Mechanisms and Treatment Strategy of Complex Regional Pain Syndromes

INTRODUCTION

Complex regional pain syndromes (CRPS) are painful disorders typically affecting the limbs that may develop as a consequence of trauma. That the sympathetic nervous system is causally involved in CRPS is based mainly on two observations: (1) the pain is spatially correlated with signs of autonomic dysfunction, that is, with abnormalities in blood flow and sweating, as well as with trophic changes, and (2) temporary blocking the efferent sympathetic supply to the affected part relieves the pain in subgroups of patients with CRPS. CRPS syndromes were originally known under the terms reflex sympathetic dystrophy or causalgia (or several other terms depending on medical speciality and country). However, these terms were thought to be inappropriate as a clinical designation: They were sloppily used to describe an extensive range of clinical presentations and the pathophysiologic mechanisms underlying these syndromes were poorly understood. The new terminology is descriptive and based entirely on elements of history, symptoms, and findings on clinical examination, with no implied pathophysiologic mechanism avoiding any mechanistic implications. According to the International Association for the Study of Pain (IASP) Classification of Chronic Pain, reflex sympathetic dystrophy and causalgia are now called Complex Regional Pain Syndromes (CRPS). In reflex sympathetic dystrophy (CRPS type I), minor injuries at the limb or lesions in remote body areas precede the onset of symptoms. CRPS type II (causalgia) may develop after injury of a major peripheral nerve.

Patients with CRPS presenting with exactly the same clinical signs and symptoms can be further divided into two subgroups by the negative or positive effect of sympathetic blockade. The pain component that is relieved by specific sympatholytic procedures is considered sympathetically maintained pain (SMP). Thus, SMP is defined as a symptom and not a clinical entity . The positive effect of a sympathetic blockade is not essential for the diagnosis of CRPS. On the other hand, the only possibility to differentiate between SMP and sympathetically independent pain (SIP) is the efficacy of a correctly applied sympatholytic intervention.

The mechanisms underlying CRPS remain controversial despite the extensive body of clinical experience and experimentation on patients and animal models. The lack of well-controlled clinical studies has been accompanied by extensive speculations about the underlying pathophysiology. Here we will first present a hypothesis stating that CRPS is a disease of the central nervous system (CNS) involving central and peripheral components and discuss the clinical and experimental data supporting this hypothesis. Based on this concept we will then describe the diagnostic approach and the treatment strategy. The clinical phenomena of CRPS ( Table 27-1 ) are described in detail in references .

Table 27-1

Diagnostic Criteria and Diagnostic Tests in Complex Regional Pain Syndrome (CRPS)

A. Diagnostic Criteria: Categories of Clinical Signs and Symptoms ^*
1. Positive sensory abnormalities • Spontaneous pain • Mechanical hyperalgesia • Thermal hyperalgesia • Deep somatic hyperalgesia 2. Vascular abnormalities • Vasodilation • Vasoconstriction • Skin temperature asymmetries • Skin color changes	4. Motor (M) and trophic changes (T) • Motor weakness (M) • Tremor (M) • Dystonia (M) • Coordination deficit (M) • Nail or hair changes (T) • Skin atrophy (T) • Joint stiffness (T) • Soft tissue changes (T)
3. Edema, sweating abnormalities • Swelling • Hyperhidrosis • Hypohidrosis
Interpretation
Clinical use
≥1 symptoms of ≥3 categories each AND ≥1 signs at the time of evaluation in ≥2 categories each Sensitivity 0.85, specificity 0.60
Research use
≥1 symptoms in each of the 4 categories AND ≥1 signs at the time of evaluation in ≥2 categories eachSensitivity 0.70, specificity 0.96

B. Diagnostic Tests
Test	Sensitivity	Specificity
1. Plain Radiograph (only chronic CRPS)	0.73	0.57
2. Three phase bone scan (only acute CRPS)	0.97	0.86 ^†
3. Quantitative sensory testing	High	Low
4. Temperature differences (during symptom stimulation)	0.76	0.93
5. MRI (e.g., skin, joint)	0.91	0.17
6. Magnet encephalogram, functional MRI (cortical reorganization)	Unknown (prob. not useful)	Probably low

MRI, magnetic resonance imaging.

From Baron R. Complex regional pain syndromes. In: Basbaum AL, Kaneko A, Shepherd GM, Westheimer G, Basbaum AI, editors. The Senses: a Comprehensive Reference. Vol 5. Pain. San Diego: Academic Press; 2008. p. 909–18.

* These diagnostic criteria have been modified compared with the original criteria as defined by Stanton-Hicks and collegues in order to obtain a higher specificity for clinical use (to avoid overdiagnosis of CRPS) and a higher sensitivity for research use.

