Radiofrequency lesions of the dorsal root ganglion |
| Chronic pain is an important problem in daily medical practice. Apart from its psychosocial consequences for the individual patient, it has enormous socio-economic impact on society. In 1991 the total cost of low back pain to the Netherlands economy was estimated at NLG 7.5 milliard (Van Tuider et al. 1995). The majority of chronic pain patients suffers from pain emanating from the spine and adjacent structures. It is this type of pain that is a frequent cause for disability and loss of mobility (Kelsey et al. 1979). About 70-80% of the total adult population experiences acute low back or neck pain at some stage of their lives. The symptoms resolve spontaneously in the majority of cases but in 8-12% of patients the pain becomes chronic (Bland 1986, Valkenburg and Haanen 1982). Social and psychological factors start playing a role making the pain syndrome more complex (Bonica 1979, Sternbach 1986). The mechanism of spinal pain is complex. In 15% of patients with low back pain a specific cause is found such as disc protrusion, infection, spondylolisthesis or a vertebral fracture (Spitzer et al. 1987). In the majority of cases the mechanism remains unclear despite thorough diagnostic investigations. When radiological abnormalities are found, they are often aspecific and fail to correlate with the patient's complaints (Friedenberg and Miller 1963, Heller et al. 1983, Jensen et al. 1994). Wyke (1987) classifies the causes of spinal pain into primary, secondary, referred and psychogenic causes:
O'Brien (1 984) further classifies primary pain, according to the source, into:
Management of pain of spinal origin is directed to the cause of the pain provided that diagnostic procedures have revealed a specific lesion that correlates with the patient's complaints. However, as mentioned, this approach only applies to a small number of patients. Patients without a specific cause of pain are managed symptomatically by non-invasive conservative treatment such as medication, transcutaneous nerve stimulation, manual therapy and physical therapy. Nevertheless, there remains an important group of patients who do not have a specific lesion and who do not respond to these forms of symptomatic treatment. Further management of their pain then depends on the presence or absence of nociception. Nociceptive pain is defined as pain that originates from somatic structures and is perceived to be commensurate with tissue damage. Diagnostic nerve blocks using a local anaesthetic solution are a useful tool in determining whether a nociceptive focus is present and if so where this is located. Even though this diagnostic technique has both technical and theoretical problems, which could lead to a false interpretation of the results, these blocks seem, at present, to be the best available method of analysing the nociceptive pattern. When a diagnostic block has been succesful a more definitive interruption of the sensory innervation may be considered. In the past a surgical approach was used to obtain more permanent pain relief. Often this resulted in aggravating the pain problem in the deafferentated dermatomes (anesthesia dolorosa) (Bertrand 1975, Loeser et al. 1968, Loeser 1972). A less destructive method is the use of high-frequency current which gives a controllable heat lesion in the adjacent tissue. Since high-frequencies were also used in radiotransmitters, the high-frequency current was named radiofrequency current (RF-current). The lesions made with this current were introduced as radiofrequency lesions (RF-lesions). The extent of the interrupting lesion may be controlled by varying the output of the RF-lesion generator. RF-lesions can be made at various locations. They may be directed at interrupting the sensory innervation from a specific nociceptive structure, such as the zygapophysial joint. To interrupt nociceptive stimuli from a segmental region rather than from a specific structure, RF-lesions at the level of the dorsal root ganglion can be produced. This was first proposed by Uematsu et al. (1974) who described a "percutaneous electrothermocoagulation of the spinal nerve trunk, ganglion and rootlets" as an alternative to open surgery. Uematsu's technique of radiofrequency thermocoagulation of the dorsal root ganglion was considered to be a relatively simple technique that could relieve pain and preserve motor function and proprioception. Sluijter and Mehta (1981) later improved this technique by introducing a small diameter temperature-monitoring electrode system. This should allow higher controllability as to the extent and localisation of the lesion. A radiofrequency lesion of the dorsal root ganglion (RF-DRG) is an intriguing site for a radiofrequency lesion for two reasons:
The aim of this thesis was to evaluate the efficacy and safety of RF-DRG in the treatment of chronic spinal pain. More specifically, we sought to answer the following questions:
These questions were approached as follows:
In chapter 1: The impact of spinal pain on society is discussed. This thesis concerns radiofrequency (RF) techniques in the treatment of spinal pain. It focusses on the efficacy and on the potential side effects of a radiofrequency lesion of the dorsal root ganglion (RF-DRG). A lesion at this location is of interest for two reasons:
The aim of the thesis is to evaluate the efficacy and safety of an RF-DRG in the treatment of spinal pain. In order to solve this problem the following questions were formulated:
In chapter 2: The history of RF-lesioning and the theoretical aspects of these lesions are discussed. A literature review concerning the size of the lesion using different types of equipment and different tip temperatures is given. The controversial aspects of the effect of an RF-lesion on nerve tissue is discussed. All morphological studies indicate that RF-current applied directly to nerve tissue destroys all unmyelinated and myelinated fibres. No studies are available concerning RF-DRG. In chapter 3: General considerations for treatment of spinal pain with RF-techniques are discussed. Several classifications of pain and spinal pain are given. The innervation of the spine is discussed and also the diagnosis in spinal pain and the role of RF-lesions in treatment. The value of diagnostic blocking and the pitfalls of these blocks are discussed. In chapter 4: Review of the clinical syndromes in the cervical, thoracic and lumbar area, which are relevant for the treatment of chronic spinal pain with RF-techniques. Also the RF-techniques are discussed. In chapter 5: Morphological changes in the dorsal root ganglion were evaluated in goats. Using fluoroscopic control, RF-DRG's were made at 5 lumbar levels, 22 G 1 00 mm SMK electrodes with a 5 mm active tip were positioned posterior to the lumbar dorsal root ganglion of Ll -L5. Sixty seconds 67°C RF-lesions were made unilaterally. On the contralateral side electrodes were similarly placed but no RF-lesion was made. The goats were sacrificed two weeks after the procedure. The lesions were studied for size as weil as for effects on proliferation and regeneration using Ki-67 (MIB-1). The single lesion made in the L5 lumbar ganglion in the two separated models were 1.8 mm-2.0 mm, respectively. These lesions located within the dorsal root ganglion, showed a total loss of myelinated fibres, as shown by light microscopy in HE-stained sections. These changes were not observed when the lesion was made adjacent to the dorsal root ganglion. However, in these ganglia, using immunohistochemical technique (MIB-1) microscopic changes were observed. There was a significantly higher MIB-1 activity in the ganglia which had had a lesion of the dorsal root ganglion when compared to ganglia where a sham lesion was made. In chapter 6: Possible denervation sequelae of RF-DRG were assessed in patients with chronic cervicobrachialgia. The patients were selected for treatment after at least three diagnostic blocks at the relevant levels (C4, C5, C6). The level providing the best temporary analgesic response, was selected for RF-DRG. Electromyography (EMG) and sensory evoked potentials (SEP) were recorded before and three weeks after RF-lesioning, in order to investigate the incidence of unwanted damage to myelinated fibres. Also side effects and the effect on pain were studied. After the procedure the EMG showed no signs of denervation. One SEP recording remained abnormal after treatment. There was relief of pain in 75% of patients 3 months after treatment and in 50% of patients 6 months after treatment. In chapter 7: Clinical evaluation of RF-DRG as a therapy for chronic pain was done on a consecutive group of patients presenting with chronic thoracic pain. Patients were selected for an RF-DRG after two or more diagnostic blocks had been performed under fluoroscopic control. The level providing the best analgesic response was selected for treatment. At this level an RF-DRG was made. An RF-DRG provided short-term (8 weeks) relief of pain in 67% and long term pain relief (>36 weeks) in 52% of patients with a limited distribution of pain. In chapter 8: Placebo-controlled treatment effects were evaluated in a prospective double blind randomised study performed in patients with unilateral chronic cervicobrachialgia. Twenty consecutive patients with a history of, at least, one year of intractable chronic cervicobrachialgia, were randomly assigned to one of two treatment groups. Group I received a 67° C RF-DRG. Group II was treated in an identical manner but no RF-lesion was made. Both one week prior to, as well as eight weeks after the procedure the patients were questioned about their pain using different tests. These tests showed that eight weeks after the procedure there was a significant number of patients with pain reduction in group I compared to Group II.
a. The morphological changes in the dorsal root ganglion following RF-DRG, An RF-lesion made inside the dorsal root ganglion resulted in an indiscriminate destruction of nervous tissue, when viewed by light microscopy in HE-stained sections, two weeks after the lesion was made. These changes were not observed in a dorsal root ganglion when the lesion was made adjacent to the dorsal root ganglion and viewed using the same technique and timespan. However, using an immunohistochemical technique in the same ganglia after two weeks, an increased MIB-1 activity was observed in the satellite cells indicating proliferation and regeneration of nervous tissue. This indicates that the extend of the effect of an RF-lesion is larger than is visible by using light microscopy. b. The signs of motor denervation after RF-DRG. The electromyography (EMG) study performed in 17 patients did not show any signs of motor denervation after an RF-DRG. One sensory evoked potential (SEP) recording was changed after treatment, indicating damage to small myelinated fibres. The most common side effect was a temporary mild burning pain in the treated dermatome. A transient hyposensibility in the dermatome was noticed in 35% of patients. Except for one patient all side effects had disappeared 6 weeks after treatment. In conclusion, this study did not reveal any signs of motor denervation after RF-DRG. There were no long term signs of deafferentation. c. Clinical effects of one single RF-DRG on pain. The clinical efficacy of one single RF-DRG was evaluated in patients with chronic thoracic pain and in the group of patients with chronic cervicobrachial pain:
d. The efficacy of one single RF-DRG on chronic cervicobrachialgia compared to placebo. The results of the prospective double blind study showed that 8 weeks after the procedure there was a significant number of patients with pain reduction after an RF-DRG compared with the placebo group of patients. This study indicates that an RF-DRG can result in a significant alleviation of chronic cervicobrachial pain.
Closing remarks On theoretical grounds an RF-DRG is a potentially elegant procedure to induce pain relief in chronic spinal pain syndromes. We undertook several studies as to the clinical applicability of this procedure. One of these studies clearly shows that the procedure does not produce any serious complications such as denervation of motor fibres. The experimental animal study demonstrated that the technique enables one to produce an effect on the ganglion, without actual positioning the electrode inside the ganglion. Clinical efficacy studies showed pain relief in 67-75% of selected patients with chronic spinal pain after one RF-DRG. On a long term basis (36 weeks) this positive effect was still present in 40-52% of patients. It appeared that the pain relief obtained by RF-DRG tends to decrease over time. The cause of this phenomenon remains obscure. It may be due either to the multisegmental innervation of the spine or due to regeneration of nerve fibres. If multisegmental innervation plays a role in the phenomenon, one could consider producing multiple RF-lesions for better long term effects. Outcome studies should be performed to evaluate this. Should regeneration of nervous tissue play a role in the gradual decrease of clinical benefit, then treatment could consist of a repeated procedure. This would be permissible since morphological and neurophysiological studies indicate that an RF-DRG can be performed without any effect on motor fibres and only temporary effect on small myelinated afferent fibres. Another possible method to reduce the gradual decrease of clinical response, is the combination of RF-lesions and rehabilitation programs. The period of pain relief could then be utilized to obtain maximum benefit from conservative treatment such as rehabilitation and improvement of function. The effect of a combination of several modalities of treatment should be the subject of future studies.
|
alpha2-adrenergic receptor agonists in myocardial ischemia |
Despite continuing improvements in anesthetic and surgical techniques, it is expected that perioperative cardiac morbidity and mortality will continue to increase because of the rapid ageing of the surgical population and greater prevalence of more advanced coronary artery disease.1 Solution of this problem requires identification of the predictors of adverse perioperative cardiac outcome, followed by therapeutic trials aimed at modifying these predictors in an effort to decrease morbidity and mortality. Perioperative outcome studies have identified preoperative, intraoperative and postoperative predictors of perioperative cardiac morbidity.1 The most important preoperative predictors of cardiac outcome are unstable coronary syndromes, decompensated congestive heart failure, significant arrhythmia’s and severe valvular disease. The classic intraoperative predictors include emergency surgery, vascular surgery, prolonged thoracic or upper abdominal surgery and hypotension and tachycardia. It was recently shown by Mangano and co-workers that the single most important predictor of adverse cardiac outcome was early postoperative myocardial ischemia.2,3 Such postoperative ischemia conferred a ninefold increase in the odds of experiencing cardiac death, a nonfatal myocardial infarction, or unstable angina. These results suggest that prevention and therapy for perioperative - and especially postoperative - ischemia may hold the key to reducing perioperative cardiac morbidity. There are very few randomized trials of medical therapy before surgery to prevent perioperative myocardial ischemia, and they do not provide enough data which to draw firm conclusions.4-12 Studies in non-surgical patients have suggested that the alpha2-adrenergic receptor agonist clonidine has anti-anginal and ischemia-limiting effects.13-14 Recently, several studies have evaluated the use of alpha2-adrenoreceptor agonists during the perioperative period.15 Most of these studies concentrated on their sedative, sympatholytic and hemodynamic stabilizing effects. The present series of investigations was initiated in order to study possible mechanisms for the potential anti-ischemic effect of the new alpha2 agonists dexmedetomidine and mivazerol, and to gain information on the possible usefulness of these drugs as anesthesia adjuvants in high-risk cardiovascular patients at risk for coronary artery disease.
