Since the 1990s, there has been a great deal of progress in the treatment of spinal cord injuries, and although there is some consensus on certain approaches, such as the early application of high-dose methylprednisolone [1], there are still many controversial and clinically pressing issues. For example, should spinal cord injuries be treated conservatively or surgically? Should surgical treatment be early or delayed? Should the surgical approach be anterior or posterior? In addition, cell transplantation has been used clinically in the treatment of spinal cord injury and has shown some results, but recovery of motor function is almost nil. In response to these results, what new ideas do researchers have for cell transplantation? This article provides a review of these issues in the treatment of spinal cord injury. 1. Conservative or surgical? The current classifications of spinal cord injury are to evaluate the spine and spinal cord separately, such as the Dennis [2] and AO [3] classifications of spinal injury and the ASIA classification of spinal cord injury, without combining spinal and spinal cord injury for comprehensive evaluation; their treatment is mainly chosen according to spinal stability, with most stable spinal injuries choosing conservative treatment [4] and unstable ones choosing surgical treatment. to prevent the deterioration of neurological function and the occurrence of secondary spinal deformities, while in fact there is a great controversy in judging the stability of the spine as well. Therefore, to answer the question of conservative or surgical, a comprehensive and scientific classification and assessment system for spinal cord injuries is needed to guide the clinic. In 2006, Moore et al [5] reported a new classification method for lower cervical spine injuries, the Cervical Spine Injury Severity Score System (CSIS). This system divides the cervical spine into four columns, the anterior column, the posterior column, and two lateral columns. The anterior column consists of the vertebral body, intervertebral disc, and anterior and posterior longitudinal ligaments; the posterior column includes the spinous process, the vertebral plate, and bony ligament complex structures such as the collateral ligament and the ligamentum flavum; the two lateral columns each include the lateral block and the articular synovial joint and joint capsule on one side. On CT 3D reconstruction, each column was scored according to fracture displacement and ligament rupture, with scores progressively increasing from 0 to 5 according to the degree of injury, with a score of 1 representing no displaced fracture and 5 indicating fracture displacement >5 mm or complete ligament rupture. The total score is a maximum of 20 points. Injuries involving multiple segments were calculated using the most severe segment.Anderson et al. analyzed this classification and found high confidence and reproducibility with mean kappa values of 0.977 and 0.883, respectively.The authors found neurological impairment in 11 of 14 patients with a total score ≥7 and recommended surgery at a total score ≥7 [6]. This scoring system quantifies the degree of cervical spine injury, but it does not introduce data from cervical spine MRI and does not take into account the status of neurological function, which has certain shortcomings. Recently, the American Spine Injury Study Group developed a lower cervical spinal cord injury classification system (Subaxial Cervical Spine Injury Classification System SLIC) [7], and this classification system includes three aspects: injury morphology, interdiscal ligamentous complex (Disco-ligamentous complex DLC), and neurological functional status. The scores of the three aspects are finally summed according to the injury status and their total scores can be used for treatment selection (Table 1).Vaccaro et al. analyzed the reliability of this method, which was 0.49, 0.57, and 0.87 according to injury morphology, DLC status, and neurological functional status, respectively, with reproducibility of 0.66, 0.75, and 0.90, respectively, for moderate confidence and consistency [8], and treatment recommendation compliance rate was 93.3%. Although this classification combines neurological function with injuries to bony, intervertebral, and ligamentous structures, this classification was evaluated using data from only 11 cases and is not yet widely used, and further clinical studies of large numbers of cases are still needed. In the thoracolumbar segment, the American Spinal Injury Study Group developed a scoring system for the extent of spinal cord injury in the thoracolumbar segment (Thoracolumbar Injury Classification and Severity Score, TLICS) [9], and the TLICS system is also divided into three areas: fracture pattern, posterior ligamentous complex ( The TLICS system is also divided into three aspects: fracture morphology, integrity of the posterior ligamentous complex, and neurological status, and the total score is calculated after scoring the items (Table 2).The greatest difficulty in the application of TLICS is the determination of the injury status of the posterior longitudinal ligamentous complex. The posterior ligamentous complex includes the supraspinous ligament, interspinous ligament, ligamentum flavum and the small joint capsule. Injury to the posterior ligament complex tends to cause instability in the spine and often requires surgical intervention because of its poorer healing ability compared to bony structures. The typical manifestations of injury are widened spinous process spacing and subluxation or subluxation of small joints, which