Spinal cord injury (SCI) is defined as damage to the nerve tissue of the spinal cord, resulting in reduction or loss of sensory and/or motor function (1). Globally, more than 15 million people are suffering from SCI, of which up to 90% of cases are of traumatic origin (2, 3). In traumatic SCI, the primary injury is caused by an external physical force affecting the spinal cord, such as a fall, motor vehicle accident, sports-related accident, or violence (4–6). This trauma results in acute cell damage and cell death in the surrounding neuronal and vascular tissues. Subsequently, a secondary injury cascade is initiated, characterized by inflammatory cell infiltration, ischemia, edema, hemorrhage, and the release of cytotoxic products, resulting in further spinal cord damage and neurological dysfunction (1, 7). In addition to the pathophysiological subdivision of traumatic SCI into primary and secondary injuries, it can also be divided into temporal phases. Although different classifications are used in literature, generally, acute and subacute phases last hours to weeks, whereas the chronic phase refers to six months post-injury and beyond (1, 8).

During the secondary injury phase, the destruction of the blood-spinal cord barrier following the initial trauma allows the recruitment of various immune cells into the injured spinal cord. Local inflammation of the spinal cord in the early stages following SCI is mediated by the innate immune system. As demonstrated in human post-mortem SCI spinal cords, neutrophils are the first immune cells to reach the areas of injury, peaking in number at 1-3 days and remaining increased up to 10 days post-injury (9, 10). Subsequently, resident microglia are activated and monocytes/macrophages infiltrate the spinal cord. At 5-10 days post-injury and beyond, microglia and macrophages are the predominant inflammatory cells in the spinal cord (9–11). Since it takes longer for the adaptive immune system to initiate an immune response via processes of antigen recognition, activation, proliferation, and clonal expansion, few lymphocytes populate the lesion site in the acute stages following injury (11). Nevertheless, at one week to months post-injury, CD8⁺ and CD4⁺ T cells could be detected in human post-mortem spinal cord lesions (9, 10). In addition, for a subset of SCI patients in the subacute phase, a pronounced infiltration of B cells and active plasma cells was observed in human autopsy spinal cord tissues at 15 days up to 36 days post-injury (10, 12).

Changes in the immune system are not limited to the accumulation and activation of immune cells at the site of injury. Instead, SCI also affects peripheral immune function and composition, as demonstrated in both SCI animal models and SCI patients. Here, we focus on human studies since species-specific differences have been reported in immune responses following SCI (13–15), contributing to a lack of clinical translation of the findings in experimental SCI models. Contradictory findings have been reported in SCI patients at various stages following the injury with, on the one hand, inflammation and autoreactivity, but on the other hand, immunosuppression. In this review, we discuss the current literature supporting both perspectives regarding the peripheral immune system in traumatic SCI patients. An overview of the studies that are covered in this review is given in Supplementary Tables 1 , 2 .

Emerging experimental and clinical data indicate that SCI triggers activation of the peripheral immune system, with signs of systemic inflammation and autoreactivity, at various stages following the initial trauma ( Figure 1A ).

Although seemingly contradictory to the immune activation seen following SCI, systemic immunosuppression has been described in SCI patients as well ( Figure 1B ). One of the main indications of reduced immune function is the frequent occurrence of infections, predominantly affecting the respiratory and urinary tract (64–67). Respiratory tract infections are common, with a reported incidence up to 60%, during the first few days after SCI (65, 66). Contrarily, the urinary tract is the primary origin for infection during rehospitalization, affecting 34% of rehospitalized SCI patients (65, 68). Moreover, large epidemiological studies have indicated infections as the leading cause of death in the subacute and chronic phase following SCI and as independent risk factors for a worse neurological recovery (69–73). Nevertheless, underlying mechanisms explaining the high susceptibility to infections are poorly understood. In recent decades, several publications have shown changes in immune cell numbers and functions that suggest the triggering of immunosuppression following SCI.