alpha2-adrenergic receptors are widely distributed in various animal and human tissues. These receptors mediate a variety of physiological functions, depending on the adrenoceptor type and the tissue. Therefore, the activation of adrenoceptors represent the net effect of sometimes conflicting actions at different sites. alpha2-adrenergic receptor agonists are drugs that activate these receptors. In 1964, the prototypical alpha2 agonist clonidine was tested for possible use as a nasal decongestant. Volunteers receiving intranasal clonidine became very sleepy and their heart rate and blood pressure decreased. It was decided therefore by the pharmaceutical company to further study clonidine for it's anti-hypertensive properties. In 1966, clonidine was introduced into clinical practice for the treatment of hypertension. The sedative and antisialic effects of clonidine certainly must have limited the popularity of the drug when used as chronic anti-hypertensive medication. In recent years, interest has been focused on the application of clonidine in the perioperative period. In addition to decreased sympathoadrenal activity to provide more cardiovascular stability, clonidine has been shown to cause sedation, anxiolysis, analgesia, decreased salivation and to reduce anesthetic requirements. These effects of clonidine might be especially beneficial in patients with severe systemic diseases. Data from preliminary studies in small patient groups suggest that clonidine may decrease the incidence of myocardial ischemia in high-risk cardiovascular patients. In this thesis, we studied whether the new, specific and selective alpha2-adrenoreceptor agonists like dexmedetomidine and mivazerol could be beneficial anesthetic adjuvants in the perioperative period in patients at risk for coronary artery disease. To this purpose, a series of laboratory investigations was designed to study potential anti-ischemic mechanisms of action of dexmedetomidine and mivazerol. Special attention was paid to their central sympatholytic effect, their systemic cardiovascular effects, and to their effects on perfusion and function of normal and ischemic myocardium. In addition, we studied – together with investigators from The Multicenter Study of Perioperative Ischemia Group (McSPI) from other European Medical Centres -the effect of mivazerol on perioperative hemodynamic stability and myocardial ischemia in 300 patients at risk for coronary artery disease. Dexmedetomidine and mivazerol consistently caused profound sympatholysis with decreases in circulating catecholamines. The clinical dose of 1 microg.kg-1 dexmedetomidine decreased heart rate by 20%, increased mean arterial pressure by 25% and decreased cardiac output by 32% (chapters 4, 6 and 7). Mivazerol decreased heart rate by 13%, increased mean arterial pressure by 13% and decreased cardiac output by 43% (chapter8). In 8 dogs we measured left ventricular pressure-volume relations with the conductance catheter technique, before and after the administration of 1 microg.kg-1 dexmedetomidine. dP/dtmax, which is a pre-and afterload dependent index of contractility, decreased approximately 10-20%, but not always significantly. End-systolic elastance and preload-recruitable stroke work, pre- and afterload independent variables of contractility, decreased 45% and 40%, respectively. Left ventricular ejection fraction decreased 35%. (R. Frietman, J.J. Schreuder, F.W. Prinzen, P.M.H.J. Roekaerts and S. de Lange: Pressure-volume relationships in the left ventricle during use of dexmedetomidine. Ned Tijdschr Geneeskd 139 (30): 1575, 1995). Thus, alpha2 agonists significantly decrease cardiac contractility. However, it is most important to note that long and extensive clinical experience with clonidine has not produced any evidence of adverse effects on cardiac and coronary function. On the contrary, clonidine has been used successfully in patients with coronary artery disease, as well as in patients with congestive heart failure. It is becoming increasingly apparent that the optimum way of anesthetizing patients with ischemic heart disease is to "reduce the determinants of myocardial oxygen demand". (WK Hamilton: Do let the blood pressure drop and do use myocardial depressants! Anesthesiology 45(3): 273-274, 1976). The reductions in heart rate and contractility after dexmedetomidine and mivazerol decrease myocardial oxygen demand and are therefore important mechanisms for the anti-ischemic potential of alpha2-adrenergic agonists. In the present experiments, dexmedetomidine and mivazerol caused coronary vasoconstriction under normal, non-ischemic, conditions (chapters 4, 6 and 8). This vasoconstriction however was not associated with changes in arterial-coronary venous oxygen saturation or lactate differences, indicating adequate adaptation of myocardial blood flow to metabolic requirements (chapter 6). In chapter 4, we showed that the systemic and coronary vasoconstriction after alpha2 agonists could be alleviated by the calcium channel blocker isradipine. If peripheral, potentially undesirable, vasoconstrictive effects of alpha2 agonists would occur initially after their administration in humans, the present study indicates that the short-lasting administration of a short-acting calcium antagonist like isradipine could be used to rapidly antagonize these vasoconstrictive effects, while having no effect on the central sympatholytic and anesthetic qualities of the alpha2 agonist. In chapter 5, in order to find out whether the decrease in cardiac function after dexmedetomidine was secondary to its peripheral vasoconstrictive effect, we administered the purinoceptor vasodilating agent adenosine triphosphate (ATP) to reverse the vasoconstrictive effects of dexmedetomidine. As the reversal of the vasoconstrictive effect of dexmedetomidine was associated with only partial restoration of cardiac function, we concluded that the decrease in cardiac function after dexmedetomidine is mainly due to a decrease in sympathetic outflow. This study also showed that alpha2-adrenergic vasoconstriction could be completely antagonized by ATP. Extracellular ATP is converted rapidly by ecto-5'-nucleotidases into ADP, AMP and adenosine, which are important mediators of metabolic coronary vasodilatation during myocardial hypoperfusion. This encouraged us to further study the interaction between alpha-adrenergic coronary vasoconstriction and metabolic coronary vasodilatation. In chapter 6, we investigated the effect of dexmedetomidine during reactive hyperemia after 2 minutes coronary artery occlusions. This hyperemia is known to be mainly caused by the metabolic vasodilatation by adenosine. Although dexmedetomidine reduced myocardial oxygen demand and decreased blood flow in normally perfused myocardium, the supranormal blood flow levels in the endocardial and midmyocardial layers during reactive hyperemia were not influenced by dexmedetomidine. Lactate release during the reperfusion phases after dexmedetomidine were significant less than after the alpha2-antagonist atipamezole. From our findings in chapters 5 and 6, it appeared that the vasoconstriction of alpha2 agonists could be overruled by metabolic vasodilatation. These findings encouraged us to further study the new alpha2 agonists during experimental myocardial ischemia. In chapters 7 and 8, we found that dexmedetomidine and mivazerol reduced myocardial oxygen demand and, in parallel, decreased blood flow in nonischemic myocardium. In ischemic endocardium however, blood flow was preserved. Dexmedetomidine was found to increase the ischemic/non-ischemic blood flow ratio. Both dexmedetomidine and mivazerol reduced oxygen deficiency of ischemic myocardium. The peripheral vasoconstrictive effect of dexmedetomidine and mivazerol is thus restricted to the epicardial layer during myocardial ischemia, with preservation of blood flow in the more vulnerable endocardial and midmyocardial layers. Preservation of blood flow in ischemic myocardium by alpha2 agonists is probably caused by local metabolic stimuli during ischemia, which overrule adrenergic vasoconstriction. As the degree of ischemia is most severe in the inner layers during hypoperfusion, adrenergic vasoconstriction in this region is inhibited to a greater extent than in the outer layer. Specific epicardial vasoconstriction distal to a flow limiting stenosis which leads to improvement of endocardial perfusion is called the "reverse steal" effect. The findings of these studies are now being applied to patient care. (chapter 9). A preliminary study on the effects of mivazerol on perioperative hemodynamic stability and myocardial ischemia was carried out in 23 centres in 300 patients and published in abstract form; the allocated patient cohort for each centre was 12-13 patients. This study showed that mivazerol, when administered for 72 hours continuously perioperatively in patients at risk for coronary artery disease, decreased the incidence of tachycardia and hypertension. The incidence of bradycardia was increased in the treatment group, but there was no difference in the treatment for bradycardia. There was no increase in the incidence of hypotension in the drug group. Intraoperative myocardial ischemia was significant lower in the alpha2 agonist group, especially during the emergence period.
Our experimental studies reveal several mechanisms by which alpha2 agonists may have beneficial actions during myocardial ischemia:
The clinical study showed that the administration of mivazerol is safe in high-risk patients and decreases measures of myocardial ischemia, particularly during high- stress periods. Evidence is thus mounting that alpha2-adrenoreceptor agonists may decrease perioperative myocardial ischemia. However, although postoperative ischemia is a predictor of adverse cardiac outcome, there are no definitive data demonstrating that prevention of the postoperative ischemia will reduce adverse outcome. Therefore, further study is indicated in large-scale trials assessing the effects of the alpha2-adrenoreceptor agonists on cardiac outcome.
|
|
Blood-pressure management by automated drug infusion |
Dexmedetomidine as anesthetic adjuvant |
| In the early days of anesthesia a single agent such as ether was used to provide surgical anesthesia. High doses of this single agent were often required, resulting in delayed recovery and persistent nausea and vomiting. In modern, so-called 'balanced' anesthesia the total dose of anesthetic agent is reduced by using separate agents to provide the triad of sleep, pain relief, and muscle relaxation. By balancing the agents used, side-effects can be reduced and the quality of post-operative recovery improved. In addition, sedative drugs may be used as premedicants or adjuvants to further reduce the dose of the primary anesthetics. Benzodiazepines are frequently used in this context. To further improve the quality of anesthesia, anesthesiologists have become interested in alpha2-adrenergic receptor agonists since these agents have been shown to have an interesting combination of effects: sedation [1,21] anxiolysis [3,4], reduction of anesthetic requirements [5,6], and improved hemodynamic stability in the peri- operative period [7-9]. In addition, they provide analgesia which is not mediated via opiate receptors [10] and which is not associated with respiratory depression [11]. In veterinary anesthesia, a2-adrenergic receptor agonists were introduced as sedatives and anesthetic adjuvants in the 1980's and are now commonly used in this field [12-15]. Clonidine, an alpha 2-adrenergic agonist, which was first introduced into clinical use 25 years ago as an anti-hypertensive agent, has been extensively studied as a potential anesthetic adjuvant in a variety of settings including aortic and coronary artery surgery [7-9,16-18]. Dexmedetomidine is one of the second generation of alphalpha2-adrenergic agonists, and is a full agonist with a high selectivity for the alphalpha2-adrenergic receptor (1600:1 alpha2:alpha1, compared to 200:1 for clonidine) [19,20]. It has been shown to possess a greater anesthetic-sparing effect than clonidine in animals [21], and to have sedative [22] and hemodynamic stabilizing effects in human volunteers [23].
While these properties suggested promising anesthetic possibilities for selective alpha2-agonists like dexmedetomidine, studies in animals showed cardiovascular effects which might be undesirable in a potential anesthetic adjuvant: Flacke et al. demonstrated the high vasoconstrictive potency of dexmedetomidine [24], and Heusch and his coworkers indicated that alpha2-adrenergic receptor stimulation could under certain conditions elicit myocardial ischemia [25-27].
Aims of the studies In order to test the hypothesis that dexmedetomidine can be beneficially and safely used during anesthesia in healthy adults we developed two main aims:
To the first purpose we designed the following studies:
To the second purpose a double-blind placebo-controlled study was designed to investigate the potential benefits of dexmedetomidine when given as, a single intravenous dose prior to minor surgery in healthy patients (Chapter 8). References
Improvements in anesthesia and perioperative patient care have allowed patients to undergo surgery with a minimum of discomfort and complications. Although anesthesia can, during less stressful surgery, protect patients against the harmful effects of noxious stimuli, it is not always possible to provide optimal protection throughout all types of surgery without recourse to invasive techniques (like epidural analgesia) or large doses of anesthetic agents which may cause hemodynamic instability and prolong recovery. In order to provide stable anesthesia without recourse to high doses of anesthetics, other pharmacological agents which are not themselves anesthetics may be applied as anesthetic adjuvants. Recently, 2-adrenergic receptor agonists have been advocated as such anesthetic adjuvants since they provide sedation, and reduce anxiety, nausea and pain (Chapter 3). These effects are more apparent following the recently developed agents with a greater selectivity for the alpha2-receptors than the "classic" alpha2-adrenergic agonist clonidine. The anesthetic effects of dexmedetomidine and other alpha2-adrenergic agonists are due to stimulation of alpha2-adrenergic receptors in centres such as the locus coeruleus in the brain. Alpha2-receptors are found also in the walls of blood vessels where they cause vasoconstriction. Some animal studies raised concerns about the potential vasoconstrictive effects of alpha2-agonists on alpha2-receptors in peripheral and coronary arteries. Therefore, before exploring the use of dexmedetomidine in clinical anesthesia, we investigated its cardiovascular effects on the normal cardiovascular system in detail. Since more invasive studies on coronary blood flow and oxygen requirements of the heart can be performed in experimental animals, we studied the effects of dexmedetomidine on the cardiovascular system in anesthetized dogs and goats. Finally, we carried out a clinical study in humans to better evaluate the benefits and cardiovascular effects of a pre-anesthetic dose of dexmedetomidine on healthy patients undergoing minor surgery. In Chapter 4, we compared the hemodynamic effects of clonidine with dexmedetomidine in dogs using a range of doses. Both agents caused a transient increase in arterial blood pressure and systemic vascular resistance and a longer-lasting decrease in heart rate and cardiac output, but dexmedetomidine was 3-10 times more potent than clonidine for the pressor effects. Within 15 min the pressor effect recovered considerably for doses < 3 : microg/kg of dexmedetomidine and clonidine. High dose clonidine (10 and 30 : microg/kg) induced prolonged coronary vasoconstriction and reduction in myocardial contractility. These results indicate that slow intravenous administration is more important for the potent and selective alpha2-adrenergic agonist dexmedetomidine than for clonidine, and that dexmedetomidine lacks the ontoward alpha1-adrenergic effects, occurring at high doses of clonidine. In Chapters 5 and 6 we describe the hemodynamic effects of dexmedetomidine in dogs anesthetized with either chloralose/urethane or fentanyl/halothane. The rapid, transient increase in blood pressure, systemic vascular resistance and left ventricular end- diastolic pressure, and longer-lasting reduction in cardiac output and heart rate was similar in both dog groups. Fifteen minutes later the effect was dependent on anesthetic technique. In chloralose/urethane anesthetized dogs, blood pressure decreased, whereas in fentanyl/halothane anesthetized dogs blood pressure remained elevated. Thus, anesthetic technique influences the cardiovascular effects of dexmedetomidine in dogs. Since sympathetic tone is known to be greater in dogs anesthetized with chloralose/urethane it is likely that the stabilizing effect of dexmedetomidine on heart rate and blood pressure is greater in situations where the initial sympathetic tone is high. This stabilizing effect in chloralose/urethane anesthetized dogs was also accompanied by decreased myocardial energy requirement, while myocardial oxygen supply decreased in parallel with demand. Myocardial oxygen extraction increased only after the highest dose of 10 : microg/kg dexmedetomidine, but even at this dose the heart continued to extract lactate, indicating the absence of global myocardial ischemia. These data indicate that for dexmedetomidine < 3 : microg/kg, the reduction in myocardial blood flow is functional; by metabolic vasoregulation rather than adrenergic vasoconstriction. Although total myocardial blood flow is reduced after dexmedetomidine, endocardial epicardial blood flow ratio is increased, safeguarding the layer most vulnerable to ischemia (Chapter 5).
In Chapter 6, a study of the distribution of cardiac output using radioactive microspheres, we show that the reduction in cardiac output following dexmedetomidine is not the result of an equal reduction of blood flow to all organs. Blood flow to the vital organs (heart, brain, kidneys) is spared and greater reductions occur in the blood flow to less vital organs such as the skin, intestine and spleen and through arterio-venous anastomoses. After studying the cardiovascular effects of dexmedetomidine in the dog, a species prone to alpha-adrenergic vasoconstriction, we investigated the goat. We found that the pressor effect of dexmedetomidine in goats was shorter than in dogs, and therefore more like that seen in humans. In goats, dexmedetomidine 10 : microg/kg reduced myocardial oxygen extraction (Chapter 7), whereas in the dog this dose increased it (Chapter 5). This indicates that the balance of myocardial oxygen supply and demand is maintained up to a higher dose of dexmedetomidine in goats. If similarity in pressor response is indicative of the coronary vasoconstictive effect in goats and humans, humans may be less prone to coronary vasoconstriction following dexmedetomidine.In the clinical study during minor surgery in healthy adults, we showed that a single intravenous dose of 2 : microg/kg dexmedetomidine given over 5 min prior to anesthetic induction reduced the requirements for isoflurane by 90%, for thiopental by 14% and for fentanyl by 33%. Plasma catecholamine levels were lower in the dexmedetomidine patients, indicating a sympatholytic effect. Dexmedetomidine reduced the hemodynamic response to intubation and emergence from anesthesia and increased hemodynamic stability in the peri-operative period. Also, there was a reduced requirement for post-operative analgetic and anti-emetic medication. While the incidence of peri-operative bradycardia was high in these patients, this did not produce adverse hemodynamic effects (Chapter 8). Therefore, we have demonstrated that the hemodynamic effects of dexmedetomidine do not represent a contraindication to its use during anesthesia. On the contrary, the central sympatholysis together with the many other beneficial effects demonstrated in this thesis suggest that dexmedetomidine is a an excellent agent for use as an anesthetic adjuvant in healthy patients.
|
Anesthetic interactions with lysophosphatidate signaling |
After more than a century of investigation, it appears that a common site of action for volatile and most non-volatile anesthetics may have been identified. The GABAA ligand-gated ion channel has for some time been recognized as the primary site of action for non-volatile drugs, such as barbiturates, benzodiazepines and steroid anesthetics. Propofol also appears to exert its primary action at this ionotropic receptor. Recent data indicate that volatile anesthetics, in clinically relevant concentrations, similarly enhance GABAA signaling, and a defined molecular target on the GABAA receptor has been localized. Thus, it appears that a unitary model of anesthesia may actually be the correct one. Ketamine, with its primary action at the NIVMA glutamate receptor, appears one of the few exceptions to this model. However, if the GABAA receptor were the only site of action for anesthetics, their side effects would be expected to be the same. This is obviously not the case. Therefore, if most anesthetics indeed induce anesthesia through the single mechanism of GABAA enhancement, their varying side effect profiles have to be explained through interactions with other systems. Particularly for volatile drugs this is not surprising, as these require very high (low millimolar) concentrations for clinical effect, and their lipophilic nature makes them likely to interact (possibly non-specifically) with a variety of lipophilic sites. In contrast, non-volatile anesthetics are usually therapeutically active in micromolar concentrations. They would be expected to interact with fewer sites, and likely in a specific manner. Thus, one of the research challenges in the years to come will be to elucidate, at the molecular level, anesthetic sites of interaction in addition to the GABAA receptor. These secondary actions of anesthetics will define their unique side effect profile and can aid in the development of drugs with fewer adverse actions. This thesis reports some of our research in this area. Clearly, the number of potential secondary molecular targets for anesthetics is overwhelming. Some have been probed in significant detail, many others have not been studied at all. In an effort to narrow the field, we focused on the physicochemical parameter that has driven anesthetic mechanisms research since the beginning of the century: the Meyer-Overton hypothesis. Most anesthetics are lipophilic, therefore lipophilic targets may be likely sites of secondary anesthetic interaction. One clear example of such lipophilic targets is G protein-coupled receptors for lipophilic mediators. These include receptors for a variety of compounds relevant to anesthesiology and critical care, including the prostaglandin and leukotriene roceptors, the thromboxane receptor and the platelet-activating factor receptor. Study of interactions between anesthetics and G protein-coupled receptors requires an appropriate model: one where receptors can be expressed functionally and in isolation, where they can be compared with each other in a similar setting, where volatile and non-volatile anesthetics can be administered in a reproducible manner, and where anesthetic effects on intracellular signaling can be studied in detail. Such a model was not established at the time, and we therefore developed one using Xenopus oocytes, already used widely for the investigation of G protein-coupled receptors, and used previously to study anesthetic effects on ion channels. As an initial test of the system, we decided to focus on a lipid mediator receptor which we had found to be already expressed in oocytes: a novel receptor for the phospholipid lysophosphatidate (LPA). Although the (patho) physiology of the compound is as yet poorly described, it is known to induce (among other actions) platelet aggregation, contraction of smooth muscle (including vascular smooth muscle), changes in cellular morphology and mitosis. We felt that it would form a useful model for the study of anesthetic interactions with lipid mediator signaling. In addition, our studies would enhance the understanding of the physiology of LPA signaling.
