)||2 (67%)||2 (66.7%)|
|Herbicide||-||3 (11.5%)||23 (88.5%)||26 (100%)|
|DNRI||-||2 (67%)||1 (33%)||2 (66.7%)|
|Central muscle relaxator||-||2 (100%)||-||1 (50%)|
|Na channel blocker||-||1 (25%)||3 (75%)||2 (50%)|
|Insecticide||-||2 (100%)||-||1 (50%)|
|Antimalarial||-||1 (100%)||-||0 (0%)|
|Antidiabetic||-||1 (100%)||-||0 (0%)|
Table 3: Given amount of lipid emulsion therapies and success rates.
The cases were stratified according to the treatment outcome as successful and failure regardless of amount of lipid given. Mean log p values (Experimental, ALOGPS, ChemAxon) were calculated for drugs, in which the lipid emulsion therapy was successful. In addition, mean log p values were calculated for all drugs regardless of treatment outcome with ILE therapy. Moreover, mean log p values were calculated according to the reasons for failure. When groups were compared, it was found that mean ChemAxon and Experimental log p values were higher in the group, in which ILE therapy failed (Table 4).
|Log P||Log P||Log P|
|Experimental (mean ± SD)*||ALOGPS (mean ± SD)*||ChemAxon (mean ± SD)*|
|Successful||1.970 ± 3.108||2.663 ± 1.621||2.364 ± 2.975|
|Unsuccessful||2.676 ± 1.131||2.495 ± 1.814||2.513 ± 1.688|
|Exitus||2.520 ± 1.372||2.259 ± 1.661||2.413 ± 1.489|
|Hypoxic ischemic encephalopathy||3.395 ± 0.559||3.725 ± 2.128||3.340 ± 2.404|
|Delirium||2.300 ± 0.866||1.680 ± 2.165||1.613 ± 2.073|
Table 4: Log P values according to success rate.
Amount of ILE therapy given was ≤ 100 ml in 14 (9.9%), 100-500 ml in 73 (51.8%) and >500 ml in 54 (38.3%) of the cases. The amount of lipid emulsion was <100 ml in only one case, in which treatment outcome was found to be successful; thus, we stratified the amount of the lipid emulsion as ≤ 500 ml and >500 ml. The success rate was 85.1% in patients received ILE therapy ≤ 500 ml, whereas the same rate was 92.6% in patients that received ILE therapy>500 ml. There was no significant difference between groups received ILE therapy ≤ 500 ml or >500 ml (p=0.142).
When amount of lipid emulsion given was below 500 ml (1-500 ml), the log p value, especially the ALOGPS and ChemAxon data, becomes more important. In cases that received ILE therapy ≤ 500 ml, the ALOGPS and ChemAxon log p values were higher in the group with successful outcome than those observed in cases, in which ILE therapy failed (p values are 0.043 and 0.008). In addition, Experimental log p value was higher, indicating a trend towards statistical significance (p=0.071). Thus, we can argue that log p value has significant effect on treatment success when amount of lipid emulsion is equal or below 500 ml. But, there is no significant effect of treatment outcome when amount of lipid emulsion is higher than 500 ml (Table 5). The summary of the 141 human case reports treated with ILE is given in Table 6.
|≤ 500 ml lipid emulsion||Experimental Log P||3.107 ± 1.644||2.577 ± 1.269||0.071|
|ALOGPS Log P||3.087 ± 1.074||2.313 ± 1.917||0.043|
|ChemAxon Log P||3.483 ± 1.745||2.285 ± 1.765||0.008|
|>500 ml lipid emulsion||Experimental Log P||0.310 ± 3.917||2.948 ± 0.681||0.577|
|ALOGPS Log P||2.040 ± 2.056||3.043 ± 1.561||0.302|
|ChemAxon Log P||0.741 ± 3.600||3.198 ± 1.410||0.286|
Table 5: Log P values according to success and given amount of lipid emulsion.
|Drugs||References [26-97]||References [26-97]|
|ILE ≤ 500 ml||ILE>500 ml|
Besides the use for nutritional purposes, intravenous lipid emulsions were introduced into treatment of serious intoxications. There is limited number of studies on the use of lipid emulsion therapy in poisoned patients as antidote, and majority of available studies are animal studies and case reports. Although it was shown that lipid emulsions are effective in many intoxication cases caused from liposoluble drugs, they were also shown to be ineffective in some drug intoxications with high liposolubility . In addition, there are also case reports indicating that lipid emulsions are successful in drug intoxications with low liposolubility . In this review, we focused on the intoxication caused from a single agent. Because, in mixed drug intoxications, it is difficult to identify which drug is responsible for clinical picture and to confirm which drug was affected by therapy given. Given this, we performed our analysis in patients with single agent ingestion according to amount of lipid given and log p values, which indicate liposolubility of drugs leading to the intoxication. According to our study until 500 ml ILE treatment liposolubulity is important and ILE therapy is more effective in highly liposoluble drugs. Liposolubility does not affect the success rate.
In a study on healthy volunteers, Litonius et al. suggested that there was no significant decrease in plasma-free bupivacaine levels before and after intravenous lipid administration and that no lipid sink effect was observed . Litonius et al. reported the lack of lipid sink phenomenon despite lipophilic structure of bupivacaine (ALOGPS log p: 3.31) but positive response to ILE therapy at toxic levels, and also we reported that over 500 ml of ILE therapy the liposolubility has no effect. When our systematic review and the study by Litonius et al. were considered together, we can argue that there may be some additional action mechanisms besides the lipid sink phenomenon.
It is thought that lipid sink effect occurs via uptake of liposoluble drug molecule from aqueous compartment to lipid compartment . In this systematic review, we compared success rate according to amount of lipid given. Successful outcome was observed in only one out of 14 patients received ILE therapy<100 ml. Based on this finding, it was concluded that ILE therapy given less than 100 ml was ineffective. Among the patients that received ILE therapy>500 ml, no significant difference was found in mean log p values of drug between cases with successful outcome and those with failure.
Among the patients that received ILE therapy ≤ 500, ChemAxon log p and ALOGPS log p values were significantly higher in the group successfully treated when compared to those with treatment failure. Experimental log p value was significantly higher. Based on these findings, ILE therapy ≤ 500 ml was more successful in drugs with high log p value, while log p value had no contribution to clinical outcomes in cases received ILE therapy>500 ml.
According to lipid sink theory, it is anticipated that ILE therapy should be associated to more effective outcomes in lipophilic group ≤ 500 ml ILE therapy. It could be thought that additional action mechanisms other than lipid sink phenomenon are more active in ILE therapy. When amount of ILE therapy given was assessed, it was found that ILE therapy<100 ml failed to achieve successful outcome, and that there was no association between success rate and lipophilicity of drug in cases receiving ILE therapy>500 ml.
Many cases were excluded due to the fact that the amount of ILE was not indicated. On the other hand, using the one database for scanning the literature is limitation of this study.
Authors warrant that no conflicts of interest.