Abstract
Background:
The authors tested the hypothesis that combined use of dexmedetomidine on fetal rats during isoflurane exposure in maternal anesthesia can attenuate the abnormal spatial learning and memory abilities in adults via the renin–angiotensin–aldosterone system.
Methods:
Fifty timed-pregnancy rats were randomly assigned to five groups (Dex+Iso, Sal+Iso, Sal+Oxy, Dex+Oxy, and a control group ) on embryonic day 14 to receive five different dispositions, i.e. combined injection of dexmedetomidine (Dex) or saline (Sal) and inhalation of isoflurane (Iso), oxygen (Oxy), or normal air for 4 h (n = 10).
Results:
The latency time(s) from day 1 to day 4 all showed a decreasing tendency in all four groups. The synaptic count of the Sal+Iso group was significantly lower than the Control group (p < 0.05), suggesting that severe neurodegeneration occurred under the influence of fetal isoflurane exposure. In contrast, the synapse count of the Dex+Iso group was near to that of Control group. The rats are protected in neurodevelopmental, normal development.
Conclusion:
Combine use of dexmedetomidine during exposure to isoflurane in utero during middle-pregnancy can attenuate the impairment of spatial learning and memory abilities for the rats in adulthood.
Introduction
Because most non-obstetric surgery and fetal intervention procedures are performed during the second trimester of pregnancy, and high concentrations of anesthetics (1.5 minimum alveolar concentration) such as isoflurane are usually required to facilitate uterine quiescence and minimize the risk of preterm labor during these surgeries, 1 the risks of fetal brains being exposed to anesthetic agents are significantly increased. As the brain is developing in such a “growth spurt” phase, neurogenesis, neuronal migration, and corticogenesis are the major neurodevelopmental events. 2 Anesthetic agents administered can cause apoptosis-mediated neurodegeneration and synapse loss, whereas they increase synaptogenesis later in neurodevelopment.3–5 Recent studies based on rodents had demonstrated that maternal anesthesia in the second trimester pregnancy for several hours can induce neurodegeneration of the fetal rats. In the second trimester of pregnancy, both the rodent and human fetus show similar neurodevelopmental profiles.2,6 Therefore, more attention should be paid to the neurodevelopmental consequences of the fetus during maternal anesthesia in this phase.
In experiments of fetal rodents, the neurotoxicity of isoflurane is indicated in three ways. First, in results from immunohistochemical tests, caspase-3 was raised significantly in the hippocampus after exposure several hours later, which suggests that abnormal apoptosis occurred. Second, in results from behavior tests, anxiety-related behaviors were changed and spatial memory abilities were decreased in adulthood. Third, as observed by transmission electron microscopy, the number of synapses was reduced significantly in the hippocampus in adulthood. However, whether humans would undergo such changes is still unclear.
The neuroprotective function of dexmedetomidine has been demonstrated in 5–7 days postnatal rats. But there is still no evidence to show that dexmedetomidine can attenuate the neurodegeneration induced by isoflurane. Accordingly, we hypothesized that combined use of dexmedetomidine in second-trimester-pregnancy exposure to isoflurane in maternal anesthesia may have adverse effects on the fetal brain that can attenuate the abnormal spatial learning and memory abilities in adulthood.
Materials and methods
Subjects
With the approval of the Animal Ethics Committee (Guangdong Province), experiments were conducted on 50 Sprague–Dawley female rats (Medical Laboratory Animal Center of Guangdong Province, Guangzhou, China) and their respective offspring. The dams, weighing between 220 and 270 g, were housed in 60 × 40 × 20 cm3 polypropylene cages and exposed to a 12-h light–dark cycle in a temperature- and humidity-regulated room, fed with standard rat chow and water. Each dam stayed with a male adult rat for one night before pregnancy, and pregnancy was assumed if sperm was found on the vaginal smear at the next day morning. The pregnant rats were fed under the same conditions solely till delivery.
Procedure for the dams
On embryonic day 14 (E14), 50 dams were randomly assigned to five groups (n = 10), including Dex+Iso, Sal+Iso, Sal+Oxy, Dex+Oxy, and control groups. For gas inhalation, all dams were placed in a 20 × 12 × 12 cm3 organic glass chamber to inhale gas. The animals randomly assigned to Dex+Iso and Sal+Iso groups received 1.5% isoflurane in 100% oxygen for 4 h, whereas animals in Sal+Oxy and Dex+Oxy groups received 100% oxygen for 4 h in identical anesthetizing chambers. The animals breathed spontaneously and the isoflurane concentration was measured continuously with an agent analyzer (Datex-Ohmeda, S/5 Compact Anesthesia Monitor). This concentration of isoflurane was referenced from the study by Arvind et al. 7 which reported that 1.4% isoflurane can cause significant spatial memory impairment. Simultaneously, this concentration approximates one minimum alveolar concentration in the pregnant rodent. 8 The control group inhaled only normal air for 4 h.
