96 Journal of the Royal Society of Medicine Volume 83 February 1990
Vicious circles in reflex sympathetic dystrophy-a hypothesis: discussion paper
Caroline Ward BSc Thermographic Unit, Department of Medical Electronics, St Bartholomew's Hospital, London EC1A 7BE E D Cooke MD
Keywords: reflex sympathetic dystrophy; thermoregulation; receptors
Introduction The syndrome of persistent pain, swelling and discolouration of an extremity following injury, or arising de novol, has over the years been called Sudeck's atrophy, painful post-traumatic osteoporosis and more recently algoneurodystrophy and reflex sympathetic dystrophy (RSD). The pathogenesis is not understood. However, a number of hypotheses2 have been suggested all of which involve altered activity of the regional sympathetic nerve supply. From our observations we suggest that sympathetic damage is not the primary cause; that the altered haemodynamics of the affected limb may be explained by the action of neuroactive and vasoactive peptides which act locally to establish a number of vicious circles leading to loss of thermoregulatory control and to persistent pain.
Observations During the past 7 years, 28 patients, in which RSD of the hand was suspected, have been referred to this laboratory for temperature and blood flow studies. Using infrared thermography to measure hand temperature, the ipsilateral and contralateral response to mild cold stress was determined. Methodological details and results have been published elsewhere3. In brief, these studies showed that loss of thermoregulatory control is a feature of RSD. At the same time microcirculatory blood flow in adjacent finger pulps was measured using infrared photoplethysmography (PPG) and laser Doppler flowmetry (LDF). In the former method, infrared light is emitted by the photoplethysmograph probe and transmitted through the tissues. A proportion of the light is absorbed by red blood cells and static tissue, the remainder of the light is reflected. This back scattered light is detected by a photoelectric cell incorporated in the probe. Variations in the intensity of the reflected light is determined by the number of red blood cells in the field of the probe4. The output of the PPG is an expression of microcirculatory volume, a large amplitude in the AC (beat by beat) configuration indicating a large red cell volume5. The depth of penetration of the light emitted by the PPG is some 3 mm thus the record includes flow in arterio-venous shunts. Laser Doppler flowmetry is a similar technique using laser light and is dependent, as the name implies, on the Doppler effect and hence is an expression of red blood cell velocity6. This is measured as a voltage output, a high output indicating rapid flow. The depth of penetration of the laser light is about one millimetre and is largely, therefore, a reflection of capillary flow though clearly arteriovenous flow may also be detected depending on
*
-
=
-_t
Figure 1. Photoplethysmography (top) and laser Doppler flowmety (LDF) (bottom). Normally, with the hands supported at heart level, the records are concordant
I
I
i_ -iI
-,
L
_
__-4_.-
-4.X
___--t = __
---
I
.,
--
--- I-__.__
V' 1L-i
-
,_
F_
t---j
EI_I ~--!FI
_- i_7-I - -' --
---~~
-
_
_
-: -,-
--+ --i J:.A--4
Figure 2. Photoplethysmography (top) showing large microcirculatory wlume, and laser Doppler flowmetry (bottom) showing low microcirculatory velocity (flux) in the right (symptomatic) hand; typical findings in RSD
0141-0768/90/
020096-04/$02.00/0 © 1990 The Royal Society of Medicine
Journal of the Royal Society of Medicine Volume 83 February 1990
skin thickness. Normally the output of PPG and LDF are concordant when the patient is seated with the hands supported at heart level7, the position adopted for this investigation. Vasomotion8, the spontaneous rhythmogenesis of vessels, is evident on both recordings. Also, the PPG recording shows a deep dicrotic notch placed midway in the downstroke ofthe PPG amplitude and provides some notion of the state of compliance of the vessel wall (Figure 1). In patients with RSD the PPG trace is typically large but featureless, ie the dicrotic notch is absent, and the LDF trace has a low voltage output (igure 2). Neither record shows vasomotion. These findings indicate a large microcirculatory volume moving at a low velocity in vessels no longer responding to rhythmic bursts of sympathetic activity.
