Paralysis currently affects 1 in 50 Americans, 16% of which are completely unable to move any part of their body (Spinal Cord Injury Zone, 2009). In addition, limb loss affects 1 in 190 Americans (Ziegler-Graham et al., 2008). In both populations, the loss of motor function imposes a significant lifelong burden. Implantable microelectrode recording arrays are a type of implantable biomedical device being developed to restore function for such patients. Penetrating microelectrode arrays implanted in the human cortex, along with additional computational equipment, have provided paralyzed patients with intentional control over computer cursors (Hochberg et al., 2006; Simeral et al., 2011) and robotic devices (Hochberg et al., 2012; Collinger et al., 2013). More recently, studies have shown that it is possible for individuals with traumatic high cervical spinal cord injury (SCI) to perform coordinated reaching and grasping movements using their own paralyzed arm and hand muscles, which are reanimated through functional electrical stimulation and controlled using volitional cortical signals through a chronically implanted intracortical recording array (Ajiboye et al., 2017).

The potential group of patients with paralysis resulting from traumatic injury or stroke that may be treated with such devices includes a significant number of older adults, 56% of which are over the age of 50, and 33.5% of which are over the age of 60 (Spinal Cord Injury Zone, 2009). In addition, 80% of individuals with limb loss are over 45, while 42% are over 65 (Ziegler-Graham et al., 2008). Moreover, most young patients will eventually reach at least middle age. For instance, in a case of high tetraplegia due to spinal cord injury in a 20-year-old patient, the life expectancy may reach 57 years (Spinal Cord Injury Zone, 2009).

During aging, the immune system undergoes changes including: a reduced number of naïve B and T cells, a decrease in stimulated phagocytosis and reactive oxygen species (ROS) production by neutrophils, and dysregulation of immune cell signaling pathways (Weiskopf et al., 2009; Ponnappan and Ponnappan, 2011). These changes result in prolonged wound healing (Gosain and DiPietro, 2004) and increased levels of proinflammatory cytokine production in cases of neuroinflammation (Weiskopf et al., 2009; Ponnappan and Ponnappan, 2011).

The brain also changes with age. Microglia, the brain’s resident macrophage, acquire a more reactive phenotype (von Bernhardi et al., 2010), become larger and less ramified, express increased markers of activation, and produce higher levels of proinflammatory cytokines such as tumor necrosis factor alpha (TNF-alpha), interleukin 1-beta (IL-1β) and interleukin 6 (IL-6) (Perry et al., 1993; Ye and Johnson, 1999; Sierra et al., 2007; Hart et al., 2012; Ritzel et al., 2015). In addition, astrocytes become more numerous and hypertrophic (Amenta et al., 1998). Neuronal loss and a reduction in the number of dendrites and synapses also occurs (Juraska and Lowry, 2012), rendering the aged mammalian brain more susceptible to such neurodegenerative diseases as Alzheimer’s (Lindsay et al., 2002) and Parkinson’s (Van Den Eeden et al., 2003). Therefore, the brain of older patients may present a uniquely challenging environment for the successful implementation of neuroprosthetic control that uses penetrating multishaft, microelectrode arrays chronically implanted into the surface of the brain. The multiple injuries associated with the implantation of a penetrating multishaft microelectrode array and the accompanying neuroinflammation associated with the chronic foreign body response (FBR), may play a more significant role in the aged mammalian brain and thus reduce chronic recording performance compared to such implants in younger individuals.

Clinical studies indicate that older patients have poorer clinical outcomes following ischemic stroke (Macciocchi et al., 1998), traumatic brain injury (Hukkelhoven et al., 2003), and aneurysmal subarachnoid hemorrhage (Lanzino et al., 1996). Similarly, older rats show increased loss of neural tissue following experimental injuries (Rosen et al., 2005; Kumar et al., 2013). A recent study using a single planar penetrating array in the rat cortex reported a significantly reduced FBR and improved recording performance over a six-week period in 4-week-old rats compared to a slightly older group (9 weeks at implantation) (Sharon et al., 2023). Moreover, clinical studies using deep brain stimulation (DBS) with devices implanted in older patients show increased rates of complications (Voges et al., 2007), decreased clinical benefit (Saint-Cyr et al., 2000; Charles et al., 2002; Welter et al., 2002), and increased rates of cognitive and behavioral impairment (Saint-Cyr et al., 2000).

To the best of our knowledge, no studies have examined whether penetrating, multishaft chronic recording arrays can function chronically after implantation in the brain of older animals as a model for their use in older patients. A survey of the literature indicates that most of the studies using rats to study recording performance of central nervous system penetrating recording microelectrodes have used males that are in the early part of adulthood (Figure 1), while a significant number of patients enrolled in clinical BCI research studies have been much older, ranging in age from 52 to 66 (Hochberg et al., 2006; Simeral et al., 2011; Hochberg et al., 2012; Collinger et al., 2013; Ajiboye et al., 2017). To address this issue, we implanted a 4×4 microelectrode recording array into the 18-month old rat motor cortex, analyzed the recording performance in the unanesthetized condition over a 3-month indwelling period, compared the recording performance to a younger cohort that received the same implant over the same indwelling period, and assessed the foreign body response (FBR) at the study endpoint using an immunohistochemical approach.