In addition to SCI-induced inflammation, as discussed above, NK cells are frequently related to immunosuppression following SCI. One study reported decreased frequencies of CD56^bright NK cells and CD56^dim NK cells at 0-3 days post-injury compared to HC, remaining significantly suppressed up to 12 months following SCI (19). In addition, the expression of NK cell genes was significantly reduced in subacute (up to 6 months post-injury) and chronic stages (≥1 year post-injury) (19, 38). Although in another study no significant differences in total and CD56^bright NK cells could be demonstrated in acute SCI patients (30.3 ± 18.9 h post-injury), significantly decreased frequencies of CD56^dim and activated NK cells were shown compared to HC and TC (74). A statistically significant decrease in the percentage of blood NK cells following SCI has also been reported in subacute/chronic SCI patients (>3 months post-injury) compared to HC (33). In contrast, we did not find significant differences in NK cell frequencies of HC, (sub)acute (≤1 month post-injury) and subacute/chronic (>1 month post-injury) SCI patients (47). Three other studies (including samples of SCI patients >3 months, 7-120 months, and >5 years post-injury) also did not observe significant differences in NK cell numbers and frequencies in the blood and bone marrow compared to HC (31, 32, 75). Despite the inconsistent results in literature concerning the frequency of NK cells post-SCI, several research groups have confirmed a decrease in their cytotoxicity. NK cell function deficiencies after SCI were detected as early as two weeks post-injury, being ultimately suppressed at two months (76–79). These deficiencies remained present for years following SCI (33, 76–80).

SCI has also been reported to negatively affect the number and function of circulating T cells, although conflicting results have been reported concerning T cell frequencies. In one study, significant reductions in T cell numbers were evident within 24 h following SCI compared to TC (81). Cell numbers remained suppressed up to 3-4 days post-injury, but recovered significantly by the end of the first week and reached plateau control levels in later stages (105-136 days post-SCI) (81). However, since all patients were treated with a high dose of the corticosteroid methylprednisolone within 24 h after SCI, a contributing effect of this therapy cannot be excluded. In contrast, comparable frequencies of CD4⁺ and CD8⁺ T cells were demonstrated in SCI patients within 24 h of injury compared to HC (30). Moreover, we and others showed significantly increased frequencies of total T cells and CD4⁺ T cells in subacute/chronic SCI (>1 or >3 months post-injury) in comparison to (sub)acute SCI (≤1 month post-injury) and/or HC (33, 47). Two other studies reported no significant differences in circulating T cell frequencies and numbers in smaller cohorts of chronic SCI patients (7-120 months and >5 years post-injury) compared to HC (31, 32). Additionally, frequencies of total, CD4⁺, and CD8⁺ T cells in bone marrow were not significantly different between chronic SCI patients (8 months to 5 years post-injury) and HC (82). Concerning T cell function, changes have been described in multiple studies. First of all, several studies analyzed the function of lymphocytes by measuring their proliferation in response to mitogens. When comparing peripheral lymphocytes from chronic SCI patients (7-120 months post-injury) to HC, significant suppression in the lymphocyte blastogenic response to three different mitogens was reported (32). Furthermore, T cell function, as measured by lymphocyte transformation induced by a T cell mitogen, showed a significant decline starting at two weeks and reaching a minimum at three months post-injury (76–79). After that, T cell function was restored gradually over the next 3 months and stayed constant between 6 and 12 months following SCI (76–79). T cell activation, as reflected by the expression levels of IL-2 receptors, showed the same pattern in time (76–79). Lastly, the ability of T cells to kill allogeneic lymphocytes was significantly reduced in chronic SCI patients (7-40 years post-injury) compared to HC (80).

When analyzing specific T cell subsets, regulatory T cells (Tregs), that were identified as CD25⁺CD127^loCD4⁺ T cells that express CCR4⁺ and/or HLA-DR⁺, were significantly increased in chronic SCI (>1 year post-injury) (46). In addition, gene set enrichment analysis using an existing RNA-sequencing dataset of peripheral blood leukocytes indicated a significant increase in the expression of Tregs in acute SCI patients (30.3 ± 18.9 h post-injury) compared to HC and TC (74). However, production of IL-10, an anti-inflammatory cytokine that can be produced by Tregs, was not significantly different between chronic SCI patients (>6 months post-injury) and HC following in vitro stimulation of PBMC with phytohemagglutinin and lipopolysaccharide (83). Another T cell subset that has been related to immunosuppression following SCI is the CD4⁺ effector memory T cell re-expressing CD45RA (termed Temra). Significantly decreased frequencies of this cell type have been reported by us in (sub)acute SCI patients (≤1 month post-injury) compared to HC (47). Recent studies indicated Temra as specialized effector memory T cells that play a role in protective cytotoxic responses against pathogen-infected cells, indicating that decreased numbers of these cells may possibly challenge the eradication of infections (84–86).

