Spinal cord injury (SCI) is an increasingly important global health factor (GBD 2016 Traumatic Brain Injury and Spinal Cord Injury Collaborators, 2019). After SCI, neurogenic bowel dysfunction (NBD) is an almost inevitable condition; its main symptoms are faecal incontinence (FI) and constipation (Glickman and Kamm, 1996; Awad, 2011). According to reports, the prevalence of constipation in patients with SCI is between 56 and 80%, and FI occurs in 42–75% of patients (Tate et al., 2016; Qi et al., 2018). Approximately 78% of interviewees pointed out that bowel dysfunction was the main cause of quality of life impairment because bowel care (a large amount of time spent defaecating, compulsory drug use, digital stimulation for regular bowel movements, risk of faecal incontinence, persistent personal assistance needs, and other factors) interfered with their social life and personal relationships, preventing them from working away from home (Glickman and Kamm, 1996; Inskip et al., 2018). Improving intestinal function has been identified as one of the highest priorities for recovery in patients with SCI (including those with quadriplegia and paraplegia) (Anderson, 2004; Simpson et al., 2012; Lo et al., 2016). Generally, neurogenic bowel and subsequent symptoms in patients with motor-complete SCI were considered more severe than those in patients with motor-incomplete SCI (Vallès and Mearin, 2009). Research by Liu et al. showed that the risk of severe NBD in patients with American Spinal Cord Injury Association (ASIA) A was 12.8 times higher than that in patients with ASIA D (OR = 12.8, 95% CI 3.3–50.1) (Liu et al., 2010). These results are consistent with the finding that people with more severe injuries tend to have more serious bowel dysfunction (Vallès et al., 2006).

Individuals with SCI are susceptible to an imbalance in the intestinal flora. Using stool samples from mice with traumatic SCI, Kigerl et al. showed that traumatic SCI could cause intestinal diseases and that intestinal dysbiosis can disrupt functional recovery (Kigerl et al., 2016). According to reports, after SCI, the abundance of Actinobacteria, Proteobacteria, Verrucomicrobia, Bacteroides, Blautia, Escherichia-Shigella, Akkermansia, Alistipes, Parabacteroides, etc., increased significantly. In contrast, that of Firmicutes, Faecalibacterium, Megamonas, Prevotella 9, Dialister, Roseburia, etc., were significantly reduced (Gungor et al., 2016; Zhang et al., 2018; Bazzocchi et al., 2021; Du et al., 2021; Yu et al., 2021). Changes in the abundance of Bifidobacterium, Lactobacillus, Bacteroides, Roseburia, and other types of intestinal flora are related to constipation in patients with SCI (Zhu et al., 2014; Kim et al., 2015; Mancabelli et al., 2017; Zhang et al., 2021; Wang et al., 2023). Intestinal flora, such as Bifidobacterium, Lactobacillus plantarum P9, and Latilactobacillus sakei, have been shown to relieve constipation symptoms (Guo et al., 2022; Ma et al., 2023; Takeda et al., 2023). Studies have indicated an interactive relationship between constipation and intestinal flora. Constipation can lead to intestinal flora disorders, while intestinal flora dysfunction can also aggravate constipation symptoms, mainly characterised by increased pathogenic bacteria in the body and decreased dominant flora (Dimidi et al., 2017; Wortelboer et al., 2019; Fu et al., 2021). These results indicate that the microbiome may be a potential therapeutic target for constipation after SCI.

Physical activity, especially walking, as is universally acknowledged, can stimulate intestinal peristalsis in the able-bodied population (Peters et al., 2001; Bi and Triadafilopoulos, 2003; De Schryver et al., 2005; Hubscher et al., 2018). Exercise may modulate intestinal permeability, motility, stool transit time, and consistency (Sohail et al., 2019). Improvement in bowel function is more likely related to activity than an upright posture (Kwok et al., 2015). Studies have found that exercise interventions tend to increase the abundance of beneficial bacterial genera, such as Blautia, Dialister, and Roseburia (Quiroga et al., 2020). The exercise training of the exoskeleton system has been applied to rehabilitate paraplegic patients in recent years. The beneficial effects of exercise on improving intestinal function and intestinal flora have been widely confirmed in clinical medicine and exercise science (Allen et al., 2018; Gao et al., 2019; Resende et al., 2021).

