Patients with reflex sympathetic dystrophy have posttraumatic pain disproportionate to the injury and spreading beyond the distribution of any single peripheral nerve. We examined sympathetic neurocirculatory function and the role of sympathetic postganglionic nerve traffic in maintaining the pain in 30 patients with reflex sympathetic dystrophy. Most had had the condition for more than 1 year, and 14 had undergone sympathectomy for the pain. Positron emission tomographic scanning after administration of 13N-ammonia was used to assess local perfusion, and 6-[18F]fluorodopamine was used to assess sympathetic innervation. Rates of entry of norepinephrine in the regional venous drainage (spillovers) and regional plasma levels of L-dihydroxyphenylalanine (the immediate product of the rate-limiting enzymatic step in norepinephrine biosynthesis) and dihydroxyphenylglycol (the main neuronal metabolite of norepinephrine) were measured with and without intravenous trimethaphan for ganglion blockade. 13N-Ammonia-derived radioactivity was less on the affected side than on the unaffected side, whereas 6-[18F]fluorodopamine-derived radioactivity was symmetrical. Thus, perfusion-adjusted 6-[18F]fluorodopamine-derived radioactivity was higher on the affected side. Norepinephrine spillover and arteriovenous increments in plasma levels of L-dihydroxyphenylalanine and dihydroxyphenylglycol did not differ significantly between affected and unaffected limbs, although 4 patients had noticeably less norepinephrine spillover and smaller arteriovenous increments in plasma dihydroxyphenylglycol on the affected side. Trimethaphan decreased the pain in only 2 of 12 nonsympathectomized patients. The results indicate that patients with chronic unilateral reflex sympathetic dystrophy have decreased perfusion of the affected limb, symmetrical sympathetic innervation and norepinephrine synthesis, variably decreased release and turnover of norepinephrine in the affected limb, and failure of ganglion blockade to improve the pain in most cases. These findings suggest augmented vasoconstriction, intact sympathetic terminal innervation, possibly impaired sympathetic neurotransmission, and pain usually independent of sympathetic neurocirculatory outflows.