RESEARCH—ANIMAL RESEARCH—ANIMAL
The Upper Limit of Cerebral Blood Flow Autoregulation Is Decreased With Elevations in Intracranial Pressure Matthew Pesek, MD* Kathleen Kibler, BS‡ R. Blaine Easley, MD*‡ Jennifer Mytar, BS§ Christopher Rhee, MD¶ Dean Andropoulos, MD‡ Ken Brady, MD*‡ *Department of Pediatrics, Division of Pediatric Critical Care, ‡Department of Anesthesiology, Division of Pediatric Anesthesiology, §College of Osteopathic Medicine, Touro University, Vallejo, California; Baylor College of Medicine, Texas Children’s Hospital, Houston, Texas; ¶Department of Pediatrics, Division of Neonatology, Baylor College of Medicine, Texas Children’s Hospital, Houston, Texas Correspondence: Ken Brady, MD, Texas Children’s Hospital, 6621 Fannin, WT 17-417B, Houston, TX. E-mail: [email protected] Received, November 22, 2013. Accepted, March 27, 2014. Published Online, April 15, 2014. Copyright © 2014 by the Congress of Neurological Surgeons.
BACKGROUND: The upper limit of cerebrovascular pressure autoregulation (ULA) is inadequately characterized. OBJECTIVE: To delineate the ULA in an infant swine model. METHODS: Neonatal piglets with sham surgery (n = 9), interventricular fluid infusion (INF) (n = 10), controlled cortical impact (CCI) (n = 10), or CCI 1 INF (n = 11) had intracranial pressure monitoring and bilateral cortical laser-Doppler flowmetry recordings during arterial hypertension to lethality using an aortic balloon catheter. An increase of red cell flux as a function of cerebral perfusion pressure was determined by piecewise linear regression, and static rates of autoregulation were determined above and below this inflection. The ULA was rendered as the first instance of an upward deflection of Doppler flux causing a static rate of autoregulation decrease greater than 0.5. RESULTS: ULA was identified in 55% of piglets after sham surgery, 70% after INF, 70% after CCI, and 91% after CCI with INF (P = .36). When identified, the median (interquartile range) ULA was as follows: sham group, 102 mm Hg (97-109 mm Hg); INF group, 75 mm Hg (52-84 mm Hg); CCI group, 81 mm Hg (69-101 mm Hg); and CCI 1 INF group, 61 mm Hg (52-57 mm Hg) (P = .01). In post hoc analysis, both groups with interventricular INF had significantly lower ULA than that observed in the sham group. CONCLUSION: Neonatal piglets without intracranial pathology tolerated acute hypertension with minimal perturbation of cerebral blood flow. Piglets with acutely increased intracranial pressure with or without trauma demonstrated loss of autoregulation when subjected to arterial hypertension. KEY WORDS: Autoregulation, Hypertension, Hydrocephalus, Pediatric, Traumatic brain injury Neurosurgery 75:163–170, 2014
DOI: 10.1227/NEU.0000000000000367
C
erebral blood flow (CBF) autoregulation is the constraint of CBF over a range of cerebral perfusion pressure (CPP) mediated by dynamic changes in cerebral vasculature resistance.1 The upper limit of autoregulation (ULA) is the CPP at which vascular constriction is unable to maintain constant CBF in the presence of arterial hypertension. Dysautoregulation due to hypertension was first termed “hypertensive breakthrough of autoregulation” ABBREVIATIONS: ABP, arterial blood pressure; CBF, cerebral blood flow; CCI, controlled cortical impact; CPP, cerebral perfusion pressure; ICP, intracranial pressure; INF, infusion; IQR, interquartile range; LLA, lower limit of autoregulation; SRoR, static rate of autoregulation; TBI, traumatic brain injury; ULA, upper limit of autoregulation
NEUROSURGERY
www.neurosurgery-online.com
by Lassen and Agnoli,2 who believed it to be related to hypertensive encephalopathy. In theory, increases in CPP beyond the ULA cause increases in CBF by forced dilation of the cerebral arterioles. In many of the original human and animal studies of the ULA, a number of animals failed to demonstrate a definitive ULA despite having injuriously high arterial blood pressures (ABPs).3,4 We sought to develop a neonatal swine model demonstrating the ULA to study dynamic autoregulation monitoring during hypertension. Previous efforts were hindered by the finding that half of the animals showed no change in CBF despite enduring lethal arterial hypertension. Rightward shifting of the ULA (ie, shifting to a higher CPP) is thought to be associated with chronic hypertension, which is not relevant to the neonatal swine model.3 Leftward shifting of the
VOLUME 75 | NUMBER 2 | AUGUST 2014 | 163
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
PESEK ET AL
ULA (ie, shifting to a lower CPP) has been observed in rats with acutely increased intracranial pressure (ICP) from infusion (INF) into the cisterna magna.5 Conversely, rightward shifting of the lower limit of autoregulation (LLA) has been observed in the neonatal swine model with an acute increase in ICP.6 This narrowing of the autoregulatory plateau is commonly thought to occur after traumatic brain injury (TBI), but was not observed in the neonatal swine after acute controlled cortical impact (CCI) with ICP less than 20 mm Hg.7,8 We hypothesized that the ULA would shift leftward from both trauma and increases in ICP. These hypotheses were tested in a systematic series of experiments to compare CBF autoregulation during induced arterial hypertension over 2 to 3 hours in neonatal swine. Four experimental groups, sham, ventricular INF, CCI, and CCI with ventricular INF, were compared during hypertension for presence of an ULA as well as the CPP at which the ULA was detected.
METHODS Approval was obtained by the Animal Care and Use Committee at the Baylor College of Medicine. A total of 40 male piglets (5-10 days old) were studied in 4 groups: sham surgery with no cortical impact (n = 9), ventricular fluid INF (n = 10), CCI (n = 10), and CCI 1 INF (n = 11).
Anesthesia Methods of anesthesia have been previously described.6,7 All piglets were induced with a mixture of inhaled 5% isoflurane, 50% nitrous oxide, and 50% oxygen. Tracheostomies were then performed, and mechanical ventilation was provided to maintain an arterial pH between 7.35 and 7.45, PaO2 between 200 and 300, and PCO2 (partial pressure of carbon dioxide) between 35 and 45. Maintenance anesthesia was administered with 0.8% isoflurane, 50% nitrous oxide, and 50% oxygen. Fentanyl (10 mg/kg bolus followed by 20 mg/kg/h INF) and vecuronium (2-mg bolus followed by 1-mg/kg/h INF) were administered after femoral central venous cannulation. ABP was measured and recorded from axillary cannulae with fluid pressure transducers zeroed to the level of the right atrium. Core body temperature was maintained with warming pads at between 38.5 and 39.5°C.
Surgery The femoral artery was cannulated, and a 5 French esophageal balloon catheter (Cooper Surgical, Trundall, Connecticut) was advanced to the distal aorta and used subsequently for controlled increasing of the ABP. All piglets, including sham, had a 10-mm craniotomy without disrupting the dura over the right hemisphere at the bregma, where CCI was performed in the relevant groups. Craniotomies were then done for placement of bilateral external ventricular drains for ICP monitoring, and artificial cerebrospinal fluid INF in the relevant groups, and bilateral laser Doppler flowmetry probes over the parietal cortex (for animals with impact craniotomy, the probe was placed 2 mm lateral to the impact craniotomy). ICP transducers were zeroed at the level of the external auditory meatus. Small craniotomies were sealed with dental cement. Baseline signal data were recorded for 1 hour before injury. Piglets in the injury groups (CCI and CCI 1 INF) had CCI with a 10-mm diameter piston, depth of 10 mm, velocity of 3 m/s, and dwell time of 300 ms. The preserved bone flap used for CCI was adhered back to the cranium after CCI with dental cement to preserve the integrity of the calvarium.
164 | VOLUME 75 | NUMBER 2 | AUGUST 2014
Injury was allowed to evolve for 6 hours before manipulation of blood pressure. Piglets that had induced intracranial hypertension (INF and CCI 1 INF) had artificial cerebrospinal fluid (KCl 3.0 mmol/L, MgCl2 0.6 mmol/L, CaCl2 1.3 mmol/L, NaCl 131.8 mmol/L, NaHCO3 24.6 mmol/L, urea 6.7 mmol/L, and glucose 3.6 mmol/L) infused into an external ventricular catheter at varying rates to achieve a steady-state ICP value of 20 to 25 mm Hg.
