Early diagnosis is a prerequisite for a successful treatment of complex regional pain syndrome (CRPS). In order to describe neurological symptoms which characterize CRPS, we evaluated 145 patients prospectively. Two-thirds of these were women, the mean age at time of investigation was 50.4 years. CRPS followed limb trauma, surgery and nerve lesion. Employing the current IASP criteria 122 patients were classified as CRPS I and 23 as CRPS II. All patients were assessed clinically pain was quantified using the McGill pain questionnaire, skin temperature was measured by an infrared thermometer and a subgroup of 57 patients was retested in order to determine thermal thresholds (QST). Of our patients 42% reported stressful life events in a close relationship to the onset of CRPS and 41% had a history of chronic pain before CRPS. The latter group of patients gave a higher rating of CRPS pain (P<0.05). The major symptoms were pain at rest in 77% and hyperalgesia in 94%. Typical pain was deep in the limb having a tearing character. Patients getting physical therapy had significantly less pain than those without (P<0.04). Autonomic signs were frequent (98%) and often changed with the duration of CRPS. Skin temperature was warmer in acute and colder in chronic stages (P<0.001). Likewise edema had a higher incidence in acute stages (P<0.001). We found no correlation between pain and autonomic dysfunction. Motor dysfunction (present in 97%) included weakness, tremor, exaggerated tendon reflexes, dystonia or myoclonic jerks. QST revealed increased warm perception thresholds (P<0.02) and decreased cold pain thresholds (P<0.03) of the affected limb. The detailed knowledge of clinical features of CRPS could help physicians early to recognize the disease and thus to improve therapy outcome.

Complex regional pain syndrome (CRPS) is characterized by a triad of sensory, motor and autonomic dysfunctions, with long-standing pain and temperature differences of the affected and contralateral limb as predominant symptoms. The pathogenesis of the disorder still remains unclear. Among the main hypotheses of an underlying pathophysiology we find inflammatory processes and dysfunction of the sympathetic nervous system. Whether the main site of dysfunction is found centrally or peripherally is not known. With psychophysical methods we studied patterns of hyperalgesia to obtain a better understanding of the neuropathic pain component in CRPS. Forty patients in an acute phase of CRPS and a median duration of the disease of 10 weeks, were included in the study. Hyperalgesia to heat was tested with a thermode providing feedback-controlled temperature increases. Two forms of mechanical hyperalgesia were examined: phasic mechanical stimuli by using a custom-made impact stimulator for the determination of individual pain thresholds, tonic mechanical stimuli were applied using a pinch-device. Additionally a 'wind-up' paradigm was used to study a pain phenomenon of presumed central origin: a defined impact stimulus was given once and five times in repetition. A subpopulation of patients was reevaluated for mechanical hyperalgesia after i.v. injection of 500 mg acetyl-salicylic acid. Hyperalgesia to heat was insignificant. We found, however, a marked mechanical hyperalgesia to phasic impact stimuli (P < 0.005), whereas, static stimulation (squeezing skin folds) results were insignificant again. Wind-up related pain was also significantly enhanced in the affected limb (P < 0.02). The anti-inflammatory agent had no effect. These results indicate a non-inflammatory pathogenesis in CRPS presumably central in origin.

The objective of the present investigation was to describe and localize autonomic dysfunction in acute and chronic stages of complex regional pain syndrome (CRPS). Patients were investigated twice: the first investigation was performed as soon as diagnosis was established during the acute stage of CRPS and the second investigation was performed about 2 years later. Twenty-one patients completed the follow-up investigation. The median duration of CRPS was 5 (range 2-21) weeks at first investigation and 94 weeks (22-148) at follow-up. Skin temperature was recorded by thermography, sudomotor function was assessed by thermoregulatory sweat test (TST) and quantitative sudomotor axon reflex test (QSART). Skin temperature was warmer on the affected side at the first investigation (P< 0.001) and colder at follow-up (P < 0.02) compared with the contralateral limb. Sudomotor output was enhanced after both TST (P < 0.005) and QSART (P < 0.05) at the first investigation on the affected side. However, at follow-up, sweating after TST was still increased (P < 0.04) while QSART responses were not different between the affected and unaffected limbs. As compared to controls there was no statistically significant difference, neither in skin temperature nor sweating, neither on the affected nor on the unaffected side. In conclusion, the present investigation proved that vasomotor and sudomotor control are substantially altered in CRPS. In the acute stage vasomotor control is decreased in the affected limb whereas sudomotor function is enhanced. This may be the result of disturbances of thermoregulation, but different secondary peripheral mechanisms, concerning vasomotor and sudomotor function, contribute to clinical presentation of CRPS and affect autonomic function at all stages of CRPS.

