OBJECTIVE To examine whether changes in circulating norepinephrine are associated with the sensory disturbances of diabetic polyneuropathy. Experimental studies have indicated that NE can excite sprouts from injured nerves, producing pain.

RESEARCH DESIGN AND METHODS We measured supine and erect plasma NE in 13 normal, nondiabetic control subjects and three groups of diabetic patients: 20 without clinical neuropathy, 20 with chronic painful neuropathy, and 15 with painless neuropathy and foot ulceration. Neuropathy was characterized bysymptom and deficit scores, sensory thresholds, electrophysiology, and cardiovascular autonomic function tests. Neuropathic pain was scored by the patients on a linear analogue scale.

RESULTS In painless neuropathy, NE levels were greatly reduced (supine, 1.3 nM; erect, 2.2 nM) compared with control subjects (supine, 2.4 nM; erect, 4.0 nM; P < 0.001) and were combined with grossly abnormal autonomic reflexes. NE also was reduced in the diabetic group without neuropathy (supine, 1.7 nM; erect, 2.7 nM; P < 0.01 vs. control subjects). By contrast, in painful neuropathy NE levels (supine, 2.2 nM; erect, 3.6 nM) were similar to control subjects and significantly higher than in painless neuropathy (P < 0.01). Furthermore, NE correlated with the severity of neuropathic pain (r = 0.46, P = 0.02). To assess whether pain, acting as a stressor, could account for the observed differences in NE, we also measured the stress hormones epinephrine and cortisol. They did not differ among the diabetic groups.

CONCLUSIONS Circulating NE is higher in painful than painless diabetic neuropathy. We suggest that painful neuropathy is associated with a relatively higher number of functioning sympathetic fibers that may contribute to pain.