Signaling Sampling ABP, ICP, and laser Doppler flowmetry were sampled with an analogdigital converter (Data Translation, Marlboro, Massachusetts) at 200 Hz using ICM1 software (Cambridge University, Cambridge, United Kingdom). Mean values across the cardiac and respiratory cycles were recorded as 10-s time-averaged means. Laser Doppler flowmetry values were normalized as a percentage of baseline values using a 30-minute recording at baseline (100%) and the flux reading obtained at demise (0%).
Determining the ULA and Static Rate of Autoregulation Autoregulation was quantified from recordings of CPP and cortical red cell flux. Laser Doppler flowmetry data were plotted as a function of CPP, and piecewise linear regression was applied to this plot to determine the intersection of 2 regression lines with minimized residual error squared. This intersection was considered the putative ULA.6 However, this method always renders a candidate ULA, even when there is no change in CBF. Therefore, we determined the static rate of autoregulation (SRoR = %D cerebrovascular resistance/%DCPP) above and below each putative ULA. A decrease in SRoR of $0.5 caused by hypertension was required to accept the ULA rendered by piecewise regression. An SRoR value of 1.0 is considered to be perfect autoregulation where changes in CPP are proportionally matched with changes in cerebrovascular resistance. An SRoR less than 0.5 is considered impaired autoregulation, so a decrease of 0.5 across the putative ULA is confirmation of a demarcation between an intact and an impaired state.6 Examples of 2 animals with and without an identifiable ULA are shown in Figure 1. Both right and left cortical flux measurements were recorded, and a detectable ULA from either side that met the specified criteria was included in the analysis. If both sides rendered a valid ULA, then the 2 breakpoints were averaged. For animals with CCI, there was a detectable ULA ipsilateral to the injury in 4 animals (40%) and contralateral to the injury in 8 animals (80%). For animals with CCI 1 INF, there was a detectable ULA ipsilateral to impact in 7 animals (64%) and contralateral to impact in 6 animals (55%). When the ULA was detectable both ipsilateral and contralateral to injury in the groups with CCI (n = 9 total), there was no trend for differences relative to the side monitored: median (interquartile range [IQR]) ipsilateral ULA was 74 mm Hg (65-95 mm Hg), and contralateral ULA was 72 mm Hg (64-92 mm Hg) (P = 1.0, Student t test).
Statistical Analysis Descriptive statistics were performed on physiological measurements for the groups and presented as median and IQR. Differences between groups and across the stages of the experiment were compared with 2-way repeatedmeasures analysis of variance. Groups were compared for the presence or absence of a ULA with the x2 test. Subjects with an identifiable ULA were further compared across experimental groups for the range of ULA using a Kruskal-Wallis analysis of variance, and post hoc comparisons were made between sham and all other groups using Dunn’s multiple comparison test. All statistical analyses were performed using GraphPad Prism version 5.04 (for Mac, GraphPad Software, La Jolla, California).
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
NEONATAL AUTOREGULATION DURING HYPERTENSION
FIGURE 1. Animals with and without an identifiable upper limit of autoregulation (ULA). The top 2 panels of both A and B show arterial blood pressure (ABP) and intracranial pressure (ICP) as a function of time. The bottom panels of both A and B show cerebral blood flow (CBF) as a function of cerebral perfusion pressure (CPP). A, a piglet with an identifiable ULA by piecewise regression at a CPP of 84. In this animal, static rate of autoregulation (SRoR) decreased from 0.6 (intact) below the ULA to 0.1 (impaired) above the ULA, confirming a shift from intact to impaired autoregulation at the putative demarcation. B, a piglet with no identifiable ULA rendered a putative inflection at a CPP of 107.5 by piecewise regression. However, SRoR above the putative ULA was 0.93 and below the putative ULA was 0.92, demonstrating intact autoregulation throughout the induced hypertension.
RESULTS A total of 40 neonatal piglets were randomized to 4 experimental groups. Two animals were excluded: 1 in the sham group did not meet experiment criteria and died before achieving any appreciable increase in CPP and 1 in the CCI group was excluded
NEUROSURGERY
due to LDF failure. Survival to a CPP of 100 mm Hg occurred in 100% of the sham group, 80% of the INF group, 90% of the CCI group, and 91% of the CCI 1 INF group. Survival to CPP of 120 mm Hg occurred in 66% of the sham group, 20% of the INF group, 80% of the CCI group, and 27% of the CCI 1 INF group. The relative inability of the experimental groups with INF to
VOLUME 75 | NUMBER 2 | AUGUST 2014 | 165
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
PESEK ET AL
survive to a higher CPP is accounted for by the difference in ICP between the infused and noninfused groups because a much higher ABP was required in these groups to achieve the same CPP. Physiological variables are compared across the experimental groups in Table. There were no significant group differences with respect to ABP and CPP over the course of the experiment. However, ICP was significantly higher in the INF and CCI 1 INF groups. Temperature was lower at baseline in the CCI group, but returned to normal values during the 6-hour recovery and during arterial hypertension. The ventilation
and fluid management strategy kept pH and PCO2 within normal range across the experiment, and there were no group differences. The CCI group had a higher PO2 (partial pressure of oxygen) than the other subject groups, but PO2 values were consistently greater than 150 mm Hg and less than 300 mm Hg for all animals. Serum sodium was lower at baseline in the CCI group, but was restored to normal in all groups during recovery and during the induced hypertension phases. Serum lactate levels were low and similar across groups and experimental phases.
TABLE. Physiological Data by Experimental Group and Phasea Sham Weight, kg Age, d ABP, mm Hg B R H ICP, mm Hg B R H CPP, mm Hg B R H Temp, °C B R H pH B R H PCO2, torr B R H PO2, torr B R H Na, mg/dL B R H Serum lactate, mg/dL B R H
2.6 (2.1-3.7) 8.0 (5-9.5) 67 (60-71) 60 (52.5-76.5) 112 (104-121.5)
Infusion 2.4 (2.3-3.2) 7.5 (4.8-10.2) 74.5 (62.5-78.2) 72 (63.8-76.5) 128 (114-131)
CCI 2.4 (2.1-2.8) 9.5 (7-12) 74 (65.8-83.5) 73 (67.8-83.2) 96 (80-119)
CCI 1 Infusion
P Valueb
3.8 (2.8-4.3) 9.0 (4-13)
.06 .50
65 (59-81) 65 (62-72) 121 (110-127)
NS, ,.0001
9 (6.3-11) 23 (21.8-25) 25 (24-32.3)
10 (6.8-12) 14.5 (9-21.3) 13 (10.5-19.3)
9 (6-12) 20 (19-25) 28 (23-32)
,.0001, ,.0001
62 (57-64) 53 (50-64.5) 103 (98.5-111.5)
64 (51.8-72) 48.5 (40-51.5) 96.5 (91.5-101.8)
65 (59-69.3) 57 (46.8-69.3) 83.5 (62.3107.8)
55 (53-67) 45 (43-47) 96 (81-105)
NS, ,.0001
38.1 (37.8-39) 38.8 (37.9-39.2) 38.7 (37.9-39.2)
38.9 (38.1-39.3) 38.7 (38.4-38.9) 38.4 (38.1-38.9)
36.5 (39.4-38.4) 38.1 (36.6-38.9) 37.8 (37.4-38.2)
38.4 (37.2-38.1) 38.1 (37.3-38.9) 37.5 (36.9-38.7)
.027, NS
7.40 (7.38-7.46) 7.46 (7.43-7.49) 7.45 (7.44-7.48)
7.42 (7.38-7.44) 7.39 (7.38-7.43) 7.39 (7.32-7.47)
7.44 (7.40-7.45) 7.41 (7.36-7.46) 7.41 (7.34-7.43)
7.40 (7.37-7.41) 7.44 (7.40-7.46) 7.42 (7.40-7.45)
NS, NS
37.8 (35-42.4) 40.2 (39.4-40.6) 38.7 (35.9-42.5)
39.9 (36.2-41) 41.3 (37.5-43.6) 39.2 (36.3-43.9)
37.8 (34.2-40.2) 37.8 (35.5-46.5) 40.4 (35.8-44.5)
40.2 (39.1-41.8) 40.4 (36.3-42.1) 39.6 (37.2-43.2)
NS, NS
206 (187-244) 185 (178-214) 202 (183-213)
209 (169-226) 199 (145-220) 206 (174-249)
247 (235-255) 227 (220-257) 223 (197-249)
206 (188-221) 202 (178-219) 202 (175-223)
.0005, NS
142 (141-144) 144 (140-147) 140 (138-147)
142 (139-145) 142 (139-145) 142 (140-147)
138 (137-140) 140 (135-142) 140 (138-143)
141 (139-143) 141 (140-142) 140 (140-141)
.001, NS
2.6 (1.7-3.2) 1.4 (1.2-3.1) 2.0 (1.5-2.9)
1.7 (1.3-2.5) 1.1 (0.9-2.2) 1.5 (0.9-2.3)
1.6 (1.3-2.5) 1.0 (0.4-3.1) 1.0 (0.6-3.2)
1.7 (1.0-1.9) 1.0 (0.8-1.4) 1.4 (1.1-2.1)
NS, NS
6 (2.5-9) 5 (4-8.5) 7 (2.5-9)
a CCI, controlled cortical impact; ABP, arterial blood pressure; NS, not significant; ICP, intracranial pressure; CPP, cerebral perfusion pressure; Temp, temperature; PCO2, partial pressure of carbon dioxide; PO2, partial pressure of oxygen; Na, serum sodium concentration. Experimental phases: B, baseline phase, R, recovery phase, H, hypertensive phase. Data are presented as median and range. b P values are Kruskal-Wallis analysis of variance for weight and age; all other analyses are 2-way repeated-measures analyses of variance reported across subject group and experimental phase.
