• Open Access

Experimental insights into age-exacerbated cognitive dysfunction after peripheral surgery



Antonio R. Fidalgo, Research Center for Advanced Science and Technology, The University of Tokyo, 4-6-1, Komaba, Meguro-ku, Tokyo 153-8904, Japan. Tel.: +81 3 5452 5247; fax: +81 3 5452 5248; e-mail: afidalgo@outlook.com


Here I comment on the recent contribution by Barrientos et al. J. Neurosci. 32, 14641–14648 (2012) addressing treatment possibilities for surgery-induced cognitive dysfunction. It has been over 15 years since the publication of a landmark study that indicated age as a major risk factor for postoperative cognitive dysfunction (POCD) (Moller et al., Lancet 351, 857–861 1998). With increasing life expectancy, surgical procedures conducted in elderly persons are becoming more common. The prevalence of POCD may mean that some patients will exchange the incapacitating condition that led them to surgery in the first instance for another such condition, which has been created by the surgical procedure itself. The report by Barrientos and collaborators (2012) is a timely and welcome study that further examines treatment possibilities for surgery-induced cognitive dysfunction. Future studies should address issues such as intensity and onset of inflammation within the brain and additional treatments possibilities beyond IL-1-ra.

Studies designed to improve our understanding of the underlying pathogenic processes in surgery-induced cognitive dysfunction and possible therapeutic interventions are increasingly relevant. Barrientos et al. (2012) reported on how hippocampal-dependent contextual freezing in a fear-conditioning task is significantly impaired up to 4 days after a laparotomy (a surgery designed to mimic abdominal surgery in adult humans) in aged but not young rats. The hippocampal-dependent memory impairment was associated with interleukin-1β (IL-1β) increases in the hippocampal structure of aged rats, once again young rats did not evidence any exacerbated response to surgery (Barrientos et al., 2012). The authors also demonstrated that a single intracisternal administration of interleukin-1 receptor antagonist (IL-1-ra) is sufficient to stop the behavioural symptoms and cognitive deficits resulting from the elevation of hippocampal IL-1β. Whilst this elegantly designed set of experiments highlight the specific role of central-acting IL-1β in POCD, one should question the reasons underlying the contradictory evidence in the occurrence of surgery-induced cognitive dysfunction in young versus old animals, as well as the prominent role of the hippocampus and the relevance of the use of IL-1-ra for therapeutic purposes.

It has been demonstrated that IL-1β is elevated in the serum and hippocampus after peripheral orthopaedic surgery in young adult mice (Cibelli et al., 2010; Fidalgo et al., 2011a). In those reports, the hippocampus was also defined as the key area affected by IL-1β elevation and microgliosis (Cibelli et al., 2010). The cytokine elevation in laparotomy, however, appears to be remarkably different (Barrientos et al., 2012). In peripheral orthopaedic surgery, the cytokine elevation appears to be much more sudden with IL-1β levels in the hippocampus peaking at around 6 h postsurgery and returning to basal levels within 24 h of the surgical procedure (Cibelli et al., 2010). In laparotomies, the cytokine elevation is delayed, with IL-1β levels in the hippocampus peaking at 24 h postsurgery (Barrientos et al., 2012). It has been demonstrated that the presence of high levels of IL-1β in the dorsal hippocampus within the initial 24 h postacquisition of a fear memory results in retrograde amnesia (Barrientos et al., 2002). The delayed elevation of hippocampal cytokines in the Barrientos et al. (2012) laparotomy model may partially explain why the authors do not report any surgery-induced cognitive dysfunction in young rats.

Barrientos et al. (2012) also did not report any amelioration of surgery-induced cognitive deficits from peripherally injected IL-1-ra. This is contrary to other reports which indicate that peripherally injected IL-1-ra ameliorates surgery-induced cognitive dysfunction (Cibelli et al., 2010). This difference may be explained by pharmacokinetic studies in rats that demonstrate the maintenance of IL-1-ra concentrations in plasma and cerebral spinal fluid after intravenous infusion for 24 h (Clark et al., 2008). Brain penetration of peripherally administered IL-1-ra in those cases is limited due to the blood–brain barrier, the relatively large size (17 kDa) of the cytokine and its hydrophilic nature (Cawthorne et al., 2011).

