【专题文献】之胸腰椎骨折-胸腰段爆裂性骨折的手术与非手术治疗

Seminars in
SPINE SURGERY
Operative versus Nonoperative Treatment of Thoracolumbar Burst Fractures
胸腰段爆裂性骨折的手术与非手术治疗

Adam L. Shimer, MD,* and Brian W. Su, MD†




High energy injuries to the thoracolumbar (TL) region are commonly encountered and have been described since the time of Hippocrates. Despite this long history and mountains of manuscripts generated on the topic, the optimal care of TL burst fractures remains controversial. There is such great heterogeneity in study designs, inclusion criteria, and interventions used that traditional treatment guidelines require a critical re-evaluation. Many outcome studies have failed to correlate radiographic indicators such a kyphosis, loss of vertebral body height, and canal compromise to long-term clinical outcomes. Furthermore, 3 large prospective, randomized trials have demonstrated outcome equivalency of operative and nonoperative treatment for TL burst fractures without neurological compromise. Surgical intervention remains the standard of care in the setting of progressive neurological deficits, fracture-dislocations, and translational instability. Semin Spine Surg 22:38-43 © 2010 Elsevier Inc. All rights reserved.
KEYWORDS thoracolumbar, burst, fracture, review, evidence-based

History
历史

The thoracolumbar (TL) region of the spine ranges by definition from T11 to L2 inclusively. This area is particularly susceptible to injury because it transitions from a fixed kyphotic spine to a mobile lordotic spine. In 1931, Jones1 described a pure flexion injury to the TL spine that could be anatomically reduced and adequately treated with hyperflexion and cast maintenance in that position. In 1963, approximately 30 years later, Holdsworth2 reexamined TL fractures both as a homage to Watson Jones and because of his overall dissatisfaction with clinical outcomes after treatment of these injuries. The study reported on his experience of treating more than 1000 patients and was later reprinted in JBJS in 1970 as a result of the excitement it generated from lectures he gave at the Albert duPont Center in 1969, the same year of his death. Holdsworth described the “two-column” model of TL stability emphasizing the importance of the posterior ligamentous complex (PLC), which he defined as a group of ligaments composed of the posterior facet joint, intraspinous ligament, ligament flavum, and the supraspinous ligament.3
脊柱的胸腰段区域被定义为T11～L2。由于处于相对固定的后凸脊柱与活动的前凸脊柱的转换区域，因此易于损伤。1931年，Jones描述了胸腰段的单纯的屈曲损伤，可以通过充分的过度屈曲实现解剖复位，并应用石膏保持复位状态。1963年，近30年后，Holdsworth一方面作为向Watson Jones致敬，另一方面也是由于对这些损伤治疗临床结果的不满意，他重新审视了胸腰段骨折。研究报道了他治疗超过1000例病人的治疗经验，随后1970年发表在JBJS杂志上。Holdsworth描述了胸腰段稳定性的“两柱”模型，强调了后侧韧带复合体的重要性，他定义后侧韧带复合体包括后侧关节面、棘间韧带、黄韧带和棘上韧带。
He asserted that a “burst” fracture was from a vertical compression force where the body is “shattered from within outward” and that it was always a stable injury by definition, thereby amenable to 3 months of casting.2 Holdsworth also described a burst variant that involved PLC disruption which rendered the fracture biomechanically unstable and required surgical stabilization. In 1983, Denis4 introduced the concept of the 3-column spine which emphasized the importance of the middle column consisting of the posterior half of the body, posterior annulus, and the posterior longitudinal ligament. The importance of the middle column rather than the posterior structures as championed by Holdsworth was based on biomechanical studies that demonstrated that the posterior longitudinal ligament as well as the posterior aspect of the annulus needed to be disrupted to create instability. Denis defined a burst fracture as failure of the anterior and middle columns under compression and defined instability as a middle column injury associated with either a PLC or anterior column injury. The mechanistic classification of TL fractures by Ferguson and Allen,5 published 1 year later, described injuries according to 3 anatomic regions and eschewed Denis’ “column” concept as anatomically and biomechanically incorrect. They determined that disruption of the posterior elements leads to instability, similar to Holdsworth’s conclusion. Of note, Ferguson and Allen were the first group to describe bony retropulsion and canal compromise as a prominent feature of TL burst fractures. Magerl et al,6 in a retrospective review of 1445 fractures, were the first to relate classification to prognosis in TL injuries. The mechanistic classification is very complex and based on the direction of force; compression, distraction, or rotation.
他评估“爆裂性”骨折是由于承受垂直压迫，椎体“向外侧粉碎”，为一种稳定性损伤，治疗采取石膏制动3个月。Holdsworth同时描述了合并后侧韧带复合体的爆裂性骨折，认为此类骨折存在生物力学不稳定，需要手术稳定。1983年，Denis介绍了三柱的概念，并强调了中柱的重要性。中柱的重要性大于后侧结构这一观点得到了Holdsworth的支持，他是基于生物力学研究显示，要导致不稳定，需要后纵韧带以及纤维环均发生破裂才行。Denis将爆裂性骨折定义为在压缩应力下前中柱破坏，将不稳定定义为中柱损伤合并后侧韧带复合体或者前柱损伤。1年后，Ferguson等提出胸腰段骨折机械性的分型，认为损伤划分3个解剖区域及Denis’ “柱”的概念在解剖学和生物力学上是不正确的。他们认为是后侧元件的破裂导致不稳定，这与Holdsworth’s的结论一致。Ferguson等首先提出骨块突入椎管导致占位是胸腰段爆裂性骨折的主要特征之一。Magerl等回顾了1445例骨折，首次将胸腰段损伤的进展同分类想联系。机械性的分型非常复杂，基于应力的方向，压缩、牵张或旋转。
It is well known that classification systems should provide a common language for clinicians, hold prognostic implications, and direct treatment. The fundamental basis of all classifications on TL burst fractures is the assessment and/or achievement of spinal stability which White and Panjabi7 define as the “ability to limit patterns of displacement so as not to damage or irritate the spinal cord or nerve roots and, in addition, to prevent incapacitating deformity or pain due to structural changes.” James et al8 conducted a landmark human cadaveric biomechanical study examining the relative contribution of the anterior, middle, and posterior columns to spinal stability. Disruptions of the columns progressed from anterior to posterior analogous to the direction of forces generated in a TL burst fracture. They determined that the posterior, not the middle column, was the key to resistance of flexion and kyphosis which agreed with the principles of Holdsworth, and Ferguson and Allen. As such, it is now well agreed upon that integrity of the PLC in the setting of a TL burst fracture is critical to biomechanical stability and is used as a major component for dictating operative or nonoperative treatment of these injuries in the Thoracolumbar Injury Classification System (TLICS) score.9,10
众所周知，分型系统应该提供有关临床、并发症、直接治疗方面的信息。胸腰段爆裂性骨折分型的主要基础是评估和/或达到脊柱稳定性，即White等定义为“限制移位以避免破坏或刺激脊髓或神经根，另外，防止由于结构改变导致的残废或疼痛”的能力。James主持的一项重要的人类尸体生物力学研究检验了前中后柱对于脊柱稳定性的关系。从前到后柱的破裂类似于产生于胸腰段爆裂性骨折的直接应力。他们确定，是后柱，而不是中柱，是预防屈曲和后凸畸形的关键，这一点复合Holdsworth和Ferguson的看法。这样，符合后侧韧带复合体的完整性在胸腰段爆裂性骨折的生物力学稳定性中是关键性，并且被用作一个主要的元素用以提示此类损伤的采用手术或非手术治疗，根据胸腰段损伤分类系统评分。