† Early differentiation from normal post-traumatic states is difficult.

COMPLEX REGIONAL PAIN SYNDROME IS A NEURONAL DISEASE INVOLVING THE CENTRAL NERVOUS SYSTEM

CRPS is characterized by sensory, sympathetic, somatomotor, and trophic changes (including swelling) that occur in variable combinations. We have hypothesized that these changes are the result of an altered processing of information in the CNS involving the somatosensory non-nociceptive and nociceptive systems, the endogenous neuronal systems controlling nociceptive impulse transmission, the sympathetic systems, and the somatomotor system. Various levels of integration probably are involved such as spinal cord, brain stem, diencephalon (hypothalamus, thalamus), and telencephalon (cortex and limbic system). Key players in generation and maintenance of CRPS are most likely the nociceptive system and the sympathetic nervous system. But CRPS cannot be reduced to a malfunctioning of one of these systems or components of them alone. Here we discuss the arguments supporting that CRPS is a CNS disease. The upper part in Table 27-2 lists clinical and experimental observations made on patients with CRPS that clearly support this contention. The lower part in Table 27-2 lists the peripheral changes observed in patients with CRPS that are also related in some yet unknown way to the central changes.

Table 27-2

Arguments for Central and for Peripheral Changes in Complex Regional Pain Syndrome *

Central Changes

Initiating events ( 1 in Fig. 27-1 )

•

Out of proportion to finding (minor trauma)
•

Events remote from affected extremity (e.g., in the visceral domain)
•

Central (e.g., after stroke; related to endogenous control systems?)

Sensory changes ( 2 in Fig. 27-1 )

•

Mechanical allodynia (quadrant, hemisensory)
•

Hypoesthesias (mechanical, cold, warm; hemisensory, quadrant)
•

Bilateral distribution of hypo- and hyperesthesias (mechanical, cold, warm, heat)

Pain relief by sympathetic blocks with local anesthetics ( 3 in Fig. 27-1 )

•

Relief of pain outlasts conduction block by an order of magnitude (i.e., a temporary block is followed by a long-lasting pain relief)
•

A few temporary blocks are sometimes sufficient to generate permanent pain relief
•

Sympathetic activity maintains a positive feedback circuit (?)

Changes of regulation by sympathetic systems ( 4 in Fig. 27-1 )

•

Thermoregulatory reflexes in cutaneous vasoconstrictor neurons reduced
•

Respiration elicited reflexes (generated by deep inspiration and expiration) in cutaneous vasoconstrictor neurons reduced
•

Changes of activity in sudomotor neurons (sweating)
•

Swelling reduced by sympathetic blocks

Somatomotor changes ( 5 in Fig. 27-1 )

•

Active motor force and active range of motion reduced
•

Physiologic tremor increased
•

Poor motor control and coordination of movement; altered gait and posture
•

Dystonia
•

Sensory-motor body perception disturbance

Peripheral Changes

Sympathetic-afferent coupling ( 6 in Fig. 27-1 )

•

After nerve lesion via norepinephrine and adrenoceptors (CRPS II)
•

Indirectly via vascular bed and other mechanisms (CRPS I; deep somatic?)
•

[Indirectly via inflammatory mediators and neurotrophic factors]
•

[Mediated by the adrenal medulla (epinephrine)]

Inflammatory changes and edema ( 7 in Fig. 27-1 )

•

Neurogenic inflammation (precapillary vasodilation, venular plasma extravasation), involvement of peptidergic afferents (?)
•

Sympathetic fibers mediating effects of inflammatory mediators (e.g., bradykinin) to venules leading to plasma extravasation (?)
•

Involvement of inflammatory cells and immune system (?)
•

Change of capillary filtration pressure (?)

Trophic changes ( 8 in Fig. 27-1 )

•

Long-range consequences of inflammatory changes and edema (?)
•

Direct (trophic?) effect of sympathetic and afferent fibers on tissue (?)
•

Endothelial damage (?)

Modified from Jänig W, Baron R. Complex regional pain syndrome: mystery explained? Lancet Neurol 2003;2:687–97.

* Arguments for central and peripheral changes based on clinical observations and quantitative measurements made in patients with CRPS. In bold are listed the phenomena observed on the patients. The numbers in brackets refer to the explanatory hypothesis outlined graphically in Figure 27-1.

The explanatory hypothesis as outlined in Figure 27-1 puts the clinical findings observed in patients with CRPS in relation to the changes in the somatosensory, autonomic and somatomotor systems and postulates that changes in the central representations of these systems must occur in order to explain the clinical findings. The events initiating the clinical symptoms are mostly associated with a trauma at the extremities but sometimes also with trauma in the viscera or in the CNS. The changes developing after these triggering events usually outlast the trauma by orders of magnitude.

