43. Continuous assessment of cerebrovascular resistance and cerebral blood flow before and following traumatic brain injury
M. Daley1, N. Narayanan1 and C. Leffler2
1The University of Memphis; 2The University of Tennessee Health Science Center, Memphis, Tennessee, USA
Background and aims: Published guidelines for the management of adult patients with severe traumatic brain injury prescribe the maintenance of a general threshold of cerebral perfusion pressure (CPP) of 60 mm Hg during intensive care monitoring of arterial blood pressure (ABP) and intracranial pressure (ICP).1 This threshold value of CPP is near the critical threshold for ischemia, the lower limit of cerebrovascular autoregulation, which is generally thought to be between 50 to 60 mm Hg.2 These guidelines point out the need for the development of methods that individualize patient CPP management and minimize secondary ischemic complications associated with traumatic brain injury.1
We have developed a laboratory method to determine model-derived assessments of cerebrovascular resistance (mCVR) and cerebral blood flow (mCBF) from cerebrovascular pressure transmission, the dynamic relationship between ABP and ICP.3 The aim of this twofold study is to:
evaluate relative changes in the model-derived parameters of mCVR and mCBF with corresponding changes in the pial arteriolar vascular parameters of pial arteriolar resistance (PAR) and relative pial arteriolar blood flow (rPABF); and
examine the efficacy of the proposed modeling methodology for continuous assessment of the state of cerebrovascular regulation by evaluating relative changes in the model-derived parameters, mCVR and mCBF, in relation to changes of CPP induced by pressor challenge prior to and following fluid percussion brain injury.
Methods: Changes of arterial blood pressure (ABP), intracranial pressure (ICP), pial arteriolar resistance (PAR) and relative blood flow (rPABF) induced by acute pressor challenge (norepinephrine (1 uk/kg/min)) were evaluated in both uninjured and fluid percussion brain injured piglets (N = 6) equipped with cranial windows. Assessments of model-derived cerebrovascular resistance (mCVR) and cerebral blood flow (mCBF) were obtained by a computational modeling method of cerebrovascular pressure transmission for each challenge.
Results: Consistent with functional autoregulation of the uninjured cerebrovascular circulation, hypertensive challenge resulted in a significant increase of PAR and mCVR; whereas, both rPABF and mCBF remained constant. For all injured piglets, hypertensive challenge resulted in a significant decrease of PAR and mCVR consistent with impaired autoregulation. Hypertensive challenge also significantly increased both rPABF, and mCBF with increased CPP with correlation values of (r = 0.96, P<0.01) and (r = 0.97, P⩽0.01) respectively.
Conclusions: Assessment of model-derived cerebrovascular resistance and cerebral blood flow with changes of CPP may provide a means to continuously monitor the state of cerebrovascular regulation and cerebral perfusion.
45. Validation of computational modeling method to assess cerebral blood flow and regulation from arterial blood pressure and intracranial pressure recordings
N. Narayanan1, C. Leffler2 and M. Daley1
1The University of Memphis; 2The University of Tennessee Health Science Center, Memphis, Tennessee, USA
Background and aims: Guidelines for the management of adult patients with severe traumatic brain injury prescribe the maintenance of cerebral perfusion pressure (CPP) near the lower limit of cerebrovascular regulation during intensive care monitoring of arterial blood pressure (ABP) and intracranial pressure (ICP).1 The lower limits of regulation for pediatric patients are age-dependent2 and generally unknown. To prevent either excessive or insufficient perfusion during management, a method designed to continuously evaluate cerebral blood flow (CBF) and cerebrovascular resistance (CVR) of each patient is needed. The aim of this study is to validate a computational method based on a biomechanical model of the dynamic relationship between ABP and ICP. This method is designed to continuously assess to model derived parameters of cerebral blood flow (mCBF) and cerebrovascular resistance (mCVR). Comparisons between experimental and model derived assessments of CBF and CVR were made to examine the validity of the proposed modeling methodology.
