- Email: [email protected]
Postoperative pain impairs subsequent performance on a spatial memory task via effects on N-methyl-d -aspartate receptor in aged rats
Life Sciences 93 (2013) 986–993
Contents lists available at ScienceDirect
Life Sciences journal homepage: www.elsevier.com/locate/lifescie
Postoperative pain impairs subsequent performance on a spatial memory task via effects on N-methyl-D-aspartate receptor in aged rats Haidong Chi a, Takashi Kawano a,⁎, Takahiko Tamura a, Hideki Iwata a, Yasuhiro Takahashi a, Satoru Eguchi b, Fumimoto Yamazaki a, Naoko Kumagai c, Masataka Yokoyama a a b c
Department of Anesthesiology and Intensive Care Medicine, Kochi Medical School, Kochi, Japan Department of Dental Anesthesiology, Tokushima University School of Dentistry, Tokushima, Japan Clinical Trial Center, Kochi Medical School, Kochi, Japan
a r t i c l e
i n f o
Article history: Received 1 July 2013 Accepted 25 October 2013 Keywords: Postoperative pain Cognition Aging NMDA receptor Memantine
a b s t r a c t Aims: Pain may be associated with postoperative cognitive dysfunction (POCD); however, this relationship remains underinvestigated. Therefore, we examined the impact of postoperative pain on cognitive functions in aged animals. Main methods: Rats were allocated to the following groups: control (C), 1.2 % isoflurane for 2 hours alone (I), I with laparotomy (IL), IL with analgesia using local ropivacaine (IL + R), and IL with analgesia using systemic morphine (IL + M). Pain was assessed by rat grimace scale (RGS). Spatial memory was evaluated using a radial maze from postoperative days (POD) 3 to 14. NMDA receptor (NR) 2 subunits in hippocampus were measured by ELISA. Finally, effects of memantine, a low-affinity uncompetitive N-methyl-D-aspartate (NMDA) receptor antagonist, on postoperative cognitive performance were tested. Key findings: Postoperative RGS was increased in Group IL, but not in other groups. The number of memory errors in Group I were comparable to that in Group C, whereas errors in Group IL were increased. Importantly, in Group IL + R and IL + M, cognitive impairment was not found. The memory errors were positively correlated with the levels of NMDA receptor 2 subunits in hippocampus. Prophylactic treatment with memantine could prevent the development of memory deficits observed in Group IL without an analgesic effect. Significance: Postoperative pain contributes to the development of memory deficits after anesthesia and surgery via up-regulation of hippocampal NMDA receptors. Our findings suggest that postoperative pain management may be important for the prevention of POCD in elderly patients. © 2013 Elsevier Inc. All rights reserved.
Introduction An increase in the aged population and advances in anesthetic and surgical techniques have increased the number of geriatric patients undergoing surgical operations under anesthesia (Chow et al., 2012). Consequently, postoperative cognitive dysfunction (POCD) has become a common problem in elderly patients (Monk et al., 2008). POCD is a temporary decline in cognitive functioning in the weeks or months following a surgical procedure. It is associated with long-term disability and even increased mortality (Steinmetz et al., 2009). Currently, however, there are few recognized intervention strategies for preventing POCD. A relationship between postoperative pain and long-term memory impairment may exist, but the data on this relationship are limited (Fong et al., 2006). Most of these data focus on the choice of ⁎ Corresponding author at: Department of Anesthesiology and Intensive Care Medicine, Kochi Medical School, Kohasu, Oko-cho, Nankoku, Kochi 783-8505, Japan. Tel.: +81 88 880 2471; fax: +81 88 880 2475. E-mail address: [email protected] (T. Kawano). 0024-3205/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.lfs.2013.10.028
postoperative analgesic rather than the pain itself. Nevertheless, pain is closely linked to postoperative delirium (Morrison et al., 2003; Vaurio et al., 2006), a condition in which patients suffer an acuteonset of cognitive decline that is distinct from POCD (Deiner and Silverstein, 2009). Although the continuum between postoperative delirium and POCD remains controversial (Deiner and Silverstein, 2009), postoperative delirium can progress to POCD (Saczynski et al., 2012). Therefore, we hypothesized that postoperative pain mediates the long-lasting memory deficits in elderly patients. The mechanisms that cause POCD in elderly patients are largely unknown. Several preclinical studies suggest that general anesthesia causes subsequent long-term memory impairment. For example, isoflurane anesthesia administered at clinically relevant doses causes long-term cognitive impairments in unoperated animals (Culley et al., 2004; Lin and Zuo, 2011; Cao et al., 2012). Other studies, however, indicate inhalation anesthesia can enhance cognitive functions after exposure (Komatsu et al., 1993; Rammes et al., 2009; Callaway et al., 2012). Cognitive improvement following isoflurane exposure was defined by a transient up-regulation of N-methyl-D-aspartate (NMDA) receptor (NR) in the hippocampus (Rammes et al., 2009). NRs play pivotal
H. Chi et al. / Life Sciences 93 (2013) 986–993
roles in learning, memory, and processing pain (Moriarty et al., 2011; Zhuo, 2009). Therefore, we further hypothesized that postoperative pain causes POCD by increasing NR expression. We tested these hypotheses by first investigating in isofluraneanesthetized aged rats the effects of postoperative pain on performance in working and reference memory tasks. Specifically, we tested 2 clinically relevant analgesics—a local anesthetic and a systemic opioid analgesic— for their ability to mitigate the effects of postoperative pain on cognitive dysfunction. Furthermore, to confirm NRs involvement in pain-related cognitive impairment, we examined whether the non-competitive NR antagonist memantine prevented the development of memory deficits after anesthesia and surgery.
987
the environment for 10 min. Facial expressions of rats were recorded by digital video cameras (HandyCam HDR-CX560, Sony, Japan) for 10 min before surgery (baseline), at 2, 4, 6, 8, 12, 24, and 48 h after inhalation period. Still-frames (front-view) were captured and cropped to display each rat's head to a non-participating assistant. Randomized and unlabeled facial images were presented on a high-resolution computer monitor, and the RGS was assigned by the experienced evaluator who was blinded to the study treatment using a 3-point scale (0 = no pain, 1 = moderate pain, 2 = intense pain) for each of the 4 RGS action units: orbital tightening, nose/cheek fluttering, ear position, and whisker change. The final RGS score is calculated by taking the average score of all 4 RGS action units.
Materials and methods Single and repeated open-field test Animals Male Wistar rats aged between 24 and 25 months, and weighing between 550 and 640 g, were used in this study. Subjects were maintained at approximately 90% of their free-feeding body weight during the entire radial arm maze procedure. All experiments were approved by the Institutional Animal Care and Use Committee of the Kochi Medical School.
A repeated open-field habituation test was conducted 1 week before the start of experiments to screen for baseline memory function in all rats. Rats were exposed to the same open field for 3 consecutive 5min trials, with an interval of 30 min between each trial. The activity counts at the third exposure were compared to those after the first exposure. Single open-field tests were subsequently performed to evaluate locomotor activity on POD 3, 7, and 14. The total accumulated counts of horizontal beam crosses were recorded for 60 min.
