Cerebrospinal fluid biochemistry reflects effects

of therapeutic hypothermia after cardiac arrest in a porcine model^?

Rong Hua MD, Chunsheng Li MD?, Ping Gong MD, Ziren Tang MD, Xue Mei MD, Hong Zhao MD

Beijing Chaoyang Hospital affiliated with Capital Medical University, Beijing 100020, China

Received 4 October 2011; revised 21 October 2011; accepted 22 October 2011

Abstract

Background: Mild induced hypothermia (MIH) is recommended to treat neurologic injury after cardiac arrest . However, clinical trials to assess MIH benefit after CA have been largely inconclusive. We investigated the subsequent changes in cerebrospinal fluid (CSF) biochemistry after MIH (33?C-34?C for 12 hours) and evaluated the importance of ongoing Fever control.

Methods: Thirty-two male Wuzhishan inbred mini pigs (n = 16/group) underwent ventricular fibrillation followed by cardiopulmonary resuscitation and were randomized into 2 groups: hypothermic and control. Upon resumption of spontaneous circulation (ROSC) from CA, the hypothermic group was treated with MIH by endovascular cooling. The control group received no temperature intervention. Core temperatures were continually monitored. At various points throughout the procedure, CSF samples were obtained to measure glutamate, lactate, and pyruvate levels.

Results: The core temperature of the hypothermic group was found to have increased postrewarming and reached levels comparable with those of the control group at ROSC 72 hours. In both groups, glutamate increased significantly after ROSC, but the glutamate levels in the hypothermic group were lower than those in the control group, except at ROSC 1 hour. The lactate-pyruvate ratio increased in the control group at ROSC 1 hour and was significantly lower in the hypothermic group (P b .05).

Conclusions: Mild induced hypothermia mitigated and delayed the CA-induced increase of CSF glutamate. Therefore, our results suggest that clinically inducing hypothermia as soon as possible after CA, or prolonging the time of MIH in combination with controlling ongoing fever, may enhance hypothermic protective effects.

(C) 2012

Introduction

^? Conflict of interest statement: None of the authors had any personal financial interest that could bias this work.

* Corresponding author. Department of Emergency Medicine, Beijing

Chaoyang Hospital affiliated with Capital Medical University, Beijing 100020, China. Tel.: +86 10 85231051; fax: +86 10 85231051.

E-mail addresses: [email protected], [email protected] (C. Li).

Cardiac arrest is usually accompanied by coma for more than 1 hour [1], and the mortality and risk of permanent severe brain damage are high. Mild induced hypothermia (MIH) has been suggested to improve neurologic outcomes of CA and the overall survival rate, and the American Heart Association (AHA) has recommended that MIH (33?C-34?C

0735-6757/$ - see front matter (C) 2012 doi:10.1016/j.ajem.2011.10.022

for 12-24 hours) be applied to all unconscious CA patients [2]. However, clinical trials to assess MIH benefit after CA have been characterized by nonnegligible risks of bias and largely inconclusive [3]. Further studies are necessary to determine the protective mechanisms of MIH against brain injury after CA to more effectively apply MIH therapy.

Studies using a rat CA model revealed some beneficial effects of MIH on excessive glutamate release and brain injury after CA [4]. However, the rodent brain differs from the larger mammalian brain in some important aspects, and these results are not readily generalizable to humans. For example, the rodent brain is lissencephalic and has distinctive rheological and metabolic properties [5]. The pig CA model is accepted as a very close representation of the mammalian physiologic experience, and this model has been extensively used in laboratory-based experimental studies. A cerebral microdialysis detection method has been used to measure glutamate in the pig CA model, but the method is localized and not able to survey the entire Cerebral injury area [6]. Furthermore, the hypothermic state can impact the recovery ability of microdialysis and may have influenced the glutamate values obtained with this method [6]. In addition, to our knowledge, the duration of MIH induced by other cooling methods in most of the animal studies was very short and did not abide by the Guideline recommendations of 12 to 24 hours.

Mild induced hypothermia efficacy is optimized using the intravascular cooling system, which allows for a target core temperature to be reached rapidly, steadily, and safely [7]. This approach also maintains the target hypothermic state more effectively than other available methods, such as water- circulating blankets and gel-coated water circulating pads [8]. Hypothermia is well-recognized for its ability to reduce brain metabolism and is believed to elicit other protective Physiologic effects as well. However, the mechanisms underlying the Neuroprotective effects are incompletely known, especially those resulting from MIH performed according to the AHA guidelines. Fever is another common occurrence in the first 72 hours after CA and is associated with unFavorable neurologic outcomes [9]. Fever control after MIH may further enhance MIH benefits.

To determine MIH influence on brain metabolism and brain injury after CA, we established a porcine CA model to evaluate the physiologic effects of AHA-recommended MIH treatment. Mild induced hypothermia-treated or nontreated CA pigs were assessed for oxygen levels in venous jugular bulb blood (SjVO2) and glutamate levels and Energy metabolism markers in cerebrospinal fluid (CSF).

