J Neurophysiol 114: 818 – 824, 2015. First published June 10, 2015; doi:10.1152/jn.00353.2015.
CALL FOR PAPERS
Neurological Disease and Autonomic Dysfunction
Impaired cerebral autoregulation and brain injury in newborns with hypoxic-ischemic encephalopathy treated with hypothermia An N. Massaro,1,2,6 R. B. Govindan,2,6 Gilbert Vezina,3,6 Taeun Chang,4,6 Nickie N. Andescavage,1,6 Yunfei Wang,5,6 Tareq Al-Shargabi,2 Marina Metzler,2 Kari Harris,4 and Adre J. du Plessis2,6 1
Division of Neonatology, Children’s National Health System, Washington, District of Columbia; 2Division of Fetal and Transitional Medicine, Children’s National Health System, Washington, District of Columbia; 3Division of Diagnostic Imaging and Radiology, Children’s National Health System, Washington, District of Columbia; 4Department of Neurology, Children’s National Health System, Washington, District of Columbia; 5Department of Biostatistics and Study Methodology, Children’s National Health System, Washington, District of Columbia; and 6Department of Pediatrics, The George Washington University School of Medicine, Washington, District of Columbia Submitted 8 April 2015; accepted in final form 9 June 2015
Massaro AN, Govindan RB, Vezina G, Chang T, Andescavage NN, Wang Y, Al-Shargabi T, Metzler M, Harris K, du Plessis AJ. Impaired cerebral autoregulation and brain injury in newborns with hypoxic-ischemic encephalopathy treated with hypothermia. J Neurophysiol 114: 818 – 824, 2015. First published 10 June 2015; doi:10.1152/jn.00353.2015.—Impaired cerebral autoregulation may contribute to secondary injury in newborns with hypoxic-ischemic encephalopathy (HIE). Continuous, noninvasive assessment of cerebral pressure autoregulation can be achieved with bedside nearinfrared spectroscopy (NIRS) and systemic mean arterial blood pressure (MAP) monitoring. This study aimed to evaluate whether impaired cerebral autoregulation measured by NIRS-MAP monitoring during therapeutic hypothermia and rewarming relates to outcome in 36 newborns with HIE. Spectral coherence analysis between NIRS and MAP was used to quantify changes in the duration [pressure passivity index (PPI)] and magnitude (gain) of cerebral autoregulatory impairment. Higher PPI in both cerebral hemispheres and gain in the right hemisphere were associated with neonatal adverse outcomes [death or detectable brain injury by magnetic resonance imaging (MRI), P ⬍ 0.001]. NIRS-MAP monitoring of cerebral autoregulation can provide an ongoing physiological biomarker that may help direct care in perinatal brain injury. hypoxic-ischemic encephalopathy; cerebral blood flow; magnetic resonance imaging; newborn PERINATAL HYPOXIC-ISCHEMIC
encephalopathy (HIE) is an important cause of neonatal mortality and long-term neurodisability (Dilenge et al. 2001; Liu et al. 2012). While therapeutic hypothermia improves outcome in HIE and is the current standard of care, a significant proportion of treated infants continue to suffer adverse outcomes after therapy (Gluckman et al. 2005; Shankaran et al. 2005). Adjuvant neuroprotective interventions are needed to further improve outcomes for this devastating condition. Unfortunately, methods to acutely identify infants with inadequate response to hypothermia and evolving brain injury are limited in the neonatal intensive care
Address for reprint requests and other correspondence: A. N. Massaro, Div. of Neonatology, Children’s National Medical Center, 111 Michigan Ave. NW, Washington, DC 20010 (e-mail: [email protected]). 818
unit (NICU). Clinicians often rely on magnetic resonance imaging (MRI) performed after hypothermia to determine the presence and extent of brain injury that has occurred. This approach does not enable therapeutic interventions aimed to prevent injury progression. Thus bedside methods to monitor risk for evolving brain injury are needed to direct clinical care and rational therapeutic interventions. Ongoing assessment of cerebral pressure autoregulation integrity can provide an assessment of the physiological response to hypoxia and/or ischemia, offering a window into evolving brain injury. Impaired cerebral pressure autoregulation has been implicated in the pathogenesis of secondary injury in HIE, as cerebral vasoparalysis following a hypoxic and/or ischemic insult has been well documented in both animal and human studies (Meek et al. 1999; Pryds et al. 1990; Short et al. 1994). Cerebral pressure passivity, or the inability to regulate cerebral blood flow (CBF) in the setting of changes in systemic blood pressure, can lead to uncoupling of CBF and cerebral energy metabolism that can contribute to secondary energy failure and structural brain injury in HIE. We previously described methods to monitor cerebral pressure autoregulation by quantifying frequency domain spectral coherence between near-infrared spectroscopy (NIRS)-derived cerebral oxyhemoglobin-deoxyhemoglobin (HbD), a surrogate measure of CBF (Tsuji et al. 1998), and systemic mean arterial blood pressure (MAP) (Govindan et al. 2014; Soul et al. 2007). Spectral coherence is a technique that allows quantification of the relationship between changes in MAP and changes in CBF (or HbD), with pressure-passive CBF indicated by increased coherence between the two physiological signals. This approach allows for quantification of both the duration [pressure passivity index (PPI)] and magnitude (gain) of cerebral pressure passivity. While cerebral pressure passivity assessed by this approach has been proposed as a risk factor for brain injury in the preterm infant (O’Leary et al. 2009), its significance in term newborns with HIE is unknown. This study aimed to investigate whether quantified changes of the duration and magnitude of cerebral pressure passivity during therapeutic hypothermia are associated with adverse neonatal outcomes
0022-3077/15 Copyright © 2015 the American Physiological Society
www.jn.org
IMPAIRED CEREBRAL AUTOREGULATION IN HIE
(i.e., death or structural brain injury) in newborns with HIE. We hypothesized that neonates who die or have moderate to severe brain injury on MRI will have worse autoregulatory function during therapeutic hypothermia than those with only mild or no brain injuries. MATERIALS AND METHODS
Study population. Newborns with HIE referred for therapeutic hypothermia to a regional NICU in a freestanding children’s hospital were prospectively enrolled in this observational study. Infants met institutional criteria for hypothermia: 1) ⬎35 wk gestational age; 2) ⬎1,800 g at birth; 3) physiological evidence of perinatal asphyxia with either demonstrated severe metabolic acidosis, with pH ⱕ 7 or base deficit ⱖ 16, or milder acidosis, with pH 7.01–7.15 or base deficit 10 –16, accompanied by Apgar score ⱕ 5 or positive-pressure ventilation at 10 min and a sentinel perinatal event; and 4) signs of moderate to severe clinical encephalopathy and/or seizures. All infants were treated with whole-body hypothermia according to the National Institute of Child Health and Human Development Neonatal Research Network protocol (Shankaran et al. 2005). Briefly, infants were cooled to 33.5°C with an infant-size cooling blanket servoregulated to esophageal temperature (Blanketrol-II, Cincinnati Subzero, Cincinnati, OH) for 72 h, followed by gradual rewarming by 0.5°C/h over 6 h. All infants undergo continuous video EEG monitoring during cooling and through the rewarming period as standard clinical care. Sedation with low-dose morphine (starting 0.05 mg/kg every 4 h) is titrated to prevent shivering and other overt signs of pain or discomfort. Blood pressure is maintained at goal MAP ⬎40 mmHg with use of dopamine as first-line vasoactive therapy, followed by the addition of epinephrine and systemic hydrocortisone as needed. The study was approved by the Children’s National Health System Institutional Review Board and was conducted in accordance with the ethical standards of the Children’s National Office for the Protection of Human Subjects, and informed consent was obtained from a parent of each participant for physiological monitoring. Data collection. Demographic and clinical data were collected from birth hospital and study site medical records. Presenting characteristics were recorded, including umbilical cord or arterial blood gas obtained in the first hour of life, Apgar scores, and the initial grade of encephalopathy as classified by modified Sarnat criteria (Sarnat and Sarnat 1976; Shankaran et al. 2005). Data from NICU hospitalizations were also collected, including use of vasoactive therapy, presence of electrographic seizures, use of sedatives, and serial measurements of core body temperature, hemoglobin, and partial arterial pressure of carbon dioxide (PaCO2) levels. Continuous recordings of NIRS cerebral oximetry (NIRO 200, Hamamatsu Photonics, Hamamatsu, Japan) and MAP from an indwelling arterial line (Philips IntelliVue MP70) were recorded in a time-locked manner at a rate of 5 Hz and upsampled to 1 kHz with custom software developed in LabVIEW (National Instruments). NIRS monitoring was performed bilaterally, with optodes placed over the right and left fronto-temporal regions. After parental consent was obtained, monitoring was initiated within 24 h of cooling initiation and continued for at least 6 h after the completion of rewarming. Physiological signal processing. Data were processed and analyzed with a power spectral estimation approach as previously described (Govindan et al. 2014). Briefly, after using an in-house pipeline for automated artifact rejection (Govindan et al. 2014), we divided our continuous data into 10-min epochs. We then quantified the spectral coherence between MAP and HbD, using a previously described mathematical algorithm (Halliday et al. 1995). Coherence is a normalized quantity and takes on a value of 1 in cases of a perfect linear relationship between the signals and 0 in the case of independence. The statistical significance of the coherence was assessed with a mathematically derived threshold of 0.384, defined by 1 ⫺ (1 ⫺ 1/(M ⫺ 1) ␣) , where M is the number of epochs involved in the coherence
819