Anesthesia and surgery Simple laparotomy was used as an acute postoperative pain model. Anesthesia was induced and maintained with 1.2% isoflurane in 100% oxygen. One and a half hours after anesthesia was induced, a 1.0 cm longitudinal midline incision was made through the skin, abdominal muscle, and the peritoneum. The muscle layers were then repaired with 5–0 Vicryl sutures, and the skin was closed with tissue adhesive glue. The surgery duration was fixed at 10 min for each procedure, and the incised muscles were retracted during this period. Isoflurane anesthesia was continued until the total exposure time reached 2 h. This duration was selected because previous studies with isoflurane indicate both positive and negative effects on cognitive function with this time frame (Komatsu et al., 1993; Culley et al., 2004; Rammes et al., 2009; Lin and Zuo, 2011; Cao et al., 2012). Animals emerged from anesthesia 20 min after the end of surgery, which was due to the onset-time of systemic analgesia with postoperative morphine doses. In a pilot study in our laboratory, the minimum alveolar concentrations (MAC) value of isoflurane in aged rats was 1.01 ± 0.08% (n = 8). Therefore, 1.2% isoflurane is equivalent to approximately 1.2 MAC. Rats were allocated to 1 of 5 experimental groups (n = 8 rats/ group): (1) 100% oxygen inhalation for 2 h without surgery (Group C); (2) isoflurane anesthesia without surgery (Group I); (3) isoflurane anesthesia with laparotomy (Group IL); (4) isoflurane anesthesia with laparotomy plus single-dose surgical wound infiltration with 0.2% ropivacaine (300 μl) after surgery (Group IL + R); and (5) isoflurane anesthesia with laparotomy plus a single subcutaneous dose administration with 0.8 mg/kg morphine after surgery (Group IL + M). The doses of the 2 analgesic regimens used in this study were determined based on our preliminary findings. During the inhalation period, noninvasive recordings of pulse rate, arterial oxygen saturation, and MAP were measured by tail-cuff plethysmography. In a separate experiment, we tested whether the 2 analgesic methods used in this study show direct effects on memory function by treating the rats in Group I with either the analgesic or an identical volume of physiological saline (n = 8 rats/group). Measurement of pain Postoperative pain intensity was measured using a rat grimace scale (RGS), as previously reported (Sotocinal et al., 2011). Briefly, rats were placed individually in a clear plastic cage and allowed to acclimate to
Radial arm maze A 12-arm radial maze was used to evaluate spatial working and reference memory performance, with some modifications to a method described previously (Culley et al., 2004). Briefly, rats were habituated to the maze daily for 10 min, for 5 consecutive days. During this habituation phase, rats were allowed to freely explore and eat randomly scattered food rewards in the maze. The maze consisted of a circular center platform (34-cm diameter) surrounded by 12 equally spaced radial arms (50-cm-long, 11-cmwide, and 20-cm-high transparent walls). A plastic food cup intended to contain 45 mg of reward pellet was located at the end of each arm. Six arms were randomly assigned to be baited, and the remaining 6 arms were never baited. The locations of baited and unbaited arms were varied between animals, but remained consistent throughout testing for each rat. The maze was surrounded by multiple visual cues: posters, a door and windows, a chair, and some desks. All experimental trials were conducted during the dark phase of the light/dark cycle. Formal radial arm testing was performed for 12 consecutive days (POD 3 to 14). During each trial, rats were individually placed at the center of the maze and allowed to freely explore the maze until all 6 reward pellets in the food cups were consumed. Each session was terminated either when the rat had retrieved all reward pellets or after 15 min had elapsed. The numbers of errors were recorded, and the following 2 parameters of memory function were determined: (1) working memory errors, defined as repeat entries into the arms that had already been visited within a trial, and (2) reference memory errors, defined as first entries into non-baited arms. The data obtained were analyzed in 2-day blocks. The averaged latency per arm choice (s/arm entry) was also calculated as a measure of motivation and motor ability. All trials were monitored by an overhead video camera, and analysis of behavior was always scored by an investigator who was blind to experimental group assignment. After the completion of the last trial, all subjects were killed by rapid decapitation under terminal anesthesia with inhaled isoflurane. Three different brain regions, the hippocampus, prefrontal cortex (PFC), and amygdala were dissected according to the dissection method described by Glowinski and Iversen (Glowinski and Iversen, 1966). Brain samples were stored at − 80 °C until required for ELISA.
988
H. Chi et al. / Life Sciences 93 (2013) 986–993
ELISA
Results
Commercially available ELISA kits for measuring rat NR 2 (NR2; MBS701426, MyBioSource), IL-1β (ER2IL1B, Thermo Scientific), and TNF-α (438207, Biolegend) were used according to the manufacturers' instructions. Absorbance was read using an ELISA microplate reader (ThermoMax; Molecular Devices, Sunnyvale, CA). The inter-assay coefficients of variation for NR2, IL-1β, and TNF-α were 9.1%, 7.6%, and 7.9%, respectively, and the intra-assay coefficients of variation for NR2, IL-1β, and TNF-α were 7.0%, 3.3%, and 4.1%, respectively.
Repeated open-field tests were conducted on all studied rats before the start of experiments to measure non-aversive memory function. During 3 repeated trials, similar pattern of habituation were observed in each group: the horizontal locomotor activity at the 3rd trial was reduced to 46–51% of that at the 1st trial (data not shown). Total locomotor counts during trial were also comparable in each group. These results imply that baseline adaptive, memory, and locomotor function between groups may be comparable.
Pharmacological experiments
Rat grimace scale and locomotor activity after anesthesia and surgery
Antagonist studies were performed to test the effects of memantine on cognitive dysfunction after anesthesia and surgery. The rats in Group C and Group IL were injected i.p. with either memantine (SigmaAldrich; loading dose of 20 mg/kg immediately after surgery followed by maintenance doses of 2 mg/kg/day until POD 14), or saline alone as control treatment injected according to the same schedule (n = 8 rats/group). The dose of memantine was determined based on that used in a previous study that demonstrated improvement of cognitive performance in memory-impaired rats (Lukoyanov and Paula-Barbosa, 2001).
No differences in arterial oxygen saturation level, pulse rate, and body temperature were observed among groups during the inhalation period (data not shown). Fig. 1A shows the RGS changes in each group. In Groups C and I, RGS data throughout the recording period were comparable to the preoperative baseline values. In contrast, in Group IL, the RGS score increased significantly from baseline until 8 h after surgery. In Groups IL + R and IL + M, however, we observed no such increase in the RGS score, which suggests acute postoperative pain was successfully managed in these 2 groups. Further, as shown in Fig. 1B, there were no intergroup differences in spontaneous locomotor activity at any time point. The averaged latency per arm choice was also comparable in each experimental group (Fig. 1C).
Statistical analysis All data, including continuous variables, are expressed as the mean ± standard error of the mean (SEM). For each dependent variable, group and/or other main effect(s) were tested with repeatedmeasures ANOVA. Huynh–Feldt correction was applied to compensate for any violations of the assumption of sphericity. Whenever ANOVA indicated statistical significance in identifying a significant main effect, post hoc comparisons between the groups were performed in a pairwise manner by using the Wilcoxon–Mann–Whitney test with Bonferroni correction. All data were analyzed using the statistical software SAS (version 9.3; SAS Institute Inc., Cary, NC) and SPSS (versions 11; SPSS Inc, Chicago, IL). P b 0.05 was considered statistically significant.
Spatial memory task performance All groups showed continual improvements in working memory (main effect of POD, F (3.58, 125.39) = 129.34, p b 0.001) and reference memory (main effect for POD, F (3.41, 119.47) = 71.53, p b 0.001) across testing periods (Fig. 2A and B). With respect to working memory errors, a repeated-measures analysis of variance (ANOVA) showed a significant main effect for group (F (4, 35) = 6.31, p b 0.001), as well as a significant POD by group interaction (F (14.33, 125.39) = 3.25, p b 0.001). In particular, pairwise comparisons with Bonferroni correction indicated the total number of working memory errors in Group I were comparable to those
Fig. 1. Rat grimace scale scores and locomotor activity for each group. (A) Rat grimace scale (RGS) scores recorded at baseline and 2, 4, 6, 8, 12, 24, and 48 h after treatments and/or procedure. *p b 0.05 vs. baseline. (B) Activity counts were determined under novel conditions for 60 min on postoperative days (POD) 3, 7, and 14. (C) The average latency per arm choice in each experimental group. The 5 study groups (n = 8 in each group) are as indicated in Materials and methods. Each vertical bar represents the mean ± SEM.
H. Chi et al. / Life Sciences 93 (2013) 986–993
989
working (Fig. 4A and D) or reference memory (Fig. 4B and E). In addition, NR2 subunit levels in the hippocampus on POD 14 were comparable among the groups (Fig. 4C and F). Effects of prophylactic memantine treatment As shown in Fig. 5A, RGS score was unchanged by memantine treatment. Furthermore, we observed no differences among groups in spontaneous locomotor activity (Fig. 5B), or the averaged latency per arm choice (Fig. 5C). These results indicate that memantine has no effects on task motivation and motor ability at the dose used in this study. In addition, the levels of NR2 subunits in the hippocampus on POD 14 were not affected by treatment with memantine compared to the vehicle control in both Groups I and IL (Fig. 5D). A repeated-measures ANOVA failed to reveal significant differences between vehicle and treatment groups in both working (Fig. 6A) and reference memory errors (Fig. 6B). Significant differences were observed in Group IL, which was indicated by a significant main effect for treatment in both working (F(1, 14) = 5.67, p = 0.032) and reference (F(1, 14) = 6.36, p = 0.024) memory errors. This finding suggests that prophylactic treatment with memantine prevented subsequent cognitive deficits after laparotomy, whereas memantine exerted no effects in the anesthesia-only animals. Discussion
Fig. 2. Working and reference memory performance in aged rats using the 12-radial arm maze task. Mean numbers of working (A) and reference (B) memory errors for all groups from POD 3 to 14 were analyzed in each group. The 5 study groups (n = 8 in each group) are as indicated in Materials and methods. Left panel: Each point represents the mean values for 2-day trial blocks. Right panel: total errors over the entire testing period. Each vertical bar represents the mean ± SEM. *p b 0.05 vs. Group C.
