Functional localization and visualization of the subthalamic nucleus from microelectrode recordings acquired during DBS surgery with unsupervised machine learning.

Microelectrode recordings are a useful adjunctive method for subthalamic nucleus localization during deep brain stimulation surgery for Parkinson's disease. Attempts to quantitate and standardize this process, using single computational measures of neural activity, have been limited by variability in patient neurophysiology and recording conditions. Investigators have suggested that a multi-feature approach may be necessary for automated approaches to perform within acceptable clinical standards. We present a novel data visualization algorithm and several unique features that address these shortcomings. The algorithm extracts multiple computational features from the microelectrode neurophysiology and integrates them with tools from unsupervised machine learning. The resulting colour-coded map of neural activity reveals activity transitions that correspond to the anatomic boundaries of subcortical structures. Using these maps, a non-neurophysiologist is able to achieve sensitivities of 90% and 95% for STN entry and exit, respectively, to within 0.5 mm accuracy of the current gold standard. The accuracy of this technique is attributed to the multi-feature approach. This activity map can simplify and standardize the process of localizing the subthalamic nucleus (STN) for neurostimulation. Because this method does not require a stationary electrode for careful recording of unit activity for spike sorting, the length of the operation may be shortened.

High-frequency oscillations (>200 Hz) in the human non-parkinsonian subthalamic nucleus.

The human basal ganglia, and in particular the subthalamic nucleus (STN), can oscillate at surprisingly high frequencies, around 300 Hz [G. Foffani, A. Priori, M. Egidi, P. Rampini, F. Tamma, E. Caputo, K.A. Moxon, S. Cerutti, S. Barbieri, 300-Hz subthalamic oscillations in Parkinson's disease, Brain 126 (2003) 2153-2163]. It has been proposed that these oscillations could contribute to the mechanisms of action of deep brain stimulation (DBS) [G. Foffani, A. Priori, Deep brain stimulation in Parkinson's disease can mimic the 300 Hz subthalamic rhythm, Brain 129 (2006) E59]. However, the physiological role of high-frequency STN oscillations is questionable, because they have been observed only in patients with advanced Parkinson's disease and could therefore be secondary to the dopamine-depleted parkinsonian state. Here, we report high-frequency STN oscillations in the range of the 300-Hz rhythm during intraoperative microrecordings for DBS in an awake patient with focal dystonia as well as in a patient with essential tremor (ET). High-frequency STN oscillations are therefore not exclusively related to parkinsonian pathophysiology, but may represent a broader feature of human STN function.

Revision of deep brain stimulator for tremor. Technical note.

The treatment of essential tremor with thalamic deep brain stimulation (DBS) is considered to be more effective and to cause less morbidity than treatment with thalamotomy. Nonetheless, implantation of an indwelling electrode, connectors, and a generator is associated with specific types of morbidity. The authors describe three patients who required revision of their DBS systems due to lead breakage. The connector between the DBS electrode and the extension wire, which connects to the subclavicular pulse generator, was originally placed subcutaneously in the cervical region to decrease the risk of erosion through the scalp and to improve cosmesis. Three patients presented with fractured DBS electrodes that were located in the cervical region near the connector, necessitating reoperation with stereotactic retargeting and placement of a new intracranial electrode. At reoperation, the connectors were placed subgaleally over the parietal region. Management of these cases has led to modifications in the operative procedure designed to improve the durability of DBS systems. The authors recommend that surgeons avoid placing the connection between the DBS electrode and the extension wire in the cervical region because patient movement can cause microfractures in the electrode. Such microfractures require intracranial revision, which may be associated with a higher risk of morbidity than the initial operation. The authors also recommend considering prophylactic relocation of the connectors from the cervical area to the subgaleal parietal region to decrease the risk of future DBS electrode fracture, which would necessitate a more lengthy procedure to revise the intracranial electrode.

Cerebral blood flow and neurological outcome in the preterm infant.