estimation, which was 20 in our case, and ␣ set to 0.999 (Govindan et al. 2014; Halliday et al. 1995). Based on the knowledge that impulse response time of cerebral autoregulation is expected to operate at 5–20 s, pressure passivity was defined as coherence greater than the threshold in the frequency band of 0.05– 0.25 Hz (O’Leary et al. 2009). The PPI was then calculated as the total percentage of pressurepassive 10-min epochs per 6 h of recording. We also calculated transfer gain (Halliday et al. 1995) within the pressure-passive epochs and averaged over 6 h of recording to match the PPI calculation. Gain reflects the change in HbD caused by 1-mmHg change in MAP. Gain function provides a measure of the magnitude of pressure passivity. All processing was done off-line with MATLAB software (MathWorks, Natick, MA). Magnetic resonance imaging. MRI was performed according to our institutional clinical protocol based on established guidelines (2014). Surviving infants underwent routine MRI at target age 3–5 days (to evaluate early diffusion changes) and 10 –12 days (to evaluate extent of lesions) of life on a 3-T scanner (Discovery MR750, GE Healthcare, Milwaukee, WI) using a 32-channel receive-only head coil (MR Instruments, Minneapolis, MN). Standard anatomical sequences included 3D T1-weighted spoiled gradient recalled, double-acquisition axial FSE T2 proton density, axial T2 propeller (in cases of patient motion), axial T2-star-weighted susceptibility imaging, coronal T1 FLAIR propeller, and axial 30-direction DTI with generation of ADC maps off-line. Images were scored according to a validated scoring system that assigns a score ranging from 0 to 4 for the degree of basal ganglia/thalamic (BG) injury and from 0 to 5 for the degree of watershed cortical (WS) injury (Barkovich et al. 1998) by an experienced neuroradiologist (G. Vezina) masked to clinical and physiological data. If an infant had more than one MRI, the higher injury score was used for analysis. Adverse outcome was defined as death during hospitalization or moderate to severe brain injury by MRI (BG score ⱖ 3 or WS score ⱖ 4). Statistical analysis. Descriptive statistics included means (standard deviations) and medians (ranges) for continuous parametric and nonparametric variables, respectively, as well as counts (percentages) for categorical data. Generalized estimating equations (GEEs) were used to evaluate the relationship between repeated measures of PPI/gain over time and outcome classification adjusting for baseline and clinical covariates including birth weight, gestational age, sex, 5-min Apgar score, presenting pH, severe encephalopathy at presentation, time from birth, presence of electrographic seizures, use of vasoactives, sedation, temperature, hemoglobin, and PaCO2 level. Models were developed for right and left PPI and gain values to evaluate for laterality of relationships. A log link function for expected values of PPI/gain was specified in the models in order to account for data that were not normally distributed. Secondarily, to explore the predictive ability of PPI and gain to distinguish infants with adverse outcomes, prediction models were developed with logistic regression. For these analyses, a summary value of average PPI or gain over the entire period of recording was used for each individual subject. Performance characteristics of the models were evaluated with receiver operating curve analyses, where an area under the curve (AUC) of 1 represents high diagnostic accuracy, in contrast to a value of 0.5 that signifies poor discrimination of outcome groups. Because of sample size limitations, these secondary analyses were not adjusted for covariates. Statistical analysis was performed with SAS 9.3 (SAS Institute, Cary, NC). RESULTS
Description of study population. Of the 59 eligible patients with moderate to severe HIE admitted during the study period (November 2012–July 2014), 9 infants were not enrolled because of extremis presentation and death prior to consent (n ⫽ 1), parental refusal (n ⫽ 3), or inability to obtain consent
J Neurophysiol • doi:10.1152/jn.00353.2015 • www.jn.org
820
IMPAIRED CEREBRAL AUTOREGULATION IN HIE
Table 1. Characteristics of study population Favorable Outcome (n ⫽ 20)
Adverse Outcome (n ⫽ 16)
P Value
38.6 ⫾ 1.7 3.5 ⫾ 1.0 12 (60) 6.91 (6.50–7.25)
39.2 ⫾ 1.3 2.9 ⫾ 0.3 46 (37.5) 6.9 (6.65–7.14)
0.294 0.016 0.180 0.814
Gestational age, wk (mean ⫾ SD) Birth weight, kg (mean ⫾ SD) Sex, n males (%) Initial pH Apgars 1 min 5 min 10 min Encephalopathy grade, n (%) Moderate Severe EEG seizures, n (%) Received vasoactives, n (%) Dopamine Epinephrine Age monitoring initiated, h
1 (0–9) 4 (1–8) 6 (2–8)†
1 (0–8)* 2 (0–9)* 3 (0–7)‡
16 (80) 4 (20) 2 (10)
0.479 0.023 0.059 0.180
10 (62.5) 6 (37.5) 8 (50)
13 (65) 2 (10) 13.4 (5.9–23.2)
0.008
11 (69) 5 (31) 13.3 (7.4–25.1)
0.983 0.109 0.987
Data presented as median (range) except where indicated. *Documented for 15/16 patients; †documented for 17/20 patients; ‡documented for 13/16 patients.
because of parental (n ⫽ 2) or investigator (n ⫽ 3) unavailability. Of the 50 patients enrolled for physiological monitoring, 14 infants did not have indwelling arterial catheters for continuous blood pressure monitoring. Thus data from 36 patients with complete MAP and NIRS data were included in the study. The adverse outcome group comprised seven infants who died (because of withdrawal of medical care based on poor neurological prognosis) and nine infants with moderate/ severe MRI brain injury (severe BG injury n ⫽ 5, severe WS injury n ⫽ 4). Infants in the adverse outcome group had lower birth weight, lower Apgar score at 5 min, and higher frequency of electrographic seizures compared with the favorable outcome group (P ⬍ 0.05). Otherwise, baseline characteristics were similar between groups (Table 1). The majority of infants received vasoactive medications (dopamine n ⫽ 23, 67%; epinephrine n ⫽ 7, 19%) and sedatives (morphine n ⫽ 31, 86%; midazolam n ⫽ 12, 33%).
MAP over time did not differentiate the two outcome groups (Fig. 1). Phenobarbital was given for seizures in 12 (33%) patients. Frequency of these medical therapies did not differ between outcome groups (P ⬎ 0.05). Temperature and laboratory data monitored through hypothermia and rewarming are presented in Table 2. Median temperatures, PaCO2, and hemoglobin levels did not differ between groups (P ⬎ 0.05). Continuous NIRS/MAP recordings were initiated at a median age of 13 (range 6 –25) h of life. The median duration of recording was 73 (range 5–91) h. Early MRI was performed in 24 infants at a median age of 4 (range 3– 6) days of life, while a later MRI (median 10, range 7–16 days) was performed in 25 infants. Seventeen infants had two scans performed. Of the nine infants with moderate/severe MRI brain injury in the adverse outcome group, the majority (n ⫽ 8, 89%) had severe BG injury while only one infant had severe WS injury. The favorable outcome group comprised 14
60 Favorable Group Adverse Group 55
Fig. 1. Median systemic mean arterial blood pressure (MAP) over time in infants with hypoxicischemic encephalopathy (HIE) who died or had moderate to severe MRI injury (solid line) compared with survivors with no or mild MRI injury (dashed line). Bars represent interquartile range.
MAP (mmHg)
50
45
40
35
30
0
10
20
30
40
50
60
70
80
90
100
Time Since Birth (hours) Hypothermia
J Neurophysiol • doi:10.1152/jn.00353.2015 • www.jn.org
Rewarming
Normothermia
110
IMPAIRED CEREBRAL AUTOREGULATION IN HIE
Table 2. Temperature and laboratory data Favorable Outcome Adverse Outcome (n ⫽ 20) (n ⫽ 16) Temperature during hypothermia, °C Min Max Average Temperature after rewarming, °C Min Max Average PaCO2, mmHg Min Max Average Hemoglobin, mg/dl Min Max Average
31.1 (27.3–32.7) 35.1 (33.7–36.7) 33 (32.5–33.5)
31.2 (29.0–32.4) 34.3 (33.4–36.4) 33.1 (32.6–33.4)
36.5 (34.7–36.8) 36.7 (36.3–37.2) 36.6 (36.1–36.9)
36.5 (35.2–36.9) 36.4 (36.2–37) 36.5 (35.8–36.9)
29.7 (14.2–45.5) 56.9 (46.6–75.2) 43.9 (38.3–52.1)
27.0 (12.0–52.5) 58.1 (33.0–104.6) 44.4 (24.6–63.1)
12.6 (9.3–17.0) 16.2 (12.7–19.3) 15.0 (11.7–17.8)
13.7 (7.2–18.1) 16.5 (12.6–22.2) 15.0 (11.0–19.5)
Data presented as median (range). PaCO2, partial arterial pressure of carbon dioxide. Differences between groups were not significant (P ⬎ 0.05).
infants with normal MRI and 6 infants with mild BG injury (score ⱕ 2). There was no laterality to injury pattern in any of the infants. Pressure passivity index. A total of 442 observations (i.e., PPI calculated in a 6-h window) were included in the PPI analyses for each cerebral hemisphere. Unadjusted median PPI values over time are shown in Fig. 2. After adjusting for baseline and clinical covariates, higher PPIs measured from both the left (LH) and right (RH) hemispheres were significantly associated with adverse outcome (P ⬍ 0.01). There was no significant time effect on these relationships, and, similarly, no interaction was detected between time and outcome group. Regression model results are summarized in Table 3. HbD-MAP spectral gain. Given that gain function can only be calculated within pressure-passive epochs, fewer observations (LH, n ⫽ 332; RH, n ⫽ 334) were included in the gain 0.9
A
0.8 0.7
analyses. While gain in the RH was significantly higher in the adverse outcome group (P ⬍ 0.001), gain from the LH was not significantly associated with outcome (P ⬎ 0.05) after adjusting for covariates. Similar to PPI, there was no significant time effect on the relationship between gain and outcome. Regression model results are included in Table 3. Secondary PPI/gain predictive models. Only average PPI (over the recording period) measured from the LH was significantly associated with adverse outcome in the logistic regression analyses (AUC ⫽ 0.7193, 95% CI: 0.5430 – 0.896, P ⫽ 0.034). There was a trend toward association between PPI from the RH and adverse outcome (AUC 0.664, 95% CI: 0.480 – 0.849, P ⫽ 0.082), while gains from both the LH and RH were not significantly predictive of adverse outcome (P ⬎ 0.05). A probability cutoff point of 0.378 for average PPI value measured from the LH had 87% sensitivity and 58% specificity for prediction of adverse outcome. DISCUSSION
We demonstrate that measurements of spectral coherence between MAP and NIRS HbD, reflecting impaired cerebral autoregulation, differ between newborns with HIE who die or have moderate to severe MRI brain injury and survivors with normal to mild injury. Specifically, infants with adverse outcomes exhibited higher PPI and gain, indicating longer duration and higher magnitude of cerebral pressure passivity following hypoxia-ischemia. Monitoring of blood pressure alone did not differentiate outcome groups. These data support the notion that quantifying cerebral autoregulation integrity can provide a viable physiological biomarker of brain injury. A bedside tool that can assess evolving cerebral injury is needed to individualize therapy in newborns with HIE. Such a tool would enable the timely identification of patients for whom hypothermia alone is providing inadequate neuroprotection and could be used to direct the rational introduction of adjuvant therapies. Continuous real-time assessment of spectral coher-
Favorable Group Adverse Group
0.6
PPI-LH
821
0.5 0.4 0.3 0.2 0.1 0 0.9
Fig. 2. Median pressure passivity index (PPI) over time in HIE infants with adverse (death or moderate to severe MRI injury, solid line) vs. favorable (survivors with no or mild MRI injury, dashed line) outcome. Bars represent interquartile range. Measurements from the left (LH; A) and right (RH; B) hemispheres are shown.
