Complications of vertebral augmentation in metastatic spinal fractures: a systematic review of multiple study designs Authors: Maria Clara Correia, Ana Margarida Vieira, Filipe Duarte, André Rodrigues Pinho, Francisco Serdoura, Maria João Leite Published: 2026 ABSTRACT BackgroundVertebral augmentation techniques, including vertebroplasty and kyphoplasty, are widely used to relieve pain and restore spinal stability in patients with metastatic vertebral lesions. Despite their widespread use, the full spectrum and clinical significance of procedure-related complications remain incompletely characterised. This systematic review aims to summarise and characterise all reported complications associated with vertebral augmentation for metastatic thoracolumbar disease.MethodsA comprehensive search of PubMed/MEDLINE, Scopus, and Web of Science databases was performed. Studies reporting complications associated with vertebral augmentation procedures for metastatic thoracolumbar spinal fractures were included. Randomised controlled trials, observational studies, case series, and case reports were eligible. Due to substantial heterogeneity across studies, a descriptive synthesis was performed.ResultsA total of 76 studies were included, comprising 3234 patients and at least 5737 vertebral levels treated. The overall weighted complication rate was 25.5%, while the weighted rate of symptomatic complications was 3.41%. Cement leakage was the most frequently reported complication, with a weighted mean incidence of 20.7%, although the majority were asymptomatic. Other complications included postprocedural pain (1.77%), pulmonary cement embolism (1.08%), and adjacent level fractures (0.79%). Studies using CT imaging reported significantly higher cement leakage rates (28.6% vs 11.9%, p = 0.004). Case reports described rare but potentially severe complications, including needle-tract tumour seeding, tumour extrusion, and cement migration to cardiac structures.ConclusionsThis review presents a comprehensive and detailed description of all complications of vertebral augmentation for treatment of metastatic spine disease and reinforce its safety. Differences in imaging protocols and reporting standards contribute to variability in reported complication rates. Further prospective studies with standardised complication reporting are needed. Introduction Advances in oncological therapies have improved survival rates and consequently increased the prevalence of metastatic disease and its complications, with 70% of patients developing spinal metastases [1], [2]. Metastatic involvement of the spine disrupts normal bone architecture, increasing its susceptibility to vertebral compression fractures (VCFs), associated with axial pain, mechanical instability, spinal deformity, and even neurological compromise. These manifestations may impair mobility, quality of life and increase analgesic use and hospitalisation [1], [2]. Treatment strategies for management of metastatic spinal disease are mainly palliative and aim to control pain, maintain or restore spinal stability, preserve neurological function, and optimise quality of life [1], [3]. Minimally invasive vertebral augmentation techniques, as percutaneous vertebroplasty and balloon kyphoplasty, have emerged as important treatment options for painful cancer-related VCFs [1], [4]. These procedures involve the percutaneous injection of polymethylmethacrylate (PMMA), into the fractured vertebral body, improving mechanical stability and load-bearing capacity [5]. Multiple studies have shown that vertebral augmentation provides effective pain control, reduces opioid consumption, and improves functional outcomes. [1], [4], [6], [7] . These benefits are particularly relevant in palliative care, where rapid symptom control and preservation of autonomy are essential. [1], [3]. The increasing use of vertebral augmentation procedures has drawn attention to a range of perioperative and postprocedural complications. These complications vary from minor and self-limiting events to rare but potentially severe and life-threatening outcomes. Cement leakage is the most frequently reported complication, with incidence rates varying widely across studies and often exceeding 20% [5], [8]. Although most cases of cement leakage are asymptomatic, it may result in clinically significant consequences, such as nerve compression, pulmonary and intracardiac embolism, adjacent vertebral fractures, and local pain, highlighting the need for careful procedural planning [8]. Recent systematic reviews have attempted to identify risk factors for cement-related complications, but findings remain inconsistent. [8] Other reported complications include infection, paravertebral haemorrhage or haematoma, pedicle