Increasing experimental evidence along with accumulating human epidemiological data suggest a link between anaesthesia exposure and subsequent neurobehavioural deficits in young patient populations.1 Given the potential public health importance of this issue, identification of cellular and molecular pathways through which anaesthetics might hamper brain development is of utmost interest. Laboratory work conducted during the past decade has revealed that, through interference with growth factor signalling pathways and/or with mitochondrial homeostasis, general anaesthetics can induce neuronal cell death, impaired neurogenesis and altered synaptic network development at critical periods of brain development.2 The ultimate goal of these experimental studies, revealing fundamental insights into anaesthesia mechanisms of actions, is to develop therapeutic options allowing well tolerated perioperative care of newborns and young infants.
Activation of inflammatory pathways is an emerging candidate mechanism to explain the pathogenesis of long-term cognitive deficits related to the perioperative period. There is ample evidence that the peripheral immune message, triggered by pro-inflammatory cytokines, is relayed to the central nervous system (CNS) via both neural and humoral mechanisms.3 The negative impact of the resulting neuroinflammation on cognitive capacities in humans is well established.3–5 Importantly, infection/inflammation during pregnancy and the early postnatal period has been associated with brain lesions in humans.6 In experimental models, activation of inflammatory cascades in neonatal animals has been shown to predispose to cognitive and neurobehavioural deficits in adulthood.7,8
Recent laboratory observations suggest that exposure to both surgery and anaesthesia can trigger inflammatory cascades in the CNS.9–13 These data, along with the deleterious influence of neuroinflammation on brain development, raise the important issue of the impact of these drugs on inflammatory responses in the immature brain. Elucidating whether and how exposure to general anaesthetics per se induces activation of neuroinflammatory cascades in the developing brain is a first step to answering this question. Recent observations indicate that sevoflurane can induce cognitive impairment and neuroinflammation in the early postnatal period in a developmental stage and an exposure time dependent manner.14 The present study was designed to evaluate the impact of the commonly used general anaesthetic propofol on cytokine expression profiles in the developing rat brain. As a primary endpoint, we measured pro-inflammatory cytokine mRNA expression patterns in the prefrontal cortex and the hippocampus following propofol anaesthesia at defined stages of the brain growth spurt. As a secondary endpoint, we evaluated the impact of propofol on chemokine mRNA patterns under the same experimental conditions.
Materials and methods
This was a prospective, randomised, naive and placebo-controlled animal trial comparing propofol, vehicle injection and control. Analyses were conducted blinded to experimental groups. Figure 1 is a timeline diagram depicting the study design.
The primary endpoint was pro-inflammatory cytokine mRNA expression in the prefrontal cortex and hippocampus following a 6-h exposure to propofol at postnatal day 10 (PND10) and PND20. The secondary endpoint was chemokine mRNA expression in the prefrontal cortex and hippocampus following a 6-h exposure to propofol at PND10 and at PND20. Mortality during propofol anaesthesia was 0%.
The experimental protocol was conducted according to the guidelines of the Swiss Federal Veterinary Office and was approved by the Cantonal Veterinary Office Geneva, Switzerland (Authorisation number 1007/3477/2, Chaired by Dr Astrid Rod; approved on 2 July 2009).
Wistar rats were group-housed and bred in the animal facilities of the University of Geneva Medical School under light (12-h light/dark cycle) and temperature (22 ± 2°C) controlled conditions. Food and water were available ad libitum. Every effort was made to minimise the number of animals used and their suffering. Rats aged 10 to 20 days (males and females) were used for all experiments.
