References

Rochester, MN population study: clinically diagnosed vertebral fracture incidence 117/100,000 person-years, women 145 vs men 73; 83% followed moderate or no trauma.

Pooled placebo arms (n=2725): 19.2% of women with an incident vertebral fracture sustained another within 1 year; prevalent fracture raised 1-year risk 5-fold (RR 5.1).

SOF cohort (9575 women, 8.3 years): prevalent vertebral fractures gave 1.23-fold age-adjusted mortality; 19 to 44 deaths/1000 woman-years with 0 vs 5+ fractures; pulmonary death HR 2.1.

NEJM Clinical Practice review of vertebral fracture epidemiology, diagnosis (VFA, radiographs), acute pain management, augmentation controversy and pharmacological secondary prevention (no PubMed abstract).

Global estimate for 2000: 9.0 million osteoporotic fractures, including 1.4 million clinical vertebral fractures; 5.8 million DALYs lost, exceeding most cancers in Europe.

Medicare 5% sample (97,142 VCF patients): mortality about twice matched controls; survival 53.9%, 30.9% and 10.5% at 3, 5 and 7 years; greater excess in men.

IMPACT study (2451 osteoporotic women): local radiology reading missed vertebral fractures in 34% (false-negative 45% North America); underdiagnosis is a worldwide problem.

934 hospitalised older women with chest radiographs: 14.1% had moderate/severe vertebral fractures but only 1.8% received the diagnosis; 18% got any osteoporosis treatment.

Statistical synthesis: prior vertebral fracture confers about 4-fold risk of subsequent vertebral fracture and roughly 2-fold risk (RR 2.0) of fracture at other sites.

FIT cohort (6459 women): prevalent vertebral deformities predicted mortality (RR 1.60) and hospitalisation (RR 1.18), with risk rising with number of deformities.

Spinal osteoporotic fractures reduced vital capacity (93.6% vs 105.6% of predicted) and FEV1 versus chronic low back pain controls; VC correlated with spinal deformity index.

19 studies of hospitalised vertebral fragility fracture patients: mean age 81, median stay 10 days, 1-year mortality 20-27%, 34-50% discharged to care facilities, one-third started osteoporosis treatment.

Genant semiquantitative grading (grade 1 mild 20-25%, 2 moderate 25-40%, 3 severe >40% height loss) is reliable for prevalent and incident vertebral fracture assessment when criteria are defined.

DGOU OF classification: OF1 oedema only; OF2 deformation with <1/5 posterior wall; OF3 >1/5 posterior wall; OF4 frame collapse/pincer; OF5 distraction/rotation; interobserver kappa 0.63.

DGOU OF-score from 707 cases: <6 points favours nonsurgical care, >6 favours surgery; minimally invasive, cement-augmented or high-purchase fixation preferred; goal is rapid painless mobilisation.

AOSpine thoracolumbar classification: type A compression, B tension-band, C translation, plus neurological grade and modifiers; morphology reliability kappa 0.72.

Review: osteoporotic fractures are half of all osteoporotic fractures, mostly thoracolumbar and stable; up to 30% develop painful nonunion, kyphosis or neurological deficit; algorithm from bracing to instrumentation.

Standing versus supine radiographs in 41 vertebroplasty patients: dynamic mobility in 44% of patients (35% of vertebrae), all with intravertebral clefts; height increased 106% when mobile.

MRI signs of metastasis: convex posterior border, pedicle/posterior element signal, epidural or paraspinal mass; osteoporotic: band-like low signal, spared marrow, retropulsed fragment, multiple fractures.

93 acute collapses: retropulsed fragment, preserved T1 signal and band-like enhancement favour osteoporosis; convex posterior cortex, epidural mass, pedicle involvement and inhomogeneous enhancement favour malignancy.

28 patients with delayed collapse and neurological deficit: intravertebral cleft seen in 79% before deficit onset; cleft and instability are risk factors warranting immobilisation and careful observation.

