Posterior short-segment instrumentation for thoracolumbar burst fracture is known for a high implant failure rate because of the lack of anterior support. Anterior body augmentation by transpedicular bone grafting has been developed as an alternative to overcome this failure. However, the efficacy of transpedicular bone grafting remains debatable.
Between August 2002 and August 2006, 31 patients with a single-level thoracolumbar fracture underwent insertion of posterior short-segment pedicle screws and transpedicular bone grafting. Twenty-one men and 10 women had a mean age of 39.7 years at the time of surgery. All patients were followed up for at least 2 years; the mean follow-up period was 52.7 months. Preoperative computed tomography showed that the mean canal encroachment was 48.1%. The kyphotic angle improved from 20.9° to 3.7° immediately postoperatively. Loss of kyphosis correction was 2.7° before implant removal and 6.2°; at final follow-up. The anterior body height was corrected from 50.9% to 86.9% by surgery, but collapsed to 82.2% before implant removal and became 78.1% at final follow-up. The failure of the surgery was defined as an increase of >10° in local kyphosis and/or implant breakage. At last follow-up, no implants had failed, but 3 patients had a loss of correction >10°; the failure rate was 9.7%. According to the Dennis functional scales, the mean pain score was 1.7, and the mean work score was 2.2 at final follow-up. All 8 patients with partial neurological deficit initially had improvement.
Analysis of the results indicated that this technique effectively corrects deformities, restores vertebral height, prevents early implant failure, and provides satisfactory clinical results.
Fractures of the spine frequently occur in the thoracolumbar region. Burst fracture is a common thoracolumbar fracture that results from the failure of the anterior and middle column by axial loading. It remains controversial whether a surgical or nonsurgical approach is the optimal management strategy for thoracolumbar burst fracture.1 Surgical treatment is generally recommended for patients with neurologic deficits2 or in those with severe instability, defined as a loss of anterior vertebral body height exceeding 50%, angulation exceeding 20°, or canal compromise exceeding 50%.3-5
Posterior instrumentation enables early correction of kyphosis and indirect reduction of the canal by ligamentotaxis. Currently, posterior short-segment instrumentation and short fusion is a popular procedure for treating unstable thoracolumbar burst fracture. Although the clinical results of this surgery are usually satisfactory, progressive kyphosis and a high failure rate of the pedicle screws remain a concern. Failure of the anterior column support is the main cause of hardware failure.6 Daniaux7 introduced transpedicular bone grafting to solve these problems. The efficacy of this method remains debatable. Transpedicular grafting to prevent implant failure showed positive results in earlier studies8,9; however, the results of several recent studies on additional transpedicular bone grafting for the thoracolumbar burst fracture were disappointing.5,10
We retrospectively evaluated the clinical and radiographic outcomes of posterior short-segment pedicle screws with additional transpedicular bone grafting as treatment for patients with acute thoracolumbar burst fractures to determine whether this technique remains a reliable surgical method for patients with thoracolumbar burst fracture.
Materials and Methods
Between August 2002 and August 2006, 38 patients with thoracolumbar fractures underwent insertion of posterior short-segment pedicle screws and transpedicular bone grafting at our institution. Thirty-one of these patients were enrolled into our retrospective study based on the following inclusion criteria: (1) single-level fracture; (2) fracture caused by high-energy trauma (fall from a height or motor vehicle accident); (3) type A3 burst fracture according to Magerl classification 11; (4) unstable burst fracture (ie, local kyphotic angle >20°, or anterior body height collapse >50%, or spinal canal encroachment >50%); and (5) at least 2 years follow-up with radiographic and clinical data.
Preoperative and final neurological impairment were evaluated using the American Spinal Injury Association (ASIA) Impairment Scale. The clinical results were assessed at final follow-up using the scale developed by Denis et al,12 a 5-point scale used to evaluate pain and work status. Pain is ranked from no pain (P1) to constant and incapacitating pain requiring chronic medication (P5). Work status is ranked from return to previous labor (W1) to completely disabled (W5).