According to the experimental protocol, general anaesthesia was induced by intraperitoneal injection of propofol (Fresenius, Bad Homburg, Germany) 40 mg kg−1 on postnatal day 10 or 20. This first dose of the drug induced sedation (i.e. loss of the righting reflex) after 45 to 60 min. Following the initial dosing regimen, animals then received an hourly intraperitoneal injection (four in total) of propofol 20 mg kg−1 to maintain sedation for 6 h. Control sham-treated animals received intraperitoneal injections of a lipid vehicle solution (Lipofundin 20%; Fresenius, Bad Homburg, Germany) at equivalent volumes and frequency in each experimental protocol. All rats underwent the same maternal separation and handling as anaesthetised animals, and were kept in individual cages for the duration of the experimental procedure. Body temperature was monitored and maintained between 37°C and 38°C by means of a heating pad (Harvard Apparatus, Holliston, Massachusetts, USA). We did not perform blood gas analyses in the present study. In experiments in which rats were euthanised at 24 h following drug exposure, they were placed beside their mother once fully recovered from anaesthesia. Naive animals, not injected with propofol or lipofundin, were used for assessing baseline mRNA expressions at PND10 and PND20.
Rats were sacrificed by intracardiac perfusion of 0.9% physiological saline containing 10 IU ml−1 heparin under deep anaesthesia with 4% isoflurane at 6 or at 24 h following the initiation of propofol anaesthesia. The prefrontal cortex and the hippocampus were rapidly microdissected, frozen in liquid nitrogen and stored at -80°C until analysis. Total RNA was extracted with the Nucleospin kit and on-column deoxyribonuclease digested (Macherey-Nagel, Düren, Germany), following the manufacturer's instructions. The concentration and purity of RNA were determined spectrophotometrically, and cDNA was prepared from 1 μg total RNA using the ImPromII cDNA biosynthesis kit (Promega Corporation, Madison, Wisconsin, USA), as suggested by the manufacturer. Cytokine mRNAs were quantified by real-time PCR (RT-PCR), using TaqMan probes (Microsynth, Balgach, Switzerland) on an StepOnePlus (Applied Biosystems, Foster, California, USA). Probes were designed using the Primer3 programme, and specificity was confirmed by Basic Local Alignment Search Tool analysis. Briefly, RT-PCR was performed in a 25-μl reaction medium containing 12.5 μl master mix (ABgene, Epsom, UK), 400 nmol l–1 primers, 100 nmol l–1 TaqMan probe and previously reverse-transcribed cDNA. The PCR thermal protocol consisted of 40 to 46 cycles (denaturating at 95°C for 15 s, and annealing/extension at 60°C for 1 min). Concentrations of mRNA were normalised to those of the house-keeping gene succinate dehydrogenase (SDHA), using the comparative 2-ΔΔCt method,15 and expressed as a function of concentrations obtained under naive conditions. For the list of primers, see Supplementary Digital Content, https://links.lww.com/EJA/A54.
Data are expressed as median (interquartile range, IQR). Multiple group comparisons were performed by Kruskal–Wallis analysis of variance followed by Mann–Whitney U-tests. All statistical analyses were carried out using SPSS statistics version 20.0 for the Mac. The Bonferroni procedure for controlling family-wise error rate was applied.16 The null hypothesis was rejected if P was 0.05/7 = 0.007 or less. Box plots show median values, lower and upper quartiles, and largest and smallest values as whiskers if the values lay within 2.5 times the IQR.
Figure 2a shows that the 6-h propofol treatment paradigm induced a modest but significant increase in tumour necrosis factor (TNF) mRNA expression in the prefrontal cortex of PND10 animals at 6 h following the initiation of propofol anaesthesia [median (IQR) 1.8-fold relative change (1.7 to 2.2) versus 0.9-fold (0.6 to 1.1) relative change in vehicle-treated animals, P = 0.004]. There was also a tendency towards increased concentrations of TNF mRNA in the hippocampus at this age, although it did not reach significance [Fig. 2b, 1.4-fold (1.2 to 2.4) relative change versus 0.7-fold (0.6 to 1.1) relative change, P = 0.014]. When evaluated at 24 h postanaesthesia, no significant difference was observed between vehicle and propofol-treated animals in any of these regions (Fig. 2a,b). Propofol exposure of PND20 animals had no effect on TNF mRNA concentrations in any of these brain regions when assessed 6 h following the onset of drug administration (Fig. 2c,d). Similarly, we found that propofol did not seem to influence the mRNA expression concentrations of two other pro-inflammatory cytokines, interleukins IL-6 and IL-1β, in the developing brain either at PND10 or at PND20 (Figs. 3 and 4). Altogether, these results suggest that propofol anaesthesia per se does not seem to have a major impact on pro-inflammatory cytokine expression profiles in the developing brain during the brain growth spurt.