Bradykinin activates phospholipase A2 on cell membranes" releasing arachidonic acid with, following the cycloxygenase pathway, the production of a number of vasodilatory prostaglandins; F2, G2, E2, A2 and C2. The lipoxygenase products, leukotrienes D4 and C4, are also formed. These produce further capillary permeability and further neutrophil activity with the release of more bradykinin (Figure 3). These reactions must clearly follow any injury. However, depending on a number offactors, eg hypersensitization topa, persistent immobility ofthe limb, either self-imposed or therapeutic, failure of inhibitors or some other unsuspected factor, the activity ofthese peptides produce a state from which the individual either escapes or becomes further embroiled. For example, it is possible that immobility ofthe limb may potentiate the low flow state and prevent the ingress of naturally occurring inhibitors in sufficient quantity to terminate vasodilatation. In these circumstances noxious metabolites may accumulate producing continuous stimulation of nociceptors.
Establishment of the vasodilated state The immediate response to tissue injury is vasodilatation due to the release of substance P and histamine9. Increased capillary dilatation and permeability results in local oedema. Neutrophils migrate along a chemotactic gradient into the tissue fluid producing a plasmin activator with the formation of bradykininl°.
Tissue response to injury
.
Histamine
.
Transmural migration of neutrophils
Establishment of RSD The essential feature of RSD is persistent pain. It follows that this is due to constant stimulation
Bradykinin
--
Arachidonic acid
I
1A
Leukotrienes (D4, C4)
CO2 t Reduced inhibition
Hypoxic stimulation of chemoreceptors
o*
j of autocoids
VASODILATATION-.A
LOW FLOW
-.
H+t
-
Immobility Oe dema Cool limb
Activation of PKC
-
a
Binds to haemoglobin
causing s Lliberation of K-
*
Plus partial agonist & decrease in density or change in receptor characteristics ( a to ,Bi)
of limb
STIMULATION OF NOCICEPTORS
Release of _/ noradrenaline
Prostaglandins,EDRF
-
[B3adkinin |
Activation of myosin _etrophils light chain kinase i Substance P
Histamine
La akaae of .-
Ca2+ TISSUE RESPONSE TO INJURY
Figure 3. (top) Vasoactive amines released as the initial response to injury (shown boxed in b). (bottom) Vicious circles in RSD maintaining persistent vasodilatation, low flow and stimulation of nociceptors
97
98
Journal of the Royal Society of Medicine Volume 83 February 1990
of nociceptors. This may occur due to the action of bradykinin, prostaglandins and endothelium derived relaxing factor (EDRF), released as response to tissue injury. However another important consequence of the release of these peptides may be persistent vasodilatation with inactivation of arterio-venous shunts, leading to loss of thermoregulatory control. Persistent dilatation of arterio-venous shunts allows the blood to be short-circuited producing tissue hypoxia with the stimulation of chemoceptors. The fall in pH, enhanced by the loss of acid-base buffer, enables hydrogen ions to bind to haemoglobin in the red cell with the release of potassium ions12. Both the liberated potassium and the stimulated chemoceptors may also activate the nociceptors to produce pain. Thus the pain of RSD may be explained on the basis ofthe establishment of a series of vicious circles in which vasodilatation, low flow and persistent stimulation of nociceptors all play a part (Figure 2).