Preoperative computed tomography (CT) of the spine was used to evaluate the degree of canal encroachment by the fractured fragment. Plain radiographs were obtained preoperatively, immediately postoperatively, before instrumentation removal, and at final follow-up. Sagittal local kyphosis was measured from the superior endplate of the cephalic intact vertebra to the inferior endplate of the caudal intact vertebra. The formulas adopted by Mumford et al13 were used to calculate the percentage of anterior body height and the percentage of canal compromise. The surgery was considered a failure if the implant failed or if radiographs obtained at last follow-up showed an increase of >10° in sagittal kyphosis compared to local kyphosis angle measured immediately postoperatively in radiographs.5
Other data collected included age, sex, duration of surgery, estimated blood loss, duration of admission, time between injury and surgery, associated injuries, and complications.
All patients were placed in the prone position. A standard posterior midline approach was used to explore the spine. Pedicle screws were inserted into the vertebra 1 level above and below the fractured vertebra. The manual lordotic maneuver was applied first. After connecting the rods and screws, distraction force was applied using spreader forceps to restore lordosis and anterior body height. A 2-cm wound was created over the posterior iliac spine to harvest 5-cm3 iliac bone grafts. A trocar was used as a pedicle finder to create a transpedicular pathway within the defect of the fractured vertebra. This transpedicular pathway was dilated simultaneously with different pedicle screw sizes.
We developed special tools, including a metallic funnel and a metallic mallet, for insertion of bone graft (Figure 1). A 6.5- or 7.0-mm pathway is enough for insertion of the metallic funnel. At the same time, an assistant morselized the harvested bone grafts. The morselized cancellous bone graft was placed in the metallic funnel and pushed into the defect of the fractured vertebra using a metallic mallet. The defect of the fractured vertebra was impacted with morselized cancellous bone graft gradually.
| || |
Figure 1: Tools for transpedicular bone grafting (A). These tools include a metallic funnel (right) and a metallic mallet (left) (B). The metallic mallet can insert into the metallic funnel to push bone graft into vertebra body.
All surgical procedures were monitored under a C-arm fluoroscopy (Figure 2). Posterior or posterolateral fusion with the bone graft was not performed. None of the patients underwent a laminectomy procedure.
Patients were encouraged to sit upright on the second postoperative day and begin rehabilitation for preparation of ambulation. All patients were protected using a Taylor brace for 3 months.
Data were analyzed using the SPSS statistical software package (SPSS, Chicago, Illinois). Radiographic parameters were compared preoperatively, postoperatively, and at final follow-up using paired t tests with a confidence interval of 95%.
The study cohort comprised 31 patients, 21 men and 10 women, with a mean age of 39.7 years at the time of surgery (range, 19-66 years). All patients were followed up for at least 2 years; the mean follow-up period was 52.7 months (range, 24-72 months). Most injuries were caused by a fall from a height (19/31). Most fractures occurred at the L1 vertebra (12/31). Twenty-three patients had intact neurologic status, and the remaining 8 patients had partial neurologic deficits preoperatively.
Twelve patients had associated injuries: 1 head injury, 1 intra-abdominal injury, 2 rib fractures with hemothorax, 3 clavicle fractures, 3 pelvic fractures and/or acetabulum fractures, and 5 calcaneal and/or foot bone fractures. The mean duration of surgery was 227.2 minutes (range, 144-310 minutes), and the estimated mean blood loss was 600.0 cc (range, 100-1900 cc.). The mean duration of hospitalization was 13.6 days (range, 6-21 days), and the mean interval between injury and surgery was 3.5 days (range, 1-8 days).
Posterior instrumentation included RF (Advanced Spinal Technology, Oakland, California) in 16 patients and Prot (Merris International, Taipei, Taiwan) in 15 patients. Posterior instrumentation was surgically removed from all patients between 12 and 18 months after initial surgery, and no implant failure occurred. Two patients had spinal complications following surgery: 1 had a superficial wound infection that responded to intravenous antibiotic treatment, and the other had a donor-site infection that was successfully treated by surgical debridement with intravenous antibiotics. Table 1 shows the detailed clinical characteristics of the patients.
