1 Department of Neurosurgery, Radcliffe Infirmary, Woodstock Road, Oxford, UK. 2 University Laboratory of Physiology, University of Oxford, Parks Road, Oxford, ...
Acta Neurochir Suppl (2007) 97(2): 521–528 # Springer-Verlag 2007 Printed in Austria
The periaqueductal grey area and the cardiovascular system A. L. Green1;2 , S. Wang2 , S. L. F. Owen2 , and T. Z. Aziz1;2 1 2
Department of Neurosurgery, Radcliffe Infirmary, Woodstock Road, Oxford, UK University Laboratory of Physiology, University of Oxford, Parks Road, Oxford, UK
Summary In this chapter, we report that blood pressure can be increased or decreased depending on whether an electrode is in ventral or dorsal PAG. We also describe that it is theoretically possible to treat orthostatic hypotension. These are exciting developments not only because they provide an example of direct translational research from animal research to humans but also because they highlight a potential for future clinical therapies. The control of essential hypertension without drugs is attractive because of the side effects of medication such as precipitation of heart failure [10]. Similarly, drug treatment of orthostatic hypotension cannot differentiate between the supine and standing positions and can therefore lead to nocturnal hypertension [22, 29]. A stimulator could be turned off at night or contain a mercury switch that reacts to posture. Keywords: Neuromodulation; deep brain stimulation; DBS; periaquaductal grey area; periventricular grey area; cardiovascular system; orthostatic hypotension.
Introduction The periaqueductal grey area (PAG) is well known to be important in the modulation of pain and is an area where deep brain stimulating electrodes are often placed for the treatment of chronic, intractable neuropathic pain [6, 16, 25]. However, in mammals, this region is known to be an important component in the defence reaction [7]. The defence reaction is an integrated response from the forebrain down to the cardiovascular system that is associated with survival in the wild [19]. For example, if escape from danger is a possibility, the response involves a ‘fight or flight’ reaction that includes raised blood pressure and heart rate, non-opioid mediated analgesia and emotional effects such as fear [9]. On the other hand, if escape is unlikely, the reaction consists of lowered blood pressure, opioid-mediated analgesia and ‘passive’ behaviour as well as fear [12, 21]. Electrical
stimulation of the PAG in animals will elicit these defence reactions and thus, it is likely that stimulation of the same area in the human will affect not only the pain modulation pathways, but also the cardiovascular components of this system. As we have patients with electrodes implanted into the rostral part of the PAG, we are in an ideal position to study the effect of PAG stimulation in the human. A limited amount of previous evidence exists to suggest that stimulation of the human PAG causes cardiovascular changes in humans [36]. Here, we characterise these effects in detail. Alteration of blood pressure with PAG stimulation We have shown that electrical stimulation of the human PAG alters blood pressure [17]. In this study of fifteen chronic neuropathic pain patients (17 electrodes), blood pressure and ECG were continuously measured in the laboratory whilst stimulation parameters were altered (either 10 or 50 Hz i.e. in the frequency range used to treat chronic pain). We found that cardiovascular responses to stimulation were consistent (on at least three occasions) for any pair of electrode contacts used. Overall, arterial blood pressure significantly decreased in seven pairs of electrode contacts in seven patients (significance was determined using one-way analysis of variance of blood pressure with time – ANOVA). Conversely, blood pressure significantly increased in six pairs of contacts. These results are summarised in Fig. 1. The average reduction in SBP (for those in whom BP was reduced) was 14.2 3.6 mmHg (range 7–25 mmHg), or 13.9%, after 300 s stimulation. Figure 1A shows that the drop in SBP is accompanied by a fall in diastolic