Methods: ABP, ICP, CPP, CBF by H2 clearance, and pial arteriolar diameter (PAD) were measured in piglets equipped with cranial windows during normocapnia (N = 6) and permissive hypercapnia (N = 4). Corresponding mCVR and mCBF were calculated. Experimental CVR was computed as CPP/CBF and pial arterial resistance (PAR) was derived from PAD. Grand mean parameter values (±s.d.) were derived for each experimental condition.
Results: A summary of the comparisons between experimentally derived values of CBF, CVR, and PAR and the model-derived values of mCBF and mCVR is given in Table 1. Experimental and model-derived values of CBF and CVR were not significantly different.
Comparison of grand mean values (±s.d.) of ABP, ICP, CPP, experimental CBF, CVR, PAR and model-derived assessments of CBF and CVR during normocapnia and permissive hypercapnia
Denotes a significant degree of difference between mean values of normocapnic and permissive hypercapnic groups at P<0.01.
Conclusions: Assessment of model-derived cerebrovascular resistance and cerebral blood flow with changes of CPP may provide a means to continuously monitor the state of cerebrovascular regulation and cerebral perfusion.
143. Cerebral hemodynamics of preterm infants during postural intervention measured with diffuse correlation spectroscopy and transcranial doppler ultrasound
E. Buckley1, N. Cook2, T. Durduran1,3, M. Kim1, C. Zhou1, R. Choe1, G. Yu1, D. Licht4, J. Detre3,4, J. Greenberg5, H. Hurt2 and A. Yodh1
1Physics and Astronomy, University of Pennsylvania; 2Neonatology, 3Radiology, 4Neurology, Hospital of the University of Pennsylvania; 5Neurology, University of Pennsylvania, Philadelphia, Pennsylvania, USA
Objective: Preterm infants are at significant risk of brain injury, in part due to their limited ability to regulate cerebral blood flow (CBF). A continuous monitor of CBF at the bedside could therefore be a valuable supplement for gathering information about a patient's condition and for guiding treatment.
In this study, diffuse correlation spectroscopy (DCS), a continuous and non-invasive optical technique to measure blood flow, and transcranial Doppler ultrasound (TCD) were employed to monitor the cerebral hemodynamics of 4 preterm neonates during a 0 to 12° postural change on 9 different days.
Methods: The DCS instrument consists of a 785 nm source laser, a single mode detector fiber located on the tissue surface 1.5 cm from the source, and detection electronics to measure the light intensity autocorrelation function. Variations in the decay time of the autocorrelation function, which we refer to as the blood flow index (BFI) are related to blood flow in the tissue.1 BFI was recorded on the forehead of each preterm infant at 7 Hz for 5 mins at each head-of-bed angle (HOB) adjustment from 0 to 12° and was repeated 3 times.
For TCD, measurements of peak systolic, end diastolic, and mean velocities (PSV, EDV, and MV respectively) were obtained on the middle cerebral artery. Three data points were taken at each HOB angle before the bed was repositioned, and the HOB angle was elevated twice. DCS and TCD measurements were not taken at the same time.
Results:Figure 1 shows the relationship between baseline PSV and BFI for TCD and DCS data taken on the same day from the same patient. A significant correlation (P<0.05) between PSV and BFI, as well as between MV and BFI, was discovered. Similar correlations were not found with BFI and EDV.
Both techniques showed no significant population averaged hemodynamic changes during HOB elevation as compared to HOB flat. In addition, no population averaged correlations between rCBF and relative ultrasound parameters were found (rCBF and rPSV, rs = 0.1, P>0.05; rCBF and rEDV, rs = 0.17, P>0.05; rCBF and rMV, rs = 0.1, P>0.05).
Conclusions: In summary, we have demonstrated the feasibility of DCS to continuously monitor changes in CBF in very low birthweight preterm infants. We also showed with DCS that when posed with a small HOB challenge, the infants maintained a constant CBF. This result was corroborated by findings from TCD measurements.
