Clinical Effect And Correlation Analysis Of Short-segment Pedicle Instrumentation For Thoracolumbar Fractures

Objective: With the development of modern science, transportation andconstruction, the incidence of traumatic spinal cord injury had risen sharply.Thoracolumbar fracture (T11~L2) is the most common of this kind injury inclinic, which accounts for7.23%of the whole body fractures and couldaccompany with the injury of spinal cord, conus medullaris and cauda equine(men are more than women, most of their ages are between41and50).Thoracic spine is relatively fixed while lumbar spine is more mobilizable. Thethoracolumbar spine is the turning point from thoracic kyphosis to lumbarlordosis, the facets of articular processes are from coronary to sagittal in thisarea. When spine suffered from violence, strength could focus onthoracolumbar segment and led to thoracolumbar fractures. This area is also atransition zone between spinal cord and cauda equine, so spinal cord andcauda equine could be injured when fractures happened. Thoracolumbarfractures are complicated and can cause a high disability rate. Without aneffective treatment, it would affect the patient’s quality of life seriously andbring a heavy economic burden to family, society and nation. So strengthenthe research of this injury is of great significance. Goals for the treatment areto stabilize the spine and protect or restore nerve function, to improve thequality of patients’ life and work ability. In clinic, no unified standard isaccepted to guide the treatment, it remains controversial. Pedicle screwfixation has become popular for thoracolumbar fractures by its goodrestoration and effect. The anatomy of posterior approach was simple and itstechnique was easy to grasp with smaller trauma and less bleeding. So pediclescrew fixation technique has been widely used in hospitals at all levels. Butthe late follow-up found that pedicle screw fixation had some failure rate,correction lose and kyphosis exacerbation, which could influence the patient’s life and work. In this study, we followed up some thoracolumbar fracturespatients treated by short-segment pedicle instrumentation (SSPI) to evaluatethe effect of SSPI and explore factors that led to fixation failure and correctionlose.Methods:59cases of thoracolumbar fractures treated by SSPI during2009.4to2012.7were followed for17to50(34.97±9.23) months, including50men and9women. The age was from15to68, the average was38.95±12.21. The mechanism of injury: high falling injury in36cases (61%),traffic accident injury in16cases (27.1%), falling down injury in5cases(8.5%) and bruise injury caused by heavy object in2cases (3.4%). There were2cases (3.4%) fractured in T11,15cases (25.4%) in T12,26cases (44.1%) inL1and16cases (27.1%) in L2. The ASIA classification was used to evaluatespinal cord injury:0case (0%) at A level,0case (0%) at B level,4cases (6.8%)at C level,10cases (16.9%) at D level and45cases (76.3%) at E level. Everypatient had thoracolumbar spine X-ray (AP and LAT), CT and MRIPre-surgery (Pre) and Po-surgery (Po), the last follow-up X-ray (AP and LAT).Anterior vertebral height (AVH), middle vertebral height (MVH), posteriorvertebral height (PVH), vertebral wedge angle and Cobb’s angle weremeasured, the difference values between Po-surgery and follow-up weredefined as correction lose. Denis et al.’s Pain scale, Work scale and ASIAclassification were used to evaluate the clinical effect. Patients were dividedinto two Groups, Group A: the correction lose of Cobb’s angle <10°, Group B:the correction lose of Cobb’s angle≥10°. SPSS16.0was used to analyze,comparing the total and within group differences between the last follow-upand Pre, the last follow-up and Po, the differences between groups wereanalyzed. T-test was used for enumeration data and chi square text for countdata. The correlation coefficients between pain scale and last follow-upCobb’s angle, work scale and last follow-up Cobb’s angle were also analyzed.Results: Imaging data:1. The overall follow-up cases’ vertebral wedgeangle correction lose was1.89°±1.35°, Cobb’s angle correction lose was7.25°±3.69°, there were statistically differences (P<0.05) between last follow-up and Pre, last follow-up and Po.2. In Group A, vertebral wedgeangle correction lose was1.57°±1.33°and Cobb’s angle correction lose was5.79°±1.34°. Compared last follow-up with Pre and Po, all the data hadsignificant differences (P<0.05).3. In Group B, vertebral wedge anglecorrection lose was3.27°±2.15°while Cobb’s angle correction lose was13.59°±4.01°. Cobb’s angle at the last follow-up compared with Pre, there wasno statistically difference (P=0.716). Apart from this, all the data at the lastfollow-up compared with Pre and Po, the differences were statisticallysignificant (P<0.05). The height of disc space was lost obviously in7cases(7/11) and in Pre MRI images, injury could be found in intervertebral discadjoined the fractured endplate. Vertebral restoration loss was significantly in4patients (4/11) and their Pre X-ray and CT showed a diminution of bonedensity.4. Between Group A and B, except Pre AVH (P=0.095), Po Cobb’s(P=0.384) and MVH correction lose (P=0.057), the values were statisticaldifference (P<0.05). The unilateral95%medical reference range of AVH,PVH and Cobb’s angle at Pre: AVH＞46.4%，PVH＞90.3%，Cobb’s angle＜20.5°. The clinical effect evaluation:1. Denis et al.’s Pain scale distribution inGroup A:23patients in G1,25in G2and no one in G3~5. In Group B,3casesin G1,5in G2and3in G3, no one in G4and G5. The distribution of painscale in Group A and B, there was a statistical difference (P<0.05). But therewas no relationship between the pain scale and Cobb’s angle at the lastfollow-up (γ=0.224, P=0.088).2. Denis et al.’s Work scale distribution inGroup A:11cases in G1,36in G2,1in G3, no one in G4and G5. In Group B:no one in G1,6in G2,5in G3, no one in G4and G5. The distribution of Workscale had a significant difference (P<0.05) between Group A and B. Thecorrelation coefficient of Work scale and Cobb’s angle at the last follow-upwas0.345, it had statistical significance (P=0.007).3. The patients who hadincomplete nerve injury before surgery had at least1level of ASIA gradeimprovement. No patient’s neurological status was aggravating at the lastfollow-up.Conclusion: In general, SSPI for Type A thoracolumbar fractures has a satisfied effect, but to strictly grasp the indications. For Type A thoracolumbarfractures, when AVH<46.4%, PVH<90.3%, Cobb’s angle>20.5°combinedwith intervertebral disc injury or osteoporosis, SSPI is not applicable.