Principles of Treatment
治疗原则

The treatment of TL burst fractures is based on two primary principles, neurological optimization and biomechanical stability. Neurological optimization includes prevention, limitation, and possible reversal of neurological compromise by eliminating neural compression through stabilization of the spinal segment. An unstable TL burst fracture carries the risk of symptomatic posttraumatic deformity, delayed mobilization, and progressive neurological deterioration.
胸腰椎爆裂性骨折的治疗主要基于两个原则，神经“最优化”和生物力学稳定性。神经“最优化”包括预防、限制以及通过减压脊柱节段稳定达到神经功能逆转。不稳定性胸腰椎爆裂性骨折带来症状性脊柱后凸、卧床时间延长以及迟发性神经功能恶化的风险。

Nonoperative Treatment
非手术治疗

Nonoperative treatment consists of pain control, preventative care (pulmonary therapy, venous thromboembolism prophylaxis), and brace immobilization. Dating back to Watson Jones, fractures secondary to flexion-compression forces are reduced and maintained by hyperextension. Traditional casting has been largely supplanted with functional bracing. Bracing options include a cruciform anterior spinal hyperextension brace, Jewett hyperextension brace, or a custom-molded polypropylene thoracolumbosacral orthosis (TLSO). Despite the increased cost of a TLSO compared with “off-the-shelf” bracing options,11 our preference is a TLSO as it has shown to have superior control in all planes,12 is easy to don and doff, and is easy to clean. Our protocol involves 24 hours per day of TLSO treatment for 3 months with routine standing lateral radiographs at 3, 6, 9, and 12 weeks to assess for progressive kyphosis. Other risks of nonoperative management which need to be closely monitored include neurological decline, skin breakdown, respiratory or intra-abdom-inal restriction, and noncompliance. It should be noted that surgical treatment does not obviate the need for bracing because many surgeons elect to brace patients postoperatively. A recent systematic review by Giele et al13 failed to demonstrate conclusive evidence to support the use of braces in TL fractures.
非手术治疗包括疼痛控制、预防措施（肺功能治疗、预防静脉血栓栓塞）以及支具制动。回溯到Watson Jones，继发于屈曲-压缩应力的骨折已经减少了，而过伸伤则逐渐增多。传统的石膏已经逐渐为功能性支具所取代。这些支具包括十字形的前侧脊柱过伸支具、Jewett过伸支具或者定制的聚丙烯材料胸腰骶支具(TLSO)。尽管TLSO相对现有的支具价格高，但TLSO的各向稳定性均更好，且易于穿脱及清洗，因此，我们仍然推荐使用该支具。我们的策略包括24小时佩戴TLSO共3个月，在3, 6, 9和12周摄常规站立位片评估脊柱后凸畸形情况。非手术治疗的其他的风险包括神经功能恶化、皮肤破溃、呼吸及腹部受限以及不顺从。应该注意到手术治疗并不能代表可以不使用支具，因为很多外科医生术后仍然选择支具保护。Giele等最近的一篇系统性回顾中没有找到关于支持胸腰椎骨折使用支具的结论性的证据。

