- Email: [email protected]
Skin temperature side differences – a diagnostic tool for CRPS?
Pain 98 (2002) 19–26 www.elsevier.com/locate/pain
Skin temperature side differences – a diagnostic tool for CRPS? Gunnar Wasner, Jo¨rn Schattschneider, Ralf Baron* Klinik fu¨r Neurologie, Universita¨tsklinikum Kiel der Christian-Albrechts-Universita¨t, Niemannsweg 147, 24105 Kiel, Germany Received 6 July 2001; received in revised form 29 October 2001
Abstract Complex regional pain syndrome type I (CRPS I) is a chronic painful disease of one extremity that may develop as a disproportionate consequence of a trauma affecting the limbs without overt nerve injury. It is clinically characterized by sensory, motor and autonomic symptoms including vascular abnormalities. Previously, we have reported that pathophysiological alterations of the ongoing sympathetic activity play a crucial role in vasomotor disturbances (Brain 124 (2001) 587). As a companion article, the aim of this study was to evaluate the diagnostic value of skin temperature side differences in consideration of the spontaneous sympathetic vasoconstrictor activity. Twentyfive patients with CRPS I were studied. Fifteen patients with painful limbs of other origin and 20 healthy individuals served as controls. Controlled thermoregulation was performed to change cutaneous sympathetic vasoconstrictor activity by the use of a thermal suit: skin sympathetic vasoconstrictor neurones were activated by whole-body cooling and nerve activity was abolished by whole-body warming. Skin temperature at the affected and unaffected limbs (infra-red thermometry) was measured under resting conditions and continuously monitored during controlled modulation of sympathetic activity. The results showed only minor skin temperature asymmetries between both limbs under resting conditions in most CRPS patients. However, during controlled thermoregulation temperature differences between both sides increased dynamically and were most prominent at a high to medium level of vasoconstrictor activity. In both control groups, there were only minor side differences in temperature both in rest and during thermoregulatory changes of sympathetic activity. When comparing the diagnostic value of skin temperature asymmetries in CRPS I, sensitivity was only 32% under resting conditions, but increased up to 76% during controlled alteration of sympathetic activity. Specificity was 100% at rest and 93% at controlled thermoregulation. We concluded that the degree of unilateral vascular disturbances in CRPS I depends critically on spontaneous sympathetic activity. Taking this into consideration, skin temperature differences in the distal limbs are capable of reliably distinguishing CRPS I from other extremity pain syndromes with high sensitivity and specificity. q 2002 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. Keywords: Complex regional pain syndrome; Sympathetic nervous system; Skin temperature; Thermoregulation
1. Introduction Complex regional pain syndrome type I (CRPS I, formerly reflex sympathetic dystrophy) is a neuropathic pain condition that may develop as a disproportionate consequence of a trauma affecting the limbs like bone fractures, sprains or skin lesions (Schwartzman and McLellan, 1987; Veldman et al., 1993; Baron et al., 1996; Wasner et al., 1998). CRPS I is distinguished from CRPS II (causalgia) in which a partial lesion of a peripheral nerve is necessary for the diagnosis (Merskey and Bogduk, 1995; StantonHicks et al., 1995). The clinical features are spontaneous pain, hyperalgesia, impairment of motor function, swelling and autonomic abnormalities in a single extremity. The
* Corresponding author. Tel.: 149-431-597-2633; fax: 149-431-5972712. E-mail address: [email protected] (R. Baron).
symptoms tend to spread distally without confining to the innervation zone of an individual nerve or root. Autonomic signs include abnormalities in skin vascular and sudomotor functions that are due to unilateral disturbances of the sympathetic innervation (Wasner et al., 1999, 2001). Clinically, the measure of skin temperature differences between affected and unaffected side has repeatedly suggested to be an important diagnostic test for CRPS (Blumberg and Ja¨nig, 1994; Bruehl et al., 1996). This is of particular relevance to distinguish this entity from extremity pain syndromes of different etiology. Recently, Harden et al. (1999) pointed out that internal validity of the International Association for the Study of Pain (IASP)/CRPS criteria can be improved by paying more attention to vasomotor signs. However, one dilemma seems to be the poor intra- and interindividual reproducibility of temperature asymmetries in CRPS when measured at different times (Sherman et al., 1994).
0304-3959/02/$20.00 q 2002 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. PII: S 0304-395 9(01)00470-5
20
G. Wasner et al. / Pain 98 (2002) 19–26
Table 1 Clinical characteristics of CRPS patients a No.
Age/Sex
Location
Inciting event
Duration (months)
DTrest (8C)
Contra Trest (8C)
DTmax (8C)
Contra Tmax (8C)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
62/f 62/m 43/f 44/f 49/f 51/f 55/m 56/f 27/f 34/f 40/f 50/f 66/m 63/f 57/m 40/m 52/f 64/f 34/f 49/f 33/f 30/m 39/f 43/f 27/m
L upper limb R upper limb L upper limb L upper limb L lower limb R upper limb L lower limb L upper limb R lower limb R upper limb R upper limb L lower limb L upper limb R upper limb R upper limb R upper limb R upper limb R upper limb R upper limb R lower limb R upper limb R lower limb L upper limb R lower limb L lower limb
Wrist fracture Wrist fracture Metacarpal fracture Colles fracture Ankle joint distortion No obvious trauma Strain Colles fracture No obvious trauma Post-elbow surgery Strain Lower leg fracture Colles fracture Tendosynovitis Shoulder torn tendon No obvious trauma Colles fracture Elbow dislocation Tendosynovitis No obvious trauma Post-tendosynovitis Metacarpal fracture No obvious trauma Ankle joint fracture Sprain
1.5 2.5 20 12 35 48 5 14 48 48 3 25 2 5 2 10 0.5 3.5 18 4.5 7 10 15 7 2
20.4 20.5 2.0 7.5 29.5 21.9 1.3 26.4 20.3 22.2 0.7 22.1 0.6 1.8 2.1 1.8 0.8 0.4 21.9 4.0 0.4 20.3 0 0.5 1.9
35.3 32.7 31.4 27.7 33.9 29.5 21.8 33.5 24.4 23.7 34.1 26.5 33.6 26.1 29.4 24.8 34.6 34.8 23.5 28 27.8 23.3 22 26.9 21.7
1.2 2.7 2.9 6.4 29.5 22.9 10.4 25.2 23.0 25.1 1.1 23.1 21.4 1.8 6.9 21.7 10.2 3.8 24.2 7.0 26.1 24.9 2.4 21.7 7.4
32 28 29 28 34 31 25 32 29 32 28 32 32 26 25 34 25 27 30 27 34 35 30 32 25
a
f, female; m, male; R, right; L, left.
Skin temperature is under the control of sympathetic vasoconstrictor neurones. Their spontaneous nerve activity depends critically on the thermoregulatory state and may change rapidly inducing alterations in skin temperature as demonstrated in animal experiments and human microneurographic recordings (Bini et al., 1980; Ha¨ bler et al., 1994, 1998). So far, skin temperature measurements when used as diagnostic criterion for CRPS were performed as a single measurement under resting conditions, e.g. defined as resting in supine position, room temperature and acclimatization for 30 min. However, the skin temperature varies dynamically and continuously depending on the thermoregulatory state of the patient as described above. Therefore, the question arises whether one single measurement of skin temperature side differences at resting conditions really reflects the maximal possible difference that occurs during different states of thermoregulation. The present investigation aims at several questions: the skin temperature dynamic and, in particular, the variation of temperature differences between both sides that occurs during different thermoregulatory states will be investigated in patients with CRPS and in controls. If skin temperature side differences are used as a diagnostic test for CRPS, is it possible to increase the sensitivity of this test by using the maximal side difference in skin temperature? As a companion paper to a previous published article data of the present study were collected from the same investigations (Wasner et al., 2001).