in Group C. The laparotomy condition (Group IL), however, significantly increased the number of working memory errors. Interestingly, rats receiving either of the 2 analgesia regimens (Groups IL + R and IL + M) failed to show laparotomy-induced working memory impairments. For reference memory errors, a repeated-measures ANOVA revealed a significant main effect for group (F (4, 35) = 4.52, p = 0.0048), but not a POD by group interaction (F (13.65, 119.47) = 1.27, p = 0.24). Pairwise comparisons indicated trends similar to those observed with working memory errors (Fig. 2B). Levels of NR2 subunits after spatial memory testing Following behavioral testing, the levels of NR2 subunits in the hippocampus, PFC, and amygdala were determined using enzyme-linked immunosorbent assay (ELISA). NR2 subunit levels in the hippocampus were significantly higher in Group IL than in Group C (p b 0.01, Fig. 3A). A regression analysis revealed the number of working memory errors on POD 13–14 were positively correlated with hippocampus NR2 subunit levels (Fig. 3B, p b 0.0001). In contrast, no significant changes and correlation in the levels of NR2 subunits in the PFC (Fig. 3C and D) or amygdala (Fig. 3E and F) were observed among the 5 groups. Furthermore, there is no difference in IL-1β (Fig. 3G) or TNF-α (Fig. 3H) levels among the experimental groups on POD 14. Effects of analgesic regimens on memory and hippocampal NR2 subunit expression Compared to vehicle-treated anesthesia-only rats (Group I), rats that received analgesic regimens failed to show significant effects in
We observed that acute postoperative pain can lead to memory deficits after anesthesia and surgery in aged animals. Our results further demonstrate that memory deficits detected 14 days after surgery were correlated with increased NR2 subunit levels in the hippocampus. In addition, prophylactic treatment with memantine could help prevent postoperative memory deficits without alleviating the effects of postoperative pain. These results suggest that acute postoperative pain could induce long-lasting pathological activation of hippocampal NRs, which contribute to the development of POCD. Despite a significant amount of research detailing the effects of isoflurane anesthesia on cognitive function in non-surgical animals, a number of conflicting results have been reported. Some studies have reported impairments in cognitive performance, whereas others have reported improvements (Komatsu et al., 1993; Culley et al., 2004; Rammes et al., 2009; Lin and Zuo, 2011; Cao et al., 2012). Although our results show that isoflurane alone had no effect, differences between experimental results were reported in previous studies; these differences may be attributable to differences in animal species and age, anesthetic type and duration, and cognitive task type. In contrast, surgical models with anesthesia, which would seem to better reflect clinical situations, have been consistently reported to induce longterm cognitive dysfunction (Wan et al., 2007; Cibelli et al., 2010; Terrando et al., 2010). Similarly, in the present study, rats that received laparotomy under isoflurane anesthesia showed impairment of subsequent radial maze task performance. More importantly, postoperative analgesia with either local ropivacaine infiltration or systemic morphine ameliorated cognitive impairment. Furthermore, neither analgesic regimen affected cognitive function in the non-surgical rats that did not experience pain (Fig. 4). These findings suggest that the two analgesics used in this study prevented the development of POCD via their painrelieving effects, rather than through any direct effects on cognitive function. A previous systematic review indicated that no significant difference in the incidence of POCD between intravenous and epidural techniques for postoperative analgesia, although it is worth noting that the conclusions were limited by the small number of well-conducted clinical trials (Fong et al., 2006). This conclusion is consistent with our results, and suggests that adequate pain relief, independent of analgesic methods, may be important for preventing the development of POCD in elderly patients.
990
H. Chi et al. / Life Sciences 93 (2013) 986–993
Fig. 3. The levels of N-methyl-D-aspartate receptor 2 subunits and cytokines. The levels of N-methyl-D-aspartate receptor 2 (NR2) subunits were assayed by enzyme-linked immunosorbent assay (ELISA) for each group. The samples of hippocampus (A, B), prefrontal cortex (C, D), and amygdala (E, F) were obtained from rats in each group after completion of behavioral testing. Correlation of the number of the working memory errors on postoperative day (POD) 13–14 with each donor rat's levels of NR2 subunits in the hippocampus (B), PFC (D), and amygdala (F). Simple linear regression curve and confidence intervals are shown. The levels of IL-1β (G) and TNF-α (H) in hippocampus were also assayed by ELISA for each group. The 5 study groups (n = 8 in each group) are as indicated in Materials and methods. Each vertical bar represents the mean ± SEM. *p b 0.05 vs. Group C.
Our results demonstrated that spatial memory impairment was correlated with the increased expression of NR2 subunit in the hippocampus (Fig. 3B). Recent evidence demonstrates that excessive activation of NRs in hippocampus leads to neuro-excitotoxicity, which has been implicated in chronic neuronal degeneration (Collingridge et al., 2013). This pathological activation of NRs may play an important role in the pathogenesis of dementia (Zajaczkowski et al., 1997; Danysz and Parsons, 2012). Therefore, these results imply that increased NR
expression in hippocampus can induce pathological activation of these receptors, which may be an underlying mechanism for memory deficits after surgery with anesthesia in our model. In addition to memory processing, NRs play a role in pain transmission, for example, in central sensitization relating to the transition between acute and chronic pain (Zhuo, 2009; Voscopoulos and Lema, 2010). Furthermore, in the central nervous system, pain-induced synaptic plasticity has been reported to occur in overlapping brain regions
H. Chi et al. / Life Sciences 93 (2013) 986–993
991
Fig. 4. Effects of 2 analgesic regimens on spatial memory performance and hippocampal N-methyl-D-aspartate receptor 2 subunits. Mean number of working and reference memory errors from POD 3 to 14 and levels of hippocampal NR2 subunits on POD 14 were analyzed in Group I in the presence or absence of either infiltration of ropivacaine (A, B, C) or systemic morphine (D, E, F). The 4 study groups are as indicated in Materials and methods. Each vertical bar represents the mean ± SEM (n = 8 in each group).
involved in cognitive processes, such as the hippocampus, PFC, and amygdala (Zhuo, 2009; Ji et al., 2010; Moriarty et al., 2011). On the other hand, the hippocampus is known to the one of most vulnerable brain regions to aging, as well as to pathological process in age-related cognitive decline including the overactivation of NMDA receptors (Koch et al., 2004; Lister and Barnes, 2009; Danysz and Parsons, 2012). Therefore, enhanced vulnerability and stimulussensitivity of NMDA receptors with aging may partly explain the hippocampus-selective and long-lasting increase in NR2 subunit in response to acute pain stimulation. However, further research is required to determine why short-term acute postoperative pain can induce long-lasting up-regulation of NR2 subunit in hippocampus of aged animals. Memantine is a low-affinity uncompetitive NR antagonist, and can improve memory and cognition in both clinical and preclinical settings (Zajaczkowski et al., 1997; Rammes et al., 2008; Danysz and Parsons, 2012). The most remarkable properties of memantine are its fast channel-blocking kinetics and strong voltage-dependence, so that the drug may block only pathological activation of NRs while allowing their normal physiological activity level. Therefore, memantine could restore impaired memory in the absence of the side effects commonly
observed with other NR antagonists. As expected, our results showed that prophylactic treatment with memantine improved postoperative spatial memory performance in Group IL without producing locomotor or memory disturbances. More important, memantine at the dose used in our experiment showed no analgesic effect on acute postoperative pain (Fig. 5A) without any change in the levels of hippocampal NR2 subunit (Fig. 5D). These findings indicate that the overactivation of NRs may contribute to the development of memory deficits observed in Group IL. Thus, treatment with memantine might be a promising strategy for prevention of POCD. Previous studies reported that surgical trauma-induced peripheral cytokines can travel to the central nervous system, especially the hippocampus, and may contribute to surgery-induced cognitive dysfunction (Wan et al., 2007; Cibelli et al., 2010; Terrando et al., 2010; Steinman, 2010). In contrast, our results showed that the levels of both cytokines in hippocampus on either POD 14 were comparable among the experimental groups (Fig. 3F and G). These findings indicate that cytokine contributions may not be a significant factor in memory deficits in our experimental model. Our study has the following limitations. First, we used a midline laparotomy without visceral manipulation as postoperative pain model.
992
H. Chi et al. / Life Sciences 93 (2013) 986–993
Fig. 5. Effects of prophylactic memantine treatment on acute postoperative pain, locomotor activity, and the levels of N-methyl-D-aspartate receptor 2 subunits. (A) Levels of pain were assessed using the RGS at baseline and 2, 4, 6, 8, and 12 h after the inhalation period. *p b 0.05 vs. baseline. (B) Activity counts were determined under novel conditions for 60 min in Group I and Group IL in the presence or absence of memantine as detailed in the Methods. (C) The average latency per arm choice in each experimental group. (D) The levels of N-methyl-D-aspartate receptor 2 (NR2) subunits in hippocampus were assayed by enzyme-linked immunosorbent assay for each group. Each vertical bar represents the mean ± SEM (n = 8 in each group).
