Neurocrit Care DOI 10.1007/s12028-010-9362-1
ORIGINAL ARTICLE
‘‘Optimal Cerebral Perfusion Pressure’’ in Poor Grade Patients After Subarachnoid Hemorrhage Philippe Bijlenga • Marek Czosnyka • Karol P. Budohoski Martin Soehle • John D. Pickard • Peter J. Kirkpatrick • Peter Smielewski
•
Ó Springer Science+Business Media, LLC 2010
Abstract Background Cerebrovascular pressure reactivity depends on cerebral perfusion pressure (CPP), with the optimal CPP (CPPopt) defined as pressure at which cerebrovascular reactivity is functioning optimally, reaching minimal value of pressure reactivity index (PRx). The study investigates the association between vasospasm, PRx, and CPPopt in poor grade patients (WFNS 4&5) after subarachnoid hemorrhage (SAH). Methods Data of intracranial pressure (ICP), arterial blood pressure (ABP), and flow velocities (FV) in the Middle Cerebral Artery (MCA) on transcranial Doppler from 42 SAH patients were analyzed retrospectively. PRx was calculated as a correlation coefficient between 10 s mean values of ABP and ICP calculated over a moving 3 min window. Data recorded during the first 48 h were available in 25 cases and during the first 3 days in 29 patients. Recordings obtained from day 4 to day 24 were available in 23 patients.
Results PRx at optimal CPP measured during the first 48 h showed better cerebrovascular reactivity in patients who were alive at 3 months after ictus than in those who died (PRx value -0.17 ± 0.05 vs. 0.1 ± 0.09; P < 0.01). PRx below zero at CPPopt during the first 48 h had 87.5% positive predictive value for survival. CPPopt was lower before than during vasospasm (78 ± 3 mmHg, N = 29 vs. 98 ± 4 mmHg; N = 17, P < 0.0001). The overall correlation between CPPopt and Lindegaard ratio was positive (R = 0.39; P < 0.01; N = 45). Conclusion Most WFNS 4&5 grade SAH patients with PRx below zero at optimal CPP during the first 48 h after ictus survived. Optimal CPP increases during vasospasm. Keywords Subarachnoid hemorrhage Vasospasm Cerebrovascular pressure reactivity Monitoring Treatment
Introduction P. Bijlenga M. Czosnyka K. P. Budohoski (&) M. Soehle J. D. Pickard P. J. Kirkpatrick P. Smielewski Academic Neurosurgical Unit, Department of Neurosurgery, Addenbrooke’s Hospital, University of Cambridge, Box 167, Cambridge CB2 0QQ, UK e-mail:
[email protected] P. Bijlenga Neurochirurgie, De´partement des neurosciences cliniques, Hoˆpital Universitaire de Gene`ve, Geneva, Switzerland M. Soehle Department of Anaesthesiology and Intensive Care Medicine, University of Bonn, Bonn, Germany K. P. Budohoski Department of Neurosurgery, Bielanski Hospital, Medical Research Centre, Polish Academy of Sciences, Warsaw, Poland
Subarachnoid hemorrhage (SAH) secondary to a ruptured intracranial aneurysm occurs in around 10/100,000 people each year [1]. It accounts for 5% of deaths from stroke and for 27% of potential life loss associated with stroke before the age of 65 [2]. The case fatality for aneurysmal SAH is 50% with 10–15% of patients dying before they reach the hospital. A further 20–30% of SAH patients suffer delayed ischemic neurological deficits (DINDs) as a result of cerebral vasospasm, a complication which is held directly responsible for death in 10% of SAH patients or severe morbidity in 11% of cases [3–8]. Cerebral vasospasm usually develops 4–9 days after SAH and resolves within 2 weeks [9]. It is diagnosed with angiography, transcranial Doppler ultrasonography (TCD),
Neurocrit Care