An important component of the hypothesis for patients with CRPS with SMP is postulated to be a positive feedback circuit consisting of afferent neurons, central neurons (spinal circuits and their supraspinal controls), sympathetic neurons, and sympathetic-afferent coupling. This circuit would maintain spontaneous pain (sympathetically maintained), hyperalgesia and allodynia and the other associated changes in the peripheral tissues (see 6-8 in Table 27-2 and Figure 27-1 ). This positive feedback circuit does not explain the (peripheral and central) mechanisms in detail and fails to explain the central changes that must occur in CRPS in view of the clinical changes observed in these patients (see below and Table 27-2 ). Furthermore, it fails to explain why CRPS I without SMP is clinically entirely indistinguishable from CRPS I with SMP.

Clinical and Experimental Observations Supporting the Hypothesis

Initiating Events

The signs and symptoms in CRPS (see Table 27-1A ) are disproportionate to the traumatic events initiating or triggering this syndrome. The local changes generated by the trauma often disappear, yet the syndrome persists. Furthermore, CRPS I in an extremity may be triggered by remote events (e.g., in the viscera) or by events in the CNS (e.g., central lesions) ( 1 in Fig. 27-1 ). It has been proposed that processes in the prefrontal, frontal, and parietal cortices that are related to psychosocial changes enhance the clinical signs and symptoms in CRPS or even may initiate them. These clinical observations argue (1) that mechanisms operating in CRPS cannot simply be explained to be caused by events in the body related to the trauma (e.g., sympathetic-afferent coupling or persistent activation of nociceptive afferents) and (2) that the CNS is important to understand the mechanisms initiating and maintaining the CRPS syndromes.

Somatic Sensations Including Pain

Pain is a salient feature of CRPS, consisting of ongoing pain, hyperalgesia and allodynia * generated by mechanical or cold stimuli. However, it must be kept in mind that about 10% or more of patients with CRPS do not have ongoing pain, that the pain is mostly not restricted to the site of trauma, that the pain is associated with other somato-sensory changes (cold, mechanical, warm hypoesthesias) in 50% of the patients, and that pain (and the other somatosensory changes) persists for long periods of time after the local changes of the trauma have subsided. Thus, pain in CRPS must be understood in a wider context in order to understand the therapeutic implications: (1) Ongoing pain and evoked pain (superficial and particular deep mechanical hyperalgesia and allodynia) are mostly entirely out of proportion to the trauma. (2) The anatomic distribution of the changed painful and nonpainful somatosensory perceptions observed in CRPS patients are likely due to changes in the central representation of somatosensory sensations in the thalamus and cortex. This is fully supported by magnetic encephalographic (MEG) and functional magnetic resonance imaging (MRI) studies showing that changes occur in various cortical areas, including the primary and secondary somatosensory, insular, frontal and parietal cortices (for discussion and references see reference ). In fact, patients with CRPS can exhibit strong body perception disturbances and show sensations referred to areas of the body immediately adjacent to the stimulated body sites. (3) Generalized somatosensory deficits are particularly found in patients with chronic CRPS, indicating long-term plastic changes in the telencephalon. To emphasize, these CNS changes occur in patients with CRPS I who have no nerve lesions. (4) CRPS patients mostly locate their spontaneous pain into deep somatic structures of the affected extremity. Furthermore, they have deep somatic mechanical hyperalgesia/allodynia.

Sympathetically Maintained Pain in Complex Regional Pain Syndrome

Pain dependent on activity in the sympathetic neurons called sympathetically maintained pain (SMP ) usually includes both spontaneous and evoked pain (i.e., allodynia evoked by mechanical or cold stimuli) and is present in about 60% of patients with acute CRPS and can persist for years. The concept that the (efferent) sympathetic nervous system is involved in the generation of pain is based on long standing clinical observations.

In healthy subjects, there is no obvious sign for direct or indirect coupling between the efferent sympathetic systems and the afferent systems in the peripheral tissues leading to pain, discomfort, or other sensations maintained by the sympathetic outflow. This may change after trauma with or without nerve lesion. Now activity in sympathetic neurons may lead to SMP. The concept is schematically exemplified in Figure 27-2 . During inflammation or following nerve lesion, the efferent (noradrenergic) innervation of the affected tissue may generate feedback to the primary afferent neurons and activate them, enhance their ongoing activity, or enhance their activity to mechanical or thermal stimuli. This, in turn, would amplify the physiologic impulse transmission in the spinal or trigeminal dorsal horn (sensitization of dorsal horn neurons) or would enhance the hyperexcitability of these neurons following nerve lesion.