Essay: Paraplegia or paraparesis after otherwise successful thoracic or thoracoabdominal aortic surgery

INTRODUCTION
Paraplegia or paraparesis after otherwise successful thoracic or thoracoabdominal aortic surgery remains a devastating complication. The incidence of postoperative spinal cord complications has been reported to be 3% -10% with various strategies to protect the spinal cord. Spinal cord ischemia and subsequent neurological complications have been shown to correlate with the duration and severity of ischemia, failure to establish adequate blood supply to the spinal cord, and ischemic reperfusion injury. Recent developments have revealed that ischemic reperfusion injury involves both innate and acquired immunity.
T lymphocytes have been shown to infiltrate into the ischemic brain by 24 h after onset (1). Involvement of T lymphocytes, representing acquired immunity, has thus been implicated in the post-ischemic inflammation and the tissue damage seen in the brain 24 h after onset (2).
In the present study, we hypothesized that suppression of acquired immune function in mice would enable us to reduce the level of ischemic injury to the spinal cord. We investigated the impact of acquired immunity on spinal cord ischemia (SCI) using wild type and SCID mouse models of paraplegia.
MATERIALS AND METHODS
Animals
Female C.B-17/lcr wild type and SCID mice (10-13 weeks old; CLEA Japan, Tokyo, Japan) were used for all experiments. The handling of laboratory animals and their use in experiments conformed to the ”Guidelines for Animal Experiment at Kobe University Graduate School of Medicine (permission number A140605)”; in addition, all animals received humane care and treatment in accordance with the ”Guide for the Care and Use of Laboratory Animals” (www.nap.edu/ catalog/5140.html).
SCI Model and Experimental Design
The operative procedures to induce SCI in mice were performed as previously described with modifications (3, 4). After the mouse was weighed, anesthesia was induced by 5 % isoflurane, and the trachea was intubated using a 20 G custom made catheter for mechanical ventilation. The mouse was positioned in a spine position and anesthesia was maintained by 1.3% to 2% isoflurane with 100% oxygen. Paravertebral muscle temperature was measured and maintained at 37.5 ?? 0.5 ??C during surgery using a heating pad. Left femoral artery blood pressure was monitored for distal aortic flow. Heparin (1000 IU/kg) was injected via the left femoral arterial catheter. First, under direct visualization, ischemic preconditioning for intestinal protection was induced by transient aortic clamping (30 s) placed between the left common carotid artery and the left subclavian artery followed by reperfusion for 60 s. The first clamp was then placed on the aortic arch between the left common carotid artery and the LSA, and the second clip was placed on the left subclavian artery at the same time. The clamps were then removed, and the chest was closed. Ten minutes after reperfusion, the arterial catheter was removed, and animals were allowed to recover from anesthesia. Sham mice underwent exposure of the aortic arch through the same procedure but did not undergo aortic cross-clamping. To define the ischemic duration which caused paraplegia, wild type mice and SCID mice were randomly assigned to receive 7 min (wild type, n=3; SCID; n=5), 9 min (wild type, n=3; SCID, n=3), 9.5 min (wild type, n=6; SCID, n=4), or 10.5 min (wild type, n= 51; SCID, n= 43) with IP. Motor function of SCID mice was compared with that of wild-type mice. For histological analysis of mice with paraplegia, wild type and SCID mice with paraplegia were euthanized at 12 h (wild type, n=6; SCID, n=5), 24 h (wild type, n=8; SCID, n=5), 48 h (wild type, n=5; SCID, n=4), 72 h (wild type, n=6; SCID, n=4) and 7 days (wild type, n=4; SCID, n=4) after reperfusion, and sham mice (wild type, n=4; SCID, n=4) were also euthanized. Spinal cords were then harvested.
