Summary and conclusions

Transcription

1 Summary and conclusions Syncope is an extremely common symptom with a life time prevalence of 30 to 40% in the general population. 52,81,212 In common with its frequent occurrence, syncope accounts for 1 to 1.5% of emergency department visits and up to 6% of all hospital admissions. 17,29,60,64,200,209 The pathophysiological hallmark of syncope is transient global cerebral hypoperfusion. History taking is the most important tool to differentiate syncope from other causes of transient loss of consciousness (TLOC), including epilepsy. 29 Detailed questioning of the circumstances and prodromal features may also provide vital clues to distinguish reflex syncope from other, less prevalent, causes of syncope. 4,41,222 Hence, the overall diagnostic yield and accuracy of history taking in combination with physical examination and ECG at initial evaluation is high. 251 Accordingly, the use of additional testing beyond taking a history, performing a physical examination, and an ECG, can be avoided in the majority of patients with TLOC. Nevertheless the fact that syncope might be caused by a deadly condition may fuel a shotgun approach aimed at investigating all disorders that may cause syncope or other causes of TLOC. 144 The shotgun approach does not have a high diagnostic yield but is still apparently not rare: admissions for syncope are common and account for large costs. US data from 2005 indicate a cost of $2.4 billion per year, with a mean cost of $5,600 per admission. 239 The reasons for such an approach are not altogether clear. The lack of a consistent terminology of syncope and related disorders might contribute to this problem. The terminology of syncope is discussed in more detail in Part I of this thesis. Part II and III address the elusive pathophysiology of syncope. In Part II the role of orthostatic fluid shifts in autonomic failure is questioned, while part III discusses the effects of hyperventilation on syncope and blood pressure regulation and reviews the potential of respiratory manoeuvres to prevent orthostatic hypotension in autonomic failure. Part IV contains various clinical studies illustrating a number of the above-mentioned factors. Part I Terminology and classification Chapter 1 systematically reviews definitions found in the current medical literature. Syncope and related conditions proved to be infrequently and inconsistently defined. The paucity of definitions is rather remarkable, given that the search primarily encompassed the 10 journals 177

2 with the highest impact in the fields of cardiology, internal medicine, and neurology. Salient features of the definition of syncope (its brief duration, its self-limiting character, and its causative mechanism) were frequently omitted. Definitions often treated syncope as synonymous with 'transient loss of consciousness', hereby including epilepsy and even conditions such as concussion as 'syncope'. Of the various keywords investigated, orthostatic hypotension was the most frequently defined one. For this term we observed an increase in number and consistency of definitions after publication of a consensus statement in 1996, illustrating the beneficial effects obtained through a consensus statement. Hence, a clear classification system for syncope and related disorders is strongly recommended. In Chapter 2, the terms in use designating transient loss of consciousness in a Dutch emergency ward were evaluated and compared with the definitions of the European Society of Cardiology (ESC). The prevalence of a non-traumatic TLOC diagnosis in the emergency ward during the two months of study was 2.9%. The Dutch word collaps was the most frequently used term, but its use was not consistent, and could span a wide range of ESC categories including syncope, TLOC, or no TLOC. At the time of writing we advised to avoid this term for specific disease entities, and to reserve its use for a condition with little specific features, i.e., a fall without an obvious external cause. At present, however, our position has evolved to the extent that it may be better to abolish its use altogether, and to use only terms that conform to ESC categories. We recommended the term wegraking for transient loss of consciousness without a clear cause, and syncope for transient loss of consciousness due to temporary global cerebral hypoperfusion. A proposal for a pathophysiological classification of syncope is presented in Chapter 3, taking into account the findings of the preceding chapters. Syncope and related conditions are infrequently and inconsistently defined in current medical literature. Salient features of the definition of syncope (its brief duration, its self-limiting character, and its causative mechanism) are frequently omitted. Definitions often treat syncope as synonymous with 'transient loss of consciousness', hereby including epilepsy and other conditions as a cause of syncope. There is a strong need for a systematic terminology for syncope and related conditions. Part I 178