Operative Treatment
手术治疗

Traditionally, operative techniques for the treatment of TL burst fractures have included a combination of anterior and/or posterior decompression, followed by anterior and/or posterior stabilization. Direct neural decompression can be performed anteriorly through a corpectomy followed by cage or allograft strut placement and subsequently by an anterior rod/screw or plate construct. In our experience, this should be followed by posterior short-segment fixation in the setting of a PLC injury. It is our experience that an anterior corpectomy within 48 hours after injury is associated with significantly more blood loss and morbidity.14 Therefore, it is our preference to clear retropulsed fragments through a posterior extracavitary approach or indirectly through posterior reduction of the fracture. The fracture is then stabilized posteriorly through short-segment pedicular fixation. This is followed up 48-72 hours later by a staged anterior corpectomy and grafting for anterior column support. It is our preference to use a humeral shaft or iliac crest allograft. Many alternative constructs have been well described, including anterior corpectomy and fusion,15 posterior alone instrumentation, and kyphoplasty reduction and cementation for anterior column support.16 Interestingly, Dai et al17 recently reported on a randomized controlled trial of 73 patients with TL burst fractures treated with fusion or nonfusion with pedicle screw instrumentation and followed up for 5 years. There were no significant differences in radiographic or clinical outcomes between the 2 techniques. Both operative time and blood loss were significantly less in the nonfusion group compared with the fusion group.
一般来说胸腰椎爆裂性骨折的手术治疗包括前路和/或后路减压、前路和/或后路固定。可以通过椎体次全切除进行前路的神经减压，随后cage或自体骨块植骨以及前路棒/板固定。以我们的经验来看，如果同时合并后侧韧带复合体损伤，应该采用后路短节段固定。我们经验是，如果在急性损伤后48小时内进行前路椎体次全切除，将带来术中出血量明显增加以及病死率增加。因此，我们认为采用后侧入路清除占位骨块减压或通过后路间接骨折复位方式比较稳妥。骨折随后采用后路短节段椎弓根钉固定。48-72小时后，采用一期前路椎体次全切除植骨支撑。植骨宜采用肱骨干以及自体髂嵴。文献报道了多种治疗策略，包括前路椎体次全切除、单纯后路器械、椎体成型术复位、骨水泥注入加强前柱。值得注意的是，戴等最近报道对73个胸腰椎爆裂性骨折的病人采用后路椎弓根钉融合或不融合的方式，并随访5年。两组间在影像学及临床结果上没有发现明显的差别。与融合组相比，非融合组在手术时间以及出血量上明显减少。

Indications for Operative Intervention
手术干预的指征

The absolute indication for surgical treatment of a TL burst fracture includes a progressive neurological deficit in the setting of neural element compression, a fracture-dislocation or translational instability, or progressive symptomatic kyphosis. Other factors, such as greater than 50% canal compromise, injury to the posterior column or PLC, greater than 50% loss of vertebral body height (LOVBH), greater than 30° of kyphosis, or any neurological compromise, remain controversial with lack of level 1 evidence despite their common use in reference texts. For example, the often cited criteria of greater than 50% canal compromise and 30° of kyphosis is from a retrospective study by Willen et al18 of 54 patients with T12 or L1 fractures, treated conservatively over a 26-year period. Over half of these patients were neurologically compromised at time of presentation with incomplete (n = 34) or complete (n = 13) paraparesis. Using measurements from plain radiographs, the study concluded that patients with canal compromise or anterior column compression exceeding 50% had a greater incidence of intractable back pain, neurologic involvement, and signs of instability.
胸腰椎爆裂性骨折的绝对手术指征包括合并占位的进展性的神经功能障碍、骨折脱位或平移性的不稳定、进展性症状性的后凸畸形。其他指征包括大于50%得椎管占位、后柱或者后侧韧带复合体损伤、伤椎高度丢失大于50%、后凸畸形大于30°或者神经压迫。以上指征目前存在争议，尽管它们广泛出现在各种文献当中。例如，Willen等回顾了采用非手术疗法治疗54例椎管占位超过50%和后凸畸形大于30° 的T12或L1骨折长达26年以上的治疗结果（本文被其他作者多次引用）。其中超过一半的病例合并神经功能障碍（不全瘫34例、全瘫13例）。通过平片观察，该研究指出合并椎管占位或前柱压缩超过50%的病例发生难治性背痛、神经损伤和不稳定的风险大增。