Methods

Animal preparation

This study was designed in accordance with the administration regulation of Beijing on Animal experimenta-

tion. Thirty-two male Wuzhishan (WZSP) inbred mini pigs, aged 10 +- 1 months and weighing 28 +- 3 kg, were used in this study (Institute of Animal Sciences, Chinese Academy of Agricultural Sciences, Beijing, China). Before experimenta- tion, pigs were fasted overnight with access to water. The pigs were randomly divided into 2 groups (n = 16 each): hypothermic and control. Anesthesia was administered by Intramuscular injection of ketamine (20 mg/kg), immediately followed by ear vein injection of propofol (1.0 mg/kg). Sedation was maintained by intravenous (IV) infusion of pentobarbital (8 mg/kg per hour) [10]. A standard lead II electrocardiogram was used to continuously monitor cardiac rhythm. Fluid losses were compensated by IV infusion of 30 mL /kg Ringer acetate solution for 1 hour before the experiment, followed by a continuous 2.5% glucose infusion (10 mL/kg per hour) [6].

Under sedation, pigs were placed in the supine position and intubated. Ventilation was performed with a volume-con- trolled ventilator (Servo 900c; Siemens, Germany) set at 15 mL/kg tidal volume, 40 L/min peak flow, and 0.21 fraction of inspired oxygen. End-tidal PCO2 was monitored with an infrared capnometer (Model NPB-75, Nellcor Puritan Bennett Inc, Boulder, CO). minute ventilation was adjusted to maintain 35 to 45 mm Hg arterial Pco2 and 7.40 to 7.50 pH.

During surgery, the right external and internal jugular veins (IJV) were exposed, followed by the left femoral artery and femoral vein. A Swan-Ganz catheter (7F; 7 Fr; Edwards Life Sciences, Irvine, CA) was advanced from the left femoral vein, and flow was directed into the pulmonary artery. Measure- ments were recorded for pulmonary artery wedge pressure (PAWP), Continuous cardiac output (CO), and core tempera- ture. An angiographic catheter was then inserted from the left femoral artery into the aortic arch to collect blood samples and measure mean arterial blood pressure (MAP) with a pressure transducer (Biosensors International Corp, Singapore). A catheter was inserted retrograde into the right IJV for vein blood collection. Another sheath was inserted into the right external jugular vein for electrode catheter placement to induce ventricular fibrillation (VF) and for subsequent Icy catheter placement (Model IC-3585AE; Alsius Corp, Irvine, CA) to induce hypothermia after VF. The Coolgard 3000 Temperature control system (Alsius Corp, Irvine, CA) was used for the MIH treatment. A bladder catheter was inserted for urine volume monitoring, whereas an epidural catheter was inserted into the lumbar spine at the 3 to 4 interval for CSF sampling and intracranial pressure monitoring.

Measurements

Cerebrospinal fluid levels of lactate, pyruvate, glucose, glycerol, and glutamate were analyzed by standard enzymatic methods on a CMA 600-Microdialysis Analyser (Carnegie Medicine AB, Sweden). The lactate/pyruvate ratio (L/P ratio) and lactate/glucose ratio (L/G ratio) were calculated.

Hemodynamic parameters were measured continually. Blood gases in arterial and IJV blood were analyzed

throughout the procedure. Lactate clearance (LC) (in percent) was calculated by the formula, ([initial lactate - delayed lactate]/initial lactate) x 100, in which initial lactate was the measurement at resumption of spontaneous circulation (ROSC) 20 minutes and delayed lactate was that at 1 hour or more after ROSC. Intracranial pressure was measured continually, and cerebral perfusion pressure (CPP) was calculated as MAP -ICP. The neurologic deficit score (NDS) [11] was determined at ROSC 24, 48, and 72 hours.

Experimental protocol

After surgery, pigs were rested for 1 hour before baseline data were collected. Ventricular fibrillation was induced with a programmable electrical stimulation instrument (GY-600A; Kai Feng Huanan Instrument Company, Kaifeng, Henan, China) [12]. When VF was achieved, mechanical ventilation was discontinued. After 8 minutes untreated VF, chest compressions and ventilations (ratio, 30:2) were initiated. At 10 minutes VF, defibrillation shock was administered at 150 J (Smart Biphasic) (M3535A, Philips Medical Systems, Holland) for the first resuscitation attempt. If the first defibrillation failed, 200 J subsequent attempts were carried out. Epinephrine (0.02 mg/kg) was administered after 2 minutes of cardiopulmonary resuscitation. Death was called if ROSC was not achieved after 4 defibrillations; 3 pigs died at this point. Resumption of spontaneous circulation was defined as systolic blood pressure maintenance of 60 mm Hg or more for 10 or more consecutive minutes.

