- Email: [email protected]
Autoregulation and brain metabolism in the perihematomal region of spontaneous intracerebral hemorrhage: An observational pilot study
Journal of the Neurological Sciences 295 (2010) 16–22
Contents lists available at ScienceDirect
Journal of the Neurological Sciences j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j n s
Autoregulation and brain metabolism in the perihematomal region of spontaneous intracerebral hemorrhage: An observational pilot study Jennifer Diedler a,⁎, Georg Karpel-Massler b,c, Marek Sykora a,d, Sven Poli a, Oliver W. Sakowitz b, Roland Veltkamp a, Thorsten Steiner a a
Department of Neurology, University of Heidelberg, Im Neuenheimer Feld 400, 69120 Heidelberg, Germany Department of Neurosurgery, University of Heidelberg, Im Neuenheimer Feld 400, 69120 Heidelberg, Germany Department of Neurosurgery, University of Ulm, Steinhövelstr.9, 89075 Ulm, Germany d Department of Neurology, Comenius University, Mickiewiczova 13, 813 69 Bratislava, Slovakia b c
a r t i c l e
i n f o
Article history: Received 23 November 2009 Received in revised form 20 April 2010 Accepted 27 May 2010 Available online 16 June 2010 Keywords: Intracerebral hemorrhage Autoregulation PRx ORx Microdialysis Tissue oxygenation
a b s t r a c t The metabolic and hemodynamic processes in the edema surrounding spontaneous, supratentorial intracerebral hemorrhage (ICH) are poorly understood. Specifically, the local metabolic effects of autoregulatory failure have not been described previously. In the current observational pilot study, microdialysis and brain tissue oxygenation probes (PbrO2) were placed in the perihemorrhagic edema using neuronavigation in five non-surgically treated patients with deep ICH. The cerebrovascular pressure reactivity index (PRx, moving correlation between mean arterial and intracranial pressure) and PbrO2 reactivity index (ORx, correlation between PbrO2 and cerebral perfusion pressure) were used to characterize cerebral autoregulation. While all five patients had ORx values indicative for severely disturbed autoregulation, assessment of PRx only in one patient was consistent with sustained failure of cerebrovascular reactivity. This patient at the same time had the worst metabolic parameters and the poorest tissue oxygenation. We conclude that multimodality monitoring in the perihemorrhagic penumbra is feasible. A study in a larger population is needed to clarify the relationship between PRx and ORx in ICH patients, the local metabolic effects of autoregulatory failure and its relation to brain edema formation and clinical outcome. © 2010 Elsevier B.V. All rights reserved.
1. Introduction The metabolic and hemodynamic processes in the perihematomal region of spontaneous intracerebral hemorrhages (ICH) remain controversial. While several studies provided evidence for a reduced perihematomal blood flow [1,2], recent imaging studies have dispelled the idea of major ischemia in the edematous zone adjacent to the hematoma [3–5]. Instead, the concept of a perihemorrhagic metabolic penumbra has been suggested, analogously to patients suffering from traumatic ICH [6,7]. Additionally, the state and significance of autoregulation of cerebral blood flow after ICH remains a matter of debate. Earlier studies did not find evidence for a disturbance of autoregulation in acute ICH. However, it has recently been shown that autoregulation may be impaired in ICH patients, possibly influencing outcome [8,9]. While the exact pathophysiological mechanisms remain undetermined, it has been suggested that disturbed autoregulation may promote additional brain injury, especially in the perihemorrhagic region. To further assess this subject, we undertook a pilot study
⁎ Corresponding author. Tel.: + 49 62215637557; fax: +49 622156467. E-mail address: [email protected] (J. Diedler). 0022-510X/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jns.2010.05.027
including patients suffering from spontaneous supratentorial ICH who did not undergo hematoma evacuation. We performed multimodality monitoring directly measuring brain metabolites and brain tissue oxygenation (PbrO2) in the edema surrounding the hematoma. Cerebral autoregulation was assessed using the pressure reactivity index (PRx) and the oxygen pressure reactivity index (ORx) as described previously [10,11]. In this hypothesis generating pilot study, we aimed to evaluate feasibility of using PbrO2 and microdialysis to assess autoregulation and local metabolism in the perihemorrhagic region. 2. Methods 2.1. Patients During April 2007 and August 2008, five prospectively collected patients suffering from spontaneous, deep supratentorial ICH underwent invasive multimodality neuromonitoring including measurement of PbrO2, cerebral microdialysis, and ICP and CPP monitoring. Criteria for invasive neuromonitoring and hence inclusion into the current study were 1) intraventricular hemorrhage extension with need for placement of an extraventricular drain (and thereby requiring intubation and transfer to the operating room) and 2) conservative therapy without
J. Diedler et al. / Journal of the Neurological Sciences 295 (2010) 16–22
evacuation of the hematoma as decided by the attending physicians because of deep hemorrhage location. The study was approved by the local ethics committee. Written, informed consent was obtained from the legal representatives of the patients. All patients were intubated and mechanically ventilated during the entire monitoring period. Intravenous midazolam and sufentanil were used for analgosedation. Therapy was aimed to keep ICP b20 mm Hg and CPP N60 mm Hg, according to current guidelines [12,13]. Arterial blood gas samples were obtained every 2 h to adjust ventilation parameters. PaCO2 was maintained between 35 and 45 mm Hg. Neurological deficit on admission was assessed by the National Institutes of Health Stroke Scale Score (NIHSSS). Hematoma volume was calculated from the first CT scan using the a × b × c × 0.5 method, as described by Broderick et al. [14]. Intraventricular bleeding was scored according to the Graeb score [15].