522
A. L. Green et al.
Fig. 1. (A) Changes in cardiovascular parameters associated with reduced blood pressure. Patterned area Period of stimulation. Grey area one standard error of the mean. SBP Systolic blood pressure, DBP diastolic blood pressure, PP pulse pressure, RR interval time period between R waves on electrocardiogram, dP=dt change of systolic blood pressure with time. (B) Changes in cardiovascular parameters associated with increased blood pressure (see text for details)
BP (DBP) of 4.9 mmHg 2.9 (p ¼ 0.03, single factor ANOVA, n ¼ 7, range 1.5–9.3), equivalent to 6%. This implies a degree of peripheral vasodilatation. However, as the systolic drops more than the diastolic BP (leading to a reduction in pulse pressure), the mechanism is unlikely to be related simply to peripheral vascular changes. We therefore measured the change of SBP with time (maximum dP=dt i.e. the slope of the blood pressure curve). This is known to be a marker of cardiac contractility [8] as the harder the myocardium contracts, the steeper the slope of this curve. This revealed a mean reduction of 222 mmHg=s 126 (19.8%, p ¼ 0.06). This is suggestive, but not absolute proof that the contractility of the myocardium was reduced. On the other hand, R–R interval (a measure of heart rate) did not change significantly throughout the stimulation period (mean change ¼ 0.01s 0.04, range 0–0.08). As heart rate is controlled via the vagal nerve, this implies that there was no change in parasympathetic activity. For those with an increase in BP, the mean rise in SBP was 16.73 mmHg 5.9 (p< 0.001, single factor ANOVA, n ¼ 6, range 16–31 mmHg), equivalent to 16.4% at the end of a 400 s period where stimulation was started at 100 s (however, the maximum rise of 22.23 mmHg occurred just before this – see Fig. 1B). Stimulation parameters required to raise BP were the same as with the episodes
of reduced BP (i.e. 10 Hz, 120 ms and up to 3.0 V), except that 50 Hz did not have the same effect in any patient. As with BP reduction, increases were accompanied by a smaller rise in DBP of 4.9 mmHg 2.8 or 6.4% (p ¼ 0.04, single factor ANOVA, n ¼ 6, range ¼ 2.4– 12.1 mmHg). There was also an increase in mean pulse pressure and again, the maximum rise of 17.33 mmHg occurred just before 400 s. Maximum dP=dt increased by 212 97 mmHg=s (p < 0.03, single factor ANOVA). As with reduction in BP, there was no significant change in R–R interval. Thus, it appears that increasing BP is accompanied by a mirror of the changes that occur during reduction in BP. Six control patients were investigated (six thalamic electrodes, one spinal cord stimulator). Despite extensive investigation using a variety of frequencies and voltages, as well as a variety of electrode contact configurations, we were unable to modulate the BP in any of these patients. In addition to the control electrodes that had no effect on BP, four patients with PVG electrodes (six electrodes in total) also had no effect. Electrode location Because blood pressure changes in animals depend on whether the electrode is in ventral or dorsal PAG, we
The periaqueductal grey area and the cardiovascular system
523
Fig. 2. (A) Sagittal positions of the electrodes in patients in whom there were changes in blood pressure. (B) Coronal positions. For clarity, patients with no changes are not shown. Note that patients #1–7 all had reduction in BP (black contacts) and have the most ventral electrodes. Conversely, #8–11 and the upper 2 contacts of #1 and #6 had a rise in BP (patterned contacts). Gray contacts are those that, when stimulated, had no effect on BP. AC Anterior commissure, PC posterior commissure, PVG periventricular gray, PAG periaqueductal gray, SC superior colliculus (the level of which is depicted by the dotted circle in 1B), RN red nucleus, III third ventricle, Aq aqueduct. Inset of A shows the AC–PC plane, inset of B shows the slice position
looked at electrode position. These were plotted on a brain atlas [23] using the post-operative MRI and a manipulation program (MRIcro version 1.38 build 1, Chris Rorden). The results are shown in Fig. 2. This shows that those electrodes that reduced blood pressure were placed ventrally, as compared to the dorsal electrodes that increased blood pressure. Patients with no blood pressure changes are not shown for clarity. However, we plotted electrode positions for five of these six electrodes (one had not had a post-operative scan). Four of the five electrodes were dorsal to the group that raised BP and were therefore probably outside the PAG=PVG. The remaining electrode was in mid-PVG. Changes in BP were compared between the two groups of ventral and dorsal electrodes (n ¼ 8 and 9, respectively – unlike the changes described above, this included all patients, even those without significant changes in BP). The mean peak change in SBP was 10.3 2.8 mmHg for the ventral group and þ10.8 3.1 mmHg for the dorsal group. Comparison using one way ANOVA showed significance (p ¼ 0.003). Similarly, the mean peak change in DBP was 4.6 1.2 and þ3.5 0.8 mmHg, respectively (p ¼ 0.007). Mean peak change in pulse pressure ranged from 8.6 3.5 mmHg for the ventral to þ7.4 2.1 mmHg for the dorsal group (p ¼ 0.01).