The salient role of the hippocampus in studies of surgery-induced cognitive dysfunction could be due to the fact that fear conditioning and water maze are the favoured behavioural tests (Rosczyk et al., 2008; Cibelli et al., 2010; Fidalgo et al., 2011a,b; Barrientos et al., 2012). Previous reports using radial-arm maze – a test known for its selective roles for hippocampal, prefrontal cortical and ventral striatal circuits (Floresco et al., 1997) – have demonstrated that anaesthesia alone can result in cognitive impairment in aged rats (Culley et al., 2003). A small variation on contextual fear-conditioning tasks can allow one to assess remote neocortical-dependent fear memories. This type of paradigm has detected behavioural differences when a control group is compared with anaesthesia, lipopolysaccharide or surgery groups (Fidalgo et al., 2011b, 2012). However, in hippocampal-dependent fear-conditioning tasks, administration of short anaesthesia regimens in young animals has not shown to cause cognitive dysfunction (Cibelli et al., 2010).

Cytokines have been described to have specific roles in hippocampal long-term potentiation (Jankowsky & Patterson, 1999; Yirmiya & Goshen, 2011), and while IL-1-β, in particular, is necessary for normal physiological regulation of hippocampal-dependent memory (Goshen et al., 2007), abnormally high levels result in hippocampal memory impairment (Barrientos et al., 2002; Goshen et al., 2007; Spulber et al., 2009). In a similar manner, transgenic over-expression of IL-1-ra in mice resulted in impaired hippocampal-dependent memory in water maze, fear-conditioning and T-maze tasks (Goshen et al., 2007; Spulber et al., 2009). More interestingly, administration of IL-1-ra during neonatal development results in long-term memory impairment even when mice reach adulthood (Goshen et al., 2007). In summary, data from animal studies indicate that IL-1-β and IL-1-ra play an important role in memory storage and formation, an element that should be taken into careful consideration when exploring new therapeutic treatments.

There are, however, at least three promising alternatives that should be explored (see Table 1 for an overview of some of the possible therapies). Inflammation within the brain has been shown to reduce the brain-derived neurotropic factor (BDNF) (Fidalgo et al., 2011b). BDNF plays an important role in synaptic plasticity and memory (Jankowsky & Patterson, 1999) and its use in therapy has lead to promising results in the treatment of both neurological and psychiatric disorders (Nagahara & Tuszynski, 2011). TNF inhibitors have also been demonstrated to restore neuronal function and reverse cognitive deficits induced by chronic inflammation within the brain (Belarbi et al., 2012). Cyclooxygenase (COX) inhibition has been reported to block the disruption of hippocampal-dependent contextual fear-conditioning freezing after pathophysiological levels of IL-1-β (Hein et al., 2007).

Table 1. Comparison table of molecules that are known to have a detrimental effect on cognitive function and possible treatments. Currently, only IL-1β, TNF-α and prostaglandins have been targeted in the scrutiny of inflammation-induced memory deficits. BDNF has yet to be applied to the study of surgery-induced cognitive decline
MoleculeDetrimental effectPossible treatment
IL-1βReduced hippocampal-dependent freezing response and decreased performance in T-mazeIL-1-ra
TNF-αImpairment in novel place recognition task and spatial learning/memory in water maze testsTNF-α protein synthesis inhibitor
ProstaglandinsReduced hippocampal-dependent freezing responseCyclooxygenase (COX) inhibitors
Amyloid precursor proteinReduced hippocampal-dependent freezing responseBDNF

In summary, Barrientos et al. (2012) report is a timely and welcome study that further examines treatment possibilities for surgery-induced cognitive dysfunction. Future studies should address issues such as intensity and onset of inflammation within the brain and additional treatments possibilities beyond IL-1-ra.


ARF research has been funded by a FCT – Portugal predoctoral fellowship and a JSPS postdoctoral fellowship.