B
0.8 0.7
PPI-RH
0.6 0.5 0.4 0.3 0.2 0.1 0 0
10
20
30
40
50
60
70
80
90
100
110
Time Since Birth (hours) Hypothermia
Rewarming
Normothermia
J Neurophysiol • doi:10.1152/jn.00353.2015 • www.jn.org
822
IMPAIRED CEREBRAL AUTOREGULATION IN HIE
Table 3. GEE model results B
SE
95% CI
P Value
Other Significant Covariates
PPI-left PPI-right
0.92 1.24
0.28 0.25
0.37–1.48 0.75–1.74
0.001 ⬍0.001
Gain-left Gain-right
0.11 1.64
0.12 0.41
⫺0.11–0.34 0.84–2.44
0.322 ⬍0.001
BW, GA, severe HIE, initial pH, 5-min Apgar, EEG Sz, Hb, PaCO2 BW, GA, severe HIE, initial pH, 5-min Apgar, EEG Sz, Hb, PaCO2, sex, vasopressor use PaCO2, vasopressor use Severe HIE, 5-min Agpar, EEG Sz, PaCO2, sex, vasopressor use
B, outcome group regression coefficient; SE, standard error; CI, confidence interval; PPI, pressure passivity index; BW, birth weight; GA, gestational age; HIE, hypoxic-ischemic encephalopathy; EEG Sz, electrographic seizure; Hb, hemoglobin level.
ence between MAP and NIRS HbD offers a promising candidate for a bedside brain injury monitor. Cerebral pressure autoregulation is a physiological compensatory mechanism that maintains CBF relatively constant during changes in systemic blood pressure. Its importance in maintaining cerebral integrity has been recognized for over 50 years (Lassen 1964). Several animal and human studies have demonstrated loss of cerebral autoregulation (i.e., pressure passivity) following hypoxia and/or ischemia (Marks et al. 1996; Meek et al. 1999; Pryds et al. 1990; Short et al. 1994). Proposed mechanisms include release of vasoactive agents such as extracellular hydrogen ions, adenosine, endothelins, prostaglandins, reactive oxygen species, and nitric oxide. These mediators lead to dysregulation of cerebral vascular tone (Lassen 1964; Meek et al. 1999). The association between cerebrovascular dysfunction and cerebral injury has been supported by studies linking CBF abnormalities with proton magnetic resonance spectroscopy markers of cerebral energy failure (Tsuji et al. 1995; Winter et al. 2009) and adverse neurological outcomes (Meek et al. 1999; Pryds et al. 1990; Toet et al. 2006; Zaramella et al. 2007). Prior human studies evaluating CBF in newborns have utilized technically challenging methods such as contrast-enhanced imaging modalities [e.g., positron emission tomography (Duncan and Ment 1982; Volpe et al. 1985), single-photon emission computed tomography (Haddad et al. 1994; Konishi et al. 1994), perfusion-weighted MRI (Laswad et al. 2009; Wintermark et al. 2008)] or other point-in-time measures [e.g., Doppler ultrasound cerebral blood velocity (Archer et al. 1986; Bennhagen et al. 1998; Fukuda et al. 2008; Ilves et al. 2004), xenon-133 clearance method (Pryds et al. 1990; Younkin et al. 1982)]. More recently, NIRS has emerged as an alternative method to interrogate CBF continuously and noninvasively at the bedside (Edwards et al. 1988; Skov et al. 1991). While prior studies have linked NIRS measures with outcomes in HIE (Ancora et al. 2013; Lemmers et al. 2013; Meek et al. 1999; Toet et al. 2006; van Bel et al. 1993; Zaramella et al. 2007), few studies have focused on evaluating the relationship between MAP and NIRS as a method to interrogate autoregulation (Howlett et al. 2013; Tekes et al. 2015). Signs of impaired cerebral pressure autoregulation may precede the presence of unregulated CBF, giving this approach the potential advantage of directing intervention before irreversible injury has occurred. To our knowledge, only one other investigative group has evaluated the relationship between impaired cerebral autoregulation and brain injury in newborns undergoing therapeutic hypothermia. Howlett and colleagues recently described the use of the hemoglobin volume index (HVx), calculated as a moving correlation coefficient between MAP and relative tis-
sue hemoglobin (a surrogate measure of cerebral blood volume obtained by NIRS) (Howlett et al. 2013). HVx was used to identify the optimal MAP where vasoreactivity was preserved. These authors demonstrated a significant relationship between increased time spent below optimal MAP and brain injury assessed by both qualitative (Howlett et al. 2013) and quantitative (Tekes et al. 2015) MRI. Of note, this approach was not able to identify optimal MAP in all patients (79%, 77%, and 86% of patients during hypothermia, rewarming, and normothermia, respectively) and evaluated changes occurring in a narrower frequency range (⬍0.1 Hz). While we similarly demonstrated a relationship between indexes of cerebral autoregulation and brain injury, we used a frequency-based spectral analysis approach. This approach provides a measure of the duration and magnitude of autoregulation impairment for each patient across a wider range of frequencies (0.05– 0.25 Hz). Both approaches require time for baseline data collection but have the potential for real-time ongoing quantification of cerebral autoregulation. Our study has limitations. Although the repeated measurements over time enabled the inclusion of a relatively large number of observations in our primary GEE analyses, the small patient sample size precluded the development of robust predictive models. While our data support the association between impaired cerebral autoregulation and adverse outcome, further study is needed to establish the predictive value of PPI and gain measures, to compare these values with other clinical indicators of severity (e.g., Apgar score, seizure), and to determine thresholds that signify brain injury with high sensitivity and specificity at various time points during treatment. While we did not detect a significant time effect in our statistical analyses, graphical representation of our unadjusted data (Fig. 1) suggests that PPI may increase in the adverse outcome group after 72 h. That this time period coincides with rewarming may reflect that PPI monitoring can demonstrate withdrawal of the therapeutic effect of hypothermia. We speculate that PPI monitoring could be useful in identifying infants who may benefit from continued hypothermia or adjuvant therapies. This potential clinical application will need validation in future studies. Additionally, future investigations will be needed to confirm whether there is vulnerability in a particular cerebral hemisphere with regard to autoregulation control (i.e., gain in the RH), as we acknowledge that technical complications leading to the observation of laterality cannot be excluded. This study did not include a control group of healthy, nonencephalopathic infants, as the feasibility of continuous and prolonged invasive blood pressure monitoring is problematic in such a population. However, since the primary goal of our study was to evaluate whether cerebral hemodynamic biomarkers could identify infants with significant brain injury among at-risk infants, inclu-
J Neurophysiol • doi:10.1152/jn.00353.2015 • www.jn.org
IMPAIRED CEREBRAL AUTOREGULATION IN HIE
sion of a healthy control group was not needed to achieve study aims. Our study cannot address how these methods would characterize cerebral autoregulation in healthy newborns. While we included in the analyses many factors that may impact CBF and/or systemic blood pressure including presence of seizure, temperature, hemoglobin level, PaCO2, and use of sedative and vasoactive medications in our primary analyses, sample size limitations precluded our ability to adjust for all factors in the predictive models. These and possibly other variables will need to be considered in future studies. Finally, we used a short-term outcome of death or MRI brain injury. Qualitative MRI injury has been established as a reliable predictor of developmental outcome, particularly moderatesevere MRI injury is sensitive and specific for later motor and cognitive impairment (Azzopardi and Edwards 2010; Barkovich et al. 1998; Rutherford et al. 2010). Validation of PPI/gain as brain injury biomarkers, however, will require correlation with long-term functional outcomes. Conclusions. In this small observational study, quantifiable measures of the duration (PPI) and magnitude (gain) of cerebral pressure passivity differentiate newborns with HIE who die or have moderate to severe MRI brain injury from survivors with mild to no detectable injury. Ongoing monitoring of spectral coherence between systemic MAP and cerebral NIRS HbD may provide a bedside tool to assess evolving brain injury in the critically ill newborn. Further study is warranted to establish the role of cerebral hemodynamic monitoring in the care of infants at risk for brain injury in the intensive care unit. ACKNOWLEDGMENTS We thank Dr. Maria Powell for her editorial assistance. GRANTS This publication was supported by the Clinical and Translational Science Institute at Children’s National [National Institutes of Health (NIH) UL1 TR-000075, KL2 TR-000076] and the Intellectual and Developmental Disabilities Research Consortium (NIH P30 HD-040677). DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONTRIBUTIONS Author contributions: A.N.M., R.B.G., T.C., and A.J.d.P. conception and design of research; A.N.M., N.N.A., M.M., and K.H. performed experiments; A.N.M., R.B.G., G.V., Y.W., T.A.-S., M.M., and K.H. analyzed data; A.N.M., R.B.G., and A.J.d.P. interpreted results of experiments; A.N.M. and R.B.G. prepared figures; A.N.M. drafted manuscript; A.N.M., R.B.G., G.V., T.C., N.N.A., Y.W., T.A.-S., M.M., K.H., and A.J.d.P. edited and revised manuscript; A.N.M., R.B.G., G.V., T.C., N.N.A., Y.W., T.A.-S., M.M., K.H., and A.J.d.P. approved final version of manuscript. REFERENCES Anonymous. Executive summary: Neonatal encephalopathy and neurologic outcome, second edition. Report of the American College of Obstetricians and Gynecologists’ Task Force on Neonatal Encephalopathy. Obstet Gynecol 123: 896 –901, 2014. Ancora G, Maranella E, Grandi S, Sbravati F, Coccolini E, Savini S, Faldella G. Early predictors of short term neurodevelopmental outcome in asphyxiated cooled infants. A combined brain amplitude integrated electroencephalography and near infrared spectroscopy study. Brain Dev 35: 26 –31, 2013.