dP=dt ranged from 181.6 28 mmHg=s for ventral and þ82 26 mmHg=s for dorsal electrodes (p ¼ 0.007). Comparison of RR interval between the two groups did not reveal any significant difference (p ¼ 0.13). Power spectral analysis of systolic blood pressure It is possible to elucidate underlying mechanisms of blood pressure changes by looking at the dominant frequencies in the blood pressure wave-form [26]. Activity in the range just under 0.1 Hz is associated with activity of the sympathetic nervous system – a wave known as Meyer’s wave [28, 13]. We, therefore, performed autoregressive power spectral analysis of the blood pressure waveform in all patients. Frequencies below 0.02 Hz were filtered out to remove the trend in the signal (see [33] for methodology). Figure 3A shows a typical example in a patient whose blood pressure could be increased or decreased depending on which contacts were used. It can be seen that with an increase in blood pressure, there was a large increase in the low frequency wave in the 0.1 Hz region. With a reduction in blood pressure, there was a corresponding decrease. This implies that increase in blood pressure is associated with an increase in sympathetic activity, and vice versa.
524
A. L. Green et al.
Fig. 3. Changes in low- and high-frequency power spectra of systolic blood pressure. (A) Example in one patient, in whom blood pressure could be increased or decreased, depending on which contacts were used. A change in the low-frequency component was associated with change in blood pressure, implying changes in sympathetic nervous system activity. (B, C) Changes for the groups in whom blood pressure decreased (n ¼ 7) or increased (n ¼ 6), respectively. Error bars denote 1 SEM
To look at the group results, we calculated the power of the low- and high-frequency components as the integral of the power spectra between 0.05 and 0.15 Hz and between 0.15 and 0.4 Hz. The logarithm of the low- and high-frequency power for the two groups of patients (blood pressure increase or decrease) ON and OFF stimulation were analysed using a paired t-test (Fig. 3C, D). This revealed that for the group as a whole, there was a change in low-frequency power spectra that corresponded to blood pressure changes. There were also changes in high-frequency power, but these were not significant (this may be due to small numbers). Can we treat essential hypertension? We have demonstrated that it is possible to increase or decrease blood pressure in humans with electrical stimulation of the PAG. Furthermore, the direction of
blood pressure change can be controlled by placing the electrode in either ventral or dorsal PAG. Essential hypertension is a significant clinical problem that has a skewed distribution and can lead to stroke or myocardial infarction [2, 34]. Approximately, 3% of these patients are refractory to treatment [1]. Reducing blood pressure with deep brain stimulation is theoretically possible but in itself poses a risk; there is approximately one in three hundred risk of stroke as well as other less serious, but nevertheless troublesome complications such as infection and hardware problems [5]. Therefore it is unlikely that deep brain stimulation could be justified at the present time. As well as further elucidating the mechanisms of DBS on blood pressure, the risk of the procedure needs to be reduced for what is essentially a prophylactic operation. This may come about with advances in technology such as nanotechnology etc. that may lead to smaller, less invasive electrodes. However,
The periaqueductal grey area and the cardiovascular system
Gotoh et al. [15] have shown that it is theoretically possible to control blood pressure at a given value by altering electrical stimulation of cardiovascular centres in the medulla according to blood pressure measurements made via an arterial line. This shows proof of principle. Can we treat orthostatic hypotension? We have shown that we are able to increase as well as decrease blood pressure with PAG stimulation. This raises the possibility that we might be able to treat orthostatic or postural hypotension (OH). In the normal subject, assumption of an upright posture leads to pooling of venous blood in the lower extremities and splanchnic circulation [14]. The resulting decrease in venous return
525
to the heart leads to a compensatory, centrally mediated increase in sympathetic and decrease in parasympathetic activity (known as the baroreceptor reflex). This activity usually results in a transient fall in systolic blood pressure of 5–10 mmHg, a small rise in diastolic blood pressure (5–10 mmHg) and a rise in heart rate of 10–25 beats per minute. In orthostatic hypotension, patients suffer troublesome low blood pressure on standing or symptoms of cerebral hypoperfusion [32]. It is present in up to 20% of people over 65 and its treatment may lead to troublesome raised blood pressure [29, 22]. Ascending projections of barosensitive adrenergic cells in the rostroventrolateral medulla project to PAG [18]. There is evidence that chemical stimulation of the PAG inhibits baroreflex vagal bradycardia in rats [20]. Thus it is conceivable that stimulation of this area in the
Fig. 4. Blood Pressure and Heart Rate changes on standing. (A–C) Mean changes in systolic blood pressure for subject #1, MOI group, and nonMOI group, respectively. (D–F) Changes in heart rate for the same groups. All traces include the mean of three sessions, averaged every ten seconds. MOI Mild orthostatic intolerance group, nonMOI no mild orthostatic intolerance group. Grey area Period when patient was sitting, white area (from 0 seconds) period of standing. & Stimulation ‘OFF’, œ stimulation ‘ON’. Error bars show 1 standard error of the mean