Case-based review of Kummel disease (post-fracture avascular necrosis with intravertebral cleft) for primary care, with radiographic and histological workup.

350 conservatively treated OVFs: nonunion 13.5% at 6 months; predictors were thoracolumbar level, middle-column injury, and T2 confined-high or diffuse-low signal patterns.

11 cohorts: intravertebral cleft and T1 total/T2 hypointense-wide patterns predict collapse; middle-column injury predicts nonunion; retropulsion >42% canal or >15 degrees dynamic wedge change predicts neurological impairment.

Lumbar CT Hounsfield units correlate significantly with DXA BMD, T-scores and compressive strength, offering opportunistic bone quality assessment at no extra cost.

10 osteoporotic burst fractures with deficit (mean canal compromise 41%, kyphosis 28 degrees) treated by decompression and fusion: 6 improved a Frankel grade, but pain-related disability and complications were common.

Review of conservative care (bed rest, analgesia, physiotherapy, orthoses): only one RCT of bracing existed; unstandardised conservative care limits comparison with cementoplasty.

60 acute OVCF patients randomised to no brace, soft or rigid brace: 12-week ODI without brace (35.95) non-inferior to soft (37.83) and rigid (33.54); pain and compression progression similar.

284 women aged 65-85 randomised to rigid versus soft brace for 12 weeks: no difference in anterior vertebral compression at 48 weeks, EQ-5D, VAS pain or JOABPEQ.

16 studies (5 RCTs): bracing appears safe, but no compelling evidence that a rigid brace is superior to a soft brace or no brace.

Six months of a thoracolumbar orthosis (Spinomed) in women with osteoporotic vertebral fractures: back extensor strength +73%, kyphosis angle -11%, pain -38%, body sway -25%.

5 RCTs (n=350) of analgesics, orthoses and rest: very low to low quality evidence; orthoses gave medium-term pain (SMD -1.47) and disability benefit versus nothing; evidence insufficient overall.

13 trials (n=589): calcitonin reduced acute pain at rest within 1 week (MD -3.39) and with mobility at 4 weeks, but showed no convincing benefit for chronic fracture pain.

Fresh OVFs treated with daily or weekly teriparatide versus risedronate or no drug: lower VAS at 8-12 weeks, less collapse and kyphosis, fewer clefts with daily teriparatide.

98 non-operatively treated single-level OVCFs: teriparatide shortened time-to-union (HR 1.86), union 89% vs 68% at 6 months versus alendronate; no teriparatide patient needed fracture-site surgery.

78 patients with fractures <12 months randomised to vertebroplasty or sham: no benefit at any time point; 3-month pain reduction 2.6 vs 1.9 (difference 0.6, 95% CI -0.7 to 1.8).

INVEST: 131 patients randomised to vertebroplasty or simulated procedure; at 1 month no difference in RDQ (0.7) or pain (0.7); crossover 51% control vs 13% by 3 months.

Vertos II: 202 acute fractures (<6 weeks, MRI oedema) randomised open-label; VAS reduction favoured vertebroplasty by 2.6 (1 month) and 2.0 (1 year); 53% of screened patients recovered spontaneously.

VERTOS IV: 180 patients with acute OVCF randomised to vertebroplasty or sham (periosteal anaesthetic); VAS differences 0.2-0.45, not significant at any point through 12 months.

VAPOUR: 120 patients with fractures <6 weeks and NRS >=7 randomised to vertebroplasty or placebo; NRS <4 at 14 days in 44% vs 21% (difference 23 percentage points, p=0.011).

FREE: 300 patients with 1-3 acute fractures randomised to balloon kyphoplasty or nonsurgical care; SF-36 PCS improved 7.2 vs 2.0 points at 1 month (difference 5.2, p<0.0001); adverse events similar.

FREE 2-year results: kyphoplasty SF-36 PCS benefit averaged 3.24 points but lost significance by 12 months; back pain still better at 24 months; new fractures 47.5% vs 44.1%.