fractures, allergic reactions to bone cement, postoperative pain, and deep vein thrombosis. [5], [9]. These events are uncommon and often described only in case reports or small case series. This systematic review aims to comprehensively evaluate both common and rare complications associated with vertebral augmentation for metastatic spinal disease, and to clarify their incidence, risk factors, and clinical relevance. Methods Study design A systematic review including multiple study designs (SR-MSD) was conducted to provide a comprehensive overview of complications associated with vertebral augmentation for metastatic spinal disease. Randomised controlled trials (RCTs), observational studies, case series, and case reports were included to capture both common and rare complications. Despite their lower level of evidence, case reports were included as they may provide valuable insights in oncologic spine care, where high-quality trials remain limited. Search strategy and research questions This systematic review was conducted in accordance with the PRISMA guidelines [10]. A systematic literature search was performed in PubMed/MEDLINE, Scopus, and Web of Science from database inception to July 2025. The search strategy combined terms related to metastatic spinal disease, vertebral augmentation procedures (including vertebroplasty and kyphoplasty), and postoperative complications. The complete search strategies for each database are provided in Appendix A. Study selection and eligibility criteria All records identified through the database search were imported into a reference management software, and duplicates were removed. Study selection was conducted independently by two reviewers (MCC and AMV), through title/ abstract screening followed by full-text review. Disagreements between reviewers were resolved by a third reviewer (MJL). No restrictions were imposed regarding year of publication or language. The inclusion criteria were patients with thoracolumbar spinal metastatic disease submitted to vertebral percutaneous augmentation procedures (vertebroplasty, kyphoplasty); studies reporting data on complications; RCTs, observational clinical studies (cohort studies, case series, case–control studies) and case reports; and investigations conducted in a clinical setting. The exclusion criteria were: primary bone tumours or osteoporotic fractures; vertebral augmentation of cervical (different surgical approach) or sacral spine (irregular shape); neurological compromise before surgery; vertebral augmentation techniques performed simultaneously with other spinal interventions (surgical fixation, bone tumour ablation procedures, conjoined use of bone filling mesh, or others); and systematic reviews, meta-analyses, scoping reviews, abstracts, editorials, book chapters, conference papers, commentaries, opinion articles, and narrative reviews without original data. If complications attributable exclusively to vertebroplasty or kyphoplasty were clearly documented, the articles were included, and only those patients were considered in data extraction. Case reports describing complications related to percutaneous vertebroplasty or kyphoplasty were included, whereas reports focusing exclusively on procedural benefits or efficacy without mention of complications were excluded. In comparative studies (RCTs, case-control studies, and two-arm cohort studies), data extraction was tailored to the relevance of the comparison groups to the outcome of this review. When appropriate, only the vertebroplasty or kyphoplasty treatment arms were included, whereas in studies where the comparison was not relevant to the evaluation of complications, the overall study population was considered. As the primary objective was to characterise complications rather than compare interventions, all studies were analysed descriptively. In one study, both arms were analysed separately and treated as independent series, once the study specifically compared vertebroplasty and kyphoplasty [11]. The absence of reported complications in some studies was interpreted as not reported rather than the absence of events, and those studies were therefore excluded. Data extraction and synthesis Data extraction was independently performed by two reviewers (MCC, AMV) using a standardised data extraction form. Any discrepancies in data extraction were resolved through discussion and consensus between the reviewers. The following variables were collected: study characteristics (author, year, country, study period, design); patient characteristics (number of patients, treated levels, age, tumour sites, distribution and characteristics of involved levels); treatment characteristics (type of vertebral augmentation procedure, anaesthesia, fluoroscopy/CT guided, volume