UNLABELLED: Cerebral blood flow (CBF) studies have provided some insight into pathophysiological mechanisms of cerebral damage in newborn children; their value in predicting brain damage, however, remains elusive. The purpose of our study was to evaluate the role of CBF measurements in predicting developmental outcome in preterm neonates at 18 months. Preterm babies with a gestational age of less than 34 weeks and a birth weight of less than 1500 g (n = 71) were enrolled in the study. CBF was measured by the noninvasive intravenous 133Xe method on three different occasions. We classified our measurements into three groups: depending on the time when performed group 1: between 2 and 36 h (n = 52); group 2: between 36 and 108 h (n = 44); group 3: between 108 and 240 h (n = 41). At the age of 18 months neurodevelopment testing was performed according to the Bayley mental and motor scales. Surviving infants had a higher mean CBF over the three groups than non surviving children (15.2 +/- 3.5 ml/100 g brain tissue/min vs 13.0 +/- 2.1 ml/100 g brain tissue/min, P < 0.05). There was no correlation of CBF with mental or motor development in our study population in either of the three groups. CONCLUSION: In preterm infants basal CBF is higher in surviving than in non surviving infants, but there is no correlation of resting CBF and later neurological outcome.

Estimation of component and parameter distributions in spectral analysis.

A method is presented for estimating the distributions of the components and parameters determined with spectral analysis when it is applied to a single data set. The method uses bootstrap resampling to simulate the effect of noise on the computed spectrum and to correct for possible bias in the estimates. A number of bootstrap procedures are reviewed, and one is selected for application to the kinetic analysis of positron emission tomography dynamic studies. The technique is shown to require minimal assumptions about noise in the measurements, and its small sample properties are established through Monte-Carlo simulations. The advantages and limitations of spectral analysis with bootstrap resampling for deriving inferences for tracer kinetic modeling are illustrated through sample analyses of time-activity curves for [18F]fluorodeoxyglucose and [15O]-labeled water.

Regional heterogeneity and differential vulnerability of cerebral and spinal vascular CO2-responsiveness during graded haemorrhagic hypotension.

Regional inhomogeneity of cerebrovascular CO2-sensitivity as well as its changes at three different levels of standardized haemorrhagic hypotension were studied in ten distinct brain and spinal cord regions of anesthetized, ventilated cats. Regional cerebral blood flow was measured with radiolabelled microspheres in hypocapnic, normocapnic, and hypercapnic conditions, and CO2-responsiveness was determined from the equation of the slopes of the best fit regression lines to the obtained flow values. It was concluded that in normotensive, normoxic cats response of the cerebral and spinal vessels to PaCO2 alterations can be assigned to four major categories. The CO2-responsiveness of a brain region is not solely determined by the rate of its basal steady state blood flow: CO2-reactivity of the hypothalamus was significantly different from that of any other investigated regions with almost identical steady state flow values. Vulnerability of the cerebrovascular CO2-sensitivity during hypotension was different from region to region, with the vessels of the pons-medulla oblongata region being the most sensitive to haemorrhage. Reduced regional cerebral and spinal CO2-responsiveness during haemorrhage is not a consequence of a reduced L-arginine supply for nitric oxide generation since administration of an excess amount of the precursor L-arginine failed to restore the haemorrhage-induced reduction of regional CO2-sensitivity at the 60 mm Hg mean arterial pressure level.

Regional differences of cerebral blood flow in the preterm infant.

The purpose of our study was to evaluate the regional distribution of the resting cerebral blood flow (CBF) pattern in preterm neonates. Sixty-eight preterm babies with a gestational age of less than 34 weeks and a birth weight of less than 1500 g were enrolled into the study. The CBF was measured by the noninvasive intravenous 133Xenon method at three different times. Depending on the age we classified our measurements into three groups. Group 1: measurement between 2-36 h (n = 46). Group 2: measurement between 36-108 h (n = 39). Group 3: measurement between 108-240 h (n = 41). In all three groups CBF was significantly lower in the occipital region than in the frontal and parietal regions (group 1: frontal region 12.8 +/- 3.5 ml/100 g/min, parietal region 12.8 +/- 3.9 ml/100 mg/min, and occipital region 11.6 +/- 3.18 ml/100 g/min; group 2: frontal region 15.4 +/- 4.2 ml/100 g/min, parietal region 15.3 +/- 4.1 ml/100 g/min, and occipital region 13.4 +/- 3.5 ml/100 g/min; group 3: frontal region 14.6 +/- 3.6 ml/100 g/min, parietal region 14.6 +/- 3.2 ml/100 g/min, and occipital region 12.8 +/- 2.7 ml/100 g/min.). CBF did not differ between the left and the right hemispheres in either of the three measured regions. No gradient was found in infants between 108 h and 240 h of age with periventricular leukomalacia and periventricular haemorrhage. CONCLUSION. In preterm neonates the antero-posterior gradient of CBF is already present. Periventricular leukomalacia as well as periventricular haemorrhage may affect the regional regulation of CBF.