166 | VOLUME 75 | NUMBER 2 | AUGUST 2014
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
NEONATAL AUTOREGULATION DURING HYPERTENSION
CBF is plotted as a percentage of baselines from the laser Doppler flowmetry recordings in Figure 2 for visual inspection of the differences between groups during hypertension. Differences between the groups with respect to the presence of a detectable ULA were not statistically significant: 55% of animals in the sham group demonstrated a detectable ULA by the criterion specified. In both the INF and CCI group, 70% of animals were determined to have a ULA. The CCI 1 INF group demonstrated a ULA in 91% of animals (P = .36). However, the leftward shifting of the ULA with addition of progressive injury that is evident in Figure 1 was verified by comparing the CPP at ULA across the groups (Figure 3). The median (IQR) ULA was detected at CPP of 102 mm Hg (97-109 mm Hg) in the sham group, 75 mm Hg (52-84 mm Hg) in the INF group, 81 mm Hg (69-101 mm Hg) in the CCI group, and 61 mm Hg (53-99 mm Hg) in the CCI 1 INF group (P = .01). Although there was a trend toward a lower ULA in the CCI group compared with the sham group, the difference was not statistically significant on post hoc analysis. The higher ULA, however, was statistically different in both INF groups when compared with the sham group.
DISCUSSION The primary objective of this study was to develop a model of injury that would reliably render a ULA. The main finding of this study is that the combination of CCI and induced ICP caused significant decreases in the ULA to near normal CPP values. Interestingly, increases in ICP without trauma caused a greater shift in the ULA than CCI alone. Although the effects of CCI produced similar results on the ULA, they were not of statistical significance. This trend could be explained in part by the primary injury itself and by the fact that the animals with CCI had a higher mean ICP compared with sham animals. However, the groups with ventricular INFs had their mean ICP maintained at greater than 20 mm Hg, which was higher than either the sham or CCI group. Our results are consistent with the findings of Hauerberg et al,5 who showed in mature rats that the ULA is left-shifted by acute increases in ICP in a dose-dependent fashion. Further, when combined with previous studies of the LLA, this dataset can estimate the change in the range of the autoregulatory plateau of neonatal swine caused by acute intracranial hypertension.
FIGURE 2. Cerebral blood flow (CBF) is shown normalized to baseline and plotted as a function of cerebral perfusion pressure (CPP) for each of the 4 groups: sham (n = 9), ventricular infusion (infusion, n = 10), controlled cortical impact (CCI, n = 10), and CCI with ventricular infusion (CCI & infusion n = 11). Data are shown as mean and SD at each 5-mm Hg increment of CPP. The median group upper limit of autoregulation is shown in each graph as a vertical dashed line.
NEUROSURGERY
VOLUME 75 | NUMBER 2 | AUGUST 2014 | 167
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
PESEK ET AL
FIGURE 3. Presence of a detectable upper limit of autoregulation (ULA) and the cerebral perfusion pressure (CPP) at ULA. The ULA was found at a median (interquartile range) CPP of 102 mm Hg (97-109 mm Hg) in the sham group (n = 9), 75 mm Hg (52-84 mm Hg) in the infusion group (Inf, n = 10), 81 mm Hg (69-101 mm Hg) in the closed cortical impact group (CCI, n = 10), and 61 mm Hg (53-99 mm Hg) in the CCI 1 Inf (n = 11) group. Box whiskers graphs show median, interquartile range, and range; P = .01.
Naive neonatal swine have an LLA near 30 mm Hg CPP, and the sham group in this study had a median ULA near 100 mm Hg, giving a plateau spanning 70 mm Hg. Swine of the same age with ICP acutely increased to 20 mm Hg have a modest increase in LLA to near 40 mm Hg, and when ICP is increased to 40 mm Hg, the LLA increases to 50 mm Hg.6 The effect of an acute increase in ICP in this study has a more profound effect on the ULA than was observed in the LLA. An ICP of 20 mm Hg results in a median ULA near 80 mm Hg. Thus, an acute increase in ICP to 20 mm Hg can be estimated to halve the width of the autoregulatory plateau from a 70-mm Hg span to a 35-mm Hg span, with most of the effect caused on the hypertensive end. We are unable to address the dose response because attempts to maintain hypertension with acute increases in ICP to 40 mm Hg in this age animal resulted in the immediate death of the animal.
168 | VOLUME 75 | NUMBER 2 | AUGUST 2014
Interestingly, acute neurological injury without an increase in ICP greater than 20 mm Hg produced a trend of decreased ULA that was of smaller magnitude than the increased ICP group and was not significant when compared with the sham group. The same CCI depth, velocity, piston diameter, dwell time, and recovery duration were used in a previous study of neonatal swine to examine the effect of acute injury on the LLA. The injury used was unilateral, but of a severity that caused an 18% acute injury volume in a similar cohort and an acute increase in ICP from 6 mm Hg to 17 mm Hg. The mean LLA of 35 mm Hg in that cohort was not statistically increased compared with a sham group. We conclude that acute brain injury without an increase in ICP greater than 20 mm Hg in this model causes less impairment of CBF autoregulation than does an atraumatic increase in ICP acutely. Taken together, these studies suggest that the ULA is left-shifted by a larger magnitude than the LLA is right-shifted for the same insult, which is provocative when common management of brain trauma resuscitation is to increase ABP aggressively before ICP is even known. However, these data imply that during acute resuscitation from neurological insult, the ability of the cerebral vasculature to autoregulate is improved by control of ICP and avoidance of prolonged hypertension. Such an interpretation is tempered by the finding in adults that dysautoregulation due to hypertension is associated with worse neurological performance at outcome, but dysautoregulation due to hypotension is associated with death.9 The most recent pediatric TBI guidelines propose a minimum CPP of 40 mm Hg for children and an age-specific CPP range of 40 to 50 mm Hg with infants being at the lower end and adolescents at the upper end of this range.10 Due to the paucity of evidence, only level III recommendations can be made regarding CPP thresholds. Numerous studies have shown that hypotension after brain injury is associated with increased mortality.11-14 During the initial resuscitation, systolic blood pressures should be maintained above fifth percentile for age to avoid ischemic injury.15 However, there are limited data to suggest a maximum threshold for systolic blood pressure. One study done by White et al16 found a 19-fold increase in survival after TBI in pediatric patients with a systolic blood pressure greater than 135 mm Hg. The data of White et al appear to contradict previous TBI studies that showed worse neurological outcomes in the setting of arterial hypertension.17,18 These seemingly conflicting data might be conceptually reconciled by the adult trauma study of Aries et al9 in which a CPP less than optimal for autoregulation was associated with death and a CPP greater than optimal for autoregulation was associated with survival but worsening neurological outcomes. Sex differences after TBI with regard to autoregulatory disturbances have been documented in piglets, with male piglets having increased activity of vasoconstrictive endothelin-1. These male piglets showed impaired pressure autoregulation during hypotension that was aggravated by phenylephrine. By contrast, female piglets had diminished endothelin-1 release after TBI and autoregulation that was improved by phenylephrine.19 We cannot speak to sex differences at the upper limit of autoregulation in this
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
NEONATAL AUTOREGULATION DURING HYPERTENSION
all-male cohort. The biochemical pathways identified in that study are related to vasodilatory responses to hypotension, and it is not known whether or how increased ICP influences mediators of vasoconstriction during hypertension. Brain trauma is expected to evolve over 3 to 5 days after injury, and these data cannot speak to potential temporal changes of the autoregulatory range. This study is limited by the sample size and the inability to detect the significance of a modest difference between the sham and CCI groups with respect to the ULA position. In part, this is because 40% of the sham group and 30% of the CCI group were excluded due to our inability to delineate the ULA before the death of the piglet. Our clinical observation was death due to acute congestive heart failure with frank pulmonary edema and hemorrhagic myocardium seen on autopsy after a sustained doubling of the ABP. By contrast, we have been able to detect the LLA in 100% of subjects studied in previous efforts.6 A reliable model demarcating the ULA would be useful in the study of autoregulation and vascular reactivity monitoring to delineate optimal perfusion pressures.9 However, the lack of consistent ULA delineation in the piglet model presents a challenge to finding a gold standard against which to test the sensitivity and specificity of these monitoring techniques during hypertension as we have done against the LLA.20-23
CONCLUSION The primary finding of this study is that acute increases in ICP impair CBF autoregulation in neonatal piglets with arterial hypertension. This effect was observed in the presence and absence of trauma. Disclosure The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.