The benefit of this model is that it preserves food-seeking behavior and locomotor activity, both of which are essential for subsequent cognitive task performance. Nevertheless, since this model replicates only minimally invasive surgery; it may not accurately reflect the entire clinical situation. Second, we measured the hippocampal levels of NR2 subunits, but not NR1 subunits. While NR1 distributes ubiquitously in the brain, the expression of NR2 subunits varies according to the developmental stage. Therefore, the identity of the NR2 subunits helps to determine the pharmacological and biophysical properties of NMDA receptor (Clarke and Johnson, 2006). However, we cannot rule out that changes in the expression of NR1 subunits could also contribute to our results. Third, it remains still under question how the intensity and duration of postoperative pain could impact the increase in hippocampal NR2 subunit. In addition, detailed time-course analysis of NR2 levels associated with cognitive impairment certainly needs to be performed to translate our findings into the clinical setting.
Conclusion Our results indicate that effective postoperative analgesia can prevent the development of spatial memory decline after isoflurane anesthesia and laparotomy in aged rats. In addition, postoperative memory deficits were accompanied by up-regulation of hippocampal NRs, and pharmacological intervention with memantine attenuated of spatial memory impairment in non-analgesic-treated rats. Our findings suggest that postoperative pain management may be important for preventing POCD in elderly patients.
Conflict of interest statement The authors declare that there are no conflicts of interest.
Fig. 6. Effects of prophylactic memantine treatment on working and reference memory performance in aged rats. Mean number of working (A) and reference (B) memory errors from POD 3 to 14 were analyzed in Group I and Group IL in the presence or absence of memantine as detailed in the Methods. Left panel: Each point represents the mean values for 2-day trial blocks and vertical bars represent SEM. Right panel: total errors over the entire testing period. Each vertical bar represents the mean ± SEM (n = 8 in each group). *p b 0.05 vs. vehicle group.
Acknowledgments This work was supported in part by a Grant-in-Aid for Scientific Research (C): 24592339 and Grant-in-Aid for Challenging Exploratory Research (C): 24659699 from the Japan Society for the Promotion of Science, Tokyo, Japan.
References Callaway JK, Jones NC, Royse AG, Royse CF. Sevoflurane anesthesia does not impair acquisition learning or memory in the Morris water maze in young adult and aged rats. Anesthesiology 2012;117:1091–101. Cao L, Li L, Lin D, Zuo Z. Isoflurane induces learning impairment that is mediated by interleukin 1β in rodents. PLoS One 2012;7:e51431. Chow WB, Rosenthal RA, Merkow RP, Ko CY, Esnaola NF, American College of Surgeons National Surgical Quality Improvement Program, et al. Optimal preoperative assessment of the geriatric surgical patient: a best practices guideline from the American College of Surgeons National Surgical Quality Improvement Program and the American Geriatrics Society. J Am Coll Surg 2012;215:453–66. Cibelli M, Fidalgo AR, Terrando N, Ma D, Monaco C, Feldmann M, et al. Role of interleukin-1β in postoperative cognitive dysfunction. Ann Neurol 2010;68:360–8. Clarke RJ, Johnson JW. NMDA receptor NR2 subunit dependence of the slow component of magnesium unblock. J Neurosci 2006;26:5825–34. Collingridge GL, Volianskis A, Bannister N, France G, Hanna L, Mercier M, et al. The NMDA receptor as a target for cognitive enhancement. Neuropharmacology 2013;64:13–26. Culley DJ, Baxter MG, Yukhananov R, Crosby G. Long-term impairment of acquisition of a spatial memory task following isoflurane-nitrous oxide anesthesia in rats. Anesthesiology 2004;100:309–14. Danysz W, Parsons CG. Alzheimer's disease, β-amyloid, glutamate, NMDA receptors and memantine — searching for the connections. Br J Pharmacol 2012;167:324–52. Deiner S, Silverstein JH. Postoperative delirium and cognitive dysfunction. Br J Anaesth 2009;103:i41–6. Fong HK, Sands LP, Leung JM. The role of postoperative analgesia in delirium and cognitive decline in elderly patients: a systematic review. Anesth Analg 2006;102:1255–66. Glowinski J, Iversen LL. Regional studies of catecholamines in the rat brain. I. The disposition of [3H]norepinephrine, [3H]dopamine and [3H]dopa in various regions of the brain. J Neurochem 1966;13:655–69.
H. Chi et al. / Life Sciences 93 (2013) 986–993 Ji G, Sun H, Fu Y, Li Z, Pais-Vieira M, Galhardo V, et al. Cognitive impairment in pain through amygdala-driven prefrontal cortical deactivation. J Neurosci 2010;30:5451–64. Koch HJ, Szecsey A, Haen E. NMDA-antagonism (memantine): an alternative pharmacological therapeutic principle in Alzheimer's and vascular dementia. Curr Pharm Des 2004;10:253–9. Komatsu H, Nogaya J, Anabuki D, Yokono S, Kinoshita H, Shirakawa Y, et al. Memory facilitation by posttraining exposure to halothane, enflurane, and isoflurane in ddN mice. Anesth Analg 1993;76:609–12. Lin D, Zuo Z. Isoflurane induces hippocampal cell injury and cognitive impairments in adult rats. Neuropharmacology 2011;61:1354–9. Lister JP, Barnes CA. Neurobiological changes in the hippocampus during normative aging. Arch Neurol 2009;66:829–33. Lukoyanov NV, Paula-Barbosa MM. Memantine, but not dizocilpine, ameliorates cognitive deficits in adult rats withdrawn from chronic ingestion of alcohol. Neurosci Lett 2001;309:45–8. Monk TG, Weldon BC, Garvan CW, Dede DE, van der Aa MT, Heilman KM, et al. Predictors of cognitive dysfunction after major noncardiac surgery. Anesthesiology 2008;108:18–30. Moriarty O, McGuire BE, Finn DP. The effect of pain on cognitive function: a review of clinical and preclinical research. Prog Neurobiol 2011;93:385–404. Morrison RS, Magaziner J, Gilbert M, Koval KJ, McLaughlin MA, Orosz G, et al. Relationship between pain and opioid analgesics on the development of delirium following hip fracture. J Gerontol A Biol Sci Med Sci 2003;58:76–81. Rammes G, Danysz W, Parsons CG. Pharmacodynamics of memantine: an update. Curr Neuropharmacol 2008;6:55–78. Rammes G, Starker LK, Haseneder R, Berkmann J, Plack A, Zieglgänsberger W, et al. Isoflurane anaesthesia reversibly improves cognitive function and long-term
993
potentiation (LTP) via an up-regulation in NMDA receptor 2B subunit expression. Neuropharmacology 2009;56:626–36. Saczynski JS, Marcantonio ER, Quach L, Fong TG, Gross A, Inouye SK, et al. Cognitive trajectories after postoperative delirium. N Engl J Med 2012;367:30–9. Sotocinal SG, Sorge RE, Zaloum A, Tuttle AH, Martin LJ, Wieskopf JS, et al. The Rat Grimace Scale: a partially automated method for quantifying pain in the laboratory rat via facial expressions. Mol Pain 2011;7:55. Steinman L. Modulation of postoperative cognitive decline via blockade of inflammatory cytokines outside the brain. Proc Natl Acad Sci U S A 2010;107:20595–6. Steinmetz J, Christensen KB, Lund T, Lohse N, Rasmussen LS, ISPOCD Group. Long-term consequences of postoperative cognitive dysfunction. Anesthesiology 2009;110:548–55. Terrando N, Monaco C, Ma D, Foxwell BM, Feldmann M, Maze M. Tumor necrosis factor-alpha triggers a cytokine cascade yielding postoperative cognitive decline. Proc Natl Acad Sci U S A 2010;107:20518–22. Vaurio LE, Sands LP, Wang Y, Mullen EA, Leung JM. Postoperative delirium: the importance of pain and pain management. Anesth Analg 2006;102:1267–73. Voscopoulos C, Lema M. When does acute pain become chronic? Br J Anaesth 2010;105: i69–85. Wan Y, Xu J, Ma D, Zeng Y, Cibelli M, Maze M. Postoperative impairment of cognitive function in rats: a possible role for cytokine-mediated inflammation in the hippocampus. Anesthesiology 2007;106:436–43. Zajaczkowski W, Frankiewicz T, Parsons CG, Danysz W. Uncompetitive NMDA receptor antagonists attenuate NMDA-induced impairment of passive avoidance learning and LTP. Neuropharmacology 1997;36:961–71. Zhuo M. Plasticity of NMDA receptor NR2B subunit in memory and chronic pain. Mol Brain 2009;2:4.