Outcome Studies
结果研究

Canal Compromise
椎管占位

Canal compromise is typically calculated as a ratio of the canal area of the injured level to the average of the canal areas of the 2 adjacent uninjured segments.19 The correlation between spinal canal compromise and neurological injury, and therefore clinical outcome is intuitively attractive yet remains unproven. Many authors have attributed neural injury to the dynamic canal encroachment that occurs at the time of injury, with the resulting retropulsed fragments in a recoiled partially reduced resting position.20-22 In his original description of 59 burst fractures, Denis stated that a direct relationship between the degree of obstruction of the canal and neurological deficit could not be established.4 This has been echoed by other investigators who have found no correlation between degree of canal compromise and neurological injury.22 Dall and Stauffer23 reported on 14 consecutive patients with T12 or L1 burst fractures, greater than 30% canal compromise, and incomplete neurological injury and also found no correlation between amount of canal compromise and neurologic recovery. Limb et al22 studied 20 patients using plain radiographs and found no statistically significant correlation between bony retropulsion on lateral radiographs and severity of neurological injury but correlated canal compromise to the injury severity score and neurological Frankel grade. In a prospective, consecutive series of 43 patients with TL burst fractures, 30 of whom were neurologically intact, Vaccaro et al demonstrated no correlation between canal cross-sectional area and neurologic compromise. The authors also found that, unlike in the cervical spine,24,25 a larger preinjury cross-sectional area did not confer neuroprotection. They did, however, find a correlation between neurological injury and a smaller sagittal-to-transverse diameter ratio, higher absolute transverse diameter, and higher injury severity score20 Similiarly, Meves and Avanzi,26 in a retrospective study of 184 patients found a similar correlation with an average midsagittal diameter loss of 52% in those with neurologic injury and 34% in those who were neuro intact. The finding that a smaller sagittal to transverse diameter ratio and higher absolute transverse diameter correlates to neurological injury implies that the morphology of the canal rather than its absolute cross-sectional area may be more significant. Hashimoto et al27 characterized canal shapes into oval, semicircle, horseshoe, and crescent and found that the crescent type (which has a large transverse and small sagittal diameter) correlated with neurological injury. Longitudinal studies of nonoperative treatment of neurologically intact TL burst fractures by Mumford et al19 and Cantor et al28 demonstrate remodeling of the canal occupying retropulsed fragment with reconstitution of the spinal canal at 1 year by 66% and 50%, respectively. In the setting of neurological injury, Willen et al18 reported that in patients
with >50% canal compromise, fragments did not “appear to resorb” and correlated with increasing low back and radiculopathy. Although it is often cited as a cause for surgical intervention when there is >50% canal compromise, better clinical and/or neurological outcomes after surgery in patients with >50% canal compromise have not been reported. On the contrary, Kim et al29 demonstrated that there was no correlation between neurological recovery and canal clearance after surgery.
椎管占位一般是指伤椎椎管的横截面积与上下相邻两个脊柱节段的椎管横截面积平均数的比值。有关椎管占位与神经损伤之间相关性仍无可靠的结论。很多学者将神经损伤归因于损伤即刻椎管的动态压迫，当处于休息位置时有部分骨块可以回缩。Denis在对59例爆裂性骨折研究之后指出椎管占位的程度与神经损伤之间没有直接关系。这一观点得到了许多学者的支持。Dall和Stauffer报道了14例连续的椎管占位超过30%合并不全性损伤的T12或L1爆裂性骨折的病例，没有发现椎管占位与神经恢复之间有何必然联系。Limb通过平片研究了20例病人，没有发现骨块占位与神经损伤严重性之间有何关联，但椎管压缩同神经损伤分数以及神经Frankel分级之间存在关联。在一个前瞻性的连续的43例胸腰椎爆裂性骨折的研究中，30例神经功能完整，Vaccaro等没有发现在椎管横截面积以及神经压迫之间有何关联。作者同时发现伤前较大的椎管横断面积并不能提供更多的神经保护，这一点和颈椎不同。但他们发现神经损伤与较小的矢状位/水平位直径比值有关，横径越大，则相应的损伤严重分数越高。Meves和Avanzi回顾了184个病例，平均中央矢状位的丢失为52%，其中34%神经功能完好。较小的矢状位/水平位直径比以及较高的横径与神经损伤之间的关联提示椎管的形态可能较之横截面积更为重要。Hashimoto等将椎管的形状分为椭圆形、半圆形、马蹄形、新月形，并发现新月形（横径较大，矢状径较小）与神经损伤之间存在关联。Mumford等和Cantor等所作的纵向调查了神经功能完整的胸腰段爆裂性骨折的病人采用非手术治疗的情况后发现1年后椎管占位校调分别为66%和50%。对于神经损伤的病人，Willen等报道椎管占位大于50%、后凸骨块未显示吸收与背痛以及神经根病有关。尽管椎管占位大于50%常常作为手术指征被引用，但并没有此类病人治疗后去的更好的临床和/或神经功能结果的报道。相反，Kim等指出，神经功能恢复与椎管占位清除之间没有相关性。
In summary, a crescent-shaped canal with a large transverse and small sagittal diameter rather than an absolute reduction in cross-sectional area may correlate with neurological injury. In the absence of a neurological deficit, canal compromise should not be considered a factor supporting surgical intervention. In the setting of a neurological deficit, >50% canal compromise may lead to increased back pain and radiculopathy with nonoperative treatment. However, there are no studies that demonstrate improved outcomes in patients with >50% canal compromise, who undergo surgical decompression.
总之，椎管为新月形合并较小的矢状位/水平位直径比较之横截面积，与神经损伤之间更有相关性。对于没有神经损伤的病例，椎管占位不应视为手术干预的指征。对于超过50%的椎管占位的神经损伤的病例，采用非手术治疗可能增加背痛和神经根病的可能。但是，也没有确切的研究证实对此类病人采用手术干预能改善功能结果。

Loss of Vertebral Body Height
椎体高度丢失

LOVBH is described by Willen et al18 as the ratio of anterior height of the fractured body to the average of the adjacent, uninjured bodies. Cantor et al28 treated 18 neurologically intact patients who had >50% LOVBH with bracing and early ambulation for a mean of 19 months and found no significant increase in LOVBH over time. Of 18 patients, 15 reported little to no pain, and there were no late neurologic sequelae. Mumford et al19 reported on 41 neurologically intact patients with 2 year follow-up and found a progression of LOVBH from 39% to 47% with no statistical correlation of LOVBH to outcome. Similarly, Willen et al found that anterior column compression tended to increase after mobilization from an average of 37%-43% 6 months after injury, with little change after that time. However, patients with >50% LOVBH had increased complications and all 6 patients who underwent surgery for neurological impairment had >50% LOVBH. However, poor clinical outcomes in patients with >50% LOVBH have not been demonstrated. On the contrary, Weinstein et al30 reported the 20-year follow-up of nonoperative treatment of 83-TL burst fractures all of whom had >50% LOVBH and found that the average back pain score was 3 with 88% of patients returning back to work.
Willen等将椎体高度丢失定义为伤椎前方的高度与上下相邻椎体高度平均数的比值。Cantor等治疗了18例椎体高度丢失超过50%神经功能完好的病例采用支具保护早期下床活动，随访平均19个月，发现椎体高度丢失没有出现显著性的增加。18个病人中，有15个自述无疼痛或微小疼痛，未合并晚期神经后遗症。Mumford等报道了41例神经功能完好的病人随访2年，椎体高度丢失由39%增加至47%，但椎体高度的矫正与功能结果之间没有发现相关性。同时，Willen等发现从开始活动到伤后6个月前柱的压缩倾向于增加37%-43%，之后就没有显著的变化了。但是，椎体高度丢失超过50%的病例的并发症增加，其中6例合并神经损伤采用了手术治疗。不过，文章并没有指出椎体高度丢失大于50%的病人临床结果不好。相反，Weinstein等报道了非手术治疗83例椎体高度丢失大于50%的病人随访20年的结果后指出重返工作的病人中88%平均背痛分数仅为3分。