REFERENCES 1. Lassen NA. Cerebral blood flow and oxygen consumption in man. Physiol Rev. 1959;39(2):183-238. 2. Lassen NA, Agnoli A. The upper limit of autoregulation of cerebral blood flow— on the pathogenesis of hypertensive encephalopathy. Scand J Clin Lab Invest. 1972; 30(2):113-116. 3. Strandgaard S, Olesen J, Skinhoj E, Lassen NA. Autoregulation of brain circulation in severe arterial hypertension. Br Med J. 1973;1(5852):507-510. 4. Strandgaard S, MacKenzie ET, Sengupta D, Rowan JO, Lassen NA, Harper AM. Upper limit of autoregulation of cerebral blood flow in the baboon. Circ Res. 1974; 34(4):435-440. 5. Hauerberg J, Xiaodong M, Willumsen L, Pedersen DB, Juhler M. The upper limit of cerebral blood flow autoregulation in acute intracranial hypertension. J Neurosurg Anesthesiol. 1998;10(2):106-112. 6. Brady KM, Lee JK, Kibler KK, et al. The lower limit of cerebral blood flow autoregulation is increased with elevated intracranial pressure. Anesth Analg. 2009; 108(4):1278-1283. 7. Mytar J, Kibler KK, Easley RB, et al. Static autoregulation is intact early after severe unilateral brain injury in a neonatal swine model. Neurosurgery. 2012;71(1): 138-145. 8. Rangel-Castilla L, Gasco J, Nauta HJ, Okonkwo DO, Robertson CS. Cerebral pressure autoregulation in traumatic brain injury. Neurosurg Focus. 2008;25(4):E7.
NEUROSURGERY
9. Aries MJ, Czosnyka M, Budohoski KP, et al. Continuous determination of optimal cerebral perfusion pressure in traumatic brain injury. Crit Care Med. 2012; 40(8):2456-2463. 10. Kochanek PM, Carney N, Adelson PD, et al. Guidelines for the acute medical management of severe traumatic brain injury in infants, children, and adolescents—second edition. Pediatr Crit Care Med. 2012;13(suppl 1): S1-S82. 11. Pigula FA, Wald SL, Shackford SR, Vane DW. The effect of hypotension and hypoxia on children with severe head injuries. J Pediatr Surg. 1993;28(3):310-314; discussion 315-316. 12. Luerssen TG, Klauber MR, Marshall LF. Outcome from head injury related to patient’s age. A longitudinal prospective study of adult and pediatric head injury. J Neurosurg. 1988;68(3):409-416. 13. Ong L, Selladurai BM, Dhillon MK, Atan M, Lye MS. The prognostic value of the Glasgow Coma Scale, hypoxia and computerised tomography in outcome prediction of pediatric head injury. Pediatr Neurosurg. 1996;24(6):285-291. 14. Mayer TA, Walker ML. Pediatric head injury: the critical role of the emergency physician. Ann Emerg Med. 1985;14(12):1178-1184. 15. Adelson PD, Bratton SL, Carney NA, et al. Guidelines for the acute medical management of severe traumatic brain injury in infants, children, and adolescents. Chapter 4. Resuscitation of blood pressure and oxygenation and prehospital brainspecific therapies for the severe pediatric traumatic brain injury patient. Pediatr Crit Care Med. 2003;4(suppl 3):S12-S18. 16. White JR, Farukhi Z, Bull C, et al. Predictors of outcome in severely head-injured children. Crit Care Med. 2001;29(3):534-540. 17. Kanter RK, Carroll JB, Post EM. Association of arterial hypertension with poor outcome in children with acute brain injury. Clin Pediatr (Phila). 1985;24(6): 320-323. 18. Brink JD, Imbus C, Woo-Sam J. Physical recovery after severe closed head trauma in children and adolescents. J Pediatr. 1980;97(5):721-727. 19. Armstead WM, Riley J, Vavilala MS. TBI sex dependently upregulates ET-1 to impair autoregulation, which is aggravated by phenylephrine in males but is abrogated in females. J Neurotrauma. 2012;29(7):1483-1490. 20. Brady KM, Easley RB, Kibler K, et al. Positive end-expiratory pressure oscillation facilitates brain vascular reactivity monitoring. J Appl Physiol (1985). 2012;113(9): 1362-1368. 21. Lee JK, Kibler KK, Benni PB, et al. Cerebrovascular reactivity measured by nearinfrared spectroscopy. Stroke. 2009;40(5):1820-1826. 22. Brady KM, Lee JK, Kibler KK, Easley RB, Koehler RC, Shaffner DH. Continuous measurement of autoregulation by spontaneous fluctuations in cerebral perfusion pressure: comparison of 3 methods. Stroke. 2008;39(9):2531-2537. 23. Brady KM, Lee JK, Kibler KK, et al. Continuous time-domain analysis of cerebrovascular autoregulation using near-infrared spectroscopy. Stroke. 2007;38 (10):2818-2825.
COMMENTS
T
he upper limit of autoregulation (ULA) has received relatively less investigation than the lower limit of autoregulation after traumatic brain injury (TBI). This study seeks to better understand the ability of the cerebral circulation to handle hypertension using a neonatal swine model. They hypothesized that the ULA would be shifted to the left by both trauma and intracranial hypertension. They found the greatest shift in ULA to occur with the combined injury and induced intracranial hypertension. These results are interesting and have clinical relevance for how aggressive the initial resuscitation of blood pressure should be in patients with TBI. It also implies that for patients with intracranial hypertension lowering intracranial pressure would be more physiological than increasing pressure to improve cerebral perfusion pressure. Claudia Robertson Houston, Texas
I
n this well-designed swine model study, the authors investigate the upper limit autoregulation (ULA) in response to traumatic brain
VOLUME 75 | NUMBER 2 | AUGUST 2014 | 169
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
PESEK ET AL
injury (TBI) and intracranial hypertension. Results were as expected; TBI and secondary intracranial hypertension resulted in a significant shift of ULA. In clinical practice, these results emphasize the adequate management of blood pressure in patients with TBI. During acute resuscitation, cerebral autoregulation is improved by avoidance of prolonged hypertension and control of intracranial pressure. Adequate
170 | VOLUME 75 | NUMBER 2 | AUGUST 2014
cerebral perfusion pressure should be obtained by lowering intracranial pressure and not by inducing systemic blood pressure. More is not always better. Leonardo Rangel-Castilla Phoenix, Arizona
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
The Upper Limit of Cerebral Blood Flow Autoregulation Is Decreased With Elevations in Intracranial Pressure Matthew Pesek, MD* Kathleen Kibler, BS‡ R. Blaine Easley, MD*‡ Jennifer Mytar, BS§ Christopher Rhee, MD¶ Dean Andropoulos, MD‡ Ken Brady, MD*‡ *Department of Pediatrics, Division of Pediatric Critical Care, ‡Department of Anesthesiology, Division of Pediatric Anesthesiology, §College of Osteopathic Medicine, Touro University, Vallejo, California; Baylor College of Medicine, Texas Children’s Hospital, Houston, Texas; ¶Department of Pediatrics, Division of Neonatology, Baylor College of Medicine, Texas Children’s Hospital, Houston, Texas Correspondence: Ken Brady, MD, Texas Children’s Hospital, 6621 Fannin, WT 17-417B, Houston, TX. E-mail: [email protected] Received, November 22, 2013. Accepted, March 27, 2014. Published Online, April 15, 2014. Copyright © 2014 by the Congress of Neurological Surgeons.
