Cushing\'s mechanism maintains cerebral perfusion pressure in experimental subarachnoid haemorrhage

Neuroscience Letters 529 (2012) 92–96

Contents lists available at SciVerse ScienceDirect

Neuroscience Letters journal homepage: www.elsevier.com/locate/neulet

Cushing’s mechanism maintains cerebral perfusion pressure in experimental subarachnoid haemorrhage Christine M. Barry a,∗ , Corinna van den Heuvel b , Stephen Helps b , Robert Vink b a b

Discipline of Anatomy and Histology, Centre for Neuroscience, Flinders University, Bedford Park, SA 5042, Australia Discipline of Anatomy and Pathology, University of Adelaide, SA 5005, Australia

h i g h l i g h t s     

The impact of ABP responses to subarachnoid haemorrhage was assessed in two models. ABP usually increased at the ictus but less than the transient, maximal ICP increase. Failure of ABP to increase at the ictus was associated with risk of poor outcome. Increased ABP at 2 h was associated with maintained cerebral perfusion pressure. Failure of ABP to increase at 2 h was associated with poor outcome.

a r t i c l e

i n f o

Article history: Received 11 July 2012 Received in revised form 27 August 2012 Accepted 28 August 2012 Keywords: Subarachnoid haemorrhage Intracranial hypertension Intracranial pressure Cerebral perfusion pressure Cushing response Blood pressure

a b s t r a c t Mortality following subarachnoid haemorrhage (SAH) is high, especially within the first 48 h. Poor outcome is predicted by high intracranial pressure which causes diminished cerebral perfusion pressure unless a compensatory increase in mean arterial blood pressure occurs. Therefore blood pressure elevation can be protective following subarachnoid haemorrhage despite the potential for rebleeding. This study investigated blood pressure responses to SAH and the impact on cerebral perfusion pressure and outcome, as demonstrated by two experimental models. Various blood pressure responses were demonstrated, both at the ictus and within the following 5 h. Elevated MABP at the ictus and at 2 h following experimental SAH was associated with maintenance of CPP in the presence of raised ICP. Poor outcome (arrest of the cerebral circulation) was predicted by failure of MABP to increase significantly above sham levels within 2 h of SAH. Rat SAH provides relatively inexpensive models to investigate physiological mechanisms that maintain cerebral perfusion in the presence of intracranial hypertension. © 2012 Elsevier Ireland Ltd. All rights reserved.

1. Introduction Following subarachnoid haemorrhage (SAH), forty per cent of patients die within a month with most deaths occurring within 48 h of the ictus [12]. Appropriate early management is therefore crucial to outcome. Poor outcome is predicted by raised intracranial pressure (ICP) [18,19], which can reduce cerebral perfusion pressure (CPP), the pressure driving blood flow to the brain. Given that CPP is the difference between mean arterial blood pressure (MABP) and ICP, elevation of ICP that is unaccompanied by a parallel increase in MABP leads to reductions in CPP that potentially result in global cerebral ischaemia. Our previous studies demonstrated that even small volume, experimental SAH consistently produces significant

∗ Corresponding author at: Discipline of Anatomy and Histology, Flinders University, Bedford Park, SA 5042, Australia. Tel.: +61 8 8204 6637. E-mail address: christine.barry@flinders.edu.au (C.M. Barry). 0304-3940/$ – see front matter © 2012 Elsevier Ireland Ltd. All rights reserved. http://dx.doi.org/10.1016/j.neulet.2012.08.057

ICP elevation [2]. Elevation of MABP following SAH may therefore be crucial for adequate cerebral perfusion. The Cushing response is an increase in systemic arterial blood pressure triggered by elevated ICP [14]. It involves ␣-adrenergic receptor mediated peripheral vasoconstriction accompanied by a ␤-receptor mediated increase in cardiac stroke volume, which minimises loss of cardiac output in the face of the increased peripheral resistance [20]. The Cushing response is widely considered a terminal response to extremely high ICP, and is linked to other components of Cushing’s triad (widening of arterial pulse pressure), bradycardia and irregular breathing [10]. However, modest increases in ICP induce proportionate changes in systemic blood pressure to maintain cerebral blood flow in normal physiological conditions. For example, ICP increases of 10–20 mmHg evoke near parallel increases in systemic MABP in conscious dogs without major changes in heart rate or breathing [8]. It has been proposed that the term “Cushing’s mechanism” be used to distinguish this homeostatic increase in systemic blood pressure from the classic, terminal Cushing response [14].