Possible role of adrenoceptors The loss of thermoregulatory control suggests that the essential change may be at adrenoceptor level. This may occur due to either down-regulation or conversion of receptors (only a-receptors are represented on the peripheral blood vessels) or depletion of the neurotransmitter noradrenaline or a combination of these factors. Though initially warmer than the contralateral limb the affected limb soon becomes cool, easily mistaken for vasoconstriction, due to the combined effects of the vasodilated low flow state, oedema and immobility. The limb equilibrates with, and finally reaches, the temperature of the surrounding air. Cooling decreases noradrenaline-release from sympathetic nerve terminals reducing vascular smooth muscle contraction mediated by the two subtypes of a-adrenoceptors. Normally this would be compensated for by an increase in the affinity of a2 subtype13. In addition to the reduction in noradrenaline release, cooling depresses the vasoconstrictive response of the a1 subtype, however their large reserve capacity normally buffers this decrease in sensitivity and augments the response of a2 receptors'4. Though only 10% of the a1 receptors need to be activated by the full agonist, noradrenaline, to produce a maximal response, if noradrenaline is depleted all of the receptors need to be activated to produce a maximal response in the presence of a partial agonist. Therefore any inactivation of the receptor sites would prevent the full response by partial agonists accumulating locally, (see below). In addition to inactivation of a-receptors by cooling'3 changes in adrenoceptor activity may occur with alteration in blood pH'5 or by conversion from a to ,B characteristics'0"16. This possibility is suggested by the action of cooling on cardiac muscle which may convert f3 to a behaviour. As well as the inhibition of noradrenaline release by cooling, locally stored noradrenaline may be depleted by persistent, but ineffectual, efforts to produce vasoconstriction as a response to heat loss. Also the slowing and diversion of the regional flow by pre-capillary shunting may depress the ingress of circulating inhibitors so that the normal balance of agonists and antagonists is altered and less potent agonists, produced locally, predominate. Partial agonists, possibly serotonin (5HT) or angiotensin II, would then have a greater probability of occupying
receptor sites. However, since many of the receptors may be inactivated, as suggested above, the large receptor reserve of a1 may be reduced so that a partial agonist may not be able to produce a maximal response since insufficient receptors are available for occupancy. Extending the argument of receptor conversion it is possible to postulate that should noradrenaline not be depleted its action may be vasodilatory. If the receptor is seen as being constructed of two interconvertable states, one accepting the agonist and the other the antagonist'0, then when the agonist configuration is stimulated adjacent receptors convert to this configuration. Changes in blood pH or the increased levels of metabolites, due to the reduced blood flow, may change the receptor site so that the state which normally accepts the agonist, noradrenaline, is converted to the antagonist configuration. Noradrenaline occupying this altered site may either produce no response or a paradoxical, vasodilatory, response. Alternatively, changes in pH may not affect receptor affinity but may alter the ability to produce intrinsic activity, for example activation of adenylate cyclase and consequently cAMP, so that though the receptor site is occupied there is no response. Clearly, none of these suggested pathogenic mechanisms are mutually exclusive; any or all may act synergistically. Discussion The possible role ofthe a-adrenoceptors in the pathogenesis of RSD was suggested by our observations of loss of thermoregulatory control and the changes in the haemodynamics ofthe extremity. As far as we are aware the possibility of vasodilatation associated with a cold cyanotic limb has only been described previously by de Takats'7. Stolte et al.'8 have, however, suggested the possibility of persistent pre-capillary arterio-venous shunting. Changes in adrenoceptors may also explain remote injury causing symptoms in the hand; the shoulderhand syndrome, in which inflammatory products released at the shoulder may reach the hand either via the blood stream or by gravitation where the vicious circles of RSD become established. If this hypothesis is correct then the changes in the circulation should be reversible by active exercise despite the increase in pain which this may produce initially, and which may be overcome by adequate analgesia. This has proved to be the case; increasing mobility is accompanied by pain reduction. We believe that the successes claimed using guanethidine' and, in a few cases, calcitonin'9'20 are due to their analgesic effect allowing the limb to be mobilized, the end point insisted on by authors claing success for these particular therapies2' and which is, de facto, an essential element for recovery. Equally, non-steroidal anti-inflammatory agents are ineffective in RSD because, although prostaglandin synthesis is inhibited, lipoxygenase activity is not and may even be enhanced, increasing the level of leukotrienes. Pain in RSD may be produced by a number of noxious products acting via pathways uninhibited by nonsteroidal anti-inflammatories. It would seem from this hypothesis that following injury there is a high risk of RSD, yet, there is clearly a spectrum ofdisease. Only a few patients develop the full syndrome21. This may be related to the severity of the injury but, since RSD may follow seemingly