The mean preoperative spinal canal encroachment as determined using CT was 48.1% (range, 10-80%). The mean preoperative kyphotic angle was 20.9° (range, 2°-37°), which was corrected to 3.7° (range, 15°-18°) immediately postoperatively (P <.001) This is a correction of 17.2°. The mean Cobb angle changed to 6.4° (range, 11°-19°) before implant removal and became a statistically significant loss of correction to 9.9° (range, 8°-29°) at final follow-up (P <.001). Loss of kyphosis correction was 2.7° before implant removal and 6.2° at final follow-up. There was still a statistically significant 11° mean correction from injury to final follow-up (P <.001).
Eighty-four percent of the initial surgical reduction was maintained before hardware removal, and only 60% of the beginning correction was preserved at final follow-up. No implants had failed, but 3 patients had a loss of correction >10°. The failure rate was 9.7% (3/31 patients). The mean preoperative anterior body height was 50.9% (range, 33-75%), which improved to 86.9% (range, 67-99%) immediately postoperatively (P <.001). Surgery restored the anterior body height by 31.0%. However, the anterior body height became 82.2% (range, 68-97%) before implant removal and collapsed significantly to 78.1% (range, 53-96%) at final follow-up (P <.001). Thus, the average loss of body height correction was 8.8%. From injury to final follow-up, there was a statistically significant 27.7% mean correction (P <.001). Table 2 shows the patients radiographic data.
According to the Dennis functional scales, the mean pain score was 1.7 (range, 1-4), and the mean work score was 2.2 (range, 1-5) at final follow-up. According to the ASIA neurologic grading system, 8 and 23 patients had grades D and E, respectively, preoperatively. Surgery did not exacerbate neurological impairment. All patients with grade D impairment preoperatively had improved to grade E at final follow-up. The neurological and functional findings for all patients are shown in Table 3.
Thoracolumbar burst fracture is a common spinal osseous injury that usually occurs after a motor vehicle accident or following a fall from a height. The advantage of surgical intervention is immediate stabilization of the injured spine and decompression of the spinal canal. In addition, surgery enables early mobilization without an uncomfortable cast and shortens hospitalization.
It remains controversial whether anterior or posterior surgery is the most effective treatment of burst fracture with neurologic deficits. Posterior surgery clears the spinal canal by ligamentotaxis but may leave behind some retropulsed fragments. Some studies show little correlation between spinal encroachment and neurologic recovery.14,15 Despite this, some authors have advised anterior surgery to remove retropulsed fragments.16,17 The principle at our institution is to perform anterior surgery for patients with moderate to severe neurological deficits (ASIA grades A, B, or C); this is why all patients in this study were neurologically intact or had minor neurological deficits (ASIA grades D and E) preoperatively. Full neurologic recovery was seen at final follow-up in 8 patients with preoperative ASIA grade D status.
After pedicle screws were developed, posterior short-segment fixation and fusion became a popular treatment for thoracolumbar burst fracture. This approach has several advantages, including incorporation of fewer motion segments, shortened duration of surgery, and less blood loss. However, posterior short-segment fixation was associated with a 20% to 40% rate of implant failure.3,18 A large defect in fractured vertebrae is created during posterior instrumentation after application of distraction force. Posterior instrumentation fails easily without anterior column support, and early implant failure may lead to re-kyphosis and worsen clinical outcomes. We noted injuries not only to the vertebra body but also to disk and endplate regions in the thoracolumbar burst fractures. Disks can creep into defects of the fractured vertebra through the injured endplate, which makes the disk degenerative and prevents vertebral healing.19 Both disk-space narrowing and body-height collapse contribute to progressive kyphosis postoperatively.