823
Archer LN, Levene MI, Evans DH. Cerebral artery Doppler ultrasonography for prediction of outcome after perinatal asphyxia. Lancet 2: 1116 –1118, 1986. Azzopardi D, Edwards AD. Magnetic resonance biomarkers of neuroprotective effects in infants with hypoxic ischemic encephalopathy. Semin Fetal Neonatal Med 15: 261–269, 2010. Barkovich AJ, Hajnal BL, Vigneron D, Sola A, Partridge JC, Allen F, Ferriero DM. Prediction of neuromotor outcome in perinatal asphyxia: evaluation of MR scoring systems. Am J Neuroradiol 19: 143–149, 1998. Bennhagen RG, Weintraub RG, Lundstrom NR, Svenningsen NW. Hypoxic-ischaemic encephalopathy is associated with regional changes in cerebral blood flow velocity and alterations in cardiovascular function. Biol Neonate 73: 275–286, 1998. Dilenge ME, Majnemer A, Shevell MI. Long-term developmental outcome of asphyxiated term neonates. J Child Neurol 16: 781–792, 2001. Duncan CC, Ment LR. Cerebral blood flow determinations by positron emission tomography. J Neurosurg Sci 26: 225–232, 1982. Edwards AD, Wyatt JS, Richardson C, Delpy DT, Cope M, Reynolds EO. Cotside measurement of cerebral blood flow in ill newborn infants by near infrared spectroscopy. Lancet 2: 770 –771, 1988. Fukuda S, Mizuno K, Kawai S, Kakita H, Goto T, Hussein MH, Daoud GA, Ito T, Kato I, Suzuki S, Togari H. Reduction in cerebral blood flow volume in infants complicated with hypoxic ischemic encephalopathy resulting in cerebral palsy. Brain Dev 30: 246 –253, 2008. Gluckman PD, Wyatt JS, Azzopardi D, Ballard R, Edwards AD, Ferriero DM, Polin RA, Robertson CM, Thoresen M, Whitelaw A, Gunn AJ. Selective head cooling with mild systemic hypothermia after neonatal encephalopathy: multicentre randomised trial. Lancet 365: 663– 670, 2005. Govindan RB, Massaro AN, Andescavage NN, Chang T, du Plessis A. Cerebral pressure passivity in newborns with encephalopathy undergoing therapeutic hypothermia. Front Hum Neurosci 8: 266, 2014. Haddad J, Constantinesco A, Brunot B, Messer J. A study of cerebral perfusion using single photon emission computed tomography in neonates with brain lesions. Acta Paediatr 83: 265–269, 1994. Halliday DM, Rosenberg JR, Amjad AM, Breeze P, Conway BA, Farmer SF. A framework for the analysis of mixed time series/point process data—theory and application to the study of physiological tremor, single motor unit discharges and electromyograms. Prog Biophys Mol Biol 64: 237–278, 1995. Howlett JA, Northington FJ, Gilmore MM, Tekes A, Huisman TA, Parkinson C, Chung SE, Jennings JM, Jamrogowicz JJ, Larson AC, Lehmann CU, Jackson E, Brady KM, Koehler RC, Lee JK. Cerebrovascular autoregulation and neurologic injury in neonatal hypoxic-ischemic encephalopathy. Pediatr Res 74: 525–535, 2013. Ilves P, Lintrop M, Metsvaht T, Vaher U, Talvik T. Cerebral blood-flow velocities in predicting outcome of asphyxiated newborn infants. Acta Paediatr 93: 523–528, 2004. Konishi Y, Kuriyama M, Mori I, Fujii Y, Konishi K, Sudo M, Ishii Y. Assessment of local cerebral blood flow in neonates with N-isopropyl-p[123I]iodoamphetamine and single photon emission computed tomography. Brain Dev 16: 450 – 453, 1994. Lassen NA. Autoregulation of cerebral blood flow. Circ Res 15, Suppl: 201–204, 1964. Laswad T, Wintermark P, Alamo L, Moessinger A, Meuli R, Gudinchet F. Method for performing cerebral perfusion-weighted MRI in neonates. Pediatr Radiol 39: 260 –264, 2009. Lemmers PM, Zwanenburg RJ, Benders MJ, de Vries LS, Groenendaal F, van Bel F, Toet MC. Cerebral oxygenation and brain activity after perinatal asphyxia: does hypothermia change their prognostic value? Pediatr Res 74: 180 –185, 2013. Liu L, Johnson HL, Cousens S, Perin J, Scott S, Lawn JE, Rudan I, Campbell H, Cibulskis R, Li M, Mathers C, Black RE. Global, regional, and national causes of child mortality: an updated systematic analysis for 2010 with time trends since 2000. Lancet 379: 2151–2161, 2012. Marks KA, Mallard EC, Roberts I, Williams CE, Sirimanne ES, Johnston B, Gluckman PD, Edwards AD. Delayed vasodilation and altered oxygenation after cerebral ischemia in fetal sheep. Pediatr Res 39: 48 –54, 1996. Meek JH, Elwell CE, McCormick DC, Edwards AD, Townsend JP, Stewart AL, Wyatt JS. Abnormal cerebral haemodynamics in perinatally asphyxiated neonates related to outcome. Arch Dis Child Fetal Neonatal Ed 81: F110 –F115, 1999. O’Leary H, Gregas MC, Limperopoulos C, Zaretskaya I, Bassan H, Soul JS, Di Salvo DN, du Plessis AJ. Elevated cerebral pressure passivity is
J Neurophysiol • doi:10.1152/jn.00353.2015 • www.jn.org
824
IMPAIRED CEREBRAL AUTOREGULATION IN HIE
associated with prematurity-related intracranial hemorrhage. Pediatrics 124: 302–309, 2009. Pryds O, Greisen G, Lou H, Friis-Hansen B. Vasoparalysis associated with brain damage in asphyxiated term infants. J Pediatr 117: 119 –125, 1990. Rutherford M, Ramenghi LA, Edwards AD, Brocklehurst P, Halliday H, Levene M, Strohm B, Thoresen M, Whitelaw A, Azzopardi D. Assessment of brain tissue injury after moderate hypothermia in neonates with hypoxic-ischaemic encephalopathy: a nested substudy of a randomised controlled trial. Lancet Neurol 9: 39 – 45, 2010. Sarnat HB, Sarnat MS. Neonatal encephalopathy following fetal distress. A clinical and electroencephalographic study. Arch Neurol 33: 696 –705, 1976. Shankaran S, Laptook AR, Ehrenkranz RA, Tyson JE, McDonald SA, Donovan EF, Fanaroff AA, Poole WK, Wright LL, Higgins RD, Finer NN, Carlo WA, Duara S, Oh W, Cotten CM, Stevenson DK, Stoll BJ, Lemons JA, Guillet R, Jobe AH. Whole-body hypothermia for neonates with hypoxic-ischemic encephalopathy. N Engl J Med 353: 1574 –1584, 2005. Short BL, Walker LK, Traystman RJ. Impaired cerebral autoregulation in the newborn lamb during recovery from severe, prolonged hypoxia, combined with carotid artery and jugular vein ligation. Crit Care Med 22: 1262–1268, 1994. Skov L, Pryds O, Greisen G. Estimating cerebral blood flow in newborn infants: comparison of near infrared spectroscopy and 133Xe clearance. Pediatr Res 30: 570 –573, 1991. Soul JS, Hammer PE, Tsuji M, Saul JP, Bassan H, Limperopoulos C, Disalvo DN, Moore M, Akins P, Ringer S, Volpe JJ, Trachtenberg F, du Plessis AJ. Fluctuating pressure-passivity is common in the cerebral circulation of sick premature infants. Pediatr Res 61: 467– 473, 2007. Tekes A, Poretti A, Scheurkogel MM, Huisman TA, Howlett JA, Alqahtani E, Lee JH, Parkinson C, Shapiro K, Chung SE, Jennings JM, Gilmore MM, Hogue CW, Martin LJ, Koehler RC, Northington FJ, Lee JK. Apparent diffusion coefficient scalars correlate with near-infrared spectroscopy markers of cerebrovascular autoregulation in neonates cooled for
perinatal hypoxic-ischemic injury. AJNR Am J Neuroradiol 36: 188 –193, 2015. Toet MC, Lemmers PM, van Schelven LJ, van Bel F. Cerebral oxygenation and electrical activity after birth asphyxia: their relation to outcome. Pediatrics 117: 333–339, 2006. Tsuji M, du Plessis A, Taylor G, Crocker R, Volpe JJ. Near infrared spectroscopy detects cerebral ischemia during hypotension in piglets. Pediatr Res 44: 591–595, 1998. Tsuji M, Naruse H, Volpe J, Holtzman D. Reduction of cytochrome aa3 measured by near-infrared spectroscopy predicts cerebral energy loss in hypoxic piglets. Pediatr Res 37: 253–259, 1995. van Bel F, Dorrepaal CA, Benders MJ, Zeeuwe PE, van de Bor M, Berger HM. Changes in cerebral hemodynamics and oxygenation in the first 24 hours after birth asphyxia. Pediatrics 92: 365–372, 1993. Volpe JJ, Herscovitch P, Perlman JM, Kreusser KL, Raichle ME. Positron emission tomography in the asphyxiated term newborn: parasagittal impairment of cerebral blood flow. Ann Neurol 17: 287–296, 1985. Winter JD, Tichauer KM, Gelman N, Thompson RT, Lee TY, St Lawrence K. Changes in cerebral oxygen consumption and high-energy phosphates during early recovery in hypoxic-ischemic piglets: a combined near-infrared and magnetic resonance spectroscopy study. Pediatr Res 65: 181–187, 2009. Wintermark P, Moessinger AC, Gudinchet F, Meuli R. Perfusion-weighted magnetic resonance imaging patterns of hypoxic-ischemic encephalopathy in term neonates. J Magn Reson Imaging 28: 1019 –1025, 2008. Younkin DP, Reivich M, Jaggi J, Obrist W, Delivoria-Papadopoulos M. Noninvasive method of estimating human newborn regional cerebral blood flow. J Cereb Blood Flow Metab 2: 415– 420, 1982. Zaramella P, Saraceni E, Freato F, Falcon E, Suppiej A, Milan A, Laverda AM, Chiandetti L. Can tissue oxygenation index (TOI) and cotside neurophysiological variables predict outcome in depressed/asphyxiated newborn infants? Early Hum Dev 83: 483– 489, 2007.
