Local Anesthetic Systemic Toxicity: The Epinephrine You're Not Supposed to Give the Usual Way
A regional block goes systemic mid-injection. The lipid emulsion isn't the hard part — the hard part is running a cardiac arrest algorithm that overturns half of what ACLS usually says to do.
D.K., a 52-year-old man, is halfway through an interscalene brachial plexus block for a planned rotator-cuff repair — the fourth injection this surgeon has done with him assisting this month, routine enough that nobody is watching the clock particularly closely — when he suddenly reports a metallic taste and circumoral tingling, then within ninety seconds becomes tonic-clonic and loses consciousness. His rhythm strip shows wide-complex tachycardia at 160, and his blood pressure, cycling automatically, reads 78/44 before the cuff even finishes its second attempt.
This isn't a seizure managed with benzodiazepines and observation; a sodium-channel blocker delivered directly into a vascular plane has produced true cardiotoxicity, and the standard ACLS reflexes that would normally follow — full-dose epinephrine, amiodarone for the wide-complex rhythm, calcium channel blockers or beta-blockers if it looks supraventricular — are each specifically wrong here. High-dose epinephrine worsens the arrhythmia and impairs lipid-emulsion resuscitation in animal LAST models; vasopressin, calcium channel blockers, and beta-blockers are excluded by ASRA's own checklist for the same reason. What actually reverses this is 20% lipid emulsion, functioning as a lipid sink that redistributes the highly lipophilic bupivacaine away from cardiac sodium channels — and the dose has to be running now, not after a first round of standard drugs has already failed.
The ultrasound machine is still on, its last frozen frame showing what looked like an unremarkable needle position relative to the plexus — the kind of image the block team will spend the debrief studying for a sign of intravascular threading that wasn't visible in real time, since aspiration before injection was negative and the injection itself was fractionated in five-milliliter increments exactly as taught. That review is a question for afterward. In the next sixty seconds, what matters is that his airway is being secured by a second physician while the first pushes lipid emulsion, and that everyone in the room has said the words "this is LAST" out loud, because the entire modified protocol only works if the whole team has already mentally discarded the standard arrest algorithm before the first drug goes in.
At the bedside, ninety seconds into the event
This is LAST until proven otherwise — call it now, out loud, so the whole team switches off standard ACLS reflexes. If epinephrine is needed at all, it's small boluses under 1 mcg/kg, not the full 1 mg push muscle memory wants to reach for. No vasopressin, no calcium channel blockers, no beta-blockers — each one either worsens the arrhythmia or blunts the response to lipid rescue in the animal data ASRA's guidance is built on.
Agreed on all of that, and I'd add: lipid emulsion starts now, in parallel, not after we see whether the modified epinephrine dosing alone turns this around. ASRA's own checklist puts 20% lipid emulsion at the first sign of a major LAST event — 1.5 mL/kg lean body weight as a bolus, roughly 120 mL for him, then an infusion at 0.25 mL/kg/min, with the bolus repeatable if he doesn't stabilize.
The proposed mechanism is a lipid sink pulling the drug off the myocardium — it works better started early than reached for as a last resort once other measures have already stalled.
I'd contest one word in that: "mechanism." The lipid-sink account is the one everybody teaches, but it's contested in the pharmacokinetic literature — sequestration alone doesn't explain the speed of recovery, and direct cardiotonic and metabolic effects are at least as likely to be doing the work. That isn't pedantry. If you believe it's purely a sink, the dose logic is "more lipid, more binding," and that's how people end up past the 12 mL/kg ceiling and into lipid overload. And I'd push back on framing epinephrine as the thing to fear. He isn't arrested. His pressure is 78/44 with a perfusing wide-complex rhythm, which is the state ASRA's own dose reduction was written for, not a state where epinephrine is withheld — the failure I've actually watched is a team so primed against ACLS reflexes that nobody gives the small dose either.
Agreed and executed within the first four minutes: lipid emulsion bolus given, infusion started, epinephrine limited to sub-microgram boluses as needed for pressure support, and the full standard-ACLS drug list (vasopressin, calcium channel blockers, beta-blockers, reflexive amiodarone) explicitly withheld. Rhythm converted to sinus tachycardia within six minutes; he regained consciousness in the ICU roughly ninety minutes later with no apparent neurologic deficit.
The surgery itself was cancelled for the day, and the block technique — ultrasound guidance was already in use — is being reviewed separately for whether intravascular injection was genuinely unavoidable or a technical miss, a question left to the department's own case review rather than resolved at the bedside.