Lipid Emulsion Improves Glasgow Coma Scale and Decreases Blood

Home , Lipid emulsion

European Review for Medical and Pharmacological Sciences 2012; 16(Suppl 1): 38-42 Lipid emulsion improves Glasgow Coma Scale and decreases blood glucose level in the setting of acute non-local anesthetic drug poisoning – a randomized controlled trial

F. TAFTACHI, H. SANAEI-ZADEH, B. SEPEHRIAN, N. ZAMANI

Department of Forensic Medicine and Toxicology, Tehran University of Medical Sciences, Hazrat Rasoul Akram Hospital, Tehran (Iran)

Abstract. – Background: To date, no Conclusions: In the setting of non-local anes- study has been performed to evaluate the anti- thetic drug overdose, intravenous intralipid infu- dotal effect of intravenous lipid emulsion on the sion can increase GCS and interestingly, de- poisoned patients’ level of consciousness and crease the blood glucose. routine metabolic profile tests in non-local anesthetic drug overdose. Key Words: Objectives: Our aim was to evaluate the ef- Intralipid, Non-local anesthetic overdose, Glascow fect of intravenous intralipid administration as Coma Scale, Blood glucose, Antidote. an antidote on the poisoned patients’ Glasgow Coma Scale (GCS), hemodynamic parameters, arterial blood gas analysis, and routine metabolic profile tests (i.e. urea, glucose, sodium, and Introduction potassium) in the setting of non-local anesthetic drug overdose. Intravenous lipid emulsion is a new method Material and Methods: In this randomized controlled trial, a total of 30 patients with non-lo- for the treatment of local anesthetic systemic tox- 1,2 cal anesthetic drug intoxication were enrolled icity . Of course, this treatment is not limited to and randomly assigned into case (n=15) and the local anesthetic toxicities. Because of recent control (n=15) groups. In the case group, all pa- published human case reports of successful re- tients received 10 cc/kg intralipid 10% infusion. suscitation3-6, there has been increasing interest The patients in the control group just received in the potential benefits of this type of treatment the supportive care. Patients' demographic and in cardiac arrest5,6, and hemodynamic instabili- clinical characteristics and results of their labo- 3,7-11 ratory tests were evaluated at presentation and 6 ty attributable to lipophilic, non-local anes- hours after that. thetic drugs. Also, it has been suggested that in- Results: Mean age was 23 ± 5 and 28 ± 11 tralipid emulsion therapy can be used as an anti- years in cases and controls, respectively. There dote for resuscitation of lipophilic non-local were no significant statistical differences be- anesthetic toxicity3,12,13. To our best knowledge, tween these two groups regarding age, gender, no randomized controlled trial (RCT) has been elapsed time between intubation and extubation, and need for intubation and/or mechanical venti- performed to evaluate the antidotal effect on the lation (p = 0.70 and p = 1.00, respectively). Also, poisoned patients’ level of consciousness and systolic blood pressure, pulse rate, mean rate routine metabolic profile tests in non-local anes- pressure product, respiratory rate, results of ar- thetic overdoses. Therefore, our aim was to eval- terial blood gas analyses, serum sodium, potas- uate the effect of intravenous intralipid adminis- sium, urea, and creatinine on presentation and tration as an antidote on the poisoned patients’ six hours later were not statistically significantly different between the two study groups. Howev- Glasgow Coma Scale (GCS), hemodynamic pa- er, a significant difference was found between rameters, arterial blood gas (ABG) analysis, and the two groups in terms of GCS difference (p = routine metabolic profile tests (i.e. urea, glucose, 0.048) and blood glucose six hours after presen- sodium, and potassium) in the setting of non-lo- tation (p = 0.04). cal anesthetic drug overdose.