C.M. Barry et al. / Neuroscience Letters 529 (2012) 92–96

Spontaneous blood pressure elevation has been observed following SAH in both humans [21] and in experimental models [1,9] and may therefore be considered an example of Cushing’s mechanism. The impact of Cushing’s mechanism on outcome following SAH is unclear. Some authors report that it causes a protective increase in CPP following SAH [21] and so is beneficial, whilst others conclude that it leads to a deleterious increase in intracranial blood volume and therefore ICP [5]. Uncertainty regarding the impact of Cushing’s mechanism in the acute phase of SAH on outcome probably relates to the challenges involved in non-invasive ICP/CPP measurement. Available data, from both patient monitoring and experimental studies, demonstrates wide variability in the magnitude of the blood pressure response to raised ICP and variability in the magnitude of CPP at which deleterious events such as plateau waves occur [6]. These waves of dramatically increased ICP are the result of cerebrovascular vasodilation responses to brain ischaemia when CPP is low and brain compliance reduced by high ICP [7]. Cerebral autoregulation is likely to fail as CPP approaches levels below 50 mmHg [13], but great variability has been observed. Ambiguity in management guidelines regarding optimal MABP and CPP following SAH [3] highlights the need for further studies. This study aimed to: 1. Describe blood pressure responses to raised ICP in two experimental models of SAH in rats. 2. Determine the impact of these responses on CPP and outcome. 2. Materials and methods 2.1. Experimental procedures All experiments were performed according to the guidelines established by the National Health and Medical Research Council of Australia for the use of laboratory animals in experimental research, and approved by the Institute of Medical and Veterinary Sciences and the University of Adelaide Animal Ethics Committees. 2.2. Experimental groups Adult male Sprague-Dawley rats (n = 20; 350–420 g) were group housed with free access to food and water, before being randomly allocated to receive injection SAH (n = 5), filament SAH (n = 5) or the corresponding sham procedure. Sham animals had the same surgical procedures but without induction of SAH. 2.3. Animal preparation Following induction anaesthesia (4% Isoflurane in 100% O2 ), animals were intubated and mechanically ventilated (1.8% Isoflurane in 30% O2 and 70% N2 O). Anaesthesia then changed to urethane (1 g/kg, i.v.). Ventilation parameters were adjusted to maintain normal arterial blood gases and temperature was maintained at 37.0 ± 0.5 ◦ C. The right femoral artery was cannulated for monitoring ABP and sampling blood for gas analysis. 2.4. Induction of injection SAH Autologous blood (200 ␮L) was injected into the subarachnoid space according to the method of Prunell and colleagues [15] as previously described [2]. Briefly, a 2 mm burr hole was drilled through top of the skull and under stereotactic guidance, a needle was advanced between the cerebral hemispheres until the tip lay at the base of the brain just anterior to the optic chiasma. After 30 min blood was taken from the femoral artery and injected into the

93

prechiasmatic cistern. Sham operated animals had needle insertion and saline injection. 2.5. Induction of filament SAH The endovascular filament model has been described in detail by Bederson and colleagues [4]. Briefly, the animal was placed supine, the neck opened and a surgical stump made from the external carotid artery. A 3-0 nylon filament was introduced through the stump into the internal carotid artery and advanced to perforate near the middle cerebral artery junction. The filament was then immediately withdrawn and the stump tied. Sham operated animals had all procedures other than vessel rupture. 2.6. ICP and CPP monitoring A Codman Microsensor probe (Codman and Shurtleff Inc., USA) was inserted just below the dura in left parietal cortex. ICP was monitored continuously until 5 h after SAH by a Codman ICP Express monitoring system and digitally recorded at 100 Hz using a PowerLab/LabChart data acquisition system (ADInstruments). MABP was monitored by a polyethylene catheter inserted into the right femoral artery. CPP was calculated from the formula: CPP = MABP − ICP. 2.7. Statistical analysis All data were analysed using two-way ANOVA and Bonferroni post-tests using PRISM (Graphpad Software, San Diego, CA, USA). Data are expressed as mean ± standard error of the mean. A p value less than 0.05 was considered significant. 3. Results 3.1. Baseline physiological parameters There were no differences amongst the groups with respect to physiological variables prior to SAH (Table 1). A non-significant difference in baseline CPP between injection and filament model animals reflects the sphinx vs. supine position of animals during monitoring. 3.2. Summary of results ICP data is summarised in Table 1. Shams demonstrated no ICP change throughout the monitoring period. SAH resulted in 2 (and sometimes 3) patterns of ICP increase: 1. an ictal increase with peak ICP followed by a decline to an elevated baseline (demonstrated by all SAH animals); 2. a secondary increase within 90 min (demonstrated by all SAH animals); and 3. delayed ICP waves, characteristic of plateau waves (demonstrated by 2 animals following filament SAH). Regarding systemic blood pressure, 2 patterns of response to SAH were observed at the ictus: 1. elevation of MABP of lower magnitude than ICP elevation (demonstrated by 5/5 injection SAH animals and 2/5 filament SAH animals); or 2. lack of MABP elevation or transient depression of MABP (demonstrated by 3 filament SAH animals).