17
example during in-house transportations). Artifacts mostly resulted from inadequate pressure signals, for example during withdrawal of blood gas probes or due to nursing interventions. After re-sampling in order to obtain one value every 6 s, PRx was calculated every 60 s as a moving linear (Pearson's) correlation between 40 consecutive values of MAP and ICP, as described by Steiner and Czosnyka [10,16]. CPP and PbrO2 values were averaged in the same moving window in order to obtain a time synchronous matrix containing PRx, CPP and PbrO2. For calculation of PRx, only data points fulfilling the criteria of systolic arterial pressure between 60–180 mm Hg, mean arterial pressure between 50 and 120 mm Hg and ICP N0 mm Hg were included in the analysis. Fisher-Z-transformation was used to calculate mean PRx values. All calculations were performed using Matlab (MathWorks, Version 7.5). 2.5. ORx
2.2. Neuromonitoring Intracranial pressure was measured using a combined ventricular drainage and piezoelectric ICP probe allowing continuous assessment of ICP irrespective of drainage (Raumedic NEUROVENT, Germany). PbrO2 was measured using an intraparenchymal probe (LICOX PMO, Integra, Germany). Microdialysis catheters (CMA 70, CMA Microdialysis, Sweden; membrane length 10 mm) were perfused with artificial cerebrospinal fluid (CMA Microdialysis, Sweden) at a rate of 0.3 μl/ min. The perfusates were collected in microvials and analyzed in 1 to 2 h intervals immediately after collection. Samples were analyzed for glucose, pyruvate, lactate, glutamate and glycerol using the ISCUS Clinicial Microdialysis Analyzer (CMA Microdialysis, Sweden). In order to ensure exact placement of the probes in the perihematomal region, an intraoperative, image-guided, frameless, localization system (BrainLAB VectorVision® compact Navigation System, BrainLAB, Feldkirchen, Germany) was used in 4 of 5 patients. Placement of the probes was guided by a freehand pointer determining the correct direction and depth of insertion. Blood pressure was measured from the radial artery (Dräger, Siemens, Germany). No complications or adverse events caused by the monitoring probes were observed. ICP, CPP, PbrO2, systolic and diastolic blood pressure, MAP, and heart rate were synchronously recorded at a sampling frequency of 1 Hz in 4 patients. Due to technical difficulties, one patient (#2) was recorded with a sampling frequency of 1/min (∼0.02 Hz). The data were stored on a bedside computer using the ICU pilot software (CMA, Sweden). 2.3. Signal analysis To estimate cerebral autoregulation and its local effects on tissue oxygenation, the indices PRx and ORx were calculated. Additionally, the coherence between the moving averages of CPP, PRx and PbrO2 were calculated as described below. 2.4. PRx The PRx investigates the correlation between slow wave changes in MAP and ICP. It may be used as a continuous surrogate measure of cerebral autoregulation. A positive PRx implies a positive association between the slow components of MAP and ICP, indicating a passive, nonreactive behavior of the cerebral vessels. A negative value reflects a normally reactive vascular bed where changes in MAP result in inversely correlated changes in ICP within a 5 to 30-second time window. PRx was calculated as described previously [8,10]. Before analysis, data files were screened and epochs containing incomplete datasets or artifacts were visually identified and excluded from the analysis. Incomplete datasets were caused by disturbed interaction between the monitoring system and the recording software or due to complete disconnection the patient from the monitoring system (for
Analogously to PRx, ORx values range between + 1 and −1. A positive correlation suggests that PbrO2 passively follows CPP, indicative for disturbed autoregulation [11,17]. The oxygen reactivity index ORx was calculated after artifact elimination and re-sampling to obtain one value every 30 s as moving linear correlation coefficient between CPP and PbrO2 over an average period of 1 h as described by Jaeger et al. [11]. For patient #2 with a sampling frequency of 1/min, ORx was calculated based on minutely values. 2.6. Coherence To assess the relation between local tissue oxygenation and CPP or PRx respectively, the coherence between PbrO2 vs. CPP and PbrO2 vs. PRx was analyzed. The window length of the 50% overlapping hamming window was chosen to obtain a frequency resolution of 0.23 cycles/h and the mean coherence in the frequency range of 0.23 to 4.2 cycles/h was calculated. Calculation of coherence was based on the low passed filtered curves of PRx and the moving averages of PbrO2 and CPP (first-order filter of 0.01 Hz). 3. Results Five non-consecutive patients were included in the study. Patient characteristics are shown in Table 1. The mean age was 61.2 years (range 40–75). The median hematoma size was 18.4 ml (range 14.1– 67). The mean time from onset of symptoms to start of monitoring was 19.6 h (range 11–32). According to our inclusion criteria, all patients had intraventricular hemorrhage extension and required ventriculostomy. Probe placement for each individual patient is illustrated in Fig. 1. One patient (#3) died during ICU stay because care was withdrawn at the request of the family. The monitoring probes were removed in this patient at the time of withdrawal of care. Fig. 2 shows the fluctuations of PbrO2, CPP and PRx for each patient individually. Each data point represents a minutely moving average summarizing 4 min of monitoring time. Phases of disturbed vasoreactivity defined as PRx N0.2 are indicated in gray. Table 2 gives an overview of the different autoregulatory indices. The data of each patient will be presented individually before summarizing the findings and comparing 24-hour averages. 3.1. Patient #1 Patient #1 (Fig. 2A) had a mean PRx of 0.14, generally indicating intact cerebrovascular reactivity. Except for two short periods (gray areas), PRx remained below the threshold of 0.2. The mean PbrO2 was 23 ± 7 mm Hg, with a mean arterial blood oxygen content (PaO2) of 119 ± 34 mm Hg. CPP was 86 ± 7 mm Hg in the mean. Of interest, despite a PRx indicating globally intact vasoreactivity, ORx was 0.73, indicative for severe autoregulatory failure. Fig. 2A shows the
18
J. Diedler et al. / Journal of the Neurological Sciences 295 (2010) 16–22
Table 1 Patient characteristics. Patient #
Hemorrhage volume [ml]
Graeb score
Baseline NIHSS
Discharge NIHSS
Age
Etiology
Time from ictus to monitoring [h]
1 2 3 4 5
18.4 15.2 67.3 31.9 14.1
5 7 11 11 6
34 34 34 34 10
17 13 42b 15 20
59 63 69 40 75
Hypertensive Coagulopathy Coagulopathy Hypertensive Coagulopathy
18 32 11 26a 11
a b
Microdialysis probe had to be replaced and worked only 55 h after onset of symptoms. Care withdrawn at request of the family on day 6 post-ictus.
synchronously fluctuating curves of CPP and PbrO2. This relation is also mirrored by the high coherence between both parameters (0.86). In contrast, coherence between PbrO2 and PRx was 0.28, indicating that slow fluctuations (frequency range of 0.23 to 4.2 cycles/h) of local tissue oxygenation were not associated with changes of PRx in this patient. 3.2. Patient #2 For patient # 2 (Fig. 2B), the mean PbrO2 was 31 ± 7 mm Hg, PaO2 was 118 ± 10 mm Hg and CPP 80 ± 10 mm Hg, respectively. The coherence between CPP and PbrO2 was 0.84, ORx was 0.77, suggestive for failure of autoregulation. 3.3. Patient #3 Patient #3 (Fig. 2C) had an overall PRx of −0.11, indicative for intact global cerebrovascular reactivity. However, this patient shows
clearly discernible periods of intact versus disturbed cerebrovascular reactivity (gray areas). Of interest, phases exceeding a PRx N0.2 were associated with simultaneous drops of PbrO2 and CPP. This observation is not mirrored by analysis of coherence between PRx and PbrO2 (0.19). However, time resolution of the coherence function was not suited to detect very slow processes. Mean PbrO2, CPP and PaO2 levels were 27 ± 9 mm Hg, 69 ± 10 mm Hg and 117 ± 27 mm Hg, respectively. The coherence between CPP and PbrO2 was 0.83 and ORx was 0.60, suggesting disturbed autoregulation.
3.3.1. Patient #4 A similar pattern of clearly discernable phases of disturbed vasoreactivity was observed in patient #4 (Fig. 2D). However, unlike in patient #3, these periods were not associated with clear drops in PbrO2. Mean PbrO2 was 30 ± 16 mm Hg, mean CPP 80 ± 8 mm Hg, and mean PaO2 99 ± 13 mm Hg respectively. ORx was 0.37. Coherence between PbrO2 and CPP was 0.82, and 0.14 between PbrO2 and PRx.
Fig. 1. Probe placement in the perihematomal edema in the 5 included patients (MD: microdialysis catheter, O2: oxygen probe).
J. Diedler et al. / Journal of the Neurological Sciences 295 (2010) 16–22
19
Fig. 2. Time course of PbrO2, CPP and corresponding PRx values plotted for each patient separately (A, B, C, D and E). Each data point represents a minutely moving average summarizing 4 min of monitoring time following artifact elimination. Gray areas indicate episodes of disturbed autoregulation (PRx N 0.2). Original data was low-pass filtered for better visualisation.