We hypothesized that different anesthetics interfere to different degrees with LPA signaling. These effects are localized at the membrane receptor rather than in the intracellular signaling pathways. Specifically, the goals of the study were to:
This thesis reports the results of investigations planned to accomplish these Aims. lt is organized as follows:
This thesis reports the development of a system for the study of molecular interactions between G protein-coupled receptors and anesthetics. Although the main site of action for most intravenous and inhaled anesthetics appears to be the GABAA receptor (ketamine's action on the NMDA glutamate receptor is an exception), the different effect and side effect profiles of these drugs is a result of their interactions with other systems. As discussed in Chapter 2, ion channels - both voltage-regulated and ligand-gated - have received most attention as sites of anesthetic action. Indeed, many channel types have been shown to be affected. The nicotinic acetylcholine receptor is an example of an ionotropic receptor significantly inhibited by low concentrations of anesthetics. Voltage-gated Ca and K channels are similarly inhibited by inhaled anesthetics. In contrast, the large superfamily of G protein-coupled receptors (GCRs) has received relatively little attention. Although scattered reports exist of anesthetic interactions with GCRs, only the effects on muscarinic acetylcholine signaling have been investigated in some detail. Part of this lack of investigative attention is due to technical difficulties in studying GCRs in isolation. Therefore, we adapted the Xenopus oocyte model, long used for the expression and cloning of GCRs, to the study of anesthetic interactions with these signaling molecules. Of particular interest would be GCRs for lipid mediators, as these would be likely targets for (lipophilic) anesthetics. As reviewed in Chapter 4, the Xenopus oocyte has a number of compelling advantages for use in such studies. The frogs are easily maintained and oocytes are obtained with a simple procedure. In addition, once oocytes are adequately defolliculated, they are remarkably devoid of endogenous GCRs, with the exception of those described in the present thesis. GCRs can be expressed either by microinjection of RNA (extracted from tissue or prepared in vitro) or cDNA. Particularly Ca2+-signaling GCRs are easily studied, as the oocyte expresses an endogenous Ca2+-dependent Cl- channel, which is routinely used as a reporter for intracellular Ca2+ release. The results of the investigations in this thesis demonstrate that anesthetics of various classes can be applied without significantly modifying the behavior of the oocyte. Chapters 5 and 6 describe our observations of two endogenous GCRs which are expressed consistently in oocytes: those for the phospholipid messengers lysophosphatidate (LPA) and sphingosine-1-phosphate (S1P). These two compounds are closely related structurally and functionally. As described in Chapter 3, LPA signaling has received much investigative attention recently. The compound has long be known to be an essential intermediate in phospholipid metabolism, but only in recent years has it become clear that it also acts as a signaling molecule. Its cellular effects are varied and dramatic. The compound acts as a contractile agent on various types of smooth muscle, including vascular smooth muscle. It is a potent mitogen for many types of fibroblasts. It activates platelets and - in turn - is released into the blood stream from activated platelets. It induces shape changes in a number of cell types, including neuronal cells. Its relevance to biological systems is shown by the fact that LPA acts as a chemoattractant in a species as far removed from mammals as Dictyostelium discoideum. A variety of studies over the past ten years have provided detailed insight into the cellular mechanism of action of LPA. Although a membrane receptor has not yet been cloned convincingly, biochemical evidence strongly supports its existence. LPA binding to this receptor leads to intracellular activation of several G proteins, inducing IP3 and DAG generation, as well as decreases in cAMP levels. The mitogenic and cytoskeletal actions of the compound are mediated by small cytoplasmic G proteins and the MAP kinase cascade. Thus, LPA is a molecule of significant interest. Some of its actions may be relevant in the perioperative period, and anesthetic effects on LPA signaling are therefore worthy of investigation. As in addition the compound forms a good model for lipid mediator signaling in general, and the receptor is endogenously expressed in oocytes, we decided to study the effects of several classes of anesthetics on LPA signaling. Chapter 5 reports our finding of the endogenous LPA response in Xenopus oocytes. LPA induced transient dose-dependent currents in uninjected cells. The response was found to be highly specific, as related compounds (phosphatidic acid, lysophosphatidylcholine, lysophosphatidylserine) were without effect. Intracellular LPA application did not induce intracellular Ca2+ release, strongly suggesting an extracellular site of action. In addition, the response to LPA could be inhibited by incubation in pertussis toxin, indicating G protein involvement. We could establish that the induced currents were Ca2+-activated Cl- currents, as they were abolished by intracellular injection of EGTA, and the reversal potential became more positive at lower extracellular Cl- concentrations. Suramin, which had been shown to inhibit LPA responses in other systems, blocked the current. Hence, we concluded that this novel oocyte response to LPA was mediated by a specific membrane receptor linked to a penussis toxin-sensitive G protein. When searching for related responses in oocytes, we tested S1P, and found it also to induce currents. As detailed in Chapter 6, these currents were similarly Ca2+-activated Cl- currents, virtually indistinguishable from those induced by LPA. Intracellular injection of S1P was without effect, as was extracellular application of several compounds structurally related to S1P (sphingosine, sphingosylphosphorylcholine and N,N-dimethylsphingosine). In order to investigate if S1P and LPA might signal through the same receptor, we employed cross-desensitization, and found that LPA and S1P cross-desensitized completely. Both responses were inhibited by suramin and dithiothreitol. Thus, this suggests the existence of a single receptor for the two com- pounds. We were able to elicit responses to both LPA and S1P in HEK293 fibroblasts, but to neither of the compounds in K562 cells. However, no cross-desensitization between the compounds was observed in HEK293 cells. As standard electrophysiology systems are not designed to function effectively with the long time courses and large currents induced by GCR signaling in oocytes, we developed a software system specifically geared to this function, as described in Chapter 7. OoClamp, as it is called, is a MS-DOS based package for the acquisition, analysis and storage of data from GCR studies in Xenopus oocytes. The system is designed to provide standardization of test conditions, rapid, on-line analysis of data, and self-documentation and compact storage of data files. All system settings are optimized for use with the Xenopus expression system. This system has been used for data acquisition in all subsequent studies reported in this thesis. In the course of these investigations, we discovered an additional endogenous oocyte response: a current induced by trypsin. As described in Chapter 8, trypsin induces Ca2+-activated Cl- currents when applied in concentrations as low as 0.1 mg/ml to defolliculated, uninjected oocytes. The response is dose-dependent and specific, as other proteases (chymotrypsin, Lys-C and Arg-C) or trypsin pretreated with soybean trypsin inhibitor, were without effect. The response was due to an extracellular site of action, as microinjected trypsin did not induce currents. We also tested if the trypsin response could be due to proteolytic activation of LPA signaling, but found this not to be the case. After these initial studies of oocyte physiology, we studied the effects of three classes of anesthetics on LPA signaling in oocytes. Chapter 9 reports the effects of propofol, a highly lipophilic intravenous anesthetic known to have its primary site of action on the GABAA receptor. Propofol in Intralipidâ dose-dependently inhibited LPA signaling (IC50 5.38 µM). Propofol 28 µM inhibited LPA by more than 80%. Intralipid® (0.01%) was without effect. To ascertain that intracellular signaling pathways and the Ca2+-activated Cl- channel were not affected by propofol, we tested the effects of propofol on currents induced by methylcholine in oocytes expressing m1 muscarinic acetylcholine receptors. No inhibition was observed. As both receptors share the same intracellular signaling pathway, we concluded that clinically relevant concentrations of propofol inhibit LPA signaling, acting most likely on the receptor or the associated G protein. Thus, lipophilic anesthetics indeed interfere significantly with signaling through lipid mediator receptors. Chapter 10 describes the effects of two volatile anesthetics, halothane and isoflurane, on LPA signaling. We found halothane to inhibit LPA signaling in a dose-dependent manner (IC50 0.23 mM). Halothane 0.34 mM inhibited responses to LPA virtually completely. This effect was fully reversible. As we had shown previously that halothane inhibits functioning of muscarinic receptors, we used instead the AT1A angiotensin receptor to test for anesthetic effects on the shared intracellular signaling pathway. Halothane, even at the highest concentrations tested, was without effect on angiotensin signaling. To assure that both receptors signaled through the same pathway, we microinjected oocytes with heparin, an IP3 receptor blocker. Responses to both LPA and angiotensin were inhibited by this treatment, confirming that both receptors signal through IP3. Currents elicited by microinjection of IP3 into oocytes were not affected by halothane. Together, these findings confirm that the effect of halothane is proximal in the signaling pathway, most likely at the LPA receptor or G protein. In contrast to halothane, isoflurane was without effect on LPA signaling. Thus, volatile anesthetics can interfere with lipid mediator signaling, but the action is specific. Finally, we studied the effect of two local anesthetics, lidocaine and bupivacaine, on LPA signaling, as described in Chapter 11. This study was of particular relevance as a role for LPA in wound healing seems likely, and local anesthetics have been shown to impair wound healing. Both compounds inhibited LPA signaling (IC50 29.9 mM for lidocaine and 4.5 mM for bupivacaine - concentrations reached after local anesthetic injection aroud wounds). At high local anesthetic concentrations, virtually compete inhibition of LPA signaling was observed. The effect of bupivacaine was completely reversible, whereas lidocaine appeared to have a partially irreversible effect. Both compounds were without effect on angiotensin or IP3 signaling. Taken together, these studies show that the Xenopus oocytes is a flexible model for the study of anesthetic actions on GCR signaling, and that selected members of the three main classes of anesthetics interfere with at least one lipid mediator receptor. Further studies suggest themselves. First, more detailed localization of the site of action is useful. We have now been able to induce Ca2+-activated Cl- currents in oocytes by microinjection of the non-hydrolyzable GTP analog GTPgammaS, and have found that halothane is without effect on these currents. Thus, the site of action of halothane is more proximal than the G protein effector domain, and therefore either involves the receptor, the receptor-G protein interface or early steps in G protein signaling (such as GDP-GTP exchange). We have obtained similar results have been obtained with local anesthetics. We have shown in addition that charged local anesthetics inhibit the LPA receptor at an intracellular site, using the non-permeable, permanently charged lidocaine analog QX314. Permanently uncharged compounds, exemplified by benzocaine, can also block LPA signaling, and as they interact with QX314 in a synergistic manner, the presence of two binding sites is suggested. It would be of great interest to test stereoisomers of some of the anesthetics, as a stereoselective effect would suggest a protein (receptor) site of action. Such studies are now in progress with the isomers of ropivacaine. In addition, it is important to extent these results to other lipid mediator receptors. We have chosen the thromboxane A2 (TXA2) receptor because of its relevance in vasoconstriction and platelet aggregation, effects similar to those of LPA. We found both general anesthetics and local anesthetics inhibit to TXA2 signaling. Both isoflurane and halothane inhibit signaling, but competition studies suggest different site of action. This is in agreement with data obtained at other receptors, as discussed in Chapter 3. The effect of local anesthetics is profoundly timedependent. After several hours incubation, concentrations of lidocaine and bupivacaine achieved in plasma during epidural anesthesia are found to inhibit TXA2 signaling significantly. The mechanism of this delayed effect is under study. It is also important to confirm findings obtained in a highly isolated model such as the oocyte to receptors in a more native environment. Therefore, studies of the effects of anesthetics on, e.g. LPA-induced platelet aggregation should be performed. The number of lipid mediator receptors is large, and many of them play important roles during injury and repair. Thus, their functioning is of great importance during the perioperative period. Our data suggest that they may be targets for anesthetic action, and that anesthetics of several classes can significantly limit their functioning. This may have consequences for the bodily response during and after surgery, and a full understanding of these anesthetic-receptor interactions and their consequences is therefore warranted.
|
|
Effects of volatile anesthetics on the function and perfusion of the collateral dependent myocardium |
|
|
|
Landouzy-Déjérin disease: a computed tomographic study of the skeletal muscles |
| Thirty-one patients with an autosomal dominant facioscapulohumeral dystrophy were investigated with regard to their clinical picture. Special attention was payed to the computed-tomographic images of the patient's different muscles. The findings of this investigation were compared to those of isolated manual muscle testing. The patient's heart was examined for involvement in the disease process by means of electrocardiography, 24-hours electrocardiography, echocardiography and the estimation of the serum CK-MB activity. Serum myoglobin concentration and CK activity were concomitantly estimated. Four computed-tomographic slices in the facioscapulohumeral region were made for each patient. The slice in the region of the head informed us about the computed-tomographic aspects of the medial and lateral pterygoid and masseter muscles. No patient showed an abnormal computed-tomographic aspect of the above-mentioned muscles A second slice was made at the level of the sixth cervical vertebra. Twenty-nine percent of the patients showed a hypodense aspect of the sternocleidomastoid muscles. The overwhelming majority (94%) of the computed-tomographic images of the extensors of the neck were normal, which is in agreement with the findings of isolated manual muscle testing. The slice through the shoulder-girdle presented the most extensive computed-tomographic abnormalities of those muscles which were found to be paretic by isolated manual muscle testing. Some striking discrepancies were found between the two kinds of investigations. To begin with, in the case of the deltoid muscle, 87% of the patients investigated showed a paresis of one or more of this muscle's constituent components. On the other hand only 9 patients (29%) showed computed-tomographic abnormalities of this muscle. Seven of the 9 patients had an illness of more than 20 years' duration. The whole deltoid muscle was computed-tomographically abnormal in only 3 patients (10%). Only the posterior part of the deltoid muscle showed computed-tomographic abnormalities in 4 patients (13%). The high incidence of a paretic deltoid muscle found in isolated manual muscle testing in our investigation as well as that of Padberg (1982) - he found a paresis of this muscle in 74% of his patients - is a result of the fact that the deltoid is hardly to test in this disease. As a result of the poor fixation of the scapulae in this disease, a diminished function of the deltoid is found rather than a diminished strength. In view of the low percentage (29%) of computed-tomographic abnormalities of the deltoid, a sparing of this muscle in the course of time is to be concluded. A second remarkable finding of the computed-tomographic slices through the shoulder-girdle was the fact that only one of the 22 slices which imaged the infraspinatus and subscapularis muscles, showed computed-tomographic abnormalities. This is in contrast to the isolated manual muscle testing findings, which showed a diminished function of these muscles in more than half of the patients. Hereby it should be noted that the subscapularis muscle is not to be tested manually while the infraspinatus can be manually tested only with the greatest difficulty, once again as a result of poor fixation of the scapula. The same was true for the teres major muscle. Fourteen patients (45%) had a paresis while only 3 patients (10%) showed computed-tomographic abnormalities of this muscle. The sparing of these muscle in the course of time is reasonable. The pectoral minor muscle, the sternocostal part of the pectoral major muscle and the rhomboid muscle showed in according to the clinical findings most computed-tomographic abnormalities. Twenty-four slices (39%) through the middle of the upper arm showed computed-tomographic abnormalities of the biceps. More than half of these slices the biceps appeared to be completely hypodense.