BACKGROUND: The upper limit of cerebrovascular pressure autoregulation (ULA) is inadequately characterized. OBJECTIVE: To delineate the ULA in an infant swine model. METHODS: Neonatal piglets with sham surgery (n = 9), interventricular fluid infusion (INF) (n = 10), controlled cortical impact (CCI) (n = 10), or CCI 1 INF (n = 11) had intracranial pressure monitoring and bilateral cortical laser-Doppler flowmetry recordings during arterial hypertension to lethality using an aortic balloon catheter. An increase of red cell flux as a function of cerebral perfusion pressure was determined by piecewise linear regression, and static rates of autoregulation were determined above and below this inflection. The ULA was rendered as the first instance of an upward deflection of Doppler flux causing a static rate of autoregulation decrease greater than 0.5. RESULTS: ULA was identified in 55% of piglets after sham surgery, 70% after INF, 70% after CCI, and 91% after CCI with INF (P = .36). When identified, the median (interquartile range) ULA was as follows: sham group, 102 mm Hg (97-109 mm Hg); INF group, 75 mm Hg (52-84 mm Hg); CCI group, 81 mm Hg (69-101 mm Hg); and CCI 1 INF group, 61 mm Hg (52-57 mm Hg) (P = .01). In post hoc analysis, both groups with interventricular INF had significantly lower ULA than that observed in the sham group. CONCLUSION: Neonatal piglets without intracranial pathology tolerated acute hypertension with minimal perturbation of cerebral blood flow. Piglets with acutely increased intracranial pressure with or without trauma demonstrated loss of autoregulation when subjected to arterial hypertension. KEY WORDS: Autoregulation, Hypertension, Hydrocephalus, Pediatric, Traumatic brain injury Neurosurgery 75:163–170, 2014
DOI: 10.1227/NEU.0000000000000367
C
erebral blood flow (CBF) autoregulation is the constraint of CBF over a range of cerebral perfusion pressure (CPP) mediated by dynamic changes in cerebral vasculature resistance.1 The upper limit of autoregulation (ULA) is the CPP at which vascular constriction is unable to maintain constant CBF in the presence of arterial hypertension. Dysautoregulation due to hypertension was first termed “hypertensive breakthrough of autoregulation” ABBREVIATIONS: ABP, arterial blood pressure; CBF, cerebral blood flow; CCI, controlled cortical impact; CPP, cerebral perfusion pressure; ICP, intracranial pressure; INF, infusion; IQR, interquartile range; LLA, lower limit of autoregulation; SRoR, static rate of autoregulation; TBI, traumatic brain injury; ULA, upper limit of autoregulation
NEUROSURGERY
www.neurosurgery-online.com
by Lassen and Agnoli,2 who believed it to be related to hypertensive encephalopathy. In theory, increases in CPP beyond the ULA cause increases in CBF by forced dilation of the cerebral arterioles. In many of the original human and animal studies of the ULA, a number of animals failed to demonstrate a definitive ULA despite having injuriously high arterial blood pressures (ABPs).3,4 We sought to develop a neonatal swine model demonstrating the ULA to study dynamic autoregulation monitoring during hypertension. Previous efforts were hindered by the finding that half of the animals showed no change in CBF despite enduring lethal arterial hypertension. Rightward shifting of the ULA (ie, shifting to a higher CPP) is thought to be associated with chronic hypertension, which is not relevant to the neonatal swine model.3 Leftward shifting of the
VOLUME 75 | NUMBER 2 | AUGUST 2014 | 163
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
PESEK ET AL
ULA (ie, shifting to a lower CPP) has been observed in rats with acutely increased intracranial pressure (ICP) from infusion (INF) into the cisterna magna.5 Conversely, rightward shifting of the lower limit of autoregulation (LLA) has been observed in the neonatal swine model with an acute increase in ICP.6 This narrowing of the autoregulatory plateau is commonly thought to occur after traumatic brain injury (TBI), but was not observed in the neonatal swine after acute controlled cortical impact (CCI) with ICP less than 20 mm Hg.7,8 We hypothesized that the ULA would shift leftward from both trauma and increases in ICP. These hypotheses were tested in a systematic series of experiments to compare CBF autoregulation during induced arterial hypertension over 2 to 3 hours in neonatal swine. Four experimental groups, sham, ventricular INF, CCI, and CCI with ventricular INF, were compared during hypertension for presence of an ULA as well as the CPP at which the ULA was detected.
METHODS Approval was obtained by the Animal Care and Use Committee at the Baylor College of Medicine. A total of 40 male piglets (5-10 days old) were studied in 4 groups: sham surgery with no cortical impact (n = 9), ventricular fluid INF (n = 10), CCI (n = 10), and CCI 1 INF (n = 11).
Anesthesia Methods of anesthesia have been previously described.6,7 All piglets were induced with a mixture of inhaled 5% isoflurane, 50% nitrous oxide, and 50% oxygen. Tracheostomies were then performed, and mechanical ventilation was provided to maintain an arterial pH between 7.35 and 7.45, PaO2 between 200 and 300, and PCO2 (partial pressure of carbon dioxide) between 35 and 45. Maintenance anesthesia was administered with 0.8% isoflurane, 50% nitrous oxide, and 50% oxygen. Fentanyl (10 mg/kg bolus followed by 20 mg/kg/h INF) and vecuronium (2-mg bolus followed by 1-mg/kg/h INF) were administered after femoral central venous cannulation. ABP was measured and recorded from axillary cannulae with fluid pressure transducers zeroed to the level of the right atrium. Core body temperature was maintained with warming pads at between 38.5 and 39.5°C.
Surgery The femoral artery was cannulated, and a 5 French esophageal balloon catheter (Cooper Surgical, Trundall, Connecticut) was advanced to the distal aorta and used subsequently for controlled increasing of the ABP. All piglets, including sham, had a 10-mm craniotomy without disrupting the dura over the right hemisphere at the bregma, where CCI was performed in the relevant groups. Craniotomies were then done for placement of bilateral external ventricular drains for ICP monitoring, and artificial cerebrospinal fluid INF in the relevant groups, and bilateral laser Doppler flowmetry probes over the parietal cortex (for animals with impact craniotomy, the probe was placed 2 mm lateral to the impact craniotomy). ICP transducers were zeroed at the level of the external auditory meatus. Small craniotomies were sealed with dental cement. Baseline signal data were recorded for 1 hour before injury. Piglets in the injury groups (CCI and CCI 1 INF) had CCI with a 10-mm diameter piston, depth of 10 mm, velocity of 3 m/s, and dwell time of 300 ms. The preserved bone flap used for CCI was adhered back to the cranium after CCI with dental cement to preserve the integrity of the calvarium.
164 | VOLUME 75 | NUMBER 2 | AUGUST 2014
Injury was allowed to evolve for 6 hours before manipulation of blood pressure. Piglets that had induced intracranial hypertension (INF and CCI 1 INF) had artificial cerebrospinal fluid (KCl 3.0 mmol/L, MgCl2 0.6 mmol/L, CaCl2 1.3 mmol/L, NaCl 131.8 mmol/L, NaHCO3 24.6 mmol/L, urea 6.7 mmol/L, and glucose 3.6 mmol/L) infused into an external ventricular catheter at varying rates to achieve a steady-state ICP value of 20 to 25 mm Hg.