Some studies have reported the effect of exoskeleton robots on patients with SCI, including ReWalk (ReWalk Robotics, Marlborough, MA, United States), Ekso (Ekso Bionics, Richmond, CA, United States), HAL (Cyberdyne, Tsukuba, Japan), and H-MEX (Hyundai Motor Company, Uiwang, Korea) Equipment (Gorgey et al., 2017; Baunsgaard et al., 2018; Chun et al., 2020; Gorman et al., 2021; Kim et al., 2021; Brinkemper et al., 2023). However, few studies have evaluated the relationship between microflora and the efficacy of EAW in treating constipation in individuals with SCI. In recent years, some scholars have suggested that the interactions between the microbiome, intestinal connectome, and brain connectome are established in the gut itself, meaning that if the input to the spinal cord or vagus nerve is cut off, the different components of ENS, mucosal immunity, and microbiome will still function (though not optimally; Macpherson et al., 2023). So is it possible that EAW training to treat constipation in SCI patients is related to intestinal flora? At the moment, there are no such reports. Therefore, in this study, we evaluated the role of EAW training in bowel function in patients with motor-complete SCI. The first to observe intestinal flora changes after EAW treatment of SCI, evaluate its correlation with intestinal function.

SCI is a sudden and catastrophic event. The severity of the injury and functional recovery vary among individuals. The most important determinant of prognosis is whether the injury is complete or incomplete. Up to 25% of complete injuries are incomplete within the first year after injury. However, functional motor recovery (e.g., weight-bearing and walking) at the distal end of the injury area after complete motor SCI is rare (Angeli et al., 2018; Fouad et al., 2021). This is because the motor center is difficult to activate in patients with complete motor spinal cord injury and may exhibit a more pronounced activation effect when using epidural stimulation, but this technology is still in the promotion stage, and the combined effect of physical rehabilitation needs to be further improved (Côté et al., 2017; Gorgey et al., 2021). AIS A and B are classified as motor-complete injuries with more serious dysfunction and life restrictions. SCI changes intestinal peristalsis and sphincter control, and the loss of mobility and movement flexibility makes intestinal dysfunction a major life-limiting factor (Lynch and Frizelle, 2006). Powered robot exoskeletons have low energy consumption, intelligence, efficiency, and power, enabling patients with complete motor SCI to achieve therapeutic walking and increase the possibility of functional recovery.

After the 8-week intervention, five participants (62.5%) in the EAW group described improvement in at least one or more items of bowel function management, compared to only three participants (37.5%) in the conventional group. At the same time, the number of glycerol enemas in EAW group decreased significantly after the intervention (p < 0.05). In addition, as our results from NBD scores indicate, the EAW group exhibited a downward trend in the NBD score, while the conventional group showed an overall upward trend. A lower NBD score indicated better bowel function. Although there was no statistically significant improvement in intestinal function before and after the intervention in either group, patients in the EAW group were more likely to benefit from the training. Our results suggest EAW improved bowel function in patients with SCI to a certain extent, but it did not have the desired effect. Consistent with previous studies, among spinal cord injury patients who underwent 25–63 ReWalk sessions, more than 50% of participants experienced a shortened time to defaecation, a decrease in bowel accidents, and a reduction in the frequency of laxative, stool softening use, or both, and overall bowel satisfaction improved (Chun et al., 2020; Brinkemper et al., 2023). Gorman et al. reported in a later large-scale randomised controlled crossover study that EAW training had a positive impact on approximately one-quarter of the participants in terms of bowel function and management, but there was no statistical difference in the frequency and duration of bowel movements, and the magnitude of the impact was not as significant as initially assumed (Gorman et al., 2021). However, contrary results were obtained by Baunsgaard et al., who found that intestinal function did not improve in patients with chronic spinal cord injury, but improved in patients with recent injuries after 24 sessions of robotic exoskeleton gait training, which may be related to different study design, geography, etc. (Baunsgaard et al., 2018). Notably, the patients involved in this study were injured for a short period. The reported intestinal function improvement effect of the EAW in this study may be partly attributed to the early self-recovery of patients with newer SCI. The SCI subjects included in this study were all motor-complete injury, with more serious intestinal function involvement and slower functional recovery, which may be related to the inobvious improvement. In addition, due to the small sample size, most of the functional changes were not statistically significant, and the sample size will be increased in the future and more rigorous protocols will be adopted to enhance the reliability of the study.