50 patients with acute/subacute fractures randomised to vertebroplasty or conservative care: pain fell from about 8 to 2 in both groups by 3 months; vertebroplasty relief was immediate.

Cochrane update, 21 trials: versus placebo, vertebroplasty gave 0.7/10 less pain and 1.5/23 less RDQ disability at 1 month (below MCID); no important benefit; harms uncertain.

Second ASBMR Task Force: vertebroplasty gives no clinically significant benefit over placebo regardless of pain duration; insufficient evidence for kyphoplasty; routine augmentation unsupported; anti-osteoporotic drugs cut subsequent vertebral fractures 40-70%.

RAND/UCLA multispecialty pathway: MRI preferred; augmentation appropriate with positive advanced imaging and worsening symptoms or 2-4 unfavourable features; time since fracture less relevant; follow-up must address bone health.

AAOS 2010 guideline summary: strong recommendation against vertebroplasty, moderate for 4 weeks of calcitonin, weak for kyphoplasty in neurologically intact patients; six recommendations inconclusive.

Medicare 2005-2008 (858,978 VCF patients): 4-year adjusted survival 60.8% operated vs 50.0% nonoperated (HR 0.63); kyphoplasty 62.8% vs vertebroplasty 57.3% (HR 0.77).

Medicare 2006 (68,752 VCFs): 3-year survival 42.3% nonoperative, 49.7% vertebroplasty, 59.9% kyphoplasty; kyphoplasty HR 0.80 vs vertebroplasty; authors caution selection bias.

Meta-analysis of >2 million patients: augmentation associated with 22% lower mortality up to 10 years (pooled HR 0.78, 95% CI 0.66-0.92); kyphoplasty HR 0.77, vertebroplasty HR 0.87 versus nonsurgical care.

11 RCTs (n=1048): vertebroplasty gave significantly lower pain than conservative/sham control at 1-2 weeks, 2-3 months and 12 months; largest pooled effect -1.4, near the 1.5 MCID; heterogeneity high.

27 prospective controlled studies: pain reduction BKP -5.07 and VP -4.55 versus NSM -2.17; subsequent fractures 22% NSM vs 11% VP/BKP; BKP gave more kyphosis correction and less leakage.

432 vertebroplasty patients: 19.9% developed new fractures; 41% were adjacent, occurring at median 55 vs 127 days for non-adjacent; adjacent-level relative risk 4.62.

VERTOS V: 80 patients with chronic (>3 months) painful OVCF and MRI oedema randomised to vertebroplasty or anaesthetic infiltration; 12-month VAS 3.9 vs 5.1 (difference 1.3, p=0.02) and better QUALEFFO.

SAKOS: 141 patients randomised to titanium implantable device (SpineJack) or balloon kyphoplasty; non-inferior on composite endpoint, better midline height restoration, adjacent fractures 12.9% vs 27.3%.

14 RCTs: vertebroplasty beat conservative treatment on all outcomes but showed slight-to-no advantage over placebo; benefit larger in fractures <6 weeks and versus skin-only sham.

26 patients with post-traumatic kyphosis and deficit: posterior closing wedge osteotomy (n=15) matched anterior-posterior surgery (n=11) for Frankel improvement (73% vs 64%) with less operative time, blood loss and complications.

83 thoracolumbar collapses with deficit: vertebroplasty-augmented short posterior fixation lost 4.6 degrees correction versus 8.6 without augmentation and 4.5 with anterior surgery; neurological outcomes equivalent.

50 osteoporotic collapses with deficit: anterior strut (n=32) versus posterior fixation with vertebroplasty (n=18); similar neurological recovery and correction, but 34% respiratory complications and revisions when BMD <0.60 g/cm2 anteriorly.

31 anterior Kaneda reconstructions for osteoporotic collapse with deficit: all recovered neurologically; kyphosis 31 to 13 to 21 degrees; 19% needed posterior reinforcement (severe osteoporosis or multilevel corpectomy).