of bone cement, number of treated vertebras per session); reported postoperative complications (frequency, severity, risk factors; leakage type and symptomatic events). Due to significant heterogeneity of the studies included information a quantitative meta-analysis was not feasible, and a descriptive synthesis of complications was performed. Risk of bias assessment Methodological quality was evaluated using the Joanna Briggs Institute (JBI) [12] critical appraisal tools according to study design (JBI checklist for randomised controlled trials, cohort studies, and case-control studies). Case series and case reports were not formally assessed for publication bias, as their study designs are considered inherently susceptible to selection and reporting bias. Results Study selection After a full electronic search, a total of 1224 references were identified. After duplicate removal, 682 records were excluded based on title and abstract screening. 134 full text articles were accessed for eligibility, of which 70 full text records were excluded according to the exclusion criteria, leaving 59 articles to be included finally. 17 additional studies were identified following backward citation. In total, 76 articles were included in this review. The study selection process is illustrated in the flow diagram in Fig. 1.Fig. 1Flowchart of study selection.Fig. 1 Flowchart of study selection. Characteristics of included studies Most of the included studies consisted of descriptive designs. According to the Oxford Centre for Evidence-Based Medicine classification [13], most included studies were Level IV evidence (41 case series and 16 case reports), followed by Level III studies (16 cohort studies and 1 case-control study), with only one Level II study (1 non-randomised clinical trial) and one Level I randomised control trial. The studies were published between 1996 and 2024 and originated from 17 different countries. The main characteristics of the included studies are summarised in Table 1, Table 2.Table 1Characteristics of included observational studies.StudyStudy typeSample sizeLevels treated & mean level per surgeryMean age & genderPrimary tumour typesProcedureFollow-up1. Barr, JD et al. [15]2000, USACohort single arm8 patients13 vertebraeThoracolumbar spine--no informationMetastatic diseaseVertebroplasty*CT or fluoroscopy guided*With DSA12 months (2–35)2. Berenson, J et al. [57]2011,Cancer Patient Fracture Evaluation (CAFE) studyRCT68 patients108 vertebraeThoracolumbar spineT5-L5-64,8 years41% menMultiple myeloma: 22Breast: 16Lung: 7Prostate: 4Other: 19Kyphoplasty-no information3. Burton, AW et al.[49]2005, USACase series3 patients6 vertebraeThoracolumbar spineT12-L52 (1–3) levels/ session64 years33% menLung: 1Breast: 1Parathyroid: 1VertebroplastyFluoroscopy guidedUp to 3 months or death4. Caudana, R et al. [16]2008, ItalyCohort single arm39 patients62 vertebraeThoracic (56,4%) and lumbar (43,6%) spine--62 years36% menBreast: 17Lung: 7Kidney: 4Multiple myeloma: 4Colon: 3Prostate: 1Oesophagus: 1Liver: 1Neuroendocrine tumours: 1VertebroplastyCT guided3,39 months (0,3-11,5)5. Chen, G et al. [22]2013, ChinaCase series6 patients6 vertebraeThoracic (50%) and lumbar (50%) spineT9-L51 level/ session59,3 years50% menBreast: 2Lung: 2Liver: 1Prostate: 1KyphoplastyFluoroscopy guided35 months (16–96)6. Chen, L et al. [58]2011, ChinaCase series4 patients4 vertebraeThoracic (25%) and lumbar (75%) spineT8-L4-57,75 years75% menLung: 2Prostate: 1Pancreas: 1VertebroplastyFluoroscopy guided18 months (14–24)7. Chen, KY et al. [40]2009, TaiwanCohort single arm31 patients42 vertebraeThoracic (35,7%) and lumbar (64,2%) spineT3-L5-67 years39% menLung: 10Breast: 7Gastrointestinal tract: 4Hepatobiliary: 3RCC: 1Cervical: 1Multiple myeloma: 1Pancreatic: 1Prostate: 1Unknown: 1Thyroid: 1Vertebroplasty-no information8. Chew, C et al. [59]2011, Scotland, UKCohort single arm128 patients264 vertebraeThoracolumbar spine-Up to 4 levels/ session60 years53% menMultiple myeloma: 41Breast: 22Lung: 16Lymphoma: 11Renal: 8Prostate: 5Others: 25VertebroplastyFluoroscopy guidedno information9. Cui, Y et al. [52]2022, ChinaComparative retrospective cohort230 patients530 vertebraeThoracic (50,6%) and lumbar spine (49,4%)--66 years57% menLung: 172Breast: 82Prostate: 78Other: 198VertebroplastyFluoroscopy guidedno information10. Da Silva Herrero, CFP et al [60]2014, BrazilCohort single arm22 patients34 vertebraeThoracic (30%) and lumbar (70%) spineT7-L5-56 years36% menMultiple Myeloma: 8Breast: 6Plasmacytoma: 2Paraganglioma: 1Oropharynx: 1Fibro histiocytoma: 1Cervix: 1Multiple angiomatosis: 1Gastric: 1KyphoplastyFluoroscopy guided8,5 months11. Diamond, TH et al. [41]2004, AustraliaCase series7 patients14 vertebraeThoracolumbar spineT5-L1Up to 7 levels/ session69 