526
A. L. Green et al.
human may affect the baroreceptor reflex. We performed a study in eleven patients with PVG=PAG stimulators for neuropathic pain, in which we continuously recorded blood pressure while sitting for 100 s, followed by 280 s of standing. Subjects were grouped into three categories as follows; subject #1 had a history of orthostatic hypotension that had resolved after the stimulator was inserted (whilst it was on). The second group (five subjects) had mild orthostatic intolerance (MOI-group) that was defined as a fall in systolic blood pressure of >20 mmHg on standing, but no clinical symptoms. The third group (five subjects) had no significant postural effects on blood pressure (non-MOI group). We showed that stimulation in subject #1 significantly reduced the postural drop in blood pressure (from 28.2 to 11.1%, p < 0.001, t-test) and in the MOI group, completely reversed it (p < 0.001, t-test) (Fig. 4A, B). In the control group (non-MOI), there was no significant difference in blood pressure between the two groups (Fig. 4C). Figure 4D–F shows that absolute heart rate changes on standing were not significantly altered with stimulation. However, Fig. 4D shows that with stimulation, the heart rate variability appears to be increased (there is a greater oscillation in heart rate with stimulation). To formally assess this, we looked at the power of RR interval spectra. The power of RR interval spectra in the high frequency band (0.15–0.4 Hz) has been shown to be a marker of cardiac vagal control [27, 31]. The low frequency band (0.04–0.15 Hz) has been associated with cardiac sympathetic activity, although it has been shown to be affected by both vagal and sympathetic nerves [4, 31]. Previous research has shown a reduction in both of these components of heart rate variability power
with head up tilt in patients with autonomic neuropathy [37] compared to the increase in low-frequency power seen on standing in normal subjects [30]. We performed auto-regressive power spectra analysis of RR interval on all patients in the study. In the MOI and Non-MOI groups, baseline low-frequency power of RR interval significantly increased with stimulation (t-test, p ¼ 0.021 and p< 0.001, respectively, Table 1). However, baseline high-frequency power in these groups was not significantly altered by the stimulation (p> 0.1, t-test). In the MOI group and subject #1, the reduction in both low and high frequency power associated with standing was prevented with stimulation (p ¼ 0.008, t-tests, ‘ON’ vs. ‘OFF’). These results suggest that stimulation may increase the cardiac sympathetic activity and enhance its response to standing. Another way of elucidating mechanisms of the effects of stimulation on postural changes is to look at baroreflex sensitivity. In young and middle-aged healthy subjects, baroreflex sensitivity decreases on standing [11, 30, 35]. In autonomic neuropathy, such as that of diabetes, it has been shown that it is lower in the supine position and there is less further reduction on standing than in normal subjects [30]. We calculated the baroreflex sensitivity index from the transfer function of systolic blood pressure and RR interval signals using bivariate autoregressive modeling [3, 35]; RRðnÞ ¼ Pp Pp k¼1 a11 ðkÞRRðn kÞ þ k¼1 a12 ðkÞSBPðn kÞ þ wðnÞ. We showed that the baroreflex sensitivity in subject #1 (i.e. orthostatic hypotension) and MOI groups were similar to those with a mild autonomic neuropathy (Table 1). We also showed that stimulation significantly raises sensitivity in the sitting position (t-test, p ¼ 0.018,
