OBJECTIVETo compare the clinical results of additional screws fixation on fractured vertebrae versus only short-segment posterior transpedicular instrumentation for A3 thoracolumbar fracture without neurologic deficit.

METHODSClinical data of 52 cases of thoracolumbar burst fracture without neurologic deficit were retrospectively analyzed. All patients were divided into 2 groups due to different instrumentation and all fractures were classified as type A3 according to AO Classification.From January 2005 to December 2006, 23 cases in group A were treated by short-segment posterior instrumentation combined with additional screws fixation on fractured vertebrae. There were 18 male and 5 female with a mean age of (35.3+/-8.3) years. The fracture segment included 1 in T11, 9 in T12, 11 in L1 and 2 in L2. From January 1999 to December 2004, 29 cases in group B were treated only by conventional short-segment posterior transpedicular instrumentation. There were 20 male and 9 female with a mean age of (37.3+/-6.8) years. The fracture segment included 1 in T11, 7 in T12, 20 in L1 and 1 in L2. The clinical effect and radiographic measurements were respectively compared preoperatively, immediate and 2 years postoperatively.

RESULTSAll patients were followed up and the mean follow-up time was (37.4+/-10.9) months (from 24 to 48 months). There was no statistic difference of mean JOA and VAS score between 2 groups preoperatively, immediate and 2 years postoperatively (P>0.05). The average immediate postoperative correction of Cobb's angle was 13.7 degrees+/-7.7 degrees in group A, which was statistically significantly higher than that of 8.8 degrees+/-5.0 degrees in group B (P<0.01). The mean kyphosis correction loss of 2.9 degrees+/-1.5 degrees in group A was statistically significantly lower than that of 5.0 degrees+/-2.9 degrees in group B 2 years postoperatively (P<0.01). The average restoration of anterior height of fractured vertebral body immediate postoperatively was (29.4+/-6.0)% and (21.7+/-6.9)% respectively. The mean correction loss of anterior height 2 years postoperatively was (3.1+/-0.8)% and (6.6+/-3.0)% respectively. The average restoration of posterior height of fractured vertebral body immediate postoperatively was (8.5+/-3.2)% and (6.1+/-1.8)% respectively. The mean correction loss of posterior height 2 years postoperatively was (2.0+/-0.8)% and (3.4+/-1.0)% respectively. There were significant differences in average restoration of anterior/posterior height immediate postoperatively and correction loss of anterior/posterior height 2 years postoperatively between the 2 groups (P<0.01). According to fracture fragments protruded into the spinal canal on immediate postoperative CT image, there were complete reduction in 11 cases (47.8%) and partial reduction in 12 cases (52.2%) in group A, which was statistically significantly better than those in group B (P<0.01). There was no severe neurologic complications and no other complications related to additional screws fixation postoperatively. Pedicle screw breakage occurred in 2 cases in group B and none in group A.

CONCLUSIONSBetter initial kyphosis correction and less loss of correction 2 years after operation can be obtained by using additional screws fixation on fractured vertebra for thoracolumbar A3 fracture without neurologic deficit.