Kyphosis
后凸畸形

Kyphosis is measured using the Cobb technique as the local sagittal angle or the regional sagittal angle. The local sagittal angle is the angle between the superior end plate and inferior end plate of the injured vertebra. The regional sagittal angle is the angle between the superior end plate of the uninjured cephalad vertebra and the inferior end plate of the uninjured caudal vertebra.18 Severe kyphosis associated with TL burst fractures has traditionally been used as a justification for operative intervention. It is critical to understand the evidence behind whether the kyphotic deformity increases over time and more importantly, if it is associated with poor clinical outcomes.
后凸畸形采用Cobb技术测量局部节段性角度或区域节段性角度。局部节段性角度位于伤椎上位终板和下位终板之间。区域性节段性角度为伤椎上一椎体上终板与伤椎下一椎体下终板之间的夹角。胸腰段爆裂性骨折合并严重的后凸畸形一般采用手术治疗。需理解后凸畸形增加的证据，更重要的是是否和不好的临床结果相关联。
Willen et al followed up 54 patients for 2 years after a TL burst fracture and found that kyphosis increased from 14° to 20° over the first 6 months, and then remained stable thereafter. It is important to note that the status of the PLC in these patients was unknown.18 Cantor et al28 excluded patients with kyphosis <30° and reported that average kyphosis increased from 19° to 20° at 19 month follow-up. Most of the patients reported little or no pain and no loss of function. Other studies have also reported no association between kyphosis and long-term outcomes.19,30 Mumford et al19 followed up 41 patients and found that even severe kyphosis (>30°) resulted in good outcomes. Therefore, although severe kyphosis (>30°) has been historically used as a relative indication for reduction and internal fixation, the link between high-grade kyphosis and the risk of a delayed neurological deficit or adverse outcome is not evidence based.
Willen等对54例胸腰段爆裂性骨折的病人进行了2年随访，发现后凸畸形在最初的6个月内从14° 进展至20°，之后则保持稳定了。不过他列出的病人的后侧韧带复合体的情况不清楚。Cantor等排除了病人后凸畸形小于30°以及平均后凸畸形在随访中从19°进展至20°。多数病人自述无背痛或轻度背痛，无功能障碍。其他的研究指出后凸畸形与长期功能结果之间并没有相关性。Mumford等随访了41例病人，发现甚至严重的后凸畸形的病人也可能有好的功能结果。因此，尽管严重的后凸畸形作为手术干预的一项相对指征，但并没有证据显示高度的后凸畸形与延迟性的神经功能障碍以及不好的功能结果之间有何明显关联。

Neurological Compromise
神经压迫

Neurological deficits are detected in 10%-38% of TL burst fractures and can include injury to the spinal cord, conus medullaris, cauda equina, or an isolated nerve root. Delamarter et al31 demonstrated in a canine model that the recovery of somatosensory-evoked potentials at 6 weeks after spinal cord injury was directly related to the duration of spinal cord compression. Although this model is relevant to spinal cord injury and less so for injury to the conus medullaris, cauda equina, or nerve root, these data have often been extrapolated to the clinical situation which has led to the dogma that any progressive neurological deficit is an absolute indication for prompt surgical reduction, decompression, and internal fixation. In a large clinical series using Scoliosis Research Society member collaboration, Gertz-bein32 reported on the prospective outcome of more than 1000 patients with spinal fractures, with 81% having some neurological deficit. The only scenario in which there was an improved neurological outcome with surgical decompression was in the setting of a progressive neurological deficit while under clinical observation.
胸腰段爆裂性骨折神经损伤的发生率为10%-38%，包括脊髓、马尾、圆锥和神经根的损伤。Delamarter等指出山羊脊髓损伤模型6周时体感诱发电位与脊髓压迫的持续时间直接相关。尽管这一实验的对象是脊髓，和马尾、圆锥和神经根的损伤不太相关，但有关的数据还是被应用于临床判定，即对于任何进展性的神经功能损伤均为积极手术复位、减压和内固定的绝对手术指征。Gertz-bein等报道了超过1000例脊柱骨折的病例的前瞻性的结果，其中81%有不同程度的神经损伤。临床观察，对进行性神经损伤的部位进行手术减压改善了神经功能。
The treatment of a nonprogressive neurological deficit in the setting of a TL burst fracture is controversial. Hartman et al33 reported on the nonoperative treatment of 32 patients with TL burst fractures, of whom 9 had incomplete and 3 had complete neurological injuries. All the patients with a neurological injury improved 1 neurological Frankel grade. Gertzbein found that after excluding American Spinal Injury Association (ASIA) A and E patients there was a similar rate of neurological improvement regardless of operative or nonoperative treatment at 2 years. These studies demonstrate that despite a neurological injury, nonoperative treatment of a stable injury leads to at least 1 ASIA grade improvement with little data to support decompression of a nonprogressive neurological deficit. However, it is our belief that a neurological picture of conus medullaris or cauda equina injury in the setting of a large compressive bony fragment is analogous to cauda equina syndrome in the setting of an acutely herniated disc.34 In this particular situation, the spinal canal should be decompressed within 48 hours regardless of whether the deficit is progressive to allow for maximal neurological recovery.
对于非进行性神经损伤的治疗仍然存在争议。Hartman等报道对32例胸腰段爆裂性骨折的病人采用非手术治疗，其中9例为不完全性损伤，3例为完全性损伤。所有病人的神经损伤改善了1级。Gertzbein发现除ASIA分级A级和E级的患者以外，手术治疗和非手术治疗2年对神经功能的改善作用是相似的。这些研究提示有神经损伤，稳定性损伤的非手术治疗能至少带来ASIA分级1级的改善，少有资料支持对非进展性神经损伤的病例采用减压处理。不过，后凸的骨块对脊髓马尾和圆锥的压迫类似于疝出的椎间盘所导致的马尾综合征。在这种特殊的情况下，我们认为，应该在伤后48小时内进行急诊减压以挽救神经功能。