In this study, we implanted 18-month-old male rats with a 4×4 UEA into the motor cortex and then examined single-unit recording performance in the implanted, freely moving animals at weekly intervals over a 12-week indwelling period. We observed single-unit action potentials over the entire indwelling period. Single-unit recording performance was most robust in the month following implantation, and then gradually declined over the subsequent 2-month period. These findings are similar to what has been reported using the same recording array in younger adult rats (Nolta et al., 2015; Black et al., 2018; Cody et al., 2018). More importantly, when we compared the single-unit recording yield of the older cohort in this study to a younger male cohort (3 months old) of the same rat strain over a 12-week period, the recording performance yield was as good as, and in some weeks significantly better than, the younger cohort (Table 2).

To the best of our knowledge, this is the first study that has examined chronic recording performance and the FBR in older rats using a penetrating recording array of any type. Based on a 25-month median lifespan for male Sprague Dawley rats (Cameron et al., 1982) and a 79-year median lifespan for American males, an approximate equivalent human age for the rats used in this study was approximately 53 years at implantation and 62 years at the study endpoint (Sengupta, 2013). This age group is similar to the patient population that has been studied in clinical trials that have evaluated the UEA as a treatment of paralysis (Hochberg et al., 2006; Simeral et al., 2011; Hochberg et al., 2012; Collinger et al., 2013; Ajiboye et al., 2017).

The chronic recording performance and aspects of the FBR observed here were somewhat unexpected given the body of literature that suggests that the effects of aging might have had a negative impact on recording performance given the older rat’s increased sensitivity to ischemic injury and neuroinflammation (Lanzino et al., 1996; Macciocchi et al., 1998; Saint-Cyr et al., 2000; Charles et al., 2002; Welter et al., 2002; Hukkelhoven et al., 2003; Rosen et al., 2005; Voges et al., 2007; Kumar et al., 2013). We found that the FBR of the older rats at the study endpoint was similar to that previously described using the same array implanted in younger rat motor cortex (Nolta et al., 2015; Black et al., 2018). Moreover, the FBR around individual microelectrode shafts near the recording tips showed a stereotypical response, as previously described in studies performed in younger rats, irrespective of the indwelling period or implanted microelectrode type (Collias and Manuelidis, 1957; Biran et al., 2005, 2007; Rennaker et al., 2007; McConnell et al., 2009; Winslow et al., 2010; Winslow and Tresco, 2010; Skousen et al., 2011; Potter et al., 2012; Prasad et al., 2014).

With regard to FBR biomarkers, we observed immunoreactivity for IgG and CD68 surrounding the UEA after a 3-month indwelling period, which also was reported using younger animals (Nolta et al., 2015; Black et al., 2018). In this study, neither biomarker correlated with recording success at the study endpoint and did not appear significant near the recording tips. Several groups have reported that the distribution of intraparenchymal IgG due to BBB leakiness decreases with time after implantation for single-shank planar silicon microelectrode arrays in the cortex of young rats (Potter et al., 2012) and in mice (Ravikumar et al., 2014). Several studies using younger cohorts of rats have reported that CD68 immunoreactivity is also reduced at longer indwelling periods compared to shorter periods (McConnell et al., 2009; Potter et al., 2012; Prasad et al., 2014; Ravikumar et al., 2014).

We observed a significant amount of connective tissue underneath the base and on the upper parts of the microelectrode shafts of the retrieved arrays. This observation also was reported in cortically implanted UEAs in younger rats (Nolta et al., 2015; Black et al., 2018; Cody et al., 2018) and in UEAs explanted from non-human primates after long indwelling periods (Barrese et al., 2013). Connective tissue was present on all explanted arrays in the older cohort. The connective tissue was NeuN and neurofilament negative and contained little GFAP, indicating that it was non-neural tissue and likely of meningeal origin. The amount of connective tissue was highly variable between animals and appeared to correlate closely with the volume of the lesion cavity. We speculate that brain tissue lost due to vascular and tissue damage that accompany device implantation caused by the multiple, closely spaced penetrating shafts is eventually filled in with non-neural, collagenous tissue. This interpretation agrees with several studies that showed that a stab wound injury performed with the 4×4 UEA resulted in a stroke-like lesion cavity in younger rats, and focal hemorrhage near and below UEAs implanted acutely and then removed in the cortex of human patients (House et al., 2006; Waziri et al., 2009; Nolta et al., 2015). In a recent study, key genes that mediate the acute injury and neuroinflammatory response, along with genes critical to the function of the BBB, were not significantly different between a UEA implanted group and a UEA stab wound group using the same implantation approach (Bennett et al., 2018). That study showed that the effect of UEA insertion-related trauma dominates early wound healing surrounding the implant.