38 Corresponding Author: Hossein Sanaei-Zadeh, MD; e-mail: [email protected] Setting of acute non-local anesthetic drug poisoning – a randomized controlled trial

Material and Methods tralipid 10% infusion in addition to the routine supportive management (i.e. cardiac monitoring, This is a randomized controlled trial (RCT: su- intubation, mechanical ventilation, volume ex- periority trial with parallel design). The mini- pansion, use of inotropic/or vasopressor drug mum required sample size per group was calcu- support, administration of sodium bicarbonate, lated according to the related formula for RCT and etc. when indicated) within the first six hours studies with alpha of 0.05, power of 80%, of hospital presentation. The patients in the con- 14 π1=0.1, and π2=0.6 at two-sided 5% . Therefore, trol group just received the supportive manage- cases and controls with documented diagnosis of ment. Secondary abovementioned vital signs, intoxication with non-local anesthetic drugs were GCS, and laboratory tests were evaluated six enrolled from all patients with suspicion of poi- hours after presentation in both groups. GCS dif- soning transferred to the toxicological Emer- ference (differences between GCS at hospital gency Department of Shohada Yaft Abad Hospi- presentation and at 6-hour follow-up) was also tal (Tehran, Iran) between October 2010 and recorded. During the study, the patients were ex- March 2011. The patients were entered into the cluded from both groups if we retrospectively study if they had GCS ≤ 9, had not received any found that their urine and/or serum drug screen type of antidote, had not responded to the anti- had been reported to be negative and our history dote therapy (e.g. naloxone), or had some kind of was wrong, even when the intralipid had previ- contraindication for receiving an antidote (e.g. ously been administered. Other patients with the flumazenil). Documentation of poisoning was same inclusion criteria were randomly included based on a positive history of non-local anesthet- into the study instead of those excluding from it. ic drug ingestion and a positive urine or serum drug screen test. The positive history of ingestion Statistical Analysis was defined as giving the name of the consumed Statistical analysis was done using SPSS soft- medication by the patient him/herself before de- ware version 17, (SPSS Inc., Chicago, IL, USA) crease in the level of consciousness (to his/her and application of Kolmogorov-Smirnov, Mann- relatives or the personnel of Emergency Medical Whitney U-test, Pearson’s chi square or Fisher’s Service) or after complete recovery from coma exact test, and Student’s t-test. p values less than and identifying and retaining any evidence of 0.05 were considered to be statistically signifi- medications (such as its bottle or packet) in the cant. Our study was approved by the Regional bedroom or household, workplace, etc. All pa- Ethics Committee. Written informed consents tients with documented head trauma were ex- were taken from all patients’ relatives entering cluded. Information including age, gender, med- into the study for administration of intralipid. ication ingested (documented by urine and/or serum drug screen test, pill ID, or patient’s history), manner of poisoning, need for intubation and mechanical ventilation, and the time elapsed be- Results tween intubation and extubation were recorded in the standardized abstraction forms. Intubation A total of 15 cases (9 men and 6 women) and had been performed due to hypoventilation, loss 15 controls (9 men and 6 women) met the inclu- of protective airway reflexes, seizures, and he- sion criteria and were entered into the study. modynamically significant dysrhythmia. Those Mean age was 23±5 and 28±11 years in the cases with inadequate oxygenation and ventilation de- and controls, respectively. A combination of spite supplemental oxygen had mechanically medications including benzodiazepines, tricyclic been ventilated. antidepressants (TCAs), anticonvulsants, anti- In both groups, initial Emergency Department cholinergics, antihistamines, muscle relaxants, GCS, systolic blood pressure, pulse rate, mean selective serotonin reuptake inhibitors, antipsy- rate pressure product (RPP, systolic pressure x chotics, acetaminophen, nonsteroidal anti-in- heart rate), and respiratory rate were recorded in flammatory drugs, salicylates, and opioids were addition to the patients’ demographic characteris- ingested in both case and control groups. Manner tics. Samples were sent for ABG analysis and of poisoning was suicidal in all patients. Nine check of serum sodium, potassium, urea, creati- and six patients in the case and control groups nine, and blood glucose on presentation. In the were intubated. In each group, four were ventilat- case group, all patients received 10 cc/kg in- ed. Information on the patients’ initial Emer-