All but one patients with a duration of illness of more than 20 years had a complete hypodensity of one or more muscles of the upper arm. Noteworthy was the fact that only 2 patients (6%) with a duration of illness of less than 20 years had computed-tomographic abnormalities of the upper arm muscles. The same was true for the clinical findings. Only 5 patients (16%) had a paresis of the upper arm muscles. The facioscapulohumeral pattern of muscle paresis - as the name of this disease suggests - was not consistently found in the patients mentioned above. The latissimus dorsi and serratus anterior muscles frequently showed bilateral complete hypodensity (61%). Five slices were made outside the facioscapulohumeral region. Those which showed the anterolateral abdominal wall muscles, demonstrated computed-tomographic abnormalities in 26 patients (84%). In contrast to the above only 62% had a paresis of these muscles. Seven patients (22%) showed computed-tomographic abnormalities of the abdominal wall muscles and no paresis. Eight of the 26 patients with computed-tomographic abnormalities of the abdominal wall muscles had a duration of illness of more than 20 years. Most abnormalities were localized in the rectus abdominis and internal oblique abdominal wall muscles. The two slices in the pelvic region yielded computed-tomographic abnormalities which were characterized by hypodense areas as far as the gluteus maximus and medius muscles are concerned. An exception to the fore-mentioned was found in the case of the gluteus minimus muscle which was mostly completely hypodense. The whole gluteal muscluculature was completly hypodense in only 3 patients (10%). The total percentage of paresis of the gluteus maximus and medius muscles (43%) corresponded to that of computed-tomographic abnormalities of these muscles. In contrast to the clinical findings (44%), computed-tomographic abnormalities were seldom found in the iliopsoas muscle (26%). An explanation for this discrepancy could be the involvement of the rectus femoris muscle as flexor of the hip. Of the remaining muscles which could be identified on the two slices in the pelvic region, the pectineus, adductor brevis and obturatorius externus muscles often had a normal aspect. In patients with a duration of illness of more than 20 years the tensor fasciae latae and sartorius muscles were involved. In 12 patients with a duration of illness of less than 20 years the slices trough the thigh showed computed-tomographic abnormalities of the hamstrings (3 times) and the rectus femoris muscles (twice). Six of the 12 patients had a paresis of the hamstrings. These very muscles showed the most extensive computed-tomographic abnormalities, which means complete hypodensity, in patients with a duration of illness of more than 20 years. The same was true for the adductor magnus and longus muscles. In view of the small number of computed-tomographic abnormalities of the adductor brevis and pectineus muscles, the paresis of the adductors of the hip is due to an involvement of the remaining two. The total percentage of paretic adductors of the hip (58%) differed slightly from the findings of the computed-tomographic investigation (53%). A paresis of the quadriceps femoris muscle was found 26 times (42%). On the contrary a computed-tomographic abnormality was found in one or more of the constituent parts of the quadriceps femoris muscle in 39 images of this muscle (63%). This difference was mainly based on an involvement of the rectus femoris muscle, which showed computed-tomographic abnormalities 8 times while no paresis was found.
A striking contrast between the severely involved hamstrings and the normal or slightly abnormal image of the vastus medialis, intermedius and lateralis muscles was found in 12 of the 19 patients with a duration of illness of more than 20 years. The adductor magnus muscles were never completely hypodense in the case in which a slight involvement of the hamstrings was found. The findings in isolated manual muscle testing confirmed the high percentage of paresis of the hamstrings (60%). The slices through the lower leg showed a typical pattern of involvement, namely a picture of a completely hypodense anterior tibial and extensor digitorum longus muscles, whereas the peroneus longus had a normal or sligthly abnormal computed-tomographic aspect. The same time the medial head of the gastrocnemial muscle could also be completely hypodense. This was not concluded from the findings of isolated manual muscle testing. In spite of a normal strength of the gastrocnemial muscle, a computed-tomographic abnormality of the medial head of this muscle was found on 16 images (26%). One could get an impression of a sequential involvement of the different muscles when comparing the results of isolated manual muscle testing and skeletal muscular computed-tomography correlated to the duration of the disease. According the findings of isolated manual muscle testing,the abdominal wall muscles, the hamstrings, the anterior tibial and extensor digitorum longus muscles were involved in early stages of the disease. The adductor magnus and longus muscles, the gluteus minimus and rectus femoris muscles were involved in the disease process later on. Moreover it became obvious that of the sural muscles - thought to be spared until a late stage of the disease - the medial head of the gastrocnemial muscle, too, was affected at an early stage. The low percentage of complete hypodensity of the gluteus maximus and medius muscles, the vastus medialis, intermedius and lateralis muscles and the long muscles of the back suggest their involvement in a late stage of the disease. A relative sparing was seen in the following muscles: adductor brevis and pectineus muscles in the,pelvis region, the peroneus longus muscle, deep flexors and the remaining muselcs of the calf. At the same time more than two thirds of the patients showed a asymmetry of computed-tomographic abnormalities between the left and right side of the body. Density measurements of the muscles of the lower extremities of 8 patients with a facioscapulohumeral dystrophy were compared with those of 6 healthy persons of the same age. As a result of the presence of bone artefacts some density measurements became unreliable. Therefore, a great dispersion of the computed-tomographic density values (CT-values) for these muscles arose. The average CT-values of the muscles with an artefact in the control was 23 HU with a standard deviation of 18 HU. The muscles without an artefact had a average CT-value of 35 HU (standard deviation 10 HU). These values were considerably lower than the average values found by Bulcke et al. (1979). These investigators found CT-values varying from 55 to 74 HU. A possible explanation for this discrepancy is the fact that their values were obtained from muscles with and without an artefact. Bulcke et al. did not constantly correct their CT-values calibrating for water density by simultaneously scanning a bag of water. A different scanning technique could also offer an explanation for the above-mentioned difference. In view of the unreliability of certain CT-values of muscles with an artefact, statistical analysis was only applied to the muscles without an artefact. No difference between the CT-values of the left and right side of the body could be established in the controls. The same applied for the patient's group. The three different values which were measured in every muscle in controls and patients did not differ among themselves. In comparing the CT-values measured-in the largest field, a significant difference between some muscles of the controls and patients became obvious. The average CT-values of the patients' group was lower (24 HU) than the CT-values of the controls (35 HU). A difference could be demonstrated for the semitendinosus, semimembranosus and the biceps longus (long head) muscles. The CT-values of the rectus femoris in both groups also differed significantly at the same time. However the average CT-values of the patients’ group (41 HU) was higher than that of the controls (35 HU). The lower CT-values in the patients’ group indicate existence of infiltration of fat in some muscles, in spite of the apparently normal aspect of the computed-tomographic images. The findings of the rectus femoris muscles, however, do not confirm this.
In view of the low number of measurements of computed-tomographically normal muscles, which were not paretic. no conclisions could be drawn. There was no difference between the CT-values of the computed-tomographic abnormal muscles with and without an artefact. The average CT-values in both groups was -65 HU with a standard deviation of 39 HU. As a result of frequent existence of bone artefacts in computed- tomography of the skeletal muscular system, the diagnostic value of density measurements is limited. Electrocardiography showed abnormalities in only one of the 31 patients with an autosomal dominant inherited facioscaputohumeral dystrophy. This patient had a RIS ratio over 1 in lead V1 as a result of a serious thorax deformity. None of the 31 patients showed rhythm disturbances by 24-hours electrocardiography. No indication for existence of a cardiomyopathy could be found by 23 echocardiographies, which were judged to be reliable. Forty-two percent of the patients showed a raised serum CK activity in laboratory investigations. A raised serum myoglobin concentration was found in 71% of the patients. Remarkable was the fact that 11 patients (35%) had a normal serum CK activity, whereas the serum myoglobin concentration was raised. This fact is in contrast with the findings of Kagen et al. (1980) and Asmark et al. (1981), who found the very opposite. The serum CK-MB activity was elevated in 7 patients (24%) and in view of the negative findings in cardiology investigations no relation exists between this and an involvement of heart muscle.
In trying to answer questions postulated in chapter V and taking account of the findings of clinical neurological, cardiological and computed-tomographical investigations, the following conclusions with regard to the autosomal dominant inherited facioscapulohumeral dystrophy can be drawn:
|
|
Local anesthetic interactions with G protein-coupled receptor signaling |
This thesis reports the interaction of local anesthetics with various signaling pathways of G protein- coupled receptors. Although local anesthetics interact primarily with the sodium channel, from which interaction derive their analgesic and antiaarhythmic effects, not all actions of these compounds can be explained by sodium channel blockade. Their interaction with GPCR signaling pathways may explain some of the "alternative" effects of local anesthetics, (e.g. antithrombotic and antiinflammatory properties). As reviewed in chapter 2, recent experimental results have modified profoundly our understanding of G protein-mediated transmembrane signaling. The concept of linear signal transduction pathways, i.e. one receptor coupling to one G protein that activates one effector, is inadequate to explain recent findings. The majority of seven transmembrane spanning receptors interact with diverse G proteins and elicit multiple intracellular signaIs. Interaction of a single G protein with a given receptor in a certain cell, however, may be regulated by a high degree of selectivity. G protein-mediated signal transduction can be seen as a complex, highly organized signaling network with diverging and converging transduction steps at the ligand-receptor, receptor G protein, and G protein-effector composition. Although the a-subunit gained most attention for its important role in interactions with receptor and effector structures, and G proteins are named according to their a subunit, our knowledge of Gß?'s role in signal transduction bas advanced from the assumption that it simply anchored and regulated Ga to the demonstration that the ß-isoform in the ß? dimer also can determine the specificity of signaling at both effectors and receptors. Chapter 3 bas discussed several aspects of local anesthetic action that are considered independent of Na channel blockade. Local anesthetics provide a modest degree of general anesthetic action, and as such can be usefully employed to 'stretch' an anesthetic (e.g. at the end of a case). They also provide a considerabIe amount of neuroprotection. In combination with the hemodynamic stability provided, they are therefore of great usefulness in the practice of neuroanesthesia. The ability to reduce bronchial reactivity is weIl recognized, even if the underlying mechanism is poorly understood. Finally, these versatile compounds have been employed with success in a variety of other clinical settings, including tinnitus, pruritus and migraine. Chapters 4 and 5 described local anesthetic inhibition of muscarinic m1 and m3 acetylcholine receptors expressed recombinantly in Xenopus oocytes. Their action on these signaling pathways might explain at least in part some of the additional effects of local anesthetics reported in chapter 3. As detailed in chapter 4, we found that local anesthetics inhibit muscarinic m1 receptors expressed recombinantly in Xenopus oocytes. Lidocaine inhibits at concentrations significantly less than those required for blocking sodium channels. We suggest that this inhibitory effect is due to superadditive interactions between an extracellular, polar, non-competitive site on the muscarinic receptor molecule, and an intracellular site probably on the coupled G protein. Whereas the intracellular site appears to be the same on muscarinic, lysophosphatidate and thromboxane A2 receptors, the lipid mediator receptors lack the extracellular polar local anesthetic binding domain.
Chapter 5 reports our finding that lidocaine reversibly inhibits signaling of m3 receptors recombinantly expressed in Xenopus oocytes. The calculated IC50 (370 nM) is significantly less than that required for blocking sodium channels. However, compared with inhibition of m1 muscarinic signaling (IC50 18 nM), m3 signaling is approximately 21-fold less sensitive to lidocaine. This discrepancy is explained most parsimoniously by absence on the m3 receptor of the major extracellular binding site for charged local anesthetics that is present on the m1 receptor. The N-terminus and third extracellular loop of the m1 receptor molecule were identified as requirements for this extracellular binding site for charged local anesthetics. In contrast, intracellular inhibition of both receptors was quite similar. As determined for m1 signaling, m3 signaling is primarily mediated by Gq and G11; of these, Gq was shown to be a likely target for intracellular inhibition by local anesthetics. In summary, this study adds to previous findings that local anesthetics at clinically relevant concentrations can inhibit G protein-coupled receptors. The interactions between the anesthetics and the receptor pathways are complex, and involve multiple sites of action, on receptor as well as coupled G protein. In view of the important roles of m1 and m3 muscarinic signaling in the brain and peripheral tissues, local anesthetic inhibition of these receptors is likely to be of relevance. The clinical implications remain to be addressed. To elucidate in more detail a common intracellular site of action for local anesthetics within the signaling pathway of various G protein-coupled receptors, we determined the G protein-a-subunits coupling to those receptors using antisense methodology. As described in chapter 6, LPA signaling was shown to be inhibited by ropivacaine stereoisomers in a concentration-dependent and stereoselective manner, strongly suggesting a protein site of action for ropivacaine. This inhibition was primarily due to a non-competitive antagonism. We also found that LPA signaling is mediated primarily by Gaq and Gao. Gaq couples to bath LPA, muscarinic m1, and trypsin receptors, and is a main target for intracellular LA inhibition of G protein-coupled receptors. In conclusion, our studies suggest that G protein-coupled receptors may be common targets for local anesthetics. The concentrations used in these investigations are routinely attained after local injection of these compounds. Inhibition of G protein-coupled receptors by LA results in part from an intracellular action, which can be largely explained by selective interference with Gaq protein function.