Transpedicular cancellous bone grafting to augment the anterior column was introduced to overcome the problems associated with traditional treatment of thoracolumbar burst fracture. Olerud et al8 reported 20 patients who underwent posterior segmental fixation followed by bone grafting to the injured vertebra through the pedicle. After a mean follow-up period of 10 months, the spinal canal had complete clearance in 85% of patients, no implants had failed, and the restored anterior body height was lost by only a few percent. Ebelke et al9 used posterior instrumentation with or without anterior body augmentation by transpedicular bone grafting to treat 21 patients with burst fracture. After 19 months, all implantations in patients with transpedicular bone grafting had survived, whereas only 50% of implantations in patients who were not treated with anterior body augmentation had survived. These findings suggest that transpedicular bone grafting to injured vertebrae in addition to short-segment posterior instrumentation appears to be an effective treatment approach for thoracolumbar burst fracture.
Not all studies, however, have showed positive results with bone grafting. Indeed, some other researchers have recently argued against the superiority of transpedicular bone grafting for patients with thoracolumbar burst fracture. Alanay et al5 conducted a prospective, randomized study that compared 2 treatments for thoracolumbar burst fracture: posterior short-segment instrumentation with or without transpedicular bone grafting. According to their criteria for failurean increase of >10° of local kyphosis and/or implant breakagethe patient groups had similar failure rates (50% vs 40%) after a mean follow-up of 32 months. Alvine et al10 reported on 40 patients with thoracolumbar burst fracture who received transpedicular bone grafting with Isola instrumentation (DePuy AcroMed Inc, Raynham, Massachusetts) or variable screw placement. Implant failure rates were as high as 39% in both treatment groups after a mean follow-up of 52 months.
In contrast to these 2 reports, only 3 patients in our study had an increase of >10° in local kyphosis without implant failure, and survivorship was 90.3%. The bone defect created inside the injured vertebra should be refilled by adequate grafting to prevent early collapse. To explain the differences in results, we suggest that the size of the defect in the fractured vertebra and the amount of bone graft are key factors to determine outcome, although there was no scientific data for comparison between these studies.
Some data suggest that transpedicular diskectomy and intercorporeal grafting for an injured upper disk is more effective than transpedicular intracorporeal bone grafting.20 However, Knop et al21 analyzed 76 patients with thoracolumbar fracture who underwent short-segment posterior instrumentation with transpedicular intercorporeal bone grafting and found that the mean loss of kyphosis correction was 10.1°, 37% of the initial surgical reduction was maintained, and only 33.3% of patients had solid interbody fusion after a mean follow-up of 3 years. In our series, the mean loss of kyphosis correction was 6.2° and 60% of the initial surgical correction was maintained after a similar period of follow-up. Thus, transpedicular interbody bone grafting is not able to obtain better results than transpedicular intrabody bone grafting for management of thoracolumbar fracture. Knop et al21 attributed this phenomenon to the relatively poor osteogenic potential of the region near the incompletely evacuated intervertebral disk space. In our opinion, this method creates a large defect for which bone grafting is inadequate to consistently support the anterior column.
Posterior long-segment instrumentation with short-segment fusion has also been studied to prevent early implant failure and correction loss. Altay et al22 reported 63 patients with thoracolumbar burst fracture treated with short- (32 patients) or long-segment (31 patients) posterior instrumentation and short-level fusion. Both groups were similar with regard to age, sex, injury level, load-sharing classification points, preoperative sagittal index, and preoperative canal compromised. The long-segment posterior fixation group acquired better radiographic results but did not provide better clinical results. Some disadvantages of this long-segment fixation short-fusion technique include requiring longer wound incision, lengthening surgery time, and increasing blood loss. Another disadvantage of this technique is the possibility of osteoarthritis at the immobilized but unfused facet joint. Persistent back pain or progressive kyphosis after implant removal is believed to originate from these instrumented unfused segments.23
Figure 3: Preoperative CT of a 30-year-old man with an L2 burst fracture showing a prominent L1-2 facet joint injury. The patient underwent posterior instrumentation and transpedicular cancellous bone grafting without posterior fusion. The alignment deteriorated quickly after removal of the implant.