Furthermore, alterations in the expression of chemokines and their receptors following SCI have been associated with immunosuppression. By analyzing chip-based RNA-sequencing data from peripheral blood leukocytes of acute SCI patients, expression of C-C chemokine receptor type 7 (CCR7) was found to be significantly downregulated in acute SCI patients (30.3 ± 18.9 h post-injury) compared to TC and HC, which was positively correlated with T follicular helper (Tfh) cell function (74). The authors hypothesized that following acute SCI, peripheral CCR7 is downregulated causing suppression of Tfh cells through the chemokine signaling pathway, contributing to immunosuppression (74). Moreover, a significant reduction in activated B cells was demonstrated following acute SCI (74). This reduction may also be linked to decreased CCR7 expression and impaired Tfh function, as Tfh cells play an important role in B cell maturation in the germinal center reaction. Additionally, CCR7 plays a role in effector T cell migration and positioning within secondary lymphoid organs (87). Secondly, a significant defect in the production of C-X-C motif chemokine ligand 10 (CXCL10) by monocytes in response to TLR7 and TLR9 stimulation was reported in chronic SCI patients (>6 months post-injury) with an injury level above T6 compared to HC (83). This defect in the innate immune response may contribute to the increased susceptibility to infections observed in chronic SCI.

Interestingly, it has been hypothesized that the reduced immune cell activity following SCI is attributed to a qualitative rather than a quantitative defect. In a study in which bone marrow aspirates of small cohorts of chronic SCI patients (7-40 years post-injury) and HC were analyzed, significantly reduced long-term colony formation of all hematopoietic cell lineages was reported following SCI (80). Notably, another study that also analyzed bone marrow aspirates, demonstrated that the percentage of hematopoietic stem cells was elevated in chronic SCI patients (8 months to 5 years post-injury) compared to HC, suggesting that it is the ability of the stem cells to form mature immune cells, rather than the proliferative capacity of the stem cells, that is impaired (82). The changes in the early maturation process of immune cells such as NK and T cells might in turn interfere with the development of their cytotoxic machinery and their functionality. Additionally, immune function deficits may be influenced by the level of injury, which has been demonstrated in rodent studies to be related to the anatomy of the sympathetic column of the spinal cord (88–91). In brief, upon SCI, sympathetic preganglionic neurons below the lesion level are removed from inhibitory supraspinal control, resulting in their overactivation. Consequently, the neurons release excessive amounts of norepinephrine and glucocorticoids in lymphoid organs, leading to recurrent or prolonged activation of β-2-adrenergic or glucocorticoid receptors on immune cells, promoting immunosuppressive responses (88, 91, 92). This effect is more pronounced for lesions at cervical or high thoracic spinal segments since the majority of sympathetic innervation of immunologically relevant organs originates from preganglionic neurons situated below the fifth thoracic segment (T5), both for experimental animal models and for humans (89). Indications for this phenomenon were also obtained in human studies. In motor complete SCI patients, injuries at high (T1–4) relative to mid (T4-8) thoracic levels and at mid relative to low (T9-12) thoracic levels were independently associated with an increased risk for pneumonia following SCI (89). In another study, the mean killing capacity of NK cells was significantly reduced in subacute/chronic tetraplegic SCI patients, whereas no significant difference was observed in subacute/chronic paraplegic (at or below T10) SCI patients (>3 months post-injury) compared to HC (75). Although no statistically significant difference could be demonstrated for neutrophils, the phagocytic activity of neutrophils showed a decreasing trend in the tetraplegic group, whereas their activity was unchanged in paraplegic patients when compared with HC (75). In contrast, other studies did not observe differences in neutrophil oxygen consumption, eosinophil activation, and the cytotoxic capability of NK and T cells between HC, chronic tetraplegics (cervical segment (C)5-7), and chronic paraplegics (T5-10) (>5 years and 7-40 years post-injury) (31, 80). In addition, no significant difference in NK cytotoxicity was observed between subacute/chronic SCI patients with injuries at T6 and above and those with injuries below T6 (>3 months post-injury) (33).