2. Methods 2.1. Patients and controls 2.1.1. Patients with CRPS I The study was performed on 25 patients, 18 women and seven men (mean age 47 years, range 27–66), with the diagnosis of unilateral CRPS I who were referred to the Interdisciplinary Pain Center of the University Clinic of Kiel from 1995 to 2000 (Table 1). The upper extremity was affected in 17 cases and the lower in eight cases. CRPS I was diagnosed according to the criteria defined by Evans (1946) and to the novel clinical criteria defined by the IASP (Merskey and Bogduk, 1995; Stanton-Hicks et al., 1995). All patients were clinically characterized by spontaneous pain (at least in their medical history) and evoked pains (e.g. deep hyperalgesia, mechanical allodynia) that were distally generalized and that were not restricted to an innervation territory of any peripheral nerve. In all cases, pain was increased by movements of the affected limb and patients had at least one symptom of motor dysfunction like impairment of muscle strength, tremor or dystonia. Furthermore, there was or there had been evidence of at least one autonomic involvement like edema, skin temperature asymmetries or sweating abnormalities. Patients who had additional diseases that might contribute to neuropathic pain, e.g. diabetes, were excluded. By this procedure the incidence of false positive diagnoses was minimized. Additional investigations like radio-
G. Wasner et al. / Pain 98 (2002) 19–26
21
Table 2 Clinical characteristics of patients with chronic painful limbs of other origin a No. Age/sex Location 1
38/m
2 3 4 5 6
47/m 39/f 57/m 49/f 56/m
7 8 9 10
38/m 42/f 32/f 18/f
11 12 13
53/f 49/m 42/m
14 15
36/f 50/f
a
Disease
Duration (months) DTrest (8C) Contra Trest (8C) DTmax (8C) Contra Tmax (8C)
L upper limb Traumatic neuralgia of ulnar nerve and 33 the internal cutaneous nerve R upper limb Lunatomalacia 0.5 R lower limb Radiculopathy 18 L upper limb Central pain after thalamic infarction 6 R lower limb Achillodynia 103 L upper limb Ischemic nerve lesion at wrist and 12 dorsal hand L upper limb Pseudoarthrosis in the wrist 10 R lower limb Traumatic neuralgia of peroneal nerve 17 L upper limb Carpal tunnel syndrome 6 R upper limb Traumatic nerve lesion at wrist and 48 ulnar hand R upper limb Traumatic neuralgia of radial nerve 25 L upper limb Neuralgia of ulnar nerve 23 L upper limb Severe shoulder trauma with lesion of 7 brachial plexus R upper limb Brachial plexopathy 38 L upper limb Traumatic nerve lesion at thumb 6
20.7
27.3
0.7
34
0.4 20.3 20.4 0.2 0.5
32 25.4 34.7 25.1 33
0.7 20.7 0.6 1.2 0.6
27 25 31 27 26
21.2 1.1 21.6 20.1
33.7 26 23.7 30.2
1.2 2.5 1.6 1.1
31 29 27 27
0.2 21.8 20.3
34.4 25 34.1
0 1.15 0.7
25 30
0 0.1
32.5 24.2
1.5 1.8
25 30
f, female; m, male; R, right; L, left.
graphy and three phase bone scan was performed. The clinical data are summarized in Table 1. 2.1.2. Patients with chronic painful limbs of other origin Fifteen patients, eight women and seven men (mean age 42 years, range 18–57), with chronic pain of one limb of other origin (patients did not meet the criteria described above) served as a control group (Table 2). Although, the patients are suffering from different diseases, the following reasons argue against the possibility, that they have early CRPS I or II. (1) In all patients with nerve injury the pain was restricted to the innervation territory of the affected nerve without tendency to spread beyond. Therefore, they were classified as having posttraumatic neuralgia (Baron et al., 1999). (2) This clinical picture was stable for several months in the control patients, therefore it was unlikely that symptoms begin to generalize like it would be expected in early CRPS II. (3) No control patient demonstrated trophic disturbances. (4) None of the patients suffered from edema and there was no history of autonomic abnormalities. All patients underwent a general physical and neurological examination. 2.1.3. Healthy controls Twenty healthy subjects (11 women and nine men) served as a control group (mean age 27 years, range 23–45). 2.1.4. Procedures The tests were performed between 3 and 6 p.m. The patients and controls were tested in supine position (room temperature 22–248C). None of the control subjects or patients was on drugs affecting vascular function. Patients suffering from cardiovascular disorders were excluded from
the study. Aims and procedures of the study were explained to all subjects according to the Helsinki Declaration. All individuals gave their informed consent to participate in the study, which was approved by the local ethical committee. The procedures followed were in accordance with institutional guidelines. 2.2. Investigations and measurements 2.2.1. Skin temperature side differences under resting conditions After resting in supine position (room temperature 22– 248C) for 30 min skin temperature was measured bilaterally at all finger or toe tips with infra-red thermometers. For each extremity the mean value of all finger and toe tips, was calculated. The absolute side difference in skin temperature under resting conditions between both extremities was determined by the following formula: Side difference in skin temperatures in patients : DTrest ¼ uskin temperature on affected side 2 skin temperature on unaffected sideu Side difference in skin temperature in healthy controls : DTrest ¼ uskin temperature on right side 2 skin temperature on left sideu 2.2.2. Thermoregulatory cycle Skin sympathetic vasoconstrictor activity is under thermoregulatory control. Controlled thermoregulatory reflexes
22
G. Wasner et al. / Pain 98 (2002) 19–26
were performed to induce a physiological tonic change of sympathetic skin nerve activity. This was achieved by changing environmental temperature with the help of a thermal suit. The subject was lying in a cotton suit supplied by tubes, in which circulating water of 12 and 508C (inflow temperature) was used to cool or warm the whole body, respectively. Both hands and feet were not covered by the suit. Wholebody cooling is the most effective stimulus to induce a massive tonic activation of cutaneous vasoconstrictor neurons as demonstrated in microneurographical recordings (Bini et al., 1980); whole-body warming leads to a complete inhibition of this activity. Skin temperature measurements (bilaterally at all finger or toe tips at 5 min intervals with infra-red thermometers, calculation of mean values for each extremity) were performed during a complete thermoregulatory cycle, i.e. during the entire spectrum of sympathetic vasoconstrictor activity. The temperature on the contralateral not affected extremity (or right side in healthy controls) was taken as a reference to estimate the level of sympathetic activity. After the patients had put on the thermal suit followed by a resting period lasting for 30 min, first, whole-body cooling was performed to induce maximal vasoconstrictor activity. The cooling session was continued until the skin temperature on the unaffected side (or right side in healthy controls) was close to room temperature (i.e. 258C). Thereafter, whole-body warming was performed until the skin temperature on the unaffected side (or right side in healthy controls) was close to body core temperature (i.e. 358C) in order to induce maximal inhibition of sympathetic activity. All data that were used in the evaluation were collected during the whole-body warming period after whole-body cooling was performed. Whole-body cooling was used to establish a baseline allowing comparability of all patients.
2.3. Data acquisition and statistical analysis For comparison of non-parametric data between CRPS patients, patients with chronic painful limbs of other origin and healthy controls the Mann–Whitney U test was used. Skin temperature side differences under resting conditions and maximal asymmetries during controlled thermoregulation were compared between the groups. Data are given as mean ^ standard deviation and median. P-values of ,0.05 were regarded as statistically significant. The 95% confidential values were calculated to determinate specificity and sensitivity of temperature side differences at rest and during the thermoregulatory cycle. 3. Results 3.1. Skin temperature side differences under resting conditions In CRPS patients skin temperature side differences at rest between the affected and healthy limbs were on average DTrest ¼ 2:1 ^ 0:58C (mean ^ SEM; median 1.88C; range 0–9.48C) after resting in supine position for 30 min (room temperature 22–248C) (Table 1). In patients with extremity pain of other origin average side differences were DTrest ¼ 0:6 ^ 0:38C (median 0.48C; range 0–1.88C) (Table 2). Healthy controls had side differences of DTrest ¼ 0:6 ^ 0:18C (median 0.58C; range 0–2.18C). The 95% confidential value was 2.08C. Temperature side differences during rest were on average significantly greater in CRPS patients in comparison to patients with extremity pain of other origin and healthy controls ðP , 0:01Þ. 3.2. Skin temperature during thermoregulatory cycle
2.2.3. Maximal skin temperature side difference during the thermoregulatory cycle The absolute maximal side difference in skin temperature that occurred during the whole thermoregulatory cycle was determined using the following formula: Maximal side difference in skin temperature in patients : DTmax ¼ uskin temperature on the affected side 2 skin temperature on unaffected sideu
Maximal side difference in skin temperature in healthy controls :
DTmax ¼ uskin temperature on right side 2 skin temperature on left sideu
3.2.1. Healthy controls and patients with extremity pain of other origin Whole-body cooling induced a symmetrical decrease in skin temperature in both limbs due to maximal tonic activation of cutaneous sympathetic vasoconstrictor neurons. Thereafter, skin temperature increased bilaterally during whole-body warming, because of inhibition of cutaneous sympathetic vasoconstrictor activity (Fig. 1A, B). The regulation pattern was identical in the healthy control group and in the group of patients with extremity pain of other origin. Only small side differences of skin temperature occurred during the entire cycle (Table 2). 3.2.2. Patients with CRPS I Patients with CRPS I were characterized by asymmetrical changes of skin temperature during the thermoregulatory cycle. In the unaffected distal extremity whole-body cooling led to an immediate sustained decrease in skin temperature very similar to the situation in healthy controls. On the
G. Wasner et al. / Pain 98 (2002) 19–26
23
Fig. 1. Characteristics of skin temperature as a measure of cutaneous sympathetic vasoconstrictor activity at the fingers of both hands in one healthy control subject (A,B) and in a patient with CRPS I (C,D) during a controlled thermoregulatory cycle (controlled alterations in cutaneous sympathetic activity). Controlled thermoregulatory changes (whole-body cooling and warming) were applied by means of a thermal suit to change environmental temperature in a standardized way and reflexly alter skin sympathetic vasoconstrictor activity. The subject was lying in a suit supplied by tubes, in which running water of 12 and 508C (inflow temperature) was used to cool or warm the whole body, respectively. During the experiment, the skin temperature of the fingers of both hands was monitored at regular intervals. (A) Temperature of the right and left hand in a healthy subject during the thermoregulatory cycle. (C) Affected extremity (CRPS) and healthy side (contralateral) in a CRPS patient. Side differences in skin temperature of the fingers of both hands in one healthy control subject (B) and in a CRPS patient (D) during a controlled thermoregulatory cycle. The arrow indicates beginning of whole-body warming. Data that were used in the evaluation were collected during this period. Maximal skin temperature differences during whole-body warming are indicated by the vertical dotted lines. Same subjects as in (A,C).