3.4. Patient #5 Patient #5 had severely disturbed cerebrovascular reactivity throughout the entire monitoring period (mean PRx 0.52). At the same time, this patient had the lowest PbrO2 values during the entire monitoring period (11 ± 7 mm Hg) despite having normal PaO2 levels (100 ± 13 mm Hg). Mean CPP was 72 ± 10 mm Hg. Coherence between PbrO2 and CPP was 0.67, ORx was 0.41, indicating failure of autoregulation. Coherence between Prx and PbrO2 was 0.45.
Fig. 3 shows the relationship between microdialysis parameters and local tissue oxygenation versus PRx, and ORx, respectively. All parameters were averaged for the first three days of monitoring for each patient individually. Comparing daily averages aims to compare absolute, steady-state level differences rather than analyzing the relationship of rapid fluctuations between two measures. Glutamate levels, L/P-ratios and local tissue oxygenation were within the physiological range except for patient #5 and patient #1 on the first day of monitoring (Fig. 3). Interestingly, brain glucose concentrations
20
J. Diedler et al. / Journal of the Neurological Sciences 295 (2010) 16–22
Table 2 Parameters to estimate autoregulation and determinants of PbrO2. Patient # 1 2 3 4 5
PRx 0.14 – −0.11 0.02 0.52
ORx
Coherence PbrO2/CPP
Coherence PbrO2/PRx
0.75 0.77 0.63 0.37 0.41
0.86 0.84 0.83 0.82 0.67
0.28 0.19 0.14 0.45
a
Care withdrawn at request of the family on day 6 post-ictus.
were below 1 mmol/l in the majority of patients. While no clear relationship is evident between ORx and microdialysis parameters or the absolute levels of tissue oxygenation, failure of cerebrovascular reactivity in patient #5 was associated with elevated glutamate levels and L/P ratios (upper right corner) and decreased brain glucose levels (lower right corner). Moreover, averaged PbrO2 levels rather seemed
to be correlated to PRx rather than ORx. Because 3 repeated measurements for each subject are represented in the plots, formal correlation analysis was not performed. 4. Discussion We have investigated different measures of cerebral autoregulation, tissue oxygenation and metabolic parameters in the perihematomal zone in five conservatively treated patients suffering from spontaneous, deep ICH. All patients had ORx values indicative for disturbed autoregulation. In contrast, only one patient had a PRx consistent with sustained failure of cerebrovascular reactivity. This patient at the same time showed the worst metabolic pattern and lowest tissue oxygenation levels. The concept of PRx has been introduced by Czosnyka et al. [10]. It is based on the idea that in case of intact cerebrovascular reactivity,
Fig. 3. Average PRx and ORx values during the first three days of monitoring plotted against glutamate concentrations, L/P ratios, brain glucose concentrations and local tissue oxygenation. Elevated glutamate concentrations and L/P ratios as well as decreased glucose levels seemed to be associated with PRx values N 0.2. For ORx there was no clear relationship to metabolic parameters or PbrO2.
J. Diedler et al. / Journal of the Neurological Sciences 295 (2010) 16–22
cerebral vessels will contract in response to raises in the MAP, reducing intracranial blood volume and thereby ICP. This concept has been clinically evaluated in several studies and Czosnyka et al. have reported that a PRx above 0.2 for more than 6 h was associated with fatal outcome in TBI patients [18]. ORx has been introduced by Jaeger et al. in patients suffering from SAH and TBI [11,17]. It is based on the assumption that when CPP and PbrO2 are highly correlated in such a way that PbrO2 passively follows CPP, autoregulation is disturbed. While Jaeger et al. have reported a high correlation between ORx and PRx [11], interestingly, Radolovich et al. recently were unable to confirm a correlation between both indices in patients suffering from traumatic brain injury [19]. Differences in the location of the oxygen sensors between studies have been suggested as a possible explanation [19]. Based on our results and those of Radolovich et al., we suggest that ORx, in contrast to PRx, might be used as a local measure of autoregulation, representing the local cerebral microvasculature. Since PbrO2 probes provide information on tissue oxygenation within a small radius of the probe (∼ 17mm2 [20]), PbrO2 values and thereby PbrO2 derived indices are highly dependent on probe location. PRx, in contrast, derived from MAP and ICP, gives a global estimate of cerebrovascular reactivity, however not necessarily reflecting the local microvasculature in the perihemorrhagic region. It could be hypothesized that while global autoregulation is intact, local autoregulation in the perihematomal region may be severely disturbed. At the same time, the overall availability of substrates might depend on the upstream supply, influenced by the vasomotor functioning of the cerebral macrovasculature. However, no definitive conclusions can be drawn from our preliminary study and the relationship between ORx and PRx and the respective pathopysiological implications need to be clarified in a larger study. Additional questions for further research include the relevance of autoregulation for brain edema formation, mass effect and ICP increases and clinical outcome. Three previous studies have employed microdialysis to investigate the neurochemistry in the perihematomal region [21–23], including patients with evacuated hematomas and not providing a link between local brain metabolism and autoregulation. Moreover, the operative procedure itself, the decompressive effect evoked by hematoma evacuation and the removal of blood and its degradation products may significantly influence the metabolic processes in the perihematomal zone. Using microdialysis, Nilsson et al. reported severely jeopardized tissue oxygenation [22]. In our study only the patient with globally impaired vasoreactivity had PbrO2 and L/P ratios indicative for tissue hypoxia; none of our patients had signs of new infarction in the follow-up imaging studies. Recently, the existence of an ischemic penumbra in ICH patients has been challenged [3–5] and a switch of concept from ischemic to metabolic penumbra was suggested [6], referring to the finding of increased glucose metabolism in the perihemorrhagic region [7]. The transient focal increases in glucose metabolism have been interpreted as signs of ongoing neuronal injury lasting for several days [7]. Of interest, we have found relatively low levels of interstitial brain glucose (Fig. 3) as compared to studies in the non-injured human brain [24,25], the patient with severely disturbed vasoreactivity having the lowest glucose levels. Since substrate supply depends on cerebral blood flow, patients suffering from impaired global autoregulation may be more vulnerable for secondary brain injury. The small sample size is the main limitation of our study. However, it must be noted that it was intended as hypothesis generating pilot study. From our experience, a large case series is difficult to obtain in a single centre. Navigated probe placement is logistically challenging, and despite the considerable experience in our centre [26–30], multimodality brain monitoring proved to be work and cost intensive, requiring a broad infrastructure. This may be contribute to the fact that so far there has no multimodality brain monitoring data been available in the literature on conservatively treated patients suffering from spontaneous ICH.