41 osteoporotic patients (291 PMMA-augmented screws): VAS 9.2 to 1.5, ODI 77.5% to 44.2%, kyphosis correction loss 3 degrees, no symptomatic leakage or significant screw migration at 22 months.

Biomechanical review: PMMA and calcium ceramics enhance pedicle screw fixation in osteoporotic and revision models; fenestrated and prefill techniques comparable; extravasation risk remains.

Review: acute OVCF care is bracing, analgesia and functional restoration; augmentation for pain beyond about 2 months; open decompression and stabilisation reserved for neural compression and progressive deformity.

AACE/ACE 2020 guideline: 52 recommendations; stratifies high versus very-high fracture risk (including recent fracture), adds dual-action romosozumab, and outlines transitions between therapies.

Endocrine Society guideline: assess fracture risk with country-specific tools, integrate patient preferences, lifestyle and fall prevention; multiple agents reduce fractures with acceptable safety.

BHOF Clinician's Guide: any fracture after 50 signals imminent risk; vertebral fracture carries 5-fold risk of further vertebral and 2-3-fold risk of other fractures; diagnosis persists despite BMD improvement.

FIT vertebral fracture arm (2027 women): alendronate reduced new morphometric vertebral fractures 8.0% vs 15.0% (RR 0.53), clinical vertebral fractures RR 0.45, hip fractures RH 0.49 over 36 months.

1637 women with prior vertebral fractures: teriparatide 20 ug reduced new vertebral fractures 14% to 5% (RR 0.35) and nonvertebral fractures 6% to 3% (RR 0.47) over median 21 months.

FREEDOM (7868 women): denosumab 60 mg 6-monthly reduced new vertebral fractures 7.2% to 2.3% (RR 0.32, 68% reduction), hip fractures by 40% and nonvertebral fractures by 20% over 36 months.

ARCH (4093 women with fragility fracture): romosozumab then alendronate cut new vertebral fractures 48% (6.2% vs 11.9%) and hip fractures 38%; more year-1 serious cardiovascular events (2.5% vs 1.9%).

VERO (1360 women with severe osteoporosis): teriparatide versus risedronate reduced new vertebral fractures 5.4% vs 12.0% (RR 0.44) and clinical fractures 4.8% vs 9.8% (HR 0.48) at 24 months.

ISCD 2019 position: orthopaedic patients with fracture after 50, inflammatory arthritis, steroids or renal disease should have DXA before surgery; opportunistic CT Hounsfield units are reliable with moderate DXA correlation.

AOA critical issues: bone health optimisation means assessing bone status, correcting metabolic deficits and starting treatment; assess all patients >=50, and consider delaying elective surgery 3 months to treat.

FRAX models 10-year probability of hip and major osteoporotic fracture from clinical risk factors with or without femoral neck BMD; probabilities range about 100-fold across age and risk factors.

96,887 US hip fracture patients: osteoporosis medication within 12 months only 28.5%, falling from 40.2% in 2002 to 20.5% in 2011; prior use was the strongest predictor (HR 7.45).

74 controlled studies: fracture liaison services raised BMD testing by 24 and treatment initiation by 20 percentage points, cut re-fracture 13.4% to 6.4% and mortality 15.8% to 10.4%.

NASS evidence-based guideline summary answering 29 questions on diagnosis and treatment of adult OVCF (literature to September 2020), with graded recommendations for imaging, bracing, augmentation and pharmacotherapy.

ACR Appropriateness Criteria 2022 update rating imaging and treatment (including augmentation) for osteoporotic, traumatic and malignant VCF scenarios using RAND/UCLA and GRADE methods.

Modified Delphi international consensus (27-31 experts, 90.7% agreement) giving biopsychosocial recommendations on pain, nutrition, bracing, safe movement and exercise for acute and chronic vertebral fractures.

ACP living guideline: bisphosphonates first line (strong, high certainty); denosumab second line; romosozumab or teriparatide followed by bisphosphonate only for females at very high fracture risk.