affected side, three different patterns of regulation were observed. 1. Patients with a ‘warm’ regulation type showed higher cutaneous temperatures in the affected limb than in the contralateral limb during the entire spectrum of sympathetic vasoconstrictor activity (Fig. 1C, D). This regulation type was found in 11 patients. 2. In patients with an ‘intermediate’ type, the direction of the temperature side difference changed during the thermoregulatory cycle. In some patients the affected side was warmer during increased sympathetic activity and colder during decreased activity and vice versa. Seven patients were characterized by regulation type. 3. In patients with a ‘cold’ regulation type, vasoconstriction was more intense on the affected side during whole-body cooling. As a result, skin temperature in the affected limb was lower than in the unaffected side during the entire
thermoregulatory cycle. This regulation type was present in seven patients. Besides these differences all CRPS patients had the following factor in common: the differences in skin temperature were not static but changed dynamically depending on the thermoregulatory state (example see Fig. 1C, D): during a very low level of sympathetic vasoconstrictor activity (intense whole-body warming) only minimal side differences could be detected in all patients. The largest side differences were found at a high to intermediate level of sympathetic activity. 3.3. Maximal skin temperature side differences during thermoregulatory cycle The maximal skin temperature difference between both sides that occurred during modulation of sympathetic vasoconstrictor activity by controlled thermoregulation was
24
G. Wasner et al. / Pain 98 (2002) 19–26
determined for each patient and for each of the controls. Independent of whether the affected limb was warmer or colder than the contralateral side, the absolute value of the difference was used for further analysis. Using this approach the mean maximal side difference of the CRPS patients was DTmax ¼ 4:5 ^ 0:68C (mean ^ SEM; median 3.88C; range 1.1–10.48C) at a contralateral temperature of 29.7 ^ 0.68C. In contrast, patients with extremity pain of other origin showed a maximal difference of DTmax ¼ 1:0 ^ 0:28C (median 0.78C; range 0–2.58C). Similar, healthy controls had a maximal difference of DTmax ¼ 1:3 ^ 0:18C (median 1.28C; range 0.2–2.28C). The 95% confidential value was 2.28C. Maximal temperature side differences during controlled thermoregulation were on average significantly greater in CRPS patients in comparison with patients with extremity pain of other origin and healthy controls ðP , 0:001Þ. 3.4. Sensitivity and specificity of temperature side differences in CRPS To determine the validity of skin temperature asymmetries in diagnosis of CRPS sensitivity and specificity values were calculated. According to the 95% confidential value of skin temperature measurements in healthy controls, side differences of DTrest , 2:08C are considered to be normal under resting conditions (room temperature 22–248C, resting in supine position for 30 min). Based on these values 17 out of 25 CRPS patients were false negative and none of the patients with limb pain of other origin was false positive. This results in a poor sensitivity of 32% and a specificity of 100%. Using the maximal asymmetries in skin temperature during the thermoregulatory cycle as reference value normal maximal side differences are DTmax , 2:28C (calculated from the healthy control group). Under these conditions only six out of 25 CRPS patients were false negative and one of the control patients was false positive. This results in a sensitivity of 76% and a specificity of 93%. The difference in sensitivity under both conditions is due to an increase in skin temperature side differences in CRPS patients during controlled modulation of sympathetic vasoconstrictor activity in comparison to values under resting conditions (Fig. 2). 4. Discussion From the present experiments several findings can be summarized: 1. Under resting conditions (supine position for 30 min, room temperature 22–248C) mean side differences in temperature in the digits pads were significantly different between CRPS I patients (DTrest ¼ 2:1 ^ 0:58C) and controls (DTrest ¼ 0:6 ^ 0:38C in patients with chronic painful limbs of other origin and DTrest ¼ 0:6 ^ 0:18C
in healthy subjects). However, concerning the validity of skin temperature side differences in diagnosis of CRPS, sensitivity was only 32%, whereas specificity was 100%. 2. During controlled changes of environmental temperature (controlled thermoregulation with a thermal suit) side differences in skin temperature between the affected and unaffected extremity were found to be typical features in CRPS I. Three different regulation types could be distinguished: a ‘warm’ regulation type with higher cutaneous temperature, ‘cold’ regulation type with ‘lower’ cutaneous temperature and an ‘intermediate’ type of regulation with a partially warmer and partially colder affected limb (Wasner et al., 2001). However, these side differences in cutaneous regulation are not static during the thermoregulatory cycle. When sympathetic vasoconstrictor activity is low or absent (intense experimental whole-body warming or warm environmental temperature and relaxing atmosphere etc.) no significant differences are detectable whereas they are most pronounced during a high to intermediate level of sympathetic activity. 3. Therefore, the maximal skin temperature difference that occurred during the thermoregulatory cycle was used as a
Fig. 2. Individual skin temperature asymmetries in 25 CRPS I patients under resting conditions (ordinate: DTrest) and during controlled thermoregulation (abscissa: DTmax) Values in 8C, temperature of contralateral extremity as reference. To each patient a filled circle is dedicated. Horizontal dotted lines indicate 95% confidential value for temperature side differences during rest and vertical dotted lines for maximal asymmetries during thermoregulatory cycle. Patient no. 7 (see Table 1) is marked as an example (*): under resting conditions skin temperature of the affected side was 1.38C warmer (ordinate) than the contralateral side. This value is in between the 95% confidential interval (filled circle in between horizontal lines) indicating that this temperature asymmetry is not sensitive for CRPS. Maximal skin temperature difference increased to 10.48C (abscissa) during controlled thermoregulation. This value is outside the 95% confidential interval (filled circle outside vertical lines) indicating a sensitive value for the diagnosis of CRPS. Note that 17 patients ( ¼ 68%) are in between horizontal lines and only six ( ¼ 24%) in between vertical lines. This results in a sensitivity of 32% under resting conditions and of 76% during alterations of sympathetic activity for unilateral vascular disturbances in CRPS.
G. Wasner et al. / Pain 98 (2002) 19–26
descriptive measure. Using this parameter patients with CRPS showed on average maximal side differences of DTmax ¼ 4:5 ^ 0:68C. In contrast, in controls only small maximal side difference values were measured on average (DTmax ¼ 1:0 ^ 0:28C in patients with other chronic limb pain and DTmax ¼ 1:3 ^ 0:18C in healthy volunteers). From these data a normal maximal side difference value of DTmax , 2:28C was calculated resulting in a sensitivity of 76% and a specificity of 93% for skin temperature side differences in CRPS diagnosis.