The average pain score in our study was 1.7, and no patients required chronic medication for incapacitating back pain. Only 2 patients had severe back pain requiring frequent medication, 4 had moderate pain, 7 had occasional pain not requiring medication, and the remaining 18 reported no back pain. Back pain was caused by local kyphosis, disk degeneration, or facet joint degeneration. The average score on the work scale was 2.2 points. Of the 31 patients, 21 (68%) could resume their previous work. Although 3 patients were completely disabled, they reported that their disabilities were caused by associated injuries (calcaneal fracture, pelvic fracture, or acetabulum fracture) not related to the spine injury.
All patients received transpedicular bone grafting and posterior short-segment pedicle screws without posterior fusion. No implant failure occurred, but 3 patients had an increase of >10° in local kyphosis. The loss of kyphosis correction was 10°, 11°, and 25° in these 3 patients. The last case is noteworthy because the kyphosis correction loss was only 4° during the first postoperative year. The local kyphosis increased dramatically after removal of the implant. The CT scan of this patient showed that the facet joint was severely injured preoperatively (Figure 3). Without performing posterior fusion, the corrected alignment could not be maintained even though the anterior column was healed. The patient reported that he was able to return to his previous employment, and his occasional back pain did not require regular medication. Therefore, revision surgery was not recommended for him.
Based on the outcomes of this study, insertion of posterior short-segment pedicle screws followed by transpedicular cancellous bone grafting for anterior body augmentation effectively corrects deformities and restores vertebral height. After a mean follow-up period of 52.7 months, no implant had failed and the clinical results were satisfactory. Although radiographic parameters (vertebral body height, local kyphosis) revealed loss of correction gradually, our data were better than that of previous studies. We conclude that this technique remains a reliable approach for most patients with thoracolumbar burst fracture. However, posterior fusion should be added to prevent quick progression of kyphosis after removal of the implant if posterior column injuries (facet joint or interspinal ligament injury) are apparent during surgery.
- Thomas KC, Bailey CS, Dvorak MF, Kwon B, Fisher C. Comparison of operative and nonoperative treatment for thoracolumbar burst fractures in patients without neurological deficit: a systematic review. J Neurosurg Spine. 2006; 4(5):351-358.
- Bradford DS, McBride GG. Surgical management of thoracolumbar spine fractures with incomplete neurologic deficits. Clin Orthop Relat Res. 1987; (218):201-216.
- Benson DR, Burkus JK, Montesano PX, Sutherland TB, McLain RF. Unstable thoracolumbar and lumbar burst fractures treated with the AO fixateur interne. J Spinal Disord. 1992; 5(3):335-343.
- Louis CA, Gauthier VY, Louis RP. Posterior approach with Louis plates for fractures of the thoracolumbar and lumbar spine with and without neurologic deficits. Spine. 1998; 23(18):2030-2039.
- Alanay A, Acaroglu E, Yazici M, Oznur A, Surat A. Short-segment pedicle instrumentation of thoracolumbar burst fractures: does transpedicular intracorporeal grafting prevent early failure? Spine. 2001; 26(2):213-217.
- Müller U, Berlemann U, Sledge J, Schwarzenbach O. Treatment of thoracolumbar burst fractures without neurologic deficit by indirect reduction and posterior instrumentation: bisegmental stabilization with monosegmental fusion. Eur Spine J. 1999; 8(4):284-289.
- Daniaux H. Transpedicular repositioning and spongioplasty in fractures of the vertebral bodies of the lower thoracic and lumbar spine [in German]. Unfallchirurg. 1986; 89(5):197-213.
- Olerud S, Karlström G, Sjöström L. Transpedicular fixation of thoracolumbar vertebral fractures. Clin Orthop Relat Res. 1988; (227):44-51.