39 F. Taftachi, H. Sanaei-Zadeh, B. Sepehrian, N. Zamani gency Department GCS, systolic blood pressure, mise and without a clear history of toxicity, an pulse rate, mean rate pressure product, and respi- unidentified drug overdose is possible. Even in ratory rate, results of ABG analyses, serum sodi- those referring with a positive history of drug um, potassium, urea, creatinine, and blood glu- overdose, the drug ingested may be unknown. In cose on presentation and six hours later is these cases, the lipid solubility characteristic of brought in Tables I and II. the drug is, therefore, unknown, as well. Howev- There were no significant statistical differences er, in all patients referring with decreased level of between these two groups regarding age, gender, consciousness, the culpable drug ingested has elapsed time between intubation and extubation, possibly high lipid solubility characteristics ex- and need for intubation and/or mechanical ventila- cept if other confounding factors such as asphyx- tion (p = 0.70 and p = 1.00, respectively; Fisher’s ia have developed. Thus, if we are due to use in- Exact test). Also, systolic blood pressure, pulse tralipid as an antidote, we have to evaluate its ef- rate, mean rate pressure product, respiratory rate, fect on the cardiovascular and neurological para- results of ABG analyses, serum sodium, potassi- meters in a setting most similar to our study. um, urea, and creatinine on presentation and six In the previously performed studies, it has hours later were not statistically significantly dif- been shown that instable hemodynamic of the pa- ferent between the two study groups. But, a signif- tients with overdose of non-local anesthetic icant difference was found between the two drugs with lipophilic properties such as beta groups in terms of GCS difference (p = 0.048, blockers, calcium channel blockers, TCAs, and MWU test) and blood glucose six hours after pre- some psychotropic agents respond to the admin- sentation (p = 0.04, MWU test). istration of intralipid15. Our study more or less covers the same drug groups. Of course, it should be mentioned that none of our patients needed advanced resuscitation (for cardiovascular col- Discussion lapse, refractory hypotension, or cardiac arrest) on presentation (Table II). In toxicity due to local anesthetic drugs, intra- This investigation shows that in poisoning set- venous intralipid infusion is a proven resuscita- ting, intravenous intralipid infusion can increase tive method in the treatment of cases refractory GCS (of the clinical characteristics) and decrease to conventional modes of resuscitation1,2. Howev- the blood glucose (of the laboratory characteris- er, in the patients presenting to the Emergency tics evaluated). The effect of intralipid in increas- Department with cardiac and neurologic compro- ing the level of consciousness has previously

Table I. The results of poisoned patients' arterial blood gas (ABG) analysis, and routine metabolic profile tests.

p value Case Control (applied statistical test)

pH 7.32 ± 0.09 7.24 ± 0.13 NS*(MWU)† 7.38 ± 0.03 7.33 ± 0.06

pCO2 (mmHg) 38.7 ± 6.5 42.6 ± 10.5 NS (MWU) 39.6 ± 6.5 39.7 ± 8.6 – HCO3 (mmol/l) 20.7 ± 2.8 20.1 ± 5.3 NS (MWU) 21.5 ± 3.6 22.0 ± 4.4 Blood glucose (mg/dL) 115 ± 26 118 ± 40 0.04 (MWU) 98 ± 10 110 ± 19 Potassium (mEq/l) 3.9 ± 0.3 3.9 ± 0.5 NS (MWU) 3.9 ± 0.4 3.8 ± 0.4 Sodium (mEq/l) 139 ± 3 140 ± 3 NS (MWU) 138 ± 3 140 ± 3 Urea (mg/dL) 29 ± 5 33 ± 9 NS (MWU) 26 ± 4 34 ± 7 Creatinine (mg/dL) 0.8 ± 0.2 1.0 ± 0.3 NS (MWU) 0.9 ± 0.2 0.9 ± 0.2

Data are presented as mean value (± standard deviation; SD); †MWU: Mann-Whitney U test; *NS: Not significant; Comparison of the mean ± SD of the patients’ data at presentation (first row for each parameter) and six hours after presentation (second row).

40 Setting of acute non-local anesthetic drug poisoning – a randomized controlled trial

Table II. The results of poisoned patients’ demographic characteristics, Glasgow Coma Scale (GCS), hemodynamic parameters, and elapsed time between intubation and extubation.

p value Case Control (applied statistical test)

Age (years) 23 ± 5 28 ± 11 NS*(MWU)† Male/female ratio 9/6 9/6 NS (P Chi2)¶ GCS difference 3 ± 1 2 ± 2 0.048 (MWU) Systolic blood pressure (mmHg) 101 ± 16 110 ± 17 NS(MWU) 112 ± 9 117 ± 23 Pulse rate (bpm) 91 ± 25 97 ± 23 NS (MWU) 91 ± 16 98 ± 15 Mean rate pressure product (RPP) 9189 ± 3184 10529 ± 2726 NS (MWU) 10236 ± 2384 11327 ± 2326 Respiratory rate (/min) 17 ± 5 19 ± 7 NS (MWU) 18 ± 2 16 ± 3 Elapsed time between intubation 28 ± 20 37 ± 20 NS (MWU) and extubation (hours)