21
We conclude that multimodality monitoring of the perihemorrhagic penumbra is feasible and may provide important information on autoregulation status and brain chemistry. Monitoring of autoregulation, tissue oxygenation and brain metabolism in the perihemorrhagic region may help to elucidate the complex pathophysiology of ICH and allow identifying future therapeutic targets. Funding None. Conflicts of interest Jennifer Diedler has received speaker's honoraria from Integra. Acknowledgments Jennifer Diedler was supported by a fellow-ship grant of the Medical Faculty of the University of Heidelberg. References [1] Mayer SA, Lignelli A, Fink ME, Kessler DB, Thomas CE, Swarup R, et al. Perilesional blood flow and edema formation in acute intracerebral hemorrhage: a SPECT study. Stroke 1998;29:1791–8. [2] Pascual AM, Lopez-Mut JV, Benlloch V, Chamarro R, Soler J, Lainez MJ. Perfusionweighted magnetic resonance imaging in acute intracerebral hemorrhage at baseline and during the 1st and 2nd week: a longitudinal study. Cerebrovasc Dis 2007;23:6–13. [3] Herweh C, Juttler E, Schellinger PD, Klotz E, Jenetzky E, Orakcioglu B, et al. Evidence against a perihemorrhagic penumbra provided by perfusion computed tomography. Stroke 2007;38:2941–7. [4] Schellinger PD, Fiebach JB, Hoffmann K, Becker K, Orakcioglu B, Kollmar R, et al. Stroke MRI in intracerebral hemorrhage: is there a perihemorrhagic penumbra? Stroke 2003;34:1674–9. [5] Zazulia AR, Diringer MN, Videen TO, Adams RE, Yundt K, Aiyagari V, et al. Hypoperfusion without ischemia surrounding acute intracerebral hemorrhage. J Cereb Blood Flow Metab 2001;21:804–10. [6] Vespa PM. Metabolic penumbra in intracerebral hemorrhage. Stroke 2009;40: 1547–8. [7] Zazulia AR, Videen TO, Powers WJ. Transient focal increase in perihematomal glucose metabolism after acute human intracerebral hemorrhage. Stroke 2009;40: 1638–43. [8] Diedler J, Sykora M, Rupp A, Poli S, Karpel-Massler G, Sakowitz O, et al. Impaired cerebral vasomotor activity in spontaneous intracerebral hemorrhage. Stroke 2009;40:815–9. [9] Reinhard M, Neunhoeffer F, Gerds TA, Niesen WD, Buttler KJ, Timmer J, et al. Secondary decline of cerebral autoregulation is associated with worse outcome after intracerebral hemorrhage. Intensive Care Med 2010;36:264–71. [10] Czosnyka M, Smielewski P, Kirkpatrick P, Laing RJ, Menon D, Pickard JD. Continuous assessment of the cerebral vasomotor reactivity in head injury. Neurosurgery 1997;41:11–7 discussion 7-9. [11] Jaeger M, Schuhmann MU, Soehle M, Meixensberger J. Continuous assessment of cerebrovascular autoregulation after traumatic brain injury using brain tissue oxygen pressure reactivity. Crit Care Med 2006;34:1783–8. [12] Broderick J, Connolly S, Feldmann E, Hanley D, Kase C, Krieger D, et al. Guidelines for the management of spontaneous intracerebral hemorrhage in adults: 2007 update: a guideline from the American Heart Association/American Stroke Association Stroke Council, High Blood Pressure Research Council, and the Quality of Care and Outcomes in Research Interdisciplinary Working Group. Circulation 2007;116:e391–413. [13] Steiner T, Kaste M, Forsting M, Mendelow D, Kwiecinski H, Szikora I, et al. Recommendations for the management of intracranial haemorrhage — part I: spontaneous intracerebral haemorrhage. The European Stroke Initiative Writing Committee and the Writing Committee for the EUSI Executive Committee. Cerebrovasc Dis 2006;22:294–316. [14] Broderick JP, Brott T, Duldner J, Tomsick T, Huster G. Volume of intracerebral hemorrhage: a powerful and easy-to-use predictor of 30-day mortality. Stroke 1993;24:987–93. [15] Graeb DA, Robertson WD, Lapointe JS, Nugent RA, Harrison PB. Computed tomographic diagnosis of intraventricular hemorrhage. Etiology and prognosis. Radiology 1982;143:91–6. [16] Steiner LA, Czosnyka M, Piechnik SK, Smielewski P, Chatfield D, Menon DK, et al. 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. [17] Jaeger M, Schuhmann MU, Soehle M, Nagel C, Meixensberger J. Continuous monitoring of cerebrovascular autoregulation after subarachnoid hemorrhage by brain tissue oxygen pressure reactivity and its relation to delayed cerebral infarction. Stroke 2007;38:981–6.
22
J. Diedler et al. / Journal of the Neurological Sciences 295 (2010) 16–22
[18] Czosnyka M, Smielewski P, Kirkpatrick P, Piechnik S, Laing R, Pickard JD. Continuous monitoring of cerebrovascular pressure-reactivity in head injury. Acta Neurochir Suppl 1998;71:74–7. [19] Radolovich DK, Czosnyka M, Timofeev I, Lavinio A, Hutchinson P, Gupta A, et al. Reactivity of brain tissue oxygen to change in cerebral perfusion pressure in head injured patients. Neurocrit Care 2009;10:274–9. [20] Kiening KL, Unterberg AW, Bardt TF, Schneider GH, Lanksch WR. Monitoring of cerebral oxygenation in patients with severe head injuries: brain tissue PO2 versus jugular vein oxygen saturation. J Neurosurg 1996;85:751–7. [21] Miller CM, Vespa PM, McArthur DL, Hirt D, Etchepare M. Frameless stereotactic aspiration and thrombolysis of deep intracerebral hemorrhage is associated with reduced levels of extracellular cerebral glutamate and unchanged lactate pyruvate ratios. Neurocrit Care 2007;6:22–9. [22] Nilsson OG, Polito A, Saveland H, Ungerstedt U, Nordstrom CH. Are primary supratentorial intracerebral hemorrhages surrounded by a biochemical penumbra? A microdialysis study. Neurosurgery 2006;59:521–8 discussion -8. [23] Wang E, Ho CL, Lee KK, Ng I, Ang BT. Changes in brain biochemistry and oxygenation in the zone surrounding primary intracerebral hemorrhage. Acta Neurochir Suppl 2008;102:293–7.
[24] Reinstrup P, Stahl N, Mellergard P, Uski T, Ungerstedt U, Nordstrom CH. Intracerebral microdialysis in clinical practice: baseline values for chemical markers during wakefulness, anesthesia, and neurosurgery. Neurosurgery 2000;47:701–9 discussion 9-10. [25] Schulz MK, Wang LP, Tange M, Bjerre P. Cerebral microdialysis monitoring: determination of normal and ischemic cerebral metabolisms in patients with aneurysmal subarachnoid hemorrhage. J Neurosurg 2000;93:808–14. [26] Berger C, Kiening K, Schwab S. Neurochemical monitoring of therapeutic effects in large human MCA infarction. Neurocrit Care 2008;9:352–6. [27] Berger C, Sakowitz OW, Kiening KL, Schwab S. Neurochemical monitoring of glycerol therapy in patients with ischemic brain edema. Stroke 2005;36:e4–6. [28] Schabitz WR, Berger C, Schellinger PD, Aschoff A, Steiner T, Schwab S. Neurometabolic changes during treatment with moderate hypothermia in a patient suffering from severe middle cerebral artery infarction. Cerebrovasc Dis 2001;12:298–302. [29] Berger C, Annecke A, Aschoff A, Spranger M, Schwab S. Neurochemical monitoring of fatal middle cerebral artery infarction. Stroke 1999;30:460–3. [30] Steiner T, Pilz J, Schellinger P, Wirtz R, Friederichs V, Aschoff A, et al. Multimodal online monitoring in middle cerebral artery territory stroke. Stroke 2001;32:2500–6.
