Background context: Stable thoracolumbar spinal fracture is a common injury, yet there remains a relative lack of evidence that would allow reliable prediction of outcome. Variation in outcome in stable thoracolumbar spine fracture without neurological deficit could not be explained by the assessment made from x-rays and CT imaging alone, which measures mainly the bony injury severity. So far, no good classification system has been developed to assess the severity of soft tissue injury (disc). Objectives: To measure short-term outcome of stable thoracolumbar fracture and analyse aspects of injury severity for their ability to predict outcome. To develop a new disc injury severity grading system in thoracolumbar spine fractures. Study design: Prospective observational. Patient sample: 44 patients with stable fractures between T11 and L5 vertebra with no neurological deficit and treated conservatively were selected. Methods: All had plain x-rays, CT and MRI scans post-injury and at one year post-injury (except CT). Bony injury severity was scored on a seven-point ordinal scale based on (a) communication, (b) apposition and (c) kyphosis. Disc injury severity was scored by the newly developed six-point ordinal scale (none to severe) based on the following variables: (a) Herniation of the disc, (b) Indentation of the end plate, (c) Change in height and (d) Change in signal. Outcome was assessed at one to two years from injury. The following outcome measures were collected: (A) Five domains of pain symptoms – intensity, duration, bothersome, interference and satisfaction. (B) Five domains of function – Physical Component Summary (PSC) measure, Mental Component Summary (MSC) measure, Oswestry disability score, return to pre-injury level of overall activities and return to employment. Non-parametric correlation coefficients were calculated between outcome variables and other variables to look for the predictors of outcome. Stepwise linear regression analysis was performed to compare the predictive values and to look at what proportion of outcome is predicted by different predictors. Results: According to AO classification, the fractures were A1, A2, A3 and B1. The Spearman correlation coefficients between outcome and injury severity were consistently higher with disc injury severity than bony. For the outcome of pain intensity, the correlation coefficients for disc and bone injury severity respectively were:.63 (p < .0001) and.28 (NS-not significant). Similarly for SF36 PSC were: .41 (p < .01) and.25 (NS). The predictive value of pain at F < .01 was 29% for disc injury severity and all other variable were not significant and excluded. At F < .05, the predictive value of disc injury severity (29%) increased further by 9%, 8% and 6% by addition of variable “Patient’s pre-injury SF36 MSC”, “Legal and Compensation issues pending” and “Physical demand of job” respectively. The predictive value of function at F < .01 was 16% for disc injury severity and it increased to 31% by the addition of “physical demand of the job” variable. At F < 0.05 the predictive value further increased by 5% by addition of variable “Legal and Compensation issues pending”. All other variables were not significant. Conclusions: A new grading system of disc injury severity was developed and it showed good predictive value for pain and functional outcome. Disc injury severity has a better predictive value for short term outcome than bony injury severity in stable thoracolumbar fractures. In the spectrum of injuries studies, the AO classification and the degree of kyphosis provided no prediction of outcome

The purpose of this study was to evaluate and compare the effect of short segment pedicle screw instrumentation and an intermediate screw (SSPI+IS) on the radiological outcome of type A thoracolumbar fractures, as judged by the load-sharing classification, percentage canal area reduction and remodelling. . We retrospectively evaluated 39 patients who had undergone hyperlordotic SSPI+IS for an AO-Magerl Type-A thoracolumbar fracture. Their mean age was 35.1 (16 to 60) and the mean follow-up was 22.9 months (12 to 36). There were 26 men and 13 women in the study group. In total, 18 patients had a load-sharing classification score of seven and 21 a score of six. All radiographs and CT scans were evaluated for sagittal index, anterior body height compression (%ABC), spinal canal area and encroachment. There were no significant differences between the low and high score groups with respect to age, duration of follow-up, pre-operative sagittal index or pre-operative anterior body height compression (p = 0.217, 0.104, 0.104, and 0.109 respectively). The mean pre-operative sagittal index was 19.6° (12° to 28°) which was corrected to -1.8° (-5° to 3°) post-operatively and 2.4° (0° to 8°) at final follow-up (p = 0.835 for sagittal deformity). No patient needed revision for loss of correction or failure of instrumentation. Hyperlordotic reduction and short segment pedicle screw instrumentation and an intermediate screw is a safe and effective method of treating burst fractures of the thoracolumbar spine. It gives excellent radiological results with a very low rate of failure regardless of whether the fractures have a high or low load-sharing classification score. Cite this article: Bone Joint J 2014;96-B:541–7