- Ebelke DK, Asher MA, Neff JR, Kraker DP. Survivorship analysis of VSP spine instrumentation in the treatment of thoracolumbar and lumbar burst fractures. Spine. 1991; 16(8 Suppl):S428-432.
- Alvine GF, Swain JM, Asher MA, Burton DC. Treatment of thoracolumbar burst fractures with variable screw placement or Isola instrumentation and arthrodesis: case series and literature review. J Spinal Disord Tech. 2004; 17(4):251-264.
- Magerl F, Aebi M, Gertzbein SD, Harms J, Nazarian S. A comprehensive classification of thoracic and lumbar injuries. Eur Spine J. 1994; 3(4):184-201.
- Denis F, Armstrong GW, Searls K, Matta L. Acute thoracolumbar burst fractures in the absence of neurologic deficit. A comparison between operative and nonoperative treatment. Clin Orthop Relat Res. 1984; (189):142-149.
- Mumford J, Weinstein JN, Spratt KF, Goel VK. Thoracolumbar burst fractures. The clinical efficacy and outcome of nonoperative management. Spine. 1993; 18(8):955-970.
- Gertzbein SD, Court-Brown CM, Marks P, et al. The neurological outcome following surgery for spinal fractures. Spine. 1988; 13(6):641-644.
- Limb D, Shaw DL, Dickson RA. Neurological injury in thoracolumbar burst fractures. J Bone Joint Surg Br. 1995; 77(5):774-777.
- Kostuik JP. Anterior fixation for burst fractures of the thoracic and lumbar spine with or without neurological involvement. Spine. 1988; 13(3):286-293.
- Kaneda K, Taneichi H, Abumi K, Hashimoto T, Satoh S, Fujiya M. Anterior decompression and stabilization with the Kaneda device for thoracolumbar burst fractures associated with neurological deficits. J Bone Joint Surg Am. 1997; 79(1):69-83.
- Butt MF, Farooq M, Mir B, Dhar AS, Hussain A, Mumtaz M. Management of unstable thoracolumbar spinal injuries by posterior short segment spinal fixation. Int Orthop. 2007; 31(2):259-264.
- Oner FC, van der Rijt RR, Ramos LM, Dhert WJ, Verbout AJ. Changes in the disc space after fractures of the thoracolumbar spine. J Bone Joint Surg Br. 1998; 80(5):833-839.
- Daniaux H, Seykora P, Genelin A, Lang T, Kathrein A. Application of posterior plating and modifications in thoracolumbar spine injuries. Indication, techniques, and results. Spine. 1991; 16(3 Suppl):S125-133.
- Knop C, Fabian HF, Bastian L, Blauth M. Late results of thoracolumbar fractures after posterior instrumentation and transpedicular bone grafting. Spine. 2001; 26(1):88-99.
- Altay M, Ozkurt B, Aktekin CN, Ozturk AM, Dogan O, Tabak AY. Treatment of unstable thoracolumbar junction burst fractures with short- or long-segment posterior fixation in magerl type a fractures. Eur Spine J. 2007; 16(8):1145-1155.
- Chen WJ, Niu CC, Chen LH, Chen JY, Shih CH, Chu LY. Back pain after thoracolumbar fracture treated with long instrumentation and short fusion. J Spinal Disord. 1995; 8(6):474-478.
Drs Liao, Fan, Chen (Wen-Jer), Chen (Lih-Hui), and Kao are from the Department of Orthopedic Surgery, Chang Gung Memorial Hospital, Chang Gung University, Taoyuan, Taiwan.
Drs Liao, Fan, Chen (Wen-Jer), Chen (Lih-Hui), and Kao have no relevant financial relationships to disclose.
Correspondence should be addressed to: Jen-Chung Liao, MD, Department of Orthopedic Surgery, Chang Gung Memorial Hospital, Chang Gung University, No. 5, Fu-Shin St, Kweishian, Taoyuan 333, Taiwan.