J Neurophysiol • doi:10.1152/jn.00353.2015 • www.jn.org
CALL FOR PAPERS
Neurological Disease and Autonomic Dysfunction
Impaired cerebral autoregulation and brain injury in newborns with hypoxic-ischemic encephalopathy treated with hypothermia An N. Massaro,1,2,6 R. B. Govindan,2,6 Gilbert Vezina,3,6 Taeun Chang,4,6 Nickie N. Andescavage,1,6 Yunfei Wang,5,6 Tareq Al-Shargabi,2 Marina Metzler,2 Kari Harris,4 and Adre J. du Plessis2,6 1
Division of Neonatology, Children’s National Health System, Washington, District of Columbia; 2Division of Fetal and Transitional Medicine, Children’s National Health System, Washington, District of Columbia; 3Division of Diagnostic Imaging and Radiology, Children’s National Health System, Washington, District of Columbia; 4Department of Neurology, Children’s National Health System, Washington, District of Columbia; 5Department of Biostatistics and Study Methodology, Children’s National Health System, Washington, District of Columbia; and 6Department of Pediatrics, The George Washington University School of Medicine, Washington, District of Columbia Submitted 8 April 2015; accepted in final form 9 June 2015
Massaro AN, Govindan RB, Vezina G, Chang T, Andescavage NN, Wang Y, Al-Shargabi T, Metzler M, Harris K, du Plessis AJ. Impaired cerebral autoregulation and brain injury in newborns with hypoxic-ischemic encephalopathy treated with hypothermia. J Neurophysiol 114: 818 – 824, 2015. First published 10 June 2015; doi:10.1152/jn.00353.2015.—Impaired cerebral autoregulation may contribute to secondary injury in newborns with hypoxic-ischemic encephalopathy (HIE). Continuous, noninvasive assessment of cerebral pressure autoregulation can be achieved with bedside nearinfrared spectroscopy (NIRS) and systemic mean arterial blood pressure (MAP) monitoring. This study aimed to evaluate whether impaired cerebral autoregulation measured by NIRS-MAP monitoring during therapeutic hypothermia and rewarming relates to outcome in 36 newborns with HIE. Spectral coherence analysis between NIRS and MAP was used to quantify changes in the duration [pressure passivity index (PPI)] and magnitude (gain) of cerebral autoregulatory impairment. Higher PPI in both cerebral hemispheres and gain in the right hemisphere were associated with neonatal adverse outcomes [death or detectable brain injury by magnetic resonance imaging (MRI), P ⬍ 0.001]. NIRS-MAP monitoring of cerebral autoregulation can provide an ongoing physiological biomarker that may help direct care in perinatal brain injury. hypoxic-ischemic encephalopathy; cerebral blood flow; magnetic resonance imaging; newborn PERINATAL HYPOXIC-ISCHEMIC
encephalopathy (HIE) is an important cause of neonatal mortality and long-term neurodisability (Dilenge et al. 2001; Liu et al. 2012). While therapeutic hypothermia improves outcome in HIE and is the current standard of care, a significant proportion of treated infants continue to suffer adverse outcomes after therapy (Gluckman et al. 2005; Shankaran et al. 2005). Adjuvant neuroprotective interventions are needed to further improve outcomes for this devastating condition. Unfortunately, methods to acutely identify infants with inadequate response to hypothermia and evolving brain injury are limited in the neonatal intensive care
Address for reprint requests and other correspondence: A. N. Massaro, Div. of Neonatology, Children’s National Medical Center, 111 Michigan Ave. NW, Washington, DC 20010 (e-mail: [email protected]). 818
unit (NICU). Clinicians often rely on magnetic resonance imaging (MRI) performed after hypothermia to determine the presence and extent of brain injury that has occurred. This approach does not enable therapeutic interventions aimed to prevent injury progression. Thus bedside methods to monitor risk for evolving brain injury are needed to direct clinical care and rational therapeutic interventions. Ongoing assessment of cerebral pressure autoregulation integrity can provide an assessment of the physiological response to hypoxia and/or ischemia, offering a window into evolving brain injury. Impaired cerebral pressure autoregulation has been implicated in the pathogenesis of secondary injury in HIE, as cerebral vasoparalysis following a hypoxic and/or ischemic insult has been well documented in both animal and human studies (Meek et al. 1999; Pryds et al. 1990; Short et al. 1994). Cerebral pressure passivity, or the inability to regulate cerebral blood flow (CBF) in the setting of changes in systemic blood pressure, can lead to uncoupling of CBF and cerebral energy metabolism that can contribute to secondary energy failure and structural brain injury in HIE. We previously described methods to monitor cerebral pressure autoregulation by quantifying frequency domain spectral coherence between near-infrared spectroscopy (NIRS)-derived cerebral oxyhemoglobin-deoxyhemoglobin (HbD), a surrogate measure of CBF (Tsuji et al. 1998), and systemic mean arterial blood pressure (MAP) (Govindan et al. 2014; Soul et al. 2007). Spectral coherence is a technique that allows quantification of the relationship between changes in MAP and changes in CBF (or HbD), with pressure-passive CBF indicated by increased coherence between the two physiological signals. This approach allows for quantification of both the duration [pressure passivity index (PPI)] and magnitude (gain) of cerebral pressure passivity. While cerebral pressure passivity assessed by this approach has been proposed as a risk factor for brain injury in the preterm infant (O’Leary et al. 2009), its significance in term newborns with HIE is unknown. This study aimed to investigate whether quantified changes of the duration and magnitude of cerebral pressure passivity during therapeutic hypothermia are associated with adverse neonatal outcomes
0022-3077/15 Copyright © 2015 the American Physiological Society
www.jn.org
IMPAIRED CEREBRAL AUTOREGULATION IN HIE
(i.e., death or structural brain injury) in newborns with HIE. We hypothesized that neonates who die or have moderate to severe brain injury on MRI will have worse autoregulatory function during therapeutic hypothermia than those with only mild or no brain injuries. MATERIALS AND METHODS
Study population. Newborns with HIE referred for therapeutic hypothermia to a regional NICU in a freestanding children’s hospital were prospectively enrolled in this observational study. Infants met institutional criteria for hypothermia: 1) ⬎35 wk gestational age; 2) ⬎1,800 g at birth; 3) physiological evidence of perinatal asphyxia with either demonstrated severe metabolic acidosis, with pH ⱕ 7 or base deficit ⱖ 16, or milder acidosis, with pH 7.01–7.15 or base deficit 10 –16, accompanied by Apgar score ⱕ 5 or positive-pressure ventilation at 10 min and a sentinel perinatal event; and 4) signs of moderate to severe clinical encephalopathy and/or seizures. All infants were treated with whole-body hypothermia according to the National Institute of Child Health and Human Development Neonatal Research Network protocol (Shankaran et al. 2005). Briefly, infants were cooled to 33.5°C with an infant-size cooling blanket servoregulated to esophageal temperature (Blanketrol-II, Cincinnati Subzero, Cincinnati, OH) for 72 h, followed by gradual rewarming by 0.5°C/h over 6 h. All infants undergo continuous video EEG monitoring during cooling and through the rewarming period as standard clinical care. Sedation with low-dose morphine (starting 0.05 mg/kg every 4 h) is titrated to prevent shivering and other overt signs of pain or discomfort. Blood pressure is maintained at goal MAP ⬎40 mmHg with use of dopamine as first-line vasoactive therapy, followed by the addition of epinephrine and systemic hydrocortisone as needed. The study was approved by the Children’s National Health System Institutional Review Board and was conducted in accordance with the ethical standards of the Children’s National Office for the Protection of Human Subjects, and informed consent was obtained from a parent of each participant for physiological monitoring. Data collection. Demographic and clinical data were collected from birth hospital and study site medical records. Presenting characteristics were recorded, including umbilical cord or arterial blood gas obtained in the first hour of life, Apgar scores, and the initial grade of encephalopathy as classified by modified Sarnat criteria (Sarnat and Sarnat 1976; Shankaran et al. 2005). Data from NICU hospitalizations were also collected, including use of vasoactive therapy, presence of electrographic seizures, use of sedatives, and serial measurements of core body temperature, hemoglobin, and partial arterial pressure of carbon dioxide (PaCO2) levels. Continuous recordings of NIRS cerebral oximetry (NIRO 200, Hamamatsu Photonics, Hamamatsu, Japan) and MAP from an indwelling arterial line (Philips IntelliVue MP70) were recorded in a time-locked manner at a rate of 5 Hz and upsampled to 1 kHz with custom software developed in LabVIEW (National Instruments). NIRS monitoring was performed bilaterally, with optodes placed over the right and left fronto-temporal regions. After parental consent was obtained, monitoring was initiated within 24 h of cooling initiation and continued for at least 6 h after the completion of rewarming. Physiological signal processing. Data were processed and analyzed with a power spectral estimation approach as previously described (Govindan et al. 2014). Briefly, after using an in-house pipeline for automated artifact rejection (Govindan et al. 2014), we divided our continuous data into 10-min epochs. We then quantified the spectral coherence between MAP and HbD, using a previously described mathematical algorithm (Halliday et al. 1995). Coherence is a normalized quantity and takes on a value of 1 in cases of a perfect linear relationship between the signals and 0 in the case of independence. The statistical significance of the coherence was assessed with a mathematically derived threshold of 0.384, defined by 1 ⫺ (1 ⫺ 1/(M ⫺ 1) ␣) , where M is the number of epochs involved in the coherence
819