Signaling Sampling ABP, ICP, and laser Doppler flowmetry were sampled with an analogdigital converter (Data Translation, Marlboro, Massachusetts) at 200 Hz using ICM1 software (Cambridge University, Cambridge, United Kingdom). Mean values across the cardiac and respiratory cycles were recorded as 10-s time-averaged means. Laser Doppler flowmetry values were normalized as a percentage of baseline values using a 30-minute recording at baseline (100%) and the flux reading obtained at demise (0%).
Determining the ULA and Static Rate of Autoregulation Autoregulation was quantified from recordings of CPP and cortical red cell flux. Laser Doppler flowmetry data were plotted as a function of CPP, and piecewise linear regression was applied to this plot to determine the intersection of 2 regression lines with minimized residual error squared. This intersection was considered the putative ULA.6 However, this method always renders a candidate ULA, even when there is no change in CBF. Therefore, we determined the static rate of autoregulation (SRoR = %D cerebrovascular resistance/%DCPP) above and below each putative ULA. A decrease in SRoR of $0.5 caused by hypertension was required to accept the ULA rendered by piecewise regression. An SRoR value of 1.0 is considered to be perfect autoregulation where changes in CPP are proportionally matched with changes in cerebrovascular resistance. An SRoR less than 0.5 is considered impaired autoregulation, so a decrease of 0.5 across the putative ULA is confirmation of a demarcation between an intact and an impaired state.6 Examples of 2 animals with and without an identifiable ULA are shown in Figure 1. Both right and left cortical flux measurements were recorded, and a detectable ULA from either side that met the specified criteria was included in the analysis. If both sides rendered a valid ULA, then the 2 breakpoints were averaged. For animals with CCI, there was a detectable ULA ipsilateral to the injury in 4 animals (40%) and contralateral to the injury in 8 animals (80%). For animals with CCI 1 INF, there was a detectable ULA ipsilateral to impact in 7 animals (64%) and contralateral to impact in 6 animals (55%). When the ULA was detectable both ipsilateral and contralateral to injury in the groups with CCI (n = 9 total), there was no trend for differences relative to the side monitored: median (interquartile range [IQR]) ipsilateral ULA was 74 mm Hg (65-95 mm Hg), and contralateral ULA was 72 mm Hg (64-92 mm Hg) (P = 1.0, Student t test).
Statistical Analysis Descriptive statistics were performed on physiological measurements for the groups and presented as median and IQR. Differences between groups and across the stages of the experiment were compared with 2-way repeatedmeasures analysis of variance. Groups were compared for the presence or absence of a ULA with the x2 test. Subjects with an identifiable ULA were further compared across experimental groups for the range of ULA using a Kruskal-Wallis analysis of variance, and post hoc comparisons were made between sham and all other groups using Dunn’s multiple comparison test. All statistical analyses were performed using GraphPad Prism version 5.04 (for Mac, GraphPad Software, La Jolla, California).
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
NEONATAL AUTOREGULATION DURING HYPERTENSION
FIGURE 1. Animals with and without an identifiable upper limit of autoregulation (ULA). The top 2 panels of both A and B show arterial blood pressure (ABP) and intracranial pressure (ICP) as a function of time. The bottom panels of both A and B show cerebral blood flow (CBF) as a function of cerebral perfusion pressure (CPP). A, a piglet with an identifiable ULA by piecewise regression at a CPP of 84. In this animal, static rate of autoregulation (SRoR) decreased from 0.6 (intact) below the ULA to 0.1 (impaired) above the ULA, confirming a shift from intact to impaired autoregulation at the putative demarcation. B, a piglet with no identifiable ULA rendered a putative inflection at a CPP of 107.5 by piecewise regression. However, SRoR above the putative ULA was 0.93 and below the putative ULA was 0.92, demonstrating intact autoregulation throughout the induced hypertension.
RESULTS A total of 40 neonatal piglets were randomized to 4 experimental groups. Two animals were excluded: 1 in the sham group did not meet experiment criteria and died before achieving any appreciable increase in CPP and 1 in the CCI group was excluded
NEUROSURGERY
due to LDF failure. Survival to a CPP of 100 mm Hg occurred in 100% of the sham group, 80% of the INF group, 90% of the CCI group, and 91% of the CCI 1 INF group. Survival to CPP of 120 mm Hg occurred in 66% of the sham group, 20% of the INF group, 80% of the CCI group, and 27% of the CCI 1 INF group. The relative inability of the experimental groups with INF to
VOLUME 75 | NUMBER 2 | AUGUST 2014 | 165
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
PESEK ET AL
survive to a higher CPP is accounted for by the difference in ICP between the infused and noninfused groups because a much higher ABP was required in these groups to achieve the same CPP. Physiological variables are compared across the experimental groups in Table. There were no significant group differences with respect to ABP and CPP over the course of the experiment. However, ICP was significantly higher in the INF and CCI 1 INF groups. Temperature was lower at baseline in the CCI group, but returned to normal values during the 6-hour recovery and during arterial hypertension. The ventilation
and fluid management strategy kept pH and PCO2 within normal range across the experiment, and there were no group differences. The CCI group had a higher PO2 (partial pressure of oxygen) than the other subject groups, but PO2 values were consistently greater than 150 mm Hg and less than 300 mm Hg for all animals. Serum sodium was lower at baseline in the CCI group, but was restored to normal in all groups during recovery and during the induced hypertension phases. Serum lactate levels were low and similar across groups and experimental phases.
TABLE. Physiological Data by Experimental Group and Phasea Sham Weight, kg Age, d ABP, mm Hg B R H ICP, mm Hg B R H CPP, mm Hg B R H Temp, °C B R H pH B R H PCO2, torr B R H PO2, torr B R H Na, mg/dL B R H Serum lactate, mg/dL B R H
2.6 (2.1-3.7) 8.0 (5-9.5) 67 (60-71) 60 (52.5-76.5) 112 (104-121.5)
Infusion 2.4 (2.3-3.2) 7.5 (4.8-10.2) 74.5 (62.5-78.2) 72 (63.8-76.5) 128 (114-131)
CCI 2.4 (2.1-2.8) 9.5 (7-12) 74 (65.8-83.5) 73 (67.8-83.2) 96 (80-119)
CCI 1 Infusion
P Valueb
3.8 (2.8-4.3) 9.0 (4-13)
.06 .50
65 (59-81) 65 (62-72) 121 (110-127)
NS, ,.0001
9 (6.3-11) 23 (21.8-25) 25 (24-32.3)
10 (6.8-12) 14.5 (9-21.3) 13 (10.5-19.3)
9 (6-12) 20 (19-25) 28 (23-32)
,.0001, ,.0001
62 (57-64) 53 (50-64.5) 103 (98.5-111.5)
64 (51.8-72) 48.5 (40-51.5) 96.5 (91.5-101.8)
65 (59-69.3) 57 (46.8-69.3) 83.5 (62.3107.8)
55 (53-67) 45 (43-47) 96 (81-105)
NS, ,.0001
38.1 (37.8-39) 38.8 (37.9-39.2) 38.7 (37.9-39.2)
38.9 (38.1-39.3) 38.7 (38.4-38.9) 38.4 (38.1-38.9)
36.5 (39.4-38.4) 38.1 (36.6-38.9) 37.8 (37.4-38.2)
38.4 (37.2-38.1) 38.1 (37.3-38.9) 37.5 (36.9-38.7)
.027, NS
7.40 (7.38-7.46) 7.46 (7.43-7.49) 7.45 (7.44-7.48)
7.42 (7.38-7.44) 7.39 (7.38-7.43) 7.39 (7.32-7.47)
7.44 (7.40-7.45) 7.41 (7.36-7.46) 7.41 (7.34-7.43)
7.40 (7.37-7.41) 7.44 (7.40-7.46) 7.42 (7.40-7.45)
NS, NS
37.8 (35-42.4) 40.2 (39.4-40.6) 38.7 (35.9-42.5)
39.9 (36.2-41) 41.3 (37.5-43.6) 39.2 (36.3-43.9)
37.8 (34.2-40.2) 37.8 (35.5-46.5) 40.4 (35.8-44.5)
40.2 (39.1-41.8) 40.4 (36.3-42.1) 39.6 (37.2-43.2)
NS, NS
206 (187-244) 185 (178-214) 202 (183-213)
209 (169-226) 199 (145-220) 206 (174-249)
247 (235-255) 227 (220-257) 223 (197-249)
206 (188-221) 202 (178-219) 202 (175-223)
.0005, NS
142 (141-144) 144 (140-147) 140 (138-147)
142 (139-145) 142 (139-145) 142 (140-147)
138 (137-140) 140 (135-142) 140 (138-143)
141 (139-143) 141 (140-142) 140 (140-141)
.001, NS
2.6 (1.7-3.2) 1.4 (1.2-3.1) 2.0 (1.5-2.9)
1.7 (1.3-2.5) 1.1 (0.9-2.2) 1.5 (0.9-2.3)
1.6 (1.3-2.5) 1.0 (0.4-3.1) 1.0 (0.6-3.2)
1.7 (1.0-1.9) 1.0 (0.8-1.4) 1.4 (1.1-2.1)
NS, NS
6 (2.5-9) 5 (4-8.5) 7 (2.5-9)
a CCI, controlled cortical impact; ABP, arterial blood pressure; NS, not significant; ICP, intracranial pressure; CPP, cerebral perfusion pressure; Temp, temperature; PCO2, partial pressure of carbon dioxide; PO2, partial pressure of oxygen; Na, serum sodium concentration. Experimental phases: B, baseline phase, R, recovery phase, H, hypertensive phase. Data are presented as median and range. b P values are Kruskal-Wallis analysis of variance for weight and age; all other analyses are 2-way repeated-measures analyses of variance reported across subject group and experimental phase.