We observed that single-unit recording performance in the older rats was inversely related to the lesion cavity size. Lesion cavities, similar to those observed here, have also been produced in laser induced experimental stroke models that ablate a single descending arteriole or ascending venule (Shih et al., 2013), as well as in both young implanted and stab-wounded rats using a 4×4 UEA (Nolta et al., 2015; Black et al., 2018). Together, these observations suggest that neural tissue loss observed after UEA implantation in rats is likely related to the amount of vascular damage caused by device implantation, and not necessarily the result of the persistent inflammation that accompanies the chronic phases of the FBR (Nolta et al., 2015; Bennett et al., 2018; Black et al., 2018; Cody et al., 2018). In general, the lesion cavity in the older cohort examined here was similar to that reported previously using younger rats of the same strain using the same array and implantation approach (Nolta et al., 2015).

Lesion or stroke-like cavities have only been reported in rats in a subset of studies that used multi-shank microelectrode arrays in younger rats (Williams et al., 2007; Ward et al., 2009; Saxena et al., 2013; Black et al., 2018; Sharon et al., 2023), and not in those using simple planar single-shank silicon microelectrode arrays or with a few planar microelectrode shafts. Williams et al. found significantly altered impedance spectra for microwires associated with lesions, but did record single units (Williams et al., 2007). Several studies have shown lesion cavities in their published work but did not specifically describe their occurrence (Ward et al., 2009; Saxena et al., 2013; Black et al., 2018). Two studies by Prasad et al. did not analyze neural tissue loss, but did report increased injury during implantation, as evidenced by bleeding that was associated with reduced recording performance (Prasad et al., 2012, 2014). Given the high vasculature density of the rat cortex, vasculature damage resulting from insertion of a UEA with its closely spaced multiple silicon microelectrode shafts seems inevitable, irrespective of the age of the rat used.

Unlike CD68 and IgG immunoreactivity, we observed that single-unit recording performance was inversely related to the level of GFAP immunoreactivity near microelectrode recording tips. We found that microelectrodes that did not record any units in the final week of recordings had significantly higher levels of GFAP immunoreactivity within a 100 μm radius of the recording tip (the presumptive recording zone). These findings agree with a previous report that used an implanted 4×4 UEA in younger rats in the same target tissue, in which higher levels of GFAP corresponded with reduced SNR for individual microelectrodes within the same array (Nolta et al., 2015). This relationship may be causal if hypertrophic astrocytes displace neuronal soma from the recording zone, or if GFAP immunoreactivity is a good indicator of persistent neuroinflammatory stimuli which may silence neuronal activity through any number of other mechanisms including demyelination.

Our results suggest that functional issues related to anchorage of the array may be underappreciated. Fibrotic tissue buildup has been proposed to cause movement of multi-shaft recording arrays following implantation injury for free floating arrays in nonhuman primates (Barrese et al., 2013) and cats (Rousche and Normann, 1998; Maynard et al., 2000; McCreery et al., 2010). In a retrospective analysis of numerous experiments with UEAs chronically implanted in macaques, the authors determined that 53% of all slowly-progressing recording failures were due to fibrotic tissue buildup that dramatically changed the orientation of the arrays from their original implantation position (Barrese et al., 2013). We observed very little movement of the arrays in the aged rat cohort examined in this study, which may have been due to their thicker skulls compared to younger rats. At the time of array retrieval, the skull bone at the craniotomy site was consistently measured to be over a millimeter in thickness. While we can only speculate, it appeared that the edges of the craniotomy surrounding the array regenerated somewhat during the 12-week indwelling period and where very close to the base of the UEA at explant, which may have helped prevent headstage movement.

These findings have implications for studies in larger animals and humans. For one, they support the idea that reducing the impact of vascular damage during implantation and the magnitude of the persistent FBR during the indwelling period will likely improve chronic single-unit recording performance. Previous studies suggest that this may be accomplished through a number of device design changes including reducing device surface area (Seymour and Kipke, 2007; Skousen et al., 2011), increasing device permeability (Skousen et al., 2014), reducing device size (Kozai et al., 2012), reducing device stiffness (Harris et al., 2011), increasing spacing between microelectrode shafts (McConnell et al., 2007, 2009), administering antiflammatory drugs locally (Zhong and Bellamkonda, 2007) or systemically (Rennaker et al., 2007; Potter-Baker et al., 2015), and minimizing vascular damage during insertion (Kozai et al., 2010). It is likely that a combination of such approaches will be needed to improve chronic recording performance and improve device biocompatibility. The results of this study are also encouraging for the future of microelectrode–based therapeutic development for paralysis across older patient populations.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.

The animal study was approved by the University of Utah Animal Care and Use Committee. The study was conducted in accordance with the local legislation and institutional requirements.

NN: Investigation, Visualization, Writing – original draft, Writing – review & editing. MC: Formal Analysis, Investigation, Visualization, Writing – original draft, Writing – review & editing. PT: Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Visualization, Writing – original draft, Writing – review & editing.