Chapter 7 reviewed the effects of local anesthetics on inflammation, coagulation and wound healing. In this chapter we have summarized several interesting and potentially important "alternative" effects of LA, not explained by their well-known antinociceptive and antiarrhythmic actions. The most remarkable observation is that LA are able to prevent pathological changes such as hypercoagulability or excessive stimulation of the inflammatory system, without inducing increased bleeding or impairment of host defense. This sets them apart from drugs currently in use for treatment of such disorders, and points the war to potential therapeutic application. Indeed, we use intravenous LA infusions in patients who would benefit from epidural anesthesia/analgesia but are not candidates for the technique. We hope that this chapter will urge same readers to investigate these effects in more detail, because much more research is needed on basic mechanisms. What does seem clear is that Na+ channel blockade plays only a limited, if any, role in these effects. Further research should determine which molecular determinants of the LA structure exert these effects and where the corresponding site of action is. This might eventually lead to development of new drugs, selective for treatment of these disorders, but without the "side effect" of Na+ channel blockade. Chapter 8 provides new insights into the mechanism of hPMN priming, and in addition suggest a mechanism by which LA may exert same of their anti-inflammatory actions. We have shown that platelet-activating factor (PAF) primes neutrophils through a pathway dependent on PTX- insensitive G proteins, PLC and PKC. PKC activation is bath necessary and sufficient for this process. In addition, we show that clinically relevant concentrations of local anesthetics selectively inhibit priming, but not fMLP-induced activation. Ester-LA exerted the most profound inhibitory effect, whereas inhibitory potency of amide-LA increased with increased uncharged fraction. The main target site for LA in the PAF priming pathway is located upstream of PKC. Finally we studied the preventive effects of epidural anesthesia on hypercoagulation in patients undergoing major orthopedic surgery, as reported in chapter 9. The major findings of this study were twofold. First, the hypercoagulable state induced by major orthopedic surgery, which is not easily demonstrated by routine coagulation tests, is reliably identified by the CSA parameters PHT and CT. The third parameter, CITF, showed a tendency towards shortening; however, this effect was not significant. Second, this hypercoagulability is completely prevented in patients receiving epidural anesthesia. We found no significant differences between knee and hip surgery. In summary, major orthopedic surgery induces a notable hypercoagulability, which is reliably determined by CSA, and prevented by epidural anesthesia. Further investigations should focus on several issues. First, selective interaction of LAs with Gaq-function has to be confirmed in other models and the mechanism of action needs to be defined. Second, the active part of the LA molecule responsible for this selective interaction with G protein function should be determined. Such studies will allow development of drugs based on LAs and possessing the beneficial effects noted, but lacking the "side effect" of Na channel blockade. Third, the non-Na channel effects of LAs in vivo should be defined in more detail, and their potential roles in modulating pathophysiologic roles should be investigated.
|
|
Optimized blood management for elective orthopaedic surgery |
General discussion and Summary The identification of different blood types by Landsteiner at the beginning of the 20th century was the first major improvement of the safety and efficacy of blood transfusion. Later on, developments in storage technology and screening for infectious agents made transfusion of blood even safer. This progress, however, has led to the paradoxical situation that, although transfusions are safer than ever, both doctors and patients are much more aware of its risks, and thus more reluctant to apply them. In terms of peri-operative patient management, one of the major challenges then, to the anaesthetist, has become to avoid blood transfusions. With this guiding principle in mind, this thesis set out to answer two questions: What is the deleterious effect of blood transfusions in elective orthopaedic surgery, and what can the anaesthetist do about it? Peri-operative blood loss is still a major problem in elective orthopaedic surgery. Homologous blood transfusion (HBT) is the standard approach to treat potentially detrimental decreases in haemoglobin (Hb) concentration. However, HBT is associated with various adverse events, including febrile reactions, transmission of infectious diseases, and an immunomodulatory effect, which is hypothesized to increase the frequency of postoperative infections. This issue is far from resolved, as observational cohort studies, randomised controlled studies and a recent meta-analysis on the subject produced conflicting results. Nonetheless, the clinical observation that patients who receive any HBT after major orthopaedic surgery do stay in the hospital significantly longer is undisputed. As postoperative infections are relatively rare (1-3%), and as the role of HBT herein is not yet established, other factors are likely to be responsible for the prolonged hospital stay. To address this issue we undertook a prospective observational study (Chapter 4). In 444 patients who underwent elective total hip arthroplasty, we studied (among other parameters) the frequency of HBT, wound disturbances, superficial and deep wound infections, and length of hospital admission. In this prospective observational study we found that HBT is associated with prolonged hospital admission. We found that this prolonged admission is not a straightforward consequence of an increased postoperative infection rate. However, HBT was the sole significant predictor of the development of wound healing disturbances, and together these two factors were the main predictors of prolongation of hospitalisation. No significant influence on wound disturbance and hospitalisation was found, either by univariate or multivariate analysis, of age, sex, length, weight, operation duration, blood loss or the use of Gentamicine cement. Although we did not study the mechanism of the wound disturbances, the more pragmatic conclusion we reach from our study is that prevention of HBT may be of relevance in limiting duration of admission after elective orthopaedic surgery. If this is indeed the case, measures to prevent perioperative blood loss, cell saving techniques and methods to enhance preoperative Hb (such as erythropoietin) might be attractive treatment options. One of the simplest and cheapest ways to optimise the use of blood transfusions would theoretically be the development of a protocol. Indeed, in the light of the risks described above and the fact that blood supply in The Netherlands is not infinite, a nationwide consensus on guidelines for packed red blood cells (PRBC) transfusions in the peri-operative period was reached in 1997. This nationwide consensus resulted in a transfusion protocol for our hospital in which the haemoglobin concentration was the sole indication for homologous blood transfusion (HBT). We investigated the HBT ratio before and after the introduction of the new protocol in our hospital (Chapter 5). Over a 33-month period 14587 patients were included in the study. We found a 50% decrease in PRBC transfusions after implementation of the guidelines. Of course is it not possible to maintain all confounding factors that can affect per-operative blood loss during a period of evaluation as long as our study. Several important factors changed in our hospital during the evaluation period. Changes occurred in the surgical staff, and the percentage of loco-regional anaesthesia increased. Also, during the evaluation period the introduction of more COX-2 selective NSAID’s took place. Undoubtedly, these factors were, to some extent, responsible for the decreased blood loss and thus for decreased peri-operative transfusion requirements. Still, our study clearly shows the benefits of employing strict peri-operative transfusion guidelines, as a 50% decrease in PRBC transfusions was seen during the observation period compared to the 25% decrease in blood loss during the same periods. Peri-operative transfusions are responsible for 60% of all transfusions in the Netherlands. According to a representative survey, the indications were poorly specified for 40% of these: blood loss, routine practice, cardiovascular changes, weakness or fatigue were cited. Current guidelines, which our study corroborates, indicate that in the peri-operative period a patient should only be transfused if the haemoglobin concentration is known.
Apart from adherence to a strict protocol, there are more specific methods with the potential to minimise the use of blood transfusions. Among these is the use of the new generation non-steroidal anti-inflammatory drugs (NSAID’s), the COX-2 antagonists with less influence on blood coagulation. Moreover, sophisticated technology now enables the anaesthetist to retrieve RBC’s peri- and post-operatively. Finally, the hormone erythropoietin may be used to boost the number and contents of the RBC’s pre-operatively. We studied all these potential methods sequentially. NSAID’s are used in the perioperative period for analgesia and reduction of edema in the surgical field. Beside these benefits there are unwanted side effects: reduction of renal blood flow, gastric complaints and increase of blood loss during surgery by influencing the coagulation cascade. Ibuprofen is a commonly used NSAID in the Netherlands, and we wished to assess its effects on perioperative blood loss in patients undergoing hip arthroplasty in a randomised double-blinded placebo controlled study in 50 patients. In this study preoperative (2 weeks)
pre-treatment with ibuprofen showed an increase in blood loss of 46% during and the first
24 hours after total hip replacement surgery. Confounding factors in our study were the
use of prophylaxis against tromboembolism by acenocoumarol, and the technique of measuring
blood loss. The study was also not geared to prove that an increase in blood loss resulted
in increased transfusion requirements or perioperative morbidity/ mortality. However,
although we thus did not prove that the increased blood loss was clinically relevant, the
fact remains that compared to a placebo, ibuprofen caused more blood loss
peri-operatively. The anti-inflammatory, analgesic and antipyretic action of NSAID’s are mediated through inhibition of prostaglandin synthesis by inhibiting cyclo-oxygenase (COX), which is now known to exists in at least two isoforms known as COX-1 and COX-2. COX-1 is important in ‘’housekeeping” functions at the gastro-intestinal mucosa, kidneys and vasculature. In contrast, the inducible isoform COX-2, mainly contributes to the pathophysiological process of inflammation. From this one could reason that selective COX-2 inhibiting NSAIDs have fewer side effects. We have taken this idea further and designed a prospective randomised study to test the hypothesis that use of more cyclooxygenase-2 (COX-2)-selective NSAIDs can reduce perioperative blood loss compared with non-selective NSAIDs. We studied 200 patients who underwent total hip replacement. Two NSAIDs were compared: conventional NSAID indomethacin (3 x 50 mg daily) and the COX-2 selective meloxicam (1x 15 mg daily). Total perioperative blood loss after meloxicam was 17% (p<0.05) less than that observed after indomethacin. However, whether this is of clinical significance and influences patient outcome remains to be determined. In addition, the potential beneficial effects of these compounds on blood loss should be weighed against potential detrimental effects (such as a potentially increased risk for cardiovascular events) before routine use can be recommended. One of the more elegant measures to avoid red blood cell transfusions in patients with a lowered haemoglobin level (10-13 g/dl) is erythropoietin (EPO). To study the effects of EPO on the number of peri-operative BTs we undertook a multi-centre randomised controlled trial (RCT). In this RCT the effects of preoperative administration of epoetin alpha (EPO) and best standard of care (BSC) in 695 orthopaedic surgery patients were compared in normal clinical routine in six countries. EPO-treated patients had higher Hb values from the day of surgery until endpoint and lower transfusion rates (12% vs. 46%). EPO treatment delivered no significant effect on postoperative recovery (time to ambulation, time to discharge, infection rate). Time to ambulation and time to discharge were, however, longer in transfused than in not-transfused patients. Side effects in both groups were comparable. Epoetin alpha increased perioperative Hb in mild-moderate anaemic patients and reduced transfusion requirements. Patients receiving blood transfusions required a longer hospitalisation than not-transfused patients.
Apart from the above EPO probably also plays an important role in post-operative rehabilitation. In many countries hip and knee surgery are performed in accelerated rehab programs, which limit hospitalisation times to less than a week. Thus postoperative higher haemoglobin levels might become more important as rehabilitation becomes more strenuous due to reduction in hospitalisation. Preoperative epoetin alpha administration will thus become more important, especially in the more compromised patients, as they suffer more from the changes in rehabilitation and lower accepted haemoglobin levels. This study leads to the conclusion that in routine daily setting of major orthopaedic surgery epoetin alpha treatment is an efficient method to decrease perioperative transfusion requirements and to increase perioperative haemoglobin levels. As hospitalisation time is severely reduced by new rehabilitation procedures, this will become even more important. The current treatment protocol of erythropoietin (EPO) consists of a fixed dose in all patients, according to Goldberg. Although easy to use, it has its limitations, since the inter-patient variability in response is considerable. In a prospective study we investigated whether pre-operative EPO dosing can be specifically tailored to the individual patient’s needs. A total of 334 patients scheduled for major orthopaedic surgery were treated for decreased serum haemoglobin levels. Erythropoietin and iron were used three weeks prior to surgery to increase the level of serum haemoglobin. Afterwards we calculated the erythropoiesis output in grams of haemoglobin after one injection of 40,000 IU of EPO (i.e. Factor EO-40,000). In our study factor EO-40,000 was 34 grams. Factor EO-40,000 can be used to predict the increase in the serum haemoglobin level in the individual patient, as the number of epoetin alpha injections (40,000 IU) needed to reach the desired serum haemoglobin level can now be calculated. With this knowledge, patients can be better prepared for surgery and homologous blood transfusions can be avoided as far as possible. As stated above another method to reduce peri-operative blood transfusions is the use of an autotransfusion system for shed blood. Cell saving techniques are an effective approach to prevent blood transfusion. Shed blood may be re-infused after filtration (unprocessed) or after treatment in a cell separator (processed). Transfusion of unprocessed shed blood is a relatively simple and inexpensive method to restore normovolaemia, in contrast to autotransfusion of processed blood, which requires an expensive cell separator and disposables. We performed a prospective observational quality assessment study of the BellovacŇ post-operative wound drainage and reinfusion system in 135 consecutive patients scheduled for elective total knee arthroplasty or total hip arthroplasty. The control group consisted of a historic group of 96 patients. Autotransfusion reduced the percentage of patients receiving HBT from 35% -22%). The reduction was more pronounced in the knee surgery patients (18% to 6%) compared with the hip surgery patients (47% to 34%). In the knee patients, transfused packed red cells per operation decreased from 0.45 HBT/operation to 0.11 HBT/operation, a reduction of 0.34 HBT (75%) in every total knee arthroplasty. This was also found in the hip surgery patients, although this did not reach statistical significance. This may be related to surgical technique: as in most hospitals, knee arthroplasty surgery was performed using a bloodless field, so that blood loss occurred only in the post-operative period, in contrast to the hip surgery where blood loss occurred during the whole procedure. Re-transfusion of shed blood is of course most effective if all lost blood can be collected, as is the case in knee surgery, but not in hip surgery, where significant blood loss occurs during the operation. Although we observed no complications during re-transfusion, our patient group may have been too small for a safety assessment. One of the most common side effects is a febrile reaction after autotransfusion of shed blood, which we did not see. Another concern would be the presence of methyl methacrylate monomer (MMM) in re-transfused blood. However, systemic blood showed no evidence of MMM after re-infusion of salvaged blood in cemented knee arthroplasty surgery. A cost-benefit analysis of the system is difficult to make, because of the complex costs involved in blood transfusion. At first glance autotransfusion is more expensive, but a future formal cost-benefit study should also take into account possible extended hospitalisation and immunomodulatory effects due to a homologous blood transfusion. We conclude that the BellovacŇ A.B.T. device reduced homologous blood transfusions in TKA. The use of the system is less complicated and less expensive than auto re-transfusion using a cell separator. This method of auto re-transfusion should therefore be considered in knee surgery. Having defined several methods to decrease peri-operative blood transfusion, we subsequently implemented these improvements in our daily clinical routine. The first step was the restriction that the Hb level should be assessed prior to red cell infusion to ensure compliance with pre-defined cut-off values. Subsequent measures included: confinement to COX-2-selective NSAID in the perioperative period; erythropoietin and iron therapy at the Hb level below 13 g/dL; consequent cell salvage during and after surgery; administration of aprotinine in cases with expected high blood loss. We then studied the effect of these alterations in our clinical routine. To this end we surveyed a relational database with data on 28.861 orthopaedic surgery patients in our clinic before and after implementation. The survey disclosed the following issues related to a high incidence of homologous red cell infusion: negligence of guidelines, the preoperative use of non-selective NSAID’s, low preoperative Hb level, non-retrieved blood loss, and high cut-off values for homologous red cell transfusion. The type of anaesthetic procedure was found to be not relevant for blood-sparing effect.