In addition to pro-inflammatory cytokines, we also investigated the effects of propofol on chemokine mRNA patterns by focusing on four particular candidates: Ccl2, Ccl3, Cxcl1 and Cx3cl1. At PND10, when assessed 6 h following the initiation of drug exposure, propofol anaesthesia induced a significant increase in Ccl2 mRNA concentrations both in the prefrontal cortex [Fig. 5a, 4.4-fold (3.8 to 5.6) relative change versus 1 (0.7 to 1.4) relative change in vehicle-treated animals, P = 0.0002] and in the hippocampus [Fig. 5b, 3.5-fold (2.8 to 5.3) relative change versus 1 (0.9 to 1.2) relative change in vehicle-treated animals, P = 0.0001]. No such changes were detected when assessed at 24 h following the onset of propofol administration (Fig. 5a,b), and this drug did not affect Ccl2 mRNA concentrations in PND20 rats (Fig. 5c,d). A very similar, transient developmental stage-dependent propofol-induced expression pattern was observed for the chemokine Ccl3. Indeed, following the 6-h long propofol exposure protocol in PND10 animals, propofol induced a 2.9-fold (2.6 to 4.31) relative increase in Ccl3 mRNA concentrations versus the vehicle treatment [1.2-fold (0.8 to 1.7) relative increase] in the prefrontal cortex (P = 0.0001; Fig. 6a) and a comparable 2.7-fold (2.2 to 3.6) relative increase following propofol administration was observed in the hippocampus versus a 1.1-fold (1.0 to 1.5) relative increase after vehicle treatment in the same region (P = 0.0003; Fig. 6b). Propofol did not affect Ccl3 mRNA concentrations when assessed at 24 h following drug treatment (Fig. 6a,b) and had no impact on Ccl3 mRNA expression in PND20 animals (Fig. 6 c,d). Propofol did not seem to affect either Cxcl1 or Cx3cl1 mRNA concentrations at any of the developmental time points and brain regions investigated (Figs. 7 and 8).
Identifying molecular and cellular mechanisms mediating the impact of anaesthetics on the developing brain is of crucial importance in light of the emerging clinical epidemiological data suggesting adverse effects of general anaesthesia in young infants.17 To study whether neuroinflammation could be a potential candidate to mediate developmental anaesthesia toxicity, we exposed newborn rats to propofol for 6 h at two distinct stages of the brain growth spurt and found that this anaesthesia protocol did not have any major impact on pro-inflammatory cytokine expression profiles either in the prefrontal cortex or in the hippocampus of these animals. In contrast, in the same brain regions, propofol anaesthesia induced an important transient and developmental stage-dependent increase in the mRNA concentrations of two members of the chemokine family of cytokines, Ccl2 and Ccl3. Altogether, these results argue against the involvement of neuroinflammatory pathways in the neurotoxicity observed following the administration of propofol in the early postnatal period. However, they also raise the possibility that this drug could transiently influence chemokine concentrations in the developing brain, the physiological relevance of which remains to be determined.