4.1. Skin temperature side differences – a diagnostic tool for CRPS I? So far CRPS I is a pure clinical diagnosis. Patients with extremity pain of other origin may meet some of the clinical criteria and may be included under the umbrella category of CRPS I (Galer et al., 1998). Therefore, it is important to find objective laboratory tests to define CRPS in order to distinguish CRPS from other extremity pain syndromes. It is suggested that inclusion of autonomic disturbances would enhance validity (Bruehl et al., 1999; Harden et al., 1999). It was demonstrated that autonomic testing including investigations of vasomotor and sudomotor functions are of great importance for diagnosing CRPS (Chelimsky et al., 1995; Sandroni et al., 1998). Unilateral vascular disturbances is a typical clinical feature in CRPS (Bonica, 1990; Bej and Schwartzman, 1991). Bruehl et al. (1996) performed thermographic studies in CRPS patients and non-CRPS patients with extremity pain by analyzing temperature asymmetries under baseline conditions (sitting position for 20 min, room temperature 208C). They could discriminate between the two groups with a sensitivity of 68% and a specificity of 67% by using an asymmetry cutoff of 0.68C. Similar results were obtained by using stress infra-red telethermography in the diagnosis of CRPS with a high sensitivity and specificity (Gulevich et al., 1997). Also thermographic recordings of many sites at the limbs increased the probability of detecting significant side differences (Chelimsky et al., 1995; Birklein et al., 1997; Sandroni et al., 1998). Our results also showed significant mean temperature asymmetries in the digits pads between CRPS patients and controls under resting conditions. Mean values of temperature side differences in healthy subjects were very similar to normal values in distal extremities described by Uematsu et al. (1988) in a thermography study. However, concerning median and 95% confidential values temperature side differences up to 2.08C has to be estimated as normal in the present study. This results in a low sensitivity of 32%, because temperature asymmetries in most CRPS patients are below that value under resting conditions. On the other hand, none of the patients with extremity pain of other origin had temperature side differences beyond 2.08C resulting in a specificity of 100%. The different results
25
between the study of Bruehl et al. (1996) and the present data might be due to distinct techniques that were used: computerized thermography is a very sensitive method that provides more exact values than thermometry. However, thermography is restricted to centres that have access to the method, whereas infra-red thermometers are easy to handle and are available for practical use. The novel approach of using maximal skin temperature asymmetries during alterations of cutaneous sympathetic nerve activity by changing environmental temperature with a thermal suit revealed that skin temperature asymmetries are not static (Fig. 1). This is caused by thermoregulatory changes of spontaneous sympathetic cutaneous vasoconstrictor activity as demonstrated by the present data. Because healthy controls show nearly constant temperature side differences, sensitivity of measurements increased to 76% when maximal thermal asymmetries during controlled thermoregulation were considered. Accordingly, Low et al. (1994) pointed out that these unstable temperatures of the affected limb are a typical sign in CRPS. 4.2. Other disorders with unilateral vascular disturbances There are other disorders that are characterized by unilateral vascular disturbances that have been considered in the differential diagnosis of CRPS. These are all kinds of inflammations or infections (e.g. rheumatism and phlegmones) with a warm affected limb. Secondly, unilateral arterial or venous occlusive diseases might show a unilateral cold or warm limb and high temperature differences between the affected and healthy side. Thirdly, psychiatric artefact syndromes that are characterized by repetitive artificial occlusion of the blood supply to one limb might induce secondary structural vascular changes. Fourthly, increased blood flow and warm skin is found in the Charcot joint as a complication of unclear cause in a diabetic foot. 4.3. Clinical implications The present data show that skin temperature side differences in CRPS are not static but changed dynamically depending on sympathetic vasoconstrictor activity. Taken this into account, skin temperature asymmetries can be used as an additional tool in diagnosis of CRPS with high sensitivity and specificity. The applied technique of controlled alterations of skin sympathetic vasoconstrictor activity by use of a thermal suit is reserved to special centres, however, the following clinical implications are derived from the present study: since ongoing sympathetic activity may change during minutes or hours skin temperature differences need to be measured several times to obtain a full picture of vasomotor disturbances (Baron and Maier, 1996). Such multiple measurements are suggested to increase the opportunity to assess the maximal asymmetry. The detection of dynamic changes in skin temperature asymmetry in combination with a maximal side difference
26
G. Wasner et al. / Pain 98 (2002) 19–26
.2.28C is estimated to be a specific and sensitive parameter contributing to the diagnosis of CRPS. To support alterations of sympathetic activity by thermoregulatory stimuli without using a thermal suit one might warm the patient with blankets (sympathetic inhibition) and cool the patient by removing his clothes (sympathetic activation). Under both conditions the patient should be in a lying position: the affected and contralateral distal extremity must not be covered and skin temperature should be recorded for a longer period (e.g. 1 h). Alternatively, cutaneous temperature could be monitored continuously for several hours with small temperature sensors to demonstrate typical dynamic temperature changes of the affected limb in CRPS patients (Maier, 2000). Although, the symptom of skin temperature asymmetries cannot decide on the diagnosis of CRPS alone, it can be utilized as a bedside test that supplements other clinical signs and additive investigations. Acknowledgements We wish to thank Beatrice Luig and Nicoleta Blunck for excellent technical assistance. This work was supported by the Deutsche Forschungsgemeinschaft (DFG Ba 1921/1-1). References Baron R, Maier C. Reflex sympathetic dystrophy: skin blood flow, sympathetic vasoconstrictor reflexes and pain before and after surgical sympathectomy. Pain 1996;67:317–326. Baron R, Blumberg H, Ja¨ nig W. Clinical characteristics of patients with complex regional pain syndromes in Germany with special emphasis on vasomotor function. In: Ja¨ nig W, Stanton-Hicks MS, editors. Reflex sympathetic dystrophy – a reappraisal, Progress in Pain Research and Management, vol. 6. Seattle, WA: IASP Press, 1996. Baron R, Levine JD, Fields HL. Causalgia and reflex sympathetic dystrophy: does the sympathetic nervous system contribute to the generation of pain? Muscle Nerve 1999;22:678–695. Bej MD, Schwartzman RJ. Abnormalities of cutaneous blood flow regulation in patients with reflex sympathetic dystrophy as measured by laser Doppler fluxmetry. Arch Neurol 1991;48:912–915. Bini G, Hagbarth KE, Hynninen P, Wallin BG. Thermoregulatory and rhythm-generating mechanisms governing the sudomotor and vasoconstrictor outflow in human cutaneous nerves. J Physiol (Lond) 1980;306:537–552. Birklein F, Sittl R, Spitzer A, Claus D, Neundo¨ rfer B, Handwerker HO. Sudomotor function in sympathetic reflex dystrophy. Pain 1997;69:49– 54. Blumberg H, Ja¨ nig W. Clinical manifestation of reflex sympathetic dystrophy and sympathetically maintained pain. In: Wall PD, Melzack R, editors. Textbook of Pain, 3rd ed. London: Churchill Livingstone, 1994. pp. 685–697. Bonica JJ. In: Bonica JJ, editor. Causalgia and other reflex sympathetic dystrophies, The management of pain, vol. 2. Philadelphia, PA: Lea and Febiger, 1990. pp. 220–253. Bruehl S, Lubenow TR, Nath H, Ivankovich O. Validation of thermography in the diagnosis of reflex sympathetic dystrophy. Clin J Pain 1996;12:316–325. Bruehl S, Harden RN, Galer BS, Saltz S, Bertram M, Backonja M, Gayles
R, Rudin N, Bhugra MK, Stanton-Hicks M. External validation of IASP diagnostic criteria for Complex Regional Pain Syndrome and proposed research diagnostic criteria. International Association for the Study of Pain. Pain 1999;81:147–154. Chelimsky TC, Low PA, Naessens JM, Wilson PR, Amadio PC, O‘Brien PC. Value of autonomic testing in reflex sympathetic dystrophy. Mayo Clin Proc 1995;70:1029–1040. Evans JA. Reflex sympathetic dystrophy. Surg Clin North Am 1946;26:435–448. Galer BS, Bruehl S, Harden RN. IASP diagnostic criteria for complex regional pain syndrome: a preliminary empirical validation study. International Association for the Study of Pain. Clin J Pain 1998;14:48– 54. Gulevich SJ, Conwell TD, Lane J, Lockwood B, Schwettmann RS, Rosenberg N, Goldman LB. Stress infrared telethermography is useful in the diagnosis of complex regional pain syndrome, type I (formerly reflex sympathetic dystrophy). Clin J Pain 1997;13:50–59. Ha¨ bler HJ, Ja¨ nig W, Krummel M, Peters OA. Reflex patterns in postganglionic neurons supplying skin and skeletal muscle of the rat hindlimb. J Neurophysiol 1994;72:2222–2236. Ha¨ bler HJ, Stegmann JU, Timmermann L, Ja¨ nig W. Functional evidence for the differential control of superficial and deep blood vessels by sympathetic vasoconstrictor and primary afferent vasodilator fibres in rat hairless skin. Exp Brain Res 1998;118:230–234. Harden RN, Bruehl S, Galer BS, Saltz S, Bertram M, Backonja M, Gayles R, Rudin N, Bhugra MK, Stanton-Hicks M. Complex regional pain syndrome: are the IASP diagnostic criteria valid and sufficiently comprehensive? Pain 1999;83:211–219. Low PA, Amadio PC, Wilson PR, McManis PG, Willner CL. Laboratory findings in reflex sympathetic dystrophy: a preliminary report. Clin J Pain 1994;10:235–239. Maier C. Sympathisch unterhaltener Schmerz (SMP) bei Patienten mit CRPS – Diagnose und therapeutische Implikationen. Akt Neurol 2000;27:S83. Merskey H, Bogduk N. Classification of chronic pain: descriptions of chronic pain syndromes and definition of terms, Seattle, WA: IASP Press, 1995. Sandroni P, Low PA, Ferrer T, Opfer-Gehrking TL, Willner CL, Wilson PR. Complex regional pain syndrome I (CRPS I): prospective study and laboratory evaluation. Clin J Pain 1998;14:282–289. Schwartzman RJ, McLellan TL. Reflex sympathetic dystrophy. A review. Arch Neurol 1987;44:555–561. Sherman RA, Karstetter KW, Damiano M, Evans CB. Stability of temperature asymmetries in reflex sympathetic dystrophy over time and changes in pain. Clin J Pain 1994;10:71–77. Stanton-Hicks M, Janig W, Hassenbusch S, Haddox JD, Boas R, Wilson P. Reflex sympathetic dystrophy: changing concepts and taxonomy. Pain 1995;63:127–133. Uematsu S, Edwin DH, Jankel WR, Kozikowski J, Trattner M. Quantification of thermal asymmetry. Part 1: normal values and reproducibility. J Neurosurg 1988;69:552–555. Veldman PH, Reynen HM, Arntz IE, Goris RJ. Signs and symptoms of reflex sympathetic dystrophy: prospective study of 829 patients. Lancet 1993;342:1012–1016. Wasner G, Backonja MM, Baron R. Traumatic neuralgias: complex regional pain syndromes (reflex sympathetic dystrophy and causalgia): clinical characteristics, pathophysiological mechanisms and therapy. Neurol Clin 1998;16:851–868. Wasner G, Heckmann K, Maier C, Baron R. Vascular abnormalities in acute reflex sympathetic dystrophy (CRPS I): complete inhibition of sympathetic nerve activity with recovery. Arch Neurol 1999;56:613– 620. Wasner G, Schattschneider J, Heckmann K, Maier C, Baron R. Vascular abnormalities in reflex sympathetic dystrophy (CRPS I): mechanisms and diagnostic value. Brain 2001;124:587–599.