The steps mentioned above do not involve a medical novelty. Rather, we show that strict rules with the appropriate steps and in sequence resulted in an 80 % reduction of use of homologous red blood cells. Of note is that the incidence of deep wound infections decreased by 40 % over the same time, but whether this is related solely to our blood-saving measures remains to be seen Daily clinical practice in our hospital was not different from that in any general hospital in the Netherlands. Systematic presentations on “how we did it” resulted in the introduction of the measures in several of the hospitals in the region and a steady 5 – 7 % decline per year in the utilisation of the use of homologous red cell transfusions. One needs to keep in mind that each blood sparing measure on its own produces but a little decline. However the combination of all measures for the individual hospital results in a tremendous decline in the use of homologous red cells. Does this decline in HBT also result in a decrease in postoperative infections, as one would expect from the literature? Analysis of our database for these data showed a decline of the postoperative infections with 42 %, but whether this temporal relation is also a causal one is unsure. Restrictive guidelines for homologous red cell infusion Automated relational databases enabling feedback on clinical practice Preoperative assessment that involves the anaesthesiologists and allows for preoperative planning along a comprehensive algorithm (Chapter 11) |
|
Complex Regional Pain Syndrome Type I
|
|
Complex regional pain
syndrome (CRPS) Type I and II are neuropathic pain syndromes accompanied with sudomotor
and vasomotor disturbances. CRPS I, which corresponds to the common image of Reflex
Sympathetic Dystrophy (RSD) is defined as a painful, disabling syndrome1. The
Consensus Conference of the International Association for Study of Pain (IASP) defined
CRPS I as a post-traumatic syndrome that presents with spontaneous pain that is not
related to the territory of a single nerve and is disproportionate to the inciting event1,
2. The diagnostic criteria include: (a) pain, allodynia, or hyperalgesia; (b) evidence at
some time of edema, vasomotor and sudomotor change in the pain region; and (c) no other
conditions that would otherwise account for the degree of pain and dysfunction. CRPS is
differentiated from other neuropathic pain syndromes by the existence of edema, vasomotor
and sudomotor disturbances. Currently
practiced treatments of CRPS I include radical scavengers3, regional intravenous
sympathetic blocks4 and neuromodulation5. Kingery et al6 reviewed existing trials for CRPS
management in 1997 and demonstrated that there is limited support for the effectiveness of
topical DMSO (dimethylsulfoxyde), epidural clonidine, intravenous regional blocks and
intranasal calcitonine. Jadad et al4 showed that there is no evidence for the efficacy of
regional intravenous sympathetic blockade (RIS). Controversy exists about the
effectiveness of therapeutic interventions for the management of CRPS I. In order to
ascertain appropriate therapies we conducted a review of existing randomised controlled
trials of therapies for this disabling disease. We
identified 27 randomised trials, of which 18 were placebo controlled. The heterogeneity of
the studies and the small sample sizes precluded the drawing of firm conclusions about the
efficacy or effectiveness of any of the interventions studied on CRPS I patients. On basis
of our review we conclude that there is limited to no evidence for efficacy of sympathetic
blocks (Stellate ganglion block or RIS block), radical scavenging, prednisolone
administration, acupuncture and manual lymph drainage. Calcium regulating drugs and Qigong
exercises seem to be promising treatment modalities. However, further high quality
studies are required before the place of these treatments in pain therapy can be
established. The search for eligible trials about prevention resulted in 2 high quality
randomised placebo controlled studies in which vitamin C and intravenous guanethidine were
investigated on CRPS I patients7, 8. Vitamin C prevented CRPS I while guanethidine did not
prevent the development of CRPS I. Both studies had small sample sizes and no other
randomised placebo controlled studies were identified. Therefore there is limited evidence
whether any interventions can significantly prevent CRPS I. The
review showed not only limited evidence for the currently used treatment and prevention
modalities of CRPS I but it also demonstrated several main problems in CRPS I studies,
which will be discussed below. Clinically
relevant pain reduction. In the CRPS I literature most clinical studies use pain ratings as the primary outcome measure. Mostly, in these studies a significant pain reduction after treatment is defined as successful4, 9-12. Some authors define a pain reduction of 30 to 50% as successful13-15. However, a clinically relevant pain reduction in patients with CRPS I is still not defined. In our study we found that in both “successful” and “non successful” patients according the Global Perceived Effect, the pain was reduced significantly. However, the CRPS I patients defined a relative pain reduction of 58% or more on a VAS as “successful”. Furthermore, a relative pain reduction of 50 % and more as cut-off point proved to be the most accurate point with the best sensitivity and specificity. For absolute pain reduction, 3 cm on VAS could be defined as an accurate cut-off point Single
pain rating vs. multiple pain rating. In clinical studies with patients suffering from CRPS I, pain is usually assessed by a single pain rating. This rating is consequently used as the primary outcome measure, assuming that it is equivalent to multiple pain ratings. However to our knowledge this assumption had never been investigated in CRPS I patients. We
compared the validity of the single pain ratings in patients with CRPS I with multiple
pain rating test and demonstrated that a single pain rating (“recalled average”
pain) correlates well with multiple pain ratings (“actual average” pain) in a
group of patients suffering from CRPS I. There was a high degree of agreement between both
methods. Furthermore, both ratings measure significant pain reduction after treatment,
however “recalled average” pain reflects greater change in pain intensity. Bio-electrical
Impedance Assessment (MFBIA) as an objective measure. The diagnoses and evaluation of CRPS I is based solely on subjective clinical symptoms including burning pain, sensory abnormalities, diminished strength, hyperhidrosis, hypertrichosis, skin colouring changes and atrophy of the involved tissue (skin, muscle, bone) 16, 17. There is a clinical need for an objective measure of the severity of the patient’s condition. In order to develop an objective measure for CRPS I we performed a pilot study on the use of Bio-electrical Impedance Assessment in patients with CRPS I. MFBIA, a
non-invasive electrical assessment, is accepted for measuring body cell mass and has been
shown to be valid in measuring the intra and extra cellular fluid volume in each body
segment18-20. Our data
demonstrate that the impedance spectrum of the affected limbs in CRPS I patients is
altered compared to the contra lateral side and to limbs in a control group. The phase
angle, which is defined as the relation between the resistance (ability of an medium to
conduct an alternating electrical current) and the reactance (resistive effect due to
capacitance produced by tissue interface and cell membrane) was significantly less in the
affected limbs than in normal limbs and this could indicate significant change in
intra- and extra cellular fluid ratios. In addition, the MFBIA measured a significant
increase of the MFBIA frequency (?0) at which the reactance and the phase angle
are maximal in the affected limbs of CRPS I patients compared with the contralateral limb
and the control group and the resistance (R) was also increased in affected limbs. The
decrease of the phase angle was mainly caused by the increase in resistance (R), which is
defined as the ability of a medium to conduct electrical current and not due to changes in
the resistive effect produced by tissue interface and cell membrane (reactance). The
entities ?0 and phase angle of the MFBIA proved to be moderate sensitive and
specific for CRPS I in the lower limbs compared with either the left or the right limbs of
the control group. The use of
Stellate Ganglion Block(SGB) and Spinal Cord Stimulation(SCS) in CRPS I. Stellate Ganglion Block(SGB)21 and Spinal Cord Stimulation(SCS) 5, 22, 23 are currently used treatment modalities in the treatment of CRPS I. However, information on the effectiveness of SGB and the long term effect SCS, is scarce. We investigated retrospectively the use of SGB and prospectively the long term effect of SCS in patients suffering from CRPS I. The retrospective study of 226 chronic pain patients treated with RF lesions of the SG demonstrated that 40% of pre-selected patients reported a reduction of pain of more than 50% for an average duration of 52 weeks. Patients with a positive prognostic blockade were selected for RF lesions. The efficacy of the RF SG blockade in the total (unselected) population of 226 patients was 15,5 %. The clinical effectiveness of RF SG blockade may be improved at following two levels. Precise anatomic localisation of the SG may lead to increased clinical effect. This could be achieved by ultrasound24, magnetic resonance imaging25 and computed tomography26. The results show that the therapeutic effects of an RF SG block vary per diagnostic subgroup. When using a cut-off value of 50 % of patients reporting more than 50% pain relief per diagnostic subgroup, only CRPS2, ischemic pain, cervico-brachialgia and post-thoracotomy pain, appears to respond to an RF SG block, which is in line with work by Wilkinson et al, using radiofrequency to target upper thoracic sympathetic ganglia26-28. The prospective study on the effect of SCS demonstrated that all patients initially reported a significant reduction of pain intensity and a relative pain reduction of at least 50%, during all follow-up periods pain reduction declined in all patients. 42% of the patients reported at least “much improvement” over the two-year follow-up. Health status of the patients measured on the EQ-5D increased after receiving the SCS device. This increase was mainly caused by the decrease of pain and discomfort, and anxiety and depression. There were no differences between patients with a cervical or lumbar SCS. CONCLUSION There is limited evidence to support the effectiveness of
commonly used interventions for treating or preventing RSD or CRPS I. More prospective
controlled trials are needed in this field. In CRPS I patients a significant pain reduction after treatment using a VAS does not necessarily imply that the patient defines the treatment as successful. A relative pain reduction of 50% or more and an absolute pain reduction of at least 3 cm on VAS are accurate in predicting a successful pain reduction after a given treatment. Furthermore as in back pain patients, a single rating of pain “on average” (“recalled average” pain) is an accurate predictor of the actual “average pain” in patients suffering of CRPS I. Both pain ratings proved also to be accurate enough to determine reliable changes in pain over time. MFBIA measurement is a promising assessment for measuring
tissue properties in effected limbs of CRPS I patients when compared with the
contra-lateral side and compared to limbs in a control group. The relation of the
impedance to these properties is indirect and has not been completely defined. The MFBIA
measures the tissue resistivity, which is affected by relative volumes and electrolytic
balances between the intra- and extra cellular fluids and these factors vary among
individuals. Although
the efficacy of an RF SG blockade appears to be in line with other SG blockade procedures
reported in the literature, its clinical effects remain to be proven in a randomised
controlled trial. Given the potential of harmful complications, it would even be
questionable if it is ethical to continue the routine clinical use of RF SG blocks without
supportive data from such a trial. SCS seems to be effective in treating chronic therapeutic resistant CRPS I. The presented results demonstrate that SCS reduces the pain intensity in the majority of patients suffering of CRPS I. Despite the substantial complications and adverse effects, the heath state increased. This was mainly due to decrease in the level of pain, discomfort, anxiety and depression. There were no significant differences between cervical and lumbar SCS. References 1.
Merskey KR, Bogduck N. Classification of Chronic Pain: Descriptions of Chronic Pain
Syndromes and Difinitions of Pain Term. Seattle, WA: IASP Press, 1994. 2.
Schurmann M, Gradl G, Andress HJ, Furst H, Schildberg FW. Assessment of peripheral
sympathetic nervous function for diagnosing early post-traumatic complex regional pain
syndrome type I. Pain 1999; 80:149-59. 3.
Zuurmond WW, Langendijk PN, Bezemer PD, Brink HE, de Lange JJ, van loenen AC. Treatment of
acute reflex sympathetic dystrophy with DMSO 50% in a fatty cream. Acta Anaesthesiol Scand 1996; 40:364-7. 4.
Jadad AR, Carroll D, Glynn CJ, McQuay HJ. Intravenous regional sympathetic blockade for
pain relief in reflex sympathetic dystrophy: a systematic review and a randomized,
double-blind crossover study. J Pain Symptom Manage
1995; 10:13-20. 5.
Kemler MA, Barendse GA, van Kleef M, et al. Spinal cord stimulation in patients with
chronic reflex sympathetic dystrophy. N Engl J Med
2000; 343:618-24. 6.
Kingery WS. A critical review of controlled clinical trials for peripheral neuropathic
pain and complex regional pain syndromes [see comments]. Pain 1997; 73:123-39. 7.
Gschwind C, Fricker R, Lacher G, Jung M. Does peri-operative guanethidine prevent reflex
sympathetic dystrophy? J Hand Surg [Br] 1995;
20:773-5. 8.
Zollinger PE, Tuinebreijer WE, Kreis RW, Breederveld RS. Effect of vitamin C on frequency
of reflex sympathetic dystrophy in wrist fractures: a randomised trial. Lancet 1999; 354:2025-8. 9.
Hanna MH, Peat SJ. Ketanserin in reflex sympathetic dystrophy. A double-blind placebo
controlled cross-over trial. Pain 1989;
38:145-50. 10. Kho HK. The
impect of acupuncture on pain in patients with reflex sympathetic dystrophy. Pain Clin 1995; 8:59-61. 11. Kettler RE,
Abram SE. Intravenous regional droperidol in the management of reflex sympathetic
dystrophy: a double-blind, placebo-controlled, crossover study. Anesthesiology 1988; 69:933-6. 12. Korpan MI,
Dezu Y, Schneider B, Leitha T, Fialka-Moser V. Acupuncture in the treatment of
posttraumatic pain syndrome. Acta Orthop Belg
1999; 65:197-201. 13. Blanchard J,
Ramamurthy S, Walsh N, Hoffman J, Schoenfeld L. Intravenous regional sympatholysis: a
double-blind comparison of guanethidine, reserpine, and normal saline. J Pain Symptom Manage 1990; 5:357-61. 14. Price DD,
Long S, Wilsey B, Rafii A. Analysis of peak magnitude and duration of analgesia produced
by local anesthetics injected into sympathetic ganglia of complex regional pain syndrome
patients. Clin J Pain 1998; 14:216-226. 15. Verdugo RJ,
Ochoa JL. 'Sympathetically maintained pain.' I. Phentolamine block questions the concept
[see comments]. Neurology 1994; 44:1003-10. 16. Bruehl S,
Harden RN, Galer BS, et al. 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;
81:147-54. 17. Harden RN,
Bruehl S, Galer BS, et al. Complex regional pain syndrome: are the IASP diagnostic
criteria valid and sufficiently comprehensive? Pain
1999; 83:211-9. 18. Van Kreel BK,
Cox-Reyven N, Soeters P. Determination of total body water by multifrequency bio-electric
impedance: development of several models. Med. Biol.
Eng. Comput. 1998; 36:337-345. 19. Sasser DC,
Gerth WA, Wu YC. Monitoring of segmental intra- and extracellular volume changes using
electrical impedance spectroscopy. J Appl Physiol
1993; 74:2180-7. 20.
Baumgartner RN, Chumlea WC, Roche AF. Estimation of body composition from bioelectric
impedance of body segments. Am J Clin Nutr 1989;
50:221-6. 21. Bonelli S,
Conoscente F, Movilia PG, Restelli L, Francucci B, Grossi E. Regional intravenous
guanethidine vs. stellate ganglion block in reflex sympathetic dystrophies: a randomized
trial. Pain 1983; 16:297-307. 22. Kemler MA,
Barendse GA, Van Kleef M, Van Den Wildenberg FA, Weber WE. Electrical spinal cord
stimulation in reflex sympathetic dystrophy: retrospective analysis of 23 patients. J Neurosurg 1999; 90:79-83. 23. Stanton-Hicks
M, Salamon J. Stimulation of the central and peripheral nervous system for the control of
pain. J Clin Neurophysiol 1997; 14:46-62. 24. Kapral S,
Krafft P, Gosch M, Fleischmann D, Weinstabl C. Ultrasound imaging for Stellate Ganglion
block: direct visualization of puncture site and local anesthetic spread. Reg Anesth 1995; 20(4):323-328. 25.
Slappendel R, Thijssen HOM, Crul BJP, Merx JL. The Stellate Ganglion in Magnetic Resonance
Imaging: A Quantification of the Anathomic Variability. anesthesiology 1995; 83:424-426. 26. Erickson SJ,
Hogan QH. CT- guided Injection of the Stellate Ganglion: Description of Technique and
Efficacy of Sympathetic Blockade. Radiology
1993; 188:707-709. 27. Wilkinson HA.
Radiofrequency percutaneous upper-thoracic sympathectomy. Technique and review of
indications. N Engl J Med 1984; 311:34-6. 28.