Journal of the Royal Society of Medicine Volume 83 February 1990
trivial injury, it seems more likely to be due to the susceptibility of the individual. Common experience shows that some patients will use an injured limb even though it is painful, maintaining blood flow, so that the vicious circles of RSD are not established. On the other hand, the nociceptors in some individuals may be more sensitive than in others, so that the pain is disproportionally distressful, the injured part being held immobile. Conversely the initial reaction may be inhibited at an early stage. For example, neurotensin is a known inhibitor of the effect of substance P on mast cells9. It may be that in some cases the balance between the two peptides may be protective and those people with increased levels of neurotensin may not suffer from RSD. It is perhaps remarkable and difficult to explain why patients having suffered from RSD rarely have a recurrent episode. Whatever the explanation of individual susceptibility to RSD it is evident that the associated abnormal thermal responses and altered haemodynamics may be explained by the activities of neuroactive and vasoactive peptides acting locally. Given these considerations we believe that changes in the regional sympathetic supply, should they occur, are a secondary phenomenon. References 1 Hannington-Kiff JG. Reflex sympathetic dystrophy. J R Soc Med 1987;80:605 2 Schwartzman RJ, McLellan TL. Reflex sympathetic dystrophy: a review. Arch Neurol 1987;44:555-61 3 Cooke ED, Glick EN, Bowcock SA, Smith RE, Ward C, Almond NE, Beacham JA. Reflex sympathetic dystrophy (algoneurodystrophy): temperature studies in the upper limb. Br J Rheumatol 1989;28:399-403 4 Challoner AVJ. Photoelectric plethysmography for estimating cutaneous blood flow. In: Non-invasive physiological measurements, vol 1. London: Academic Press, 1979:125-48 5 Cooke ED, Bowcock SA, Smith AT. Photoplethysmography of the distal pulp in the assessment of the vasospastic hand. Angiology 1985;36:33-40 6 Nilsson GE, Tenland T, Akeoberg P. Evaluation of a laser Doppler flowmeter for measurement of tissue blood flow. IEEE Trans Biomed Eng 1980;27:597-604
7 Almond NE, Jones DP, Cooke ED. High quality photoplethysmograph signals from a laser Doppler flowmeter: preliminary studies of two simultaneous outputs from the finger. JBiomed Eng 1988;10:458-62 8 Funk W, Intaglietta M. Spontaneous arteriolar vasomotion. In: Vasomotion and quantitative capillaroscopy. Basle: Karger, 1983:66-82 9 Foreman J, Jordan C. Histamine release and vascular change induced by neuropeptides. Agents Actions 1983;13:105-16 10 Grundy HF. Lecture notes on pharmacology. London: Blackwell Scientific, 1985 11 Singer HA, Peach MJ. Endothelium-dependent relaxation ofrabbit aorta. 1. Relaxation stimulated by arachidonic acid. J Pharmacol Exp Ther 1983;226:790-801 12 Somjen G. Neurophysiology - the essentials. Baltimore: Williams & Wilkins, 1983 13 Flavahan NA, Vanhoutte PM. Thermosensitivity of cutaneous and deep veins. Phlebology 1988;3:suppl. 1: 41-5 14 Vanhoutte PM, Flavahan NA. Effects of temperature on alpha adrenoceptors in limb veins: role of receptor reserve. Fedn Proc 1986;45:2347-54 15 Stene-Larsen G, Helle KB. Evidence against a transformation of the beta2-adrenoceptor in the frog heart by changes in temperature or metabolic state. Life Scie
1978;23:2681-8 16 Kunos G, Nickerson M. Temperature-induced interconversion of alpha- and beta-adrenoceptors in the frog heart. J Physiol 1976;256:23-40 17 De Takats G, Miller DS. Post-traumatic dystrophy ofthe extremities. A chronic vasodilator mechanism. Arch Surg 1943;46:469 18 Stolte BH, Stolte JB, Leyten JF. De Pathofysiologie von ist Schoulder-Hand Syndroom. Ned Tijdschr Geneeskd 1970;114:1208 19 Martin JL. Clinical course of algoneurodystrophy ofthe limbs treated with salmon calcitonin. Sem Hop Paris 1985;61:1503-7 20 Gobelet C, Meier J-L, Schaffer W, et al. Calcitonin and reflex sympathetic dystrophy syndrome. Clin Rheumatol
1986;5:382-8 21 Glick EN, Algodystrophy. In: The foot, vol 2. Edinburgh: Churchill Livingstone, 1988:1078-87
(Accepted 6 July 1989)
99
Vicious circles in reflex sympathetic dystrophy-a hypothesis: discussion paper
Caroline Ward BSc Thermographic Unit, Department of Medical Electronics, St Bartholomew's Hospital, London EC1A 7BE E D Cooke MD
Keywords: reflex sympathetic dystrophy; thermoregulation; receptors
Introduction The syndrome of persistent pain, swelling and discolouration of an extremity following injury, or arising de novol, has over the years been called Sudeck's atrophy, painful post-traumatic osteoporosis and more recently algoneurodystrophy and reflex sympathetic dystrophy (RSD). The pathogenesis is not understood. However, a number of hypotheses2 have been suggested all of which involve altered activity of the regional sympathetic nerve supply. From our observations we suggest that sympathetic damage is not the primary cause; that the altered haemodynamics of the affected limb may be explained by the action of neuroactive and vasoactive peptides which act locally to establish a number of vicious circles leading to loss of thermoregulatory control and to persistent pain.