estimation, which was 20 in our case, and ␣ set to 0.999 (Govindan et al. 2014; Halliday et al. 1995). Based on the knowledge that impulse response time of cerebral autoregulation is expected to operate at 5–20 s, pressure passivity was defined as coherence greater than the threshold in the frequency band of 0.05– 0.25 Hz (O’Leary et al. 2009). The PPI was then calculated as the total percentage of pressurepassive 10-min epochs per 6 h of recording. We also calculated transfer gain (Halliday et al. 1995) within the pressure-passive epochs and averaged over 6 h of recording to match the PPI calculation. Gain reflects the change in HbD caused by 1-mmHg change in MAP. Gain function provides a measure of the magnitude of pressure passivity. All processing was done off-line with MATLAB software (MathWorks, Natick, MA). Magnetic resonance imaging. MRI was performed according to our institutional clinical protocol based on established guidelines (2014). Surviving infants underwent routine MRI at target age 3–5 days (to evaluate early diffusion changes) and 10 –12 days (to evaluate extent of lesions) of life on a 3-T scanner (Discovery MR750, GE Healthcare, Milwaukee, WI) using a 32-channel receive-only head coil (MR Instruments, Minneapolis, MN). Standard anatomical sequences included 3D T1-weighted spoiled gradient recalled, double-acquisition axial FSE T2 proton density, axial T2 propeller (in cases of patient motion), axial T2-star-weighted susceptibility imaging, coronal T1 FLAIR propeller, and axial 30-direction DTI with generation of ADC maps off-line. Images were scored according to a validated scoring system that assigns a score ranging from 0 to 4 for the degree of basal ganglia/thalamic (BG) injury and from 0 to 5 for the degree of watershed cortical (WS) injury (Barkovich et al. 1998) by an experienced neuroradiologist (G. Vezina) masked to clinical and physiological data. If an infant had more than one MRI, the higher injury score was used for analysis. Adverse outcome was defined as death during hospitalization or moderate to severe brain injury by MRI (BG score ⱖ 3 or WS score ⱖ 4). Statistical analysis. Descriptive statistics included means (standard deviations) and medians (ranges) for continuous parametric and nonparametric variables, respectively, as well as counts (percentages) for categorical data. Generalized estimating equations (GEEs) were used to evaluate the relationship between repeated measures of PPI/gain over time and outcome classification adjusting for baseline and clinical covariates including birth weight, gestational age, sex, 5-min Apgar score, presenting pH, severe encephalopathy at presentation, time from birth, presence of electrographic seizures, use of vasoactives, sedation, temperature, hemoglobin, and PaCO2 level. Models were developed for right and left PPI and gain values to evaluate for laterality of relationships. A log link function for expected values of PPI/gain was specified in the models in order to account for data that were not normally distributed. Secondarily, to explore the predictive ability of PPI and gain to distinguish infants with adverse outcomes, prediction models were developed with logistic regression. For these analyses, a summary value of average PPI or gain over the entire period of recording was used for each individual subject. Performance characteristics of the models were evaluated with receiver operating curve analyses, where an area under the curve (AUC) of 1 represents high diagnostic accuracy, in contrast to a value of 0.5 that signifies poor discrimination of outcome groups. Because of sample size limitations, these secondary analyses were not adjusted for covariates. Statistical analysis was performed with SAS 9.3 (SAS Institute, Cary, NC). RESULTS
Description of study population. Of the 59 eligible patients with moderate to severe HIE admitted during the study period (November 2012–July 2014), 9 infants were not enrolled because of extremis presentation and death prior to consent (n ⫽ 1), parental refusal (n ⫽ 3), or inability to obtain consent
J Neurophysiol • doi:10.1152/jn.00353.2015 • www.jn.org
820
IMPAIRED CEREBRAL AUTOREGULATION IN HIE
Table 1. Characteristics of study population Favorable Outcome (n ⫽ 20)
Adverse Outcome (n ⫽ 16)
P Value
38.6 ⫾ 1.7 3.5 ⫾ 1.0 12 (60) 6.91 (6.50–7.25)
39.2 ⫾ 1.3 2.9 ⫾ 0.3 46 (37.5) 6.9 (6.65–7.14)
0.294 0.016 0.180 0.814
Gestational age, wk (mean ⫾ SD) Birth weight, kg (mean ⫾ SD) Sex, n males (%) Initial pH Apgars 1 min 5 min 10 min Encephalopathy grade, n (%) Moderate Severe EEG seizures, n (%) Received vasoactives, n (%) Dopamine Epinephrine Age monitoring initiated, h
1 (0–9) 4 (1–8) 6 (2–8)†
1 (0–8)* 2 (0–9)* 3 (0–7)‡
16 (80) 4 (20) 2 (10)
0.479 0.023 0.059 0.180
10 (62.5) 6 (37.5) 8 (50)
13 (65) 2 (10) 13.4 (5.9–23.2)
0.008
11 (69) 5 (31) 13.3 (7.4–25.1)
0.983 0.109 0.987
Data presented as median (range) except where indicated. *Documented for 15/16 patients; †documented for 17/20 patients; ‡documented for 13/16 patients.
because of parental (n ⫽ 2) or investigator (n ⫽ 3) unavailability. Of the 50 patients enrolled for physiological monitoring, 14 infants did not have indwelling arterial catheters for continuous blood pressure monitoring. Thus data from 36 patients with complete MAP and NIRS data were included in the study. The adverse outcome group comprised seven infants who died (because of withdrawal of medical care based on poor neurological prognosis) and nine infants with moderate/ severe MRI brain injury (severe BG injury n ⫽ 5, severe WS injury n ⫽ 4). Infants in the adverse outcome group had lower birth weight, lower Apgar score at 5 min, and higher frequency of electrographic seizures compared with the favorable outcome group (P ⬍ 0.05). Otherwise, baseline characteristics were similar between groups (Table 1). The majority of infants received vasoactive medications (dopamine n ⫽ 23, 67%; epinephrine n ⫽ 7, 19%) and sedatives (morphine n ⫽ 31, 86%; midazolam n ⫽ 12, 33%).
MAP over time did not differentiate the two outcome groups (Fig. 1). Phenobarbital was given for seizures in 12 (33%) patients. Frequency of these medical therapies did not differ between outcome groups (P ⬎ 0.05). Temperature and laboratory data monitored through hypothermia and rewarming are presented in Table 2. Median temperatures, PaCO2, and hemoglobin levels did not differ between groups (P ⬎ 0.05). Continuous NIRS/MAP recordings were initiated at a median age of 13 (range 6 –25) h of life. The median duration of recording was 73 (range 5–91) h. Early MRI was performed in 24 infants at a median age of 4 (range 3– 6) days of life, while a later MRI (median 10, range 7–16 days) was performed in 25 infants. Seventeen infants had two scans performed. Of the nine infants with moderate/severe MRI brain injury in the adverse outcome group, the majority (n ⫽ 8, 89%) had severe BG injury while only one infant had severe WS injury. The favorable outcome group comprised 14
60 Favorable Group Adverse Group 55
Fig. 1. Median systemic mean arterial blood pressure (MAP) over time in infants with hypoxicischemic encephalopathy (HIE) who died or had moderate to severe MRI injury (solid line) compared with survivors with no or mild MRI injury (dashed line). Bars represent interquartile range.
MAP (mmHg)
50
45
40
35
30
0
10
20
30
40
50
60
70
80
90
100
Time Since Birth (hours) Hypothermia
J Neurophysiol • doi:10.1152/jn.00353.2015 • www.jn.org
Rewarming
Normothermia
110
IMPAIRED CEREBRAL AUTOREGULATION IN HIE
Table 2. Temperature and laboratory data Favorable Outcome Adverse Outcome (n ⫽ 20) (n ⫽ 16) Temperature during hypothermia, °C Min Max Average Temperature after rewarming, °C Min Max Average PaCO2, mmHg Min Max Average Hemoglobin, mg/dl Min Max Average
31.1 (27.3–32.7) 35.1 (33.7–36.7) 33 (32.5–33.5)
31.2 (29.0–32.4) 34.3 (33.4–36.4) 33.1 (32.6–33.4)
36.5 (34.7–36.8) 36.7 (36.3–37.2) 36.6 (36.1–36.9)
36.5 (35.2–36.9) 36.4 (36.2–37) 36.5 (35.8–36.9)
29.7 (14.2–45.5) 56.9 (46.6–75.2) 43.9 (38.3–52.1)
27.0 (12.0–52.5) 58.1 (33.0–104.6) 44.4 (24.6–63.1)
12.6 (9.3–17.0) 16.2 (12.7–19.3) 15.0 (11.7–17.8)
13.7 (7.2–18.1) 16.5 (12.6–22.2) 15.0 (11.0–19.5)
Data presented as median (range). PaCO2, partial arterial pressure of carbon dioxide. Differences between groups were not significant (P ⬎ 0.05).
infants with normal MRI and 6 infants with mild BG injury (score ⱕ 2). There was no laterality to injury pattern in any of the infants. Pressure passivity index. A total of 442 observations (i.e., PPI calculated in a 6-h window) were included in the PPI analyses for each cerebral hemisphere. Unadjusted median PPI values over time are shown in Fig. 2. After adjusting for baseline and clinical covariates, higher PPIs measured from both the left (LH) and right (RH) hemispheres were significantly associated with adverse outcome (P ⬍ 0.01). There was no significant time effect on these relationships, and, similarly, no interaction was detected between time and outcome group. Regression model results are summarized in Table 3. HbD-MAP spectral gain. Given that gain function can only be calculated within pressure-passive epochs, fewer observations (LH, n ⫽ 332; RH, n ⫽ 334) were included in the gain 0.9
A
0.8 0.7
analyses. While gain in the RH was significantly higher in the adverse outcome group (P ⬍ 0.001), gain from the LH was not significantly associated with outcome (P ⬎ 0.05) after adjusting for covariates. Similar to PPI, there was no significant time effect on the relationship between gain and outcome. Regression model results are included in Table 3. Secondary PPI/gain predictive models. Only average PPI (over the recording period) measured from the LH was significantly associated with adverse outcome in the logistic regression analyses (AUC ⫽ 0.7193, 95% CI: 0.5430 – 0.896, P ⫽ 0.034). There was a trend toward association between PPI from the RH and adverse outcome (AUC 0.664, 95% CI: 0.480 – 0.849, P ⫽ 0.082), while gains from both the LH and RH were not significantly predictive of adverse outcome (P ⬎ 0.05). A probability cutoff point of 0.378 for average PPI value measured from the LH had 87% sensitivity and 58% specificity for prediction of adverse outcome. DISCUSSION
We demonstrate that measurements of spectral coherence between MAP and NIRS HbD, reflecting impaired cerebral autoregulation, differ between newborns with HIE who die or have moderate to severe MRI brain injury and survivors with normal to mild injury. Specifically, infants with adverse outcomes exhibited higher PPI and gain, indicating longer duration and higher magnitude of cerebral pressure passivity following hypoxia-ischemia. Monitoring of blood pressure alone did not differentiate outcome groups. These data support the notion that quantifying cerebral autoregulation integrity can provide a viable physiological biomarker of brain injury. A bedside tool that can assess evolving cerebral injury is needed to individualize therapy in newborns with HIE. Such a tool would enable the timely identification of patients for whom hypothermia alone is providing inadequate neuroprotection and could be used to direct the rational introduction of adjuvant therapies. Continuous real-time assessment of spectral coher-
Favorable Group Adverse Group
0.6
PPI-LH
821
0.5 0.4 0.3 0.2 0.1 0 0.9
Fig. 2. Median pressure passivity index (PPI) over time in HIE infants with adverse (death or moderate to severe MRI injury, solid line) vs. favorable (survivors with no or mild MRI injury, dashed line) outcome. Bars represent interquartile range. Measurements from the left (LH; A) and right (RH; B) hemispheres are shown.