Contents lists available at ScienceDirect
Life Sciences journal homepage: www.elsevier.com/locate/lifescie
Postoperative pain impairs subsequent performance on a spatial memory task via effects on N-methyl-D-aspartate receptor in aged rats Haidong Chi a, Takashi Kawano a,⁎, Takahiko Tamura a, Hideki Iwata a, Yasuhiro Takahashi a, Satoru Eguchi b, Fumimoto Yamazaki a, Naoko Kumagai c, Masataka Yokoyama a a b c
Department of Anesthesiology and Intensive Care Medicine, Kochi Medical School, Kochi, Japan Department of Dental Anesthesiology, Tokushima University School of Dentistry, Tokushima, Japan Clinical Trial Center, Kochi Medical School, Kochi, Japan
a r t i c l e
i n f o
Article history: Received 1 July 2013 Accepted 25 October 2013 Keywords: Postoperative pain Cognition Aging NMDA receptor Memantine
a b s t r a c t Aims: Pain may be associated with postoperative cognitive dysfunction (POCD); however, this relationship remains underinvestigated. Therefore, we examined the impact of postoperative pain on cognitive functions in aged animals. Main methods: Rats were allocated to the following groups: control (C), 1.2 % isoflurane for 2 hours alone (I), I with laparotomy (IL), IL with analgesia using local ropivacaine (IL + R), and IL with analgesia using systemic morphine (IL + M). Pain was assessed by rat grimace scale (RGS). Spatial memory was evaluated using a radial maze from postoperative days (POD) 3 to 14. NMDA receptor (NR) 2 subunits in hippocampus were measured by ELISA. Finally, effects of memantine, a low-affinity uncompetitive N-methyl-D-aspartate (NMDA) receptor antagonist, on postoperative cognitive performance were tested. Key findings: Postoperative RGS was increased in Group IL, but not in other groups. The number of memory errors in Group I were comparable to that in Group C, whereas errors in Group IL were increased. Importantly, in Group IL + R and IL + M, cognitive impairment was not found. The memory errors were positively correlated with the levels of NMDA receptor 2 subunits in hippocampus. Prophylactic treatment with memantine could prevent the development of memory deficits observed in Group IL without an analgesic effect. Significance: Postoperative pain contributes to the development of memory deficits after anesthesia and surgery via up-regulation of hippocampal NMDA receptors. Our findings suggest that postoperative pain management may be important for the prevention of POCD in elderly patients. © 2013 Elsevier Inc. All rights reserved.
Introduction An increase in the aged population and advances in anesthetic and surgical techniques have increased the number of geriatric patients undergoing surgical operations under anesthesia (Chow et al., 2012). Consequently, postoperative cognitive dysfunction (POCD) has become a common problem in elderly patients (Monk et al., 2008). POCD is a temporary decline in cognitive functioning in the weeks or months following a surgical procedure. It is associated with long-term disability and even increased mortality (Steinmetz et al., 2009). Currently, however, there are few recognized intervention strategies for preventing POCD. A relationship between postoperative pain and long-term memory impairment may exist, but the data on this relationship are limited (Fong et al., 2006). Most of these data focus on the choice of ⁎ Corresponding author at: Department of Anesthesiology and Intensive Care Medicine, Kochi Medical School, Kohasu, Oko-cho, Nankoku, Kochi 783-8505, Japan. Tel.: +81 88 880 2471; fax: +81 88 880 2475. E-mail address: [email protected] (T. Kawano). 0024-3205/$ – see front matter © 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.lfs.2013.10.028
postoperative analgesic rather than the pain itself. Nevertheless, pain is closely linked to postoperative delirium (Morrison et al., 2003; Vaurio et al., 2006), a condition in which patients suffer an acuteonset of cognitive decline that is distinct from POCD (Deiner and Silverstein, 2009). Although the continuum between postoperative delirium and POCD remains controversial (Deiner and Silverstein, 2009), postoperative delirium can progress to POCD (Saczynski et al., 2012). Therefore, we hypothesized that postoperative pain mediates the long-lasting memory deficits in elderly patients. The mechanisms that cause POCD in elderly patients are largely unknown. Several preclinical studies suggest that general anesthesia causes subsequent long-term memory impairment. For example, isoflurane anesthesia administered at clinically relevant doses causes long-term cognitive impairments in unoperated animals (Culley et al., 2004; Lin and Zuo, 2011; Cao et al., 2012). Other studies, however, indicate inhalation anesthesia can enhance cognitive functions after exposure (Komatsu et al., 1993; Rammes et al., 2009; Callaway et al., 2012). Cognitive improvement following isoflurane exposure was defined by a transient up-regulation of N-methyl-D-aspartate (NMDA) receptor (NR) in the hippocampus (Rammes et al., 2009). NRs play pivotal
H. Chi et al. / Life Sciences 93 (2013) 986–993
roles in learning, memory, and processing pain (Moriarty et al., 2011; Zhuo, 2009). Therefore, we further hypothesized that postoperative pain causes POCD by increasing NR expression. We tested these hypotheses by first investigating in isofluraneanesthetized aged rats the effects of postoperative pain on performance in working and reference memory tasks. Specifically, we tested 2 clinically relevant analgesics—a local anesthetic and a systemic opioid analgesic— for their ability to mitigate the effects of postoperative pain on cognitive dysfunction. Furthermore, to confirm NRs involvement in pain-related cognitive impairment, we examined whether the non-competitive NR antagonist memantine prevented the development of memory deficits after anesthesia and surgery.
987
the environment for 10 min. Facial expressions of rats were recorded by digital video cameras (HandyCam HDR-CX560, Sony, Japan) for 10 min before surgery (baseline), at 2, 4, 6, 8, 12, 24, and 48 h after inhalation period. Still-frames (front-view) were captured and cropped to display each rat's head to a non-participating assistant. Randomized and unlabeled facial images were presented on a high-resolution computer monitor, and the RGS was assigned by the experienced evaluator who was blinded to the study treatment using a 3-point scale (0 = no pain, 1 = moderate pain, 2 = intense pain) for each of the 4 RGS action units: orbital tightening, nose/cheek fluttering, ear position, and whisker change. The final RGS score is calculated by taking the average score of all 4 RGS action units.
Materials and methods Single and repeated open-field test Animals Male Wistar rats aged between 24 and 25 months, and weighing between 550 and 640 g, were used in this study. Subjects were maintained at approximately 90% of their free-feeding body weight during the entire radial arm maze procedure. All experiments were approved by the Institutional Animal Care and Use Committee of the Kochi Medical School.
A repeated open-field habituation test was conducted 1 week before the start of experiments to screen for baseline memory function in all rats. Rats were exposed to the same open field for 3 consecutive 5min trials, with an interval of 30 min between each trial. The activity counts at the third exposure were compared to those after the first exposure. Single open-field tests were subsequently performed to evaluate locomotor activity on POD 3, 7, and 14. The total accumulated counts of horizontal beam crosses were recorded for 60 min.