Stable thoracolumbar fracture is a common injury. The factors that determine its outcome are unclear. Aspects of injury severity were analysed for their ability to predict outcome by controlling other outcome-affecting factors (patient's pre-injury health status, legal aspects, associated injuries, etc.). No reliable disc injury severity grading system was available and therefore a new system was developed. A prospective observational study of 44 conservatively treated patients with stable fractures between T11 and L5 was conducted. Bony injury severity was scored based on comminution, apposition and kyphosis parameters. Disc injury severity was scored by the new scale based on variables – Herniation, Indentation, Height decrease and Signal change – seen in MRI. Ten outcome domains (five domains of pain and function each) were assessed at 1 to 2 years from injury. The data was analysed by non-parametric correlation and stepwise-linear regression analysis to assess the predictive value of different variables (patient factors, injury factors and social factor) to outcome. The correlation coefficients between injury severity and outcome were consistently higher with disc injury severity than bony. Disc injury severity showed highest predictive value for both pain (29%) and functional (16%) outcomes, whereas the bony injury severity parameters (kyphosis, etc.) and the posterior ligament injury severity provided no prediction of outcome. According to AO classification, the fractures were A1, A2, A3 and B1; in this spectrum of injuries, the AO classification had no prediction of outcome. The disc injury score also had a good predictive value for final disc degeneration. Disc injury severity should be gauged in advising prognosis and treatment. The new disc injury severity grading system showed good construct validity

Study Design/Objectives: A pilot study to predict thoracolumbar kyphosis progression secondary to fracture in non-operatively treated patients. Summary of Background Data: Progressive saggital plane deformity can cause persistent pain after thoracolumbar vertebral fractures. Little data exists to suggest at what interval after the index injury the patient attains a low risk of developing further angular deformity in non-operatively treated patients. Methods: Supine and erect radiographs were assessed and the degree of fracture kyphosis was determined using an Oxford Cobbometer. The fracture kyphosis was recorded for each follow up appointment along with time after the fracture. A time/data analysis was performed using the Blyth-Still-Casella exact interval. Results: This study included 22 patients (13 male, 9 female) with average age 67.2 years (range, 14–87 years). The average length of follow up was 11.5 months (range, 5.3–19.9 months) and the average number of radiographs taken within this period was 4 (range, 2–6). The change in fracture kyphosis was plotted against time following fracture. Based on 15 patients with data extending to 200 days follow up, it was observed that the rate of change in fracture kyphosis between two time points of 100 and 150 days predicted the trend in kyphosis progression until the end of follow up in 14 out of the 15 patients. This observed rate of 14/15 (0.93) has a 95% confidence interval of 0.7 to 0.99 (Blyth-Still-Casella exact interval). Conclusions: The standing lateral radiograph of patients with conservatively treated thoracolumbar fractures at 3 and 4.5 months post injury can be used to predict fracture kyphosis progression. Using this protocol, patients can be safely discharged earlier from outpatient follow up reducing radiological exposure

We assessed narrowing of the spinal canal in 39 burst fractures and fracture-dislocations of thoracolumbar vertebrae treated by the AO Internal Spinal Fixator, using CT preoperatively and at various stages postoperatively. Computer-aided planimetry was used to measure the narrowing, and its restoration shortly after instrumentation, or at 15 months. The mean initial reduction of canal area was to 63.7% +/- 18.8% of normal; this was restored to a mean of 95.4% +/- 21.2% of normal when measured either soon after surgery or at 15 months (p < 0.001 for both groups). There was more improvement in cases assessed later. For fractures from D12 to L3, the mean canal area was restored to 99.4% of normal; but at L4 or L5 the mean restitution was to only 60.9% (p < 0.05). We found no correlation between preoperative loss of area and amount of restoration, or severity of neurological deficit. Nor was there any correlation between the delay before surgery and the improvement achieved. The mechanism of fracture reduction appears to be a combination of distraction ligamentotaxis and forced hyperextension.