B
0.8 0.7
PPI-RH
0.6 0.5 0.4 0.3 0.2 0.1 0 0
10
20
30
40
50
60
70
80
90
100
110
Time Since Birth (hours) Hypothermia
Rewarming
Normothermia
J Neurophysiol • doi:10.1152/jn.00353.2015 • www.jn.org
822
IMPAIRED CEREBRAL AUTOREGULATION IN HIE
Table 3. GEE model results B
SE
95% CI
P Value
Other Significant Covariates
PPI-left PPI-right
0.92 1.24
0.28 0.25
0.37–1.48 0.75–1.74
0.001 ⬍0.001
Gain-left Gain-right
0.11 1.64
0.12 0.41
⫺0.11–0.34 0.84–2.44
0.322 ⬍0.001
BW, GA, severe HIE, initial pH, 5-min Apgar, EEG Sz, Hb, PaCO2 BW, GA, severe HIE, initial pH, 5-min Apgar, EEG Sz, Hb, PaCO2, sex, vasopressor use PaCO2, vasopressor use Severe HIE, 5-min Agpar, EEG Sz, PaCO2, sex, vasopressor use
B, outcome group regression coefficient; SE, standard error; CI, confidence interval; PPI, pressure passivity index; BW, birth weight; GA, gestational age; HIE, hypoxic-ischemic encephalopathy; EEG Sz, electrographic seizure; Hb, hemoglobin level.
ence between MAP and NIRS HbD offers a promising candidate for a bedside brain injury monitor. Cerebral pressure autoregulation is a physiological compensatory mechanism that maintains CBF relatively constant during changes in systemic blood pressure. Its importance in maintaining cerebral integrity has been recognized for over 50 years (Lassen 1964). Several animal and human studies have demonstrated loss of cerebral autoregulation (i.e., pressure passivity) following hypoxia and/or ischemia (Marks et al. 1996; Meek et al. 1999; Pryds et al. 1990; Short et al. 1994). Proposed mechanisms include release of vasoactive agents such as extracellular hydrogen ions, adenosine, endothelins, prostaglandins, reactive oxygen species, and nitric oxide. These mediators lead to dysregulation of cerebral vascular tone (Lassen 1964; Meek et al. 1999). The association between cerebrovascular dysfunction and cerebral injury has been supported by studies linking CBF abnormalities with proton magnetic resonance spectroscopy markers of cerebral energy failure (Tsuji et al. 1995; Winter et al. 2009) and adverse neurological outcomes (Meek et al. 1999; Pryds et al. 1990; Toet et al. 2006; Zaramella et al. 2007). Prior human studies evaluating CBF in newborns have utilized technically challenging methods such as contrast-enhanced imaging modalities [e.g., positron emission tomography (Duncan and Ment 1982; Volpe et al. 1985), single-photon emission computed tomography (Haddad et al. 1994; Konishi et al. 1994), perfusion-weighted MRI (Laswad et al. 2009; Wintermark et al. 2008)] or other point-in-time measures [e.g., Doppler ultrasound cerebral blood velocity (Archer et al. 1986; Bennhagen et al. 1998; Fukuda et al. 2008; Ilves et al. 2004), xenon-133 clearance method (Pryds et al. 1990; Younkin et al. 1982)]. More recently, NIRS has emerged as an alternative method to interrogate CBF continuously and noninvasively at the bedside (Edwards et al. 1988; Skov et al. 1991). While prior studies have linked NIRS measures with outcomes in HIE (Ancora et al. 2013; Lemmers et al. 2013; Meek et al. 1999; Toet et al. 2006; van Bel et al. 1993; Zaramella et al. 2007), few studies have focused on evaluating the relationship between MAP and NIRS as a method to interrogate autoregulation (Howlett et al. 2013; Tekes et al. 2015). Signs of impaired cerebral pressure autoregulation may precede the presence of unregulated CBF, giving this approach the potential advantage of directing intervention before irreversible injury has occurred. To our knowledge, only one other investigative group has evaluated the relationship between impaired cerebral autoregulation and brain injury in newborns undergoing therapeutic hypothermia. Howlett and colleagues recently described the use of the hemoglobin volume index (HVx), calculated as a moving correlation coefficient between MAP and relative tis-
sue hemoglobin (a surrogate measure of cerebral blood volume obtained by NIRS) (Howlett et al. 2013). HVx was used to identify the optimal MAP where vasoreactivity was preserved. These authors demonstrated a significant relationship between increased time spent below optimal MAP and brain injury assessed by both qualitative (Howlett et al. 2013) and quantitative (Tekes et al. 2015) MRI. Of note, this approach was not able to identify optimal MAP in all patients (79%, 77%, and 86% of patients during hypothermia, rewarming, and normothermia, respectively) and evaluated changes occurring in a narrower frequency range (⬍0.1 Hz). While we similarly demonstrated a relationship between indexes of cerebral autoregulation and brain injury, we used a frequency-based spectral analysis approach. This approach provides a measure of the duration and magnitude of autoregulation impairment for each patient across a wider range of frequencies (0.05– 0.25 Hz). Both approaches require time for baseline data collection but have the potential for real-time ongoing quantification of cerebral autoregulation. Our study has limitations. Although the repeated measurements over time enabled the inclusion of a relatively large number of observations in our primary GEE analyses, the small patient sample size precluded the development of robust predictive models. While our data support the association between impaired cerebral autoregulation and adverse outcome, further study is needed to establish the predictive value of PPI and gain measures, to compare these values with other clinical indicators of severity (e.g., Apgar score, seizure), and to determine thresholds that signify brain injury with high sensitivity and specificity at various time points during treatment. While we did not detect a significant time effect in our statistical analyses, graphical representation of our unadjusted data (Fig. 1) suggests that PPI may increase in the adverse outcome group after 72 h. That this time period coincides with rewarming may reflect that PPI monitoring can demonstrate withdrawal of the therapeutic effect of hypothermia. We speculate that PPI monitoring could be useful in identifying infants who may benefit from continued hypothermia or adjuvant therapies. This potential clinical application will need validation in future studies. Additionally, future investigations will be needed to confirm whether there is vulnerability in a particular cerebral hemisphere with regard to autoregulation control (i.e., gain in the RH), as we acknowledge that technical complications leading to the observation of laterality cannot be excluded. This study did not include a control group of healthy, nonencephalopathic infants, as the feasibility of continuous and prolonged invasive blood pressure monitoring is problematic in such a population. However, since the primary goal of our study was to evaluate whether cerebral hemodynamic biomarkers could identify infants with significant brain injury among at-risk infants, inclu-
J Neurophysiol • doi:10.1152/jn.00353.2015 • www.jn.org
IMPAIRED CEREBRAL AUTOREGULATION IN HIE
sion of a healthy control group was not needed to achieve study aims. Our study cannot address how these methods would characterize cerebral autoregulation in healthy newborns. While we included in the analyses many factors that may impact CBF and/or systemic blood pressure including presence of seizure, temperature, hemoglobin level, PaCO2, and use of sedative and vasoactive medications in our primary analyses, sample size limitations precluded our ability to adjust for all factors in the predictive models. These and possibly other variables will need to be considered in future studies. Finally, we used a short-term outcome of death or MRI brain injury. Qualitative MRI injury has been established as a reliable predictor of developmental outcome, particularly moderatesevere MRI injury is sensitive and specific for later motor and cognitive impairment (Azzopardi and Edwards 2010; Barkovich et al. 1998; Rutherford et al. 2010). Validation of PPI/gain as brain injury biomarkers, however, will require correlation with long-term functional outcomes. Conclusions. In this small observational study, quantifiable measures of the duration (PPI) and magnitude (gain) of cerebral pressure passivity differentiate newborns with HIE who die or have moderate to severe MRI brain injury from survivors with mild to no detectable injury. Ongoing monitoring of spectral coherence between systemic MAP and cerebral NIRS HbD may provide a bedside tool to assess evolving brain injury in the critically ill newborn. Further study is warranted to establish the role of cerebral hemodynamic monitoring in the care of infants at risk for brain injury in the intensive care unit. ACKNOWLEDGMENTS We thank Dr. Maria Powell for her editorial assistance. GRANTS This publication was supported by the Clinical and Translational Science Institute at Children’s National [National Institutes of Health (NIH) UL1 TR-000075, KL2 TR-000076] and the Intellectual and Developmental Disabilities Research Consortium (NIH P30 HD-040677). DISCLOSURES No conflicts of interest, financial or otherwise, are declared by the author(s). AUTHOR CONTRIBUTIONS Author contributions: A.N.M., R.B.G., T.C., and A.J.d.P. conception and design of research; A.N.M., N.N.A., M.M., and K.H. performed experiments; A.N.M., R.B.G., G.V., Y.W., T.A.-S., M.M., and K.H. analyzed data; A.N.M., R.B.G., and A.J.d.P. interpreted results of experiments; A.N.M. and R.B.G. prepared figures; A.N.M. drafted manuscript; A.N.M., R.B.G., G.V., T.C., N.N.A., Y.W., T.A.-S., M.M., K.H., and A.J.d.P. edited and revised manuscript; A.N.M., R.B.G., G.V., T.C., N.N.A., Y.W., T.A.-S., M.M., K.H., and A.J.d.P. approved final version of manuscript. REFERENCES Anonymous. Executive summary: Neonatal encephalopathy and neurologic outcome, second edition. Report of the American College of Obstetricians and Gynecologists’ Task Force on Neonatal Encephalopathy. Obstet Gynecol 123: 896 –901, 2014. Ancora G, Maranella E, Grandi S, Sbravati F, Coccolini E, Savini S, Faldella G. Early predictors of short term neurodevelopmental outcome in asphyxiated cooled infants. A combined brain amplitude integrated electroencephalography and near infrared spectroscopy study. Brain Dev 35: 26 –31, 2013.