After ROSC, MIH was immediately induced in the hypothermic group by using the endovascular cooling technique. According to guidelines [2], the target core temperature was set at 33.0?C and maintained for 12 hours. Active rewarming was carried out at 0.5?C per hour until the 37.0?C core temperature was recovered. Pancuronium bromide (0.1 mg/kg per hour) was administered in both groups after ROSC for 3 hours. In the control group, no temperature regulation intervention was carried out. Blood glucose and electrolytes were tested, and IV fluids were administered to maintain a 3 to 8 mmol/L blood glucose level and 4.0 mmol/L serum potassium level. Cefuroxime antibiotic prophylaxis was administered first at anesthesia induction, then at ROSC 12 hours, and finally before extubation. In addition, blood lactate was measured at ROSC 20 minutes (initial lactate level) and used to calculate LC in arterial and IJV blood.

Core temperatures were recorded at 1 hour intervals for 24 hours after ROSC. Hemodynamic data, ICP, blood gases, and CSF samples were obtained at 1, 6, 12, and 24 hours after ROSC. Once recovered from anesthesia, pigs were placed in observation cages until ROSC 72 hours, at which time pigs were reanesthetized to facilitate data recording of temperature, hemodynamic parameters, blood gases, and ICP and CSF sampling. Afterwards, the pigs were euthanized with IV potassium chloride, and then the inferior parietal lobe was dissected for pathology. Histopathologic injury

of brain tissue was identified according to established descriptions [4,13].

Statistical analysis

Data were expressed as mean +- SD for continuous variables. One-way analysis of variance was used to compare differences in LC, CSF biochemistry parameters, SjVO2, hemodynamic parameters, arterial lactate, IJV lactate, and core temperature values at selected time points between the 2 groups and between the amounts of anesthetics given to each group. Repeated measure analysis of variance was used to analyze differences between the repeated recording times for CSF biochemistry parameters, SjVO2, hemodynamic parameters, arterial lactate, IJV lactate, and core temperature values. Nonparametric data (such as NDS) were expressed as median (interquartile range) and analyzed by Mann-Whitney U test. Discrete variables (ROSC number and survival after ROSC) were compared by Fisher exact test. P b .05 was considered statistically significant. All statistical analysis was carried out with SPSS v17.0 software (SPSS Inc, Chicago, IL).

Results

Outcomes

Of the 32 pigs used, 3 did not achieve ROSC after VF and were excluded from further analysis. Fourteen (88%)

Table 1 Comparison of outcomes between the hypothermic group and the control group

Outcomes	CG (n = 16)	HG (n = 16)	P
ROSC	15	14	.50
Shocks before ROSC,	2.00 +- 0.66	2.29 +- 0.47	.19
mean +- SD
Duration of VF	752 +- 93.1	746 +- 87.1	.85
before ROSC (s),
mean +- SD
1 h survival	15	14	NA b
6 h survival	12	11	.64
12 h survival	11	10	.62
24 h survival	8	10	.27
48 h survival	6	10	.09
72 h survival	5	10	.05
NDS a at baseline	0	0	NA
NDS in 24 h	215 (163-235)	125 (99-145)	.01
NDS in 48 h	135 (95-180)	83 (59-105)	.03
NDS in 72 h	95 (75-118)	40 (20-63)	.01
CG indicates control group; HG, hypothermic group; ROSC, restoration of spontaneous circulation; VF, ventricular fibrillation; NDS, neuro- logical deficit score; NA, not applicable. a NDS are expressed as median (interquartile range). b According to analysis by Fisher exact test, no animal died at ROSC 1 hour.

pigs in the hypothermic group and 15 (94%) in the control group achieved ROSC. There were no significant differences in survival rates between the 2 groups at ROSC 1, 6, 12, or 24 hours (Table 1). Ten (71%) pigs in the hypothermic group and 5 (33%) in the control group survived out to 72 hours, and this difference was statistically significant (P b .05). The animals that survived for the entire 72 hours were further analyzed. Neurologic deficit score in the hypothermic group was significantly lower than in the control group at 24, 48, and 72 hours (P b .01, P b .05, and P b .01, respectively). There were no significant differences in shocks and VF duration before ROSC between the 2 groups (Table 1). There were no significant differences observed between the baseline physiologic variables and the amounts of anesthetics.

The representative histopathology alterations of pallium are shown in Fig. 1. The brain injury observed in the control group (having received no MIH) was more serious than in the hypothermic group.

Core temperature

In the control group, core temperature gradually increased after ROSC and was significantly higher than baseline from ROSC 2 hours to ROSC 72 hours. In the hypothermic group,

Fig. 2 Core temperature of pigs during the experimental protocol. B indicates baseline. ROSC, restoration of spontaneous circulation. Temperature recorded vs baseline: ?P b .05; ??P b .01. Control vs hypothermic: ^#P b .05; ^##P b .01.

core temperature gradually decreased and reached target temperature after ROSC 3 to 4 hours and then remained for 12 hours. Upon rewarming, core temperature reached the target temperature (37.0?C) at ROSC 22 to 23 hours. After rewarming, core temperature increased to baseline at ROSC 24 hours (37.7?C +- 0.22?C, P = .07) and rose higher than baseline at ROSC 72 hours (38.4?C +- 0.34?C, P b .01) (Fig. 2).