Cerebral blood flow, vascular resistance, and oxygen metabolism in acute brain trauma: redefining the role of cerebral perfusion pressure?

OBJECTIVES: To evaluate normal or high cerebral perfusion pressure in relation to cerebral blood flow and oxygen metabolism, as well as other multivariate cerebral hemodynamic and metabolic interrelationships, in acute brain trauma in humans. DESIGN: Prospective, observational study. SETTING: Neuroscience intensive care unit of a university hospital. PATIENTS: Adults (n = 66) with severe acute brain trauma (Glasgow Coma Scale scores from 4 to 8), undergoing multivariate physiologic studies involving cerebral perfusion pressure, cerebral blood flow, cerebral metabolic rate of oxygen consumption, total hemoglobin content, arterio-jugular oxygen content difference, and cerebral vascular resistance, along with other routine procedures. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Statistical analysis did not demonstrate any correlation between cerebral perfusion pressure and cerebral blood flow, between cerebral perfusion pressure and arterio-jugular oxygen content difference, and between cerebral perfusion pressure and cerebral metabolic rate of oxygen consumption, over a broad spectrum of perfusion pressures ranging from 60 to 130 mm Hg. In contrast, a significant negative correlation was found between cerebral vascular resistance and cerebral blood flow, where higher values of cerebral vascular resistance were associated with lower blood flow levels, and vice versa. CONCLUSIONS: In severe acute brain trauma, cerebral hemodynamic and oxygen metabolic variables are not necessarily correlated with normal or even high levels of cerebral perfusion pressure. Under these circumstances, cerebral vascular resistance (not perfusion pressure) is more closely correlated with different patterns of cerebral blood flow and metabolism.

Effect of non-steady-state perfusion on xenon-133 cerebral blood flow measurements: an analytical study.

Activation studies employing the noninvasive xenon-133 technique are widely used to investigate the cerebral circulation. Typical examples are the investigation of hemispheral specialization of higher cortical function with cognitive activation or the assessment of the hemodynamic reserve in occlusive cerebrovascular disease by CO2 inhalation. Traditionally, in studies using this technique, there is the requirement of a circulatory steady state during the measurement. Due to limitations in the duration of the stimulus or habituation to the stimulus, the basic assumption is often violated. In this study we investigated with the aid of a computer model to what extent blood flow measurement results are affected by non-steady-state blood flow. The findings indicate that cortical activation need not extend throughout the whole measurement to be detectable. Maintenance of activation for at least 5 min is sufficient for a successful measurement. In addition, the results show that the activation should be fully established when the measurement starts to achieve maximal sensitivity. Delay in activating the circulation will result in attenuated responses, especially if the stimulus is delayed beyond 2 min.

Estimated cerebral metabolic rate of oxygen in severely brain-injured patients: a valuable tool for clinical monitoring.

OBJECTIVE: To evaluate a novel parameter of global cerebral oxygen metabolism, the estimated cerebral metabolic rate of oxygen, for its clinical utility in monitoring acute, severely brain-injured patients. DESIGN: Prospective, observational study. SETTING: Neuroscience intensive care unit of a university hospital. PATIENTS: Sixty-six adults with acute brain trauma undergoing 133Xe regional cerebral blood flow and global cerebral oxygen metabolic studies, in conjunction with other routine monitoring techniques. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Estimated cerebral metabolic rate of oxygen was calculated as the product of PaCO2 and arterio-jugular oxygen content difference, and compared with the measured cerebral metabolic rate of oxygen (the product of mean regional cerebral blood flow and arterio-jugular oxygen content difference). Good correlations were found between the estimated and the true values (r2 = .56/.67, p < .0001) in 91 studies performed in 66 patients. CONCLUSIONS: These findings suggest that the estimated cerebral metabolic rate of oxygen is a useful substitute for the traditional cerebral metabolic rate of oxygen when cerebral blood flow information is not available. In addition, with this parameter, the patient's metabolic status can be assessed more frequently than with cerebral blood flow studies. These measurements do not require special instrumentation and can be done in any intensive care setting.

Cerebral blood flow in preterm infants affected by sex, mechanical ventilation, and intrauterine growth.