For drugs injection, the animals randomly assigned to the Dex+Iso and Dex+Oxy groups received dexmedetomidine 10 μg/kg (2 μg/ml) by intramuscular injection at two time points, including 15 min before inhalation and 2 h after the start of inhalation. Animals in the Sal+Iso, Sal+Oxy, and control groups received two doses of saline at the same time points. The dose of dexmedetomidine was referenced from the report by Sanders et al. 9 which concluded that dexmedetomidine 10 μg/kg and 25 μg/kg gave significant neuroprotective effects while 1 μg/kg has almost no effect. In our pre-experiment, we initially set two doses of dexmedetomidine. But during the procedure, we found that dexmedetomidine 25 μg/kg may induce severe hypotension at subsequent isoflurane inhalation; this consequence may lead to hypoxia of the fetal rat brains and then affect the accuracy of our experiment. Although α1-adrenoceptor agonists such as phenylephrine can revert the blood pressure, this may counterbalance the effect of the α2-adrenoceptor agonist. 10 and thus interfere the neuroprotective effect of dexmedetomidine. For this reason, we only used a dose of 10 μg/kg for dexmedetomidine in our experiment.
After gas inhalation, the dams recovered in 100% oxygen till the return of the righting reflex, and returned to their cages. We injected the first dose of dexmedetomidine 15 min before gas inhalation. This setting was chosen by considering that the target organs of the drugs were the brains of fetal rats. Because the lipid solubility of dexmedetomidine is lower than isoflurane, it needs more duration to pass the placental barrier and to onset at the fetal brains. If both dexmedetomidine and isoflurane are received simultaneously, neuroimpairment induced by isoflurane may occur before neuroprotection by dexmedetomidine. We expected that advanced injection of dexmedetomidine can produce enough neuroprotective effect before the impairment. In addition, we found that the onset time of dexmedetomidine by intramuscular injection may be more rapid than by intraperitoneal injection from the animal’s behavioral response, and we conjectured that the onset time of neuroprotective effect may be shorter if dexmedetomidine was administered by intramuscular injection.
Furthermore, in our pre-experiment, 10 dams on E14 had a needle inserted through the arteria cruralis after chloral hydrate intraperitoneal injection. Each of these dams received all four methods above. Through the needles, we monitored the continual arterial blood pressures and took the blood samples for blood gas analysis during the whole procedures. The result of pre-experiment suggested that the dams can tolerate all the methods well. No significant hypotension or heart rate decrease occurred. Also, the blood gas analysis results suggested no hypoxemia or hypoglycemia. Considering the resistances of the rats without anesthesia when faced with invasive procedures, and the interference if we used other anesthetics such as chloral hydrate before the experiment, we did not monitor the blood pressure and blood gas analysis directly during our experiment.
Behavioral test
After delivery, each offspring was group-housed with the corresponding dam until postnatal day 21, when pups were weaned. One male and one female offspring of each dam were used for the Morris water maze (MWM) test. All testing sessions were performed between 9 a.m. and 5 p.m. in the same room, homogenously illuminated by normal fluorescent room light at 50 lux. Animals were singly housed till the end of behavioral test.
The MWM test was conducted in a pool with black internal coating (120 cm diameter, 60 cm altitude) filled with water (25°C). Animal movement was tracked using viewer software (Smart 2.5) by a camera placed 2.5 m above. The spatial cues were present around the pool, and the pool was divided into four quadrants (northwest, northeast, southeast, southwest). The test was divided into two parts. For the spatial learning part, an escape platform (10 cm diameter) was placed in the second quadrant (2 cm submerged; 30 cm from edge). Animals were trained to find the location of the platform in four acquisition trials (maximal swimming time 120 s; 30 s on the platform; inter-trial interval 60 min) per day during four consecutive days. If the rats could not find the platform in 120 s, they were directed to the platform with a short stick. Starting positions during the four trails day were: SE, SW, NW, and NE. The latency time(s) to find the hidden platform from the SE start position was scored. Latency time(s) represents the spatial learning ability. After the 120 s swim the animals were placed back in their home cage, and a towel was available inside the cage for additional drying. For the spatial memory part, all animals performed a single probe trial starting from the SE start position at day 5, in which the platform was removed from the swimming pool. Animals were also allowed to swim for 120 s. The time(s) of first platform crossing (at the former platform location) and the frequency of platform crossing were recorded. First platform crossing time(s) and frequency of platform crossing represent the spatial memory ability.