166 | VOLUME 75 | NUMBER 2 | AUGUST 2014
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
NEONATAL AUTOREGULATION DURING HYPERTENSION
CBF is plotted as a percentage of baselines from the laser Doppler flowmetry recordings in Figure 2 for visual inspection of the differences between groups during hypertension. Differences between the groups with respect to the presence of a detectable ULA were not statistically significant: 55% of animals in the sham group demonstrated a detectable ULA by the criterion specified. In both the INF and CCI group, 70% of animals were determined to have a ULA. The CCI 1 INF group demonstrated a ULA in 91% of animals (P = .36). However, the leftward shifting of the ULA with addition of progressive injury that is evident in Figure 1 was verified by comparing the CPP at ULA across the groups (Figure 3). The median (IQR) ULA was detected at CPP of 102 mm Hg (97-109 mm Hg) in the sham group, 75 mm Hg (52-84 mm Hg) in the INF group, 81 mm Hg (69-101 mm Hg) in the CCI group, and 61 mm Hg (53-99 mm Hg) in the CCI 1 INF group (P = .01). Although there was a trend toward a lower ULA in the CCI group compared with the sham group, the difference was not statistically significant on post hoc analysis. The higher ULA, however, was statistically different in both INF groups when compared with the sham group.
DISCUSSION The primary objective of this study was to develop a model of injury that would reliably render a ULA. The main finding of this study is that the combination of CCI and induced ICP caused significant decreases in the ULA to near normal CPP values. Interestingly, increases in ICP without trauma caused a greater shift in the ULA than CCI alone. Although the effects of CCI produced similar results on the ULA, they were not of statistical significance. This trend could be explained in part by the primary injury itself and by the fact that the animals with CCI had a higher mean ICP compared with sham animals. However, the groups with ventricular INFs had their mean ICP maintained at greater than 20 mm Hg, which was higher than either the sham or CCI group. Our results are consistent with the findings of Hauerberg et al,5 who showed in mature rats that the ULA is left-shifted by acute increases in ICP in a dose-dependent fashion. Further, when combined with previous studies of the LLA, this dataset can estimate the change in the range of the autoregulatory plateau of neonatal swine caused by acute intracranial hypertension.
FIGURE 2. Cerebral blood flow (CBF) is shown normalized to baseline and plotted as a function of cerebral perfusion pressure (CPP) for each of the 4 groups: sham (n = 9), ventricular infusion (infusion, n = 10), controlled cortical impact (CCI, n = 10), and CCI with ventricular infusion (CCI & infusion n = 11). Data are shown as mean and SD at each 5-mm Hg increment of CPP. The median group upper limit of autoregulation is shown in each graph as a vertical dashed line.
NEUROSURGERY
VOLUME 75 | NUMBER 2 | AUGUST 2014 | 167
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
PESEK ET AL
FIGURE 3. Presence of a detectable upper limit of autoregulation (ULA) and the cerebral perfusion pressure (CPP) at ULA. The ULA was found at a median (interquartile range) CPP of 102 mm Hg (97-109 mm Hg) in the sham group (n = 9), 75 mm Hg (52-84 mm Hg) in the infusion group (Inf, n = 10), 81 mm Hg (69-101 mm Hg) in the closed cortical impact group (CCI, n = 10), and 61 mm Hg (53-99 mm Hg) in the CCI 1 Inf (n = 11) group. Box whiskers graphs show median, interquartile range, and range; P = .01.
Naive neonatal swine have an LLA near 30 mm Hg CPP, and the sham group in this study had a median ULA near 100 mm Hg, giving a plateau spanning 70 mm Hg. Swine of the same age with ICP acutely increased to 20 mm Hg have a modest increase in LLA to near 40 mm Hg, and when ICP is increased to 40 mm Hg, the LLA increases to 50 mm Hg.6 The effect of an acute increase in ICP in this study has a more profound effect on the ULA than was observed in the LLA. An ICP of 20 mm Hg results in a median ULA near 80 mm Hg. Thus, an acute increase in ICP to 20 mm Hg can be estimated to halve the width of the autoregulatory plateau from a 70-mm Hg span to a 35-mm Hg span, with most of the effect caused on the hypertensive end. We are unable to address the dose response because attempts to maintain hypertension with acute increases in ICP to 40 mm Hg in this age animal resulted in the immediate death of the animal.
168 | VOLUME 75 | NUMBER 2 | AUGUST 2014
Interestingly, acute neurological injury without an increase in ICP greater than 20 mm Hg produced a trend of decreased ULA that was of smaller magnitude than the increased ICP group and was not significant when compared with the sham group. The same CCI depth, velocity, piston diameter, dwell time, and recovery duration were used in a previous study of neonatal swine to examine the effect of acute injury on the LLA. The injury used was unilateral, but of a severity that caused an 18% acute injury volume in a similar cohort and an acute increase in ICP from 6 mm Hg to 17 mm Hg. The mean LLA of 35 mm Hg in that cohort was not statistically increased compared with a sham group. We conclude that acute brain injury without an increase in ICP greater than 20 mm Hg in this model causes less impairment of CBF autoregulation than does an atraumatic increase in ICP acutely. Taken together, these studies suggest that the ULA is left-shifted by a larger magnitude than the LLA is right-shifted for the same insult, which is provocative when common management of brain trauma resuscitation is to increase ABP aggressively before ICP is even known. However, these data imply that during acute resuscitation from neurological insult, the ability of the cerebral vasculature to autoregulate is improved by control of ICP and avoidance of prolonged hypertension. Such an interpretation is tempered by the finding in adults that dysautoregulation due to hypertension is associated with worse neurological performance at outcome, but dysautoregulation due to hypotension is associated with death.9 The most recent pediatric TBI guidelines propose a minimum CPP of 40 mm Hg for children and an age-specific CPP range of 40 to 50 mm Hg with infants being at the lower end and adolescents at the upper end of this range.10 Due to the paucity of evidence, only level III recommendations can be made regarding CPP thresholds. Numerous studies have shown that hypotension after brain injury is associated with increased mortality.11-14 During the initial resuscitation, systolic blood pressures should be maintained above fifth percentile for age to avoid ischemic injury.15 However, there are limited data to suggest a maximum threshold for systolic blood pressure. One study done by White et al16 found a 19-fold increase in survival after TBI in pediatric patients with a systolic blood pressure greater than 135 mm Hg. The data of White et al appear to contradict previous TBI studies that showed worse neurological outcomes in the setting of arterial hypertension.17,18 These seemingly conflicting data might be conceptually reconciled by the adult trauma study of Aries et al9 in which a CPP less than optimal for autoregulation was associated with death and a CPP greater than optimal for autoregulation was associated with survival but worsening neurological outcomes. Sex differences after TBI with regard to autoregulatory disturbances have been documented in piglets, with male piglets having increased activity of vasoconstrictive endothelin-1. These male piglets showed impaired pressure autoregulation during hypotension that was aggravated by phenylephrine. By contrast, female piglets had diminished endothelin-1 release after TBI and autoregulation that was improved by phenylephrine.19 We cannot speak to sex differences at the upper limit of autoregulation in this
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
NEONATAL AUTOREGULATION DURING HYPERTENSION
all-male cohort. The biochemical pathways identified in that study are related to vasodilatory responses to hypotension, and it is not known whether or how increased ICP influences mediators of vasoconstriction during hypertension. Brain trauma is expected to evolve over 3 to 5 days after injury, and these data cannot speak to potential temporal changes of the autoregulatory range. This study is limited by the sample size and the inability to detect the significance of a modest difference between the sham and CCI groups with respect to the ULA position. In part, this is because 40% of the sham group and 30% of the CCI group were excluded due to our inability to delineate the ULA before the death of the piglet. Our clinical observation was death due to acute congestive heart failure with frank pulmonary edema and hemorrhagic myocardium seen on autopsy after a sustained doubling of the ABP. By contrast, we have been able to detect the LLA in 100% of subjects studied in previous efforts.6 A reliable model demarcating the ULA would be useful in the study of autoregulation and vascular reactivity monitoring to delineate optimal perfusion pressures.9 However, the lack of consistent ULA delineation in the piglet model presents a challenge to finding a gold standard against which to test the sensitivity and specificity of these monitoring techniques during hypertension as we have done against the LLA.20-23
CONCLUSION The primary finding of this study is that acute increases in ICP impair CBF autoregulation in neonatal piglets with arterial hypertension. This effect was observed in the presence and absence of trauma. Disclosure The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.