Sex differences in cerebral blood flow (CBF) values have been demonstrated in adults but not in newborns. This study evaluated the influence of sex, intrauterine growth, and need of mechanical ventilation on resting cerebral blood flow in preterm neonates. Sixty-eight preterm infants with gestational ages of less than 34 weeks and birth weights of less than 1,500 gm were enrolled into the study. Cerebral blood flow was measured by the noninvasive intravenous xenon 133 method 3 times. Measurements were classified into 3 groups: group 1: measurement at 2-36 hours (n = 46); group 2: measurement at 36-108 hours (n = 39); and group 3: measurement at 108-240 hours (n = 41). In all 3 groups, the CBF in girls was significantly lower than in boys (group 1: 11.5 +/- 2.8 ml/100 gm/min vs 14.0 +/- 4.1 ml/100 gm/min; group 2: 13.4 +/- 2.9 ml/100 gm/min vs 16.3 +/- 4.3 ml/100 gm/min; group 3: 12.9 +/- 3.2 ml/100 gm/min vs 15.3 +/- 3.1 ml/100 gm/min). In group 1, the CBF in neonates requiring mechanical ventilation was significantly lower (P < .05) than in patients who were spontaneously breathing (11.5 +/- 3.7 ml/100 gm/min vs 14.2 +/- 3.1 ml/100 gm/min), and the CBF in neonates who were too small for gestational age was significantly higher (P < .005) than in children with appropriate intrauterine growth (16.1 +/- 4.1 ml/100 gm/min vs 11.5 +/- 2.6 ml/100 gm/min). It is concluded that in preterm neonates CBF is substantially affected by sex, intrauterine growth retardation, and the need of mechanical ventilation.

Cerebral lactate-oxygen index in acute brain injury with acute anemia: assessment of false versus true ischemia.

OBJECTIVE: To evaluate the occurrence of global cerebral ischemia in acute brain trauma with acute anemia by combined measurements of cerebral hemodynamics, oxygenation, and lactate production. DESIGN: Prospective, intervention study. SETTING: Neuroscience intensive care unit of a university hospital. PATIENTS: Adults (n = 22) with severe acute brain trauma (Glasgow Coma Scores ranging from 4 to 8), undergoing frequent serial measurements of total hemoglobin content, jugular oxyhemoglobin saturation, arteriojugular oxygen content difference, arteriojugular lactate concentration difference, lactate-oxygen index, and cerebral blood flow, along with other routine procedures. MEASUREMENTS AND MAIN RESULTS: Acute anemia (disclosed by a total hemoglobin content of < 11 g/dL in at least three measurements) was found in 19 (86%) of 22 patients. In 211 serial multivariate physiologic observations, only one (0.4%) disclosed abnormally negative arteriojugular lactate difference consistent with global cerebral ischemia. However, in 18 (8.5%) studies in seven (31.8%) patients, acute anemia resulted in markedly decreased values of arteriojugular oxygen content difference. The latter, in turn, yielded abnormally high values of lactate-oxygen index despite normal cerebral lactate production (arteriojugular lactate difference) and oxygenation (jugular oxyhemoglobin saturation). CONCLUSIONS: In acute brain injury with acute anemia, global cerebral ischemia is a rare finding. However, false cerebral ischemia may be frequently found, if assessed by the lactate-oxygen index, because the denominator of the index (the arteriojugular oxygen content difference) frequently decreases as a function of decreasing hemoglobin, thus yielding false calculated ischemic high values for lactate-oxygen index despite normal cerebral oxygenation and lactate production.

Differential effects of mood on cortical cerebral blood flow: a 133xenon clearance study.

Studies of healthy and clinical populations have suggested valence-specific cortical and subcortical neural systems regulating emotions. In a study of 12 normal volunteers, the 133xenon clearance method for measuring regional cerebral blood flow (CBF) was used to study the effects of experimentally controlled mood states on regional brain activity within superficial cortex. CBF was measured with 254 detectors and bolus infusion during a happy mood induction task, a sad mood induction task, a sex differentiation task, and a resting baseline condition. CBF increased during sad and decreased during happy mood induction, relative to the activated (sex differentiation) and the nonactivated (resting) nonemotional control conditions. Increased CBF during sad mood induction was correlated with greater negative mood changes. Conversely, increased CBF was associated with a stronger subjective experience of positive affect during happy mood induction. This suggests that cortical arousal may serve to intensify the conscious experience of emotion. Heart rate accelerated during happy and sad mood induction and during sex differentiation relative to a pretask baseline condition. Some regional specificity of effects was also observed. The occipital temporal region showed higher overall CBF during sad mood induction than during happy mood induction. The only region that showed specific lateralized changes in CBF which differentiated sad from happy states was the frontal pole, with left CBF being higher during sad and lower during happy mood induction relative to right CBF. For sad mood induction, there were significant regional differences among correlations between CBF and self-ratings. These were attributable to higher negative correlations (i.e., higher CBF correlates with negative self-rating) in midtemporal, occipital temporal, and postcentral regions. These correlations did not vary across the 15 regions for happy mood induction. For sad mood induction, heart rate correlated positively with CBF increase and with negative affect. Correlations were opposite for happy mood induction. The results suggest high cortical and autonomic arousal during negative/sad mood and low cortical and high autonomic arousal during positive/happy mood. They underscore the value of integrating emotional experience with physiologic measures in neuroimaging activation studies.