After SCI, the influence of the spinal cord on autonomic nervous system (ANS) is destroyed, which leads to sympathetic passivation and parasympathetic innervation, resulting in multiple organ disorders (Henke et al., 2022). The sympathetic nervous system (SNS) originates from the T1-L2 spinal segment, sympathetic preganglionic neurons (SPN) from T9-T12 innervate the ascending colon, large intestine, and small intestine, and SPN from T12-L2 innervate the descending colon and rectum (Browning and Travagli, 2014; Wulf and Tom, 2023). Thus, both lower thoracic and lumbar spinal cord injuries involve the sympathetic nervous system, leading to neurogenic intestinal dysfunction characterized by constipation, abnormal excretory reflexes, and decreased colonic contractions. Only one subject with L1 injury was included in this trial, and his bowel function was not better than that of all other patients with thoracic injury. In the future, patients with spinal cord injury at different segments will be included for stratified analysis of intestinal function.

The tens of trillion microbes in the intestine regulate the metabolism of the host, and physical exercise seems to increase the abundance of some healthy microbes (Barton et al., 2018; Lensu and Pekkala, 2021; Moitinho-Silva et al., 2021). In our study, EAW altered the gut microbiota of SCI patients, with increased abundance of Dialister (p < 0.05), Faecalibacterium, Alistipes and Bifidobacterium. Previous studies have reported that Faecalibacterium are correlated with faster colonic transit (Parthasarathy et al., 2016), and Dialister is reduced in SCI patients, which may aggravate the symptoms of NBD (Zhang et al., 2018). Dialister and Faecalibacterium belong to Firmicutes. In the correlation analysis, we found that the abundance of Firmicutes negatively correlated with external assistance. Faecalibacterium, Dialister and Bifidobacterium, is a producer of butyrate, a short-chain fatty acid. Previous studies have confirmed that butyrate induces the release of serotonin, thereby activating the maturation of the enteric nervous system, restoring colon movement, and relieving constipation (Reigstad et al., 2015; Yano et al., 2015; Ge et al., 2017). Although the specific mechanisms by which the gut microbiota regulates 5-HT synthesis have not been explored, gut microbiota-derived metabolites, such as tryptophan, secondary bile acids, and short-chain fatty acids (SCFAs), are all involved in triggering 5-HT biosynthesis in EC cells (Sun et al., 2018; Rosario et al., 2021; Bai et al., 2022). In addition, it has been found that exercise can increase the concentration of butyricogenes in mice and humans to enrich the concentration of SCFAs. SCFAs can enhance fecal water content and colonic contractility, and reduced colon transit time (Wang et al., 2017; Zhao et al., 2018). It has been reported that Alistipes is negatively correlated with defecation time (Zhang et al., 2018). No such relationship was found in this study, but Alistipes increased slightly and defecation time decreased slightly after EAW training. In addition, Alistipes is a producer of propionate and acetate (El-Salhy, 2023), Butyrate has been proved to induce the expression of AHR and CYP1A1, while propionate can only induce the expression of AHR (Korecka et al., 2016). Some studies have shown that AHR expressed by intestinal cells can enhance intestinal motility function and form an automatic regulatory feedback loop with CYP1A1, which can metabolize AHR ligands and terminate AHR signal transduction, while the expression of CYP1A1 is dependent on AHR (Korecka et al., 2016; Schiering et al., 2017; Obata et al., 2020). Increasing the concentration of butyrate and propionate can activate the expression of AHR and CYP1A1, which can increase intestinal peristalsis and improve constipation. In short, the increase of beneficial bacteria such as Faecalibacterium, Dialister, Bifidobacterium and Alistipes may be one of the potential reasons for EAW to alleviate intestinal dysfunction. We will also confirm the role of intestinal flora and its metabolites SCFAs in future studies.

It is worth noting that we found that the level of Blautia in the two groups has declined, and it is positively related to the defecation time. This is inconsistent with the increase in 12-week exercise training reported by Quiroga et al. (2020). In addition, Megamonas has been reported to have beneficial properties as its reduction exacerbates NBD symptoms (Zhang et al., 2018), whereas our study found that Megamonas showed a downward trend after EAW intervention. One reason for these inconsistencies may be that Blautia and Megamonas are not particularly sensitive to athletic training. It seems to take a specific intensity or amount of exercise to improve.