Observations During the past 7 years, 28 patients, in which RSD of the hand was suspected, have been referred to this laboratory for temperature and blood flow studies. Using infrared thermography to measure hand temperature, the ipsilateral and contralateral response to mild cold stress was determined. Methodological details and results have been published elsewhere3. In brief, these studies showed that loss of thermoregulatory control is a feature of RSD. At the same time microcirculatory blood flow in adjacent finger pulps was measured using infrared photoplethysmography (PPG) and laser Doppler flowmetry (LDF). In the former method, infrared light is emitted by the photoplethysmograph probe and transmitted through the tissues. A proportion of the light is absorbed by red blood cells and static tissue, the remainder of the light is reflected. This back scattered light is detected by a photoelectric cell incorporated in the probe. Variations in the intensity of the reflected light is determined by the number of red blood cells in the field of the probe4. The output of the PPG is an expression of microcirculatory volume, a large amplitude in the AC (beat by beat) configuration indicating a large red cell volume5. The depth of penetration of the light emitted by the PPG is some 3 mm thus the record includes flow in arterio-venous shunts. Laser Doppler flowmetry is a similar technique using laser light and is dependent, as the name implies, on the Doppler effect and hence is an expression of red blood cell velocity6. This is measured as a voltage output, a high output indicating rapid flow. The depth of penetration of the laser light is about one millimetre and is largely, therefore, a reflection of capillary flow though clearly arteriovenous flow may also be detected depending on
*
-
=
-_t
Figure 1. Photoplethysmography (top) and laser Doppler flowmety (LDF) (bottom). Normally, with the hands supported at heart level, the records are concordant
I
I
i_ -iI
-,
L
_
__-4_.-
-4.X
___--t = __
---
I
.,
--
--- I-__.__
V' 1L-i
-
,_
F_
t---j
EI_I ~--!FI
_- i_7-I - -' --
---~~
-
_
_
-: -,-
--+ --i J:.A--4
Figure 2. Photoplethysmography (top) showing large microcirculatory wlume, and laser Doppler flowmetry (bottom) showing low microcirculatory velocity (flux) in the right (symptomatic) hand; typical findings in RSD
0141-0768/90/
020096-04/$02.00/0 © 1990 The Royal Society of Medicine
Journal of the Royal Society of Medicine Volume 83 February 1990
skin thickness. Normally the output of PPG and LDF are concordant when the patient is seated with the hands supported at heart level7, the position adopted for this investigation. Vasomotion8, the spontaneous rhythmogenesis of vessels, is evident on both recordings. Also, the PPG recording shows a deep dicrotic notch placed midway in the downstroke ofthe PPG amplitude and provides some notion of the state of compliance of the vessel wall (Figure 1). In patients with RSD the PPG trace is typically large but featureless, ie the dicrotic notch is absent, and the LDF trace has a low voltage output (igure 2). Neither record shows vasomotion. These findings indicate a large microcirculatory volume moving at a low velocity in vessels no longer responding to rhythmic bursts of sympathetic activity.