or the occurrence of otherwise unexplainable DINDs (as cerebral vasospasm is neither necessary not sufficient by itself to explain the occurrence of DINDs [8]). Cerebral angiography, a gold standard in the diagnosis of vasospasm, allows assessment of irregularities and narrowing of intracranial cerebral arteries with a 0.1% procedure-associated complication rate. Perfusion CT with CT angiography is another potentially promising screening test [10]. TCD is commonly used as second choice, and has a 99% specificity, 67% sensitivity, 97% positive predictive value, and 78% negative predictive value [11]. Vasospasm remains a sole angiography or transcranial Doppler ultrasonography finding, without any clinical sequealae in almost half of the cases studied [5, 12, 13]. In poor clinical grade SAH patients (WFNS 4&5), both coma and pharmacological sedation obscure clinical diagnosis of DINDs. As in most cerebrovascular conditions, the efficacy of any vasospasm treatment (triple-H therapy, angioplasty, etc.,) on clinical outcome highly relies on the clarity and accuracy of the diagnosis allowing for a correction and restoration of cerebral blood flow before irreversible damage occurs. CPP-oriented therapy proved to be efficient in head injury. A modification was suggested to target CPP at the level where cerebrovascular pressure reactivity works best, termed ‘‘optimal CPP’’ (CPPopt) [14]. In this study, we would like to investigate if (1) it is possible to stratify poor grade SAH patients according to initial pressure reactivity index (PRx) values, (2) optimal CPP could be established in patients after SAH and (3) vasospasm is associated with PRx and CPPopt in poor grade SAH.
Materials and Methods Patients The study is a retrospective analysis of prospectively recorded data in the environment of standard neurointensive care. All patients gave formal written consent and some data were recorded from the placebo group of the pravastatin trial in patients with aneurysmal SAH [15]. Simultaneous recordings of intracranial pressure (ICP), arterial blood pressure (ABP), and transcranial Doppler (TCD) of the middle cerebral artery (MCA) flow velocities (FV) measured at different time-points after the ictus were analyzed from 42 comatose patients admitted to the Neurosciences Critical Care Unit, Addenbrooke’s Hospital with aneurysmal subarachnoid hemorrhage WFNS grade 4 and 5. The diagnosis was confirmed with computed tomography (CT) and computed tomography angiography (CTA) or digital subtraction angiography (DSA). Acute recordings, during the first 48 h after ictus, were obtained
from 25 patients. Baseline recordings showing no vasospasm on TCD during the first 3 days following SAH were obtained from 29 patients. Data were recorded during the period of high risk of vasospasm (days 4–24) in 23 patients of which 17 had TCD-diagnosed vasospasm. Paired analysis of data between baseline and later time-points could be performed in 10 patients, 8 with vasospasm and 2 without vasospasm. Methods Intracranial pressure was recorded either via external ventricular drain or via intraparenchymal probes (Codman MicroSensors ICP Transducer, Codmann & Shurtleff, Raynham, Maryland, USA). When measurements were performed using external ventricular drain, only values recorded during periods with closed drainage lasting at least 20 min were taken into account. Arterial blood pressure was monitored from the radial artery using standard monitoring kits (Pressure Monitoring Set, Edwards Lifesiences; Irvine, Calfornia, USA) with 18–20 G arterial catheters. All ICP and blood pressure line manipulation artifacts were removed from the recordings retrospectively. Bilateral TCD of the MCA was performed using two 2 MHz probes held in place with a Lam head rack (Neurogard, Medasonics, Fremont, CA, USA or DWL, Stuttgart, Germany) focused at a depth between 4.5 and 5 cm. Vasospasm on TCD was defined as mean FV > 120 cm/s on at least one side and a Lindegaard index (mean FV in the MCA divided by mean FV in the ipsilateral extracranial internal carotid artery) above 3 on at least one side [16]. Analogue signals from ABP, ICP, and FV were digitized with an analogue-to-digital converter (Data Translation, Marlboro, MA, USA) and sampled at a frequency of 30 or 50 Hz on a PC laptop computer using in-house developed software (www.neurosurg.cam.ac.uk/icmplus) [17]. PRx was calculated every 3 min using a 10-s moving window, as a Pearson’s correlation coefficient between mean ABP and mean ICP averaged over 10 s periods. This averaging is performed to isolate the slow