Local anesthetic additives are the drugs you mix into a block to make it start faster, last longer, or feel better. They are everywhere in regional practice, and they are easy to reach for. But here is the part that matters at the bedside: with one exception, an additive changes the block, not the dose limit. Your maximum safe milligrams and your LAST risk are set by the local anesthetic itself, not by what you add to it.
The one that changes the math: epinephrine
Epinephrine is the exception. By constricting local vessels it slows how fast the anesthetic is absorbed into the bloodstream, which prolongs the block and raises the maximum dose you can give for most agents. Ropivacaine is the holdout: it is intrinsically vasoconstrictive, so epinephrine does not meaningfully lift its ceiling. Commercial mixes run 1:100,000 (10 mcg/mL) or 1:200,000 (5 mcg/mL). Skip it in end-arterial regions such as digits, and be cautious in cardiac disease.
Sodium bicarbonate: speeds onset, but watch bupivacaine
Alkalinizing a solution raises the nonionized fraction of the drug, so more of it crosses the nerve membrane and the block sets up faster. Lidocaine and 2-chloroprocaine tolerate this well. Bupivacaine does not: it precipitates out of solution after only a small amount of bicarbonate, so the two effectively cannot be mixed. Ropivacaine sits in between, where dilute 0.2% can be carefully alkalinized but higher concentrations crash 25 to 30 percent of the drug out of solution. This is a compatibility issue, not a toxicity one, and it changes nothing about the dose math.
The alpha-2 agonists: clonidine and dexmedetomidine
Both prolong a block by acting on alpha-2 receptors. Clonidine adds roughly two hours to a peripheral block at about 0.5 mcg/kg (up to ~150 mcg); push the dose and you trade duration for hypotension, bradycardia, and fainting. Dexmedetomidine is the stronger of the two: perineural doses of 0.5 to 1 mcg/kg can extend analgesia by hours, and intrathecal 5 to 10 mcg dramatically lengthens a spinal. The cost is dose-dependent, reversible bradycardia, plus one caveat worth remembering: in already-compromised nerves, alpha-2 agonists may add to neurotoxicity.
Dexamethasone: and the perineural-versus-IV question
A few milligrams of dexamethasone is one of the most reliable ways to stretch a single-shot block. It roughly doubles the duration of a long-acting local, on the order of 4 to 8 extra hours, without the hemodynamic baggage of the alpha-2 agonists. The effect is dose-dependent up to about 4 mg perineurally: 1 mg does little, and going above 4 mg adds nothing.
Here is the question that actually matters at the bedside: do you mix it into the block, or give it IV? The recent evidence says it barely matters which. A 2024 network meta-analysis of 118 trials (about 9,300 patients) found both routes reach the same ceiling of effect, with perineural beating IV by only about 2 to 2.7 hours, which falls below what patients reliably notice. IV is simply a little less potent milligram for milligram, so it takes a bit more to match it, roughly 8 mg IV to equal 4 mg perineural.
Because that perineural edge is so small, the trend is to give dexamethasone IV. Intravenous use is on-label, it doubles as PONV prophylaxis, it sidesteps compatibility concerns with ropivacaine, and it avoids the unsettled questions about injecting steroid around a nerve. Two caveats: nearly all of this evidence is in long-acting locals such as bupivacaine and ropivacaine, so do not assume the same prolongation with intermediate or short-acting agents, and expect a transient rise in blood glucose that matters most in diabetic patients.
Opioids
Intrathecal fentanyl, around 25 mcg, is a standard spinal adjuvant. Perineural buprenorphine can extend a long-acting block by 6 to 12 hours, but it drives nausea and vomiting up, so plan antiemetic prophylaxis if you use it.
The additives at a glance
Effect, typical dose, duration, whether it moves the dose ceiling, and the catch for each:
- Epinephrine — vasoconstricts to slow absorption; ~1:200,000 (5 mcg/mL). Prolongs the block and is the one additive that raises the max LA dose (except ropivacaine). Avoid end-arterial sites; caution in cardiac disease.
- Sodium bicarbonate — speeds onset by raising the nonionized fraction; ~1 mEq per 10 mL lidocaine. Changes onset, not duration, and never the dose limit. Precipitates bupivacaine (don't mix); limited with ropivacaine.
- Clonidine — alpha-2 agonist; 0.5 mcg/kg (up to ~150 mcg). Adds ~2 h to a peripheral block; no effect on the dose limit. Hypotension, bradycardia, and sedation at higher doses.
- Dexmedetomidine — the strongest alpha-2 agonist; 0.5-1 mcg/kg perineural (5-10 mcg intrathecal). Adds several hours; no effect on the dose limit. Reversible bradycardia, and possible neurotoxicity in already-injured nerves.
- Dexamethasone — steroid; 4 mg perineural or ~8 mg IV. Roughly doubles a long-acting block, and the route barely matters (perineural beats IV by only ~2-2.7 h). No effect on the dose limit. IV is on-label, covers PONV, and avoids ropivacaine precipitation; watch glucose.
- Opioids — intrathecal fentanyl 25 mcg is a standard spinal adjuvant; perineural buprenorphine adds 6-12 h but raises PONV. No effect on the dose limit.
The bottom line
None of these additives except epinephrine belongs in your dose-limit calculation. They shape the block; the local anesthetic sets the ceiling. And one more thing worth saying out loud: no perineural adjuvant is actually licensed for that use. Every one of them is off-label, so the decision to add it is yours to own.