Local anaesthetic toxicity

1. Historical background 

Long before the first clinical use of local anaesthetics, the native South Americans enjoyed the leaves of Erythroxylum Coca for its energizing and mood enhancing effects. It was not until 1860 in Germany, that the chemist Niemann isolated cocaine from erythroxylin extract. At the time he noticed on tasting, that it caused the tip of his tongue to go numb. Its mood enhancing effects were soon noted and before long it was a constituent of many elixirs and beverages, the most famous of which was Coca-Cola.

In 1884, Carl Koller discovered its potential for topical anaesthesia by applying a solution of the crystals to the surface of the eye. In the following years it was successfully used for infiltration, peripheral nerve blockade, caudal and spinal anaesthesia. However toxic effects soon became apparent as convulsions and arrythmias in patients, and addiction amongst medical staff were reported.1 The search for an alternative agent resulted in the discovery of other esters, and later amides. All were less toxic then cocaine but had differing cardiac and neurotoxic effects. Of particular note was the development of the amide bupivacaine that has a longer duration of action than lignocaine but is less toxic then cocaine. In the 1960s bupivacaine rapidly became globally popular especially for epidural pain relief in labour. In 1979 Albright2 reported six cases of cardiovascular collapse after presumed intravascular injection of bupivacaine or etidocaine (related to lignocaine). One caudal injection resulted in the death of a pregnant woman. In 1984, further reports of the toxic effects of bupivacaine were published by Plumer3 who collected 35 incidents involving intravenous injection of bupivacaine in pregnant women, of whom only seven survived. Later, in 1991, Chadwick et al.4 presented data from the North American closed claims study, which reported between 1975 and 1985. He listed 12 maternal deaths after regional anaesthesia, and 19 incidences of convulsions, of which 18 occurred in patients with epidural anaesthesia, and 17 were thought to be due to local anaesthetic toxicity. Outcomes were poor with 83% of convulsions resulting in neurological injury or death to the mother, fetus, or both.

In the UK, the complications seen with epidural anaesthesia in North America did not occur. This may have been due to the common UK practice of fractionating the administered dose. However in 1982, Heath5 reported five deaths (including two children) during intravenous regional anaesthesia (IVRA) with bupivacaine by non-anaesthetists.

Following these reports, the use of bupivacaine for IVRA was prohibited and bupivacaine 0.75% solution was barred from obstetric use.

2. Modern local anaesthetic toxicity 

More then a century after its introduction, cocaine deaths still occur, but now following social use rather than clinical. Non-anaesthetists still contribute to local anaesthetic related deaths through practices as diverse as local infiltration during liposuction, topical anaesthesia in healthy volunteers having bronchoscopy for research purposes, and advising a customer prior to a hair removal treatment to apply local anaesthetic gel to her legs and then to wrap them in cling film. Interestingly, in anaesthetic practice toxicity is a rare occurrence despite recommended dose maxima routinely being exceeded, for example, in the extension of epidural pain relief for caesarean section. Catheter techniques enable fractionation of the dose, allowing intermittent assessment for signs of toxicity.

The modern face of local anaesthetic toxicity in anaesthetic practice is the high dose single bolus injection around a neurovascular bundle to provide peripheral nerve blockade. This has the risk of inadvertent intravascular injection/absorption causing toxicity.

Another important cause is illustrated by three high profile deaths following intravenous administration of local anaesthetic intended for epidural use. This has prompted a reconsideration of local anaesthetic toxicity and its management.

3. Mechanism of local anaesthetic toxicity6 

Local anaesthetics produce their effect by blocking Na+ channels in axonal lipid membranes. Unionized lipid soluble local anaesthetic molecules pass through the membrane, where, within the more acidic cell cytoplasm, the molecule dissociates into its ionized base form which then binds to the Na+ channel, locking it in the inactive state. As a result, cell ion fluxes are inactivated, and the cell membrane can no longer depolarise, halting the propagation/generation of any further action potentials. This interrupts the transmission of afferent and efferent impulses and so provides analgesia and anaesthesia. Ion channels are not exclusive to axons so local anaesthetics can produce cardiac effects if high systemic concentrations are achieved.

To appreciate how high systemic concentrations of local anaesthetic occur, some understanding of pharmacokinetics is needed.

3.1. Absorption 

Absorption of local anaesthetic into the systemic circulation is influenced by the site of injection, characteristics of the agent and the presence of added vasoconstrictors.

3.1.1. Site of injection 

Rapid uptake of local anaesthetic by the circulation can be achieved if large doses are injected into well perfused sites. Even doses within recommended maxima can result in toxic symptoms if given as a single bolus in vascular sites such as the intercostal space Fig. 1.

View Large Image Download to PowerPoint

Fig. 1 Relative absorption of local anaesthetics from different sites.

