toxicity is 2-3 µg/ml of total (protein bound and unbound) and 0.1-0.2 µg/ml of unbound bupivacaine, levobupivacaine and ropivacaine [1,2]. Notably, the intra-arterial concentration of unbound bupivacaine for inducing CNS toxicity is approximately 50 % of unbound ropivacaine, both of which are three-and four-fold higher, respectively, than the intravenous concentration [1]. Cumulative case reports and animal experiments have also shown that the systemic toxicity of levobupivacaine and ropivacaine is less than bupivacaine [3,4]. Infants are prone to develop CNS toxicity by bupivacaine at total and unbound concentrations lower than these values [5]. Although the threshold plasma concentration for cardiac toxicity is higher than for CNS, symptoms such as dysrhythmia and QRS widening due to decreased intraventricular conduction by long-acting local anesthetics may appear prior to any neurological manifestations in infants [6]. As neonates and infants have a higher heart rate than adults, the intensity of the block is also higher (use-dependent block) and they are more prone to the toxic effects of bupivacaine, levobupivacaine and ropivacaine than adults. Hypoxia, acidosis, hypothermia and electrolyte disorders increase cardiac toxicity [3].
Protein binding and dispositionAmide local anesthetics are predominantly bound to plasma alpha 1 -acid glycoprotein (AAG), and to a minor extent to albumin. The plasma concentration of AAG at birth is approximately 20-50 % of that in adults. During the first 6-9 months of life, it progressively increases to reach adult levels by the end of the first year [7] (Fig. 1a). Although the levels of albumin in neonates are also lower than in adults, its affinity to local anesthetics is approximately 5,000-10,000 times lower than AAG, suggesting that albumin