(Br J Anaesth. 2018;121:406–416) Neonatal anesthesia-induced abnormalities and their mechanisms are poorly understood, even in exposed animals. Some studies have shown that rats, especially male rats, develop behavioral deficiencies and exacerbated hypothalamic-pituitary adrenal responses to stress when exposed as neonates to anesthetic. Furthermore, recent studies suggest that these deleterious effects could be carried into the next generation via epigenetic mechanisms (noncoding RNAs, DNA methylation). There is a pressing need for further research, in order to establish safety guidelines for anesthesia in children, The authors of the present study exposed neonatal rats to sevoflurane and evaluated their progeny for inherited behavioral and molecular alterations.

No abstract

Background We tested the hypothesis that the effects of traumatic brain injury, surgery and sevoflurane interact to induce neurobehavioral abnormalities in adult male rats and/or in their offspring (an animal model of intergenerational perioperative neurocognitive disorder). Methods Sprague-Dawley male rats (F0 generation) underwent a traumatic brain injury on postnatal day 60 that involved craniectomy (surgery) under 3% sevoflurane for 40 min followed by 2.1% sevoflurane for 3 h on postnatal days 62, 64, and 66 (injury group). The surgery group had craniectomy without traumatic brain injury, whereas the sevoflurane group had sevoflurane only. On postnatal day 90, F0 males and control females were mated to generate offspring (F1 generation). Results Acutely, F0 injury rats exhibited the greatest increases in serum corticosterone and interleukins 1β and 6, and activation of the hippocampal microglia. Long term, compared to controls, F0 injury rats had the most exacerbated corticosterone levels at rest (2.21 ± 0.64 vs. 7.28 ± 1.95 ng/ml, n = 8 - 7; P < 0.001) and 10 min after restraint (133.12 ± 33.98 vs. 232.83 ± 40.71 ng/ml, n =8 - 7; P < 0.001), increased interleukins 1β and 6, and reduced expression of hippocampal glucocorticoid receptor (Nr3c1) (0.53 ± 0.08 fold change relative to control, P < 0.001, n = 6) and brain-derived neurotrophic factor genes. They also exhibited greater behavioral deficiencies. Similar abnormalities were evident in their male offspring, whereas F1 females were not affected. The reduced Nr3c1 expression in F1 male, but not female, hippocampus was accompanied by corresponding Nr3c1 promoter hypermethylated CpG sites in F0 spermatozoa and F1 male, but not female, hippocampus. Conclusions These findings in rats suggest that young adult males with traumatic brain injury are at an increased risk of developing perioperative neurocognitive disorder, as are their unexposed male but not female offspring.

BACKGROUND: Sevoflurane (SEVO) increases neuronal excitation in neonatal rodent brains through alteration of gamma aminobutyric acid (GABA)(A) receptor signaling and increases corticosterone release. These actions may contribute to mechanisms that initiate the anesthetic’s long-term neuroendocrine and neurobehavioral effects. Dexmedetomidine (DEX), a non-GABAergic α2-adrenergic receptor agonist, is likely to counteract SEVO-induced neuronal excitation. We investigated how DEX pretreatment may alter the neurodevelopmental effects induced by SEVO in neonatal rats. METHODS: Postnatal day (P) 5 Sprague-Dawley male rats received DEX (25 µg/kg, intraperitoneal) or vehicle before exposure to 2.1% SEVO for 6 hours (the DEX + SEVO and SEVO groups, respectively). Rats in the DEX-only group received DEX without exposure to SEVO. A subcohort of P5 rats was used for electroencephalographic and serum corticosterone measurements. The remaining rats were sequentially evaluated in the elevated plus maze on P80, prepulse inhibition of the acoustic startle response on P90, Morris water maze (MWM) starting on P100, and for corticosterone responses to physical restraint for 30 minutes on P120, followed by assessment of epigenomic DNA methylation patterns in the hippocampus. RESULTS: Acutely, DEX depressed SEVO-induced electroencephalogram-detectable seizure-like activity (mean ± SEM, SEVO versus DEX + SEVO, 33.1 ± 5.3 vs 3.9 ± 5.3 seconds, P < .001), but it exacerbated corticosterone release (SEVO versus DEX + SEVO, 169.935 ± 20.995 versus 280.853 ± 40.963 ng/mL, P = .043). DEX diminished, but did not fully abolish, SEVO-induced corticosterone responses to restraint (control: 11625.230 ± 877.513, SEVO: 19363.555 ± 751.325, DEX + SEVO: 15012.216 ± 901.706, DEX-only: 12497.051 ± 999.816; F[3,31] = 16.878, P < .001) and behavioral deficiencies (time spent in the target quadrant of the MWM: control: 31.283% ± 1.722%, SEVO: 21.888% ± 2.187%, DEX + SEVO: 28.617% ± 1.501%, DEX-only: 31.339% ± 3.087%; F[3,67] = 3.944, P = .012) in adulthood. Of the 391 differentially methylated genes in the SEVO group, 303 genes in the DEX + SEVO group had DNA methylation patterns that were not different from those in the control group (ie, they were normal). DEX alone did not cause acute or long-term functional abnormalities. CONCLUSIONS: This study suggests that the ability of DEX to depress SEVO-induced neuronal excitation, despite increasing corticosterone release, is sufficient to weaken mechanisms leading to long-term neuroendocrine/neurobehavioral abnormalities. DEX may prevent changes in DNA methylation in the majority of genes affected by SEVO, epigenetic modifications that could predict abnormalities in a wide range of functions.