Of a total of 905 patients with fracture or fracture-dislocation of the thoracolumbar spine admitted from 1969 to 1982, a neurological deficit was present in 334 (37%). All unstable injuries were initially treated by reduction and posterior fusion. In 79 of these patients, an anterolateral decompression was undertaken later because of persistent neurological deficit and radiographic demonstration of encroachment on the spinal canal. One patient died of pulmonary embolism; 78 were reviewed after a mean period of four years. Of these 78 patients 18 made a complete neurological recovery while 53 appeared to have benefited from the procedure; 25 remained unchanged. The best results were obtained in burst fractures at thoracolumbar and lumbar levels when a solitary detached fragment of a vertebral body had been displaced into the spinal canal. These results indicate that anterolateral decompression of the spinal canal should be considered, after careful evaluation, for certain injuries of the spine in which there is severe neural involvement

Introduction. The management of thoracolumbar burst fractures is controversial. The goal of our study was to evaluate whether the psychological factors or the late spinal deformities influence outcome and in particular quality of life following surgical treatment of burst fractures of the thoracolumbar spine. Material and methods. In a retrospective analysis, we evaluated outcome in 45 patients in whom burst fractures of the thoracolumbar spine without neurological deficits were surgically treated between April 2001 and November 2004. For this purpose, patient charts, surgery reports and x-ray images were analyzed consecutively. 29 patients could be examined physically and the outcome could be evaluated with VAS spine core, quality of life according to short-form 36 (SF36) and Beck Depression Inventory (BDI) with a minimum follow up of 30 months. Results. Mean VAS spine score was 60±26. Neither VAS spine score, nor quality of life results correlated with the following radiological findings: vertebral body angle, sagittal index and height of cranial disc space of the vertebra. Beck Depression Inventory (BDI) correlated with SF-36 score and VAS spine score (p< 0.05). Patients who were depressed showed significantly worse results in relation to the VAS spine score and the SF36 score (p< 0.01). Conclusion. For the first time, we could show, that psychological factors have high influence on functional outcome and health related quality of life in operative treated thoracolumbar burst fracture independent of x-ray findings. Therefore, we recommend inclusion of psychological components in the treatment and outcome-evaluation of the thoracolumbar burst fracture in future

The aim of this study was first, to determine whether CT scans undertaken to identify serious injury to the viscera were of use in detecting clinically unrecognised fractures of the thoracolumbar vertebrae, and second, to identify patients at risk of ‘missed injury’.

We retrospectively analysed CT scans of the chest and abdomen performed for blunt injury to the torso in 303 patients. These proved to be positive for thoracic and intra-abdominal injuries in only 2% and 1.3% of cases, respectively. However, 51 (16.8%) showed a fracture of the thoracolumbar vertebrae and these constituted our subset for study. There were eight women and 43 men with mean age of 45.2 years (15 to 94). There were 29 (57%) stable and 22 (43%) unstable fractures. Only 17 fractures (33.3%) had been anticipated after clinical examination. Of the 22 unstable fractures, 11 (50%) were anticipated. Thus, within the whole group of 303 patients, an unstable spinal injury was missed in 11 patients (3.6%); no harm resulted as they were all protected until the spine had been cleared. A subset analysis revealed that patients with a high Injury Severity Score, a low Glasgow Coma Scale and haemodynamic instability were most likely to have a significant fracture in the absence of positive clinical findings. This is the group at greatest risk.

Clinical examination alone cannot detect significant fractures of the thoracolumbar spine. It should be combined with CT imaging to reduce the risk of missed injury.

Many authors recommend surgery to remove retropulsed bone fragments from the canal in burst fractures to 'decompress' the spinal canal. We believe, however, that neurological damage occurs at the moment of injury when the anatomy is most distorted, and is not due to impingement in the resting positions observed afterwards. We studied 20 consecutive patients admitted to our spinal injuries unit over a two-year period with a T12 or L1 burst fracture. There was no correlation between bony or canal disruption and the degree of neurological compromise sustained but there was a significant correlation between the energy of the injury (as gauged by the Injury Severity Score) and the neurological status (p < 0.001). This suggests that neurological injury occurs at the time of trauma rather than being a result of pressure from fragments in the canal afterwards and questions the need to operate simply to remove these fragments.

In 139 patients with burst fractures of the thoracic, thoracolumbar or lumbar spine, the least sagittal diameter of the spinal canal at the level of injury was measured by computerised tomography. By multiple logistic regression we investigated the joint correlation of the level of the burst fracture and the percentage of spinal canal stenosis with the probability of an associated neurological deficit. There was a very significant correlation between neurological deficit and the percentage of spinal canal stenosis; the higher the level of injury the greater was the probability. The severity of neurological deficit could not be predicted.

