Clinical diagnosis of complex regional pain syndrome type I (CRPS I) in post-traumatic patients is often delayed since the clinical appearance of this disease resembles normal post-traumatic states to a certain extent (pain, edema, loss of function). The purpose of this study was to assess the incidence of specific clinical features in CRPS I patients and normal post-traumatic patients and to evaluate the diagnostic value of a bedside test that measures the sympathetic nervous function. Fifty patients with post-traumatic CRPS I of the upper limb and 50 patients 8 weeks after distal radius fracture with an undisturbed course of disease were subjected to a detailed clinical examination. Pain was assessed using the VAS (visual analog scale), skin temperature measured with an infrared camera and grip-strength with a pneumatic manometer. In CRPS I patients, motor disturbances defined as an impaired active range of motion of the hand, were most frequent (96%, fracture patients: 40%), followed by edema (88%, fracture patients: 80%) and spontaneous pain (VAS 4.0 +/- 2.3, fracture patients: VAS 1.3 +/- 0.6). Systematic temperature differences (>1 degree C) between the affected and unaffected limbs were seen in only 42% of CRPS I patients and in 34% of the fracture patients. Further sensory, sudomotor or trophic changes of the hands were rare. As expected, there were significant differences in the quantity of edema, motor disturbances and sensory disturbances between CRPS I patients and normal fracture patients. However, normal fracture patients still suffered from several of the evaluated symptoms 8 weeks after trauma, which makes an early clinical diagnosis of the complication more difficult. Using a newly developed bedside test, the peripheral sympathetic nervous function was assessed in both groups of patients and in 50 age-matched healthy controls. The decrease in skin blood flow following sympathetic provocation maneuvers, detected by laser Doppler flowmetry, was quantified as sympathetic reactivity. In the affected hands of CRPS I patients, as well as in the contralateral hands, the sympathetic reactivity was obliterated or diminished in contrast to the age-matched controls and normal fracture patients. A multivariate analysis did not reveal any correlation between sympathetic function and the severity of any clinical symptom. Sympathetic reactivity seems to be an independent variable in CRPS I and the test presented may facilitate the difficult clinical diagnosis of this disease.