Operative versus Nonoperative Treatment of TL Burst Fractures
胸腰段爆裂性骨折的手术与非手术治疗

In 1984, Gaines and Humphreys35 called for an improved scientific evaluation and rationale for the treatment of TL injuries stating “spinal surgeons must be aware of the morbidity that is potentially provided by even the most carefully performed spinal instrumentation.” This is particularly true for the management of TL burst fractures in patients without a neurological deficit. Thomas et al36 performed a systematic review studies investigating operative versus nonoperative treatment of TL burst fractures in neurologically intact patients and identified 7 studies as “very good” representing 277 patients and an additional 14 studies as “good” representing 287 patients. They concluded that “there is a lack of evidence demonstrating the superiority of one approach over the other as measured using generic and disease-specific health-related quality of life scales.” In a 2008 Cochrane review, Yi et al37 identified 3 adequate trials for review. The studies by Shen et al,38 Wood et al,39 and Siebenga et al40 constitute the best level of evidence for operative versus nonoperative treatment of TL burst fractures in neurologically intact patients published in the last decade. All the studies are randomized control trials with a minimum of 2 years follow-up. These studies will be critically reviewed in regard to the significance of associated PLC injury, kyphosis, clinical outcome scores, and complications related to treatment.
1984年，Gaines和Humphreys要求改进治疗胸腰段损伤的科学的评估和原理，“脊柱外科医生必须意识到即使是最小心的脊柱器械操作也可能导致潜在的致病率”。这对于未合并神经损伤的病人来说是非常关键的。Thomas等完成了一项系统性的回顾，调查了胸腰段手术与非手术治疗胸腰段爆裂性骨折未合并神经损伤的“非常好”的研究共7项277个病人，14项“好的”研究共287个病人。他们的结论是“缺乏证据证明某种治疗在特异性健康相关生活质量指标评分上优于其他方法”。2008年的一项系统性回顾中，Yi等确定了3项研究纳入回顾。这3项研究是最近十年中有关胸腰段爆裂性骨折未合并神经损伤手术治疗或非手术治疗最佳的证据。所有的研究均为随机对照研究，且至少随访2年时间。这些研究都严格根据后侧韧带复合体、后凸畸形、临床结果分数和治疗相关并发症的显著性来进行回顾。
The importance of the PLC has been extensively discussed in this article and clearly plays a role in the biomechanical stability of a TL burst fracture. The status of the PLC was not reported in the studies by Shen et al38 and Siebenga et al.40 However, Wood et al39 specifically excluded patients with PLC injuries with an average of 11° of kyphosis at the time of admission implying that the injuries were biomechanically stable. As such, the outcomes of operative versus nonoperative treatment of the subset of patients with significant PLC injuries remain unknown.
后侧韧带复合体的重要性已经在文献中得到了充分了讨论，该复合体在胸腰段爆裂性骨折的生物力学稳定性中扮演着一定的角色。后侧韧带复合体的情况没有被纳入上3项研究中的2项，但是另一项研究专门排除了入院时合并后侧韧带复合体损伤合（平均11°后凸畸形）的病例，默认这类损伤在生物力学上是稳定的。因此，手术治疗和非手术治疗合并严重后侧韧带复合体损伤的病人的结果仍然不明了。
The range of kyphosis across the studies at the time of injury in both nonoperative and operative groups was 10° to 23° (Table 1). Nonoperative treatment led to an increase in kyphosis in all 3 studies with the greatest amount of increase in the study by Siebenga et al 13.1° to 19.5° at final follow-up, consistent with the other nonrandomized observational studies previously discussed in this article. Operative treatment led to a reduction in kyphosis angle in all the studies except for a study by Wood et al which showed a 3° increase in kyphosis after surgical treatment. It can be gathered from these studies that surgical treatment likely decreases or halts progression of increasing kyphosis. However, the studies also demonstrate that the degree of final kyphosis does not correlate with increasing pain or a worse functional outcome. This agrees with multiple prior studies that clinical outcomes are independent of residual kyphosis.30,41-43

以上手术治疗和非手术治疗后凸畸形的范围是10° ~23°。在所有3项研究中，采用非手术治疗均导致后凸畸形增加，与之前的非随机观察研究的结果相似。所有的研究均显示手术治疗后后凸畸形矫正，除了Wood报道的采用手术治疗后后凸畸形增加3度。从以上研究可以得出结论，手术治疗倾向于减少或停止后凸畸形的进展。但是，这些研究也显示最终后凸畸形的度数与疼痛增加以及功能恶化没有必然的联系。这符合之前的研究，即临床结果独立于后凸畸形。
All the studies recorded pain visual analog scale (VAS), with none of the studies demonstrating a difference in pain at final follow-up between the operative and nonoperative groups. The functional outcome scores collected included the Greenough low back score (Shen), Oswestry Disability Index (Wood), SF-36 (Wood), and Roland Morris Disability Questionnaire (RMDQ-24) disability score (Siebenga). As seen in Table 2, at final follow-up there were no differences in functional outcome between the operatively and nonoperatively treated groups. Interestingly, Wood et al demonstrated an increase in SF-36 physical function score with nonoperative treatment. The percentage of patients returning to work was examined in all 3 studies. Wood et al showed that significantly more patients returned to work in the nonoperative group at 6 months.39 However, longer-term follow-up by Siebenga et al40 at 4.3 years demonstrated more patients in the operative group being able to return to work. The discrepancy between these 2 results is most likely due to the differences in length of follow-up between the 2 studies.

所有的研究均记录了疼痛可视评分，没有研究显示手术和非手术病人最终随访时在疼痛方面有何差异。功能结果分数的选择包括Greenough腰痛评分、Oswestry残障指数、SF-36和Roland Morris残障量表。正如表2所示，手术与非手术治疗最终的随访结果没有显著性的差异。值得注意的是，Wood等指出非手术患者SF-36躯体功能评分更高。所有3项研究均提及病人重返工作的情况。Wood等显示非手术治疗病人在6个月时重返工作的比例明显较高。但是Siebenga所作的超过4.3年的长期随访提示手术治疗的病人重返工作岗位的更多。两项研究结果的偏差很可能是由于随访时间的差异所造成的。
In all the studies, there were more complications in the operatively treated group. This was particularly significant in the study by Wood et al, with 79% in the operative group and 9% in the nonoperative group. Complications in the operative group included wound dehiscence, instrumentation breakdown, instrumentation removal, etc.
在所有的研究中均显示手术治疗并发症增多。这一点在Wood等的研究中尤其显著，手术组为79%，而非手术组为9%。手术组并发症包括伤口裂开、器械失效、器械脱出等。
The cost of operative intervention, particularly in the United States, is significant when compared with nonopera-tive bracing (Table 3). A study by Wood et al, the only study performed in the United States, reported the cost of operative treatment to be $49,063 versus $11,264 in the nonoperative group.39 Notably, despite being conducted in a different country, Shen, et al also demonstrated that the cost of operative treatment was 4 times that of nonoperative treatment.38 On the basis of these studies, it is evident that in the absence of a neurological deficit and definitive PLC injury, TL burst fractures should be treated nonoperatively in a TLSO brace. In an effort to guide surgical management of TL burst fractures, Vaccaro et al10 developed the Thoracolumbar Injury Classification Score.The TLICS is based on the expert opinions of 40 spine surgeons from 15 institutions. It is a point system based on injury morphology, PLC integrity, and neurological status. Any score >5 drives operative treatment. The TLICS classification, despite being reliable between surgeons, still needs to be validated in a clinical setting that demonstrates that those patients who undergo surgical treatment have a better functional outcome.