The aim of our study was to gain insights into a potential functional link between anaesthesia exposure, neuroinflammation and subsequent cognitive dysfunction in the developing brain. Several lines of clinical and experimental observations argue for such a possibility. Indeed, both neuroinflammation and anaesthesia exposure in the early postnatal period have been reported to be associated with long-term neurocognitive impairment in young patient populations.6,17,18 It is now well established that activation of inflammatory pathways in the periphery and the related increase in circulating pro-inflammatory cytokine concentrations rapidly leads to disruption of the blood-brain barrier.3 The subsequent entry of peripheral immune cells along with other pro-inflammatory modulators into the CNS will, in turn, trigger central neuroinflammation through activation of resident microglia and the local production of inflammatory mediators by these cells.3 The resulting structural alterations in neural circuitry will then lead to functional disconnection and are generally considered to underline impaired neurocognitive function.3–5 Of utmost interest, many of the morphofunctional changes triggered by central neuroinflammation are also important hallmarks characterising the response of the developing brain to anaesthesia exposure in the early postnatal period. For example, apoptosis of neural cells, a cardinal feature of developmental anaesthesia toxicity,1 has been described both in immature and adult brains in the context of neuroinflammation.19 Altered synaptic architecture, and thereby impaired neural circuitry function, has been reported in experimental models of general anaesthesia during the brain growth spurt, and similar changes can be triggered by pro-inflammatory signalling cascades. Impaired neurogenesis, trophic factor release along with signalling pathways leading to excitotoxicity have also been shown in the case of neuroinflammation as well as following general anaesthesia in the developing CNS.2,19 Last but not least, recent experimental observations demonstrate that anaesthesia exposure can induce pro-inflammatory cytokine expression in the nervous system.12–14 These data thus raise the hypothesis that neuroinflammation plays a central role in the causal link between anaesthesia exposure and cognitive deficits.
As a primary endpoint, we focused on pro-inflammatory cytokine mRNA expression profiles following a 6-h anaesthesia exposure. Our study revealed no impact of this treatment paradigm, applied at two functionally distinct stages of the brain growth spurt, on pro-inflammatory cytokine mRNA concentrations in the prefrontal cortex and hippocampus of young rats. These results suggest that neuroinflammation as a mechanism is most probably not involved in the morphofunctional effects that propofol exerts on the developing brain. Our data are in agreement with recent observations showing the lack of effects of general anaesthetics per se on neuroinflammation. In 6-day-old mice, a single 2-h exposure to sevoflurane (3%) anaesthesia did not influence pro-inflammatory cytokine concentrations,14 and similar results were obtained in 3-month-old mice, corresponding to young adults, following isoflurane anaesthesia.10 In contrast to the immature and to the young adult brain, a 2-h isoflurane (1.4%) exposure in 5 to 8-month-old mice rapidly induced a significant increase in pro-inflammatory cytokine concentrations of TNFα, IL-1β and IL-6.12 These results suggest that the neuroinflammatory potential of general anaesthetics is age-dependent. The molecular mechanisms underlying these age-dependent differences are yet to be determined. Importantly, the isoflurane anaesthesia protocol induced a significantly higher inflammatory response in the brains of Alzheimer's disease-like transgenic mice overexpressing human amyloid precursor protein and presenilin than in wild-type animals.12 Altogether, these observations argue for an increased risk of neuroinflammation following anaesthesia exposure in the elderly, especially in those presenting with neurodegenerative disease.
The secondary endpoint of the study focused on chemokine expression profiles following propofol anaesthesia. These investigations revealed that propofol induced a developmental stage-dependent transient increase in the expression of two members of the chemokine family: Ccl2 and Ccl3. The physiological significance of these results remains to be determined. Chemokines are small, secreted proteins that can attract and activate both immune and nonimmune cells.20,21 These molecules and their receptors are widely expressed in the CNS.20 In addition to their pathological implication in a variety of CNS disorders,19 chemokines may also have physiological functions including intercellular communication and neuromodulation.21 Notably, they were shown to regulate neurotransmitter release and intracellular Ca2+ levels and, thereby, can exert an important effect on neuronal activity.22,23 Because neuronal activity has a decisive role in brain circuitry development, and even short changes in neuronal activity patterns during critical periods of development can induce a long-term impact on neuronal networks, the propofol-induced transient increases in Ccl2 and Ccl3 expression could plausibly lead to changes in synaptic connectivity. In line with this hypothesis, we and others have recently shown that single exposure to general anaesthetics during the early postnatal period can induce long-term alterations of dendritic spines and synaptogenesis.24,25 Clearly, further studies are needed to determine a potential functional link between propofol exposure, chemokine expression and synaptogenesis.