Bradykinin activates phospholipase A2 on cell membranes" releasing arachidonic acid with, following the cycloxygenase pathway, the production of a number of vasodilatory prostaglandins; F2, G2, E2, A2 and C2. The lipoxygenase products, leukotrienes D4 and C4, are also formed. These produce further capillary permeability and further neutrophil activity with the release of more bradykinin (Figure 3). These reactions must clearly follow any injury. However, depending on a number offactors, eg hypersensitization topa, persistent immobility ofthe limb, either self-imposed or therapeutic, failure of inhibitors or some other unsuspected factor, the activity ofthese peptides produce a state from which the individual either escapes or becomes further embroiled. For example, it is possible that immobility ofthe limb may potentiate the low flow state and prevent the ingress of naturally occurring inhibitors in sufficient quantity to terminate vasodilatation. In these circumstances noxious metabolites may accumulate producing continuous stimulation of nociceptors.
Establishment of the vasodilated state The immediate response to tissue injury is vasodilatation due to the release of substance P and histamine9. Increased capillary dilatation and permeability results in local oedema. Neutrophils migrate along a chemotactic gradient into the tissue fluid producing a plasmin activator with the formation of bradykininl°.
Tissue response to injury
.
Histamine
.
Transmural migration of neutrophils
Establishment of RSD The essential feature of RSD is persistent pain. It follows that this is due to constant stimulation
Bradykinin
--
Arachidonic acid
I
1A
Leukotrienes (D4, C4)
CO2 t Reduced inhibition
Hypoxic stimulation of chemoreceptors
o*
j of autocoids
VASODILATATION-.A
LOW FLOW
-.
H+t
-
Immobility Oe dema Cool limb
Activation of PKC
-
a
Binds to haemoglobin
causing s Lliberation of K-
*
Plus partial agonist & decrease in density or change in receptor characteristics ( a to ,Bi)
of limb
STIMULATION OF NOCICEPTORS
Release of _/ noradrenaline
Prostaglandins,EDRF
-
[B3adkinin |
Activation of myosin _etrophils light chain kinase i Substance P
Histamine
La akaae of .-
Ca2+ TISSUE RESPONSE TO INJURY
Figure 3. (top) Vasoactive amines released as the initial response to injury (shown boxed in b). (bottom) Vicious circles in RSD maintaining persistent vasodilatation, low flow and stimulation of nociceptors
97
98
Journal of the Royal Society of Medicine Volume 83 February 1990
of nociceptors. This may occur due to the action of bradykinin, prostaglandins and endothelium derived relaxing factor (EDRF), released as response to tissue injury. However another important consequence of the release of these peptides may be persistent vasodilatation with inactivation of arterio-venous shunts, leading to loss of thermoregulatory control. Persistent dilatation of arterio-venous shunts allows the blood to be short-circuited producing tissue hypoxia with the stimulation of chemoceptors. The fall in pH, enhanced by the loss of acid-base buffer, enables hydrogen ions to bind to haemoglobin in the red cell with the release of potassium ions12. Both the liberated potassium and the stimulated chemoceptors may also activate the nociceptors to produce pain. Thus the pain of RSD may be explained on the basis ofthe establishment of a series of vicious circles in which vasodilatation, low flow and persistent stimulation of nociceptors all play a part (Figure 2).