Anesthesia and surgery Simple laparotomy was used as an acute postoperative pain model. Anesthesia was induced and maintained with 1.2% isoflurane in 100% oxygen. One and a half hours after anesthesia was induced, a 1.0 cm longitudinal midline incision was made through the skin, abdominal muscle, and the peritoneum. The muscle layers were then repaired with 5–0 Vicryl sutures, and the skin was closed with tissue adhesive glue. The surgery duration was fixed at 10 min for each procedure, and the incised muscles were retracted during this period. Isoflurane anesthesia was continued until the total exposure time reached 2 h. This duration was selected because previous studies with isoflurane indicate both positive and negative effects on cognitive function with this time frame (Komatsu et al., 1993; Culley et al., 2004; Rammes et al., 2009; Lin and Zuo, 2011; Cao et al., 2012). Animals emerged from anesthesia 20 min after the end of surgery, which was due to the onset-time of systemic analgesia with postoperative morphine doses. In a pilot study in our laboratory, the minimum alveolar concentrations (MAC) value of isoflurane in aged rats was 1.01 ± 0.08% (n = 8). Therefore, 1.2% isoflurane is equivalent to approximately 1.2 MAC. Rats were allocated to 1 of 5 experimental groups (n = 8 rats/ group): (1) 100% oxygen inhalation for 2 h without surgery (Group C); (2) isoflurane anesthesia without surgery (Group I); (3) isoflurane anesthesia with laparotomy (Group IL); (4) isoflurane anesthesia with laparotomy plus single-dose surgical wound infiltration with 0.2% ropivacaine (300 μl) after surgery (Group IL + R); and (5) isoflurane anesthesia with laparotomy plus a single subcutaneous dose administration with 0.8 mg/kg morphine after surgery (Group IL + M). The doses of the 2 analgesic regimens used in this study were determined based on our preliminary findings. During the inhalation period, noninvasive recordings of pulse rate, arterial oxygen saturation, and MAP were measured by tail-cuff plethysmography. In a separate experiment, we tested whether the 2 analgesic methods used in this study show direct effects on memory function by treating the rats in Group I with either the analgesic or an identical volume of physiological saline (n = 8 rats/group). Measurement of pain Postoperative pain intensity was measured using a rat grimace scale (RGS), as previously reported (Sotocinal et al., 2011). Briefly, rats were placed individually in a clear plastic cage and allowed to acclimate to
Radial arm maze A 12-arm radial maze was used to evaluate spatial working and reference memory performance, with some modifications to a method described previously (Culley et al., 2004). Briefly, rats were habituated to the maze daily for 10 min, for 5 consecutive days. During this habituation phase, rats were allowed to freely explore and eat randomly scattered food rewards in the maze. The maze consisted of a circular center platform (34-cm diameter) surrounded by 12 equally spaced radial arms (50-cm-long, 11-cmwide, and 20-cm-high transparent walls). A plastic food cup intended to contain 45 mg of reward pellet was located at the end of each arm. Six arms were randomly assigned to be baited, and the remaining 6 arms were never baited. The locations of baited and unbaited arms were varied between animals, but remained consistent throughout testing for each rat. The maze was surrounded by multiple visual cues: posters, a door and windows, a chair, and some desks. All experimental trials were conducted during the dark phase of the light/dark cycle. Formal radial arm testing was performed for 12 consecutive days (POD 3 to 14). During each trial, rats were individually placed at the center of the maze and allowed to freely explore the maze until all 6 reward pellets in the food cups were consumed. Each session was terminated either when the rat had retrieved all reward pellets or after 15 min had elapsed. The numbers of errors were recorded, and the following 2 parameters of memory function were determined: (1) working memory errors, defined as repeat entries into the arms that had already been visited within a trial, and (2) reference memory errors, defined as first entries into non-baited arms. The data obtained were analyzed in 2-day blocks. The averaged latency per arm choice (s/arm entry) was also calculated as a measure of motivation and motor ability. All trials were monitored by an overhead video camera, and analysis of behavior was always scored by an investigator who was blind to experimental group assignment. After the completion of the last trial, all subjects were killed by rapid decapitation under terminal anesthesia with inhaled isoflurane. Three different brain regions, the hippocampus, prefrontal cortex (PFC), and amygdala were dissected according to the dissection method described by Glowinski and Iversen (Glowinski and Iversen, 1966). Brain samples were stored at − 80 °C until required for ELISA.
988
H. Chi et al. / Life Sciences 93 (2013) 986–993
ELISA
Results
Commercially available ELISA kits for measuring rat NR 2 (NR2; MBS701426, MyBioSource), IL-1β (ER2IL1B, Thermo Scientific), and TNF-α (438207, Biolegend) were used according to the manufacturers' instructions. Absorbance was read using an ELISA microplate reader (ThermoMax; Molecular Devices, Sunnyvale, CA). The inter-assay coefficients of variation for NR2, IL-1β, and TNF-α were 9.1%, 7.6%, and 7.9%, respectively, and the intra-assay coefficients of variation for NR2, IL-1β, and TNF-α were 7.0%, 3.3%, and 4.1%, respectively.
Repeated open-field tests were conducted on all studied rats before the start of experiments to measure non-aversive memory function. During 3 repeated trials, similar pattern of habituation were observed in each group: the horizontal locomotor activity at the 3rd trial was reduced to 46–51% of that at the 1st trial (data not shown). Total locomotor counts during trial were also comparable in each group. These results imply that baseline adaptive, memory, and locomotor function between groups may be comparable.
Pharmacological experiments
Rat grimace scale and locomotor activity after anesthesia and surgery
Antagonist studies were performed to test the effects of memantine on cognitive dysfunction after anesthesia and surgery. The rats in Group C and Group IL were injected i.p. with either memantine (SigmaAldrich; loading dose of 20 mg/kg immediately after surgery followed by maintenance doses of 2 mg/kg/day until POD 14), or saline alone as control treatment injected according to the same schedule (n = 8 rats/group). The dose of memantine was determined based on that used in a previous study that demonstrated improvement of cognitive performance in memory-impaired rats (Lukoyanov and Paula-Barbosa, 2001).
No differences in arterial oxygen saturation level, pulse rate, and body temperature were observed among groups during the inhalation period (data not shown). Fig. 1A shows the RGS changes in each group. In Groups C and I, RGS data throughout the recording period were comparable to the preoperative baseline values. In contrast, in Group IL, the RGS score increased significantly from baseline until 8 h after surgery. In Groups IL + R and IL + M, however, we observed no such increase in the RGS score, which suggests acute postoperative pain was successfully managed in these 2 groups. Further, as shown in Fig. 1B, there were no intergroup differences in spontaneous locomotor activity at any time point. The averaged latency per arm choice was also comparable in each experimental group (Fig. 1C).
Statistical analysis All data, including continuous variables, are expressed as the mean ± standard error of the mean (SEM). For each dependent variable, group and/or other main effect(s) were tested with repeatedmeasures ANOVA. Huynh–Feldt correction was applied to compensate for any violations of the assumption of sphericity. Whenever ANOVA indicated statistical significance in identifying a significant main effect, post hoc comparisons between the groups were performed in a pairwise manner by using the Wilcoxon–Mann–Whitney test with Bonferroni correction. All data were analyzed using the statistical software SAS (version 9.3; SAS Institute Inc., Cary, NC) and SPSS (versions 11; SPSS Inc, Chicago, IL). P b 0.05 was considered statistically significant.
Spatial memory task performance All groups showed continual improvements in working memory (main effect of POD, F (3.58, 125.39) = 129.34, p b 0.001) and reference memory (main effect for POD, F (3.41, 119.47) = 71.53, p b 0.001) across testing periods (Fig. 2A and B). With respect to working memory errors, a repeated-measures analysis of variance (ANOVA) showed a significant main effect for group (F (4, 35) = 6.31, p b 0.001), as well as a significant POD by group interaction (F (14.33, 125.39) = 3.25, p b 0.001). In particular, pairwise comparisons with Bonferroni correction indicated the total number of working memory errors in Group I were comparable to those
Fig. 1. Rat grimace scale scores and locomotor activity for each group. (A) Rat grimace scale (RGS) scores recorded at baseline and 2, 4, 6, 8, 12, 24, and 48 h after treatments and/or procedure. *p b 0.05 vs. baseline. (B) Activity counts were determined under novel conditions for 60 min on postoperative days (POD) 3, 7, and 14. (C) The average latency per arm choice in each experimental group. The 5 study groups (n = 8 in each group) are as indicated in Materials and methods. Each vertical bar represents the mean ± SEM.
H. Chi et al. / Life Sciences 93 (2013) 986–993
989
working (Fig. 4A and D) or reference memory (Fig. 4B and E). In addition, NR2 subunit levels in the hippocampus on POD 14 were comparable among the groups (Fig. 4C and F). Effects of prophylactic memantine treatment As shown in Fig. 5A, RGS score was unchanged by memantine treatment. Furthermore, we observed no differences among groups in spontaneous locomotor activity (Fig. 5B), or the averaged latency per arm choice (Fig. 5C). These results indicate that memantine has no effects on task motivation and motor ability at the dose used in this study. In addition, the levels of NR2 subunits in the hippocampus on POD 14 were not affected by treatment with memantine compared to the vehicle control in both Groups I and IL (Fig. 5D). A repeated-measures ANOVA failed to reveal significant differences between vehicle and treatment groups in both working (Fig. 6A) and reference memory errors (Fig. 6B). Significant differences were observed in Group IL, which was indicated by a significant main effect for treatment in both working (F(1, 14) = 5.67, p = 0.032) and reference (F(1, 14) = 6.36, p = 0.024) memory errors. This finding suggests that prophylactic treatment with memantine prevented subsequent cognitive deficits after laparotomy, whereas memantine exerted no effects in the anesthesia-only animals. Discussion
Fig. 2. Working and reference memory performance in aged rats using the 12-radial arm maze task. Mean numbers of working (A) and reference (B) memory errors for all groups from POD 3 to 14 were analyzed in each group. The 5 study groups (n = 8 in each group) are as indicated in Materials and methods. Left panel: Each point represents the mean values for 2-day trial blocks. Right panel: total errors over the entire testing period. Each vertical bar represents the mean ± SEM. *p b 0.05 vs. Group C.