Because the primary aim of our study was to evaluate the neuroinflammatory potential of propofol, our data do not give us indications as to whether and how this drug modulates systemic and/or central inflammatory responses elicited by peripheral surgery. In recent experimental studies, surgery under general anaesthesia but not general anaesthesia alone induced microglial activation and an increase in pro-inflammatory cytokines in the rodent brain.9–11 Because, for obvious reasons, no surgery was performed in the absence of anaesthesia, these studies do not provide us with insights into the immunomodulatory role of general anaesthesia in the context of neuroinflammation. Experimental induction of neuroinflammation without surgery and the combination of this model with the administration of general anaesthetics is needed to obtain more in-depth insights into this issue. These experiments would be of great value because the immunomodulatory role of general anaesthesia, either through specific mechanisms or through nonspecific attenuation of the perioperative stress response, has been an area of particular interest for a long time.26,27 Indeed, a large number of in-vitro and in-vivo studies demonstrate that general anaesthetics can disturb the function of various immune-competent cells and, thereby, can impair the immune response to aggression (for review see Schneemilch et al.27). However, the clinical significance of these experimental observations remains to be determined.28
There are several limitations of our study. First, because we focused on propofol, it does not allow us to draw conclusions on the neuroinflammatory potential of other general anaesthetics. This might be important because recent observations suggest that sevoflurane and desflurane have different characteristics in respect of inducing pro-inflammatory cytokine expression in the young brain.14 Moreover, we cannot exclude the possibility that longer or repeated exposure to propofol might still increase pro-inflammatory cytokines in the CNS. Indeed, although a single 2-h sevoflurane treatment did not induce neuroinflammation in the early postnatal period, repeated daily exposure to sevoflurane on three consecutive days did induce an increase in pro-inflammatory cytokines in the brain of young animals.14 It is nevertheless important to note that a 6-h propofol exposure, as used in our study, in the early postnatal period can already be considered as long in terms of translational relevance.29 A second potential limitation of the present work is that we cannot exclude the possibility that propofol anaesthesia at other time points during the brain growth spurt and/or in other brain regions may still induce neuroinflammatory responses. To strengthen the impact of our study, we focused on two developmental time points, depicting two functionally distinct stages of the early postnatal period when GABAA receptor mediated neurotransmission is either of excitatory or of inhibitory nature. We also investigated two brain regions, the prefrontal cortex and the hippocampus, in which the sensitivity to general anaesthetics during the early postnatal period is well established. Because we only measured mRNA concentrations, we cannot formally exclude, although highly improbable, changes at the protein concentration. Finally, given that in the present study we did not perform blood-gas analyses in animals undergoing propofol anaesthesia, we cannot formally exclude the possibility that our results are affected by nonphysiological alterations of systemic homeostasis. However, it is important to note that we have previously demonstrated that the propofol anaesthesia protocol performed in this work does not affect physiological patterns of homeostasis in these young animals.25
In conclusion, our study suggests that propofol anaesthesia does not impact on pro-inflammatory cytokine expression profiles in the developing CNS during the brain growth spurt. These results raise arguments against the involvement of neuroinflammatory pathways in the neurotoxicity observed following the administration of this drug in the early postnatal period. Further studies are needed to determine whether propofol and other general anaesthetics exert modulatory roles in neuroinflammatory processes triggered by peripheral surgery in the neonate.
Acknowledgements relating to this article
Assistance with the study: none.
Financial support and sponsorship: this study was supported by Swiss National Science Foundation Grants 31003A_130625 (to LV) and 32003B_134963/1 (to FMo) as well as FP7-INNOVATION I HEALTH-F2-2013-602114 (to FMa).
Conflicts of interests: none.
1. Vutskits L. General anesthetics in brain injury: friends or foes? Curr Pharm Des
2013; [Epub ahead of print].
