To the Editor:—
Ladd et al. 1showed in conscious sheep that the administration of incremental doses of local anesthetics in the central nervous system bloodstream induces minimal systemic recirculation, dose-dependent central nervous system excitatory behavior, and electroencephalographic changes. These authors also observed that bupivacaine is more neurotoxic than levobupivacaine, which is also more neurotoxic than ropivacaine. Finally, they stated that there are no differences among the drugs concerning their arrhythmogenicity.
It is well established that large doses of local anesthetics slow cardiac conduction velocity in a dose- and use-dependent manner by inhibiting the fast-inward sodium current. 2The slowing of ventricular conduction velocity is responsible for a proportional QRS widening 3,4and facilitates the occurrence of reentrant ventricular arrhythmias around arcs of conduction block. 5,6In other words, the increase in heart rate by enhancing the slowing of ventricular conduction is one of the main mechanisms facilitating the occurrence of serious reentrant ventricular arrhythmias in the case of high plasma concentration. Moreover, Mazoit et al. 7and our team 8reported that bupivacaine is more use-dependent than levobupivacaine and ropivacaine. This fact explains at least in part the differences in direct cardiotoxicity in these three local anesthetics. As demonstrated by Bernards and Artru 9,10and our team, 11an overdose of local anesthetics induces an increase in the sympathetic outflow by their direct neurotoxicity and therefore an increase in heart rate, facilitating the occurrence of serious reentrant arrhythmias when toxic cardiac concentrations are reached.
In the elegant study published by Ladd et al. , 1RR and PR intervals are obviously shortened. In contrast and surprisingly, QRS is shortened by the administration of any local anesthetics at any dose. One might argue that the baseline values are somewhat widened and that the method of measurement, including the paper speed, is not specified in the method section. Nevertheless, because QRS is not widened, one could specially argue that there is no direct cardiotoxicity and that the observed arrhythmias are only due to the pre- and convulsant effects of local anesthetics as any seizures can provoke. In conclusion, the sentence in the summary that “no differences were found in their arrhythmogenic potential”1may be confusing because the analyzed arrhythmogenicity in the present study is only in terms of indirect and neurally mediated cardiotoxicity and not in terms of direct cardiotoxicity.