3 Part II Orthostatic fluid shifts In Chapter 4, we evaluated the use of strain gauge plethysmography to estimate the time course of orthostatic fluid shifts in the 'free hanging' position. Our method provided stable results, based on the good agreement and significant correlation between the measurements of both legs. In addition, the occlusion test provided evidence that the changes during head-up tilt with a parachute harness were indeed predominantly caused by orthostatic fluid shifts. We found an unexpected gender difference in orthostatic pooling: calf volume during head-up tilt increased more in men than in women. This difference was only apparent during the acute phase (0-1 min) of tilting. In the prolonged phase (1-5 min) the slope of the time/volume relationship was similar for the male and female groups. The gender differences in orthostatic pooling in the calf may be explained by a higher calf venous compliance (i.e., more distensible veins) in men than in women, together with a greater hydrostatic pressure due to a greater height in men. With this newly described method the amount of orthostatic pooling in patients with autonomic failure is compared to that in healthy controls in Chapter 5. Somewhat surprisingly, we found no evidence for excessive pooling at the level of the calf in patients with autonomic failure during the course of the day despite marked orthostatic hypotension. No correlation was found between the degree of orthostatic hypotension and the orthostatic calf volume change in autonomic failure. In one hospitalised patient an additional measurement was performed in the early morning directly after getting up, demonstrating a larger, but still small increase of orthostatic fluid shifts. This slight increase of pooling may be explained by the absence of edema in the early morning. Collectively, these findings suggest that orthostatic hypotension in autonomic failure should not be ascribed to excessive pooling at the level of the calf. Venous pooling in the thighs, buttocks or the splanchnic vascular bed might be of greater importance to orthostatic hypotension in autonomic failure. Orthostatic fluid shifts at the level of the calf are greater in men than in women. This gender difference in orthostatic calf volume changes may be explained by a higher calf venous compliance in men than in women, together with a greater hydrostatic pressure due to a greater height in men. Orthostatic calf volume changes are not augmented in autonomic failure. Therefore, orthostatic hypotension in autonomic failure should not be ascribed to excessive pooling at the level of the calf. Part II 179

4 Part III Effects of hyperventilation Chapter 6 questions the concept of hyperventilation-induced syncope. In eleven healthy subjects the cardiovascular effects of prolonged hyperventilation versus normal ventilation were examined for 15 min in the supine as well as 15 min in the head-up tilted position. Prolonged hyperventilation did not provoke syncope in any of the subjects. In both positions hyperventilation significantly decreased mean cerebral blood flow velocity. Hyperventilation caused mild changes of blood pressure in the supine but a significant augmentation of blood pressure in the tilted position. Collectively, these data underscore that hyperventilation does not act as a significant trigger for syncope in healthy subjects, since prolonged hyperventilation did neither induce syncope nor hypotension. Hence, these findings do not support the inclusion of hyperventilation in the differential diagnosis of syncope. Chapter 7 elaborates on the physiological mechanisms related to the blood pressure changes during hyperventilation. To identify factors affecting the blood pressure response, we dissected the effects of hypocapnic and isocapnic hyperventilation so as to distinguish mechanical from biochemical effects and evaluated the effects of acute vs. prolonged hyperventilation. We found that hyperventilation had a time dependent effect on blood pressure: prolonged hyperventilation led to a gradual augmentation of blood pressure. Furthermore, the increase of blood pressure was higher when hypocapnia was prevented. These findings suggest that hyperventilation has two contrasting effects on blood pressure control: one is that hypocapnia causes blood pressure to decrease, presumably through peripheral vasodilatation; the second is that the mechanical effect of hyperventilation tends to increase blood pressure, presumably because the increased pumping not merely affects the flow of air, but also that of blood. The thought that this increased additional circulatory pump is under volitional control prompted the experiment described in Chapter 8, where we studied the efficacy of respiratory manoeuvres to reduce orthostatic hypotension in autonomic failure. Blood pressure after standing up was recorded in ten patients in five conditions: normal standing, leg muscle tensing, inspiratory pursed lips breathing, inspiratory sniffing, and a device causing inspiratory obstruction. The manoeuvres caused significant differences in standing blood pressure. Inspiratory obstruction and leg muscle tensing increased blood pressure to a comparable degree. The effect of inspiratory pursed lips breathing and inspiratory sniffing depended on concomitant hypocapnia. Respiratory manoeuvres can indeed reduce orthostatic 180