Network meta-analysis of 69 RCTs (>80,000 women): all agents reduce vertebral fractures; anabolics beat bisphosphonates for clinical (OR 1.49) and vertebral fractures irrespective of baseline risk.

USPSTF 2025: screen women >=65 and postmenopausal women <65 at increased risk for osteoporosis (B recommendations); evidence insufficient for men (I statement).

DGOU systematic review of 13 studies (5 RCTs) and 8 reviews: no superiority of spinal orthoses over no orthosis in OVF; non-rigid orthoses favoured over rigid; general recommendation not possible.

AO Spine Knowledge Forum Trauma validation of the AO Spine-DGOU OF classification (21 raters, 33 cases): inter-rater kappa 0.48-0.50 (moderate), intra-rater kappa 0.74 (substantial).

EOFTT: 707 inpatients in 16 German centres; mean age 75, 51% recalled no trauma, neurological deficit 3.1%, deformity progression linked to T-score <-3 (OR 1.97); 81.6% treated surgically, mostly kyphoplasty.

6 RCTs in patients with prevalent OVF: anabolic agents (teriparatide, romosozumab) reduced new vertebral fractures versus bisphosphonates (RR 0.57, high certainty) without proven faster fracture healing (RR 1.23).

20 studies of cement-augmented pedicle screws in osteoporotic spines: screw loosening 2.0%, any cement leakage 21.8%, symptomatic leakage 1.2%, symptomatic pulmonary embolism 0.8%, adjacent fracture 3.3%.

22 studies including 2 RCTs: pooled mortality HR 0.91 (95% CI 0.73-1.15) for cement augmentation versus conservative care, not significant; registry survival advantage likely reflects confounding.

Meta-analysis for Kummell disease: short-segment fixation shortens operation (-18 min) and blood loss (-82 mL); long-segment better maintains kyphosis (3.2 degrees less loss); complications, VAS and ODI similar.

8 studies (289 patients) of kyphoplasty alone for osteoporotic thoracolumbar burst fractures: VAS 6.8 to 1.1, kyphosis 21.7 to 3.2 degrees; cement leakage 7.7-45.4%; burst morphology not an absolute contraindication.

16 RCTs: vertebroplasty improved short-term pain 1.31 points more than nonoperative care; vertebroplasty and kyphoplasty equivalent; no increase in adjacent-level fractures (log OR -0.16); QOL similar.

NOGG 2024 UK guideline update: clarifies very-high-risk definition, adds abaloparatide, endorses HRT first line in younger high-risk postmenopausal women, and covers vertebral fracture management and fracture liaison models.

Three-column concept from 412 CT-imaged injuries: compression, burst, seat-belt and fracture-dislocation; instability begins with failure of two columns.

Comprehensive AO classification of 1,445 injuries into types A compression, B distraction and C rotation with 27 subgroups; A1 fractures were 34.7% of all injuries.

Load-sharing classification scoring comminution, fragment apposition and kyphosis correction to predict failure of short-segment posterior fixation.

TLICS: injury morphology, posterior ligamentous complex integrity and neurological status scored and summed; ≤ 3 nonoperative, ≥ 5 operative, 4 indeterminate.

Five surgeons scored 71 cases with the TLISS: over 90% agreement with the recommended treatment; MRI and morphology suggested to improve reliability.

25 injuries scored with TLISS then TLICS by 5 surgeon groups: both reproducible and valid; TLICS predicted treatment in 95% of cases.

Interobserver reliability of the TLISS improved significantly after 7 months of use at one institution, showing the system can be taught.

49 surgically treated patients: TLICS recommendation matched surgery in 96%; TLICS correlated with neurological status and Magerl/AO type.

Thoracolumbar AOSpine Injury Score from 100 surgeons: A1 1, A2 2, A3 3, A4 and B1 5 points; N3 elevated to 4 points; M1 adds 1 point.