Fig. 1 A representative comparison of the inferior parietal lobe from hypothermic (A and C) and control (B and D) pigs. Hematoxylin and eosin staining showed that tissue from MIH-treated pigs (A and C) exhibited few neuron cells with loss of Nissl substance, development of cytoplasmic eosinophilia, and shrinkage of the perikaryon (thin arrow). The tissue from control pigs (B and D) exhibited most of neuron cells having a diffuse loss of Nissl substance, development of cytoplasmic eosinophilia, and shrinkage of the perikaryon. In addition, neuronal nucleus disappeared, leaving a homogeneous, eosinophilic silhouette of the cell. Multiple neurons showed these features, and 3 representative neurons are denoted here (thick arrow). Original magnification: A and B x200, C and D x400.

Cerebrospinal fluid parameters

No differences were observed between the control and hypothermic groups at baseline. In both groups, glutamate gradually increased after ROSC and continued in this manner until experiment end. Levels detected in the hypothermic group, however, were significantly lower than those in the control group for every ROSC time point examined, except 1 hours. Glutamate peak values were detected in the control group at ROSC 24 hours (123.43 +- 17.27 umol/L) and in the hypothermic group at ROSC 72 hours (33.70 +- 12.68 umol/L) (Fig. 3A).

Glycerol levels in control group were increased at ROSC 1, 6, and 12 hours and then slowly declined to baseline at ROSC 24 hours. Although glycerol values in the hypothermic group were also increased at ROSC 1 and 6 hours, the values declined to baseline at ROSC 12 hours and were significantly lower than those in the control group at ROSC 6 and 12 hours (Fig. 3B).

Lactate significantly increased in both groups after ROSC, which lasted until the end of experiment. Lactate levels detected in the hypothermic group were significantly lower than in the control group at ROSC 1 hour. Lactate peak values in the control group were detected at ROSC 1 hour

Fig. 3 Cerebrospinal fluid parameters during the experimental protocol. A, Glutamate; B, glycerol; C, lactate; D, L/P ratio; E, pyruvate; F, glucose; G, L/G ratio. B indicates baseline; ROSC, restoration of spontaneous circulation. Parameter vs baseline: ?P b .05; ??P b .01.

Hypothermic vs control group: ^#P b .05; ^##P b .01.

and in the hypothermic group at ROSC 6 hours (Fig. 3C). The L/P ratio in the control group was significantly increased at ROSC 1 hour but reached baseline values in the hypothermic group and was significantly lower than in the control group at ROSC 1 hour (Fig. 3D).

Compared with baseline values, pyruvate levels were significantly increased in the control group at ROSC 6, 12, and 24 hours, and in the hypothermic group, at ROSC 12 and 24 hours. However, at ROSC 12 and 24 hours, pyruvate was significantly lower in the hypothermic group than in the control group (Fig. 3E).

Glucose in both groups was sustained at baseline levels throughout the experiment. However, glucose levels in the hypothermic group were significantly higher than in the control group at ROSC 1, 6, and 12 hours (Fig. 3F). The L/G ratio was sustained at high levels out to ROSC 24 hours in the control group and returned to baseline at ROSC 72 hours. In the hypothermic group, the L/G ratio was only elevated at ROSC 1 and 6 hours. Furthermore, the L/G ratio in the hypothermic group was significantly lower than that in the control group at ROSC 1, 6, 12, and 24 hours (Fig. 3G).

Hemodynamic findings

Heart rate (HR) in the control group stabilized after ROSC. In contrast, HR declined in the hypothermic group at ROSC 6 hours and returned to baseline at ROSC 12 hours, but which were lower compared with the control group (Table 2). Mean arterial blood pressure in the control group was decreased at ROSC 6 hours and then returned to baseline. Meanwhile, MAP in the hypothermic group was stabilized after ROSC and higher than in the control group at ROSC 6 hours. No significant differences were detected in PAWP between the control and hypothermic groups at any time point examined (Table 2).

Table 2 Hemodynamic parameters during the experimental protocol^? Baseline ROSC

1 h 6 h

12 h

24 h

72 h

In the control group, CO initially declined at ROSC 1 hour and then returned to baseline by 6 hours and remained at normal levels until experiment end. In the hypothermic group, CO decreased gradually to nadir at ROSC 6 hours (P b .01) then increased to baseline at ROSC 24 hours. Furthermore, when compared with control group CO levels, the hypothermic group had significantly lower CO levels at ROSC 6 hours (Table 2).