Study design. Retrospective study. Objectives. To identify patient outcomes, in particular employment, >5-years following traumatic thoracolumbar fracture. Methods. 235 patients between the ages of 18–65 were identified from the hospital radiology database having sustained a traumatic thoracolumbar fracture on CT or MRI between 01/01/2013 and 31/01/2017. Questionnaires were sent out via post and available emails, with a reminder letter and phone calls. Retrospective data was gathered about employment status pre-fracture and >5-years post injury. Results. 26 patients had died at follow up leaving 209 patients. 108 (52%) were treated surgically and 101 (48%) conservatively. 106 replies were received with 85 (80%) opting in and 21 (20%) opting out. 68 (80%) patients completed the full questionnaire with 17 (20%) filling out a shortened questionnaire via a phone conversation. 52 (61%) patients underwent surgery and 33 (39%) were treated conservatively. The average follow up was 8 years. Prior to injury 66 (78%) were employed and 19 (22%) unemployed (of which 6 were full time students and 8 were retired). 49 (74%) of the previously employed patients returned to work at follow up with 35 (53%) working the same or increased hours. Regarding employment, there was no significant difference between surgically and conservatively treated patients (P=0.355) or the classification of the fracture (P=0.303). 16 (19%) patients reported back pain prior to their injury whilst 69 (81%) did not. There were 58 (68%) cases of new pain at follow up with the most affected area being the lumbar region in 43 (51%) patients. 32 (38%) patients reported neurological deficit post injury: 19 with subjective symptoms, 9 with objective symptoms and 4 suffered paralysis. Conclusion. >5-years following a traumatic thoracolumbar fracture most individuals return to employment. There was no significant difference between the severity of the fracture or how patients are treated on their employment outcomes

Objectives. This presentation discusses the experience at our Centre with treating traumatic thoracolumbar fractures using percutaneous pedicle screw fixation and also looks at clinical and radiological outcomes as well as complications. Design. This is a retrospective study reviewing all cases performed between Jan 2013 and June 2019. Subjects. In our study there were 257 patients in total, of which there were 123 males and 134 females aged between 17 and 70. Methods. We reviewed the case notes and imaging retrospectively to obtain the relevant data. Results. A total of 257 patients were included, 123 males and 134 females; the mean age was 47.6 years. The majority of injuries were from fall from significant height. In 98 cases the fracture involved a thoracic vertebra and in 159 cases a lumbar vertebra. Percutaneous pedicle screw fixation was performed either one level above and below fracture or Two levels above and below the fracture depending upon the level of injury. Forty two cases were treated with additional short pedicle screws at the level of fracture. More than 15% (39) of patients presented with a neurological deficit on admission and more than 80% (32) of those showed post-operative improvement in their neurology as per Frankel Grading system. The mean Operative time was 117minutes +− 45, and mean length of hospital stay was 7.2 +− 3.8 days, with significant improvement in Visual analogue score. Percutaneous fixation achieved a satisfactory improvement in radiological parameters including sagittal Cobb angle (SCA) post-operatively in all patients. The vast majority of patients achieved a good functional outcome according to modified Macnab criteria. Follow up was for a maximum of two years, with relevant imaging at each stage. Ten (3.8%) patients had wound infection with three patients requiring wound debridement. Four patients had upper level screws pulled out and in Four cases one screw was misplaced. All eight had revision surgery. Conclusions. Percutaneous pedicle screw fixation is a safe surgical option with comparable outcomes to open surgery and a potential reduction in perioperative morbidity. Percutaneous pedicle screw fixation is the primary surgical technique to treat traumatic thoracolumbar fractures at our Centre. There were no major complications in our series, with good functional outcome following surgery

Thoracolumbar injury classification systems are not used or researched extensively in paediatric population yet. This systematic review aims to explore the validity and reliability of the two main thoracolumbar injury classification systems in the paediatric population (age ≤ 18). It also aims to explore the transferability of adult classification systems to paediatrics. The Thoracolumbar Injury Classification System (TLICS) published in 2005 and the AO Spine published in 2013 were assessed in this paper because they both provide guidance for the assessment of the severity of an injury and recommend management strategies. A literature search was conducted on the following databases: Medline, EMBASE, Ovid during the period November 2020 to December 2020 for studies looking at the reliability and validity of the TLICS and AO Spine classification systems in paediatric population. Data on validity (to what extent TLICS/ AO Spine recommended treatment matched the actual treatment) and reliability (inter-rater and intra-rater reliability) was extracted. There is an “almost perfect validity” for TLICS. There is a “strong association” between the validity of TLICS and AO Spine. The intra-rater reliability is “moderate” for TLICS and “substantial” for AO Spine. The intra-rater reliability is “substantial” for TLICS and “almost perfect” for AO Spine. The six studies show a good overall validity and reliability for the application of TLICS and AO Spine in pediatric thoracolumbar fractures. However, implication of treatment and anatomical differences of the growing spine should be explored in detail. Therefore, AO Spine can be used in absence of any other classification system for paediatrics