手术干预的费用，特别是在美国，明显高于非手术治疗。Wood等所作的研究，唯一一项在美国完成的研究，报道手术治疗的费用为$49,063，同时非手术治疗的费用为$11,264。值得注意的是，尽管国别不同，Shen等的研究指出手术治疗的费用是非手术治疗的4倍。在这些研究的基础上，对于缺乏神经损伤和确定性后侧韧带复合体损伤的证据的胸腰段爆裂性骨折应采用支具保守治疗。Vaccaro为指导胸腰段爆裂性骨折的手术处理，发展了胸腰段损伤分型评分。该评分基于15个中心的40名脊柱外科医生的经验，是一个基于损伤形态学、后侧韧带复合体完整性和神经功能状态的评分系统。任何大于5分的病例均应考虑手术治疗。尽管该系统适用于外科医生，但仍需要临床验证以证明接受手术治疗的病人能取得更好的功能结果。

Conclusions
结论

A critical, best evidence review of the available published data places the often cited surgical triggers of 50% canal compromise, 50% LOVBH, and 30° of kyphosis into question. In the neurologically intact patient without PLC injury, canal compromise, LOVBH, and kyphosis are not predictive of outcome and should not be used as indications for surgery. Three prospective, randomized control trials have demonstrated increased complications with operative treatment compared with nonoperative treatment with no difference in clinical outcomes, including VAS pain, functional outcome, and return to work status. In the patient with a neurological injury, nonoperative treatment has demonstrated at least 1 ASIA grade in the setting of a nonprogressive neurological deficit. Operative treatment has only been shown to be beneficial in the setting of progressive neurological deficits. It is our opinion that patients who demonstrate symptoms of conus medullaris or cauda equina syndrome with a compressive lesion should be decompressed within 48 hours regardless of whether there is a progressive neurological deficit. Biomechanical studies have punctuated the importance of PLC integrity to maintain stability and avoid progressive, post-traumatic kyphosis. Consideration of PLC integrity and neurological status are key drivers for the surgical treatment of TL burst fractures.
一项严格的最优的证据性回顾质疑了目前常常用到的诸如50%椎管占位、50%椎体高度丢失、30度的后凸畸形等手术指征。对于未合并后侧韧带复合体及神经损伤的病例，椎管占位、椎体高度及后凸畸形并非预测结果的指标，不应用于手术干预的指征。3项前瞻性的随机对照控制实验显示手术治疗同非手术治疗相比，临床结果相似，而并发症增多。对于合并神经损伤的病例，非手术治疗也显示了至少ASIA分级1级的改善。手术治疗仅仅对于进展性的神经损伤病例显示出优越性。我们的经验认为对于合并脊髓及马尾综合症且存在骨块压迫的病人，无论是否存在进展性的神经损伤，均应在伤后48小时内进行减压。生物力学研究强调后侧韧带复合体完整性对于保持稳定和避免进展性创伤后后凸畸形的重要性。因此，后侧韧带复合体完整性以及神经功能状态是胸腰段爆裂性骨折采取手术治疗的首要考虑因素。

