Mild hypothermia alone or in combination with anesthetic post-conditioning reduces expression of inflammatory cytokines in the cerebral cortex of pigs after cardiopulmonary resuscitation

Abstract

Introduction: Hypothermia improves survival and neurological recovery after cardiac arrest. Pro-inflammatory cytokines have been implicated in focal cerebral ischemia/reperfusion injury. It is unknown whether cardiac arrest also triggers the release of cerebral inflammatory molecules, and whether therapeutic hypothermia alters this inflammatory response. This study sought to examine whether hypothermia or the combination of hypothermia with anesthetic post-conditioning with sevoflurane affect cerebral inflammatory response after cardiopulmonary resuscitation.

Methods: Thirty pigs (28 to 34 kg) were subjected to cardiac arrest following temporary coronary artery occlusion. After seven minutes of ventricular fibrillation and two minutes of basic life support, advanced cardiac life support was started according to the current American Heart Association guidelines. Return of spontaneous circulation was achieved in 21 animals who were randomized to either normothermia at 38 degrees C, hypothermia at 33 degrees C or hypothermia at 33 degrees C combined with sevoflurane (each group: n = 7) for 24 hours. The effects of hypothermia and the combination of hypothermia with sevoflurane on cerebral inflammatory response after cardiopulmonary resuscitation were studied using tissue samples from the cerebral cortex of pigs euthanized after 24 hours and employing quantitative RT-PCR and ELISA techniques.

Results: Global cerebral ischemia following resuscitation resulted in significant upregulation of cerebral tissue inflammatory cytokine mRNA expression (mean +/- SD; interleukin (IL)-1beta 8.7 +/- 4.0, IL-6 4.3 +/- 2.6, IL-10 2.5 +/- 1.6, tumor necrosis factor (TNF)alpha 2.8 +/- 1.8, intercellular adhesion molecule-1 (ICAM-1) 4.0 +/- 1.9-fold compared with sham control) and IL-1beta protein concentration (1.9 +/- 0.6-fold compared with sham control). Hypothermia was associated with a significant (P < 0.05 versus normothermia) reduction in cerebral inflammatory cytokine mRNA expression (IL-1beta 1.7 +/- 1.0, IL-6 2.2 +/- 1.1, IL-10 0.8 +/- 0.4, TNFalpha 1.1 +/- 0.6, ICAM-1 1.9 +/- 0.7-fold compared with sham control). These results were also confirmed for IL-1beta on protein level. Experimental settings employing hypothermia in combination with sevoflurane showed that the volatile anesthetic did not confer additional anti-inflammatory effects compared with hypothermia alone.

Conclusions: Mild therapeutic hypothermia resulted in decreased expression of typical cerebral inflammatory mediators after cardiopulmonary resuscitation. This may confer, at least in part, neuroprotection following global cerebral ischemia and resuscitation.

Figure 1

Experimental time line. Thirty pigs were subjected to cardiac arrest following left anterior descending (LAD) coronary artery ischemia. Ventricular fibrillation (VF) was electrically induced twenty minutes after LAD occlusion. After seven minutes of VF, pigs were resuscitated (CPR). After successful return of spontaneous circulation (ROSC; n = 21), coronary perfusion was reestablished after 60 minutes of LAD occlusion, and animals were randomized either to normothermia at 38°C, hypothermia at 33°C or hypothermia at 33°C combined with sevoflurane (each group n = 7) for 24 hours. Five animals were sham operated. In the normothermia group, five out of seven animals survived for 24 hours compared to all animals in the hypothermia and hypothermia combined with sevoflurane group.

Figure 2

Cerebral cytokine mRNA expression. Transcript levels of the cerebral cytokines interleukin (IL)-1β, IL-6, IL-10, tumor necrosis factor (TNF)α and intercellular adhesion molecule (ICAM)-1 were determined by quantitative RT-PCR. NT, normothermia; HT, hypothermia; HT+SEV, hypothermia combined with sevoflurane. Data are expressed as mean ± SD (x-fold upregulation compared with Sham control). * P < 0.05, † P < 0.01 vs. Sham. §P < 0.05, #P < 0.01 vs. NT. RT-PCR data analysis was performed using two-sided Pair-wise fixed Reallocation Randomisation Test.

Figure 3

Protein concentration of interleukin-1β. Protein concentration of interleukin (IL)-1β was determined by a swine specific enzyme-linked-immunosorbent assay. NT, normothermia; HT, hypothermia; HT+SEV, hypothermia combined with sevoflurane. Data are expressed as mean ± SD (x-fold upregulation compared with Sham control). *P < 0.05 vs. Sham. §P < 0.05, #P < 0.01 vs. NT (using Mann-Whitney test).

Figure 4

Cerebral Bcl-2 and Bax mRNA expression. Transcript levels of the cerebral apoptosis-associated proteins Bcl-2 and Bax were determined by quantitative RT-PCR. NT, normothermia; HT, hypothermia; HT+SEV, hypothermia combined with sevoflurane. Data are expressed as mean ± SD (x-fold upregulation compared with Sham control). *P < 0.05 vs. Sham. §P < 0.05 vs. NT. RT-PCR data analysis was performed using two-sided Pair-wise fixed Reallocation Randomisation Test.