vascular reactions to systemic arterial pressure changes that take place with a few seconds delay. PRx values range between -1 and +1. Negative values express an active cerebrovascular reactivity and are observed when a strong cerebrovascular contraction is induced by a systemic increase of mean ABP. It results in a significant reduction of the intracranial blood volume which causes a decrease in ICP: a negative PRx indicates good cerebrovascular reactivity. In the absence of any vascular wall reactions, the arterial pressure slow waves are passively transmitted to the intracranial compartment and a perfect correlation between ABP and
Neurocrit Care
ICP is observed: a positive PRx indicates impaired cerebrovascular reactivity (Fig. 1). In order to identify the optimal vascular reactivity PRx values were calculated over periods of 20 min up to 6 h. Therefore, each PRx value was associated with the mean
A 180
CPP value recorded at the same time. The range of spontaneous CPP fluctuations was divided in 5 mmHg bins and all PRx values corresponding to each CPP bin were averaged. Mean PRx values and standard errors of the mean were plotted as a function of the corresponding CPP. The
D
-5
140
Amplitude [dB]
Pressure [mmHg]
160
0
120 100 80 60 40
-10 -15 -20 -25 -30
20 0
4
2
6
8
10
0
20
40
Time [sec]
60
80
Frequency [b/min]
E
B 200
-10
160
Amplitude [dB]
Pressure [mmHg]
180 140 120 100 80 60
-15
-20
-25
40 -30
20 0
50
100
150
200
250
300
-1
0
Time [sec]
2
3
4
5
6
Frequency [b/min]
C 140
F 36
120
34
ICP [mmHg]
Averaged Pressure[mmHg]
1
100 80 60 40
R = 0.87 = PRx
32 30 28 26
20 0
50
100
150
200
250
300
Time [sec]
Fig. 1 Pressure reactivity index measurement. Left side a Ten second simultaneous recording at 30 Hz of arterial blood pressure (dotted line) and intracranial pressure (continuous line) showing pulse waves. b Three hundred seconds simultaneous recording of arterial blood pressure and intracranial pressure at 30 Hz. c Filtering of the above recording by performing a moving average over 10 s revealing the slow variations of blood pressure, that are, in this particular patient with lost cerebral autoregulation, almost directly transmitted to the intracranial pressure. d Fourier transform of intracranial pressure recording from the data set in A showing the cardiac pulse varying around 60 beats per minute, waves at precisely 11, 22 and 33 bpm corresponding to mechanical ventilation induced waves and harmonics. Below 5 bpm waves are observed corresponding to slow waves.
24 120
122
124
126
128
130
132
Arterial Blood Pressure [mmHg]
e To keep a constant flow through the brain, the intracranial vessel bed adapts by, for instance, contracting to increase resistance in case of systemic pressure rise. In turn, this leads to a reduction in intracranial blood volume and intracranial pressure reducing, thus decreasing the impact of the systemic blood pressure rise on the intracranial pressure. When autoregulation is lost, systemic pressure changes are transmitted to the intracranial pressure. f Scatter graph of the ICP averaged over 10 s versus the concomitant arterial blood pressure averaged over 10 s. The PRx is calculated as the Pearson’s correlation coefficient of a regression line passing through the cloud of points calculated over 5 min. A PRx of 0.87, in this example, is an indicator of poor autoregulation. PRx pressure reactivity index, ICP intracranial pressure
Neurocrit Care
optimal vascular pressure reactivity (PRxopt) was defined as the lowest PRx observed within a range of CPP (usually 50–90 mmHg). The optimal CPP (CPPopt) was defined as the CPP at PRxopt. All results are expressed as mean and standard errors of the mean (SEM). Mean values obtained from different groups of patients were tested using the non null hypothesis of an unpaired 2-tailed Student t-test. A paired 2-tailed Student t-test was used when comparing mean values observed from the same group of patients at two different times. Probability values less than 0.05 were considered statistically significant.