Lactate clearance and SjVO2

The initial lactate levels were similar between the control group and hypothermic group, both for arterial blood (12.48 +- 1.44 mmol/L and 12.21 +- 1.09 mmol/L, respectively) and IJV blood (13.60 +- 1.48 mmol/L and 13.38 +- 1.49 mmol/L, respectively). arterial lactate levels for both groups were significantly increased at ROSC 1 and 6 hours but decreased to baseline at ROSC 12 hours; in addition, the levels in the hypothermic group were lower than in the control group at ROSC 1 and 6 hours. Internal jugular veins lactate in both groups was significantly increased at ROSC 1 hour; however, the IJV lactate (IJVLac) at ROSC 6 hours in the control group also increased, whereas IJVLac in the hypothermic group remained at baseline (Table 3). Compared with the control group, LC in the hypothermic group was significantly higher at ROSC 1 and 6 hours (P b .05) in arterial blood and significantly higher at ROSC 1, 6, and 12 hours in IJV blood (Table 3).

SjVO2 in the control group stabilized after ROSC. SjVO2 in the hypothermic group increased after ROSC 6, 12, and 24 hours and then decreased to baseline at ROSC 72 hours (Table 3). Moreover, SjVO2 levels in the hypothermic group were significantly higher than in the control group at ROSC 6, 12, and 24 hours (Table 3).

HR (beats per minute)
CG (n = 5) 124 +- 6	125 +- 6	112 +- 8	118 +- 8	118 +- 9	119 +- 6
HG (n = 10) 113 +- 11	120 +- 11	104 +- 6?+	106 +- 7+	110 +- 12	112 +- 8
MAP (mm Hg)
CG (n = 5) 109 +- 4.0	103 +- 8.7	101 +- 2.3??	106 +- 7.0	110 +- 4.2	107 +- 6.0
HG (n = 10) 109 +- 5.4	104 +- 7.0	106 +- 6.0+	108 +- 5.4	107 +- 7.0	106 +- 3.5
CO (L/min)
CG (n = 5) 3.06 +- 0.44	2.49 +- 0.38?	2.66 +- 0.39	2.98 +- 0.32	3.45 +- 0.38	3.59 +- 0.41
HG (n = 10) 3.61 +- 0.58	2.52 +- 0.42??	1.85 +- 0.43??++	2.48 +- 0.78??	3.62 +- 1.09	3.85 +- 0.56
PAWP (mm Hg)
CG (n = 5) 11 +- 3.8	11 +- 2.5	10 +- 2.3	11 +- 2.7	11 +- 2.6	11 +- 1.6
HG (n = 10) 11 +- 2.6	11 +- 2.5	12 +- 2.3	12 +- 3.2	11 +- 2.4	11 +- 1.5
?All data are expressed as mean +- SD. ROSC, restoration of spontaneous circulation; CG, control group; HG, hypothermic group; HR, heart rate; MAP, mean aortic pressure; CO, cardiac output; PAWP, pulmonary arterial wedge pressure. Hemodynamic value vs baseline. *P b .05. **P b .01. CG vs HG. +P b .05. ++P b .01.

aLac (mmol/L)
CG (n = 5)	1.62 +- 0.69	8.38 +- 1.36?	4.42	+- 0.88??	3.20 +- 1.27	1.82 +- 0.51	1.64 +- 0.47
HG (n = 10)	1.58 +- 0.53	5.91 +- 1.63??+	2.92	+- 0.65?++	2.49 +- 0.90	1.50 +- 0.45	1.48 +- 0.35
IJVLac (mmol/L)
CG (n = 5)	1.88 +- 0.67	10.48 +- 1.49??	7.30	+- 1.11?	4.18 +- 1.32	2.06 +- 0.61	1.38 +- 0.47
HG (n = 10)	1.78 +- 0.62	6.26 +- 1.19??++	2.87	+- 1.03++	2.04 +- 1.14++	1.75 +- 0.92	1.43 +- 0.40
aLC (%)
CG (n = 5)	NA	32.50 +- 11.40	63.82	+- 9.91	73.58 +- 11.18	85.17 +- 4.93	87.28 +- 4.37
HG (n = 10)	NA	50.91 +- 15.61+	75.95	+- 5.42++	79.62 +- 6.90	87.56 +- 4.05	87.76 +- 3.29
IJVLC (%)
CG (n = 5)	NA	23.04 +- 5.46	46.05	+- 8.22	69.71 +- 6.46	84.87 +- 3.82	89.77 +- 3.68
HG (n = 10)	NA	52.95 +- 9.46++	78.62	+- 7.08++	84.96 +- 7.61++	87.18 +- 5.71	89.34 +- 2.54
SjVO2 (%)
CG (n = 5)	65 +- 6	61 +- 8	60	+- 9	62 +- 6	63 +- 7	61 +- 6
HG (n = 10)	64 +- 6	69 +- 10	82	+- 3??++	81 +- 3??++	72 +- 5??+	64 +- 5
?All data are expressed as mean +- SD. ROSC, restoration of spontaneous circulation; CG indicates control group; HG, hypothermic group; aLac, arterial lactate; IJVLac, internal jugular vein lactate; aLC, arterial lactate clearance; IJVLC, internal jugular vein lactate clearance; SjVO2, venous jugular bulb oxygen saturation; NA, not applicable because the baseline value was taken before CA. Parameter vs baseline. *P b .05. ** P b .01. CG vs HG. +P b .05. ++P b .01.