2. Vutskits L. General anesthesia: a gateway to modulate synapse formation and neural plasticity? Anesth Analg
3. Konsman JP, Parnet P, Dantzer R. Cytokine-induced sickness behaviour: mechanisms and implications. Trends Neurosci
4. Capuron L, Lamarque D, Dantzer R, Goodall G. Attentional and mnemonic deficits associated with infectious disease in humans. Psychol Med
5. Dantzer R. Cytokine-induced sickness behaviour: a neuroimmune response to activation of innate immunity. Eur J Pharmacol
6. Hagberg H, Mallard C. Effect of inflammation on central nervous system development and vulnerability. Curr Opin Neurol
7. Bilbo SD, Biedenkapp JC, Der-Avakian A, et al. Neonatal infection-induced memory impairment after lipopolysaccharide in adulthood is prevented via caspase-1 inhibition. J Neurosci
8. Zerrate MC, Pletnikov M, Connors SL, et al. Neuroinflammation and behavioral abnormalities after neonatal terbutaline treatment in rats: implications for autism. J Pharmacol Exp Ther
9. Terrando N, Eriksson LI, Ryu JK, et al. Resolving postoperative neuroinflammation and cognitive decline. Ann Neurol
10. Cibelli M, Fidalgo AR, Terrando N, et al. Role of interleukin-1 in postoperative cognitive dysfunction. Ann Neurol
11. Terrando N, Monaco C, Ma D, et al. Tumor necrosis factor-alpha triggers a cytokine cascade yielding postoperative cognitive decline. Proc Natl Acad Sci U S A
12. Wu X, Lu Y, Dong Y, et al. The inhalation anesthetic isoflurane increases levels of proinflammatory TNF-α, IL-6, and IL-1β. Neurobiol Aging
13. Zhang L, Zhang J, Yang L, et al. Isoflurane and sevoflurane increase interleukin-6 levels through the nuclear factor-kappa B pathway in neuroglioma cells. Br J Anaesth
2013; 110 (Suppl 1):i82–i91.
14. Shen X, Dong Y, Xu Z, et al. Selective anesthesia-induced neuroinflammation in developing mouse brain and cognitive impairment. Anesthesiology
15. Livak KJ, Schmittgen TD. Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method. Methods
16. Montana G. Statistical methods in genetics. Brief Bioinform
17. DiMaggio C, Sun LS, Ing C, Li G. Pediatric anesthesia and neurodevelopmental impairments: a Bayesian meta-analysis. J Neurosurg Anesthesiol
18. Vutskits L, Davis PJ, Hansen T. Anesthetics and the developing brain: time for a change in practice? A pro/con debate. Paediatr Anaesth
19. Ramesh G, MacLean AG, Philipp MT. Cytokines and chemokines at the crossroads of neuroinflammation, neurodegeneration, and neuropathic pain. Mediators Inflamm
20. Bajetto A, Bonavia R, Barbero S, et al. Chemokines and their receptors in the central nervous system. Front Neuroendocrinol
21. Reaux-Le Goazigo A, Van Steenwinckel J, Rostene W, Mélik Parsadaniantz S. Current status of chemokines in the adult CNS. Prog Neurobiol
22. Biber K, Zuurman MW, Dijkstra IM, Boddeke HW. Chemokines in the brain: neuroimmunology and beyond. Curr Opin Pharmacol
23. Rostene W, Dansereau MA, Godefroy D, et al. Neurochemokines: a menage a trois providing new insights on the functions of chemokines in the central nervous system. J Neurochem
24. Lunardi N, Ori C, Erisir A, Jeftovic-Todorovic V. General anesthesia causes long-lasting disturbances in the ultrastructural properties of developing synapses in young rats. Neurotox Res
25. Briner A, Nikonenko I, De Roo M, et al. Developmental stage-dependent persistent impact of propofol anesthesia on dendritic spines in the rat medial prefrontal cortex. Anesthesiology
26. Scholl R, Bekker A, Babu R. Neuroendocrine and immune responses to surgery. Internet J Anesthesiol
27. Schneemilch CE, Schilling T, Bank U. Effects of general anaesthesia on inflammation. Best Pract Res Clin Anaesthesiol
28. Wagner F, Radermacher P, Stahl W. Anesthesia and the immune response: evidence for an ‘isoflurane paradox’? Shock
29. Soriano SG, Anand KJ, Rovnaghi CR, Hickey PR. Of mice and men: should we extrapolate rodent experimental data to the care of human neonates? Anesthesiology