Results Severity of Initial Insult can be Assessed by Using Early Optimal PRx Measurements In 25 patients, the PRxopt and CPPopt were measured using recordings performed during the first 48 h after the ictus. The mean PRxopt observed in the group of patients who survived the first 3 months was significantly lower than the mean PRxopt value measured in the group of patients who died (-0.17 ± 0.054 N = 16 vs. 0.10 ± 0.085 N = 9; P < 0.01) (Fig. 2). Modified Rankin score measured at 3 months after ictus could be obtained from the clinical files in 16 cases. PRxopt measured during the first 48 h correlated positively with the Modified Rankin score observed at 3 months (r2 = 0.35, P = 0.015, N = 16), suggesting that worse Modified Rankin score associated with poor vascular reactivity. Values of PRx below zero measured at CPPopt had a positive predictive value for survival of 87.5% (77% sensitivity and 71% specificity). The negative predictive value (death when PRx is greater than 0) was 55% (N = 25; P = 0.06, Fisher exact test).
Group Comparison: Spasm and No Spasm By comparing the baseline ‘‘optimal curves’’ (PRx vs. CPP) with recordings performed during vasospasm, we observed a shift of the U-shaped curve toward the grater CPP values. PRxopt and CPPopt were measured in 29 patients during the baseline period and compared either with later recordings performed in patients that never reached criteria for vasospasm on TCD (N = 6) or with recordings obtained from patients when vasospasm criteria were reached (N = 17). If multiple recordings were performed sequentially in the same patient, the data obtained on the worst day when the highest velocities and Lindegaard indexes were recorded were used for the analysis. PRxopt was not significantly affected by vasospasm (baseline -0.04 ± 0.05; vasospasm 0.10 ± 0.06; P = 0.08). In contrast, CPPopt shifted significantly from 78 ± 2.6 mmHg during baseline to 98 ± 3.6 mmHg when vasospasm was observed on TCD (P < 0.001). In the group of patients in which vasospasm criteria were never reached CPPopt showed a trend toward higher values (84.6 ± 5.5 mmHg; P = 0.28) with no changes in optimal PRxopt (Fig. 3a, b). The correlation of CPPopt and vasospasm intensity was assessed by checking the correlation beween CPPopt and mean Lindegaard index measured on the left and right side. 45 concomitant measurements of the Lindegaard index and CPPopt were performed and plotted on a scatter graph (Fig. 4). The linear regression line increased from 73.5 ± 8 mmHg at a Lindegaard index of 1 to 92.5 ± 8 mmHg at a Lindegaard index of 6. The correlation coefficient was 0.39 (r2 = 0.15; P < 0.01; N = 45). Effect of Vasospasm Evolving in Time Complete data sets of recordings measured during both baseline and vasospasm periods were obtained from 10 patients. 8 patients fulfilled criteria for vasospasm and 2
0.6
140
p < 0.01
p = 0.11
CPPopt [mmHg]
0.4
PRxopt
0.2 0.0 -0.2 -0.4
120 100 80 60
-0.6 40
Survived
Dead
Fig. 2 A significant difference of the mean optimal PRx is observed between the group of patients that survived 3 months and the group of patients who died within the first 3 months after ictus. The box represents the range from the 25th to 75th percentile. The line
Survived
Dead
dividing the box represents the median. The whiskers show the 5th– 95th percentiles. The square shows the mean and the stars extreme values. PRx pressure reactivity index, PRxopt optimal PRx
Neurocrit Care
A
B
1.0
CPPopt [mmHg]
120
PRxopt
0.6 0.2 -0.2
100 80 60 40
-0.6
Baseline
No vasospasm Vasospasm
Fig. 3 PRxopt and CPPopt measured before vasospasm, during the first 3 days after the ictus (N = 29) and during vasospasm. For each patient the values of PRxopt and CPPopt were obtained from the data sets recorded on the day when the highest Lindegaard index was measured. Patients were classified as suffering from vasospasm if
Baseline
No vasospasm Vasospasm
reaching at on least 1 day TCD criteria for vasospasm (N = 17) or as having never suffered from vasospasm (N = 6). CPPopt increased significantly in the vasospasm group. PRxopt optimal PRx, CPPopt optimal cerebral prefusion pressure
observed. In contrast, CPPopt increased in both cases. (Fig. 5a, b).