Data are presented as mean value (± standard deviation; SD); †MWU: Mann-Whitney U test; *NS: Not significant; ¶P Chi2: Pearson’s chi square test; Comparison of the mean ± SD of the patients' clinical characteristics at presentation (first row for each parameter) and six hours after presentation (second row). been reported in the literature4,16. It has been cal anesthetic poisonings15. We used a continuous shown that intravenous infusion of lipid (an acute infusion of the intralipid 10% for the treatment of increase in plasma free fatty acids – FFA) in nor- our patients while according to the recommenda- mal volunteers causes no change in the plasma tions of American Society of Regional Anaesthe- glucose concentration as well as the rates of en- sia, intralipid 20% should have been administered1. dogenous glucose production (EGP). However, Maybe, if we acted according to the above men- plasma C-peptide concentrations rise suggesting tioned recommendations, the patients’ GCS would the stimulation of insulin secretion. Additionally, increase more or the intubated patients would be when somatostatin prevents the increase in in- extubated faster. Also, this is the probable cause of sulin, the rise in plasma FFA is accompanied by lack of a difference between the GCS increment in an acute increase in EGP and plasma glucose25. It the case and control groups in our investigation. has also been shown that under such experimen- However, the mean time elapsed between intuba- tal conditions, a rise in gluconeogenesis is seen tion and extubation in our study (28.11±19.86 with an acute increase in plasma FFA26. In con- hours) is similar to the period mentioned in a case trast to these investigations, our study shows that report (18 hours) by Weinberg et al4 who had ad- by administration of intralipid, the level of blood ministered intralipid 20% to their patient. glucose decreases. This cannot be justified by the previously mentioned mechanisms of lipid effect on the blood glucose. The probable mechanisms for the action of in- Conclusions tralipid emulsion infusion in local anesthetic drug toxicity include lipid sink phenomenon17-19, in- The present study shows that in the setting of creasing intracellular fatty acid content (overcom- non-local anesthetic drug overdose, intravenous ing the reduction in the ATP production) in car- intralipid infusion can increase GCS and interest- diomyocytes20-22, and increase in the intramyocyte ingly, decrease the blood glucose. calcium level (leading to a direct positive inotropic effect)23. Presumably, intralipid emulsion exerts the same lipid sink effect on lipophilic, non-local anesthetic drug toxicity24. References While protocols exist for administration of intravenous intralipid in the setting of local anesthetic 1) WEINBERG GL. Treatment of local anesthetic sys- systemic toxicity1, no optimal regimen has been es- temic toxicity (LAST). Reg Anesth Pain Med tablished to date for the treatment of acute non-lo- 2010; 35: 188-193.