Introduction:. Most of the published papers on AI based diagnosis have focused on the algorithm's diagnostic performance in a ‘binary’ setting (i.e. disease vs no disease). However, no study evaluated the actual value for the clinicians of an AI based approach in diagnostic. Detection of Traumatic thoracolumbar (TL) fractures is challenging on planar radiographs, resulting in significant rates of missed diagnoses (30-60%), thus constituting a field in which a performance improvement is needed. Aim of this study is therefore to evaluate the value provided by AI generated saliency maps (SM), i.e. the maps that highlight the AI identified region of interests. Methods:. An AI model aimed at identifying TL fractures on plain radiographs was trained and tested on 567 single vertebrae images. Three expert spine surgeons established the Ground Truth (GT) using CT and MRI to confirm the presence of the fracture. From the test set, 12 cases (6 with a GT of fracture and 6 with a GT of no fracture, associated with varying levels of algorithm confidence) were selected and the corresponding SMs were generated and shown to 7 independent evaluators with different grade of experience; the evaluators were requested to: (1) identify the presence or absence of a fracture before and after the saliency map was shown; (2) grade, with a score from 1 (low) to 6 (high) the pertinency (correlation between the map and the human diagnosis), and the utility (the perceived utility in confirming or not the initial diagnosis) of the SM. Furthermore, the usefulness of the SM was evaluated through the rate of correct change in diagnosis after the maps had been shown. Finally, the obtained scores were correlated with the algorithm confidence for the specific case. Results:. Of the selected maps, 8 had an agreement between the AI diagnosis and the GT, while in 4 the diagnosis was discordant (67% accuracy). The pertinency of the map was found higher when the AI diagnosis was the same as the GT and the human diagnosis (respectively p-value = .021 and <.000). A positive and significant correlation between the AI confidence score and the perceived utility (Spearman: 27%, p-value=.0-27) was found. Furthermore, evaluator with experience < 5 year found the maps more useful than the experts (z-score=2.004; p-value=.0455). Among the 84 evaluation we found 12 diagnostic errors in respect to the GT, 6 (50%) of which were reverted after the saliency map evaluation (z statistic = 1.25 and p-value = .21). Discussion:. The perceived utility of AI generated SM correlate with the model confidence in the diagnosis. This highlights the fact that to be considered helpful, the AI must provide not only the diagnosis but also the case specific confidence. Furthermore, the perceived utility was higher among less experienced users, but overall, the SM were useful in improving the human diagnostic accuracy. Therefore, in this setting, the AI enhanced approach provides value in improving the human performance

Study design: retrospective clinical study . Objective: To study radiological late results after posterior stabilization of thoracolumbar fractures with internal fixation. To know factors related with loss of correction and hardware failure. Summary of background data: The posterior approach using an internal fixator is a standard procedure for stabilizing the injured thoracolumbar spine. None of the surgical techniques used was able to maintain the corrected the kyphosis angle. Methods: Forty-five patients with thoracolumbar fractures were included in the study. The inclusion criterion was the presence of fracture through the T11-L3 vertebrae without neurologic compromise. The Load-sharing classification has been used for all patients to determine the fracture severity. Surgical techniques (short or long instrumentation) , preoperative and postoperative radiographs ( Cobb technique) and follow-up records of all patients were reviewed carefully from the time of surgery until final follow-up assessment. Results: 13 patients were treated using short-segment instrumentation (two disc spaces) and 32 patients with long-segment instrumentation (more than two disc spaces). The mean follow-up was 3.4 years (range 1 to 11 years). The mean preoperative Cobb angle was 16.1 degrees and after surgery the mean angle was 6.8° representing an average correction of 9.2 ° . At follow-up assessments the mean Cobb angle was 13.2° representing a loss of correction of 6.4°. Implant failure ( 5 loosening and 8 breakage) was seen in 28.8% of patients: 6/14 (42%) of patients receiving short instrumentation and 7/31 (22%) of patients with long instrumentation. Hardware failure was seen in 53.3% of patients with Cobb angle preoperative more than 20° and in 16.6% of patients with Cobb angle less than 20°. Conclusions: Radiological behaviour of thoracolumbar fractures treated with posterior instrumentation without anterior support was worse than expected. Hardware failure was related with Cobb angle fracture > 20°, postoperative correction superior than 10° and short pedicular instrumentation technique