5 hypotension in autonomic failure through activation of the respiratory pump, with the important caveat that hypocapnia must be avoided. Hyperventilation is not a significant trigger to induce syncope in healthy subjects and should therefore not be included in the differential diagnosis of syncope. Hyperventilation has two contrasting effects on blood pressure regulation: hypocapnia causing vasodepression and the activation of the respiratory pump (hyperpnea without hypocapnia) acting as a vasopressor. Respiratory countermanoeuvres activating the respiratory pump are effective measures to reduce orthostatic hypotension in autonomic failure provided hypocapnia is avoided. Part III Part IV Clinical studies Chapter 9 examines the association between migraine and syncope-related autonomic nervous system symptoms. We conducted a population-based study among migraineurs with and without aura (n=323) and control subjects (n=153). A systematic questionnaire and cardiovascular measurements during rest, while standing and after venipuncture addressed the prevalence of syncope, orthostatic intolerance, orthostatic hypotension (OH) and the postural tachycardia syndrome (POTS) in migraineurs and controls. The lifetime prevalence of syncope in all participants was 41%, more often in women. Compared to controls, migraineurs had a significantly higher lifetime prevalence of syncope (46 vs. 31%), frequent syncope (five or more attacks) (13 vs. 5%) and orthostatic intolerance (32 vs. 12%). There was no association between autonomic nervous system symptoms and the severity of migraine or migraine subtype. Cardiovascular measurements and the prevalence of POTS and OH did not differ significantly between migraineurs and controls. This population-based study thus demonstrated an elevated prevalence of syncope and orthostatic intolerance in migraineurs without clear interictal signs of autonomic nervous system dysfunction. The mechanism underlying the susceptibility to syncope in migraine remains to be elucidated. Chapter 10 assesses the accuracy of eyewitness accounts of TLOC. Videos of a generalised tonic-clonic seizure and of reflex syncope were unexpectedly shown to 125 and

6 psychology students respectively during a lecture on an unrelated subject. Directly afterwards, the students filled in a multiple-choice questionnaire regarding muscle tone, twitches, head deviation, eye closure, gaze deviation, drooling and facial colour. The consensus of experienced neurologists served as a gold standard. Even though not all items could be ascertained from the videos, the full range of questions was included to simulate clinical practice. Of all responses to the observable items on the syncope video 44% were correct, 28% erroneous and 29% concerned I do not know responses. The observable items on the epilepsy video yielded 60% correct responses, 18% erroneous responses and 22% "I do not know" responses. Regarding features that were not visible on the videos, 77% of the responses were accurate ("I do not know"), while 23% erroneously provided an observation. These findings suggest that an eyewitness account of a single episode of TLOC should be interpreted with caution as salient features are frequently overlooked or inaccurately recalled. However, the accuracy of the eyewitness observations of TLOC differs however per item; muscle tone was reported with high accuracy. In Chapter 11 a sportsman is reported with frequent exercise-related syncope. A trial of rapid ingestion of 1000 ml tap water prior to exercise proved successful to prevent such episodes. Because of the short lasting pressor effect and its minimal side effects, we suggest water drinking as a simple and possibly effective therapy for idiopathic exercise-related syncope. Chapter 12 describes a patient with pure autonomic failure suffering from hypotensive complaints during emotional stress. During a tilt table test, hyperventilation was shown to cause a marked fall of blood pressure. It is hypothesized that emotional stress may trigger hypotension through hyperventilation in subjects with autonomic failure. In Chapter 13 a patient is presented with rapidly sequential vasovagal syncope. Inspired by the similar occurrence of repeated epileptic seizures, the attacks are labelled a status vasovagalis. Syncope and orthostatic intolerance are more prevalent in migraine. An eyewitness account of a single episode of transient loss of consciousness should be interpreted with caution as salient features are frequently overlooked or inaccurately recalled. However, the accuracy of eyewitness observations of transient loss of consciousness differs per item; muscle tone was reported with high accuracy. Part IV 182