years43% menMultiple Myeloma: 7VertebroplastyFluoroscopy guided6 weeks12. Dudeney, BS et al [42]2002, USACohort single arm18 patients55 vertebraeThoracolumbar spineT6-L5-63,5 years-Multiple Myeloma: 18KyphoplastyFluoroscopy guidedno information13. Eleraky, M et al [61]2011, GermanyCohort single arm14 patients30 vertebraeThoracic spineT1-T51–3 levels/session62 years57% menMultiple Myeloma: 6Lung: 3Breast: 2Melanoma: 1Leiomiosarcoma: 1Bladder: 1KyphoplastyFluoroscopy guided18 months(minimum 12 months)14. Farrokhi, M et al. [23]2012, IranCohort single arm25 patients25 vertebraeThoracic (28%) and lumbar (72%) spineT3-L4-53,5 years44% menBreast: 14Lung: 9Prostate: 2VertebroplastyFluoroscopy guidedno information15. Fourney, DR et al. [38]2003, USACohort single arm56 patients97 vertebraeThoracolumbar spineT3-L51,7 (1–5) levels/ session64 years (median)55% menMultiple Myeloma: 21Breast: 9Lung: 6Lymphoma: 6Prostate: 5Renal cell: 2Gastric: 2Other: 5Kyphoplasty andVertebroplastyFluoroscopy guidedMedian 4,5 months16. García-Maroto, R et al. [62]2015, SpainCohort single arm75 patients122 vertebraeThoracic (36,9%) and lumbar (63,1%) spine--68 years33,3% menMultiple Myeloma: 30Breast: 14Lung: 8Colon: 8Prostate: 3Others: 12KyphoplastyFluoroscopy guided11 months (3–36)17. Garland, P et al. [43]2011, England, UKCohort single arm26 patients57 vertebraeThoracolumbar spineT6-L51–3 levels/ session59 years-Multiple Myeloma: 26VertebroplastyFluoroscopy guided19 months18. Guzik, G [44]2017, PolandCohort single arm73 patients112 vertebraeThoracolumbar spine-Up to 3 levels/ session71 years36% menMultiple Myeloma: 73VertebroplastyFluoroscopy guidedno information19. Ha KY et al[24]2015, KoreaComparative cohort27 patients48 vertebraeThoracolumbar spine--66,6 years41% menMetastatic diseaseKyphoplasty andVertebroplastyFluoroscopy guided16,8 months (6–33)20. He, S et al. [63]2020, ChinaComparative cohort147 patients190 vertebraeThoracolumbar spineT6-L5-70 years35% menMetastatic diseaseKyphoplastyFluoroscopy guided12 months21. Huber FX et al. [25]2009, GermanyCohort single arm76 patients190 vertebraeThoracolumbar spineT6-L52,5 levels/ sessionUp of 5 levels/ session62 years (median)59% menMultiple Myeloma: 76KyphoplastyFluoroscopy guidedno information22. Jang JS et al.[45]2002, South KoreaCase series27 patients72 vertebraeThoracolumbar spine---67% menMultiple Myeloma: 27VertebroplastyFluoroscopy guidedno information23. Julka, A et al. [46]2014, USACohort single arm32 patients76 vertebraeThoracic (45%) and lumbar (55%) spine--64,3 years56% menMultiple Myeloma: 32KyphoplastyFluoroscopy guided24 months24. Kirkegaard, AO et al. [64]2018, DenmarkCohort single arm30 patients81 vertebraeThoracic (45,7%) and lumbar (54,3%) spine-Single level/session in 20% of patients and multiple levels/ session in 80% of patients69,1 years67% menMultiple Myeloma: 13Lung: 6Chronic lymphocytic leukaemia: 5Lymphoma: 2Gastric: 1Prostate: 1Colon: 1Ovarian: 1VertebroplastyFluoroscopy guidedno information25. Kırcelli et al. [65]2018, TurkeyCohort single arm72 patients82 vertebraeThoracolumbar spine-(1–3) levels/ session78,93 years24% menMetastatic diseaseKyphoplastyFluoroscopy guided18,91 months26. Kobayashi, T et al. [17]2009, JapanProspective multi-institutional single-arm Phase I/II clinical study33 patients42 vertebraeThoracic (42,9%) and lumbar (57,1%) spine-1–3 levels/ session62 years48% menLung: 7Breast: 7Colorectal: 7Liver: 4Multiple Myeloma: 3Pancreatic: 2Tongue: 1Oesophageal: 1Skin: 1VertebroplastyFluoroscopy and CT guidedno information27. Kodama, H et al. [18]2007, JapanCohort single arm6 patients8 vertebraeThoracic (50%) and lumbar (50%) spineT1-L11–2 levels/ session65,2 years100% menHepatocellular: 6VertebroplastyCT guidedno information28. König MA et al. [39]2012, UKCase series11 patients23 vertebraeThoracolumbar spineT2-L3-65 years36% menBreast: 3Non-Hodgkin lymphoma: 2Gastric: 1 Cervix:1Prostate: 1Lung: 1Bladder: 1Multiple Myeloma: 1KyphoplastyFluoroscopy guided74,4 weeks29. Lee, B et al [19]2009, England, UKCohort single arm18 patients33 vertebraeThoracolumbar spine-1–3 levels/ session69,3 years39% menBreast: 7Prostate: 4Renal: 2Lung: 1Oesophageal:1Adenoid cystic carcinoma: 1Cervix: 1Other: 1VertebroplastyFluoroscopy and CT guided12 months (3–38)30. 