Skin temperature side differences – a diagnostic tool for CRPS? Gunnar Wasner, Jo¨rn Schattschneider, Ralf Baron* Klinik fu¨r Neurologie, Universita¨tsklinikum Kiel der Christian-Albrechts-Universita¨t, Niemannsweg 147, 24105 Kiel, Germany Received 6 July 2001; received in revised form 29 October 2001
Abstract Complex regional pain syndrome type I (CRPS I) is a chronic painful disease of one extremity that may develop as a disproportionate consequence of a trauma affecting the limbs without overt nerve injury. It is clinically characterized by sensory, motor and autonomic symptoms including vascular abnormalities. Previously, we have reported that pathophysiological alterations of the ongoing sympathetic activity play a crucial role in vasomotor disturbances (Brain 124 (2001) 587). As a companion article, the aim of this study was to evaluate the diagnostic value of skin temperature side differences in consideration of the spontaneous sympathetic vasoconstrictor activity. Twentyfive patients with CRPS I were studied. Fifteen patients with painful limbs of other origin and 20 healthy individuals served as controls. Controlled thermoregulation was performed to change cutaneous sympathetic vasoconstrictor activity by the use of a thermal suit: skin sympathetic vasoconstrictor neurones were activated by whole-body cooling and nerve activity was abolished by whole-body warming. Skin temperature at the affected and unaffected limbs (infra-red thermometry) was measured under resting conditions and continuously monitored during controlled modulation of sympathetic activity. The results showed only minor skin temperature asymmetries between both limbs under resting conditions in most CRPS patients. However, during controlled thermoregulation temperature differences between both sides increased dynamically and were most prominent at a high to medium level of vasoconstrictor activity. In both control groups, there were only minor side differences in temperature both in rest and during thermoregulatory changes of sympathetic activity. When comparing the diagnostic value of skin temperature asymmetries in CRPS I, sensitivity was only 32% under resting conditions, but increased up to 76% during controlled alteration of sympathetic activity. Specificity was 100% at rest and 93% at controlled thermoregulation. We concluded that the degree of unilateral vascular disturbances in CRPS I depends critically on spontaneous sympathetic activity. Taking this into consideration, skin temperature differences in the distal limbs are capable of reliably distinguishing CRPS I from other extremity pain syndromes with high sensitivity and specificity. q 2002 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. Keywords: Complex regional pain syndrome; Sympathetic nervous system; Skin temperature; Thermoregulation
1. Introduction Complex regional pain syndrome type I (CRPS I, formerly reflex sympathetic dystrophy) is a neuropathic pain condition that may develop as a disproportionate consequence of a trauma affecting the limbs like bone fractures, sprains or skin lesions (Schwartzman and McLellan, 1987; Veldman et al., 1993; Baron et al., 1996; Wasner et al., 1998). CRPS I is distinguished from CRPS II (causalgia) in which a partial lesion of a peripheral nerve is necessary for the diagnosis (Merskey and Bogduk, 1995; StantonHicks et al., 1995). The clinical features are spontaneous pain, hyperalgesia, impairment of motor function, swelling and autonomic abnormalities in a single extremity. The
* Corresponding author. Tel.: 149-431-597-2633; fax: 149-431-5972712. E-mail address: [email protected] (R. Baron).
symptoms tend to spread distally without confining to the innervation zone of an individual nerve or root. Autonomic signs include abnormalities in skin vascular and sudomotor functions that are due to unilateral disturbances of the sympathetic innervation (Wasner et al., 1999, 2001). Clinically, the measure of skin temperature differences between affected and unaffected side has repeatedly suggested to be an important diagnostic test for CRPS (Blumberg and Ja¨nig, 1994; Bruehl et al., 1996). This is of particular relevance to distinguish this entity from extremity pain syndromes of different etiology. Recently, Harden et al. (1999) pointed out that internal validity of the International Association for the Study of Pain (IASP)/CRPS criteria can be improved by paying more attention to vasomotor signs. However, one dilemma seems to be the poor intra- and interindividual reproducibility of temperature asymmetries in CRPS when measured at different times (Sherman et al., 1994).
0304-3959/02/$20.00 q 2002 International Association for the Study of Pain. Published by Elsevier Science B.V. All rights reserved. PII: S 0304-395 9(01)00470-5
20
G. Wasner et al. / Pain 98 (2002) 19–26
Table 1 Clinical characteristics of CRPS patients a No.
Age/Sex
Location
Inciting event
Duration (months)
DTrest (8C)
Contra Trest (8C)
DTmax (8C)
Contra Tmax (8C)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
62/f 62/m 43/f 44/f 49/f 51/f 55/m 56/f 27/f 34/f 40/f 50/f 66/m 63/f 57/m 40/m 52/f 64/f 34/f 49/f 33/f 30/m 39/f 43/f 27/m
L upper limb R upper limb L upper limb L upper limb L lower limb R upper limb L lower limb L upper limb R lower limb R upper limb R upper limb L lower limb L upper limb R upper limb R upper limb R upper limb R upper limb R upper limb R upper limb R lower limb R upper limb R lower limb L upper limb R lower limb L lower limb
Wrist fracture Wrist fracture Metacarpal fracture Colles fracture Ankle joint distortion No obvious trauma Strain Colles fracture No obvious trauma Post-elbow surgery Strain Lower leg fracture Colles fracture Tendosynovitis Shoulder torn tendon No obvious trauma Colles fracture Elbow dislocation Tendosynovitis No obvious trauma Post-tendosynovitis Metacarpal fracture No obvious trauma Ankle joint fracture Sprain
1.5 2.5 20 12 35 48 5 14 48 48 3 25 2 5 2 10 0.5 3.5 18 4.5 7 10 15 7 2
20.4 20.5 2.0 7.5 29.5 21.9 1.3 26.4 20.3 22.2 0.7 22.1 0.6 1.8 2.1 1.8 0.8 0.4 21.9 4.0 0.4 20.3 0 0.5 1.9
35.3 32.7 31.4 27.7 33.9 29.5 21.8 33.5 24.4 23.7 34.1 26.5 33.6 26.1 29.4 24.8 34.6 34.8 23.5 28 27.8 23.3 22 26.9 21.7
1.2 2.7 2.9 6.4 29.5 22.9 10.4 25.2 23.0 25.1 1.1 23.1 21.4 1.8 6.9 21.7 10.2 3.8 24.2 7.0 26.1 24.9 2.4 21.7 7.4
32 28 29 28 34 31 25 32 29 32 28 32 32 26 25 34 25 27 30 27 34 35 30 32 25
a
f, female; m, male; R, right; L, left.
Skin temperature is under the control of sympathetic vasoconstrictor neurones. Their spontaneous nerve activity depends critically on the thermoregulatory state and may change rapidly inducing alterations in skin temperature as demonstrated in animal experiments and human microneurographic recordings (Bini et al., 1980; Ha¨ bler et al., 1994, 1998). So far, skin temperature measurements when used as diagnostic criterion for CRPS were performed as a single measurement under resting conditions, e.g. defined as resting in supine position, room temperature and acclimatization for 30 min. However, the skin temperature varies dynamically and continuously depending on the thermoregulatory state of the patient as described above. Therefore, the question arises whether one single measurement of skin temperature side differences at resting conditions really reflects the maximal possible difference that occurs during different states of thermoregulation. The present investigation aims at several questions: the skin temperature dynamic and, in particular, the variation of temperature differences between both sides that occurs during different thermoregulatory states will be investigated in patients with CRPS and in controls. If skin temperature side differences are used as a diagnostic test for CRPS, is it possible to increase the sensitivity of this test by using the maximal side difference in skin temperature? As a companion paper to a previous published article data of the present study were collected from the same investigations (Wasner et al., 2001).