Transmission electron microscopy (TEM)
Hippocampal tissue were fixed in 2.5% glutaraldehyde for 1 h, treated with 1% osmium tetroxide, dehydrated and embedded in Durcupan (Sigma-Aldrich). Samples were then sectioned (60 nm), mounted on Cu-grids, and contrasted with uranyl acetate and lead citrate and examined by electron microscopy (EM-1200EX, JEOL).
Statistical analysis
All results were computed in SPSS 17.0. Measurement data were expressed as mean±SD. Data for the latency time(s) to find the hidden platform from day 1 to day 4 in the MWM test were analyzed using two-way ANOVA repeat measurements because of different time points. The other data were analyzed using one-way ANOVA, followed by Tukey post hoc multiple comparison tests. In all experiments, differences were considered statistically significant at p < 0.05.
Results
Morris water maze test
The latency time(s) from day 1 to day 4 all revealed a decreasing tendency in all four groups. However, in day 3, the data in the Sal+Oxy, Dex+Oxy, and control group were similar and all lower than the Sal+Iso group, while the data for the Dex+Iso group did not show significant difference to that for the Sal+Iso group or control group. In day 4, the data in the Dex+Iso, Sal+Oxy, Dex+Oxy, and control groups have no significant difference, but were all significantly different to the Sal+Iso group. The data for the Sal+Iso group were significantly longer than that of the control group (Table 1).
The latency time(s) from day 1 to day 4 (mean±SD).
Compared with data in Sal+Iso group, p < 0.05.
Compared with data in control group, p < 0.05.
The time(s) of first platform crossing and the frequency of platform crossing in the Dex+Iso, Sal+Oxy, Dex+Oxy, and control groups were similar, but all showed significant difference to the Sal+Iso group. The times of first platform crossing in the Sal+Iso group were significantly longer than in the control group, and the frequency of platform crossing in the Sal+Iso group was also significantly lower than for the control group (Table 2).
Time(s) of first platform crossing and frequency of platform crossing (mean±SD).
Compared with data in Sal+Iso group, p < 0.05.
Compared with data in control group, p < 0.05.
We analyzed the results obtained by electron microscopy. As shown in Table 3, we found that the synaptic count of Sal+Iso group rats was significantly lower than the for the control group, suggesting that severe neurodegeneration occurred under the influence of fetal isoflurane exposure. In contrast, the synapse count of the Dex+Iso group was nearly that of the control group, although there were significant differences (p < 0.05). Thus, the rats were protected in neurodevelopmental, normal development.
The number of synapses (SVD), synaptic cleft width (WSC), and the thickness of postsynaptic density (PSD) (mean±SD).
Compared with data in Sal+Iso group, p < 0.05.
Compared with data in control group, p < 0.05.
Discussion
Recently, growing evidence has shown that general anesthetics may be harmful at various stages of neurodevelopment.11–13 The neurodegeneration induced by exposure to isoflurane had both “short-term” and “long-term” influences. Wang et al. found that caspase-3 densities significantly increased both in the hippocampal CA1 region and retrosplenial cortex. 14 Arvind et al. found that rats exposed to isoflurane in utero may have impaired spatial memory acquisition and reduced anxiety. 7 Because E14–16 fetal rats show similar neurodevelopmental profiles to human in the second trimester of pregnancy,6,15 the period when most non-obstetric surgery and fetal interventions are performed, this sparked concerns for anesthesia safety. Therefore, we expect to find a way to abate this damage.
Dexmedetomidine has been shown to exert potent neuroprotective effects by different mechanisms. First, dexmedetomidine can attenuate the consecutively triggered excitotoxicity by glutamate via the α2A-adrenoceptor subtype in a model of neonatal excitotoxic rat brain damage. 16 Second, dexmedetomidine can reduce the abnormal increase of caspase-3 density induced by isoflurane in the developing brain. 9 Third, dexmedetomidine increased astrocyte expression of BDNF through an extracellular signal-regulated kinase-dependent pathway and induced subsequent neuroprotective effects. 17 However, all these evidences were based on experiments on neonatal rats; the evidences on fetal rats had still not been reported.
We evaluated whether dexmedetomidine has certainly neuroprotective effects by behavior tests, mainly because we have no evidence yet available. Behavior is the most important function of the body and is controlled by the nervous system. The behavior of the experimental animals would change subsequently with the environmental change, and it can reflect the adapting ability and the neural function sensitively. Learning and memory are the inner psychological behaviors of both humans and animals. These processes occur in the brain and can only be speculated indirectly by the determination of their operating results or reaction time after learning or performing a task interval a period. In our study, the MWM test was employed to evaluate the spatial learning and memory abilities.