120
100
CPPopt [mmHg]
Discussion 80
60
40
20
0 1
2
3
4
5
6
7
8
Lindegaard index
Fig. 4 CPPopt versus concomitant Lindegaard index recorded in each patient on the day of highest Lindegaard index. Regression line with 95% confidence limit curves and 95% prediction limits lines. CPPopt optimal cerebral prefusion pressure
never reached the criteria although an increase in the MCA mean FV was observed. In this subset of patients, paired analysis of both CPPopt and PRxopt showed an increase. For the 8 patients that had vasospasm on TCD, the mean PRxopt was 0.17 ± 0.06 and the mean CPPopt was 73.2 ± 3.4 mmHg during baseline measurements. PRx increased to 0.21 ± 0.09 (P < 0.003) and CPPopt to 99 ± 3.6 mmHg (P < 0.002). The mean increase in PRx (dPRxopt) calculated as the difference between PRxopt measured during vasospasm and PRxopt measured at baseline was 0.38 ± 0.085. The median PRxopt increase was 0.41. The mean increase in CPPopt (dCPPopt) was of 26 ± 5.5 mmHg. The median CPPopt increase was 25.5 mmHg. For the 2 patients that did not fulfill criteria for vasospasm on TCD, PRxopt value slightly increased in one case and decreased in the other between baseline and recordings measured when the highest velocities were
Rapid and accurate identification of delayed ischemic neurologic deficits induced by cerebral vasospasm after subarachnoid hemorrhage is difficult especially in poor grade patients when the usefulness of clinical evaluation is decreased due to the overall poor neurological status and the frequent use of pharmacological sedation. Alterations of cerebral blood flow during vasospasm are primarily due to the impairment of cerebrovascular autoregulation with the autoregulatory curve shifting toward the right [3, 18–22]. Additionally, cerebrovascular chemoregulation in those patients may be disturbed [23]. These observations were the basis for introducing vasopressor to prevent DINDs in patients developing vasospasm [24–26]. The source of autoregulation failure and the pathophysiology of vasospasm are still poorly understood. It is most probably the result of the loss of both chemoregulation [27] at the endothelial level and disruption of the neuronal control of conductive vessels [28], however, the inflammatory response and brain edema should play a part as well. Cerebral autoregulation can be assessed using external stimuli such as the cuff deflation test (CDT) [29] (a reaction of CBF or blood flow velocities measured with TCD to a modification of the arterial blood pressure) or the transient hyperemic response test (THRT) [19, 30]. A less invasive method, suitable for monitoring over longer periods, evaluates the reaction of systolic and mean flow velocities in the middle cerebral artery to slow waves of ABP (Sx and Mx, respectively), has been introduced and shown to be an indicator of autoregulation and a predictor of outcome in patients suffering head injury [31]. Furthermore, a preliminary study showed that changes in Mx
Neurocrit Care
B
1.0
CPP opt [mmHg]
A
PRx
0.6 0.2 -0.2 -0.6
120 100 80 60 40
Baseline
No vasospasm Vasospasm
Baseline
No vasospasm Vasospasm
Fig. 5 Difference between paired PRxopt and CPPopt measured before the initiation of vasospasm during the first 3 days after the ictus and during the vasospasm period in the same patients (N = 10). Two patients never reached the TCD criteria for vasospasm (stars). Eight patients suffered vasospasm and both PRx (a P < 0.003) and
CPPopt (b P < 0.002) increased significantly from baseline to vasospasm. PRx pressure reactivity index, PRxopt optimal, dPRxopt difference in PRxopt before and during vasospasm, CPPopt optimal cerebral prefusion pressure