REFERENCES 1. Lassen NA. Cerebral blood flow and oxygen consumption in man. Physiol Rev. 1959;39(2):183-238. 2. Lassen NA, Agnoli A. The upper limit of autoregulation of cerebral blood flow— on the pathogenesis of hypertensive encephalopathy. Scand J Clin Lab Invest. 1972; 30(2):113-116. 3. Strandgaard S, Olesen J, Skinhoj E, Lassen NA. Autoregulation of brain circulation in severe arterial hypertension. Br Med J. 1973;1(5852):507-510. 4. Strandgaard S, MacKenzie ET, Sengupta D, Rowan JO, Lassen NA, Harper AM. Upper limit of autoregulation of cerebral blood flow in the baboon. Circ Res. 1974; 34(4):435-440. 5. Hauerberg J, Xiaodong M, Willumsen L, Pedersen DB, Juhler M. The upper limit of cerebral blood flow autoregulation in acute intracranial hypertension. J Neurosurg Anesthesiol. 1998;10(2):106-112. 6. Brady KM, Lee JK, Kibler KK, et al. The lower limit of cerebral blood flow autoregulation is increased with elevated intracranial pressure. Anesth Analg. 2009; 108(4):1278-1283. 7. Mytar J, Kibler KK, Easley RB, et al. Static autoregulation is intact early after severe unilateral brain injury in a neonatal swine model. Neurosurgery. 2012;71(1): 138-145. 8. Rangel-Castilla L, Gasco J, Nauta HJ, Okonkwo DO, Robertson CS. Cerebral pressure autoregulation in traumatic brain injury. Neurosurg Focus. 2008;25(4):E7.
NEUROSURGERY
9. Aries MJ, Czosnyka M, Budohoski KP, et al. Continuous determination of optimal cerebral perfusion pressure in traumatic brain injury. Crit Care Med. 2012; 40(8):2456-2463. 10. Kochanek PM, Carney N, Adelson PD, et al. Guidelines for the acute medical management of severe traumatic brain injury in infants, children, and adolescents—second edition. Pediatr Crit Care Med. 2012;13(suppl 1): S1-S82. 11. Pigula FA, Wald SL, Shackford SR, Vane DW. The effect of hypotension and hypoxia on children with severe head injuries. J Pediatr Surg. 1993;28(3):310-314; discussion 315-316. 12. Luerssen TG, Klauber MR, Marshall LF. Outcome from head injury related to patient’s age. A longitudinal prospective study of adult and pediatric head injury. J Neurosurg. 1988;68(3):409-416. 13. Ong L, Selladurai BM, Dhillon MK, Atan M, Lye MS. The prognostic value of the Glasgow Coma Scale, hypoxia and computerised tomography in outcome prediction of pediatric head injury. Pediatr Neurosurg. 1996;24(6):285-291. 14. Mayer TA, Walker ML. Pediatric head injury: the critical role of the emergency physician. Ann Emerg Med. 1985;14(12):1178-1184. 15. Adelson PD, Bratton SL, Carney NA, et al. Guidelines for the acute medical management of severe traumatic brain injury in infants, children, and adolescents. Chapter 4. Resuscitation of blood pressure and oxygenation and prehospital brainspecific therapies for the severe pediatric traumatic brain injury patient. Pediatr Crit Care Med. 2003;4(suppl 3):S12-S18. 16. White JR, Farukhi Z, Bull C, et al. Predictors of outcome in severely head-injured children. Crit Care Med. 2001;29(3):534-540. 17. Kanter RK, Carroll JB, Post EM. Association of arterial hypertension with poor outcome in children with acute brain injury. Clin Pediatr (Phila). 1985;24(6): 320-323. 18. Brink JD, Imbus C, Woo-Sam J. Physical recovery after severe closed head trauma in children and adolescents. J Pediatr. 1980;97(5):721-727. 19. Armstead WM, Riley J, Vavilala MS. TBI sex dependently upregulates ET-1 to impair autoregulation, which is aggravated by phenylephrine in males but is abrogated in females. J Neurotrauma. 2012;29(7):1483-1490. 20. Brady KM, Easley RB, Kibler K, et al. Positive end-expiratory pressure oscillation facilitates brain vascular reactivity monitoring. J Appl Physiol (1985). 2012;113(9): 1362-1368. 21. Lee JK, Kibler KK, Benni PB, et al. Cerebrovascular reactivity measured by nearinfrared spectroscopy. Stroke. 2009;40(5):1820-1826. 22. Brady KM, Lee JK, Kibler KK, Easley RB, Koehler RC, Shaffner DH. Continuous measurement of autoregulation by spontaneous fluctuations in cerebral perfusion pressure: comparison of 3 methods. Stroke. 2008;39(9):2531-2537. 23. Brady KM, Lee JK, Kibler KK, et al. Continuous time-domain analysis of cerebrovascular autoregulation using near-infrared spectroscopy. Stroke. 2007;38 (10):2818-2825.
COMMENTS
T
he upper limit of autoregulation (ULA) has received relatively less investigation than the lower limit of autoregulation after traumatic brain injury (TBI). This study seeks to better understand the ability of the cerebral circulation to handle hypertension using a neonatal swine model. They hypothesized that the ULA would be shifted to the left by both trauma and intracranial hypertension. They found the greatest shift in ULA to occur with the combined injury and induced intracranial hypertension. These results are interesting and have clinical relevance for how aggressive the initial resuscitation of blood pressure should be in patients with TBI. It also implies that for patients with intracranial hypertension lowering intracranial pressure would be more physiological than increasing pressure to improve cerebral perfusion pressure. Claudia Robertson Houston, Texas
I
n this well-designed swine model study, the authors investigate the upper limit autoregulation (ULA) in response to traumatic brain
VOLUME 75 | NUMBER 2 | AUGUST 2014 | 169
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
PESEK ET AL
injury (TBI) and intracranial hypertension. Results were as expected; TBI and secondary intracranial hypertension resulted in a significant shift of ULA. In clinical practice, these results emphasize the adequate management of blood pressure in patients with TBI. During acute resuscitation, cerebral autoregulation is improved by avoidance of prolonged hypertension and control of intracranial pressure. Adequate
170 | VOLUME 75 | NUMBER 2 | AUGUST 2014
cerebral perfusion pressure should be obtained by lowering intracranial pressure and not by inducing systemic blood pressure. More is not always better. Leonardo Rangel-Castilla Phoenix, Arizona
www.neurosurgery-online.com
Copyright © Congress of Neurological Surgeons. Unauthorized reproduction of this article is prohibited.