Wilkinson HA. Percutaneous
radiofrequency upper thoracic sympathectomy. Neurosurgery
1996; 38:715-25. SUMMARY In chapter 1 we conducted a review of existing randomised controlled trials of therapies forCRPS I. Eligible trials were identified from the Cochrane, Pubmed, Embase and MEDLINE databases from 1966 through June 2000, from references in retrieved reports and from references in review articles. Twenty-six studies concerning treatment modalities were identified. Eighteen studies were randomised placebo-controlled trials and eight studies were randomised active- controlled trials. Three independent investigators reviewed articles for inclusion criteria using a 15-item check list. Seventeen of the trials were of high quality according to the 15-item criteria. There was limited evidence for the effectiveness of these interventions because of the heterogeneity of treatment modalities. The search for trials concerning prevention of RSD/CRPS I resulted in 2 eligible studies. Both were of high quality and dealt with two different interventions. There is limited evidence for their preventive effect. The aim of the study presented in chapter 2 was to investigate the degree of painreduction in patients with CRPS I that can be defined as lssuccessfull.. Sixty-one CRPS I patients rated their pain on a Visual Analogue Scale (VAS, 0-10) before treatment and on three occasions after treatment at 6 months, 1 year and 2 years. Patients also rated a Global Perceived Effect (GPE) for their pain relief at the same time periods. The GPE items were classified as issuccessfull. or irunsuccessfulll. The mean absolute and relative pain reduction (VAS) was calculated for both iosuccessfulll and inunsuccessfulll GPE classifications for each time period. Sensitivity and specificity analyses were performed. The patients defined a relative pain reduction of 58% (s.d.: 23.4) or more as issuccessfulln, whereas in both iosuccessfulln and inunsuccessfulll patients the pain was reduced significantly on the VAS. Furthermore, sensitivity and specificity analyses demonstrated that a cut-off point of 50% relative pain reduction and a 3 cm absolute pain reduction on VAS have the highest likelihood that patients report iesuccessfulll on the GPE. We conclude that a relative pain reduction of 50% or more and an absolute pain reduction of at least 3 cm on VAS are accurate in predicting a successful pain reduction after a given treatment. Chapter 3 describes comparison of multiple and single pain ratings in patients withCRPS I. Correlation, agreement and reliability analyses were performed between the average pain intensity measured 3 times a day over a course of 4 days and one single pain rating (designated as isrecalled averagele pain by the patient) before treatment and 1-, 3- and 6 months after treatment. The patient population consisted of 54 patients suffering from CRPS I in a randomized trial. The results show that both measurements correlate and agree excellently . Furthermore, both ratings measure significant pain reduction after treatment. However iorecalled averagele pain reflects greater change in pain intensity. In patients with CRPS I a single pain rating is an accurate predictor of the average pain measured by a multiple pain rating test. Moreover, both assessments are accurate enough to determine changes in pain over time with an effective treatment. In chapter 4 we evaluated the value of Multi-Frequency Bioelectrical ImpedanceAnalysis (MFBIA) in patients suffering from Complex Regional Pain Syndrome Type I (CRPS1). We hypothesised that, patients with CRPS I have altered tissue properties in the affected limbs compared with the non-affected limbs and with limbs from unaffected individuals (control limbs). 28 patients suffering from CRPS I and in 18 healthy right-handed participated in the study. Receiver operating characteristic (ROC) analyses and areas under the curve (AUC) were performed to analyse the sensitivity and specificity of the MFBIA to detect CRPS I. Our data showed the phase angle ( ), which is defined as the relation betweenthe resistance (ability of an medium to conduct an alternating electrical current) and the reactance (resistive effect due to capacitance produced by tissue interface and cell membrane) was decreased in the CRPS I limbs compared with both the contralateral side (p = 0.004) and limbs (p = 0.02) from unaffected individuals. The MFBIA frequency ( 0) at which thereactance and the phase angle are maximal was increased in CRPS I limbs compared with both the contralateral side (p = 0.02) and the control limbs (p = 0.01). The sensitivity and specificity of ů 0 for detecting CRPS I in the lower limb were 0.91 and 0.85 compared withlower limbs in unaffected individuals. In the upper limb, however, sensitivity and specificity were best when the right hand in unaffected individuals was used as the control. The sensitivity and specificity of ů 0 0.75 and 0.82 compared with right arm controls. The Ô valueswere 0.82 and 0.70 compared with the right arm controls. In conclusion the tissue properties are changed in the affected limb according the MFBIA. In chapter 5 we evaluated the outcome of RF-SG as a therapy for CRPS I and otherchronic pain syndromes. We reviewed 86 RF stellate ganglion procedures. Medical records containing treatment information were reviewed systematically. In our clinic, 39,5% of 221 patients who received a prognostic SG block subsequently underwent RF-SG. Of these patients 40,7% noted more then 50% reduction of pain, 54,7% had no effect on pain and 4,7% showed worsening of pain. Mean follow-up was 52 weeks. A systematic literature review search in MEDLINE on SG blockades was also performed. The computer assisted literature search resulted in thirty-one studies, twelve about complications and nineteen about the efficacy of stellate ganglia block. The analysis of these studies showed a partial pain relief in 41,3 % of patients, complete pain relief in 37,8% and no pain relief in 20,9%. The efficacy of the RF-SG blockade appears to be in line with other SG block procedures reported in the literature. Our retrospective study shows that a RF-SG block is most likely to be of benefit for patients suffering from CRPS2, ischemic pain, cervico-brachialgia, or post-thoracotomy pain. However, clinical efficacy remains to be proven in a randomised controlled trial. The long term effect of cervical and lumbar Spinal Cord Stimulation (SCS) in patients with CRPS I is presented in chapter 6. Thirty-six patients with a definitive implant wereincluded in this study. A pain diary was obtained in all patients prior to treatment, 6 month, 1- and 2 year post treatment. Further, all patients were asked to complete a 7-point Global Perceived Effect (GPE) scale and the Euroqol- 5D (EQ-5D) at each post - treatment assessment period. The pain intensity was reduced 6 month, 1- and 2 year post treatment (p < 0.05). However, the repeated measures ANOVA showed a statistically significant, linear increase of the VAS (p=0.03) over this period. According to the GPE at least 42% of the cervical SCS patients and 47% of the lumbar SCS patients reported at least iemuch improvemently during measurements. Furthermore, the health state of the patients measured on the EQ-5D was increased after treatment (P < 0.05). This increase was noted both from a social as from a patient™s perspective. Complications and adverse effects occurred in 64% of the patients and consisted mainly of technical defects. There were no differences between cervical and lumbar groups with regard to outcome measures. In conclusion SCS reduces the pain intensity in the majority of the CRPS I patients and increases the health state but produces a high incidence of substantial complications and adverse effects.
|
| Jan van
Zundert
The use of pulsed radiofrequency in the treatment of chronic pain
|
| Even in
the 21st century the management of chronic pain remains a difficult chal-
lenge, because of the variety in factors interfering with its clinical presentation, cop- ing and consequently the pain behavior. The evolution in the understanding and appreciation of this complexity led to the integrated multidisciplinary approach, where cognitive behavioral treatment and physical rehabilitation programs are im- plemented alongside the algorithm for pharmacological and interventional pain management as proposed in chapter II. This is well described in the literature for low back pain1 but it is less well documented for pain originating from the cervical spine. Although many treatment modalities for cervical radicular pain are described in the literature, the available evidence for efficacy does not allow to formulate definitive conclusions on the optimal therapy. The aim of this thesis was to evaluate the role of pulsed radiofrequency treatment for chronic pain management, with special attention to PRF application at the cervi- cal dorsal root ganglion in chronic cervical radicular pain. In chapter I an overview of the evolution of radiofrequency (RF) techniques in the management of chronic pain is given. The first use of RF for spinal pain was re- ported by Shealy 2 for the treatment of lumbar zygapophyseal joint pain. Although Uematsu 3 first described RF lesions of the dorsal root ganglion (DRG), the tech- nique only gained in popularity in some countries after the introduction of small di- ameter electrodes (22 G disposable cannula with a thermocouple inside) known as Sluijter Mehta Kit (SMK) systems 4. However, the RF lesioning of the DRG remains controversial due to the potential risk for deafferentation syndromes. For this reason a modification of the technique resulting in less damage would be more attractive for chronic radicular pain syndromes. In 1998 Sluijter, Cosman, Rittman, and van Kleef introduced a modified RF technique known as pulsed radiofrequency (PRF) 5 In order to evaluate the role of PRF it is imperative to make a review on the current status of continuous RF in pain management, which is presented in chapter II. Two systematic reviews on RF treatment of spinal pain issued similar conclusions: mod- erate evidence that radiofrequency lumbar facet denervation is more effective for chronic low back pain than placebo; limited evidence for efficacy of radiofrequency neurotomy in chronic cervical zygapophyseal joint pain after whiplash; limited evi- dence that radiofrequency heating of the dorsal root ganglion is more effective than placebo in chronic cervicobrachialgia 6-8. It is concluded from this literature review that all efforts should be made to narrow the gap between the Evidence Based Medicine (EBM) data and clinical practice using standardized protocols for patient selection and treatment. In line with the experience of conducting clinical trials with pharmacotherapy a multi-center approach using international accepted protocols should be considered. In this chapter we propose an algorithm for the management of chronic pain, whereby pharmacological and interventional pain treatments are presented as being complementary within a multidisciplinary approach. Accurate use of RF treatment as part of a multimodal and multidisciplinary approach may avoid the use of more invasive and often more expensive treatment options. The current evolution of RF from the initial neuro-ablative thermocoagulation towards the less or minimally neurodestructive PRF may have additional advantages in the management of chronic pain, if its value can be demonstrated.
In particular, PRF adjacent to the (cervical) DRG may be an appealing application as compared to RF, basically due to the hypothesis that it is less neurodestructive. In order to understand the relation between DRG and cervical radicular pain, the first research question should be addressed. In chapter III we tried to clarify the role of the DRG in cervical radicular pain. The results of an extensive literature review are reported. The exact pathophysiological mechanisms underlying cervical radicu- lar pain in humans are not yet fully understood. In line with the available literature on the pathophysiology of lumbar radicular pain, two major mechanisms are thought to induce cervical radicular pain: (1) nucleus pulposus material leaking onto the nerve root, and or (2) compression of the nerve root by anatomic abnormalities. Either of these pathogenic mechanisms will induce two processes in the nerve: (1) an in- flammatory reaction and – related to this – (2) changes in ion channel functioning. Eventually, these effects cause a pattern of hyperexcitability and spontaneous ec- topic activity in the DRG, which is interpreted as pain. These discharges enter the spinal cord and induce central sensitization at the synapses located in the dorsal horn9. Many available treatment modalities for cervical radicular pain may interfere with the pathophysiological processes at the DRG. The pharmacological treatment modali- ties can be subdivided into those targeting the inflammatory changes10 and those targeting the ion channel modulation11-14, but their value has not been investigated in the management of cervical radicular pain. Several new molecules may result in alternative treatment options in the future15-21. The interventional pain management modalities targeting the cervical DRG and/or nerve root consist of epidural steroid administration22,23, (P)RF6,7,24, and neck surgery25,26. The available evidence for effi- cacy does not allow us to formulate definitive conclusions on the optimal therapy. However, based on the risk/benefit ratio there is weak recommendation against the use of neck surgery for chronic cervical radicular pain. For this reason less invasive percutaneous interventional pain management techniques have gained interest dur- ing the last decade, and amongst them the RF techniques are well described. The mode of action of RF was, for a long time, attributed to the effects of heat, but since the introduction of the modified PRF it was assumed that different mecha- nisms other than thermocoagulation might also be involved. Therefore it is neces- sary to discuss some electrophysiological principles underlying PRF, which are pre- sented in chapter IV. During the administration of high frequency current adjacent to the neuronal tissue an electric field is generated that accounts for most of the effects, such as forces on and movement of ions in electrolytes, current densities and stresses on cellular substructures and membranes, which in turn produce ionic friction and heat. When delivering the current in a continuous way the temperature increases in the tissue surrounding the electrode tip and decreases gradually away from the electrode. In PRF during the "active" burst, higher voltages of RF current are delivered than with continuous RF. This results in a higher density electric field around the electrode for PRF, which is probably capable of altering the cell mem- brane27. Exposure of cultured cells to an electric field is associated with up- regulation of nonspecific intermediate-early gene marker, c-Fos, and also transcrip- tion of other, as yet unidentified, genes28. C-Fos based functional anatomic mapping has been validated as a powerful technique to detect activated neurons29. Evidence is accumulating to suggest that changes in early gene expression within the nervous system signal long-term adaptation within particular neural pathways30. This issue was further addressed in the second research question. "Does pulsed radiofre- quency of the cervical dorsal root ganglion result in changes at the dorsal horn of the spinal cord in an experimental setting?"
An early (3h) effect of pulsed radiofrequency as measured by an increase of c-Fos in the pain processing neurons of the dorsal horn of rats has been reported in the literature31. This effect was not mediated by tissue heating. In chapter V, we pub- lished a late effect of three different radiofrequency modalities; continuous RF, PRF during 120 s and PRF during 8 min compared to sham, using the same experimen- tal rat model as described by Higuchi et. al.31. The number of c-Fos immunoreactive cells, as a marker for neuronal activity, in the dorsal horn was significantly increased in the three different radiofrequency groups. No significant difference was demon- strated between the three active intervention groups. The observation that c-Fos is still present 7 days after the intervention suggest sustained activity, because the duration of its activity exceeded the expected length of time for c-Fos expression caused by the effects of surgery and electrical stimulation. These findings may be physiologically relevant for transmission and integration of sensory information, in line with the reported long-term depression of synaptic transmission in substantia gelatinosa neurons that can be induced by low-frequency stimulation of primary af- ferent A- fibers32. The demonstrated late and temperature independent cellular activity in the rat spinal cord after application of different RF modalities is suggested to be part of the underlying mechanism. As an extension to the c-Fos study we used Bromodeoxyuridine (BrdU) labeling to investigate if RF and PRF treatment adjacent to the cervical dorsal root ganglion would lead to an increased proliferation of cells in the spinal cord. The BrdU labeled cells were recognizable due to the intense staining of the nucleus. Immunopositive cells were randomly distributed throughout the spinal cord section. Our protocol did not result in any differences between the 4 treatment groups. Further experiments, using alternative BrdU labeling protocols are needed33. In an accompanying editorial related to the c-Fos article, the authors find the con- cept that pulsed radiofrequency may produce inhibition of excitatory C-fibers using a phenomenon such as long-term depression in the spinal cord is an attractive hy- pothesis. However this hypothesis needs to be confirmed in clinical research and randomized controlled trials to evaluate the clinical value of PRF34. After the initial publications on the clinical use of PRF from Sluijter5 and Munglani35 we reported in chapter VI the first 5 high-risk patients with idiopathic trigeminal neu- ralgia36. They were treated with PRF after multidisciplinary assessment, with a mean follow-up of 19.2 months (range 10 –26). These patients were at high risk due to age, co-morbidities or previous interventional and/or surgical treatments. An excel- lent long-term effect was achieved in 3 of the 5 patients, a partial effect in 1 patient and a short-term effect in another patient. No neurological complications were re- ported. To come up with final conclusions regarding the efficacy of PRF of the Gas- serian ganglion based on case series is impossible; there is need for well-controlled trials relying on standardized diagnostic criteria37. A clinical note is, however a good format to communicate on first findings with a new technique. To build up the evi- dence according to the rules of EBM, clinical research should be continued with re- trospective analysis on larger numbers of patients, prospective trials and finally ran- domized clinical trials and systematic reviews38. We all know it takes many years before all this work is done, especially taking into consideration a 3-year follow-up as recommended by Zakrzewska and Lopez39. As with other new treatments it takes generally 10 years from accumulation of clinical data to final confirmation or rejec- tion of an intervention40. The possibility of repeating the procedure during the same session, was questioned by some colleagues41,42. In fact, with continuous RF, ther- mocoagulation of the Gasserian ganglion is performed when the needle position is considered to be optimal, based on the results of the stimulation. The needle cannot be repositioned in order to limit the extend of the lesion. Needle repositioning after a first treatment with PRF may result in a lower sensory stimulation threshold and in this case a second treatment should be performed in order to optimize the results43. Also the use of the terminology to describe side effects and complications after the continuous RF trigeminal rhizotomy was subject to some debate. Sensory loss was stated not to be a side effect but an expected and desired outcome and the conditio sine qua non to obtain the wanted long-lasting pain relief42. It can indeed be confus- ing, because, the goal of the intervention is to provide a long-lasting pain relief with- out disturbing neurological function. In clinical practice the border between the de- sired sensory loss and the undesired dysesthesia is narrow and for the long term difficult to predict44
Because our research focused on cervical radicular pain, we postulate the third re- search question: "What are the therapeutic effects and side effects of pulsed radiof- requency adjacent to the cervical dorsal root ganglion in cervical radicular pain?". We studied prospectively the effect of PRF adjacent to the cervical DRG and re- ported the results in Chapter VII. This extended pilot study was performed to evalu- ate the outcome of PRF for short- and long-term relief of chronic cervical pain such as cervicogenic headache and cervicobrachialgia/cervical radicular pain. An inde- pendent evaluator reviewed the medical records of the first 18 patients treated with PRF at the cervical DRG. Patients with good clinical results at 8 weeks were evalu- ated for long-term effect (> 6 months), based on a 7-point Likert scale. Thirteen pa- tients (72 %) showed short-term clinical success (> 50 % pain relief). More than 1 year after treatment, 6 patients (33 %) rated treatment outcome as good or very good. Kaplan-Meier analysis illustrated that 50 % of patients experienced success 3 months after treatment. We could not identify predictive variables for clinical out- come. None of the patients reported post-treatment neuritis or complications. Ac- cording to the EBM guidelines, the outcome of this clinical audit supported the set- up of a randomized controlled trial. The results of a double blind sham controlled RCT on the efficacy and side effects of PRF adjacent to the cervical DRG in patients with chronic cervical radicular pain are presented in chapter VIII. In order to have a homogeneous patient population, the patient selection was based on an extensive protocol of physical examination including a positive Spurling test to judge the radicular involvement. Twenty-three patients were randomly allocated to PRF treatment of the cervical DRG (n=11) or sham intervention (n= 12). Assessment was done by an independent evaluator be- fore treatment and at 4 weeks and 3 months. Patients who had a favorable outcome at 3 months were again evaluated 6 months after the intervention. In the primary outcome success was defined as minimal 50 % improvement on the global per- ceived effect and 20 point reduction of the VAS pain score. Our results indicate that PRF treatment of the cervical DRG for chronic cervical radicular pain provides better pain relief than sham intervention 3 and 6 months after the procedure, but signifi- cance was only reached after 3 months. The need for pain medication, which was defined as rescue treatment in the study protocol, was reduced in the PRF group and increased in the sham group. Statistical significance was reached after 6 months. There is a tendency towards better outcome for sham after 4 weeks of quality of life measured by SF-36 and Euroqol. But, after 3 and 6 months, there was a trend towards a better improvement of quality of life in the PRF group, though sta- tistical significance was only reached for the domain vitality. No side effects or neu- rological complications were noted in both groups during the study period. The on- going slow inclusion rate forced us to stop the study before the calculated sample size of 42 patients was reached. This problem is frequently reported in sham con- trolled randomized clinical trials on interventional pain management tech- niques6,7,45,46. The results of an extensive literature search supplemented with the control of the reference lists of relevant papers and the abstracts presented during international congresses are reported in chapter IX. We could detect 58 reports on the clinical use of pulsed radiofrequency in different applications: 32 full publications and 26 abstracts. Because this is a new technique, a substantial part of these results are reported in the abstract books of international scientific congresses. Their number is increasing yearly and they are progressively published in peer reviewed indexed journals. We also retrieved six basic science reports, 5 full publications and one ab- stract. Although the mechanism of action has not been completely elucidated, labo- ratory reports suggest a genuine neurobiological phenomenon altering the pain sig- naling, which some have described as neuromodulatory47. No complications related to the PRF technique were reported to date. If indeed the latter is the case, the use of PRF in neuropathic and mixed pain states may be of interest. Further research in the clinical and biological effects is justified.