in Group C. The laparotomy condition (Group IL), however, significantly increased the number of working memory errors. Interestingly, rats receiving either of the 2 analgesia regimens (Groups IL + R and IL + M) failed to show laparotomy-induced working memory impairments. For reference memory errors, a repeated-measures ANOVA revealed a significant main effect for group (F (4, 35) = 4.52, p = 0.0048), but not a POD by group interaction (F (13.65, 119.47) = 1.27, p = 0.24). Pairwise comparisons indicated trends similar to those observed with working memory errors (Fig. 2B). Levels of NR2 subunits after spatial memory testing Following behavioral testing, the levels of NR2 subunits in the hippocampus, PFC, and amygdala were determined using enzyme-linked immunosorbent assay (ELISA). NR2 subunit levels in the hippocampus were significantly higher in Group IL than in Group C (p b 0.01, Fig. 3A). A regression analysis revealed the number of working memory errors on POD 13–14 were positively correlated with hippocampus NR2 subunit levels (Fig. 3B, p b 0.0001). In contrast, no significant changes and correlation in the levels of NR2 subunits in the PFC (Fig. 3C and D) or amygdala (Fig. 3E and F) were observed among the 5 groups. Furthermore, there is no difference in IL-1β (Fig. 3G) or TNF-α (Fig. 3H) levels among the experimental groups on POD 14. Effects of analgesic regimens on memory and hippocampal NR2 subunit expression Compared to vehicle-treated anesthesia-only rats (Group I), rats that received analgesic regimens failed to show significant effects in
We observed that acute postoperative pain can lead to memory deficits after anesthesia and surgery in aged animals. Our results further demonstrate that memory deficits detected 14 days after surgery were correlated with increased NR2 subunit levels in the hippocampus. In addition, prophylactic treatment with memantine could help prevent postoperative memory deficits without alleviating the effects of postoperative pain. These results suggest that acute postoperative pain could induce long-lasting pathological activation of hippocampal NRs, which contribute to the development of POCD. Despite a significant amount of research detailing the effects of isoflurane anesthesia on cognitive function in non-surgical animals, a number of conflicting results have been reported. Some studies have reported impairments in cognitive performance, whereas others have reported improvements (Komatsu et al., 1993; Culley et al., 2004; Rammes et al., 2009; Lin and Zuo, 2011; Cao et al., 2012). Although our results show that isoflurane alone had no effect, differences between experimental results were reported in previous studies; these differences may be attributable to differences in animal species and age, anesthetic type and duration, and cognitive task type. In contrast, surgical models with anesthesia, which would seem to better reflect clinical situations, have been consistently reported to induce longterm cognitive dysfunction (Wan et al., 2007; Cibelli et al., 2010; Terrando et al., 2010). Similarly, in the present study, rats that received laparotomy under isoflurane anesthesia showed impairment of subsequent radial maze task performance. More importantly, postoperative analgesia with either local ropivacaine infiltration or systemic morphine ameliorated cognitive impairment. Furthermore, neither analgesic regimen affected cognitive function in the non-surgical rats that did not experience pain (Fig. 4). These findings suggest that the two analgesics used in this study prevented the development of POCD via their painrelieving effects, rather than through any direct effects on cognitive function. A previous systematic review indicated that no significant difference in the incidence of POCD between intravenous and epidural techniques for postoperative analgesia, although it is worth noting that the conclusions were limited by the small number of well-conducted clinical trials (Fong et al., 2006). This conclusion is consistent with our results, and suggests that adequate pain relief, independent of analgesic methods, may be important for preventing the development of POCD in elderly patients.
990
H. Chi et al. / Life Sciences 93 (2013) 986–993
Fig. 3. The levels of N-methyl-D-aspartate receptor 2 subunits and cytokines. The levels of N-methyl-D-aspartate receptor 2 (NR2) subunits were assayed by enzyme-linked immunosorbent assay (ELISA) for each group. The samples of hippocampus (A, B), prefrontal cortex (C, D), and amygdala (E, F) were obtained from rats in each group after completion of behavioral testing. Correlation of the number of the working memory errors on postoperative day (POD) 13–14 with each donor rat's levels of NR2 subunits in the hippocampus (B), PFC (D), and amygdala (F). Simple linear regression curve and confidence intervals are shown. The levels of IL-1β (G) and TNF-α (H) in hippocampus were also assayed by ELISA for each group. The 5 study groups (n = 8 in each group) are as indicated in Materials and methods. Each vertical bar represents the mean ± SEM. *p b 0.05 vs. Group C.
Our results demonstrated that spatial memory impairment was correlated with the increased expression of NR2 subunit in the hippocampus (Fig. 3B). Recent evidence demonstrates that excessive activation of NRs in hippocampus leads to neuro-excitotoxicity, which has been implicated in chronic neuronal degeneration (Collingridge et al., 2013). This pathological activation of NRs may play an important role in the pathogenesis of dementia (Zajaczkowski et al., 1997; Danysz and Parsons, 2012). Therefore, these results imply that increased NR
expression in hippocampus can induce pathological activation of these receptors, which may be an underlying mechanism for memory deficits after surgery with anesthesia in our model. In addition to memory processing, NRs play a role in pain transmission, for example, in central sensitization relating to the transition between acute and chronic pain (Zhuo, 2009; Voscopoulos and Lema, 2010). Furthermore, in the central nervous system, pain-induced synaptic plasticity has been reported to occur in overlapping brain regions
H. Chi et al. / Life Sciences 93 (2013) 986–993
991
Fig. 4. Effects of 2 analgesic regimens on spatial memory performance and hippocampal N-methyl-D-aspartate receptor 2 subunits. Mean number of working and reference memory errors from POD 3 to 14 and levels of hippocampal NR2 subunits on POD 14 were analyzed in Group I in the presence or absence of either infiltration of ropivacaine (A, B, C) or systemic morphine (D, E, F). The 4 study groups are as indicated in Materials and methods. Each vertical bar represents the mean ± SEM (n = 8 in each group).
involved in cognitive processes, such as the hippocampus, PFC, and amygdala (Zhuo, 2009; Ji et al., 2010; Moriarty et al., 2011). On the other hand, the hippocampus is known to the one of most vulnerable brain regions to aging, as well as to pathological process in age-related cognitive decline including the overactivation of NMDA receptors (Koch et al., 2004; Lister and Barnes, 2009; Danysz and Parsons, 2012). Therefore, enhanced vulnerability and stimulussensitivity of NMDA receptors with aging may partly explain the hippocampus-selective and long-lasting increase in NR2 subunit in response to acute pain stimulation. However, further research is required to determine why short-term acute postoperative pain can induce long-lasting up-regulation of NR2 subunit in hippocampus of aged animals. Memantine is a low-affinity uncompetitive NR antagonist, and can improve memory and cognition in both clinical and preclinical settings (Zajaczkowski et al., 1997; Rammes et al., 2008; Danysz and Parsons, 2012). The most remarkable properties of memantine are its fast channel-blocking kinetics and strong voltage-dependence, so that the drug may block only pathological activation of NRs while allowing their normal physiological activity level. Therefore, memantine could restore impaired memory in the absence of the side effects commonly
observed with other NR antagonists. As expected, our results showed that prophylactic treatment with memantine improved postoperative spatial memory performance in Group IL without producing locomotor or memory disturbances. More important, memantine at the dose used in our experiment showed no analgesic effect on acute postoperative pain (Fig. 5A) without any change in the levels of hippocampal NR2 subunit (Fig. 5D). These findings indicate that the overactivation of NRs may contribute to the development of memory deficits observed in Group IL. Thus, treatment with memantine might be a promising strategy for prevention of POCD. Previous studies reported that surgical trauma-induced peripheral cytokines can travel to the central nervous system, especially the hippocampus, and may contribute to surgery-induced cognitive dysfunction (Wan et al., 2007; Cibelli et al., 2010; Terrando et al., 2010; Steinman, 2010). In contrast, our results showed that the levels of both cytokines in hippocampus on either POD 14 were comparable among the experimental groups (Fig. 3F and G). These findings indicate that cytokine contributions may not be a significant factor in memory deficits in our experimental model. Our study has the following limitations. First, we used a midline laparotomy without visceral manipulation as postoperative pain model.
992
H. Chi et al. / Life Sciences 93 (2013) 986–993
Fig. 5. Effects of prophylactic memantine treatment on acute postoperative pain, locomotor activity, and the levels of N-methyl-D-aspartate receptor 2 subunits. (A) Levels of pain were assessed using the RGS at baseline and 2, 4, 6, 8, and 12 h after the inhalation period. *p b 0.05 vs. baseline. (B) Activity counts were determined under novel conditions for 60 min in Group I and Group IL in the presence or absence of memantine as detailed in the Methods. (C) The average latency per arm choice in each experimental group. (D) The levels of N-methyl-D-aspartate receptor 2 (NR2) subunits in hippocampus were assayed by enzyme-linked immunosorbent assay for each group. Each vertical bar represents the mean ± SEM (n = 8 in each group).