Possible role of adrenoceptors The loss of thermoregulatory control suggests that the essential change may be at adrenoceptor level. This may occur due to either down-regulation or conversion of receptors (only a-receptors are represented on the peripheral blood vessels) or depletion of the neurotransmitter noradrenaline or a combination of these factors. Though initially warmer than the contralateral limb the affected limb soon becomes cool, easily mistaken for vasoconstriction, due to the combined effects of the vasodilated low flow state, oedema and immobility. The limb equilibrates with, and finally reaches, the temperature of the surrounding air. Cooling decreases noradrenaline-release from sympathetic nerve terminals reducing vascular smooth muscle contraction mediated by the two subtypes of a-adrenoceptors. Normally this would be compensated for by an increase in the affinity of a2 subtype13. In addition to the reduction in noradrenaline release, cooling depresses the vasoconstrictive response of the a1 subtype, however their large reserve capacity normally buffers this decrease in sensitivity and augments the response of a2 receptors'4. Though only 10% of the a1 receptors need to be activated by the full agonist, noradrenaline, to produce a maximal response, if noradrenaline is depleted all of the receptors need to be activated to produce a maximal response in the presence of a partial agonist. Therefore any inactivation of the receptor sites would prevent the full response by partial agonists accumulating locally, (see below). In addition to inactivation of a-receptors by cooling'3 changes in adrenoceptor activity may occur with alteration in blood pH'5 or by conversion from a to ,B characteristics'0"16. This possibility is suggested by the action of cooling on cardiac muscle which may convert f3 to a behaviour. As well as the inhibition of noradrenaline release by cooling, locally stored noradrenaline may be depleted by persistent, but ineffectual, efforts to produce vasoconstriction as a response to heat loss. Also the slowing and diversion of the regional flow by pre-capillary shunting may depress the ingress of circulating inhibitors so that the normal balance of agonists and antagonists is altered and less potent agonists, produced locally, predominate. Partial agonists, possibly serotonin (5HT) or angiotensin II, would then have a greater probability of occupying
receptor sites. However, since many of the receptors may be inactivated, as suggested above, the large receptor reserve of a1 may be reduced so that a partial agonist may not be able to produce a maximal response since insufficient receptors are available for occupancy. Extending the argument of receptor conversion it is possible to postulate that should noradrenaline not be depleted its action may be vasodilatory. If the receptor is seen as being constructed of two interconvertable states, one accepting the agonist and the other the antagonist'0, then when the agonist configuration is stimulated adjacent receptors convert to this configuration. Changes in blood pH or the increased levels of metabolites, due to the reduced blood flow, may change the receptor site so that the state which normally accepts the agonist, noradrenaline, is converted to the antagonist configuration. Noradrenaline occupying this altered site may either produce no response or a paradoxical, vasodilatory, response. Alternatively, changes in pH may not affect receptor affinity but may alter the ability to produce intrinsic activity, for example activation of adenylate cyclase and consequently cAMP, so that though the receptor site is occupied there is no response. Clearly, none of these suggested pathogenic mechanisms are mutually exclusive; any or all may act synergistically. Discussion The possible role ofthe a-adrenoceptors in the pathogenesis of RSD was suggested by our observations of loss of thermoregulatory control and the changes in the haemodynamics ofthe extremity. As far as we are aware the possibility of vasodilatation associated with a cold cyanotic limb has only been described previously by de Takats'7. Stolte et al.'8 have, however, suggested the possibility of persistent pre-capillary arterio-venous shunting. Changes in adrenoceptors may also explain remote injury causing symptoms in the hand; the shoulderhand syndrome, in which inflammatory products released at the shoulder may reach the hand either via the blood stream or by gravitation where the vicious circles of RSD become established. If this hypothesis is correct then the changes in the circulation should be reversible by active exercise despite the increase in pain which this may produce initially, and which may be overcome by adequate analgesia. This has proved to be the case; increasing mobility is accompanied by pain reduction. We believe that the successes claimed using guanethidine' and, in a few cases, calcitonin'9'20 are due to