94

C.M. Barry et al. / Neuroscience Letters 529 (2012) 92–96

Table 1 Physiological parameters at baseline and ICP following SAH: mean (SEM). Injection SAH Baseline ABP (mmHg) CPP (mmHg) PO2 (mmHg) PCO2 (mmHg) pH Temperature at time of SAH (◦ C) ICP (mmHg) Pre-SAH Peak Post-SAH 15 min 60 min 90 min 300 min

123.8 (6.2) 119 (4.9) 128 (8.5) 40 (1.3) 7.4 (0.01) 36.7 (0.1)

Injection sham

Filament SAH

Filament sham

116.7 (6.9) 108 (7.6) 146 (3.1) 40 (1.8) 7.5 (0.01) 36.7 (0.1)

111.8 (2.9) 104 (2.7) 141 (4.5) 40 (1.3) 7.4 (0.01) 36.7 (0.1)

7.2 (1.1) 100.4 (4.9)

6 (0.9) 49.8 (15.0)

8.8 (1.2) 92.8 (56.6)

6.6 (1.0) 7.8 (1.6)

11.6 (1.0) 14.8 (1.3) 15.6 (1.2) 15.0 (1.6)

5.0 (1.4) 6.8 (1.5) 7.0 (1.4) 4.0 (1.2)

27.2 (3.7) 30.2 (3.3) 34.2 (3.8) 39.2 (4.5)

6.6 (1.4) 6.6 (1.5) 6.6 (1.4) 7.2 (1.7)

All sham animals demonstrated no change in MABP during the monitoring period, whilst SAH animals demonstrated either: 1. elevation of MABP in proportion to ICP elevation; or 2. lack of MABP elevation. The impact of these MABP responses on CPP after SAH is described below. 3.2.1. Transient CPP decline at the ictus The profound ICP elevation demonstrated at the ictus in both models, often 90–100 mmHg above pre-SAH baseline, was accompanied by modest MABP increase of ∼15 mmHg in 5/5 injection SAH animals and 2/5 filament SAH animals (Fig. 1A and B). Therefore in both models, CPP was substantially lower than the threshold for autoregulation (∼50 mmHg) for a period at the ictus. In injection SAH (Fig. 1A), this period was brief as most of the decline from peak ICP to post-ictal baseline occurred within 1 min, minimising the potential for ischaemic brain injury. The potential for ischaemic brain injury was greater following filament SAH (Fig. 1B), as ICP remained elevated for a longer period at the ictus and was higher at all subsequent time points compared to injection SAH, consistent with the larger haemorrhage volume produced by this model [2].

104.6 (7.8) 98 (7.7) 135 (6.7) 39 (1.3) 7.4 (0.02) 36.6 (0.04)

Two filament SAH animals demonstrated no MABP change at the ictus and one demonstrated a transient depression.

3.2.2. Increased MABP at 2 h following SAH and CPP maintained MABP was significantly higher than sham levels at 2 h after injection SAH (Fig. 2A, p < 0.05). This ABP increase maintained CPP after injection SAH despite elevated ICP (Fig. 2B). Likewise, 3 of 5 filament SAH animals demonstrated increased MABP at 2 h compared to sham levels (Fig. 2C). This group demonstrated a trend towards increased MABP at this time point compared to levels at baseline and at 1 h after SAH, and CPP was not significantly diminished at 1–5 h following SAH (Fig. 2D).