can be determined by palpation of the spinous process gap, radiographs, or 3D CT reconstruction. In the absence of signs of complete rupture of the posterior ligament complex (increased spinous process gap), but the presence of injury on MRI can be defined as an indeterminate injury.TLICS should be corrected for clinical applications, such as significant retroconvex deformity of the fracture portion, significant collapse of the vertebral body , concurrent multiple rib fractures, sternal fractures, presence of ankylosing spondylitis, diffuse idiopathic osteophytes (DISH), osteoporosis The greatest advantage of TLICS is that it incorporates the status of nerve injury and posterior longitudinal ligament complex into the assessment system, trying to return the question “conservative or surgical?” with a specific score. question. This method has only just begun to be applied in China, and has not yet been reported in the literature. 2. Early or deferred surgery? Animal experiments have found that early decompression can promote the recovery of neurological function [11]. However, early surgery can cause deterioration of respiratory function, hemodynamics, and neurological function at the same time. In patients with acute trauma, repositioning and fusion are more difficult or cannot be performed due to the lack of special instruments and experienced surgeons [12].Fehlings and Tator had [13] reviewed more than 300 articles in the literature from 1966 to 1998, and experimental studies proved that early surgery is effective, while in clinical cases, there is no clear conclusion. However, according to the current situation, early decompression has been used as a routine choice [14]. Xu Shaoting reviewed the treatment of thoracolumbar segment burst fractures from 1987-1995, with 64 cases operated early and 54 cases operated late, and the recovery results were similar in both. In contrast, another group treated 51 cases of cervical spinal cord injury with extended hemilaminectomy decompression from 1995-1998, and early surgery was superior to late surgery. In fact, although Tarlov and Klinger’s experimental animal study affirmed the role of early decompression, several clinical studies did not confirm the superiority of early decompression over delayed decompression surgery.Wagner and Chehrazi, Levi et al [12] compared patients with early and delayed decompression and there was no significant difference in neurological recovery.In a prospective study by Duh et al, it was found that patients treated surgically within 25 h and after 200 h had better outcomes, but no clear evidence was given to support early or delayed decompression surgery. a prospective randomized study by Vaccaro et al [15] found no significant difference between early and delayed decompression. Although they analyzed the surgical time window using data from Class I evidence, he defined the concept of early as within 72h. La Rosa G et al [16] used Meta-analysis to search the literature on indications, principles of surgery, and timing of decompression after spinal cord injury in Medline from 1966 to 2000, supplemented by a manual search. The 1687 eligible patients were analyzed, and the results revealed that decompression within 24 h led to better outcomes compared with conservative treatment and delayed surgical decompression (>24 h). However, a homogeneous analysis of the sample found that early surgical decompression was reliable only for incomplete spinal cord injury. 2006 Fehlings [17] reviewed 66 publications from the last 10 years, especially the last 5 years, and concluded that no definite time window for decompression of spinal cord injury could be established, but the authors noted that urgent surgery is necessary for patients with bilateral cervical synovial interlock with incomplete spinal cord injury or spinal cord injury Patients with progressive deterioration must be treated with urgent surgery. Surgery should be performed as early as possible in the cervical spine, and there is evidence that surgery within 24 h reduces ICU length of stay and complication rates. However, there is still a lack of prospective randomized bulk clinical studies to analyze the timing of surgical decompression. 3. Anterior or posterior surgery? The choice of surgical access for spinal and spinal cord injury is more controversial and problematic. The general principle is to choose the anterior approach for compression from the anterior and the posterior approach for compression from the posterior. However, choosing a single approach in different segments of the spine can often release the anterior and posterior compressions, and in this case, a reasonable choice of a single approach reduces patient trauma and shortens the recovery period. The controversy is more prominent in the selection of anterior and posterior approaches to the thoracolumbar segment. McCormack et al [18] proposed Load Sharing Classification, which is based on the degree of vertebral body comminution and severity of kyphosis. The classification is based on the severity of comminution and kyphosis and is quantified by the score to determine whether posterior decompression and fixation alone or anterior reconstruction should be performed at the same time. The scoring system is based on plain films and CT, and is divided into 3 parts: extent of fracture involvement, degree of fracture displacement, and size of the kyphotic deformity. The fracture involvement on the cephalad side of the vertebral body was scored as 1 on lateral films, 2 on 30-60%, and 3 on >60%; the degree of fracture displacement (axial CT) was classified