1 Li, Z et al. [11]2014, ChinaComparative cohort1° (KP)42 patients83 vertebraeThoracolumbar spine-1–4 levels/ session68,4 years43% menBreast: 16Lung: 9Prostate: 6Stomach: 6Cervix: 3 Rectal: 2KyphoplastyFluoroscopy guided1 year30.2 Li, Z et al. [11]2014, ChinaComparative cohort2° (VP)38 patients76 vertebraeThoracolumbar spine-1–4 levels/ session65,2 years50% menBreast: 14Lung: 9Prostate: 8Stomach: 5Ovarian: 2VertebroplastyFluoroscopy guided1 year31. Lim, B et al. [66]2009, South KoreaComparative cohort102 patients185 vertebraeThoracic (21%) and lumbar (79%) spineT6-L5-55 years59% menBreast: 24Multiple Myeloma: 19Lung: 16Liver: 10Rectal: 6Unknown primary: 5Stomach: 4Colon: 4Cervix: 2Other: 12VertebroplastyFluoroscopy guided12,2 months32. Liu, H et al. [67]2017, ChinaComparative cohort234 patients288 vertebraeThoracolumbar spine--63,7 years74% menLung: 100Breast: 97Prostate: 14Stomach: 11Rectal: 8Liver: 5Renal: 5Cervix: 2Ovarian: 1KyphoplastyFluoroscopy guided12 months33. Liu, Y et al. [20]2023, ChinaCase-control30 patients30 vertebraeThoracic (20%) and lumbar (80%) spine--59,5 years67% menLung: 30VertebroplastyCT guidedno information34. Masala, S et al. [47]2008, ItalyCohort single arm64 patients198 vertebraeThoracic (47%) and lumbar (53%) spine-3,09 levels/ session71,4 years (median)53% menMultiple Myeloma: 64VertebroplastyFluoroscopy guided6 months35. Moulin, B et al. [26]2020, JapanCohort single arm50 patients397 vertebraeThoracic (60%) and lumbar (40%) spine-7,9 (6–13) levels/ session66 years27% menMultiple Myeloma: 13Breast: 12Prostate: 6Lung: 6Other: 13VertebroplastyFluoroscopy guided401 days36. Nas, OF et al. [68]2015, TurkeyCohort single arm52 patients95 vertebraeThoracic (43,2%) and lumbar (56,8%) spine-1–5 levels/ session63,98 years46% menBreast: 14Lung: 9Prostate: 7 Renal: 5Stomach: 5Others: 12VertebroplastyFluoroscopy guided3 months37. Pflugmacher, R et al [48]2005, GermanyCohort single arm20 patients48 vertebraeThoracolumbar spineT6-L5-62,4 years100% menMultiple Myeloma: 20KyphoplastyFluoroscopy guidedMinimum of 1 year38. Pflugmacher, R et al2007, [69]GermanyCohort single arm31 patients64 vertebraeThoracolumbar spineT6-L5-64,5 years42% menProstate: 7Cervix: 5Rectum: 5Ovary: 5Lung: 3Pancreas: 2Breast: 2Stomach: 2KyphoplastyFluoroscopy guided1 year in 28 patients39. Pflugmacher, R et al2008, [70]GermanyCohort single arm65 patients99 vertebraeThoracolumbar spine-1–4 levels/ session66 years57% menMetastatic diseaseKyphoplastyFluoroscopy guided2 years in 41 patients40. Qian, Z et al.[27]2011, ChinaCohort single arm48 patients97 vertebraeThoracic (64%) and lumbar (36%) spineT5-L52,6 levels/ session68,5 years40% menBreast: 16 Lung: 10 Prostate: 6 Stomach: 6Cervical: 4Rectal: 3Ovarian: 3KyphoplastyFluoroscopy guided2 years in 39 patients41. Rajah, G et al. [71]2015, USAComparative cohort45 patients95 vertebrae in study groupThoracic (55%) and lumbar (45%) spine--38% menMultiple Myeloma: 21Lung: 9Breast: 6Polycythaemia: 2Colon: 1Neck adenocarcinoma: 1Prostate: 1Renal: 1Non-Hodgkin lymphoma: 1Chronic myeloid leukaemia: 1Lymphoma: 1KyphoplastyFluoroscopy guidedno information42. Ramos, L et al [28]2005, SpainCase series12 patients19 vertebraeThoracic (47%) and lumbar (53%) spineT9-L41–3 levels/ session66 years75% menMultiple Myeloma: 12VertebroplastyFluoroscopy guided3,2 years (2–56 months)43. Reyad, RM et al. [29]2017, EgyptCohort single arm63 patients92 vertebraeThoracic (39%) and lumbar (61%) spineT8-L51–4 levels/ session58 years 48% menBreast: 15Lung: 11Hepatocellular: 11Prostate: 9Multiple Myeloma: 5Thyroid: 5Lymphoma: 3Colon: 3Rectal: 1VertebroplastyFluoroscopy guided12 weeks44. Saracen, A et al [72]2016, PolandCohort single arm90 patients137 vertebraeThoracolumbar spine-1–3 levels/ session-30% menProstate: 36Multiple Myeloma: 32Breast: 20Lung: 2VertebroplastyFluoroscopy guidedno information45. Shi X et al. [14]2021, ChinaComparative cohort309 patients309 vertebraeThoracolumbar spine---59% menMetastatic diseaseVertebroplasty-no information46. Shimony, JS et al. [73]2024, USAComparative cohort50 patients129 vertebraeThoracolumbar spineT1-L52,6 (1–7) levels/ session66,7 years50% menMultiple Myeloma: 14Lung: 13Breast: 8Prostate: 4Lymphoma: 3Unknown primary: 3Adenocarcinoma: 1Pancreas: 1Neuroendocrine: 1Bladder: 1Uterine: 1VertebroplastyFluoroscopy guided23,8 weeks47. Tekin, SB et al. [74]2020, TurkeyCohort single arm42 patients76 vertebraeThoracolumbar spineT8-L5-64,2 years36% menMultiple Myeloma: 42VertebroplastyFluoroscopy guided12,6 months48. Tian, Q et al. [30]2016, ChinaCohort single arm39 patients51 vertebraeThoracolumbar spine-1–3 levels/ session60,1 years62% menLung: 15Prostate: 11Breast: 9Liver: 3Colon: 1VertebroplastyFluoroscopy guided14,5 months49. Trumm, CG et al. [31]2008, GermanyCohort single arm53 patients76 vertebraeThoracic (72%) and lumbar (28%) spine-1–3 levels/ session62 years2% menBreast: 53VertebroplastyFluoroscopy guided9,2 months50. Vrionis, FR et al [75]2005, USACohort single arm50 patients128 vertebraeThoracic (65%) and lumbar (35%) spineT3-L52,56 (1–6) levels/ session63 years34% menMultiple Myeloma: 23Lung: 11Breast: 8Prostate: 4Colon: 1Oesophageal: 1Uterine: 1Sarcoma: 1KyphoplastyFluoroscopy guided9 months51. Wang, H et al. [76]2023, ChinaComparative cohort36 