100 naive surgeons classified 25 cases twice: overall κ 0.56; type A 0.80, B 0.68, C 0.72; A4 least reliable (κ 0.19); intraobserver κ 0.68.

Introduces TLICS as a simpler, more reliable alternative to the Denis and AO systems for surgeons and non-surgeons.

Places TLICS in the Audigé three-phase validation framework: phases 1 and 2 complete, prospective clinical validation outstanding.

19 surgeons classified 31 fractures twice: Denis and AO systems showed only moderate reliability, falling with subclassification.

Five observers, 53 injuries on CT and MRI: fair-to-moderate reliability for both AO and Denis, with Denis more reproducible than AO.

Textbook chapter reviewing the evolution from Denis and AO to TLICS, the scoring domains, the surgical-approach table and the validation studies.

Elderly, osteoporotic, ankylosed, obese, polytrauma and paediatric patients alter TLICS-based decisions: fixation failure, transverse fractures, brace failure and mobilisation needs.

Chapter on augmentation, combined instrumentation, osteotomy and complications; a T-score below −3.5 to −4 makes any fixation difficult and medical treatment is mandatory.

Chapter on Magerl A1 impaction fractures: stable with intact posterior column; osteoporosis is the leading cause; bracing has very low quality evidence.

VERT-NA (2,458 women, 3 y): risedronate 5 mg daily reduced new vertebral fractures 16.3% to 11.3% (41%; 65% in year 1) and nonvertebral fractures 8.4% to 5.2% (39%).

HORIZON Pivotal Fracture Trial (7,765 women, 3 y): yearly zoledronic acid 5 mg reduced morphometric vertebral fractures 10.9% to 3.3% (RR 0.30, 70%), hip fractures 2.5% to 1.4% (HR 0.59, 41%) and nonvertebral fractures 25%; serious atrial fibrillation 1.3% vs 0.5%.

BONE (2,946 women, 3 y): daily oral ibandronate reduced new vertebral fractures 9.6% to 4.7% (62%), intermittent dosing 50%; no reduction in nonvertebral fractures overall.

FREEDOM and extension post hoc (1,001 denosumab discontinuers): vertebral fracture rate rose from 1.2 to 7.1 per 100 participant-years after stopping, with multiple vertebral fractures in 60.7% of those who fractured; prior vertebral fracture was the main risk factor.

ACTIVE (2,463 women, 18 mo): abaloparatide 80 ug daily reduced new vertebral fractures 4.22% to 0.58% (86%) and nonvertebral fractures 43% (HR 0.57); hypercalcaemia 3.4% vs 6.4% with teriparatide (fracture rates from the paper's results table; the PubMed abstract shows them as a table).

FRAME (7,180 women): romosozumab 210 mg monthly for 12 months reduced new vertebral fractures 1.8% to 0.5% (73%) and clinical fractures 36%; nonvertebral fractures not significantly reduced at 12 months; after transition to denosumab, vertebral fractures 2.5% vs 0.6% at 24 months (75%).

MORE (7,705 women, 3 y): raloxifene 60 mg reduced new vertebral fractures 10.1% to 6.6% (RR 0.7) with no reduction in nonvertebral fractures (RR 0.9); venous thromboembolism RR 3.1.

Kaiser Permanente cohort (196,129 women, 277 atypical femoral fractures): risk rose with bisphosphonate duration, HR 8.9 at 3 to <5 years and 43.5 at 8 years or more, fell rapidly after stopping, and was higher in Asian than White women; after 3 years of use, 149 hip fractures were prevented per 2 atypical fractures in White women.

FLEX (1,099 women after about 5 years of alendronate): a further 5 years vs placebo; stopping caused modest BMD decline, no difference in nonvertebral fractures (RR 1.00), but more clinical vertebral fractures (5.3% vs 2.4%, RR 0.45 for continuing).

International Task Force consensus on osteonecrosis of the jaw: incidence 0.001% to 0.01% at osteoporosis doses of bisphosphonate or denosumab versus 1% to 15% at oncology doses; risk factors include glucocorticoids, dental extraction and poor oral health.