3.1.2. Agent characteristics 

The pKa of the agent will determine what fraction is unionized lipid soluble moiety, and so available to cross the phospholipid membrane. As local anaesthetic drugs are basic, at normal pH an agent with a lower pKa will have a higher fraction of unionized moiety available for systemic absorption. The addition of bicarbonate will raise the pH, and increase the proportion of the unionized moiety, assisting systemic absorption. Lipid solubility also improves uptake from the injection site.

3.1.3. Added vasoconstrictor 

The addition of a vasoconstrictor is intended to reduce local perfusion, and thereby reduce systemic absorption. For lidocaine this permits an increase in the recommended maximum dose.

4. Clinical features of local anaesthetic toxicity 

The clinical manifestations of toxicity are generally neurological, cardiac or both in origin. How an individual presents with toxicity is variable as they may demonstrate either cardiac and neurological symptoms together, or symptoms solely from one system. A typical clinical course is shown in Fig. 2.

View Large Image Download to PowerPoint

Fig. 2 Clinical course of local anaesthetic toxicity.

5. Treating local anaesthetic toxicity 

This should be supportive, treating convulsions with anti-epileptics, and managing cardiac arrhythmias according to established guidelines. However, it is well recognized that the cardiac complications are resistant to treatment and require prolonged cardiopulmonary resuscitation (CPR) reputedly with limited success. Cardiopulmonary bypass was possibly the only effective method of treatment. It is important to note however, that the reputation for poor outcome in pregnant women may in part be due to the poor survival from Plumer's series3 who were resuscitated before the importance of caval compression was understood.

The last 10 years has seen the emergence of a new treatment for local anaesthetic toxicity. After a patient deficient in carnitine, essential for the intracellular transport of fatty acids, the main myocardial energy substrate, demonstrated an increased sensitivity to bupivacaine-induced cardiac dysrrhythmias, investigators considered whether intracellular lipid affected sensitivity to local anaesthetic toxicity.7 It had previously been established that ischaemic myocardial cells accumulated intracellular fatty acids. It was therefore proposed that an increased level of lipid rendered ischaemic cells more susceptible to arrhythmias. By infusing lipid it was hypothesized that an increased sensitivity to bupivacaine toxicity would be observed. In fact the opposite was shown. Infusing lipid into rats rendered them more resistant to bupivacaine-induced asystole. This effect was confirmed in anaesthetised rats which were pre-treated and resuscitated from bupivacaine-induced asystole with lipid infusions.8 A subsequent study treating dogs with bupivacaine-induced cardiac toxicity showed that lipid infusion successfully resuscitated all subjects whether given immediately or after several minutes of CPR.9

Given that there was no effective and widely available treatment for local anaesthetic toxicity, and clinical trials would be impossible to carry out, the authors of the original studies offered lipid infusion therapy as a possible treatment in refractory arrests. They set up a website (www.lipidrescue.org) to promote its use and provide a forum to present reports of its use in humans as part of resuscitative measures in local anaesthetic toxicity. To date there are 11 such case reports of successful resuscitation of local anaesthetic toxicity, both cardiac and neurological sequelae, using lipid infusions.

The exact mechanism for its action is unclear. One theory is that the lipid emulsion acts as a ‘lipid sink’ in the plasma compartment, capturing the local anaesthetic molecules and making them unavailable to the tissue. Indeed this premise is the basis for current research into its application in other overdose/poisoning scenarios with lipophilic drugs. If this simple theory held true, one would expect that plasma concentrations of local anaesthetic agent would increase after lipid administration. However a recent case report of its successful use in local anaesthetic toxicity demonstrated that plasma concentrations actually decreased and more rapidly then would be explained by pharmacokinetic principles.10 This would suggest that the mechanism may be one of increasing metabolism and distribution of the local anaesthetic. Another theory suggests that as bupivacaine can inhibit fatty acid utilisation by cardiac mitochondria, lipid could act to counter this effect.

As a result of the evidence and successful case reports, the use of lipid infusions has been recommended by the Association of Anaesthetists of Great Britain and Ireland (AAGBI) in its guidelines for treating local anaesthetic toxicity11 a summary of which is presented in Fig. 3.

View Large Image Download to PowerPoint

Fig. 3 AAGBI Guidelines for the treatment of local anaesthetic toxicity.11

6. Summary 

With the increase in popularity of regional techniques and the increasing use of local anaesthetic agents outside of anaesthetic practice, the need for a better understanding of toxicity is as great now as it was 100 years ago. The advent of lipid emulsion therapy offers a treatment for toxicity, and its widespread adoption has highlighted the dangers. Never the less, prevention remains the best treatment and new recommendations for labelling, storing and administrating local anaesthetics are in force.12 Good practice includes checking for intravascular insertion of needle/catheter, fractionating the dose and close observation of the patient. If toxicity does occur after the control of seizures and the institution of CPR, the use of lipid emulsion therapy will hopefully improve the outcome of refractory arrests.