patients36 vertebraeThoracic (39%) and lumbar (61%) spineT5-L5-61,1 years44% menBreast: 11Lung: 7Prostate: 6Stomach: 3Liver: 2Kidney: 2Thyroid: 1 Colon: 1Oesophagus: 1 Nasopharynx: 1Cervical: 1Kyphoplasty-Up to 24 months52. Wang, L et al. [32]2022, ChinaComparative retrospective cohort64 patients113 vertebraeThoracic (42%) and lumbar (58%) spine--65 years56% menMetastatic diseaseVertebroplastyFluoroscopy guided3 months53. Wang, Y et al. [33]2016, ChinaComparative cohort41 patients73 vertebraeThoracolumbar spine--56,2 years44% menBreast: 13Lung: 9Prostate: 6Stomach: 5Liver: 4Rectal: 3Cervical: 2Renal: 2Ovarian: 1KyphoplastyFluoroscopy guided1 year54. Weill, A et al. [34]1996, FranceCase series21 patients27 vertebraeThoracic (44,4%) and lumbar (55,6%) spineT6-L41–3 levels/ session57,1 years48% menBreast: 8Lung: 3Kidney: 3Pheochromocytoma: 2Prostate: 2Uterus: 1Oesophagus: 1Bladder: 1VertebroplastyFluoroscopy guided7,1 months55. Wu, W et al. [35]2022, ChinaCohort single arm31 patients58 vertebraeThoracic (32,8%) and lumbar (67,2%) spine-1–6 levels/ session58,5 years39% menLung: 11Breast: 7Kidney: 4Prostate: 2Rectal: 1Oesophagus: 1Stomach: 1Colon: 1Ovarian: 1KyphoplastyFluoroscopy guided1 year in 27 patients56. Wu, W et al.[36]2022, ChinaComparative cohort59 patients59 vertebraeThoracic (36%) and lumbar (64%) spine-1 level/ session60,07 years59% menMetastatic diseaseKyphoplastyFluoroscopy guided30,33 months57. Yamada, K et al. [21]2004, JapanCase series4 patients6 vertebraeThoracolumbar spineT12-L51–2 levels/ session58,8 years75% menMandibular: 1Gastric: 1Liver: 1Colon: 1VertebroplastyCT guided10–18 weeks58. Yan, B C et al. [37]2022, ChinaComparative cohort87 patients111 vertebraeThoracolumbar spineT1-L5-58 years48,3% menLung: 42Liver: 17Breast: 13Thyroid: 10Kidney: 7Others: 7Colon: 6Multiple Myeloma: 3Bladder: 2Oesophageal: 2Haemangioma: 2VertebroplastyFluoroscopy guidedUp to 12 months59. Zhang, H et al. [77]2017, ChinaCase series13 patients13 vertebraeThoracolumbar spine--58,3 years31% menBreast: 4Lung: 4Prostate: 2Bowel: 1Kidney: 1Liver: 1KyphoplastyFluoroscopy guidedno information60. Zhang, J et al. [50]2023, ChinaComparative cohort28 patients37 vertebraeThoracic (49%) and lumbar (51%) spineT7-L5-64,96 years54% menLung: 28KyphoplastyFluoroscopy guidedno informationTable 2Characteristics of included case-reports.StudyYearPerformed procedureComplicationTime until symptomsOutcome1. Amoretti, N et al. [78]2007VertebroplastyCement migration to the aortaPrevertebral cement leakage during the procedure (intra-operative CT demonstrated cement migration through the lumbar artery into the abdominal aorta)Intra-operativeAsymptomatic2. Chen, YJ et al. [55]2007VertebroplastyNeedle tract tumour seedingMRI showed epidural and paraspinal metastasis along the trajectory of the needle3 monthsPersistent paraparesis despite decompression surgery3. Chevallier, M et al. [79]2021VertebroplastyCement migration to cardiac chambersIntra-operative paravertebral venous cement leakage (CT demonstrated a cement filament extending into the right heart chambers with pulmonary emboli)1 monthFailed endovascular removalSubsequent pulmonary embolism and death4. Cosar M et al. [56]2009VertebroplastyEpidural cement leakageCement extravasation into the epidural space, causing spinal cord compressionPost-operativeDecompressive laminectomyPersistent paraplegia5. Cruz, JP et al. [80]2014KyphoplastyTumour progressionMRI showed tumour extension along the cannula trajectory2,5 weeksTreated with palliative radiotherapy6. Esmende, SM et al. [81]2013KyphoplastyTumour retropulsion with spinal canal compressionCT/MRI showed a tumour mass displaced into the ventral spinal canalPost-operativeEmergent decompression surgery with full recovery7. Gorospe, L et al. [82]2018VertebroplastyNeedle tract tumour seedingSubcutaneous metastatic nodule along the needle path6 monthsAsymptomatic lesion8. Lee, CH et al. [83]2014VertebroplastyNeedle tract tumour seedingSkin metastasis and epidural tumour extension along the needle tract5 monthsSurgical excision and spinal fusionDeath due to disease progression9. Li, Z et al. [11]2013VertebroplastyCement leakage into the inferior vena cavaVenography showed communication between vertebral veins and IVC, leading to cement migrationIntra-operativeAsymptomaticIVC filter captured cement10. Mao, HQ et al. [84]2011VertebroplastyExtradural aracnoyd cystRadiculopathy caused by cyst formation related to dural injury5 monthsSurgical decompression with full recovery11. Mercer, J et al. [85]2020VertebroplastyPulmonary cement embolismCT demonstrated multiple cement emboli in pulmonary arteries2 monthsAsymptomatic12. Morghen, I et al. [86]2009VertebroplastyTumour extrusion into the spinal canalMRI showed extrusion of tumour tissue