Intracranial pressure and CPP

Table 3 Lactate clearance and SjV ^?

Baseline

O2

ROSC

1 h

6 h

12 h

24 h

72 h

The ICP in both groups increased at ROSC 1, 6, and 12 hours. This trend was followed by a decline to baseline at ROSC 24 hours. The ICP in the hypothermic group was lower than that in the control group at ROSC 1, 6, and 12 hours (Table 4). Cerebral perfusion pressure in both groups declined significantly at ROSC 1, 6, and 12 hours, after which the levels returned to baseline. However, CPP in the hypothermic group at ROSC 6 hours was significantly higher than in the control group (P b .05) (Table 4).

Discussions

By investigating the effects of induced therapeutic hypothermia via the intravascular cooling system after CA in a porcine model, we found that 12 hours of MIH at ROSC onset was able to mitigate and delay brain metabolism disorders normally associated with CA outcome. Moreover,

MIH led to improved survival, promoted Neurologic recovery, and lessened brain tissue injury extent after CA.

Glutamate levels after ROSC can spike significantly [4]. In our study, we observed that glutamate levels in pigs treated with MIH were remarkably suppressed during the hypothermia maintenance phases. It appeared that MIH could depress the severity and delay the time to increase of glutamate that normally results from CA. A major mecha- nism of neuronal protection elicited by MIH is the inhibition of excitatory amino acid release, including glutamate. brain temperature changes are known to have a major impact on the glutamate reuptake system [14]. Mild induced hypother- mia is believed to lessen the potential for exposure to high glutamate levels during postischemia by protecting the glutamate reuptake system [15], thereby decreasing the excitotoxic effect on neuronal cells [4]. In a related study by others using the pig CA model, the excessive glutamate release normally elicited by CA was not appreciably reduced by MIH [6]. However, the cerebral microdialysis detection method used to measure glutamate was localized and not

Table 4 Intracranial pressure and CPP during the experimental protocol?
		Baseline	ROSC
			1 h	6 h	12 h	24 h	72 h
	ICP (mm Hg) CG (n = 5)	10 +- 1.5	15 +- 0.8??	18 +- 1.1??	19 +- 2.7??	13 +- 1.5	11 +- 1.5
	HG (n = 10)	10 +- 2.2	11 +- 2.1?++	14 +- 2.0??++	14 +- 2.3??++	11 +- 1.8	10 +- 0.9
	CPP (mm Hg) CG (n = 5)	99 +- 5.4	88 +- 8.5?	84 +- 2.1??	87 +- 8.0?	97 +- 5.0	95 +- 6.6
	HG (n = 10)	99 +- 5.3	93 +- 6.1?	92 +- 8.1?+	94 +- 7.4?	95 +- 7.0	95 +- 3.9
?All data are expressed as mean +- SD. ROSC, restoration of spontaneous circulation; CG, control group; HG, hypothermic group. ICP: intracranial pressure; CPP: cerebral perfusion pressure. Parameter vs baseline: *P b .05. **P b .01. CG vs HG. +P b .05. ++P b .01.

able to survey the entire cerebral injury area [6]. In addition, the relatively short time of cooling used in that study (180 minutes) may have affected the glutamate elevation response. In our study, we used the porcine CA model but conducted the MIH treatment according to the published AHA guidelines and sustained the treatment for 12 hours to mimic the clinical setting. Glutamate values were investi- gated in the CSF that more accurately reflect the global cerebral injury after resuscitation. Nevertheless, in our model, MIH appeared to have no significant effect on glutamate increase that was experienced during the hypo- thermia induction phases, and the postrewarming glutamate levels were still abnormal. This finding may reflect insufficiently Low Temperatures during the hypothermia induction phases or that fever was experienced postrewarm- ing. In our pigs, the pyrexic state persisted for 72 hours after ROSC, which may have been a response to CA-induced hypothalamus damage [16]. However, delayed hyperthermia aggravates global ischemic pathology through multiple mechanisms, including increased glutamate release [17]. Considered together with our results, it is possible that MIH protective effects may be enhanced if the procedure is initiated as soon as possible after CA. Likewise, prolonging MIH time or combining the treatment with ongoing fever control may extend MIH protective effects.