Assesment of Functional Motor Neuron
Motor function was quantified serially at 0, 6, 12, 24, 36, 48, and 72 h after SCI using the Basso Mouse Scale (BMS) (5). Scores for the Basso Mouse Scale range from a score of 0 for complete paraplegia to a score of 9 for normal function. A Basso Mouse Scale score < 6 (0 to 5) was taken as indicating paraplegia. Tissue Sampling and Immunohistochemistry and Neuronal Quantification Mice were deeply anesthetized and transcardially perfused with 4% paraformaldehyde at 12, 24, 48,or 72 h or 7 days after reperfusion. Lumbar enlargement of the spinal cord located at the T10 through L3 vertebral column were removed en bloc and embedded in paraffin. Spinal cord embedded in paraffin was sectioned to a thickness of 5??m and 50 sections of each spinal cord were divided on ten slide glasses (5 sections each on one slide glass). Immunohistochemistry was performed using antibodies to identify viable neurons (NeuN, 1:5000; Millipore, Billerica, MA), activated microglia (Iba1, 1:2500; Proteintech, Chicago, IL), CD3+ T cells (CD3, 1:2000; Spring Bioscience, Pleasanton, CA), CD21+ B cells (CD21, 1:2500; Abcam, Cambridge, UK) and reactive astrocytes (GFAP, 1:1; DAKO, Carpinteria, CA). For quantitative analysis, the number of NeuN positive neurons and the number of inflammatory cells (activated microglia, CD3+ T cells, and CD21+ B cells) in the spinal ventral horn were counted in a randomly chosen section. Statistical Analysis Parametric data are presented as the mean ?? standard deviation. To compare parametric data among groups, Student't test was used. Differences in numbers of viable neurons and inflammatory cells were compared between groups at individual time points using the Mann-Whitney U test. Differences were considered statistically significant for values of p < 0.05. All data analyses were performed using JMP for Macintosh version 9.0.2 (SAS Institute, Cary, NC). REAULTS Neurological Outcomes after Reperfusion Both wild type and SCID mice developed immediate paraplegia after occlusion for 10.5 min with IP (Table 1). All wild type and SCID mice subjected to 7 and 9 min of SCI with IP exhibited rapid recovery (within 3 h) to almost normal appearance (Fig. 1). In the group subjected to 10.5 min with IP, immediate paraplegia was caused in 56% of SCID mice (n= 24/43) and in 75% of wild type mice (38/51). However, delayed paraplegia was not observed in any wild-type mice or SCID mice. No significant difference in the frequency of paraplegia was observed between wild type and SCID mice that were subjected to 10.5 min of SCI with ischemic preconditioning. Physiological Variables In SCID and wild type mice, femoral arterial pressure after reperfusion was higher in mice without paraplegia than that in mice with paraplegia (Table 2). No significant difference in femoral arterial pressure was observed between SCID mice and wild-type mice with paraplegia, excluding femoral arterial pressure at 4 and 5 min after reperfusion. Neuronal Viability Neurons in the spinal ventral horn started to degenerate at 12 h after reperfusion in SCID mice as well as in wild-type mice with paraplegia. SCID and wild-type sham mice displayed significantly larger numbers of viable neurons in the spinal ventral horn compared with the mice at individual time points after reperfusion (Fig. 2). SCID mice did not show significantly more viable neurons in the ventral horn compared to the wild-type mice at any time points after reperfusion. Expression of Inflammatory Cell Microglial activation in the spinal ventral horn was assessed by Iba1 immunostaining at 12, 24, 48, and 72 h,and 7 days after reperfusion and in sham mice. Both in wild type and SCID mice with paraplegia, activated microglia were increased at 12 h and 48 h after reperfusion, with relatively little activation at 24 h after reperfusion. At 72 h after reperfusion, activated microglia were decreased, and at 7 days after reperfusion, little infiltration of activated microglia was observed. Normal spinal cord tissue (sham mice) showed almost no microglial activation (Fig 3). Immunostaining for CD3+ T cells and CD21+ B cells were used to determine the activation of lymphocytes in the spinal ventral horn at 12, 24, 48,and 72h,and 7 days after reperfusion and in sham mice. In SCID mice with paraplegia, no infiltration of CD3+ T-cells or CD21+ B-cell into the ventral horn of spinal cord was observed. Wild-type mice with paraplegia also showed scant infiltration of CD3+ T cells and CD21+ B cells into the ventral horn at all points after reperfusion. Reactive Astrocytes In wild type and SCID mice with paraplegia, the reactive astrocytes in the ventral horn were observed after the loss of spinal ventral horn cells. Seven days after reperfusion, the gliosis was confirmed in the gray matter of both wild type and SCID mice with paraplegia (Fig. 4). Discussion The present study found that no significant difference in the duration of aortic arch clamping required to produce the SCI model between the SCID mice and the wild-type mice. Furthermore, regarding the number of degenerated neurons in the ventral horn of the spinal cord in mice with paraplegia, no time points after reperfusion showed significant differences between SCID and wild type mice. SCID mice after reperfusion did not display significantly more viable neurons in the ventral horn compared to wild type mice. Paraplegia is a significant and dreadful complication that can occur after successful thoracic or thoracoabdominal aortic interventions. Although various strategies, such as preoperative identification of critical segmental arteries, intraoperative monitoring of evoked potentials, reduced the duration of ischemia, distal aortic perfusion, maintenance of preoperative higher systemic blood pressure, control of back-flow from the patent intercostal artery, cerebrospinal fluid drainage, and hypothermia have been developed to minimize the incidence of neurological complications (6), optimal pharmacotherapies for neuroprotection have not been elucidated. Ischemic reperfusion injury is a common and crucial problem in many different organs, including the kidneys, heart, liver, lungs, intestine, and brain. IRI is now recognized as a highly intricate cascade of events that includes interactions between