into spinal canal without neurological deficit24 hRadiotherapyPain resolution13. Ratliff J et al. [87]2001VertebroplastyCement leakage into the spinal canalCement extravasation through neural foramina, causing neurological deficitIntra-operativeSurgical cement removal with full recovery14. Sabuncuoğlu, H et al. [88]2008VertebroplastyIntradural cement leakageIncorrect cannula trajectory causing dural injury and intradural cement depositionIntra-operativeLaminectomy and cement removal with neurological improvement15. Shi, G et al. [89]2020VertebroplastyTransient paraparesisLikely due to anaesthetic infiltration into the spinal canal rather than cement leakagePost-operativeComplete resolution after 4 h16. Tropeano, MP et al. [90]2017VertebroplastySubdural haematomaIatrogenic dural injury3 weeksSurgical evacuation with full recovery Table 1Characteristics of included observational studies.StudyStudy typeSample sizeLevels treated & mean level per surgeryMean age & genderPrimary tumour typesProcedureFollow-up1. Barr, JD et al. [15]2000, USACohort single arm8 patients13 vertebraeThoracolumbar spine--no informationMetastatic diseaseVertebroplasty*CT or fluoroscopy guided*With DSA12 months (2–35)2. Berenson, J et al. [57]2011,Cancer Patient Fracture Evaluation (CAFE) studyRCT68 patients108 vertebraeThoracolumbar spineT5-L5-64,8 years41% menMultiple myeloma: 22Breast: 16Lung: 7Prostate: 4Other: 19Kyphoplasty-no information3. Burton, AW et al.[49]2005, USACase series3 patients6 vertebraeThoracolumbar spineT12-L52 (1–3) levels/ session64 years33% menLung: 1Breast: 1Parathyroid: 1VertebroplastyFluoroscopy guidedUp to 3 months or death4. Caudana, R et al. [16]2008, ItalyCohort single arm39 patients62 vertebraeThoracic (56,4%) and lumbar (43,6%) spine--62 years36% menBreast: 17Lung: 7Kidney: 4Multiple myeloma: 4Colon: 3Prostate: 1Oesophagus: 1Liver: 1Neuroendocrine tumours: 1VertebroplastyCT guided3,39 months (0,3-11,5)5. Chen, G et al. [22]2013, ChinaCase series6 patients6 vertebraeThoracic (50%) and lumbar (50%) spineT9-L51 level/ session59,3 years50% menBreast: 2Lung: 2Liver: 1Prostate: 1KyphoplastyFluoroscopy guided35 months (16–96)6. Chen, L et al. [58]2011, ChinaCase series4 patients4 vertebraeThoracic (25%) and lumbar (75%) spineT8-L4-57,75 years75% menLung: 2Prostate: 1Pancreas: 1VertebroplastyFluoroscopy guided18 months (14–24)7. Chen, KY et al. [40]2009, TaiwanCohort single arm31 patients42 vertebraeThoracic (35,7%) and lumbar (64,2%) spineT3-L5-67 years39% menLung: 10Breast: 7Gastrointestinal tract: 4Hepatobiliary: 3RCC: 1Cervical: 1Multiple myeloma: 1Pancreatic: 1Prostate: 1Unknown: 1Thyroid: 1Vertebroplasty-no information8. Chew, C et al. [59]2011, Scotland, UKCohort single arm128 patients264 vertebraeThoracolumbar spine-Up to 4 levels/ session60 years53% menMultiple myeloma: 41Breast: 22Lung: 16Lymphoma: 11Renal: 8Prostate: 5Others: 25VertebroplastyFluoroscopy guidedno information9. Cui, Y et al. [52]2022, ChinaComparative retrospective cohort230 patients530 vertebraeThoracic (50,6%) and lumbar spine (49,4%)--66 years57% menLung: 172Breast: 82Prostate: 78Other: 198VertebroplastyFluoroscopy guidedno information10. Da Silva Herrero, CFP et al [60]2014, BrazilCohort single arm22 patients34 vertebraeThoracic (30%) and lumbar (70%) spineT7-L5-56 years36% menMultiple Myeloma: 8Breast: 6Plasmacytoma: 2Paraganglioma: 1Oropharynx: 1Fibro histiocytoma: 1Cervix: 1Multiple angiomatosis: 1Gastric: 1KyphoplastyFluoroscopy guided8,5 months11. Diamond, TH et al. [41]2004, AustraliaCase series7 patients14 vertebraeThoracolumbar spineT5-L1Up to 7 levels/ session69 years43% menMultiple Myeloma: 7VertebroplastyFluoroscopy guided6 weeks12. Dudeney, BS et al [42]2002, USACohort single arm18 patients55 vertebraeThoracolumbar spineT6-L5-63,5 years-Multiple Myeloma: 18KyphoplastyFluoroscopy guidedno information13. Eleraky, M et al [61]2011, GermanyCohort single arm14 patients30 vertebraeThoracic spineT1-T51–3 levels/session62 years57% menMultiple Myeloma: 6Lung: 3Breast: 2Melanoma: 1Leiomiosarcoma: 1Bladder: 1KyphoplastyFluoroscopy guided18 months(minimum 12 months)14. Farrokhi, M et al. [23]2012, IranCohort single arm25 patients25 vertebraeThoracic (28%) and lumbar (72%) spineT3-L4-53,5 years44% menBreast: 14Lung: 9Prostate: 2VertebroplastyFluoroscopy guidedno information15. Fourney, DR et al. [38]2003, USACohort single arm56 patients97 vertebraeThoracolumbar spineT3-L51,7 (1–5) levels/ session64 years (median)55% menMultiple Myeloma: 21Breast: 9Lung: 6Lymphoma: 6Prostate: 5Renal cell: 2Gastric: 2Other: 5Kyphoplasty andVertebroplastyFluoroscopy guidedMedian 4,5 months16. García-Maroto, R et al. [62]2015, SpainCohort single arm75 patients122 vertebraeThoracic (36,9%) and lumbar (63,1%) spine--68 years33,3% menMultiple