Twenty patients suffering from complex regional pain syndrome (CRPS) and 21 healthy control subjects were examined to evaluate sympathetic reflex vasoconstriction. The mean age of the 12 female and eight male patients was 48.9 (21-72) years. At the time of investigation the median duration of the disease was 8.5 weeks (2-70). Twenty-one healthy subjects were investigated for control. Different maneuvers, such as the veno-arteriolar reflex (VAR), inspiratory gasp (IG), cold pressor test (CP) and mental arithmetic (MA), were employed to induce vasoconstriction while the cutaneous blood flow of the affected and the contralateral limb was recorded. In addition, the skin temperature of both limbs was measured by infrared thermography. In 14 of 20 patients and in 14 of 21 control subjects vasoconstriction due to the provocation tests could be measured, while the remaining six patients and seven controls showed vasodilatation in at least one test, and by that they were excluded from analysis of vasoconstrictor reflex pattern. After thermoregulatory adaptation skin temperature was not different between the affected and the unaffected limb. Sympathetic reflex vasoconstriction triggered by MA which represents cortical generated, moderate vasoconstrictor stimulus, was significantly reduced on the affected limb (102.9% of prestimulus period) when compared to the control limb (85.0%, P < 0.002) or to controls (84.8%, P < 0.001). VAR (pure postganglionic), IG and CP (both spinal and supraspinal), representing stronger vasoconstrictor stimuli, revealed no significant side to side difference of sympathetic vasoconstriction and no significant difference as compared to controls. In conclusion our findings prove impairment of sympathetic vasoconstrictor activity after central vasoconstrictor stimulation in CRPS, and possible mechanisms are discussed.

The aim of this study was to investigate the role of local acidosis in the generation of pain in complex regional pain syndrome (CRPS). We investigated ten patients with CRPS of the upper extremity with a mean duration of the disease of 43 weeks (range 4-280 weeks) and ten control subjects for sensitivity to infusion of fluids with low pH (pH 6.1). Another group of five CRPS patients and three healthy controls was investigated using the same protocol but neutral infusion fluid (pH 7.4). A motorized syringe pump was installed for a constant infusion of synthetic interstitial fluid (SIF, either acidified (pH 6.1) or neutral) into the skin at the back of the hands and, thereafter, into the interosseus I muscle on both sides. A flow rate of 30 ml/h was chosen for intradermal and 7.5 ml/h for intramuscular infusion over a period of 10 min. The magnitude of pain was rated on an electronic visual analogue scale. Patients were requested to give their ratings every 10 s during the whole stimulation period. The ratings were normalized as fractions of individual grand mean values. We found significantly increased pain perception during infusion of acidified SIF on the affected side in CRPS patients. Low pH fluid into the skin was significantly more painful between 4 and 6 min (ipsi 1.27 normalized rating (NR) (0. 19-1.94), contra 0.31 NR (0.03-0.51), P<0.02) and between 8 and 10 min (ipsi 1.38 NR (0.19-1.94), contra 0.08 NR (0-0.27), P<0.03) on the affected side, while analysis over the whole stimulation period just failed to reach statistical significance (ipsi 281 area under the curve (AUC) (187-834), contra 87 AUC (28-293), P=0.059). Low pH infusion into the muscle was significantly more painful on the affected side during the whole infusion time (ipsi 861 AUC (308-1377), contra 190 AUC (96-528), P<0.01). The quality of the deep pain during infusion into the muscle was described by the patients as very similar to the CRPS-related pain. In controls we found no side differences of pain intensity during low pH stimulation. Neutral SIF evoked no pain at all, neither in CRPS patients (ipsi 0 AUC, contra 0 AUC) nor in healthy controls. Our results suggest that hyperalgesia to protons is present in patients with CRPS. Further, we could demonstrate that pain is not only restricted to the skin but is also generated in deep somatic tissue of the affected limb.