Contents lists available at ScienceDirect
Journal of the Neurological Sciences j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / j n s
Autoregulation and brain metabolism in the perihematomal region of spontaneous intracerebral hemorrhage: An observational pilot study Jennifer Diedler a,⁎, Georg Karpel-Massler b,c, Marek Sykora a,d, Sven Poli a, Oliver W. Sakowitz b, Roland Veltkamp a, Thorsten Steiner a a
Department of Neurology, University of Heidelberg, Im Neuenheimer Feld 400, 69120 Heidelberg, Germany Department of Neurosurgery, University of Heidelberg, Im Neuenheimer Feld 400, 69120 Heidelberg, Germany Department of Neurosurgery, University of Ulm, Steinhövelstr.9, 89075 Ulm, Germany d Department of Neurology, Comenius University, Mickiewiczova 13, 813 69 Bratislava, Slovakia b c
a r t i c l e
i n f o
Article history: Received 23 November 2009 Received in revised form 20 April 2010 Accepted 27 May 2010 Available online 16 June 2010 Keywords: Intracerebral hemorrhage Autoregulation PRx ORx Microdialysis Tissue oxygenation
a b s t r a c t The metabolic and hemodynamic processes in the edema surrounding spontaneous, supratentorial intracerebral hemorrhage (ICH) are poorly understood. Specifically, the local metabolic effects of autoregulatory failure have not been described previously. In the current observational pilot study, microdialysis and brain tissue oxygenation probes (PbrO2) were placed in the perihemorrhagic edema using neuronavigation in five non-surgically treated patients with deep ICH. The cerebrovascular pressure reactivity index (PRx, moving correlation between mean arterial and intracranial pressure) and PbrO2 reactivity index (ORx, correlation between PbrO2 and cerebral perfusion pressure) were used to characterize cerebral autoregulation. While all five patients had ORx values indicative for severely disturbed autoregulation, assessment of PRx only in one patient was consistent with sustained failure of cerebrovascular reactivity. This patient at the same time had the worst metabolic parameters and the poorest tissue oxygenation. We conclude that multimodality monitoring in the perihemorrhagic penumbra is feasible. A study in a larger population is needed to clarify the relationship between PRx and ORx in ICH patients, the local metabolic effects of autoregulatory failure and its relation to brain edema formation and clinical outcome. © 2010 Elsevier B.V. All rights reserved.
1. Introduction The metabolic and hemodynamic processes in the perihematomal region of spontaneous intracerebral hemorrhages (ICH) remain controversial. While several studies provided evidence for a reduced perihematomal blood flow [1,2], recent imaging studies have dispelled the idea of major ischemia in the edematous zone adjacent to the hematoma [3–5]. Instead, the concept of a perihemorrhagic metabolic penumbra has been suggested, analogously to patients suffering from traumatic ICH [6,7]. Additionally, the state and significance of autoregulation of cerebral blood flow after ICH remains a matter of debate. Earlier studies did not find evidence for a disturbance of autoregulation in acute ICH. However, it has recently been shown that autoregulation may be impaired in ICH patients, possibly influencing outcome [8,9]. While the exact pathophysiological mechanisms remain undetermined, it has been suggested that disturbed autoregulation may promote additional brain injury, especially in the perihemorrhagic region. To further assess this subject, we undertook a pilot study
⁎ Corresponding author. Tel.: + 49 62215637557; fax: +49 622156467. E-mail address: [email protected] (J. Diedler). 0022-510X/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.jns.2010.05.027
including patients suffering from spontaneous supratentorial ICH who did not undergo hematoma evacuation. We performed multimodality monitoring directly measuring brain metabolites and brain tissue oxygenation (PbrO2) in the edema surrounding the hematoma. Cerebral autoregulation was assessed using the pressure reactivity index (PRx) and the oxygen pressure reactivity index (ORx) as described previously [10,11]. In this hypothesis generating pilot study, we aimed to evaluate feasibility of using PbrO2 and microdialysis to assess autoregulation and local metabolism in the perihemorrhagic region. 2. Methods 2.1. Patients During April 2007 and August 2008, five prospectively collected patients suffering from spontaneous, deep supratentorial ICH underwent invasive multimodality neuromonitoring including measurement of PbrO2, cerebral microdialysis, and ICP and CPP monitoring. Criteria for invasive neuromonitoring and hence inclusion into the current study were 1) intraventricular hemorrhage extension with need for placement of an extraventricular drain (and thereby requiring intubation and transfer to the operating room) and 2) conservative therapy without
J. Diedler et al. / Journal of the Neurological Sciences 295 (2010) 16–22
evacuation of the hematoma as decided by the attending physicians because of deep hemorrhage location. The study was approved by the local ethics committee. Written, informed consent was obtained from the legal representatives of the patients. All patients were intubated and mechanically ventilated during the entire monitoring period. Intravenous midazolam and sufentanil were used for analgosedation. Therapy was aimed to keep ICP b20 mm Hg and CPP N60 mm Hg, according to current guidelines [12,13]. Arterial blood gas samples were obtained every 2 h to adjust ventilation parameters. PaCO2 was maintained between 35 and 45 mm Hg. Neurological deficit on admission was assessed by the National Institutes of Health Stroke Scale Score (NIHSSS). Hematoma volume was calculated from the first CT scan using the a × b × c × 0.5 method, as described by Broderick et al. [14]. Intraventricular bleeding was scored according to the Graeb score [15].