as small: 1 on displacement <2 mm, 2 on medium: ≥2 mm and involvement of <50% of the circumference of the vertebral body, and 3 on large: ≥2 mm and involvement of >50% of the circumference; posterior convexity deformity: 1 on ≤30, 40-90 on 2 points, ≥100 is 3 points. When there is no dislocation of the thoracolumbar segment fracture, the posterior approach is chosen for ≤6 points; the anterior approach is chosen for ≥7 points; but the posterior approach is chosen for ≤6 points in the presence of dislocation; and the combined anterior-posterior approach is chosen for ≥7 points.Wang used a biomechanical approach to validate the load-sharing classification [19]. Although this classification method is not widely used in clinical practice in China, it is more applied abroad, and it provides at least some basis for the selection of surgical access for thoracolumbar segment injuries. 4. New thinking about cell transplantation The initial success of stem cell therapy for Parkinson’s disease at the beginning of this century led some scholars to believe that the use of stem cell transplantation could be used to treat other diseases of neurological disorders, and spinal cord injury is among them. T.S. Sun [20] pointed out that this idea seems logical, but the pathological basis of Parkinson’s disease is primitive adult ganglion cells, whereas other diseases such as spinal cord injury involve advanced highly developed motor neurons such as Bates cells. Mature nerve cells are separated from adult ganglion cells by millions of years in the evolutionary spectrum, and as early as the 1970s, the famous biologist Theodosius Dobrzynski (1928-1989) was a member of the Society for the Study of Neurology. Theodosius Dobzhansky hit the nail on the head when he pointed out that everything in biology would be meaningless if it were not based on evolutionary theory [21]. The repair of the spinal cord after injury should also follow the principles of evolutionary theory, and the restoration of function is strictly evolutionary, i.e., the repair capacity of the less evolved structures is greater than that of the more evolved structures. The spinal cord contains all the structures that connect the brain to the peripheral nervous system, so it covers a wide range of the nervous system from the lowest primitive structures (reticular formation) to the highest developed neurons (Bates cells) and their conduction tracts (pyramidal tracts). The general sequence in which the spinal cord is repaired according to its principles is reticular formation, cerebellar connections, spinal thalamic connections, and corticospinal connections. Although animal studies on spinal cord injury have confirmed that cell transplantation therapy helps to restore spinal cord injury, including motor function, this does not fully correspond to the results of clinical trials. Currently, more than 1000 spinal cord injury patients worldwide have undergone OEG transplantation [22], and most of the cases that received transplantation mainly showed the repair of less evolved structures in the spinal cord such as: improvement of temperature, color and bladder function and bowel function below the plane of spinal cord injury (vegetative function), reduction of muscle tone (spinal cerebellum, red nucleus spinal cord connections), and the recovery of these functions in the clinic often overlooked and not easily measured, some patients also show significant recovery of sensory function, with a decrease in sensory planes by 3-10 spinal cord segments, resulting in a significant increase in sensory scores. Sun T.S. et al [23] obtained similar results in 11 patients with spinal cord injury treated with OEG transplantation. Animals with spinal cord injury receiving olfactory sheath cell transplantation could achieve more satisfactory functional recovery, including motor, while patients with spinal cord injury receiving olfactory sheath cell transplantation could only achieve mild to moderate functional improvement, with the degree and likelihood of functional recovery ranging from high to low as 1, skin nutritional status, 2, spasticity, 3, bladder and bowel function, 4, superficial sensation (up to 10 segments), and 5, motor ( limited to the area of injury only); this is in perfect agreement with the evolutionary theory. This does not allow us to dismiss lightly the role of olfactory sheath cell transplantation in the repair of spinal cord injuries, but it is just a long way from our expectations. The complexity of the spinal cord determines that no single therapeutic intervention can solve all problems. Therefore, although cells have the roles of bridging, support, secretion of growth factors, and replacement, which play a multifaceted role in spinal cord injury repair and can be said to cover all aspects of spinal cord injury repair, more and more scholars emphasize the comprehensive intervention of multiple therapeutic approaches, including cell transplantation. Therefore, there is still a long way to go and a lot of work to be done in the research of spinal cord injury treatment including cell transplantation. We should encourage more and more in-depth research, but all experiments should be based on evolutionary theory and strictly follow the principles of scientific methodology. The above mainly reviews the progress of surgical treatment of spinal cord injury and the new concept of cell transplantation, but the treatment of spinal cord injury should be a combination of multiple approaches, including drugs and rehabilitation, and the restoration of spinal cord function requires a multidisciplinary effort.