The effects of xenon on sevoflurane anesthesia‐induced acidosis and brain cell apoptosis in immature rats

Anesthesia-induced neurodegeneration (AIN) occurs in newborn large animal modelswhere physiological homeostasis is maintained. However, most immature rodent models of AINreport mortality ratesand hypercarbia alone, mimicking that produced by anesthesia, increases neuro-apoptosis1 .Thepotential influence of physiological derangement is one reason why rodent models of AIN are notidealfor predicting clinical effect.However, they remain necessary for theinitial pre-clinical investigation of toxicity or treatment.


| INTRODUC TI ON
Anesthesia-induced neurodegeneration (AIN) occurs in newborn large animal models where physiological homeostasis is maintained. However, most immature rodent models of AIN report mortality rates, and hypercarbia alone, mimicking that produced by anesthesia, increases neuro-apoptosis. 1 The potential influence of physiological derangement is one reason why rodent models of AIN are not ideal for predicting clinical effect. However, they remain necessary for the initial pre-clinical investigation of toxicity or treatment.
Xenon, a cardio-stable sedative with analgesic properties and shown to reduce isoflurane-AIN in an immature rodent model, 2 was recently administered to critically ill babies in UK research trials and was shown to reduce the requirement for sevoflurane during anesthesia in another European trial. Co-administration during anesthesia for babies and young children potentially offers improved hemodynamics and neuroprotection.
We aimed to compare the effect of co-administering Xenon with sevoflurane on neuroapoptosis and acid-base homeostasis in immature rats.

| ME THOD
These experiments were carried out under Home Office License, with University ethical approval. Mixed sex, Wistar rats were randomly assigned on postnatal day 8 to 6-hour exposures to equipotent mixtures of sevoflurane ± xenon in 30% oxygen (equivalent to thrice the effective inhaled concentration of sevoflurane or xenon preventing cold-stimulated vocalization in 95% [EiC95 CSV] 3 ): 2.7% sevoflurane alone (Sevo), 1.8% sevoflurane in 35% xenon (SevoXe35) or 0.9% sevoflurane in 70% xenon (SevoXe70) (n = 9 per group). Two control groups were used (Naïve: anesthetized and culled on removal from the home cage, and Sham: expose to 30% oxygen alone [n = 6 per group]). Gasses were delivered using calibrated, low flow, rotameters and monitoring ensured that CO2 rebreathing was limited to 2%. Immediate decapitation allowed collection of mixed arteriovenous blood for analysis. There was no effect of control group (P = .645), therefore, differences in apoptotic cell count between Control, Sevo, SevoXe35, and SevoXe70 were tested using Kruskal-Wallis with Bonferroni correction for multiple comparisons. The relationships between RR and PCO2, and pH and induced apoptosis were estimated using  Lactate remained normal in our animals suggesting severe circulatory failure was not present, and the pH we report is in keeping with a recent study showing induced apoptosis in rats exposed to six hours of 2.5% sevoflurane (mean arterial blood pH: 7.04). 4 The relationships between RR and PCO2, and pH and induced apoptosis were estimated using Spearman's Rank correlation coefficient. There was a very strong negative correlation between respiratory rate and pCO2 of mixed arterio-venous blood (−0.99, P < .001). There was less acidosis with increased xenon/decreased sevoflurane (P < .001). Lactate was not raised in any animal (<3 mmol/L), and there was no difference between the three groups (P = .282).

| RE SULTS
There was a moderate negative correlation between pH and apoptotic cell count in the sensory cortex (−0.50, P = .028), but not in the hippocampus or the pre-frontal cortex (P = .092 and P = .342, respectively).

| CON CLUS IONS
Our study shows that xenon-co-administration in neonatal rats, used to reduce sevoflurane exposure, lessened physiological disturbances and this was accompanied by reduced apoptosis in the brain. Further pre-clinical studies are needed where cardiorespiratory derangement is minimized and equal between groups.
F I G U R E 1 Median (IQR) apoptotic cell count in four brain areas according to treatment group (Panels, A-D): There were few apoptotic cells in any control animals (<8/mm 2 )(A-D). Compared to controls, there were more apoptotic cells in the Hippocampus, Sensory Barrel Cortex and Pre-frontal Cortex in the Sevo and SevoXe35 groups, but not SevoXe70 (B-D). In the Hippocampus there were fewer apoptotic cells in SevoXe70 compared to Sevo group (B). In contrast, in the Caudate Putamen there were very few apoptotic cells seen and no difference between groups (A). Differences were tested using Kruskall-Wallis (P value above each panel) with Bonferroni correction for multiple comparisons (bracketed P values within panels). Median (IQR) Respiratory rate, PCO2 and pH of mixed arterio-venous blood taken immediately after 6-hour exposures (right sided panels): All animals exposed to sevoflurane alone or combined with xenon had lower respiratory rates compared to sham animals and were acidotic (pH < 7.35) [Colour figure can be viewed at wileyonlinelibrary.com]

CO N FLI C T O F I NTE R E S T
HG and AEP have no conflicts of interest.