Cerebral blood flow in schizophrenia: effects of memory processing on regional activation.

Regional cerebral blood flow (rCBF) was measured with the 133Xenon clearance technique and a high resolution (254 detectors) scanner during the performance of a verbal and a facial memory task in 18 patients with schizophrenia and 18 sociodemographically matched controls. Patients and controls had comparable resting rCBF, but differed in global and hemispheric rCBF changes induced by the memory tasks. Patients had less global increase, which was relatively higher in the left hemisphere, and this was more pronounced for the verbal task. Although controls showed appropriate laterality changes (L > R for verbal and R > L for facial memory) in the midtemporal region, patients failed to show such a focal pattern. They did not show appropriate laterality change in the midtemporal region, but instead showed such changes in other regions. Patients showed greatest impairment in specificity of verbal recognition performance, and this correlated with severity of hallucinations and delusions. This supports a model of left temporal lobe dysfunction in schizophrenia.

Effects of memory processing on regional brain activation: cerebral blood flow in normal subjects.

The role of the temporal lobe in memory has been implicated in lesion studies, which have also suggested the hypothesis of greater left hemispheric involvement in verbal, and right hemispheric involvement in facial memory. We tested these hypotheses in a sample of 27 normal right-handed subjects using the 133Xenon clearance method for measuring cerebral blood flow (CBF). The CBF was measured during resting baseline, word recognition, and face recognition conditions in counterbalanced order. CBF increased during recognition compared to baseline, and for the midtemporal lobe this increase was asymmetric to the left hemisphere for words and to the right for faces. While overall CBF levels and task related increases in CBF were uncorrelated either with performance or with delta performance (excess performance relative to basal memory abilities), laterality of task-related CBF correlated with both performance indices, showing regional specificity of correlations. This neurobehavioral probe paradigm can be applied in the study of neural substrates of normal and disturbed memory.

Cerebral blood flow and oxygen consumption in acute brain injury with acute anemia: an alternative for the cerebral metabolic rate of oxygen consumption?

OBJECTIVE: To comparatively evaluate cerebral metabolic rate of oxygen consumption and a modification of it, cerebral consumption of oxygen, in patients with acute brain injury with acute anemia. DESIGN: Prospective, observational study. SETTING: Neuroscience intensive care unit (ICU) of a university hospital. PATIENTS: Adults (n = 62) with acute brain trauma, undergoing serial 133xenon studies of regional cerebral blood flow and global cerebral oxygen metabolism, along with other routine monitoring techniques. MEASUREMENTS AND MAIN RESULTS: In 173 combined studies of blood flow and oxygen metabolism, in the presence of spontaneous decreases in hemoglobin, cerebral metabolic rate of oxygen consumption and cerebral consumption of oxygen were comparatively evaluated in three groups with different hemoglobin levels. Cerebral metabolic rate of oxygen consumption was calculated as the product of averaged regional cerebral blood flow and arterio-jugular oxygen content difference, while cerebral consumption of oxygen was calculated as the product of averaged regional cerebral blood flow and the arterio-jugular oxyhemoglobin saturation difference, i.e., cerebral extraction of oxygen. Results indicated that a decrease of hemoglobin content is paralleled by a decrease in cerebral metabolic rate of oxygen consumption, even though the level of consciousness (coma score) is essentially unchanged across three hemoglobin groups. On the other hand, cerebral consumption of oxygen does not follow the decrease in hemoglobin and cerebral metabolic rate of oxygen consumption, thus demonstrating better stability to changing hemoglobin content. The low cerebral metabolic rate of oxygen consumption is due to a decrease in arterio-jugular oxygen content difference in anemia, while the cerebral extraction of oxygen does not follow the trend of the arterio-jugular oxygen content difference. CONCLUSIONS: In acute brain trauma with acute anemia, calculated arterio-jugular oxygen content difference and cerebral metabolic rate of oxygen consumption tend to be progressively lower, depending on the extent of anemia, which is in disagreement with coma scores. These changes in hemoglobin tend to have an inverse influence on cerebral consumption of oxygen, which, therefore, constitutes an alternative and independent measure of cerebral oxygen and independent measure of cerebral oxygen consumption under these limiting circumstances.