Another important finding from our study was that CA pigs’ high CSF lactate and pyruvate levels were Sustained AFter ROSC. The L/P ratio in the control group increased almost immediately upon ROSC, at 1 hour. Moreover, the L/G ratio was sustained at high levels for 24 hours. These results may reflect persistently inhibited pyruvate dehydro- genase activity and no change in lactate dehydrogenase after ROSC [18]. Furthermore, MIH treatment in our pigs was associated with reduced and delayed lactate increase, maintenance of normal or near to normal pyruvate levels, elevated glucose levels, and synchronously decreased L/P ratio and L/G ratio. Extracellular glucose, lactate, pyruvate, L/P ratio, and L/G ratio are important biomarkers of brain tissue hypoxia because they each reflect the intracellular redox state and are related to mitochondrion function [19]. Mitochondrial dysfunction can result in metabolic failure [20]. In our pigs, high glycerol levels were sustained for 12 hours after ROSC. Extracellular glycerol is a biomarker of cellular injury, released upon breakdown of structural membrane components [21]. Disordered energy metabo- lism has also been implicated in increased glycerol levels present in injured tissue [21]. Interestingly, we observed that MIH significantly inhibited the glycerol increase normally associated with CA and ROSC during the hypothermia maintenance phases. Our results suggest that MIH was able to protect cerebral energy metabolism and mitigate cellular injury; whether these effects resulted from protected mitochondrion function or pyruvate dehydroge- nase activity requires further study. During the hypother- mia induction phases in our study, MIH reduced energy metabolism disorder but did not affect glycerol increase,

indicating that MIH could resolve metabolism disorder and subsequently mitigate brain cell injury. Therefore, these results support clinical induction of hypothermia as soon as possible after CA.

Other prominent findings from our study were that SjVO2 only increased significantly during the hypothermia mainte- nance phase and early postrewarming phase, which may reflect MIH-induced reduction of the cerebral metabolic requirements [6]. In our study, MIH not only inhibited both arterial lactate and IJVLac to abruptly increase after ROSC but also improved LC significantly in both IJV and arterial blood. Lactate clearance represents a technically simple method to assess tissue oxygen delivery [22] and indirectly indicates tissue perfusion. Effective LC indicates improved tissue perfusion and is associated with decreased mortality in post-CA patients [23]. Thus, we speculate that improved cerebral perfusion and effectively eliminating acidic metab- olites may be 2 of the key MIH protective effects.

Finally, in the present study, we discovered that ICP was elevated and CPP was decreased within 12 hours after ROSC. Mild induced hypothermia depressed ICP elevation after ROSC and maintained MAP at baseline levels after ROSC. Eventually, CPP decrease was ameliorated at ROSC

6 hours. A previous study by others showed that ICP elevation is caused by postischemic hyperperfusion [24]. Mild induced hypothermia has also been shown to suppress both postischemic hyperperfusion and hypoperfusions [25], which may contribute to depression of ICP elevation. Moreover, autoregulation of cerebrovascular pressure is lost after CA, making the cerebral perfusion CPP dependent. Mild induced hypothermia improvement of cerebral perfu- sion, although early and transient, may contribute to improved survival rates and decreased brain injury and NDS. Collectively, our study has suggested that MIH, performed according to the AHA guidelines, has many beneficial effects. Specifically, it can mitigate and delay the increase of glutamate, ameliorate abnormal cerebral energy metabolism, improve cerebral perfusion and reduce the cerebral metabolic requirements, and decrease ICP elevation. Although some protective effects were early or transient, MIH eventually improved survival rates and Neurologic dysfunction. Thus, it appears that inducing hypothermia as soon as possible after CA may be an effective and feasible approach to improving CA outcomes. In addition, prolonging MIH time or combining MIH with ongoing fever control may enhance

the hypothermia protective effects.

Study limitations

(1) In the clinical setting of human CA patients, various therapeutic interventions are usually performed after patient rewarming. However, to obtain comparable results between the control and hypothermic group, no intervention was performed after rewarming of the study animals. (2) Cardiac arrest results in dispersed cerebral tissue injury. In our study, we focused on the brain cortex because it is particularly

sensitive to ischemia and only sampled this tissue for histologic observation. (3) Anesthesia was ketamine/propofol for induction and pentobarbital for maintenance. Although these anesthetics may alter cerebral metabolism and also provide some neuroprotection, there were no significant differences in the amounts of anesthetics used between the 2 groups. Thus, anesthetics did not influence the results of the study.

Conclusions

In conclusion, MIH induced at ROSC onset and sustained for 12 hours can mitigate and delay disorders in brain metabolism, such as CSF glutamate increase caused by CA in a porcine model. Mild induced hypothermia hypothermia protective effects may be enhanced by inducing hypothermia as soon as possible, prolonging the MIH time, or combining MIH with ongoing fever control.

Acknowledgments

This project was supported by grants from the National Natural Science Foundation of China (no. 30972863). The authors are deeply grateful for their financial support. We thank Profs Huiling Huang and Dongli Muo for their expert assistance with detection and analysis of CSF paramaters using the CMA 600 Microdialysis Analyser.