3.2.3. No elevation of MABP following the ictus Two filament SAH animals failed to demonstrate elevation of MABP both during and following the ictus. The secondary ICP increase, demonstrated within 90 min after the ictus in all SAH animals (Fig. 1), contributed to critical loss of CPP these two (Fig. 2D). Disturbed cerebral autoregulation was indicated in both animals by ICP elevations characteristic of plateau waves [16] within 90 min of SAH (Fig. 3).

Fig. 1. MABP and ICP from 10 min pre-SAH to 90 min following (A) injection SAH and (B) filament SAH illustrating (1) a rapid ICP elevation during SAH and (2) a modest concurrent MABP increase. (3) ICP fell rapidly after injection SAH and more slowly after filament SAH. A secondary ICP increase was observed between 10 min and 90 min after injection SAH (***p < 0.001) and between 15 min and 90 min after filament SAH (*p < 0.05).

C.M. Barry et al. / Neuroscience Letters 529 (2012) 92–96

95

Fig. 2. (A) MABP at 1–5 h following injection SAH. MABP was increased at 2 h (*p < 0.05). (B) CPP 1–5 h following injection SAH. Note: CPP was maintained at sham levels at all time points, due to the MABP response. (C) MABP at 1–5 h following filament SAH. Three animals demonstrated an ABP response consistent with Cushing’s mechanism after filament SAH and had increased MABP compared to shams at 2 h (**p < 0.01). (D) CPP at 1–5 h after filament SAH. Cushing’s mechanism was protective against significant CPP loss compared to shams but animals that failed to demonstrate Cushing’s mechanism (n = 2) developed loss of CPP (***p < 0.001).

3.2.4. Plateau waves Two filament SAH animals demonstrated multiple plateau waves in which ICP was elevated and CPP dramatically reduced for a period (average duration 9.6 min) and then ICP declined to previous levels. CPP recovered initially from several waves in each animal, but both animals eventually developed arterial hypotension and sustained collapse of the cerebral circulation (CPP ∼ 0 mmHg). The negligible CPP in these circumstances inevitably leads to brain ischaemia and brain death. The physiological responses initiated by low CPP and leading to arrest of the cerebral circulation are demonstrated in the ICP and MABP traces illustrated in Fig. 3. The increase in MABP that immediately preceded the onset of arterial

hypotension in this trace appears to represent the commencement of a Cushing response (tachycardia and increased cardiac contractility) preceding the development of cerebral circulatory failure (with vascular baroreflexes triggering vasodilation and hypotension). 4. Discussion Both SAH models produced a profound ICP increase at the ictus (demonstrated by all animals), accompanied by a modest, transient MABP increase in 7 of 10 animals at the ictus. All animals also demonstrated a secondary ICP increase within 2 h of SAH, which,

Fig. 3. Physiological events leading to cerebral circulatory failure after filament SAH. Trace recordings of ABP (top) and ICP (below) demonstrate: (1) ABP is initially within normal range but high ICP compromises CPP; (2) ICP rises steeply, typical of plateau wave commencement; (3) a sustained period of elevated ICP; (4) transient ABP rise followed by onset of arterial hypotension; (5) ABP is near ICP level so that CPP is minimal.

96

C.M. Barry et al. / Neuroscience Letters 529 (2012) 92–96

as we have previously reported [2], may have several underlying mechanisms. A MABP increase at 2 h following SAH was demonstrated by 7 of 10 animals and was associated with improved CPP throughout the monitoring period. The 2 filament SAH animals that failed to demonstrate this secondary MABP increase both failed to demonstrate an MABP increase at the ictus (one demonstrated MABP depression) and both later exhibited multiple plateau waves followed by collapse of the cerebral circulation. A clinical challenge is presented by the need to manage MABP at increased levels following SAH so as to maintain adequate CPP [3] in a period when the risk of rebleeding is high [11]. The results of this study suggest that clinical deterioration early after the ictus could also be due to declining CPP caused by secondary ICP increase and/or plateau wave development. Management maintaining CPP above 70 mmHg has been recommended to maintain CPP safely above the threshold required for cerebral autoregulation and to avoid cerebrovascular vasodilation responses that may increase intracranial blood volume and thus further increases ICP, establishing a vicious cycle [17]. Interestingly, within 3–5 h following filament SAH, one animal exhibited periods during which CPP was