17
example during in-house transportations). Artifacts mostly resulted from inadequate pressure signals, for example during withdrawal of blood gas probes or due to nursing interventions. After re-sampling in order to obtain one value every 6 s, PRx was calculated every 60 s as a moving linear (Pearson's) correlation between 40 consecutive values of MAP and ICP, as described by Steiner and Czosnyka [10,16]. CPP and PbrO2 values were averaged in the same moving window in order to obtain a time synchronous matrix containing PRx, CPP and PbrO2. For calculation of PRx, only data points fulfilling the criteria of systolic arterial pressure between 60–180 mm Hg, mean arterial pressure between 50 and 120 mm Hg and ICP N0 mm Hg were included in the analysis. Fisher-Z-transformation was used to calculate mean PRx values. All calculations were performed using Matlab (MathWorks, Version 7.5). 2.5. ORx
2.2. Neuromonitoring Intracranial pressure was measured using a combined ventricular drainage and piezoelectric ICP probe allowing continuous assessment of ICP irrespective of drainage (Raumedic NEUROVENT, Germany). PbrO2 was measured using an intraparenchymal probe (LICOX PMO, Integra, Germany). Microdialysis catheters (CMA 70, CMA Microdialysis, Sweden; membrane length 10 mm) were perfused with artificial cerebrospinal fluid (CMA Microdialysis, Sweden) at a rate of 0.3 μl/ min. The perfusates were collected in microvials and analyzed in 1 to 2 h intervals immediately after collection. Samples were analyzed for glucose, pyruvate, lactate, glutamate and glycerol using the ISCUS Clinicial Microdialysis Analyzer (CMA Microdialysis, Sweden). In order to ensure exact placement of the probes in the perihematomal region, an intraoperative, image-guided, frameless, localization system (BrainLAB VectorVision® compact Navigation System, BrainLAB, Feldkirchen, Germany) was used in 4 of 5 patients. Placement of the probes was guided by a freehand pointer determining the correct direction and depth of insertion. Blood pressure was measured from the radial artery (Dräger, Siemens, Germany). No complications or adverse events caused by the monitoring probes were observed. ICP, CPP, PbrO2, systolic and diastolic blood pressure, MAP, and heart rate were synchronously recorded at a sampling frequency of 1 Hz in 4 patients. Due to technical difficulties, one patient (#2) was recorded with a sampling frequency of 1/min (∼0.02 Hz). The data were stored on a bedside computer using the ICU pilot software (CMA, Sweden). 2.3. Signal analysis To estimate cerebral autoregulation and its local effects on tissue oxygenation, the indices PRx and ORx were calculated. Additionally, the coherence between the moving averages of CPP, PRx and PbrO2 were calculated as described below. 2.4. PRx The PRx investigates the correlation between slow wave changes in MAP and ICP. It may be used as a continuous surrogate measure of cerebral autoregulation. A positive PRx implies a positive association between the slow components of MAP and ICP, indicating a passive, nonreactive behavior of the cerebral vessels. A negative value reflects a normally reactive vascular bed where changes in MAP result in inversely correlated changes in ICP within a 5 to 30-second time window. PRx was calculated as described previously [8,10]. Before analysis, data files were screened and epochs containing incomplete datasets or artifacts were visually identified and excluded from the analysis. Incomplete datasets were caused by disturbed interaction between the monitoring system and the recording software or due to complete disconnection the patient from the monitoring system (for
Analogously to PRx, ORx values range between + 1 and −1. A positive correlation suggests that PbrO2 passively follows CPP, indicative for disturbed autoregulation [11,17]. The oxygen reactivity index ORx was calculated after artifact elimination and re-sampling to obtain one value every 30 s as moving linear correlation coefficient between CPP and PbrO2 over an average period of 1 h as described by Jaeger et al. [11]. For patient #2 with a sampling frequency of 1/min, ORx was calculated based on minutely values. 2.6. Coherence To assess the relation between local tissue oxygenation and CPP or PRx respectively, the coherence between PbrO2 vs. CPP and PbrO2 vs. PRx was analyzed. The window length of the 50% overlapping hamming window was chosen to obtain a frequency resolution of 0.23 cycles/h and the mean coherence in the frequency range of 0.23 to 4.2 cycles/h was calculated. Calculation of coherence was based on the low passed filtered curves of PRx and the moving averages of PbrO2 and CPP (first-order filter of 0.01 Hz). 3. Results Five non-consecutive patients were included in the study. Patient characteristics are shown in Table 1. The mean age was 61.2 years (range 40–75). The median hematoma size was 18.4 ml (range 14.1– 67). The mean time from onset of symptoms to start of monitoring was 19.6 h (range 11–32). According to our inclusion criteria, all patients had intraventricular hemorrhage extension and required ventriculostomy. Probe placement for each individual patient is illustrated in Fig. 1. One patient (#3) died during ICU stay because care was withdrawn at the request of the family. The monitoring probes were removed in this patient at the time of withdrawal of care. Fig. 2 shows the fluctuations of PbrO2, CPP and PRx for each patient individually. Each data point represents a minutely moving average summarizing 4 min of monitoring time. Phases of disturbed vasoreactivity defined as PRx N0.2 are indicated in gray. Table 2 gives an overview of the different autoregulatory indices. The data of each patient will be presented individually before summarizing the findings and comparing 24-hour averages. 3.1. Patient #1 Patient #1 (Fig. 2A) had a mean PRx of 0.14, generally indicating intact cerebrovascular reactivity. Except for two short periods (gray areas), PRx remained below the threshold of 0.2. The mean PbrO2 was 23 ± 7 mm Hg, with a mean arterial blood oxygen content (PaO2) of 119 ± 34 mm Hg. CPP was 86 ± 7 mm Hg in the mean. Of interest, despite a PRx indicating globally intact vasoreactivity, ORx was 0.73, indicative for severe autoregulatory failure. Fig. 2A shows the
18
J. Diedler et al. / Journal of the Neurological Sciences 295 (2010) 16–22
Table 1 Patient characteristics. Patient #
Hemorrhage volume [ml]
Graeb score
Baseline NIHSS
Discharge NIHSS
Age
Etiology
Time from ictus to monitoring [h]
1 2 3 4 5
18.4 15.2 67.3 31.9 14.1
5 7 11 11 6
34 34 34 34 10
17 13 42b 15 20
59 63 69 40 75
Hypertensive Coagulopathy Coagulopathy Hypertensive Coagulopathy
18 32 11 26a 11
a b
Microdialysis probe had to be replaced and worked only 55 h after onset of symptoms. Care withdrawn at request of the family on day 6 post-ictus.
synchronously fluctuating curves of CPP and PbrO2. This relation is also mirrored by the high coherence between both parameters (0.86). In contrast, coherence between PbrO2 and PRx was 0.28, indicating that slow fluctuations (frequency range of 0.23 to 4.2 cycles/h) of local tissue oxygenation were not associated with changes of PRx in this patient. 3.2. Patient #2 For patient # 2 (Fig. 2B), the mean PbrO2 was 31 ± 7 mm Hg, PaO2 was 118 ± 10 mm Hg and CPP 80 ± 10 mm Hg, respectively. The coherence between CPP and PbrO2 was 0.84, ORx was 0.77, suggestive for failure of autoregulation. 3.3. Patient #3 Patient #3 (Fig. 2C) had an overall PRx of −0.11, indicative for intact global cerebrovascular reactivity. However, this patient shows
clearly discernible periods of intact versus disturbed cerebrovascular reactivity (gray areas). Of interest, phases exceeding a PRx N0.2 were associated with simultaneous drops of PbrO2 and CPP. This observation is not mirrored by analysis of coherence between PRx and PbrO2 (0.19). However, time resolution of the coherence function was not suited to detect very slow processes. Mean PbrO2, CPP and PaO2 levels were 27 ± 9 mm Hg, 69 ± 10 mm Hg and 117 ± 27 mm Hg, respectively. The coherence between CPP and PbrO2 was 0.83 and ORx was 0.60, suggesting disturbed autoregulation.
3.3.1. Patient #4 A similar pattern of clearly discernable phases of disturbed vasoreactivity was observed in patient #4 (Fig. 2D). However, unlike in patient #3, these periods were not associated with clear drops in PbrO2. Mean PbrO2 was 30 ± 16 mm Hg, mean CPP 80 ± 8 mm Hg, and mean PaO2 99 ± 13 mm Hg respectively. ORx was 0.37. Coherence between PbrO2 and CPP was 0.82, and 0.14 between PbrO2 and PRx.
Fig. 1. Probe placement in the perihematomal edema in the 5 included patients (MD: microdialysis catheter, O2: oxygen probe).
J. Diedler et al. / Journal of the Neurological Sciences 295 (2010) 16–22
19
Fig. 2. Time course of PbrO2, CPP and corresponding PRx values plotted for each patient separately (A, B, C, D and E). Each data point represents a minutely moving average summarizing 4 min of monitoring time following artifact elimination. Gray areas indicate episodes of disturbed autoregulation (PRx N 0.2). Original data was low-pass filtered for better visualisation.