Although the use of isoflurane in human medicine is now on the decline and replaced by sevoflurane, or desflurane gradually, a certain proportion of using as an inhalation anesthetic agent still exists for its uterus relaxation effects. Furthermore, the neurodegeneration effects on fetal animals induced by isoflurane have been widely reported, while few are reported for sevoflurane or desflurane. However, we needed an unquestionable neurodegeneration animal model to evaluate the protecting effect of dexmedetomidine. So we still chose isoflurane as the inhaling anesthetic agent.
From our MWM results, we find that the rats in all groups show approximate spatial learning abilities according to the similar data of the latency time(s) in the first two days. But in day 3 and day 4, the latency time(s) of the rats in Sal+Iso group were significant longer than the Sal+Oxy, Dex+Oxy and control groups. Simultaneously, the time(s) of first platform crossing and the frequency of platform crossing were significantly different compared with these three groups. This suggests that the spatial learning and memory abilities decreased in the Sal+Iso group, and that serious neurodegeneration had occurred in these rats. This result was consistent with the conclusion made by Arvind et al. 7 From the data for the Dex+Iso group, although the latency time(s) in day 3 were between those of the Sal+Iso and control groups and did not show any significant difference, the data in day 4 and day 5 were similar to the control group and significantly different to that for the Sal+Iso group. This suggests that the impairment of spatial learning and memory abilities induced by isoflurane had been attenuated, even though the effect may not be thorough. However, because we did not perform any biochemical or anatomical tests, the way in which neurodegeneration induced by isoflurane developed after combined administration of dexmedetomidine remains unclear.
We analyzed the results from electron microscopy. We found that the synaptic count of Sal+Iso group rats was significantly lower than for the control group, suggesting that severe neurodegeneration occurred under the influence of fetal isoflurane exposure. In contrast, the synapse count of the Dex+Iso group was nearly that of the control group, although there were significant differences (p < 0.05). Thus, the rats are protected in neurodevelopmental, normal development.
Oxygen has also been reported for neurotoxicity. 18 But we had to use oxygen for it is the only driven air source of our animal anesthesia machine. Similarly, a great number of human anesthesia machines only use oxygen and it is common practice to use 100% oxygen during maternal anesthesia in humans. Further, the fetal brain was probably not exposed to a high oxygen concentration because of the arterial–venous admixture in the placenta. But we still set a group for only inhaling oxygen, in order to rule out this interference in our experiment. From the result, oxygen exposure in utero did not lead to any significantly change of the behavior of the rats in adulthood.
The safety of using dexmedetomidine in the second trimester of pregnancy has been reported as being doubled in some articles. The usage guide of the US Food and Drug Administration (FDA) suggests that administering dexmedetomidine subcutaneously to pregnant rats at 8 and 32 μ/kg (representing a dose less than the maximum recommended human intravenous dose based on a body surface area comparison) from gestation day 16 through weaning, lower offspring weights were observed. Also Öcal et al. found that dexmedetomidine might increase the spontaneous uterine contraction-forces and contraction-frequencies in early- and middle-pregnancy term rats in vitro (in Krebs and Ca2+-free solutions), and consequently increase the risk of miscarriage. 19 However, Tariq et al. found that a single acute dose of dexmedetomidine (20 μg/kg) administered at the anticipated delivery time does not exert any adverse effects on perinatal morphology of pups, 20 including their birth weight, crown–rump length, physical growth, and postnatal behavioral performances. Podocyte injury has also been studied.21,22 However, in our experiment, none of the dams suffered a miscarriage. Simultaneously, exposure to a single dose of dexmedetomidine with oxygen did not lead to any behavior abnormally in our results. So we deduced that dexmedetomidine 10 μg/kg is still a safe dose for the pregnant and the fetal animals in the second trimester of pregnancy in vivo.
Recently, there have been a numbers of studies based on neonatal rats that focus on how to attenuate the neuroimpairment induced by anesthetic agents in the developing brain. Liu et al. found that concurrent surgery and procedural pain could attenuate ketamine-induced neuroapoptosis. 23 Xenon could prevent isoflurane-induced neonatal neuronal apoptosis. 24 Unfortunately, the offspring in utero may not all receive noxious stimulation and use of xenon is limited by the special equipment and source required, so that promotion of these conclusions is limited. By contrast, dexmedetomidine is both conveniently and widely available. Furthermore, the significant effect at attenuating the impairments of spatial learning and memory ability is worth a more thorough study. In our experiment, we were limited at the behavior test and lacked the ability to explore the mechanism. We expect that the mechanism of the neuroprotection by dexmedetomidine for fetal rats can be studied subsequently.
Footnotes
Conflict of interest
The authors declare that there is no conflict of interest.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.