Endocrine Society 2020 update: romosozumab for up to 1 year is recommended for postmenopausal women at very high fracture risk (severe osteoporosis or multiple vertebral fractures), followed by an antiresorptive; not within 1 year of myocardial infarction or stroke because of the cardiovascular signal in ARCH; the label carries a boxed warning.

FREEDOM open-label extension (up to 10 years of denosumab, 4,550 women): yearly new vertebral fracture incidence stayed at 0.9 to 1.9% and nonvertebral at 0.8 to 2.6%; lumbar spine BMD +21.7% and total hip +9.2% from baseline; 13 cases of osteonecrosis of the jaw and 2 atypical femoral fractures.

DATA-Switch (94 women, 4 y): teriparatide then denosumab raised spine BMD 18.3% and total hip 6.6%; denosumab then teriparatide caused transient bone loss at the hip and progressive loss at the radius; sequence matters, anabolic first.

FRAME post hoc: 1 year of romosozumab raised spine BMD 13% and total hip 7%; the T-score gain after 1 year of romosozumab plus 1 year of denosumab approximated that achieved by 7 years of continuous denosumab.

STRUCTURE (436 women transitioning from 3 or more years of oral bisphosphonate): total hip BMD +2.6% with romosozumab versus -0.6% with teriparatide at 12 months; romosozumab also improved estimated hip strength.

57 osteoporotic women with 1- to 2-level instrumented posterolateral fusion: fusion rate 82% with daily teriparatide versus 68% with risedronate; mean time to union 8 versus 10 months.

Multicentre RCT (75 osteoporotic women, PLIF or TLIF): 6 months of weekly teriparatide after surgery gave significantly higher CT fusion rates at 4 and 6 months than no teriparatide.

RCT of 47 stable thoracolumbar burst fractures without deficit, mean follow-up 44 months: kyphosis and canal remodelling similar (canal 39% to 22% operative vs 34% to 19% nonoperative), pain and return to work similar, less disability and fewer complications with nonoperative care.

80 thoracolumbar junction burst fractures without deficit, brace (47) vs short-segment posterior fixation (33): surgery gave less pain to 3 months and better function to 6 months, then no difference; 17 degrees of correction was gradually lost; nonoperative retropulsion remodelled from 34% to 15%; hospital charges four-fold with surgery.

Multicentre RCT of 34 thoracolumbar fractures without deficit (16 nonoperative, 18 short-segment posterior), mean 4.3 years: operative group had less kyphosis, better pain and function scores and more return to the original job; the dissenting trial.

Multicentre equivalence RCT of 96 AO A3 burst fractures T11 to L3 without deficit, TLSO vs no orthosis: Roland-Morris disability at 3 months 6.8 vs 7.7, within the equivalence margin; 6 patients needed surgery.

18 burst fractures with intact posterior column treated by early ambulation in a TLSO: hospital stay 10 days, kyphosis 19 to 20 degrees at 19 months, 15 of 18 with little or no pain, retropulsed bone resorbed on CT, no neurological deterioration.

18 thoracolumbar burst fractures, short-segment (9) vs long-segment (9) posterior fixation: long segment held kyphosis, sagittal index and height better; short-segment implant failure 55%; clinical outcome equal; long segment took longer with more blood loss.

MRI at 18 or more months of 75 thoracolumbar fractures: six disc patterns; recurrent kyphosis after posterior fixation arose from disc material creeping into the central end-plate depression left by ligamentotaxis, not from degeneration or instability.

Phase II study of 20 thoracolumbar burst fractures: posterior short-segment fixation plus balloon reduction and calcium phosphate cement restored central and anterior body height to 78% and 91% of intact; cement leakage in 5 of 20 without sequelae.

Cadaveric L1 burst model with short-segment pedicle screws: transpedicular calcium phosphate cement reduced screw bending moments 59% in flexion and 38% in extension and raised flexion-extension stiffness 40%.