References

1. Holzer M. Targeted temperature management for comatose survivors of cardiac arrest. N Engl J Med 2010;363:1256-64.
2. Peberdy MA, Callaway CW, Neumar RW, et al. Part 9: post-cardiac arrest care: 2010 American Heart Association guidelines for cardiopulmonary resuscitation and emergency cardiovascular care. Circulation 2010;122:S768-86.
3. Nielsen N, Friberg H, Gluud C, et al. Hypothermia after cardiac arrest should be further evaluated-a systematic review of randomised trials with meta-analysis and trial sequential analysis. Int J Cardiol 2011; 151:333-41.
4. Takata K, Takeda Y, Sato T, et al. Effects of hypothermia for a short period on histologic outcome and extracellular glutamate concentration during and after cardiac arrest in rats. Crit Care Med 2005;33:1340-5.
5. van Zanten AR, Polderman KH. Early induction of hypothermia: will sooner be better? Crit Care Med 2005;33:1449-52.
6. Nordmark J, Enblad P, Rubertsson S. Cerebral energy failure following Experimental cardiac arrest hypothermia treatment reduces secondary lactate/pyruvate-ratio increase. Resuscitation 2009;80:573-9.
7. Arrich J. Clinical application of mild therapeutic hypothermia after cardiac arrest. Crit Care Med 2007;35:1041-7.
8. Hoedemaekers CW, Ezzahti M, Gerritsen A, et al. Comparison of cooling methods to induce and maintain normo- and hypothermia in

intensive care unit patients: a prospective intervention study. Crit Care 2007;11:R91.

1. Takasu A, Saitoh D, Kaneko N, et al. Hyperthermia: is it an ominous sign after cardiac arrest? Resuscitation 2001;49:273-7.
2. Grund F, Tjomsland O, Sjaastad I, et al. Pentobarbital versus medetomidine-ketamine-fentanyl anaesthesia: effects on haemody- namics and the incidence of ischaemia-induced ventricular fibrillation in swine. Lab Anim 2004;38:70-8.
3. Wenzel V, Lindner KH, Krismer AC, et al. Survival with full neurologic recovery and no cerebral pathology after prolonged cardiopulmonary resuscitation with vasopressin in pigs. J Am Coll Cardiol 2000;35:527-33.
4. Hamer AW, Karagueuzian HS, Sugi K, et al. Factors related to the induction of ventricular fibrillation in the normal canine heart by programmed electrical stimulation. J Am Coll Cardiol 1984;3:751-9.
5. Alam HB, Chen Z, Ahuja N, et al. Profound hypothermia protects neurons and astrocytes, and preserves Cognitive functions in a Swine model of lethal hemorrhage. J Surg Res 2005;126:172-81.
6. Asai S, Zhao H, Takahashi Y, et al. Minimal effect of brain temperature changes on glutamate release in rat following severe global brain ischemia: a dialysis electrode study. Neuroreport 1998;9: 3863-8.
7. Asai S, Zhao H, Kohno T, et al. Quantitative evaluation of extracellular glutamate concentration in postischemic glutamate re-uptake, depen- dent on brain temperature, in the rat following severe global brain ischemia. Brain Res 2000;864:60-8.
8. Sharma HS, Miclescu A, Wiklund L. Cardiac arrest-induced regional blood-brain barrier breakdown, edema formation and brain pathology: a light and electron microscopic study on a new model for neurodegeneration and neuroprotection in porcine brain. J Neural Transm 2011;118:87-114.
9. Baena RC, Busto R, Dietrich WD, et al. Hyperthermia delayed by 24 hours aggravates neuronal damage in rat hippocampus following global ischemia. Neurology 1997;48:768-73.
10. Bogaert YE, Rosenthal RE, Fiskum G. Postischemic inhibition of cerebral cortex pyruvate dehydrogenase. Free Radic Biol Med 1994; 16:811-20.
11. Persson L, Hillered L. Chemical monitoring of neurosurgical intensive care patients using intracerebral microdialysis. J Neurosurg 1992;76: 72-80.
12. Robertson CL, Scafidi S, McKenna MC, et al. Mitochondrial mechanisms of cell death and neuroprotection in pediatric ischemic and traumatic brain injury. Exp Neurol 2009;218:371-80.
13. Reinstrup P, Stahl N, Mellergard P, et al. Intracerebral microdialysis in clinical practice: baseline values for chemical markers during wakefulness, anesthesia, and neurosurgery. Neurosurgery 2000;47: 701-9 [discussion 709-10].
14. Jones AE, Shapiro NI, Trzeciak S, et al. Lactate clearance vs Central venous oxygen saturation as goals of early sepsis therapy: a randomized clinical trial. JAMA 2010;303:739-46.
15. Donnino MW, Miller J, Goyal N, et al. Effective lactate clearance is associated with improved outcome in post-cardiac arrest patients. Resuscitation 2007;75:229-34.
16. Snyder JV, Nemoto EM, Carroll RG, et al. Global ischemia in dogs: intracranial pressures, brain blood flow and metabolism. Stroke 1975;6:21-7.
17. Noguchi K, Matsumoto N, Shiozaki T, et al. Effects of timing and duration of hypothermia on survival in an experimental gerbil model of global ischaemia. Resuscitation 2011;82:481-6.