and Sx were associated with the incidence of cerebral vasospasm [32]. Cerebrovascular autoregulation may also be evaluated using the pressure reactivity index (PRx) calculated as the correlation coefficient between ICP and ABP recordings filtered to remove both the heart and respiratory components of the signals. When the cerebrovascular bed is normally reactive, any change in ABP produces an inverse change in cerebral blood volume and hence ICP. When reactivity is disturbed, changes in ABP are passively transmitted to ICP. A negative value of PRx reflects a normally reactive vascular bed, whereas a positive PRx means a passive behavior corresponding to a non-reactive vascular bed. Abnormal values of PRx have been observed during temporary loss of autoregulatory reserve during ICP-plateau waves and permanent autoregulatory failure during refractory intracranial hypertension [33]. High values of PRx have been demonstrated to be predictive of a poor outcome following head injury. When PRx is plotted against CPP it shows a ‘‘U-shape’’ curve [14, 34] indicating, in the majority of cases, a patient specific value of CPP at which pressure reactivity is optimal (CPPopt). Again, the absolute value of the difference between the CPP measured most of the time in a patient and the actual CPPopt has been demonstrated to be a prognostic factor of worse outcome [14]. The approach to monitor PRx in SAH patients is justified by the disruption of cerebral autoregulation accompanying the initial insult and later during cerebral vasospasm. This method has a limitation, however, that it is only feasible in patients who have invasive intraparenchymal ICP monitoring or external ventricular drainage, thus being confined only to poor grade SAH patients. Monitoring PRx may be used in those patients to titrate CPP, especially during triple-H therapy, when clinical signs are too poor, or damped by pharmacological sedation, to be assessed precisely. Interestingly, PRx values equal or below zero during the first 48 h after SAH correlated with survival at 3 months. The positive predictive value for this was 87.5%. Although
the study comprised only a small group of patients (N = 25—cohort with available recordings from the initial 48 h after the ictus), the observed correlation raises a question, whether assessment of initial PRx could be used to aid in stratification of the initial insult severity and prognosis in patients with poor grade SAH. Moreover, during vasospasm higher values of optimal CPP were observed. This may partially explain efficiency of triple-H therapy. Although matching CPP to individual needs seems to be conceptually acceptable, any final conclusions regarding the use of optimal CPP to guide diagnosis and therapy of vasospasm after SAH cannot be drawn on the basis of this observational, retrospective study. However, this concept should be evaluated in a prospective, properly randomized trial.
Conclusion Initial optimal cerebrovascular reactivity measurements allow a refined assessment of the initial insult severity as survival at 3 months after ictus was correlated with good initial PRx. Pressure reactivity index also increased during periods of vasospasm indicating a loss of autoregulation during that time. In patients with poor grade SAH, optimal CPP shifts toward higher values during vasospasm. Acknowledgments The authors are in debt to all the team participating in data collection : Mrs. Pippa Al-Rawi, Dr. Ming-Yuan Tseng, Mrs. Dott Chatfield, Mrs. Joanne Outtrim, Mrs. Anne Manktelow, Mrs. Helen Seeley, Mrs. Carole Turner, Dr. Marcella Balestreri, Dr. Magda Hiler, Dr. Luzius Steiner, Dr. Eric Schmidt, Dr. Stefan Piechnik, Dr. Andreas Raabe, Mr. Eric Guazzo, Prof. David Menon, Prof. Arun Gupta, Dr. Basil Matta, Mr. Peter Kirkpatrick, Mr. Ivan Timofeev, Mr. Pwawanjit Minhas, and all nursing and research staff of NCCU and Wolfson Brain Imaging Centre. The project was supported by National institute of Health Research Biomedical Research Centre, Cambridge University Hospital Foundation Trust—Neurosciences Theme and Senior Investigaor Award (JDP). PB was supported by ‘‘Fond de perfectionnement’’ of the Geneva University Hospital, The ‘‘Fond Ernst and Lucie Schmidheiny’’ and the ‘‘Socie´te´ Acade´mique de Gene`ve’’.