41 F. Taftachi, H. Sanaei-Zadeh, B. Sepehrian, N. Zamani

2) PICARD J, WARD SC, ZUMPE R, MEEK T, B ARLOW J, HAR- 16) FINN SD, UNCLES DR, WILLERS J, SABLE N. Early ROP-GRIFFITHS W. Guidelines and the adoption of treatment of a quetiapine and sertraline over- ‘lipid rescue’ therapy for local anaesthetic toxicity. dose with Intralipid. Anaesthesia 2009; 64: 191- Anaesthesia 2009; 64: 122-125. 194. 3) SIRIANNI AJ, OSTERHOUDT KC, CALELLO DP, MULLER AA, 17) WEINBERG GL, VADEBONCOUER T, R AMARAJU GA, GAR- WATERHOUSE MR, GOODKIN MB, WEINBERG GL, HEN- CIA-AMARO MF, CWIK MJ. Pretreatment or resuscita- RETIG FM. Use of lipid emulsion in the resuscitation tion with a lipid infusion shifts the dose response of a patient with prolonged cardiovascular col- to bupivacaine-induced asystole in rats. Anesthe- lapse after overdose of bupropion and lamotrig- siology 1998; 88: 1071-1075. ine. Ann Emerg Med 2008; 51: 412-415. 18) MAZOIT JX, LE GUEN R, BELOEIL H, BENHAMOU D. 4) WEINBERG G, DI GREGORIO G, HILLER D, HEWETT A, Binding of long-lasting local anesthetics to lipid SIRIANNI A. Reversal of haloperidol-induced cardiac emulsions. Anesthesiology 2009; 110: 380-386. arrest by using lipid emulsion. Ann Intern Med 19) WEINBERG GL, RIPPER R, MURPHY P, E DELMAN LB, 2009; 150: 737-738. HOFFMAN W, S TRICHARTZ G, FEINSTEIN DL. Lipid in- 5) PICARD J, HARROP-GRIFFITHS W. Lipid emulsion to fusion accelerates removal of bupivacaine and treat drug overdose: past, present and future. recovery from bupivacaine toxicity in the isolat- Anaesthesia 2009; 64: 119-121. ed rat heart. Reg Anesth Pain Med 2006, 31: 6) JAMATY C, BAILEY B, LAROCQUE A, NOTEBAERT E, SANOGO 296-303. K, CHAUNY JM. Lipid emulsions in the treatment of 20) COLLINS-NAKAI RL, NOSEWORTHY D, LOPASCHUK GD. acute poisoning: a systematic review of human and Epinephrine increases ATP production in hearts animal studies. Clin Toxicol (Phila) 2010; 48: 1-27. by preferentially increasing glucose metabolism. 7) YOUNG AC, VELEZ LI, KLEINSCHMIDT KC. Intravenous Am J Physiol 1994; 267: H1862-1871. fat emulsion therapy for intentional sustained-re- 21) VAN DE VELDE M, WOUTERS PF, ROLF N, VAN AKEN H, lease verapamil overdose. Resuscitation 2009; FLAMENG W, V ANDERMEERSCH E. Long-chain triglyc- 80: 591-593. erides improve recovery from myocardial stunning in conscious dogs. Cardiovasc Res 1996; 32: 8) CHU J, MEDLEJ K, BANIA T, P EREZ E, MOURAVEV R. The effect of intravenous fat emulsions in nifedipine 1008-1015. toxicity. Acad Emerg Med 2009; 16: S226. 22) ELEDJAM JJ, DE LA COUSSAYE JE, BRUGADA J, BASSOUL AGNOL ABREGAT ASSE ASSINE In 9) LIPID RESCUE. 2010 [http://www.lipidrescue.org]. B, G JP, F JR, M C, S A. vitro study on mechanisms of bupivacaine in- 10) ENGELS PT, DAVIDOW JS. Intravenous fat emulsion to duced depression of myocardial contractility. reverse haemodynamic instability from intentional Anesth Analg 1989; 69: 732-735. amitriptyline overdose. Resuscitation 2010; 81: 1037-1039. 23) HUANG JM, XIAN H, BACANER M. Long-chain fatty acids activate calcium channels in ventricular my- 11) HAN SK, JEONG J, YEOM S, RYU J, PARK S. Use of a ocytes. Proc Natl Acad Sci USA 1992; 89: 6452- lipid emulsion in a patient with refractory hypoten- 6456. sion caused by glyphosate-surfactant herbicide. Clin Toxicol 2010; 48: 566-568. 24) LESKIW U, WEINBERG GL. Lipid resuscitation for local anesthetic toxicity: is it really life saving? Curr 12) LIPID REGISTRY. 2010 [http://www.lipidregistry.org]. Opin Anaesthesiol 2009; 22: 667-671. 13) TURNER-LAWRENCE DE, KERNS II W. Intravenous fat 25) BODEN G, JADALI F. Effects of lipid on basal carbo- emulsion: a potential novel antidote. J Med Toxi- hydrate metabolism in normal men. Diabetes col 2008; 4: 109-114. 1991; 40: 686-692. 14) CHAN YH. Randomised controlled trials (RCTs)-- 26) RODEN M, STINGL H, CHANDRAMOULI V, S CHUMANN sample size: the magic number? Singapore Med WC, HOFER A, LANDAU BR, NOWOTNY P, W ALDHAUSL J 2003; 44: 172-174. W, S HULMAN GI. Effects of free fatty acid elevation 15) ROTHSCHILD L, BERN S, OSWALD S, WEINBERG G. Intra- on postabsorptive endogenous glucose produc- venous lipid emulsion in clinical toxicology. Scand tion and gluconeogenesis in humans. Diabetes J Trauma Resusc Emerg Med 2010; 18: 51. 2000; 49: 701-707.