the vascular endothelium, interstitial compartments, circulating cells, and numerous biochemical entities (7). Immune cells such as T cells and macrophages and their cytokines have been shown to play a pivotal role in the immunomodulation of IRI (8-11). Both innate immunity and acquired immunity have been considered as critical components in ischemic reperfusion injury. The regulation of these immune responses could hold promise for neuroprotective therapy. The importance of innate immune responses is exemplified by the fact that some anti-inflammatory strategies have been revealed to ameliorate post-ischemic damage (12-14). Inflammatory responses to ischemic damage are initiated by the microglia. These responses are further promoted by infiltrating neutrophils and macrophages, resulting in the production of inflammatory cytokines and other cytotoxic mediators. The cells that engage in innate immune response, representing leukocytes and microglia, express toll-like receptors (15, 16). It is increasingly clear that toll-like receptors, particularly toll-like receptor 2 and toll-like receptor 4, play a pivotal role in the mechanisms underlying cerebral ischemic damage (17-19). Furthermore, for the first time, Bell et al (20) reported toll-like receptor 4 was critical in the mechanisms of paraplegia as well as stroke using a mouse model of SCI. Regarding the acquired immunity, some experimental data have demonstrated an important role for lymphocytes in cerebral ischemic reperfusion injury. Hurn et al. (8) evaluated the infarction volumes using wild type and SCID mouse models of stroke. They found that infarction volumes in SCID mice were markedly reduced compared to wild-type mice, and concluded that T and B lymphocytes acted as mediators of stroke. Yilmaz et al also demonstrated that compared to wild-type mice, cerebral infarct size and neurological deficit were significantly reduced in a Rag1-/- mouse model of cerebral ischemia, which is lacking in both T and B lymphocytes, and that mice deficient in B lymphocytes failed to show improvement of ischemic brain injury. The authors found that T lymphocyte, not B lymphocytes, contributed to the inflammatory response in stroke. On the other hand, although Tachibana et al (21) presented the involvement of T cells in a rabbit model of SCI, few reports have described the contribution of lymphocytes to SCI in a murine model. In the present study, contrary to our hypothesis, SCID mice did not show a significantly larger number of viable ventral horn neurons compared to wild-type mice. Moreover, even in the wild-type mice, the infiltration of T and B lymphocytes into the spinal ventral horn was hardly observed. 'The blood brain barrier (BBB) plays an important role in the homeostatic regulation of the brain microenvironment for the proper functioning of the neurons (22). The BBB also maintains the immune-privileged status of the brain by restricting the entry of lymphocytes (23). Once the BBB is collapsed by hypoxia such as stroke, the central nervous system loses its immune-privileged status, and activated glial and endothelial cells produce chemokines and cytokines that enhance the infiltration of lymphocyte and leukocytes across the BBB (2, 24, 25). Lymphocytes are ordinarily excluded from the central nervous system, but can be seen in the postischemic brain (26). In the onset and progession of experimental autoimmune encephalomyelitis, the involvement of T lymohocytes infiltrating into the brain and spinal cord has been suggested (27, 28). However, the present study scarcely detected lymphocytes in the spinal ventral horn even in wild type mice. Interestingly, a biphasic response of activated microglia was observed in our mouse model of immediate paraplegia. Smith et al. reported that delayed-onset paraplegia correlated with the bimodal distribution of both microglial activation and chemokine production (29). In our study, early expression of activated microglia was likely due to the resident microglia, and later expression may have been due to recruitment of leucocytes across the disrupted BBB. Immediate paraplegia is thought to be caused by irreversible and excessive ischemic neuronal injury in the spinal cord. Our findings suggested that immediate paraplegia, as a huge ischemic injury, had more to do with microglial cells rather than lymphocytes. To protect against intestinal necrosis, ischemic preconditioning was induced by transient aortic clamping between the left common carotid artery and left subclavian artery, was performed. Although a paraplegia model in C.B-17/lcr mice was induced by 10.5 min of aortic clamping, most of such mice become critically ill with intestinal necrosis. We have performed ischemic preconditioning for intestinal protection to improve the survival of SCI model mice (30) (31) (32). As a result, no mice with intestinal necrosis were observed in this study among mice subjected to 10.5 min of SCI with ischemic preconditioning. This study has a few limitations. First, the animals we used were female C.B-17/Icr mice. In immunological studies, male C57BL/6J mice are generally selected to exclude any potential effects of the estrous cycle. Second, ischemic preconditioning was performed to protect against the ischemic injury to the intestine. However, this technique may have influenced the immunological response of the spinal cord. Finally, the follow-up period in this study was only 1 week. A longer follow-up may have identified differences between the wild type and SCID mice. In conclusion, no significant difference was identified between wild type and SCID mice in the duration of aortic cross-clamping required to induce paraplegia. SCID mice with paraplegia did not have significantly more viable neurons in the ventral horn compared to the wild type mice with paraplegia at any time point after reperfusion. Even wild type mice with paraplegia showed scant infiltration of T cells and B cells into the ventral horn after reperfusion. Neuronal damage in the ventral horn among SCID mice with paraplegia was observed to a similar extent to that seen in wild type mice with paraplegia. In our mice model of immediate paraplegia, in the acute phase after SCI, the acquired immunity might not have involved in SCI and reperfusion injury.