Correlation between TCD and DCS.
336. Non-invasive measurements of cerebral blood flow with diffuse optics in patients after severe head injury
M.N. Kim1, T. Durduran1,2, S. Frangos3, B.L. Edlow4, E.M. Buckley1, H. Moss4, C. Zhou1, G. Yu1,5, R. Choe1, E. Maloney-Wilensky3, R.L. Wolf2, J.H. Woo2, M.S. Grady3, J.H. Greenberg4, J. Levine3, A.G. Yodh1, J.A. Detre2,4 and W.A. Kofke3,6
1Physics and Astronomy; 2Radiology; 3Neurosurgery; 4Neurology, University of Pennsylvania, Philadelphia, Pennsylvania; 5Biomedical Engineering, University of Kentucky, Lexington, Kentucky; 6Anesthesiology and Critical Care, University of Pennsylvania, Philadelphia, Pennsylvania, USA
Objectives: Patients suffering from severe head injury are a particularly challenging population clinically because of their heterogeneous condition and susceptibility to secondary injury. A continuous and non-invasive measure of microvascular CBF could help with individualizing care and alleviation of secondary injury. Current modalities for continuous monitoring provide only surrogate measurements of microvascular CBF, and some, like intracranial pressure, are invasive. We have developed and tested a novel, optical bedside monitor of microvascular CBF.
Methods: Six patients with traumatic brain injury (TBI), subarachnoid hemorrhage (SAH), or ischemic stroke were included in a validation study of diffuse correlation spectroscopy (DCS) against portable xenon-CT. CBF was measured continuously throughout two CT-scans: a baseline scan and a scan after pressor administration. Regions-of-interest under the DCS probes (placed bilaterally on the forehead) were drawn on the xenon-CT CBF maps, and CBF values were compared to values from DCS. Eleven TBI/SAH patients were recruited under a separate protocol to test DCS-CBF response to head-of-bed (HOB) elevation changes. DCS measured frontal cortical CBF while HOB position changed from 30° to supine, each for 5 mins. Results were compared against cerebral perfusion pressure (CPP) to assess cerebral autoregulation. Damaged autoregulation was defined as lack of correlation between CPP and CBF (P>0.05).
Results: Relative CBF measurements from DCS and xenon-CT comparing baseline and post-intervention values showed good correlation (R = 0.72, P = 0.02). Bland-Altman analysis comparing the two methods also showed good agreement. When HOB went from 30° to supine, CBF responses were highly variable both among patients as well as between hemispheres in a given patient. Figure below shows a CBF time-series with this type of discrepancy, in a 45-year-old male with subarachnoid hemorrhage from anterior communicating artery- and left anterior choroidal artery-origin aneurysms. Autoregulation is often damaged by neurotraumatic events, and the correlation (P = 0.03) between this patient's left hemisphere CBF and CPP may be interpreted as damaged autoregulation on that side. As a contrast, right hemisphere CBF did not correlate with CPP (P = 0.29). Averaged over the entire population, we saw a insignificant CBF decrease (−1.1%+−7.8%) with head flat, but found that individual responses did significantly change from baseline CBF (P<0.05).
Conclusions: We have validated DCS against Xenon-CT as a measure of local, microvascular CBF. We have also observed damaged cerebral autoregulation in patients after severe head injury which was confirmed by a large correlation between CBF and CPP during a simple intervention. Our results demonstrate the potential for DCS to provide continuous, non-invasive bedside monitoring of CBF for the purpose of CBF management and individualized care.
Changes in CBF and CPP during HOB manipulation.
778. A novel multimodal approach to track neural progenitor cells in vivo
A. Pendharkar1, A. De2, H. Wang2, X. Gaeta1, N. Wang1, J.Y. Chua1, R. Andres1, X. Chen2, S.S. Gambhir2 and R. Guzman1
1Department of Neurosurgery; 2Molecular Imaging Program at Stanford, Stanford University School of Medicine, Stanford, California, USA
Objectives: Stem cell transplantation represents a promising experimental therapeutic avenue for central nervous system disorders. Before these therapies can come to fruition, however, in depth studies must be conducted to assess stem cell survival and biodistribution, especially in developing minimally invasive intravascular delivery techniques.