823
Archer LN, Levene MI, Evans DH. Cerebral artery Doppler ultrasonography for prediction of outcome after perinatal asphyxia. Lancet 2: 1116 –1118, 1986. Azzopardi D, Edwards AD. Magnetic resonance biomarkers of neuroprotective effects in infants with hypoxic ischemic encephalopathy. Semin Fetal Neonatal Med 15: 261–269, 2010. Barkovich AJ, Hajnal BL, Vigneron D, Sola A, Partridge JC, Allen F, Ferriero DM. Prediction of neuromotor outcome in perinatal asphyxia: evaluation of MR scoring systems. Am J Neuroradiol 19: 143–149, 1998. Bennhagen RG, Weintraub RG, Lundstrom NR, Svenningsen NW. Hypoxic-ischaemic encephalopathy is associated with regional changes in cerebral blood flow velocity and alterations in cardiovascular function. Biol Neonate 73: 275–286, 1998. Dilenge ME, Majnemer A, Shevell MI. Long-term developmental outcome of asphyxiated term neonates. J Child Neurol 16: 781–792, 2001. Duncan CC, Ment LR. Cerebral blood flow determinations by positron emission tomography. J Neurosurg Sci 26: 225–232, 1982. Edwards AD, Wyatt JS, Richardson C, Delpy DT, Cope M, Reynolds EO. Cotside measurement of cerebral blood flow in ill newborn infants by near infrared spectroscopy. Lancet 2: 770 –771, 1988. Fukuda S, Mizuno K, Kawai S, Kakita H, Goto T, Hussein MH, Daoud GA, Ito T, Kato I, Suzuki S, Togari H. Reduction in cerebral blood flow volume in infants complicated with hypoxic ischemic encephalopathy resulting in cerebral palsy. Brain Dev 30: 246 –253, 2008. Gluckman PD, Wyatt JS, Azzopardi D, Ballard R, Edwards AD, Ferriero DM, Polin RA, Robertson CM, Thoresen M, Whitelaw A, Gunn AJ. Selective head cooling with mild systemic hypothermia after neonatal encephalopathy: multicentre randomised trial. Lancet 365: 663– 670, 2005. Govindan RB, Massaro AN, Andescavage NN, Chang T, du Plessis A. Cerebral pressure passivity in newborns with encephalopathy undergoing therapeutic hypothermia. Front Hum Neurosci 8: 266, 2014. Haddad J, Constantinesco A, Brunot B, Messer J. A study of cerebral perfusion using single photon emission computed tomography in neonates with brain lesions. Acta Paediatr 83: 265–269, 1994. Halliday DM, Rosenberg JR, Amjad AM, Breeze P, Conway BA, Farmer SF. A framework for the analysis of mixed time series/point process data—theory and application to the study of physiological tremor, single motor unit discharges and electromyograms. Prog Biophys Mol Biol 64: 237–278, 1995. Howlett JA, Northington FJ, Gilmore MM, Tekes A, Huisman TA, Parkinson C, Chung SE, Jennings JM, Jamrogowicz JJ, Larson AC, Lehmann CU, Jackson E, Brady KM, Koehler RC, Lee JK. Cerebrovascular autoregulation and neurologic injury in neonatal hypoxic-ischemic encephalopathy. Pediatr Res 74: 525–535, 2013. Ilves P, Lintrop M, Metsvaht T, Vaher U, Talvik T. Cerebral blood-flow velocities in predicting outcome of asphyxiated newborn infants. Acta Paediatr 93: 523–528, 2004. Konishi Y, Kuriyama M, Mori I, Fujii Y, Konishi K, Sudo M, Ishii Y. Assessment of local cerebral blood flow in neonates with N-isopropyl-p[123I]iodoamphetamine and single photon emission computed tomography. Brain Dev 16: 450 – 453, 1994. Lassen NA. Autoregulation of cerebral blood flow. Circ Res 15, Suppl: 201–204, 1964. Laswad T, Wintermark P, Alamo L, Moessinger A, Meuli R, Gudinchet F. Method for performing cerebral perfusion-weighted MRI in neonates. Pediatr Radiol 39: 260 –264, 2009. Lemmers PM, Zwanenburg RJ, Benders MJ, de Vries LS, Groenendaal F, van Bel F, Toet MC. Cerebral oxygenation and brain activity after perinatal asphyxia: does hypothermia change their prognostic value? Pediatr Res 74: 180 –185, 2013. Liu L, Johnson HL, Cousens S, Perin J, Scott S, Lawn JE, Rudan I, Campbell H, Cibulskis R, Li M, Mathers C, Black RE. Global, regional, and national causes of child mortality: an updated systematic analysis for 2010 with time trends since 2000. Lancet 379: 2151–2161, 2012. Marks KA, Mallard EC, Roberts I, Williams CE, Sirimanne ES, Johnston B, Gluckman PD, Edwards AD. Delayed vasodilation and altered oxygenation after cerebral ischemia in fetal sheep. Pediatr Res 39: 48 –54, 1996. Meek JH, Elwell CE, McCormick DC, Edwards AD, Townsend JP, Stewart AL, Wyatt JS. Abnormal cerebral haemodynamics in perinatally asphyxiated neonates related to outcome. Arch Dis Child Fetal Neonatal Ed 81: F110 –F115, 1999. O’Leary H, Gregas MC, Limperopoulos C, Zaretskaya I, Bassan H, Soul JS, Di Salvo DN, du Plessis AJ. Elevated cerebral pressure passivity is
J Neurophysiol • doi:10.1152/jn.00353.2015 • www.jn.org
824
IMPAIRED CEREBRAL AUTOREGULATION IN HIE
associated with prematurity-related intracranial hemorrhage. Pediatrics 124: 302–309, 2009. Pryds O, Greisen G, Lou H, Friis-Hansen B. Vasoparalysis associated with brain damage in asphyxiated term infants. J Pediatr 117: 119 –125, 1990. Rutherford M, Ramenghi LA, Edwards AD, Brocklehurst P, Halliday H, Levene M, Strohm B, Thoresen M, Whitelaw A, Azzopardi D. Assessment of brain tissue injury after moderate hypothermia in neonates with hypoxic-ischaemic encephalopathy: a nested substudy of a randomised controlled trial. Lancet Neurol 9: 39 – 45, 2010. Sarnat HB, Sarnat MS. Neonatal encephalopathy following fetal distress. A clinical and electroencephalographic study. Arch Neurol 33: 696 –705, 1976. Shankaran S, Laptook AR, Ehrenkranz RA, Tyson JE, McDonald SA, Donovan EF, Fanaroff AA, Poole WK, Wright LL, Higgins RD, Finer NN, Carlo WA, Duara S, Oh W, Cotten CM, Stevenson DK, Stoll BJ, Lemons JA, Guillet R, Jobe AH. Whole-body hypothermia for neonates with hypoxic-ischemic encephalopathy. N Engl J Med 353: 1574 –1584, 2005. Short BL, Walker LK, Traystman RJ. Impaired cerebral autoregulation in the newborn lamb during recovery from severe, prolonged hypoxia, combined with carotid artery and jugular vein ligation. Crit Care Med 22: 1262–1268, 1994. Skov L, Pryds O, Greisen G. Estimating cerebral blood flow in newborn infants: comparison of near infrared spectroscopy and 133Xe clearance. Pediatr Res 30: 570 –573, 1991. Soul JS, Hammer PE, Tsuji M, Saul JP, Bassan H, Limperopoulos C, Disalvo DN, Moore M, Akins P, Ringer S, Volpe JJ, Trachtenberg F, du Plessis AJ. Fluctuating pressure-passivity is common in the cerebral circulation of sick premature infants. Pediatr Res 61: 467– 473, 2007. Tekes A, Poretti A, Scheurkogel MM, Huisman TA, Howlett JA, Alqahtani E, Lee JH, Parkinson C, Shapiro K, Chung SE, Jennings JM, Gilmore MM, Hogue CW, Martin LJ, Koehler RC, Northington FJ, Lee JK. Apparent diffusion coefficient scalars correlate with near-infrared spectroscopy markers of cerebrovascular autoregulation in neonates cooled for
perinatal hypoxic-ischemic injury. AJNR Am J Neuroradiol 36: 188 –193, 2015. Toet MC, Lemmers PM, van Schelven LJ, van Bel F. Cerebral oxygenation and electrical activity after birth asphyxia: their relation to outcome. Pediatrics 117: 333–339, 2006. Tsuji M, du Plessis A, Taylor G, Crocker R, Volpe JJ. Near infrared spectroscopy detects cerebral ischemia during hypotension in piglets. Pediatr Res 44: 591–595, 1998. Tsuji M, Naruse H, Volpe J, Holtzman D. Reduction of cytochrome aa3 measured by near-infrared spectroscopy predicts cerebral energy loss in hypoxic piglets. Pediatr Res 37: 253–259, 1995. van Bel F, Dorrepaal CA, Benders MJ, Zeeuwe PE, van de Bor M, Berger HM. Changes in cerebral hemodynamics and oxygenation in the first 24 hours after birth asphyxia. Pediatrics 92: 365–372, 1993. Volpe JJ, Herscovitch P, Perlman JM, Kreusser KL, Raichle ME. Positron emission tomography in the asphyxiated term newborn: parasagittal impairment of cerebral blood flow. Ann Neurol 17: 287–296, 1985. Winter JD, Tichauer KM, Gelman N, Thompson RT, Lee TY, St Lawrence K. Changes in cerebral oxygen consumption and high-energy phosphates during early recovery in hypoxic-ischemic piglets: a combined near-infrared and magnetic resonance spectroscopy study. Pediatr Res 65: 181–187, 2009. Wintermark P, Moessinger AC, Gudinchet F, Meuli R. Perfusion-weighted magnetic resonance imaging patterns of hypoxic-ischemic encephalopathy in term neonates. J Magn Reson Imaging 28: 1019 –1025, 2008. Younkin DP, Reivich M, Jaggi J, Obrist W, Delivoria-Papadopoulos M. Noninvasive method of estimating human newborn regional cerebral blood flow. J Cereb Blood Flow Metab 2: 415– 420, 1982. Zaramella P, Saraceni E, Freato F, Falcon E, Suppiej A, Milan A, Laverda AM, Chiandetti L. Can tissue oxygenation index (TOI) and cotside neurophysiological variables predict outcome in depressed/asphyxiated newborn infants? Early Hum Dev 83: 483– 489, 2007.
J Neurophysiol • doi:10.1152/jn.00353.2015 • www.jn.org