The benefit of this model is that it preserves food-seeking behavior and locomotor activity, both of which are essential for subsequent cognitive task performance. Nevertheless, since this model replicates only minimally invasive surgery; it may not accurately reflect the entire clinical situation. Second, we measured the hippocampal levels of NR2 subunits, but not NR1 subunits. While NR1 distributes ubiquitously in the brain, the expression of NR2 subunits varies according to the developmental stage. Therefore, the identity of the NR2 subunits helps to determine the pharmacological and biophysical properties of NMDA receptor (Clarke and Johnson, 2006). However, we cannot rule out that changes in the expression of NR1 subunits could also contribute to our results. Third, it remains still under question how the intensity and duration of postoperative pain could impact the increase in hippocampal NR2 subunit. In addition, detailed time-course analysis of NR2 levels associated with cognitive impairment certainly needs to be performed to translate our findings into the clinical setting.
Conclusion Our results indicate that effective postoperative analgesia can prevent the development of spatial memory decline after isoflurane anesthesia and laparotomy in aged rats. In addition, postoperative memory deficits were accompanied by up-regulation of hippocampal NRs, and pharmacological intervention with memantine attenuated of spatial memory impairment in non-analgesic-treated rats. Our findings suggest that postoperative pain management may be important for preventing POCD in elderly patients.
Conflict of interest statement The authors declare that there are no conflicts of interest.
Fig. 6. Effects of prophylactic memantine treatment on working and reference memory performance in aged rats. Mean number of working (A) and reference (B) memory errors from POD 3 to 14 were analyzed in Group I and Group IL in the presence or absence of memantine as detailed in the Methods. Left panel: Each point represents the mean values for 2-day trial blocks and vertical bars represent SEM. Right panel: total errors over the entire testing period. Each vertical bar represents the mean ± SEM (n = 8 in each group). *p b 0.05 vs. vehicle group.
Acknowledgments This work was supported in part by a Grant-in-Aid for Scientific Research (C): 24592339 and Grant-in-Aid for Challenging Exploratory Research (C): 24659699 from the Japan Society for the Promotion of Science, Tokyo, Japan.
References Callaway JK, Jones NC, Royse AG, Royse CF. Sevoflurane anesthesia does not impair acquisition learning or memory in the Morris water maze in young adult and aged rats. Anesthesiology 2012;117:1091–101. Cao L, Li L, Lin D, Zuo Z. Isoflurane induces learning impairment that is mediated by interleukin 1β in rodents. PLoS One 2012;7:e51431. Chow WB, Rosenthal RA, Merkow RP, Ko CY, Esnaola NF, American College of Surgeons National Surgical Quality Improvement Program, et al. Optimal preoperative assessment of the geriatric surgical patient: a best practices guideline from the American College of Surgeons National Surgical Quality Improvement Program and the American Geriatrics Society. J Am Coll Surg 2012;215:453–66. Cibelli M, Fidalgo AR, Terrando N, Ma D, Monaco C, Feldmann M, et al. Role of interleukin-1β in postoperative cognitive dysfunction. Ann Neurol 2010;68:360–8. Clarke RJ, Johnson JW. NMDA receptor NR2 subunit dependence of the slow component of magnesium unblock. J Neurosci 2006;26:5825–34. Collingridge GL, Volianskis A, Bannister N, France G, Hanna L, Mercier M, et al. The NMDA receptor as a target for cognitive enhancement. Neuropharmacology 2013;64:13–26. Culley DJ, Baxter MG, Yukhananov R, Crosby G. Long-term impairment of acquisition of a spatial memory task following isoflurane-nitrous oxide anesthesia in rats. Anesthesiology 2004;100:309–14. Danysz W, Parsons CG. Alzheimer's disease, β-amyloid, glutamate, NMDA receptors and memantine — searching for the connections. Br J Pharmacol 2012;167:324–52. Deiner S, Silverstein JH. Postoperative delirium and cognitive dysfunction. Br J Anaesth 2009;103:i41–6. Fong HK, Sands LP, Leung JM. The role of postoperative analgesia in delirium and cognitive decline in elderly patients: a systematic review. Anesth Analg 2006;102:1255–66. Glowinski J, Iversen LL. Regional studies of catecholamines in the rat brain. I. The disposition of [3H]norepinephrine, [3H]dopamine and [3H]dopa in various regions of the brain. J Neurochem 1966;13:655–69.
H. Chi et al. / Life Sciences 93 (2013) 986–993 Ji G, Sun H, Fu Y, Li Z, Pais-Vieira M, Galhardo V, et al. Cognitive impairment in pain through amygdala-driven prefrontal cortical deactivation. J Neurosci 2010;30:5451–64. Koch HJ, Szecsey A, Haen E. NMDA-antagonism (memantine): an alternative pharmacological therapeutic principle in Alzheimer's and vascular dementia. Curr Pharm Des 2004;10:253–9. Komatsu H, Nogaya J, Anabuki D, Yokono S, Kinoshita H, Shirakawa Y, et al. Memory facilitation by posttraining exposure to halothane, enflurane, and isoflurane in ddN mice. Anesth Analg 1993;76:609–12. Lin D, Zuo Z. Isoflurane induces hippocampal cell injury and cognitive impairments in adult rats. Neuropharmacology 2011;61:1354–9. Lister JP, Barnes CA. Neurobiological changes in the hippocampus during normative aging. Arch Neurol 2009;66:829–33. Lukoyanov NV, Paula-Barbosa MM. Memantine, but not dizocilpine, ameliorates cognitive deficits in adult rats withdrawn from chronic ingestion of alcohol. Neurosci Lett 2001;309:45–8. Monk TG, Weldon BC, Garvan CW, Dede DE, van der Aa MT, Heilman KM, et al. Predictors of cognitive dysfunction after major noncardiac surgery. Anesthesiology 2008;108:18–30. Moriarty O, McGuire BE, Finn DP. The effect of pain on cognitive function: a review of clinical and preclinical research. Prog Neurobiol 2011;93:385–404. Morrison RS, Magaziner J, Gilbert M, Koval KJ, McLaughlin MA, Orosz G, et al. Relationship between pain and opioid analgesics on the development of delirium following hip fracture. J Gerontol A Biol Sci Med Sci 2003;58:76–81. Rammes G, Danysz W, Parsons CG. Pharmacodynamics of memantine: an update. Curr Neuropharmacol 2008;6:55–78. Rammes G, Starker LK, Haseneder R, Berkmann J, Plack A, Zieglgänsberger W, et al. Isoflurane anaesthesia reversibly improves cognitive function and long-term
993
potentiation (LTP) via an up-regulation in NMDA receptor 2B subunit expression. Neuropharmacology 2009;56:626–36. Saczynski JS, Marcantonio ER, Quach L, Fong TG, Gross A, Inouye SK, et al. Cognitive trajectories after postoperative delirium. N Engl J Med 2012;367:30–9. Sotocinal SG, Sorge RE, Zaloum A, Tuttle AH, Martin LJ, Wieskopf JS, et al. The Rat Grimace Scale: a partially automated method for quantifying pain in the laboratory rat via facial expressions. Mol Pain 2011;7:55. Steinman L. Modulation of postoperative cognitive decline via blockade of inflammatory cytokines outside the brain. Proc Natl Acad Sci U S A 2010;107:20595–6. Steinmetz J, Christensen KB, Lund T, Lohse N, Rasmussen LS, ISPOCD Group. Long-term consequences of postoperative cognitive dysfunction. Anesthesiology 2009;110:548–55. Terrando N, Monaco C, Ma D, Foxwell BM, Feldmann M, Maze M. Tumor necrosis factor-alpha triggers a cytokine cascade yielding postoperative cognitive decline. Proc Natl Acad Sci U S A 2010;107:20518–22. Vaurio LE, Sands LP, Wang Y, Mullen EA, Leung JM. Postoperative delirium: the importance of pain and pain management. Anesth Analg 2006;102:1267–73. Voscopoulos C, Lema M. When does acute pain become chronic? Br J Anaesth 2010;105: i69–85. Wan Y, Xu J, Ma D, Zeng Y, Cibelli M, Maze M. Postoperative impairment of cognitive function in rats: a possible role for cytokine-mediated inflammation in the hippocampus. Anesthesiology 2007;106:436–43. Zajaczkowski W, Frankiewicz T, Parsons CG, Danysz W. Uncompetitive NMDA receptor antagonists attenuate NMDA-induced impairment of passive avoidance learning and LTP. Neuropharmacology 1997;36:961–71. Zhuo M. Plasticity of NMDA receptor NR2B subunit in memory and chronic pain. Mol Brain 2009;2:4.