Systematic review of 69 clinical studies of vertebroplasty and kyphoplasty: pain relief in 87% and 92%, kyphosis correction 6.6 degrees with kyphoplasty, cement leakage per vertebra 41% vertebroplasty vs 9% kyphoplasty, subsequent fracture rates similar to untreated osteoporotic patients with a prior fracture.

Updated systematic review and meta-analysis of balloon kyphoplasty (8 comparative studies, 35 case series): better pain, function, height and kyphosis than medical management to 3 years, at least as effective as vertebroplasty; serious adverse events rare (cement leakage 8.1%, symptomatic 0.09%; new fractures 11.1%, 9.4% adjacent).

67 thoracolumbar fractures in 49 patients: plain radiographs classified the fracture correctly in 88%; 25% of injuries would have been misdiagnosed on radiographs alone, mostly bursts read as compression fractures; CT recommended.

28 T11 to L2 fractures imaged supine then erect: Cobb angle 11 to 18 degrees and anterior compression 34% to 46% on standing; 7 of 28 (25%) converted to surgical management.

Systematic review of 93 articles, 345 ankylosing spondylitis and 55 DISH patients with spinal fractures: neurological deficit on admission 67% and 40%, complications 51% and 33%, 3-month mortality 18% and 20%.

299 spine fractures fixed within 3 days (174) or later (125): hospital stay 14 vs 21 days; pneumonia similar; mortality 6.9% early vs 2.5% late (non-significant trend); early fixation shortened ICU stay and cut pneumonia in thoracic fractures with cord injury.

Whole-spine MRI in 127 patients with acute vertebral body trauma: 77% had a second injured level; about a third had non-contiguous fractures.

50 postmenopausal women, 10-year follow-up of a 2-year back-extensor strengthening programme: vertebral fractures 1.6% (6 of 378 vertebrae) vs 4.3% (14 of 322) in controls, relative risk 2.7 in controls; back-extensor strength and BMD higher with exercise.

Prospective nonrandomised study, 40 kyphoplasty vs 20 conservative, painful osteoporotic fractures including chronic ones, 6 months: vertebral height +12.1% vs -8.2%, pain score improved 26 to 44 vs unchanged, function improved, doctor visits 3.3 vs 8.6.

38 patients (47 levels) after kyphoplasty, mean 8 months: 26% had new fractures (17 in 10 patients), 8 within 2 months and all adjacent; later fractures were remote; front-loaded adjacent risk.

38 vertebroplasty patients: 14 new fractures, 10 with cement leakage into the disc vs 4 without; 58% of vertebrae adjacent to a leaked disc fractured vs 12% adjacent to an intact disc.

235 thoracolumbar fractures, 117 instrumented: postoperative wound infection 10% (9 deep), 41% in patients with complete neurological injury; 67% polymicrobial.

Five-year follow-up of open (11) vs percutaneous (10) posterior stabilisation of thoracolumbar burst fractures: lower blood loss percutaneous, identical operative time, loss of correction, Hannover score and SF-36.

18 patients, mean age 64, lumbar burst and severe compression fractures: minimal-invasive short posterior instrumentation plus balloon kyphoplasty with calcium phosphate: kyphosis 16 to 2 degrees, anterior height 0.57 to 0.87, canal 25% to 19%, 45 minutes, 70 mL; cement leakage 4 of 18; no implant failure or loss of correction at 22 months.

Cadaveric study and 12 patients: adding pedicle screws at the fractured level to short-segment fixation increased torsional stiffness; in patients kyphosis improved 15 degrees with 5 degrees loss and anterior height 58% to 89% to 78%.

Textbook chapter reviewing the randomised and comparative studies of burst fractures without deficit (Wood, Shen, Alanay, Tezeren) and the Spine Trauma Study Group recommendations: bracing 6 to 12 weeks is a reasonable option, either approach is acceptable when surgery is chosen, and posterior approaches are favoured in the elderly with poor bone.