References
1.Jones RW: Manipulative reduction of crush fractures of the spine. Br Med J 1:300-302, 1931
2.Holdsworth FW: Fractures, dislocations, and fracture-dislocations of the spine. J Bone Joint Surg Am 52:1534-1551, 1970
3.Holdsworth FW: Fractures, dislocations, and fracture-dislocations of the spine. J Bone Joint Surg Br 45:6-20, 1963
4.Denis F: The three column spine and its significance in the classification of acute thoracolumbar spinal injuries. Spine 8:817-831, 1983
5.Ferguson RL, Allen BL Jr: A mechanistic classification of thoracolumbar spine fractures. Clin Orthop Relat Res 189:77-88, 1984
6.Magerl F, Aebi M, Gertzbein SD, et al: A comprehensive classification of thoracic and lumbar injuries. Eur Spine J 3:184-201, 1994
7.White AA, Panjabi M: Clinical biomechanics of the spine. Philadelphia, PA, Lippincott, 1978
8.James KS, Wenger KH, Schlegel JD, et al: Biomechanical evaluation of the stability of thoracolumbar burst fractures. Spine 19:1731-1740, 1994
9.Vaccaro AR, Baron EM, Sanfilippo J, et al: Reliability of a novel classification system for thoracolumbar injuries: the thoracolumbar injury severity score. Spine 31:S62-S69, 2006; discussion S104
10.Vaccaro AR, Zeiller SC, Hulbert RJ, et al: The thoracolumbar injury severity score: a proposed treatment algorithm. J Spinal Disord Tech 18:209-215, 2005
11.Spinal Orthotics. eMedicine, 2008. Available at: http://emedicine.medscape. com/article/314921-overview
12.Axelsson P, Johnsson R, Stromqvist B: Effect of lumbar orthosis on intervertebral mobility. A roentgen stereophotogrammetric analysis. Spine 17:678-681, 1992
13.Giele BM, Wiertsema SH, Beelen A, et al: No evidence for the effectiveness of bracing in patients with thoracolumbar fractures. Acta Orthop 80:226-232, 2009
14.Wood KB, Bohn D, Mehbod A: Anterior versus posterior treatment of stable thoracolumbar burst fractures without neurologic deficit: a prospective, randomized study. J Spinal Disord Tech 18:S15-S23, 2005 (suppl)
15.Ghanayem AJ, Zdeblick TA: Anterior instrumentation in the management of thoracolumbar burst fractures. Clin Orthop Relat Res 335:89-100, 1997
16.Marco RA, Kushwaha VP: Thoracolumbar burst fractures treated with posterior decompression and pedicle screw instrumentation supplemented with balloon-assisted vertebroplasty and calcium phosphate reconstruction. J Bone Joint Surg Am 91:20-28, 2009
17.Dai LY, Jiang LS, Jiang SD: Posterior short-segment fixation with or without fusion for thoracolumbar burst fractures. A five to seven-year prospective randomized study. J Bone Joint Surg Am 91:1033-1041, 2009
18.Willen J, Anderson J, Toomoka K, et al: The natural history of burst fractures at the thoracolumbar junction. J Spinal Disord 3:39-46, 1990
19.Mumford J, Weinstein JN, Spratt KF, et al: Thoracolumbar burst fractures. The clinical efficacy and outcome of nonoperative management. Spine 18:955-970, 1993
20.Vaccaro AR, Nachwalter RS, Klein GR, et al: The significance of thoracolumbar spinal canal size in spinal cord injury patients. Spine 26:371-376, 2001
21.Wilcox RK, Boerger TO, Hall RM, et al: Measurement ofcanal occlusion during the thoracolumbar burst fracture process. J Biomech 35:381-384, 2002
22.Limb D, Shaw DL, Dickson RA: Neurological injury in thoracolumbar burst fractures. J Bone Joint Surg Br 77:774-777, 1995
23.Dall BE, Stauffer ES: Neurologic injury and recovery patterns in burst fractures at the T12 or L1 motion segment. Clin Orthop Relat Res 233:171-176, 1988
24.Kang JD, Figgie MP, Bohlman HH: Sagittal measurements of the cervical spine in subaxial fractures and dislocations. An analysis of two hundred and eighty-eight patients with and without neurological deficits. J Bone Joint Surg Am 76:1617-1628, 1994
25.Matsuura P, Waters RL, Adkins RH, et al: Comparison of computerized tomography parameters of the cervical spine in normal control subjects and spinal cord-injured patients. J Bone Joint Surg Am 71:183-188, 1989
26.Meves R, Avanzi O: Correlation among canal compromise, neurologic deficit, and injury severity in thoracolumbar burst fractures. Spine 31:2137-2141, 2006
27.Hashimoto T, Kaneda K, Abumi K: Relationship between traumatic spinal canal stenosis and neurologic deficits in thoracolumbar burst fractures. Spine 13:1268-1272, 1988
28.Cantor JB, Lebwohl NH, Garvey T, et al: Nonoperative management of stable thoracolumbar burst fractures with early ambulation and bracing. Spine 18:971-976, 1993
29.Kim NH, Lee HM, Chun IM: Neurologic injury and recovery in patients with burst fracture of the thoracolumbar spine. Spine 24:290-293, 1999; discussion 294
30.Weinstein JN, Collalto P, Lehmann TR: Thoracolumbar “burst” fractures treated conservatively: a long-term follow-up. Spine 13:33-38, 1988
31.Delamarter RB, Sherman J, Carr JB: Pathophysiology of spinal cord injury. Recovery after immediate and delayed decompression. J Bone Joint Surg Am 77:1042-1049, 1995
32.Gertzbein SD: Scoliosis Research Society. Multicenter spine fracture study. Spine 17:528-540, 1992
33.Hartman MB, Chrin AM, Rechtine GR: Non-operative treatment of thoracolumbar fractures. Paraplegia 33:73-76, 1995
34.Kostuik JP, Harrington I, Alexander D, et al: Cauda equina syndrome and lumbar disc herniation. J Bone Joint Surg Am 68:386-391, 1986
35.Gaines RW, Humphreys WG: A plea for judgment in management of thoracolumbar fractures and fracture-dislocations. A reassessment of surgical indications. Clin Orthop Relat Res 189:36-42, 1984
36.Thomas KC, Bailey CS, Dvorak MF, et al: Comparison of operative and nonoperative treatment for thoracolumbar burst fractures in patients without neurological deficit: a systematic review. J Neurosurg Spine 4:351-358, 2006
37.Yi L, Jingping B, Gele J, et al: Operative versus non-operative treatment for thoracolumbar burst fractures without neurological deficit. Cochrane Database Syst Rev 4:CD005079, 2006
38.Shen WJ, Liu TJ, Shen YS: Nonoperative treatment versus posterior fixation for thoracolumbar junction burst fractures without neurologic deficit. Spine 26:1038-1045, 2001
39.Wood K, Buttermann G, Mehbod A, et al: Operative compared with nonoperative treatment of a thoracolumbar burst fracture without neurological deficit. A prospective, randomized study. J Bone Joint Surg Am 85:773-781, 2003
40.Siebenga J, Leferink VJ, Segers MJ, et al: Treatment of traumatic thoracolumbar spine fractures: a multicenter prospective randomized study of operative versus nonsurgical treatment. Spine 31:2881-2890, 2006
41.Domenicucci M, Preite R, Ramieri A, et al: Thoracolumbar fractures without neurosurgical involvement: surgical or conservative treatment? J Neurosurg Sci 40:1-10, 1996
42.Hazel WA Jr, Jones RA, Morrey BF, et al: Vertebral fractures without neurological deficit. A long-term follow-up study. J Bone Joint Surg Am 70:1319-1321, 1988
43.Kraemer WJ, Schemitsch EH, Lever J, et al: Functional outcome of thoracolumbar burst fractures without neurological deficit. J Orthop Trauma 10:541-544, 1996