3.4. Patient #5 Patient #5 had severely disturbed cerebrovascular reactivity throughout the entire monitoring period (mean PRx 0.52). At the same time, this patient had the lowest PbrO2 values during the entire monitoring period (11 ± 7 mm Hg) despite having normal PaO2 levels (100 ± 13 mm Hg). Mean CPP was 72 ± 10 mm Hg. Coherence between PbrO2 and CPP was 0.67, ORx was 0.41, indicating failure of autoregulation. Coherence between Prx and PbrO2 was 0.45.
Fig. 3 shows the relationship between microdialysis parameters and local tissue oxygenation versus PRx, and ORx, respectively. All parameters were averaged for the first three days of monitoring for each patient individually. Comparing daily averages aims to compare absolute, steady-state level differences rather than analyzing the relationship of rapid fluctuations between two measures. Glutamate levels, L/P-ratios and local tissue oxygenation were within the physiological range except for patient #5 and patient #1 on the first day of monitoring (Fig. 3). Interestingly, brain glucose concentrations
20
J. Diedler et al. / Journal of the Neurological Sciences 295 (2010) 16–22
Table 2 Parameters to estimate autoregulation and determinants of PbrO2. Patient # 1 2 3 4 5
PRx 0.14 – −0.11 0.02 0.52
ORx
Coherence PbrO2/CPP
Coherence PbrO2/PRx
0.75 0.77 0.63 0.37 0.41
0.86 0.84 0.83 0.82 0.67
0.28 0.19 0.14 0.45
a
Care withdrawn at request of the family on day 6 post-ictus.
were below 1 mmol/l in the majority of patients. While no clear relationship is evident between ORx and microdialysis parameters or the absolute levels of tissue oxygenation, failure of cerebrovascular reactivity in patient #5 was associated with elevated glutamate levels and L/P ratios (upper right corner) and decreased brain glucose levels (lower right corner). Moreover, averaged PbrO2 levels rather seemed
to be correlated to PRx rather than ORx. Because 3 repeated measurements for each subject are represented in the plots, formal correlation analysis was not performed. 4. Discussion We have investigated different measures of cerebral autoregulation, tissue oxygenation and metabolic parameters in the perihematomal zone in five conservatively treated patients suffering from spontaneous, deep ICH. All patients had ORx values indicative for disturbed autoregulation. In contrast, only one patient had a PRx consistent with sustained failure of cerebrovascular reactivity. This patient at the same time showed the worst metabolic pattern and lowest tissue oxygenation levels. The concept of PRx has been introduced by Czosnyka et al. [10]. It is based on the idea that in case of intact cerebrovascular reactivity,
Fig. 3. Average PRx and ORx values during the first three days of monitoring plotted against glutamate concentrations, L/P ratios, brain glucose concentrations and local tissue oxygenation. Elevated glutamate concentrations and L/P ratios as well as decreased glucose levels seemed to be associated with PRx values N 0.2. For ORx there was no clear relationship to metabolic parameters or PbrO2.
J. Diedler et al. / Journal of the Neurological Sciences 295 (2010) 16–22
cerebral vessels will contract in response to raises in the MAP, reducing intracranial blood volume and thereby ICP. This concept has been clinically evaluated in several studies and Czosnyka et al. have reported that a PRx above 0.2 for more than 6 h was associated with fatal outcome in TBI patients [18]. ORx has been introduced by Jaeger et al. in patients suffering from SAH and TBI [11,17]. It is based on the assumption that when CPP and PbrO2 are highly correlated in such a way that PbrO2 passively follows CPP, autoregulation is disturbed. While Jaeger et al. have reported a high correlation between ORx and PRx [11], interestingly, Radolovich et al. recently were unable to confirm a correlation between both indices in patients suffering from traumatic brain injury [19]. Differences in the location of the oxygen sensors between studies have been suggested as a possible explanation [19]. Based on our results and those of Radolovich et al., we suggest that ORx, in contrast to PRx, might be used as a local measure of autoregulation, representing the local cerebral microvasculature. Since PbrO2 probes provide information on tissue oxygenation within a small radius of the probe (∼ 17mm2 [20]), PbrO2 values and thereby PbrO2 derived indices are highly dependent on probe location. PRx, in contrast, derived from MAP and ICP, gives a global estimate of cerebrovascular reactivity, however not necessarily reflecting the local microvasculature in the perihemorrhagic region. It could be hypothesized that while global autoregulation is intact, local autoregulation in the perihematomal region may be severely disturbed. At the same time, the overall availability of substrates might depend on the upstream supply, influenced by the vasomotor functioning of the cerebral macrovasculature. However, no definitive conclusions can be drawn from our preliminary study and the relationship between ORx and PRx and the respective pathopysiological implications need to be clarified in a larger study. Additional questions for further research include the relevance of autoregulation for brain edema formation, mass effect and ICP increases and clinical outcome. Three previous studies have employed microdialysis to investigate the neurochemistry in the perihematomal region [21–23], including patients with evacuated hematomas and not providing a link between local brain metabolism and autoregulation. Moreover, the operative procedure itself, the decompressive effect evoked by hematoma evacuation and the removal of blood and its degradation products may significantly influence the metabolic processes in the perihematomal zone. Using microdialysis, Nilsson et al. reported severely jeopardized tissue oxygenation [22]. In our study only the patient with globally impaired vasoreactivity had PbrO2 and L/P ratios indicative for tissue hypoxia; none of our patients had signs of new infarction in the follow-up imaging studies. Recently, the existence of an ischemic penumbra in ICH patients has been challenged [3–5] and a switch of concept from ischemic to metabolic penumbra was suggested [6], referring to the finding of increased glucose metabolism in the perihemorrhagic region [7]. The transient focal increases in glucose metabolism have been interpreted as signs of ongoing neuronal injury lasting for several days [7]. Of interest, we have found relatively low levels of interstitial brain glucose (Fig. 3) as compared to studies in the non-injured human brain [24,25], the patient with severely disturbed vasoreactivity having the lowest glucose levels. Since substrate supply depends on cerebral blood flow, patients suffering from impaired global autoregulation may be more vulnerable for secondary brain injury. The small sample size is the main limitation of our study. However, it must be noted that it was intended as hypothesis generating pilot study. From our experience, a large case series is difficult to obtain in a single centre. Navigated probe placement is logistically challenging, and despite the considerable experience in our centre [26–30], multimodality brain monitoring proved to be work and cost intensive, requiring a broad infrastructure. This may be contribute to the fact that so far there has no multimodality brain monitoring data been available in the literature on conservatively treated patients suffering from spontaneous ICH.