Altogether, the studies discussed above hypothesize that changes in lymphocyte frequencies and functions contribute to the increased risk of infections following SCI. These changes are thought to result from impaired maturation of immune cells and have been related to lesions at higher levels along the spinal column, potentially affecting immune regulation and response.

SCI adversely affects the immune system, contributing to changes in immune system composition and function both at the site of injury and in the periphery. Seemingly contrasting findings of peripheral immune system activation and suppression have been reported in traumatic SCI patients and these phenomena seem to emerge simultaneously in both (sub)acute and chronic stages following the injury. In response to SCI, a systemic inflammatory host response is triggered. In the (sub)acute stages, immune activation manifests prominently through increased neutrophil and monocyte frequencies and their enhanced oxidative and migratory capabilities, whereas in chronic stages, increased levels of cytokines, acute phase proteins, TLR signaling genes, and T and B cell subsets related to activation and inflammation were reported. In parallel, SCI-induced immune suppression develops quickly and extends into chronic stages, possibly to avert autoimmunity against self-antigens that are released or expressed following SCI. Immune suppression post-SCI is most importantly marked by significant decreases in T cell and NK cell frequencies and functions and is influenced by the level of injury, although this should be confirmed in further studies. Notably, a similar paradox of immune activation and suppression has been described in other conditions involving neuronal injury, including stroke and traumatic brain injury (93–95). Additionally, deficient or ineffective initiation of SCI-induced immune suppression may result in autoimmunity, in which MBP-reactive and memory T cells, B cells, and autoantibodies have been reported to play a role. Presumably, a balance between immune system activation and suppression is necessary to reestablish homeostasis in the affected tissues. As a consequence, an imbalance may result in an exaggerated manifestation of one of both responses. The resulting peripheral changes in immune function could complicate SCI recovery or lead to an aberrant immune reaction long after the initial trauma, for example, leading to an increased susceptibility to infections or systemic organ failure.

The peripheral changes after SCI have only been studied in small patient cohorts and numerous clinical characteristics, such as the spinal injury level, the time after injury, and the injury severity, can influence the progression of SCI and study results. Therefore, future studies should focus on large patient cohorts with varied clinical characteristics that accurately represent the diversity of the SCI patient population. Furthermore, the association of clinical characteristics with peripheral immune reactions following SCI should be investigated. A more detailed understanding of alterations in immune cell function, their temporal occurrence, and potential correlations with clinical characteristics could facilitate the clinical assessment and therapeutic management of SCI. Additionally, many of the reviewed publications lack information on the included treatments that could potentially affect the immune system. In this regard, whilst their clinical utility remains controversial, traumatic SCI patients are frequently treated with immunomodulatory drugs, including corticosteroids such as methylprednisolone (7, 96). Other therapeutic options for SCI that interfere with the immune response include non-steroidal anti-inflammatory drugs (NSAIDs) and antibiotics (7). In case these immunotherapies are not listed as exclusion criteria, their potential impact on study outcomes cannot be excluded. Hence, participants receiving treatments that impact the immune response should either be excluded or their treatments should be clearly mentioned in future studies. Moreover, examining the correlation between these treatments and the immune response to SCI would be valuable.

Finally, the reports discussed in this review highlight the need for the development of novel, more efficient therapeutic approaches to address the immune system-related consequences of SCI that extend beyond the spinal cord. Notably, it is important that these strategies control peripheral immunosuppression without aggravating immune cell-mediated damage to the spinal cord and other organs. Moreover, therapies that could ameliorate SCI recovery and prevent further damage by addressing autoreactivity merit attention. Reevaluating peripheral changes in immune system composition and function following SCI from these perspectives could open doors to more specific and personalized therapies for SCI.