7 Future perspectives The findings described in this thesis provide new opportunities for further research. Hyperventilation affects vessels both directly and by influencing the vasomotor centres and neurohumoral systems, resulting in a set of complex interacting and possibly conflicting effects. While the net effect in healthy controls is a net increase in blood pressure, this is definitely not the case in subjects with sympathetic denervation, in whom hyperventilation invariably decreases blood pressure through hypocapnic vasodilatation (Figure 1). 37,145,192 This profound contrast in blood pressure responses to hyperventilation between healthy subjects and those with autonomic failure underscore the importance, as well as the complexity, of the integrity of the sympathetic nervous system in the maintenance of normotension during hyperventilation. While we stress that hyperventilation by itself is not a significant trigger to evoke syncope in healthy subjects, hyperventilation might have act as an additional negative influence in patients with frequent vasovagal syncope, in view of the acute withdrawal of sympathetic outflow prior to syncope. 88,182 Recently, Norcliffe-Kaufmann et al. demonstrated that individuals prone to vasovagal syncope have a greater reduction in cerebral blood flow and a greater dilatation of the forearm vasculature in response to hypocapnia compared with healthy control subjects. 187 These data strongly suggest that hyperventilation may lower the threshold to syncope in patients with frequent vasovagal syncope. In addition, it would be of interest to determine whether patients with panic disorder are more prone to syncope through hyperventilation, since hyperventilation is an important feature of a panic attack. Martinez et al. reported that prolonged hyperventilation without feedback of actual carbon dioxide levels caused significant increases of blood pressure in panicking patients, compared with nonpanicking patients or normal controls. 166 Yet a definite proof of the effects of hyperventilation in patients with panic disorder would however require a more elaborate manipulation of carbon dioxide levels, since it is well known that nonstandardised hyperventilation tests result in considerable variation in the degree of hypocapnia. 19 While we established the contrasting effects of acute vs. prolonged hyperventilation in healthy subjects, we stress that the precise mechanism underlying the time course of these changes is as yet unknown. The fairly long latency of the vasopressive effect of prolonged hyperventilation, in the order of about five minutes, suggests involvement of neurohumoral control mechanisms. 95 As the circulatory response to hyperventilation is modulated by the opioid and the histaminergic system, it would be of interest to study whether changes in these systems explain how hyper- 183

8 Figure 1 The very first observations of the effects of CO 2 on blood pressure regulation in autonomic failure were made by Laplace in A patient with autonomic failure ( Bradbury Eggleston syndrome ) breathed through 1 meter long tubes, hereby increasing dead space and thus alveolar CO 2. Increasing CO 2 caused a striking increase of blood pressure (systolic blood pressure (s), diastolic blood pressure (d)). Conversely, lowering CO 2 by ten deep breaths caused a large fall of blood pressure with a rapid recovery after cessation of hyperpnea. Despite these marked blood pressure changes pulse rate (p, dots) remained relatively stable, indicating vagal denervation. Reproduced with permission from the American College of Physicians. 145 ventilation increases blood pressure over time. 75,201 In addition, the vasopressive effect of isocapnic hyperventilation needs further study. In our study this blood pressure augmentation was related to an increase of total peripheral resistance. Further experiments relating blood pressure changes to sympathetic outflow and addressing the influence of different ventilatory patterns may provide more insight into the effects of isocapnic hyperventilation on blood pressure regulation. An example would be the comparison of a 'fast shallow' pattern of a low tidal volume and high breathing frequency with a 'slow deep' mode, concerning high tidal volume together with a low breathing frequency In Chapter 8 we established the beneficial 184

9 effects of activation of the respiratory pump in autonomic failure. These findings have been confirmed by Melby et al. 173 In that study, Melby et al. made use of an impedance threshold device (ITD) and elaborated on the underlying physiological mechanisms: increasing inspiratory impedance augmented blood pressure in patients with orthostatic hypotension by an increase in total peripheral resistance without altering stroke volume. 173 The increase of total peripheral resistance is surprising as it contrasts with previous studies by Convertino et al. demonstrating a predominant effect of the ITD on cardiac output in healthy controls This discrepancy thus warrants further research. Nevertheless, if respiratory manoeuvres increase total peripheral resistance in autonomic failure, it would be of interest to study whether respiratory manoeuvres and leg crossing may act synergistically, since leg crossing augments standing blood pressure solely through an increase of cardiac output in patients with vasovagal syncope and autonomic failure. 138,241,252 Further studies on the effects of combining respiratory manoeuvres and muscle tensing might thus prove fruitful. Although respiratory manoeuvres have been shown to raise blood pressure in laboratory conditions, it is yet unclear whether respiratory manoeuvres improve orthostatic symptoms and prevent falls in daily life. Hence, a prospective, randomized clinical trial with a design similar to the physical counterpressure manoeuvres trial, 253 is an essential next step to establish the effectiveness of respiratory manoeuvres in autonomic failure. 185