Neurocrit Care Disclosure ICM+ software (www.neurosurg.cam.ac.uk/icmplus) is licensed by University of Cambridge and PS and MC have financial interest in a fraction of licensing fee.
18.
19.
References 1. Linn FH, Rinkel GJ, Algra A, van Gijn J. Incidence of subarachnoid hemorrhage: role of region, year, and rate of computed tomography: a meta-analysis. Stroke. 1996;27:625–9. 2. Johnston SC, Selvin S, Gress DR. The burden, trends, and demographics of mortality from subarachnoid hemorrhage. Neurology. 1998;50:1413–8. 3. Heros RC, Zervas NT, Varsos V. Cerebral vasospasm after subarachnoid hemorrhage: an update. Ann Neurol. 1983;14:599–608. 4. Kassell NF, Torner JC, Haley EC Jr, Jane JA, Adams HP, Kongable GL. The international cooperative study on the timing of aneurysm surgery. Part 1: overall management results. J Neurosurg. 1990;73:18–36. 5. Pluta RM. Delayed cerebral vasospasm and nitric oxide: review, new hypothesis, and proposed treatment. Pharmacol Ther. 2005; 105:23–56. 6. Schievink WI. Intracranial aneurysms. N Engl J Med. 1997;336: 28–40. 7. Seiler R, Newell D. Subarachnoid hemorrhage and vasospasm. In: Newell D, Aaslid R, editors. Transcranial Doppler. New York: Raven Press; 1992. p. 101–7. 8. van Gijn J, Kerr SR, Rinkel GJ. Subarachnoid haemorrhage. Lancet. 2007;396:306–18. 9. Weir B, Grace M, Hansen J, Rothberg C. Time course of vasospasm in man. J Neurosurg. 1978;48:173–8. 10. Wintermark M, Ko NU, Smith WS, Liu S, Higashida RT, Dillon WP. Vasospasm after subarachnoid hemorrhage: utility of perfusion CT and CT angiography on diagnosis and management. AJNR Am J Neuroradiol. 2006;27:26–34. 11. Lysakowski C, Walder B, Costanza MC, Tramer MR. Transcranial Doppler versus angiography in patients with vasospasm due to a ruptured cerebral aneurysm: a systematic review. Stroke. 2001;32:2292–8. 12. Aaslid R, Huber P, Nornes H. Evaluation of cerebrovascular spasm with transcranial Doppler ultrasound. J Neurosurg. 1984; 60:37–41. 13. Vora YY, Suarez-Almazor M, Steinke DE, Martin ML, Findlay JM. Role of transcranial Doppler monitoring in the diagnosis of cerebral vasospasm after subarachnoid hemorrhage. Neurosurgery. 1999;44:1237–47. 14. Steiner LA, Czosnyka M, Piechnik SK, Smielewski P, Chatfield D, Menon DK, Pickard JD. Continuous monitoring of cerebrovascular pressure reactivity allows determination of optimal cerebral perfusion pressure in patients with traumatic brain injury. Crit Care Med. 2002;30:733–8. 15. Tseng MY, Czosnyka M, Richards H, Pickard JD, Kirkpatrick PJ. Effects of acute treatment with pravastatin on cerebral vasospasm, autoregulation, and delayed ischemic deficits after aneurysmal subarachnoid hemorrhage: a phase II randomized placebo-controlled trial. Stroke. 2005;36:1627–32. 16. Lindegaard KF. The role of transcranial Doppler in the management of patients with subarachnoid haemorrhage—a review. Acta Neurochir Suppl. 1999;72:59–71. 17. Smielewski P, Czosnyka M, Steiner LA, Belestri M, Piechnik SK, Pickard JD. ICM+: software for on-line analysis of bedside
20.
21.
22.
23.
24.
25.
26.
27.
28.
29. 30.
31.
32.
33.