2. Methods 2.1. Patients and controls 2.1.1. Patients with CRPS I The study was performed on 25 patients, 18 women and seven men (mean age 47 years, range 27–66), with the diagnosis of unilateral CRPS I who were referred to the Interdisciplinary Pain Center of the University Clinic of Kiel from 1995 to 2000 (Table 1). The upper extremity was affected in 17 cases and the lower in eight cases. CRPS I was diagnosed according to the criteria defined by Evans (1946) and to the novel clinical criteria defined by the IASP (Merskey and Bogduk, 1995; Stanton-Hicks et al., 1995). All patients were clinically characterized by spontaneous pain (at least in their medical history) and evoked pains (e.g. deep hyperalgesia, mechanical allodynia) that were distally generalized and that were not restricted to an innervation territory of any peripheral nerve. In all cases, pain was increased by movements of the affected limb and patients had at least one symptom of motor dysfunction like impairment of muscle strength, tremor or dystonia. Furthermore, there was or there had been evidence of at least one autonomic involvement like edema, skin temperature asymmetries or sweating abnormalities. Patients who had additional diseases that might contribute to neuropathic pain, e.g. diabetes, were excluded. By this procedure the incidence of false positive diagnoses was minimized. Additional investigations like radio-
G. Wasner et al. / Pain 98 (2002) 19–26
21
Table 2 Clinical characteristics of patients with chronic painful limbs of other origin a No. Age/sex Location 1
38/m
2 3 4 5 6
47/m 39/f 57/m 49/f 56/m
7 8 9 10
38/m 42/f 32/f 18/f
11 12 13
53/f 49/m 42/m
14 15
36/f 50/f
a
Disease
Duration (months) DTrest (8C) Contra Trest (8C) DTmax (8C) Contra Tmax (8C)
L upper limb Traumatic neuralgia of ulnar nerve and 33 the internal cutaneous nerve R upper limb Lunatomalacia 0.5 R lower limb Radiculopathy 18 L upper limb Central pain after thalamic infarction 6 R lower limb Achillodynia 103 L upper limb Ischemic nerve lesion at wrist and 12 dorsal hand L upper limb Pseudoarthrosis in the wrist 10 R lower limb Traumatic neuralgia of peroneal nerve 17 L upper limb Carpal tunnel syndrome 6 R upper limb Traumatic nerve lesion at wrist and 48 ulnar hand R upper limb Traumatic neuralgia of radial nerve 25 L upper limb Neuralgia of ulnar nerve 23 L upper limb Severe shoulder trauma with lesion of 7 brachial plexus R upper limb Brachial plexopathy 38 L upper limb Traumatic nerve lesion at thumb 6
20.7
27.3
0.7
34
0.4 20.3 20.4 0.2 0.5
32 25.4 34.7 25.1 33
0.7 20.7 0.6 1.2 0.6
27 25 31 27 26
21.2 1.1 21.6 20.1
33.7 26 23.7 30.2
1.2 2.5 1.6 1.1
31 29 27 27
0.2 21.8 20.3
34.4 25 34.1
0 1.15 0.7
25 30
0 0.1
32.5 24.2
1.5 1.8
25 30
f, female; m, male; R, right; L, left.
graphy and three phase bone scan was performed. The clinical data are summarized in Table 1. 2.1.2. Patients with chronic painful limbs of other origin Fifteen patients, eight women and seven men (mean age 42 years, range 18–57), with chronic pain of one limb of other origin (patients did not meet the criteria described above) served as a control group (Table 2). Although, the patients are suffering from different diseases, the following reasons argue against the possibility, that they have early CRPS I or II. (1) In all patients with nerve injury the pain was restricted to the innervation territory of the affected nerve without tendency to spread beyond. Therefore, they were classified as having posttraumatic neuralgia (Baron et al., 1999). (2) This clinical picture was stable for several months in the control patients, therefore it was unlikely that symptoms begin to generalize like it would be expected in early CRPS II. (3) No control patient demonstrated trophic disturbances. (4) None of the patients suffered from edema and there was no history of autonomic abnormalities. All patients underwent a general physical and neurological examination. 2.1.3. Healthy controls Twenty healthy subjects (11 women and nine men) served as a control group (mean age 27 years, range 23–45). 2.1.4. Procedures The tests were performed between 3 and 6 p.m. The patients and controls were tested in supine position (room temperature 22–248C). None of the control subjects or patients was on drugs affecting vascular function. Patients suffering from cardiovascular disorders were excluded from
the study. Aims and procedures of the study were explained to all subjects according to the Helsinki Declaration. All individuals gave their informed consent to participate in the study, which was approved by the local ethical committee. The procedures followed were in accordance with institutional guidelines. 2.2. Investigations and measurements 2.2.1. Skin temperature side differences under resting conditions After resting in supine position (room temperature 22– 248C) for 30 min skin temperature was measured bilaterally at all finger or toe tips with infra-red thermometers. For each extremity the mean value of all finger and toe tips, was calculated. The absolute side difference in skin temperature under resting conditions between both extremities was determined by the following formula: Side difference in skin temperatures in patients : DTrest ¼ uskin temperature on affected side 2 skin temperature on unaffected sideu Side difference in skin temperature in healthy controls : DTrest ¼ uskin temperature on right side 2 skin temperature on left sideu 2.2.2. Thermoregulatory cycle Skin sympathetic vasoconstrictor activity is under thermoregulatory control. Controlled thermoregulatory reflexes
22
G. Wasner et al. / Pain 98 (2002) 19–26
were performed to induce a physiological tonic change of sympathetic skin nerve activity. This was achieved by changing environmental temperature with the help of a thermal suit. The subject was lying in a cotton suit supplied by tubes, in which circulating water of 12 and 508C (inflow temperature) was used to cool or warm the whole body, respectively. Both hands and feet were not covered by the suit. Wholebody cooling is the most effective stimulus to induce a massive tonic activation of cutaneous vasoconstrictor neurons as demonstrated in microneurographical recordings (Bini et al., 1980); whole-body warming leads to a complete inhibition of this activity. Skin temperature measurements (bilaterally at all finger or toe tips at 5 min intervals with infra-red thermometers, calculation of mean values for each extremity) were performed during a complete thermoregulatory cycle, i.e. during the entire spectrum of sympathetic vasoconstrictor activity. The temperature on the contralateral not affected extremity (or right side in healthy controls) was taken as a reference to estimate the level of sympathetic activity. After the patients had put on the thermal suit followed by a resting period lasting for 30 min, first, whole-body cooling was performed to induce maximal vasoconstrictor activity. The cooling session was continued until the skin temperature on the unaffected side (or right side in healthy controls) was close to room temperature (i.e. 258C). Thereafter, whole-body warming was performed until the skin temperature on the unaffected side (or right side in healthy controls) was close to body core temperature (i.e. 358C) in order to induce maximal inhibition of sympathetic activity. All data that were used in the evaluation were collected during the whole-body warming period after whole-body cooling was performed. Whole-body cooling was used to establish a baseline allowing comparability of all patients.
2.3. Data acquisition and statistical analysis For comparison of non-parametric data between CRPS patients, patients with chronic painful limbs of other origin and healthy controls the Mann–Whitney U test was used. Skin temperature side differences under resting conditions and maximal asymmetries during controlled thermoregulation were compared between the groups. Data are given as mean ^ standard deviation and median. P-values of ,0.05 were regarded as statistically significant. The 95% confidential values were calculated to determinate specificity and sensitivity of temperature side differences at rest and during the thermoregulatory cycle. 3. Results 3.1. Skin temperature side differences under resting conditions In CRPS patients skin temperature side differences at rest between the affected and healthy limbs were on average DTrest ¼ 2:1 ^ 0:58C (mean ^ SEM; median 1.88C; range 0–9.48C) after resting in supine position for 30 min (room temperature 22–248C) (Table 1). In patients with extremity pain of other origin average side differences were DTrest ¼ 0:6 ^ 0:38C (median 0.48C; range 0–1.88C) (Table 2). Healthy controls had side differences of DTrest ¼ 0:6 ^ 0:18C (median 0.58C; range 0–2.18C). The 95% confidential value was 2.08C. Temperature side differences during rest were on average significantly greater in CRPS patients in comparison to patients with extremity pain of other origin and healthy controls ðP , 0:01Þ. 3.2. Skin temperature during thermoregulatory cycle
2.2.3. Maximal skin temperature side difference during the thermoregulatory cycle The absolute maximal side difference in skin temperature that occurred during the whole thermoregulatory cycle was determined using the following formula: Maximal side difference in skin temperature in patients : DTmax ¼ uskin temperature on the affected side 2 skin temperature on unaffected sideu
Maximal side difference in skin temperature in healthy controls :
DTmax ¼ uskin temperature on right side 2 skin temperature on left sideu
3.2.1. Healthy controls and patients with extremity pain of other origin Whole-body cooling induced a symmetrical decrease in skin temperature in both limbs due to maximal tonic activation of cutaneous sympathetic vasoconstrictor neurons. Thereafter, skin temperature increased bilaterally during whole-body warming, because of inhibition of cutaneous sympathetic vasoconstrictor activity (Fig. 1A, B). The regulation pattern was identical in the healthy control group and in the group of patients with extremity pain of other origin. Only small side differences of skin temperature occurred during the entire cycle (Table 2). 3.2.2. Patients with CRPS I Patients with CRPS I were characterized by asymmetrical changes of skin temperature during the thermoregulatory cycle. In the unaffected distal extremity whole-body cooling led to an immediate sustained decrease in skin temperature very similar to the situation in healthy controls. On the
G. Wasner et al. / Pain 98 (2002) 19–26
23
Fig. 1. Characteristics of skin temperature as a measure of cutaneous sympathetic vasoconstrictor activity at the fingers of both hands in one healthy control subject (A,B) and in a patient with CRPS I (C,D) during a controlled thermoregulatory cycle (controlled alterations in cutaneous sympathetic activity). Controlled thermoregulatory changes (whole-body cooling and warming) were applied by means of a thermal suit to change environmental temperature in a standardized way and reflexly alter skin sympathetic vasoconstrictor activity. The subject was lying in a suit supplied by tubes, in which running water of 12 and 508C (inflow temperature) was used to cool or warm the whole body, respectively. During the experiment, the skin temperature of the fingers of both hands was monitored at regular intervals. (A) Temperature of the right and left hand in a healthy subject during the thermoregulatory cycle. (C) Affected extremity (CRPS) and healthy side (contralateral) in a CRPS patient. Side differences in skin temperature of the fingers of both hands in one healthy control subject (B) and in a CRPS patient (D) during a controlled thermoregulatory cycle. The arrow indicates beginning of whole-body warming. Data that were used in the evaluation were collected during this period. Maximal skin temperature differences during whole-body warming are indicated by the vertical dotted lines. Same subjects as in (A,C).