In the present study, we have designed a multi-modality approach with which intravascular transplants can be studied for their biodistribution and survival. We have created a mouse neural progenitor cell line harboring a triple-fusion reporter gene encoding a luciferase, red fluorescent protein, and thymidine kinase PET reporter gene. To combine high sensitivity and spatial resolution, we have also included SPIO labeling for MR imaging to create a clinically applicable system to monitor neural stem cells after transplantation.
Methods: C17.2 mouse neural progenitor cells were transduced by a Lentivirus harboring a tri-fusion reporter gene containing a synthetic Renilla luciferase, monomeric RFP, and a truncated version of sr39 thymidine kinase (TK). Cells were also transfected with SPIO particles. Following transduction, RFP positive cells were selected using FACS.
Cells were analyzed for TK and Luciferase activity 48 h after FACS. Negative controls for both assays were performed using the parental C17.2 cells.
Adult nude mice were injected with either FACS isolated C17.2 cells positive for the triple-fusion reporter gene or parental C17.2 neural progenitor cells. 5 × 105 FACS-sorted or wild-type cells were transplanted into the right or left striatum, respectively. Control injections of the respective cells were also performed subcutaneously into the shoulders.
In vivo bioluminescence imaging and T2*-weighted MRI was performed 3 days following transplantation. FHBG-PET imaging was also performed.
Results: FACS analysis confirmed the presence of RFP, and thus the triple-fusion reporter gene in the C17.2 mouse neural progenitor cell line. In vitro assays also demonstrated luciferase and thymidine kinase activity in the RFP sorted cell line. Wild type cells did not exhibit any thymidine kinase and luciferase activity.
The in vivo data validated the activity of the triple-fusion reporter gene after transplant, both in subcutaneous and intracerebral locations. Figure 1 demonstrates specific luciferase signal emission from the bolus of transduced cells from the shoulder, as well as the brain after intracerebral transplantation. Wild type cells transplanted on the contralateral shoulder and striatum did not emit any signal above background levels. PET confirmed the presence of the transduced cells and activity of the triple-fusion gene in vivo. SPIO transfection resulted in a typical susceptibility artifact on T2 and T2* weighted MRI.
Conclusions: We describe a novel method for tracking cells after transplant. With this approach we obtain high spatial resolution from MRI and high sensitivity with PET. This method will enable studies of biodistribution and cell migration in both short and long term studies of CNS disorders.
References
1.
BullockMRPovlishshockJT. Guidelines for the management of severe traumatic brain injury. J Neurotrauma2007;24(1):S59–S64.
2.
HowellsTElfKJonesPRonne-EngstromEPiperINilsonPAndrewsPEnbaldP. Pressure reactivity as a guide in the treatment of cerebral perfusion pressure in patients with brain trauma. J. Neurosurg.2005;102(2):311–7.
3.
NarayananNLefflerCWDaleyML. Influence of hypercapnic vasodilation on cerebrovascular autoregulation and pial arteriolar bed resistance in piglets. J Appl Physiology2008 [e-pub ahead of print].
4.
BullockMRPovlishshockJT. Guidelines for the management of severe traumatic brain injury. J Neurotrauma2007;24(1):S59–64.
5.
ChambersIRJonesPALoTYMForsythRJFultonBAndrewsPJDMendelowADMinnsRA. Critical thresholds of intracranial pressure and cerebral perfusion pressure related to age in paediatric head injury. J Neurol Neurosurg Psychiatry2006;77:234–40.
6.
BoasDACampbellLEYodhAG. Scattering and imaging with diffusing temporal field correlations. Physical Review Letters,1995;75(9):1855–8.