34.
monitoring data after severe head trauma. Acta Neurochir Suppl. 2005;95:43–9. Krasznai L, Grote EH. Acute vasoparalysis after subarachnoid haemorrhage and cerebral trauma: general reflex phenomenon? Neurol Res. 1994;16:40–4. Lam JM, Smielewski P, Czosnyka M, Pickard JD, Kirkpatrick PJ. Predicting delayed ischemic deficits after aneurysmal subarachnoid hemorrhage using a transient hyperemic response test of cerebral autoregulation. Neurosurgery. 2000;47:819–25. Tran Dinh YR, Lot G, Benrabah R, Baroudy O, Cophignon J, Seylaz J. Abnormal cerebral vasodilation in aneurysmal subarachnoid hemorrhage: use of serial 133Xe cerebral blood flow measurement plus acetazolamide to assess cerebral vasospasm. J Neurosurg. 1993;79:490–3. Voldby B, Enevoldsen EM. Intracranial pressure changes following aneurysm rupture. Part 1: clinical and angiographic correlations. J Neurosurg. 1982;56:186–96. Lang EW, Diehl RR, Mehdorn HM. Cerebral autoregulation testing after aneurysmal subarachnoid hemorrhage: the phase relationship between arterial blood pressure and cerebral blood flow velocity. Crit Care Med. 2001;29:158–63. Jakubowski J, Bell BA, Symon L, Zawirski MB, Francis DM. A primate model of subarachnoid hemorrhage: change in regional cerebral blood flow, autoregulation carbon dioxide reactivity, and central conduction time. Stroke. 1982;13:601–11. Denny -Brown D. The treatment of recurrent cerebrovascular symptoms and the question of ‘‘vasospasm’’. Med Clin North Am. 1951;35:1457–74. Kassell NF, Peerless SJ, Durward QJ, Beck DW, Drake CG, Adams HP. Treatment of ischemic deficits from vasospasm with intravascular volume expansion and induced arterial hypertension. Neurosurgery. 1982;11:337–43. Pickard JD, Boisvert DP, Graham DI, Fitch W. Late effects of subarachnoid haemorrhage on the response of the primate cerebral circulation to drug-induced changes in arterial blood pressure. J Neurol Neurosurg Psychiatry. 1979;42:899–903. Iuliano BA, Pluta RM, Jung C, Oldfield EH. Endothelial dysfunction in a primate model of cerebral vasospasm. J Neurosurg. 2004;100:287–94. Pluta RM, Thompson BG, Dawson TM, Snyder SH, Boock RJ, Oldfield EH. Loss of nitric oxide synthase immunoreactivity in cerebral vasospasm. J Neurosurg. 1996;84:648–54. Aaslid R, Lindegaard KF, Sorteberg W, Nornes H. Cerebral autoregulation dynamics in humans. Stroke. 1989;20:45–52. Giller CA. A bedside test for cerebral autoregulation using transcranial Doppler ultrasound. Acta Neurochir (Wien). 1991; 108:7–14. Czosnyka M, Smielewski P, Kirkpatrick P, Menon DK, Pickard JD. Monitoring of cerebral autoregulation in head-injured patients. Stroke. 1996;27:1829–34. Soehle M, Czosnyka M, Pickard JD, Kirkpatrick PJ. Continuous assessment of cerebral autoregulation in subarachnoid hemorrhage. Anesth Analg. 2004;98:1133–9. table. Balestreri M, Czosnyka M, Steiner LA, Schmidt E, Smielewski P, Matta B, Pickard JD. Intracranial hypertension: what additional information can be derived from ICP waveform after head injury? Acta Neurochir (Wien). 2004;146:131–41. Steiner LA, Coles JP, Johnston AJ, Chatfield DA, Smielewski P, Fryer TD, Aigbirhio FI, Clark JC, Pickard JD, Menon DK, Czosnyka M. Assessment of cerebrovascular autoregulation in head-injured patients: a validation study. Stroke. 2003;34:2404– 9. 34.