affected side, three different patterns of regulation were observed. 1. Patients with a ‘warm’ regulation type showed higher cutaneous temperatures in the affected limb than in the contralateral limb during the entire spectrum of sympathetic vasoconstrictor activity (Fig. 1C, D). This regulation type was found in 11 patients. 2. In patients with an ‘intermediate’ type, the direction of the temperature side difference changed during the thermoregulatory cycle. In some patients the affected side was warmer during increased sympathetic activity and colder during decreased activity and vice versa. Seven patients were characterized by regulation type. 3. In patients with a ‘cold’ regulation type, vasoconstriction was more intense on the affected side during whole-body cooling. As a result, skin temperature in the affected limb was lower than in the unaffected side during the entire
thermoregulatory cycle. This regulation type was present in seven patients. Besides these differences all CRPS patients had the following factor in common: the differences in skin temperature were not static but changed dynamically depending on the thermoregulatory state (example see Fig. 1C, D): during a very low level of sympathetic vasoconstrictor activity (intense whole-body warming) only minimal side differences could be detected in all patients. The largest side differences were found at a high to intermediate level of sympathetic activity. 3.3. Maximal skin temperature side differences during thermoregulatory cycle The maximal skin temperature difference between both sides that occurred during modulation of sympathetic vasoconstrictor activity by controlled thermoregulation was
24
G. Wasner et al. / Pain 98 (2002) 19–26
determined for each patient and for each of the controls. Independent of whether the affected limb was warmer or colder than the contralateral side, the absolute value of the difference was used for further analysis. Using this approach the mean maximal side difference of the CRPS patients was DTmax ¼ 4:5 ^ 0:68C (mean ^ SEM; median 3.88C; range 1.1–10.48C) at a contralateral temperature of 29.7 ^ 0.68C. In contrast, patients with extremity pain of other origin showed a maximal difference of DTmax ¼ 1:0 ^ 0:28C (median 0.78C; range 0–2.58C). Similar, healthy controls had a maximal difference of DTmax ¼ 1:3 ^ 0:18C (median 1.28C; range 0.2–2.28C). The 95% confidential value was 2.28C. Maximal temperature side differences during controlled thermoregulation were on average significantly greater in CRPS patients in comparison with patients with extremity pain of other origin and healthy controls ðP , 0:001Þ. 3.4. Sensitivity and specificity of temperature side differences in CRPS To determine the validity of skin temperature asymmetries in diagnosis of CRPS sensitivity and specificity values were calculated. According to the 95% confidential value of skin temperature measurements in healthy controls, side differences of DTrest , 2:08C are considered to be normal under resting conditions (room temperature 22–248C, resting in supine position for 30 min). Based on these values 17 out of 25 CRPS patients were false negative and none of the patients with limb pain of other origin was false positive. This results in a poor sensitivity of 32% and a specificity of 100%. Using the maximal asymmetries in skin temperature during the thermoregulatory cycle as reference value normal maximal side differences are DTmax , 2:28C (calculated from the healthy control group). Under these conditions only six out of 25 CRPS patients were false negative and one of the control patients was false positive. This results in a sensitivity of 76% and a specificity of 93%. The difference in sensitivity under both conditions is due to an increase in skin temperature side differences in CRPS patients during controlled modulation of sympathetic vasoconstrictor activity in comparison to values under resting conditions (Fig. 2). 4. Discussion From the present experiments several findings can be summarized: 1. Under resting conditions (supine position for 30 min, room temperature 22–248C) mean side differences in temperature in the digits pads were significantly different between CRPS I patients (DTrest ¼ 2:1 ^ 0:58C) and controls (DTrest ¼ 0:6 ^ 0:38C in patients with chronic painful limbs of other origin and DTrest ¼ 0:6 ^ 0:18C
in healthy subjects). However, concerning the validity of skin temperature side differences in diagnosis of CRPS, sensitivity was only 32%, whereas specificity was 100%. 2. During controlled changes of environmental temperature (controlled thermoregulation with a thermal suit) side differences in skin temperature between the affected and unaffected extremity were found to be typical features in CRPS I. Three different regulation types could be distinguished: a ‘warm’ regulation type with higher cutaneous temperature, ‘cold’ regulation type with ‘lower’ cutaneous temperature and an ‘intermediate’ type of regulation with a partially warmer and partially colder affected limb (Wasner et al., 2001). However, these side differences in cutaneous regulation are not static during the thermoregulatory cycle. When sympathetic vasoconstrictor activity is low or absent (intense experimental whole-body warming or warm environmental temperature and relaxing atmosphere etc.) no significant differences are detectable whereas they are most pronounced during a high to intermediate level of sympathetic activity. 3. Therefore, the maximal skin temperature difference that occurred during the thermoregulatory cycle was used as a
Fig. 2. Individual skin temperature asymmetries in 25 CRPS I patients under resting conditions (ordinate: DTrest) and during controlled thermoregulation (abscissa: DTmax) Values in 8C, temperature of contralateral extremity as reference. To each patient a filled circle is dedicated. Horizontal dotted lines indicate 95% confidential value for temperature side differences during rest and vertical dotted lines for maximal asymmetries during thermoregulatory cycle. Patient no. 7 (see Table 1) is marked as an example (*): under resting conditions skin temperature of the affected side was 1.38C warmer (ordinate) than the contralateral side. This value is in between the 95% confidential interval (filled circle in between horizontal lines) indicating that this temperature asymmetry is not sensitive for CRPS. Maximal skin temperature difference increased to 10.48C (abscissa) during controlled thermoregulation. This value is outside the 95% confidential interval (filled circle outside vertical lines) indicating a sensitive value for the diagnosis of CRPS. Note that 17 patients ( ¼ 68%) are in between horizontal lines and only six ( ¼ 24%) in between vertical lines. This results in a sensitivity of 32% under resting conditions and of 76% during alterations of sympathetic activity for unilateral vascular disturbances in CRPS.
G. Wasner et al. / Pain 98 (2002) 19–26
descriptive measure. Using this parameter patients with CRPS showed on average maximal side differences of DTmax ¼ 4:5 ^ 0:68C. In contrast, in controls only small maximal side difference values were measured on average (DTmax ¼ 1:0 ^ 0:28C in patients with other chronic limb pain and DTmax ¼ 1:3 ^ 0:18C in healthy volunteers). From these data a normal maximal side difference value of DTmax , 2:28C was calculated resulting in a sensitivity of 76% and a specificity of 93% for skin temperature side differences in CRPS diagnosis.