The abundance of Proteobacteria is significantly decreased in patients with constipation (Guo et al., 2020). We found that the relative abundance of Proteobacteria increased in the EAW group but decreased in the conventional group. However, at baseline, Proteobacteria abundance in the EAW group was lower than that in healthy men, while that in the conventional group was higher. Munukka et al. reported that 6 weeks of endurance training attenuated Proteobacteria in overweight women (Munukka et al., 2018). So the hypothesis is that exercise might adjust Proteobacteria to a healthy level rather than cause an absolute increase or decrease. These results indicated that Proteobacteria may be particularly sensitive to intervention. Ralstonia belongs to the phylum Proteobacteria and is inversely associated with the severity of antipsychotic-related constipation in patients aged 18–23 (Zheng et al., 2021). In addition, the abundance of Ralstonia was reported to be severely reduced in the rectal samples of patients with irritable bowel syndrome compared to healthy controls, which seems to indicate that the reduction of Ralstonia has a negative effect on bowel function (Loranskaia et al., 2013). In this study, the abundance of Ralstonia increased significantly in the EAW group compared with that of the control group. The beneficial effects of the EAW training may be mediated in part by Ralstonia, but there are few reports about Ralstonia and more studies are needed to confirm it.

Remarkably, the abundance of Verrucomicrobia decreased in both groups, with Akkermansia being one of the main representatives of Verrucomicrobia. This study found that participants with a high abundance of Akkermansia at baseline showed a more pronounced reduction in the conventional group, although Akkermansia also decreased in the EAW group. According to previous reports, the aggravation of constipation is related to the increase in Akkermansia (Cao et al., 2017; Fu et al., 2021; Ma et al., 2023). Surprisingly, an increasing number of studies have shown that Akkermansia plays a key role in keeping the gastrointestinal tract intact and is considered the next generation of probiotics (Everard et al., 2013; Li et al., 2016; Munukka et al., 2018; Macchione et al., 2019; Wegierska et al., 2022). Exercise has opposite effects on Akkermansia in different populations (Munukka et al., 2018; Mokhtarzade et al., 2021). As microbes can have multiple effects on humans, it was difficult to determine whether the increase in the abundance of Akkermansia after SCI was negative or positive. It’s not clear how exercise affects Akkermansia, but at least it can be regulated.

First of all, the sample quantity of this experiment is very small, so it is necessary to increase the amount of samples to improve the statistical power, and to determine whether there is any difference in intestinal function recovery of patients with different injury segments by EAW. In addition, Gorman’s experiments showed that men’s intestinal function improved more than women’s after using EAW. The SCI patients in this study were all men, and whether gender was a factor remains to be investigated. Finally, the link between changes in gut microbiota composition and exercise regulation of intestinal function has not been fully demonstrated, and whether SCFAs play a role in it, more experiments are needed to support this relationship. The results of our pilot study must be confirmed in a larger scale in the future.

The implementation of 40 exoskeleton-assisted walking programs in patients with spinal cord injury who were unable to walk provided a limited degree of improvement in intestinal function. The effect of EAW training on intestinal function may be related to changes in the abundance of intestinal flora, especially the increase of beneficial bacteria. However, further research is needed to fully understand the changes in microbial groups caused by EAW training, and all their associated effects, especially gut microbiota metabolites. At the same time, AIDER robots with affordable exoskeletons need to be applied in the home to observe the impact of long-term maintenance training on SCI patients.

The datasets used and analysed in this study are available from the corresponding author upon reasonable request.

The studies involving humans were approved by the Ethics Committee of the General Hospital of Western Theater Command, Care Alliance Jinchen Rehabilitation Hospital of Chengdu, and the Rehabilitation Hospital of Sichuan Province. The studies were conducted in accordance with the local legislation and institutional requirements. The participants provided their written informed consent to participate in this study.

XH: Writing – original draft, Investigation, Methodology. JF: Writing – original draft. JLu: Data curation, Writing – review & editing. RP: Supervision, Writing – review & editing. AZ: Methodology, Supervision, Writing – review & editing. JLi: Data curation, Writing – review & editing. XG: Investigation, Writing – review & editing. XB: Investigation, Writing – review & editing. JW: Supervision, Writing – review & editing. CC: Investigation, Writing – review & editing. JY: Investigation, Writing – review & editing. YW: Investigation, Writing – review & editing. HX: Methodology, Writing – review & editing. QW: Supervision, Writing – review & editing. HC: Supervision, Writing – review & editing. YC: Supervision, Writing – review & editing. WW: Methodology, Project administration, Supervision, Writing – review & editing.