their analgesic effect allowing the limb to be mobilized, the end point insisted on by authors claing success for these particular therapies2' and which is, de facto, an essential element for recovery. Equally, non-steroidal anti-inflammatory agents are ineffective in RSD because, although prostaglandin synthesis is inhibited, lipoxygenase activity is not and may even be enhanced, increasing the level of leukotrienes. Pain in RSD may be produced by a number of noxious products acting via pathways uninhibited by nonsteroidal anti-inflammatories. It would seem from this hypothesis that following injury there is a high risk of RSD, yet, there is clearly a spectrum ofdisease. Only a few patients develop the full syndrome21. This may be related to the severity of the injury but, since RSD may follow seemingly
Journal of the Royal Society of Medicine Volume 83 February 1990
trivial injury, it seems more likely to be due to the susceptibility of the individual. Common experience shows that some patients will use an injured limb even though it is painful, maintaining blood flow, so that the vicious circles of RSD are not established. On the other hand, the nociceptors in some individuals may be more sensitive than in others, so that the pain is disproportionally distressful, the injured part being held immobile. Conversely the initial reaction may be inhibited at an early stage. For example, neurotensin is a known inhibitor of the effect of substance P on mast cells9. It may be that in some cases the balance between the two peptides may be protective and those people with increased levels of neurotensin may not suffer from RSD. It is perhaps remarkable and difficult to explain why patients having suffered from RSD rarely have a recurrent episode. Whatever the explanation of individual susceptibility to RSD it is evident that the associated abnormal thermal responses and altered haemodynamics may be explained by the activities of neuroactive and vasoactive peptides acting locally. Given these considerations we believe that changes in the regional sympathetic supply, should they occur, are a secondary phenomenon. References 1 Hannington-Kiff JG. Reflex sympathetic dystrophy. J R Soc Med 1987;80:605 2 Schwartzman RJ, McLellan TL. Reflex sympathetic dystrophy: a review. Arch Neurol 1987;44:555-61 3 Cooke ED, Glick EN, Bowcock SA, Smith RE, Ward C, Almond NE, Beacham JA. Reflex sympathetic dystrophy (algoneurodystrophy): temperature studies in the upper limb. Br J Rheumatol 1989;28:399-403 4 Challoner AVJ. Photoelectric plethysmography for estimating cutaneous blood flow. In: Non-invasive physiological measurements, vol 1. London: Academic Press, 1979:125-48 5 Cooke ED, Bowcock SA, Smith AT. Photoplethysmography of the distal pulp in the assessment of the vasospastic hand. Angiology 1985;36:33-40 6 Nilsson GE, Tenland T, Akeoberg P. Evaluation of a laser Doppler flowmeter for measurement of tissue blood flow. IEEE Trans Biomed Eng 1980;27:597-604
7 Almond NE, Jones DP, Cooke ED. High quality photoplethysmograph signals from a laser Doppler flowmeter: preliminary studies of two simultaneous outputs from the finger. JBiomed Eng 1988;10:458-62 8 Funk W, Intaglietta M. Spontaneous arteriolar vasomotion. In: Vasomotion and quantitative capillaroscopy. Basle: Karger, 1983:66-82 9 Foreman J, Jordan C. Histamine release and vascular change induced by neuropeptides. Agents Actions 1983;13:105-16 10 Grundy HF. Lecture notes on pharmacology. London: Blackwell Scientific, 1985 11 Singer HA, Peach MJ. Endothelium-dependent relaxation ofrabbit aorta. 1. Relaxation stimulated by arachidonic acid. J Pharmacol Exp Ther 1983;226:790-801 12 Somjen G. Neurophysiology - the essentials. Baltimore: Williams & Wilkins, 1983 13 Flavahan NA, Vanhoutte PM. Thermosensitivity of cutaneous and deep veins. Phlebology 1988;3:suppl. 1: 41-5 14 Vanhoutte PM, Flavahan NA. Effects of temperature on alpha adrenoceptors in limb veins: role of receptor reserve. Fedn Proc 1986;45:2347-54 15 Stene-Larsen G, Helle KB. Evidence against a transformation of the beta2-adrenoceptor in the frog heart by changes in temperature or metabolic state. Life Scie
1978;23:2681-8 16 Kunos G, Nickerson M. Temperature-induced interconversion of alpha- and beta-adrenoceptors in the frog heart. J Physiol 1976;256:23-40 17 De Takats G, Miller DS. Post-traumatic dystrophy ofthe extremities. A chronic vasodilator mechanism. Arch Surg 1943;46:469 18 Stolte BH, Stolte JB, Leyten JF. De Pathofysiologie von ist Schoulder-Hand Syndroom. Ned Tijdschr Geneeskd 1970;114:1208 19 Martin JL. Clinical course of algoneurodystrophy ofthe limbs treated with salmon calcitonin. Sem Hop Paris 1985;61:1503-7 20 Gobelet C, Meier J-L, Schaffer W, et al. Calcitonin and reflex sympathetic dystrophy syndrome. Clin Rheumatol
1986;5:382-8 21 Glick EN, Algodystrophy. In: The foot, vol 2. Edinburgh: Churchill Livingstone, 1988:1078-87
(Accepted 6 July 1989)
99