4.1. Skin temperature side differences – a diagnostic tool for CRPS I? So far CRPS I is a pure clinical diagnosis. Patients with extremity pain of other origin may meet some of the clinical criteria and may be included under the umbrella category of CRPS I (Galer et al., 1998). Therefore, it is important to find objective laboratory tests to define CRPS in order to distinguish CRPS from other extremity pain syndromes. It is suggested that inclusion of autonomic disturbances would enhance validity (Bruehl et al., 1999; Harden et al., 1999). It was demonstrated that autonomic testing including investigations of vasomotor and sudomotor functions are of great importance for diagnosing CRPS (Chelimsky et al., 1995; Sandroni et al., 1998). Unilateral vascular disturbances is a typical clinical feature in CRPS (Bonica, 1990; Bej and Schwartzman, 1991). Bruehl et al. (1996) performed thermographic studies in CRPS patients and non-CRPS patients with extremity pain by analyzing temperature asymmetries under baseline conditions (sitting position for 20 min, room temperature 208C). They could discriminate between the two groups with a sensitivity of 68% and a specificity of 67% by using an asymmetry cutoff of 0.68C. Similar results were obtained by using stress infra-red telethermography in the diagnosis of CRPS with a high sensitivity and specificity (Gulevich et al., 1997). Also thermographic recordings of many sites at the limbs increased the probability of detecting significant side differences (Chelimsky et al., 1995; Birklein et al., 1997; Sandroni et al., 1998). Our results also showed significant mean temperature asymmetries in the digits pads between CRPS patients and controls under resting conditions. Mean values of temperature side differences in healthy subjects were very similar to normal values in distal extremities described by Uematsu et al. (1988) in a thermography study. However, concerning median and 95% confidential values temperature side differences up to 2.08C has to be estimated as normal in the present study. This results in a low sensitivity of 32%, because temperature asymmetries in most CRPS patients are below that value under resting conditions. On the other hand, none of the patients with extremity pain of other origin had temperature side differences beyond 2.08C resulting in a specificity of 100%. The different results
25
between the study of Bruehl et al. (1996) and the present data might be due to distinct techniques that were used: computerized thermography is a very sensitive method that provides more exact values than thermometry. However, thermography is restricted to centres that have access to the method, whereas infra-red thermometers are easy to handle and are available for practical use. The novel approach of using maximal skin temperature asymmetries during alterations of cutaneous sympathetic nerve activity by changing environmental temperature with a thermal suit revealed that skin temperature asymmetries are not static (Fig. 1). This is caused by thermoregulatory changes of spontaneous sympathetic cutaneous vasoconstrictor activity as demonstrated by the present data. Because healthy controls show nearly constant temperature side differences, sensitivity of measurements increased to 76% when maximal thermal asymmetries during controlled thermoregulation were considered. Accordingly, Low et al. (1994) pointed out that these unstable temperatures of the affected limb are a typical sign in CRPS. 4.2. Other disorders with unilateral vascular disturbances There are other disorders that are characterized by unilateral vascular disturbances that have been considered in the differential diagnosis of CRPS. These are all kinds of inflammations or infections (e.g. rheumatism and phlegmones) with a warm affected limb. Secondly, unilateral arterial or venous occlusive diseases might show a unilateral cold or warm limb and high temperature differences between the affected and healthy side. Thirdly, psychiatric artefact syndromes that are characterized by repetitive artificial occlusion of the blood supply to one limb might induce secondary structural vascular changes. Fourthly, increased blood flow and warm skin is found in the Charcot joint as a complication of unclear cause in a diabetic foot. 4.3. Clinical implications The present data show that skin temperature side differences in CRPS are not static but changed dynamically depending on sympathetic vasoconstrictor activity. Taken this into account, skin temperature asymmetries can be used as an additional tool in diagnosis of CRPS with high sensitivity and specificity. The applied technique of controlled alterations of skin sympathetic vasoconstrictor activity by use of a thermal suit is reserved to special centres, however, the following clinical implications are derived from the present study: since ongoing sympathetic activity may change during minutes or hours skin temperature differences need to be measured several times to obtain a full picture of vasomotor disturbances (Baron and Maier, 1996). Such multiple measurements are suggested to increase the opportunity to assess the maximal asymmetry. The detection of dynamic changes in skin temperature asymmetry in combination with a maximal side difference
26
G. Wasner et al. / Pain 98 (2002) 19–26
.2.28C is estimated to be a specific and sensitive parameter contributing to the diagnosis of CRPS. To support alterations of sympathetic activity by thermoregulatory stimuli without using a thermal suit one might warm the patient with blankets (sympathetic inhibition) and cool the patient by removing his clothes (sympathetic activation). Under both conditions the patient should be in a lying position: the affected and contralateral distal extremity must not be covered and skin temperature should be recorded for a longer period (e.g. 1 h). Alternatively, cutaneous temperature could be monitored continuously for several hours with small temperature sensors to demonstrate typical dynamic temperature changes of the affected limb in CRPS patients (Maier, 2000). Although, the symptom of skin temperature asymmetries cannot decide on the diagnosis of CRPS alone, it can be utilized as a bedside test that supplements other clinical signs and additive investigations. Acknowledgements We wish to thank Beatrice Luig and Nicoleta Blunck for excellent technical assistance. This work was supported by the Deutsche Forschungsgemeinschaft (DFG Ba 1921/1-1). References Baron R, Maier C. Reflex sympathetic dystrophy: skin blood flow, sympathetic vasoconstrictor reflexes and pain before and after surgical sympathectomy. Pain 1996;67:317–326. Baron R, Blumberg H, Ja¨ nig W. Clinical characteristics of patients with complex regional pain syndromes in Germany with special emphasis on vasomotor function. In: Ja¨ nig W, Stanton-Hicks MS, editors. Reflex sympathetic dystrophy – a reappraisal, Progress in Pain Research and Management, vol. 6. Seattle, WA: IASP Press, 1996. Baron R, Levine JD, Fields HL. Causalgia and reflex sympathetic dystrophy: does the sympathetic nervous system contribute to the generation of pain? Muscle Nerve 1999;22:678–695. Bej MD, Schwartzman RJ. Abnormalities of cutaneous blood flow regulation in patients with reflex sympathetic dystrophy as measured by laser Doppler fluxmetry. Arch Neurol 1991;48:912–915. Bini G, Hagbarth KE, Hynninen P, Wallin BG. Thermoregulatory and rhythm-generating mechanisms governing the sudomotor and vasoconstrictor outflow in human cutaneous nerves. J Physiol (Lond) 1980;306:537–552. Birklein F, Sittl R, Spitzer A, Claus D, Neundo¨ rfer B, Handwerker HO. Sudomotor function in sympathetic reflex dystrophy. Pain 1997;69:49– 54. Blumberg H, Ja¨ nig W. Clinical manifestation of reflex sympathetic dystrophy and sympathetically maintained pain. In: Wall PD, Melzack R, editors. Textbook of Pain, 3rd ed. London: Churchill Livingstone, 1994. pp. 685–697. Bonica JJ. In: Bonica JJ, editor. Causalgia and other reflex sympathetic dystrophies, The management of pain, vol. 2. Philadelphia, PA: Lea and Febiger, 1990. pp. 220–253. Bruehl S, Lubenow TR, Nath H, Ivankovich O. Validation of thermography in the diagnosis of reflex sympathetic dystrophy. Clin J Pain 1996;12:316–325. Bruehl S, Harden RN, Galer BS, Saltz S, Bertram M, Backonja M, Gayles
R, Rudin N, Bhugra MK, Stanton-Hicks M. External validation of IASP diagnostic criteria for Complex Regional Pain Syndrome and proposed research diagnostic criteria. International Association for the Study of Pain. Pain 1999;81:147–154. Chelimsky TC, Low PA, Naessens JM, Wilson PR, Amadio PC, O‘Brien PC. Value of autonomic testing in reflex sympathetic dystrophy. Mayo Clin Proc 1995;70:1029–1040. Evans JA. Reflex sympathetic dystrophy. Surg Clin North Am 1946;26:435–448. Galer BS, Bruehl S, Harden RN. IASP diagnostic criteria for complex regional pain syndrome: a preliminary empirical validation study. International Association for the Study of Pain. Clin J Pain 1998;14:48– 54. Gulevich SJ, Conwell TD, Lane J, Lockwood B, Schwettmann RS, Rosenberg N, Goldman LB. Stress infrared telethermography is useful in the diagnosis of complex regional pain syndrome, type I (formerly reflex sympathetic dystrophy). Clin J Pain 1997;13:50–59. Ha¨ bler HJ, Ja¨ nig W, Krummel M, Peters OA. Reflex patterns in postganglionic neurons supplying skin and skeletal muscle of the rat hindlimb. J Neurophysiol 1994;72:2222–2236. Ha¨ bler HJ, Stegmann JU, Timmermann L, Ja¨ nig W. Functional evidence for the differential control of superficial and deep blood vessels by sympathetic vasoconstrictor and primary afferent vasodilator fibres in rat hairless skin. Exp Brain Res 1998;118:230–234. Harden RN, Bruehl S, Galer BS, Saltz S, Bertram M, Backonja M, Gayles R, Rudin N, Bhugra MK, Stanton-Hicks M. Complex regional pain syndrome: are the IASP diagnostic criteria valid and sufficiently comprehensive? Pain 1999;83:211–219. Low PA, Amadio PC, Wilson PR, McManis PG, Willner CL. Laboratory findings in reflex sympathetic dystrophy: a preliminary report. Clin J Pain 1994;10:235–239. Maier C. Sympathisch unterhaltener Schmerz (SMP) bei Patienten mit CRPS – Diagnose und therapeutische Implikationen. Akt Neurol 2000;27:S83. Merskey H, Bogduk N. Classification of chronic pain: descriptions of chronic pain syndromes and definition of terms, Seattle, WA: IASP Press, 1995. Sandroni P, Low PA, Ferrer T, Opfer-Gehrking TL, Willner CL, Wilson PR. Complex regional pain syndrome I (CRPS I): prospective study and laboratory evaluation. Clin J Pain 1998;14:282–289. Schwartzman RJ, McLellan TL. Reflex sympathetic dystrophy. A review. Arch Neurol 1987;44:555–561. Sherman RA, Karstetter KW, Damiano M, Evans CB. Stability of temperature asymmetries in reflex sympathetic dystrophy over time and changes in pain. Clin J Pain 1994;10:71–77. Stanton-Hicks M, Janig W, Hassenbusch S, Haddox JD, Boas R, Wilson P. Reflex sympathetic dystrophy: changing concepts and taxonomy. Pain 1995;63:127–133. Uematsu S, Edwin DH, Jankel WR, Kozikowski J, Trattner M. Quantification of thermal asymmetry. Part 1: normal values and reproducibility. J Neurosurg 1988;69:552–555. Veldman PH, Reynen HM, Arntz IE, Goris RJ. Signs and symptoms of reflex sympathetic dystrophy: prospective study of 829 patients. Lancet 1993;342:1012–1016. Wasner G, Backonja MM, Baron R. Traumatic neuralgias: complex regional pain syndromes (reflex sympathetic dystrophy and causalgia): clinical characteristics, pathophysiological mechanisms and therapy. Neurol Clin 1998;16:851–868. Wasner G, Heckmann K, Maier C, Baron R. Vascular abnormalities in acute reflex sympathetic dystrophy (CRPS I): complete inhibition of sympathetic nerve activity with recovery. Arch Neurol 1999;56:613– 620. Wasner G, Schattschneider J, Heckmann K, Maier C, Baron R. Vascular abnormalities in reflex sympathetic dystrophy (CRPS I): mechanisms and diagnostic value. Brain 2001;124:587–599.