References 1. Carragee EJ. Clinical practice. Persistent low back pain. N Engl J Med 2005; 352: 1891-1898 2. Shealy CN. Percutaneous radiofrequency denervation of spinal facets. J Neurosurg 1975; 43: 448-451 3. Uematsu S, Udvarhelyi GB, Benson DW, Siebens AA. Percutaneous radiofrequency rhizotomy. Surg Neurol 1974; 2: 319-325 4. Sluijter M, Mehta M. Treatment of chronic back and neck pain by percutaneous thermal le- sions. Persistent pain, modern methods of treatment, Edited by Lipton S. London, Academic Press,1981, pp 141-179, 5. Sluijter ME, Cosman ER, Rittman IIWB, van Kleef M. The effects of pulsed radiofrequency field applied to the dorsal root ganglion - a preliminary report. The Pain Clinic 1998; 11: 109-117 6. Geurts J, van Wijk RM, Stolker R, Groen GJ. Efficacy of radiofrequency procedures for the treatment of spinal pain: a systematic review of randomized clinical trials. Reg Anesth Pain Med 2001; 26: 394-400 7. Niemisto L, Kalso E, Malmivaara A, Seitsalo S, Hurri H. Radiofrequency denervation for neck and back pain. A systematic review of randomized controlled trials. Cochrane Database Syst Rev 2003; CD004058 8. Niemisto L, Kalso E, Malmivaara A, Seitsalo S, Hurri H. Radiofrequency denervation for neck and back pain: a systematic review within the framework of the Cochrane collaboration back review group. Spine 2003; 28: 1877-1888 9. Guo W, Zou S, Guan Y, Ikeda T, Tal M, Dubner R, Ren K. Tyrosine phosphorylation of the NR2B subunit of the NMDA receptor in the spinal cord during the development and mainte- nance of inflammatory hyperalgesia. J Neurosci 2002; 22: 6208-6217 10. van Tulder MW, Koes BW, Assendelft WJ, Bouter LM, Maljers LD, Driessen AP. [Chronic low back pain: exercise therapy, multidisciplinary programs, NSAID's, back schools and behavioral therapy effective; traction not effective; results of systematic reviews]. Ned Tijdschr Geneeskd 2000; 144: 1489-1494 11. McQuay H, Carroll D, Jadad AR, Wiffen P, Moore A. Anticonvulsant drugs for management of pain: a systematic review. BMJ 1995; 311: 1047-1052 12. Rowbotham M, Harden N, Stacey B, Bernstein P, Magnus-Miller L. Gabapentin for the treat- ment of postherpetic neuralgia: a randomized controlled trial. JAMA 1998; 280: 1837-1842 13. Backonja MM. Anticonvulsants (antineuropathics) for neuropathic pain syndromes. Clin J Pain 2000; 16: S67-72 14. Sabatowski R, Galvez R, Cherry DA, Jacquot F, Vincent E, Maisonobe P, Versavel M. Pre- gabalin reduces pain and improves sleep and mood disturbances in patients with post-herpetic neuralgia: results of a randomised, placebo-controlled clinical trial. Pain 2004; 109: 26-35 15. Saklatvala J. The p38 MAP kinase pathway as a therapeutic target in inflammatory disease. Curr Opin Pharmacol 2004; 4: 372-377 16. Lazzeri M, Spinelli M, Beneforti P, Malaguti S, Giardiello G, Turini D. Intravesical infusion of resiniferatoxin by a temporary in situ drug delivery system to treat interstitial cystitis: a pilot study. Eur Urol 2004; 45: 98-102 17. Mitchell VA, White DM, Cousins MJ. The long-term effect of epidural administration of butam- ben suspension on nerve injury-induced allodynia in rats. Anesth Analg 1999; 89: 989-994 18. Gerner P, Nakamura T, Quan CF, Anthony DC, Wang GK. Spinal tonicaine: potency and dif- ferential blockade of sensory and motor functions. Anesthesiology 2000; 92: 1350-1360 19. Carter GT, Sullivan MD. Antidepressants in pain management. Curr Opin Investig Drugs 2002; 3: 454-458 20. Feighner JP. Mechanism of action of antidepressant medications. J Clin Psychiatry 1999; 60 Suppl 4: 4-11; discussion 12-13 21. Strumper D, Durieux ME. Antidepressants as long-acting local anesthetics. Reg Anesth Pain Med 2004; 29: 277-285 22. Peloso P, Gross A, Haines T, Trinh K, Goldsmith CH, Aker P. Medicinal and injection therapies for mechanical neck disorders. Cochrane Database Syst Rev 2005; CD000319 23. Stav A, Ovadia L, Sternberg A, Kaadan M, Weksler N. Cervical epidural steroid injection for cervicobrachialgia. Acta Anaesthesiol Scand 1993; 37: 562-566 24. Van Zundert J, Lamé IE, de Louw A, Jansen J, Kessels F, Patijn J, van Kleef M. Percutaneous Pulsed Radiofrequency Treatment of the Cervical Dorsal Root Ganglion in the Treatment of Chronic Cervical Pain Syndromes: A Clinical Audit. Neuromodulation 2003; 6: 6-14 25. Fouyas IP, Statham PF, Sandercock PA, Lynch C. Surgery for cervical radiculomyelopathy. Cochrane Database Syst Rev 2001; CD001466 26. Persson LC, Carlsson CA, Carlsson JY. Long-lasting cervical radicular pain managed with surgery, physiotherapy, or a cervical collar. A prospective, randomized study. Spine 1997; 22: 751-758 27. Cosman ER, Cosman ER. Electric and thermal field effects in tissue around radiofrequency electrodes. Submitted 28. Archer S, Li TT, Evans AT, Britland ST, Morgan H. Cell reactions to dielectrophoretic manipula- tion. Biochem Biophys Res Comm 1999; 257: 687-698 29. Kovacs KJ. c-Fos as a transcription factor: a stressful (re)view from a functional map. Neuro- chem Int 1998; 33: 287-297 30. Munglani R, Hunt SP. Proto-oncogenes: basic concepts and stimulation induced changes in the spinal cord. Prog Brain Res 1995; 104: 283-298 31. Higuchi Y, Nashold BS, Sluijter ME, Cosman E, Pearlstein RD. Exposure of the dorsal root ganglion in rats to pulsed radiofrequency currents activates dorsal horn lamina I and II neurons. Neurosurgery 2002; 50: 850-856 32. Sandkühler J, Chen JG, Cheng G, Randic M. Low-Frequency Stimulation of Afferent A -fibersinduces Long-Term Depression at Primary afferent Synapses with Substantia gelatinosa neu- rons in the rat. J Neurosci 1997; 17: 6483-6491 33. Van Zundert J, Joosten EA, Dederen PJ, Honig W, Veening JG, van Kleef M. Does (Pulsed) Radiofrequency treatment adjacent to the cervical dorsal root ganglion of the rat affect the number of newly formed cells in the spinal cord? Reg Anesth Pain Med 2005; Supplement: In press 34. Richebe P, Rathmell JP, Brennan TJ. Immediate early genes after pulsed radiofrequency treatment: neurobiology in need of clinical trials. Anesthesiology 2005; 102: 1-3 35. Munglani R. The longer term effect of pulsed radiofrequency for neuropathic pain. Pain 1999; 80: 437-439 36. Van Zundert J, Brabant S, Van de Kelft E, Vercruyssen A, Van Buyten JP. Pulsed radiofre- quency treatment of the Gasserian ganglion in patients with idiopathic trigeminal neuralgia. Pain 2003; 104: 449-452 37. Zakrezewska J, Lopez BC. New techniques for surgical management of trigeminal neuralgia. Pain 2004; 109: 520 38. Van Zundert J, Brabant S, Van de Kelft E, Vercruyssen A, Van Buyten JP. Response to Zakrzewska's Letter to the Editor. Pain 2004; 109: 520-522 39. Zakrzewska JM, Lopez BC. Quality of reporting in evaluations of surgical treatment of trigemi- nal neuralgia: recommendations for future reports. Neurosurgery 2003; 53: 110-120; discussion 120-112 40. Lau J, Antman EM, Jimenez-Silva J, Kupelnick B, Mosteller F, Chalmers TC. Cumulative meta- analysis of therapeutic trials for myocardial infarction. N Engl J Med 1992; 327: 248-254 41. Devulder J, Crombez G, Brusselmans G. Comment on "Pulsed radiofrequency treatment of the gasserian ganglion in patients with idiopathic trigeminal neuralgia". Pain 2004; 107: 195 42. Orlandini G. Pulsed percutaneous radiofrequency treatment of the Gasserian ganglion for ther- apy of trigeminal neuralgia: technical notes, validity of the method and selection of the patients. Pain 2004; 108: 297-298 43. Van Zundert J. Response to 'Comment on Pulsed radiofrequency treatment of the Gasserian Ganglion in patients with idiopathic trigeminal neuralgia'. Pain 2004; 107: 196 44. Van Zundert J, Brabant S, Van de Kelft E, Vercruyssen A. Response to Orlandini's letter to the editor, regarding Pulsed radiofrequency treatment of the Gasserian ganglion. Pain 2004; 108: 298-299 45. Nelemans PJ, deBie RA, deVet HC, Sturmans F. Injection therapy for subacute and chronic benign low back pain. Spine 2001; 26: 501-515 46. Koes BW, Scholten RJPM, Mens JMA, Bouter LM. Epidural Steroid Injections for Low Back Pain and Sciatica: An updated Systematic Review of Randomized Clinical Trials. Pain Digest 1999; 9: 241-247 47. Abejon D, Reig E. Is Pulsed Radiofrequency a Neuromodulation Technique? Neuromodulation 2003; 6: 1-3
. |
|
CRPS tryptich studies on diagnosis, pathogenesis and treatment of the Complex Regional Pain Syndrome. Measurements and treatment
|
|
Complex Regional Pain Syndrome
(CRPS) is being called Reflex Sympathetic Dystrophy. There are many synonyms for this
syndrome, and even more definitions. In general, one can say that CRPS is a chronic pain
syndrome with sensory, autonomic and motor dysfunction. Chapter 1 summarizes the
contemporary view on its diagnosis, treatment and hypotheses on etiopathogenesis. The
scientific relevance of chapter 2 - 7 of this thesis is pointed out by comparing our
studies with the literature. Numerous signs and symptoms
are being attributed to the CRPS syndrome. So far, there is no agreement which and in
which way these need to be found in a patient to diagnose CRPS. CRPS, as a diagnosis, is
not valid. Chapter 2 describes our study on this problem. We found that this is not only
caused by different definitions for CRPS, but that also the assessment of signs and
symptoms in patients is problematic. Important signs like edema, dystrophic changes and
changed hair growth have a low interobserver agreement, even between CRPS experts (chapter
2). There is also a discrepancy
between signs and symptoms in CRPS patients. Many patients complain about swelling,
discoloration and heat/cold sensations without these being confirmed by a physician during
physical examination. The most objective signs like swelling or changed hair and nail
growth are the least often confirmed in comparison with behaviour dependent signs like
hypersensitive skin and limb dysfunction (chapter 4). This discrepancy between outcomes of
patient interview and physical examination is also described in other diagnoses like
irritable bowel syndrome, fibromyalgia, chronic fatigue syndrome and somatoform disorder. Physicians are being taught
implicitly to consider psychopathology as a differential diagnosis when a patient is
presented with chronic, idiopathic pain with relatively few objective symptoms, changing
signs in time and multiple diagnoses being addressed to in the past by other physicians.
It therefore thinkable this diagnosis is sometimes being given to CRPS patients and vice
versa. In our study with 6 CRPS specialists (surgeons, anaesthesiologists and a
neurologist) somatoform disorder was an important differential diagnosis for chronic CRPS
patients (chapter 3), especially for those physicians who also often treated other chronic
pain patients. Of course, this is no proof for CRPS being a subtype somatoform disorder,
but it is likely that patients with a somatoform disorder sometimes are misdiagnosed as
having CRPS, Clinical research in CRPS is not very useful as
long as there are no generally accepted diagnostic criteria. For the time being, CRPS
criteria in research need to be thorough, even though it may minimize the possible study
population. Chapter 5 describes a study on
HLA antigen subtypes. We found an association of CRPS with MHC class II antigen HLA-DQ1.
This antigen was prevalent in 42 % of the controls in comparison to 69% in CRPS patients.
This implies a possible genetic predisposition for (a subtype) of CRPS and this
predisposition might involve the immune system. An association for HLA-DQ1 is also
described for narcolepsy and multiple sclerosis. Chapter 6 describes the study
of antibodies to viruses and bacteria in CRPS. An association between CRPS and Parvovirus
B19 infection was found. It is speculated that this virus might start an immune reactivity
in predisposed patients, and thus result in a painful post infectious reaction after a
trauma of a limb (read: after disposing antigen to the immune system). This would
partially explain why only some people develop this syndrome after trauma. Of course, this
model is incomplete as not every patient with HLA DQ1 and Parvovirus B19 infection
develops CRPS and our results are not yet confirmed by other researchers. Now only
symptomatic treatment is possible, with limited effect. In this thesis’ last
chapter a patient study on gabapentin treatment is described. In a placebo controlled
research we found that in chronic CRPS patients gabapentin can fulfil a small part in pain
management, but can also reverse sensory deficit. Conclusions These study results show that
CRPS is a syndrome with vague boundaries. The physical examination is not valid on most
relevant aspects. An important differential diagnosis of CRPS is somatoform disorder.
Nevertheless, it should not be forgotten that CRPS is an invalidating illness, which has
huge impact on daily life. There are findings that points toward a genetic and/or
inflammatory cause. A good therapy is not yet available. Gabapentin can be a possible aid
in pain management and reversing of sensory deficits. |