To the Editor, I read with interest the recent study concerning autonomic, sensory, and motor effects of lumbar epidural anesthesia in patients randomized to receive varying bupivacaine doses and concentrations. At lower concentrations, sensory block was observed in the mid-thoracic dermatomes with little change in the temperature or pulse oximeter perfusion index of the upper and lower extremities despite a concomitant decrease in systemic arterial pressure. These data were interpreted to suggest that sensory neural function may, in some cases, ‘‘be more susceptible than sympathetic neural function to local anesthetic blockade’’, in contrast to widely accepted dogma that the reverse is the case. As alluded to in the Discussion, these data must be interpreted in the context of direct (sympatholysis secondary to local anesthetic effects on spinal sympathetic outflow) and indirect (e.g., baroreceptor-mediated sympathetic activation secondary to fall in systemic arterial pressure) effects on autonomic tone. Thus, the response of sympatho-target organs, as described in this report, will reflect the balance of these opposing effects. A further confounding variable rarely addressed in the literature is that a dermatome may not reflect the segmental organization of sympathetic outflow to that area. A simple example to consider is the thumb, which has sensory innervation from spinal segments C6, but sympathetic preganglionic neurons are not found rostral to T1. In fact, there is extensive convergence and divergence of input from sympathetic preganlionic axons onto postganglionic cells contained in the sympathetic chain, and sympathetic target organs may receive input from numerous spinal cord segments. This may be one mechanism to facilitate a ‘‘widespread diffusion of nervous impulses’’ associated with the sympathetic response to stress. Of note, vasoconstriction of the upper extremity is mediated by sympathetic preganglionic neurons from the middle to the caudal thoracic spinal segment, the axons of which course in the sympathetic chain before synapsing onto sympathetic (vasoconstrictor) postganlionic neurons in the stellate ganglion. Thus, neuraxial anesthesia producing a sensory deficit in dermatome T8 may indeed be accompanied by vascular resistance changes in the upper extremity simply on the basis of anatomy: both afferent activity and autonomic outflow are blocked at the same segmental level. To conclude that this is a result of extraordinary sensitivity of sympathetic outflow to local anesthetic action, as has been decided previously, ignores these anatomical features that have been known for more than a century. A further consideration involves the unique physiological properties of sympathetic preganglionic neurons, such as tonic activity, refractory period, and axonal fibre diameter, which distinguish them from spinal sensory and motor neurons. These properties may account, in part, for the differential blocking effects of local anesthetics on specific types of spinal neurons. Moreover, functionally specific subgroups of sympathetic neurons may themselves be differentiated on the basis of their physiological properties that are particularly matched to the behaviour (e.g. frequency-response characteristics) of their sympathotarget organs. The above points may be relevant when contemplating the differential effects of neuraxial (spinal or epidural) anesthesia on autonomic, sensory, and motor responses.