Spinal cord injuries have direct and indirect effects on sympathetic outflow below the lesion. Loss of descending drive markedly reduces sympathetic discharge leading to hypotension unless the lesion is caudal to T6 when the intact supply above the lesion helps stabilize blood pressure. Injuries to the thoracic cord that directly damage preganglionic neurones may denervate postganglionic vasoconstrictor neurones. Although these neurones receive multiple inputs arising from several spinal segments, ganglionic transmission depends mainly on only one “strong” input that produces a very large suprathreshold postsynaptic potential, as at the neuromuscular junction. The loss of preganglionic inputs was investigated after resecting the caudal lumbar paravertebral chain of guinea pigs above L4 white ramus (under anaesthesia with 60 mg/kg ketamine and 10 mg/kg xylazine i.p.) (Ireland, 1999). Recordings of synaptic responses to graded stimulation in the isolated chain distal to the lesion site showed that only one in 10 neurones in L5 ganglion normally receive a strong input via L4 white ramus but, a few weeks after the lesion, nearly 60% of neurones had a strong input as well as some weak ones. These new inputs arose by sprouting of the few remaining L4 preganglionic axons and enabled many postganglionic neurones again to relay impulses to their peripheral targets. These novel connections, if inappropriate for the targets, might account for undifferentiated sympathetic activation during autonomic dysreflexia, when not only serious hypertensive episodes but also excessive sweating can be triggered in people with spinal cord injury. The hypertensive episodes are initiated by stimuli below the lesion, such as a distended bladder or bowel, and have been suggested to result from exaggerated spinal reflexes following the expansion of afferent inputs within the damaged spinal cord (Weaver et al., 2006), exacerbated by the loss of baroreflex compensation. However, recordings from sympathetic axons in people with spinal cord injury (Stjernberg et al., 1986) show only transient activation of vasoconstrictor neurones from bladder afferents but prolonged vasoconstriction, implying enhanced vascular responses to sympathetic activation. This was investigated in isolated segments of rat arteries taken from above and below a spinal transection (conducted under anaesthesia with 60 mg/kg ketamine and 10 mg/kg xylazine i.p.). In the tail artery (Yeoh et al., 2004a), responses to 1 Hz stimulation were enhanced 2.5x and those to lower frequencies >20x. The effects were similar with T4 or T8 lesions, neither of which damages the sympathetic outflow. Responses to applied agonists and antagonists revealed unchanged postjunctional α1- and slightly increased α2-sensitivity, reduced prejunctional α2-sensitivity and unchanged activity of the prejunctional noradrenaline transporter (NAT) up to 8 weeks after the lesion. Contractions to raised [K+]o were larger and prolonged. When ongoing sympathetic activity was abolished by decentralizing the neurones supplying the tail artery, the changes in neurovascular transmission were almost identical to those following spinal transection (Yeoh et al., 2004b). The potentiation of neurally-evoked vasoconstriction in the tail artery was not unique. Neurovascular transmission was similarly enhanced after T4 transection in the saphenous artery but, in this case, postjunctional α-sensitivity was reduced and prejunctional α2-sensitivity, NAT activity and muscle reactivity were unchanged. Potentiation of NA release is suspected in both the saphenous and tail artery but this needs to be tested. After inactivating the perivascular peptidergic afferents with capsaicin, the responses of mesenteric arteries to 1 Hz were enhanced 8x after spinal transection (Brock et al., 2006). The underlying mechanisms were unlike those in the cutaneous arteries, with unchanged postjunctional sensitivity or reactivity but a reduction in NAT activity leading to higher junctional concentrations of noradrenaline. In contrast to these changes below a spinal transection, nerve-evoked contractions in the median artery were very similar to control, consistent with the idea that the changes below a spinal transection follow the drop in ongoing sympathetic discharge. Although the adjustments of different vascular beds to reduced nerve impulses are distinct, the widespread enhancement of vascular responses is likely to contribute to autonomic dysreflexia.