Cerebral oxygenation monitoring.

OBJECTIVES: To discuss theoretical and practical aspects of cerebral oxygenation, from isolated observational measurements to continuous interventional monitoring. DATA SOURCES: Relevant articles from the pertinent literature, as well as a multivariate physiologic diagram developed in this article. STUDY SELECTION: Theoretical, experimental, and clinical information that further clarifies the physiologic relevance of cerebral oxygenation. DATA EXTRACTION: All basic concepts of cerebral hemometabolism were used, up to cerebral hemodynamic reserve, and were interrelated accordingly. DATA SYNTHESIS: Cerebral perfusion pressure alone does not allow global cerebral hemometabolic optimization. The same situation is true for cerebral blood flow. The reason for this limitation is because these variables lack metabolic information. Cerebral arteriovenous differences do allow global cerebral hemometabolic optimization, because they reflect the exchange between the capillary and the tissue. In addition, cerebral hemodynamic reserve allows quantification of the cerebral microcirculatory tolerance to increases in intracranial "tightness" (decreases in intracranial compliance). CONCLUSIONS: Multivariate optimization of global cerebral oxygenation, primarily based on global cerebral oxygen delivery and extraction, may strongly affect outcome in a variety of predominantly diffuse, acute intracranial disorders. Cerebral hemodynamic reserve may be further explored in experimental and clinical areas, involving not just acute brain trauma.

Signal analysis of noninvasive Xenon-133 cerebral blood flow measurements.

An anatomical model in conjunction with experimentally determined absorption data provides a framework to simulate signals as obtained from the noninvasive Xenon-133 cerebral blood flow technique. The contribution of individual tissue compartments to the total signal as well as the effect on the computed results were investigated under normal conditions. The introduction of physiological abnormalities into the model allowed the determination of sensitivity of the technique with respect to size, position, and perfusion level of the lesion. In addition, effects of cross-talk between hemispheres and signal overlap of adjacent detectors were quantified. It was found that the change of externally measured blood flow is proportional to the decrement/increment of flow in the lesion. Contrary to earlier reports, the effects of cross-talk and signal overlap were not found to be serious limitations in identifying lesions.

Relationship of early cerebral blood flow and metabolism to outcome in acute head injury.

Cerebral blood flow (CBF) measurements were obtained acutely in 96 comatose patients with closed head injury, using the intravenous 133Xe technique. Arteriojugular venous oxygen differences and cerebral metabolic rate for oxygen (CMRO2) were determined in a subgroup of 66 patients. The relationship between each of these variables and outcome at 6 months was analyzed, using the Glasgow Outcome Scale. The CMRO2 was significantly depressed in patients who subsequently died or remained in a vegetative state, whereas higher values were obtained in patients who later regained consciousness. Although CBF was not predictive of outcome in the total sample, omission of patients with acute hyperemia resulted in a significant relationship that paralleled the metabolic findings. Follow-up studies in the survivors revealed a correlation between CBF and degree of functional recovery, the lowest blood flows being obtained among patients with severe disability. Age, initial Glasgow Coma Scale score, and occurrence of intracranial hypertension were each found to be predictive of outcome, thus confirming previous reports. When these variables were combined with CMRO2 in a logistic regression analysis, the probability of recovery was correctly predicted in 82% of the cases. The CMRO2 was relatively independent of the other prognostic indicators and, next to age, contributed most to the prediction.

The effect of hematocrit and systolic blood pressure on cerebral blood flow in newborn infants.

The effects of hematocrit and systolic blood pressure on cerebral blood flow were measured in 15 stable, low birth weight babies. CBF was measured with a modification of the xenon-133 (133Xe) clearance technique, which uses an intravenous bolus of 133Xe, an external chest detector to estimate arterial 133Xe concentration, eight external cranial detectors to measure cephalic 133Xe clearance curves, and a two-compartmental analysis of the cephalic 133Xe clearance curves to estimate CBF. There was a significant inverse correlation between hematocrit and CBF, presumably due to alterations in arterial oxygen content and blood viscosity. Newborn CBF varied independently of systolic blood pressure between 60 and 84 mm Hg, suggesting an intact cerebrovascular autoregulatory mechanism. These results indicate that at least two of the factors that affect newborn animal CBF are operational in human newborns and may have important clinical implications.