Background: Thoraco-lumbar fractures without a neurological deficit are usually suitable for non-operative treatment. The main area of clinical interest is the deformity at the injured levels. The deformity may be evident at the time of presentation, though could be expected to progress in time. Objective: Accurate assessment of the temporal behaviour in the geometry of the injured segments in non-operatively treated thoracolumbar fractures with normal neurology. Materials: 102 patients with thoracolumbar fractures without a neurological deficit were treated non-operatively at our unit between June 2003 and May 2006. The mean age of our patient cohort was 46.9 yrs (16–90 yrs). Strict criteria were followed to determine suitability for non-operative treatment. Supine radiographs were performed at the initial assessment. Erect radiographs were performed when trunk control was achieved and at follow-up assessments thereafter. Methods: Quality Motion Analysis (QMA) software (Medical Metrics Inc, Houston, Tx) was used to measure rotational and translation changes between the end plates using a validated protocol. The radiographs were standardised for magnification and superimposed from different time points. Transformation matrices were used to track the changes. The AO classification was used to classify the fractures by 2 independent observers. Results: A median of 4 radiographs were analysed for each patient (range 2–9), at a mean follow-up of 5.6 mo (95% CI 4.1–7.1 mo). 92% of the cohort had sustained a 1 level injury. 76% of the injuries were between T12 and L2; 19% were in the thoracic spine. An inter-observer rating of 0.58 was obtained for the classification of the primary fracture type. The mean rotational change was −1.4855° ± 0.248° (95% CI: −0.994° to–1.976°). The mean anterior vertebral body height collapse was −4.3444° ± 0.6938 (95% CI: −2.695 to −5.724). The mean posterior vertebral height collapse was −0.7987 ± 0.259 (95% CI: −0.284 to −1.313). Conclusions: We report the use of QMA software to track changes in the vertebral body geometry accurately. This has implications on the clinical aspects of management of thoracolumbar fractures based to progression of deformity that could be explored in future studies

Introduction: Spinal fractures are associated with pain, disability, neurological dysfunction and mortality. Osteoporosis and risky leisure time activities are increasing in the population. New treatment options have been introduced. However, only a few international studies have reported its descriptive epidemiology. There are no clear consensuses regarding the choice of operative interventions versus non-operative treatment in patients with thoracolumbar fractures. Treatment is often based on local traditions, skills and experiences. The aim of this nationwide study is to analyse the incidence, the characteristics of the patients, the subsequent development, surgical incidence and mortality rate among hospitalized patients with thoracolumbar fractures in Sweden. Methods: All discharges between 1997 and 2000 with diagnoses of thoracic or lumbar vertebrae fracture according to ICD 10 classification were selected from the National Inpatient Register. Surgery in these patients was categorised by procedure codes indicating spine operations. In order to calculate the risk of patients dying, linkage was performed to the Swedish Death Register using the unique personal identification number. Results: We identified 13, 496 admissions during the study period. This corresponds to an incidence rate of 30 per 100, 000 person-years and the occurrence was stable during 1997 and 2001 for patients younger than 50 years but decreased for patients above 50 years. The incidence of spine surgery among all patients was 1.9 per 100, 000 person-years, ranging from 1.6 per 100, 000 inhabitants per year to 2.6 per 100, 000 in the different region of Sweden. The most common causes of the fracture were falls (53 %) followed by vehicle accidents (37%). Almost two thirds of the patients operated on were men (63 %) and two thirds had lumbar vertebral fractures (66 %). The median age of the patients operated on was 42 years. The median length of stay was eight days. Among those who was operated the 30 days case fatality rate was 0.7 %. Median age for death was 64 years. Operations were less common in women (OR 0.79). The number of patients operated on during the last study year 2001 was significantly increased (OR 1.29). The probability of being operated on was highest in the more urban and densely populated regions of Stockholm and southern Sweden. Discussion: This national study based study showed a stable cumulative incidence of thoracolumbar fractures over the year 1997–2000. However, the two last year of this study an increased incidence of operation was observed. A possible explanation of this finding could be the new percutaneus technique for verterbroplasty. We find it of most importance to further investigate the reasons behind the gender differences in surgical incidence in patients with vertebral fractures