21
We conclude that multimodality monitoring of the perihemorrhagic penumbra is feasible and may provide important information on autoregulation status and brain chemistry. Monitoring of autoregulation, tissue oxygenation and brain metabolism in the perihemorrhagic region may help to elucidate the complex pathophysiology of ICH and allow identifying future therapeutic targets. Funding None. Conflicts of interest Jennifer Diedler has received speaker's honoraria from Integra. Acknowledgments Jennifer Diedler was supported by a fellow-ship grant of the Medical Faculty of the University of Heidelberg. References [1] Mayer SA, Lignelli A, Fink ME, Kessler DB, Thomas CE, Swarup R, et al. Perilesional blood flow and edema formation in acute intracerebral hemorrhage: a SPECT study. Stroke 1998;29:1791–8. [2] Pascual AM, Lopez-Mut JV, Benlloch V, Chamarro R, Soler J, Lainez MJ. Perfusionweighted magnetic resonance imaging in acute intracerebral hemorrhage at baseline and during the 1st and 2nd week: a longitudinal study. Cerebrovasc Dis 2007;23:6–13. [3] Herweh C, Juttler E, Schellinger PD, Klotz E, Jenetzky E, Orakcioglu B, et al. Evidence against a perihemorrhagic penumbra provided by perfusion computed tomography. Stroke 2007;38:2941–7. [4] Schellinger PD, Fiebach JB, Hoffmann K, Becker K, Orakcioglu B, Kollmar R, et al. Stroke MRI in intracerebral hemorrhage: is there a perihemorrhagic penumbra? Stroke 2003;34:1674–9. [5] Zazulia AR, Diringer MN, Videen TO, Adams RE, Yundt K, Aiyagari V, et al. Hypoperfusion without ischemia surrounding acute intracerebral hemorrhage. J Cereb Blood Flow Metab 2001;21:804–10. [6] Vespa PM. Metabolic penumbra in intracerebral hemorrhage. Stroke 2009;40: 1547–8. [7] Zazulia AR, Videen TO, Powers WJ. Transient focal increase in perihematomal glucose metabolism after acute human intracerebral hemorrhage. Stroke 2009;40: 1638–43. [8] Diedler J, Sykora M, Rupp A, Poli S, Karpel-Massler G, Sakowitz O, et al. Impaired cerebral vasomotor activity in spontaneous intracerebral hemorrhage. Stroke 2009;40:815–9. [9] Reinhard M, Neunhoeffer F, Gerds TA, Niesen WD, Buttler KJ, Timmer J, et al. Secondary decline of cerebral autoregulation is associated with worse outcome after intracerebral hemorrhage. Intensive Care Med 2010;36:264–71. [10] Czosnyka M, Smielewski P, Kirkpatrick P, Laing RJ, Menon D, Pickard JD. Continuous assessment of the cerebral vasomotor reactivity in head injury. Neurosurgery 1997;41:11–7 discussion 7-9. [11] Jaeger M, Schuhmann MU, Soehle M, Meixensberger J. Continuous assessment of cerebrovascular autoregulation after traumatic brain injury using brain tissue oxygen pressure reactivity. Crit Care Med 2006;34:1783–8. [12] Broderick J, Connolly S, Feldmann E, Hanley D, Kase C, Krieger D, et al. Guidelines for the management of spontaneous intracerebral hemorrhage in adults: 2007 update: a guideline from the American Heart Association/American Stroke Association Stroke Council, High Blood Pressure Research Council, and the Quality of Care and Outcomes in Research Interdisciplinary Working Group. Circulation 2007;116:e391–413. [13] Steiner T, Kaste M, Forsting M, Mendelow D, Kwiecinski H, Szikora I, et al. Recommendations for the management of intracranial haemorrhage — part I: spontaneous intracerebral haemorrhage. The European Stroke Initiative Writing Committee and the Writing Committee for the EUSI Executive Committee. Cerebrovasc Dis 2006;22:294–316. [14] Broderick JP, Brott T, Duldner J, Tomsick T, Huster G. Volume of intracerebral hemorrhage: a powerful and easy-to-use predictor of 30-day mortality. Stroke 1993;24:987–93. [15] Graeb DA, Robertson WD, Lapointe JS, Nugent RA, Harrison PB. Computed tomographic diagnosis of intraventricular hemorrhage. Etiology and prognosis. Radiology 1982;143:91–6. [16] Steiner LA, Czosnyka M, Piechnik SK, Smielewski P, Chatfield D, Menon DK, et al. 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. [17] Jaeger M, Schuhmann MU, Soehle M, Nagel C, Meixensberger J. Continuous monitoring of cerebrovascular autoregulation after subarachnoid hemorrhage by brain tissue oxygen pressure reactivity and its relation to delayed cerebral infarction. Stroke 2007;38:981–6.
22
J. Diedler et al. / Journal of the Neurological Sciences 295 (2010) 16–22
[18] Czosnyka M, Smielewski P, Kirkpatrick P, Piechnik S, Laing R, Pickard JD. Continuous monitoring of cerebrovascular pressure-reactivity in head injury. Acta Neurochir Suppl 1998;71:74–7. [19] Radolovich DK, Czosnyka M, Timofeev I, Lavinio A, Hutchinson P, Gupta A, et al. Reactivity of brain tissue oxygen to change in cerebral perfusion pressure in head injured patients. Neurocrit Care 2009;10:274–9. [20] Kiening KL, Unterberg AW, Bardt TF, Schneider GH, Lanksch WR. Monitoring of cerebral oxygenation in patients with severe head injuries: brain tissue PO2 versus jugular vein oxygen saturation. J Neurosurg 1996;85:751–7. [21] Miller CM, Vespa PM, McArthur DL, Hirt D, Etchepare M. Frameless stereotactic aspiration and thrombolysis of deep intracerebral hemorrhage is associated with reduced levels of extracellular cerebral glutamate and unchanged lactate pyruvate ratios. Neurocrit Care 2007;6:22–9. [22] Nilsson OG, Polito A, Saveland H, Ungerstedt U, Nordstrom CH. Are primary supratentorial intracerebral hemorrhages surrounded by a biochemical penumbra? A microdialysis study. Neurosurgery 2006;59:521–8 discussion -8. [23] Wang E, Ho CL, Lee KK, Ng I, Ang BT. Changes in brain biochemistry and oxygenation in the zone surrounding primary intracerebral hemorrhage. Acta Neurochir Suppl 2008;102:293–7.
[24] Reinstrup P, Stahl N, Mellergard P, Uski T, Ungerstedt U, Nordstrom CH. Intracerebral microdialysis in clinical practice: baseline values for chemical markers during wakefulness, anesthesia, and neurosurgery. Neurosurgery 2000;47:701–9 discussion 9-10. [25] Schulz MK, Wang LP, Tange M, Bjerre P. Cerebral microdialysis monitoring: determination of normal and ischemic cerebral metabolisms in patients with aneurysmal subarachnoid hemorrhage. J Neurosurg 2000;93:808–14. [26] Berger C, Kiening K, Schwab S. Neurochemical monitoring of therapeutic effects in large human MCA infarction. Neurocrit Care 2008;9:352–6. [27] Berger C, Sakowitz OW, Kiening KL, Schwab S. Neurochemical monitoring of glycerol therapy in patients with ischemic brain edema. Stroke 2005;36:e4–6. [28] Schabitz WR, Berger C, Schellinger PD, Aschoff A, Steiner T, Schwab S. Neurometabolic changes during treatment with moderate hypothermia in a patient suffering from severe middle cerebral artery infarction. Cerebrovasc Dis 2001;12:298–302. [29] Berger C, Annecke A, Aschoff A, Spranger M, Schwab S. Neurochemical monitoring of fatal middle cerebral artery infarction. Stroke 1999;30:460–3. [30] Steiner T, Pilz J, Schellinger P, Wirtz R, Friederichs V, Aschoff A, et al. Multimodal online monitoring in middle cerebral artery territory stroke. Stroke 2001;32:2500–6.