Myeloma: 30Breast: 14Lung: 8Colon: 8Prostate: 3Others: 12KyphoplastyFluoroscopy guided11 months (3–36)17. Garland, P et al. [43]2011, England, UKCohort single arm26 patients57 vertebraeThoracolumbar spineT6-L51–3 levels/ session59 years-Multiple Myeloma: 26VertebroplastyFluoroscopy guided19 months18. Guzik, G [44]2017, PolandCohort single arm73 patients112 vertebraeThoracolumbar spine-Up to 3 levels/ session71 years36% menMultiple Myeloma: 73VertebroplastyFluoroscopy guidedno information19. Ha KY et al[24]2015, KoreaComparative cohort27 patients48 vertebraeThoracolumbar spine--66,6 years41% menMetastatic diseaseKyphoplasty andVertebroplastyFluoroscopy guided16,8 months (6–33)20. He, S et al. [63]2020, ChinaComparative cohort147 patients190 vertebraeThoracolumbar spineT6-L5-70 years35% menMetastatic diseaseKyphoplastyFluoroscopy guided12 months21. Huber FX et al. [25]2009, GermanyCohort single arm76 patients190 vertebraeThoracolumbar spineT6-L52,5 levels/ sessionUp of 5 levels/ session62 years (median)59% menMultiple Myeloma: 76KyphoplastyFluoroscopy guidedno information22. Jang JS et al.[45]2002, South KoreaCase series27 patients72 vertebraeThoracolumbar spine---67% menMultiple Myeloma: 27VertebroplastyFluoroscopy guidedno information23. Julka, A et al. [46]2014, USACohort single arm32 patients76 vertebraeThoracic (45%) and lumbar (55%) spine--64,3 years56% menMultiple Myeloma: 32KyphoplastyFluoroscopy guided24 months24. Kirkegaard, AO et al. [64]2018, DenmarkCohort single arm30 patients81 vertebraeThoracic (45,7%) and lumbar (54,3%) spine-Single level/session in 20% of patients and multiple levels/ session in 80% of patients69,1 years67% menMultiple Myeloma: 13Lung: 6Chronic lymphocytic leukaemia: 5Lymphoma: 2Gastric: 1Prostate: 1Colon: 1Ovarian: 1VertebroplastyFluoroscopy guidedno information25. Kırcelli et al. [65]2018, TurkeyCohort single arm72 patients82 vertebraeThoracolumbar spine-(1–3) levels/ session78,93 years24% menMetastatic diseaseKyphoplastyFluoroscopy guided18,91 months26. Kobayashi, T et al. [17]2009, JapanProspective multi-institutional single-arm Phase I/II clinical study33 patients42 vertebraeThoracic (42,9%) and lumbar (57,1%) spine-1–3 levels/ session62 years48% menLung: 7Breast: 7Colorectal: 7Liver: 4Multiple Myeloma: 3Pancreatic: 2Tongue: 1Oesophageal: 1Skin: 1VertebroplastyFluoroscopy and CT guidedno information27. Kodama, H et al. [18]2007, JapanCohort single arm6 patients8 vertebraeThoracic (50%) and lumbar (50%) spineT1-L11–2 levels/ session65,2 years100% menHepatocellular: 6VertebroplastyCT guidedno information28. König MA et al. [39]2012, UKCase series11 patients23 vertebraeThoracolumbar spineT2-L3-65 years36% menBreast: 3Non-Hodgkin lymphoma: 2Gastric: 1 Cervix:1Prostate: 1Lung: 1Bladder: 1Multiple Myeloma: 1KyphoplastyFluoroscopy guided74,4 weeks29. Lee, B et al [19]2009, England, UKCohort single arm18 patients33 vertebraeThoracolumbar spine-1–3 levels/ session69,3 years39% menBreast: 7Prostate: 4Renal: 2Lung: 1Oesophageal:1Adenoid cystic carcinoma: 1Cervix: 1Other: 1VertebroplastyFluoroscopy and CT guided12 months (3–38)30. 1 Li, Z et al. [11]2014, ChinaComparative cohort1° (KP)42 patients83 vertebraeThoracolumbar spine-1–4 levels/ session68,4 years43% menBreast: 16Lung: 9Prostate: 6Stomach: 6Cervix: 3 Rectal: 2KyphoplastyFluoroscopy guided1 year30.2 Li, Z et al. [11]2014, ChinaComparative cohort2° (VP)38 patients76 vertebraeThoracolumbar spine-1–4 levels/ session65,2 years50% menBreast: 14Lung: 9Prostate: 8Stomach: 5Ovarian: 2VertebroplastyFluoroscopy guided1 year31. Lim, B et al. [66]2009, South KoreaComparative cohort102 patients185 vertebraeThoracic (21%) and lumbar (79%) spineT6-L5-55 years59% menBreast: 24Multiple Myeloma: 19Lung: 16Liver: 10Rectal: 6Unknown primary: 5Stomach: 4Colon: 4Cervix: 2Other: 12VertebroplastyFluoroscopy guided12,2 months32. Liu, H et al. [67]2017, ChinaComparative cohort234 patients288 vertebraeThoracolumbar spine--63,7 years74% menLung: 100Breast: 97Prostate: 14Stomach: 11Rectal: 8Liver: 5Renal: 5Cervix: 2Ovarian: 1KyphoplastyFluoroscopy guided12 months33. Liu, Y et al. [20]2023, ChinaCase-control30 patients30 vertebraeThoracic (20%) and lumbar (80%) spine--59,5 years67% menLung: 30VertebroplastyCT guidedno information34. Masala, S et al. [47]2008, ItalyCohort single arm64 patients198 vertebraeThoracic (47%) and lumbar (53%) spine-3,09 levels/ session71,4 years (median)53% menMultiple Myeloma: 64VertebroplastyFluoroscopy guided6 months35. Moulin, B et al. [26]2020, JapanCohort single arm50 patients397 vertebraeThoracic (60%) and lumbar (40%) spine-7,9 (6–13) levels/ session66 years27% menMultiple Myeloma: 13Breast: 12Prostate: 6Lung: 6Other: 13VertebroplastyFluoroscopy guided401 days36. Nas, OF et al. [68]2015, TurkeyCohort single arm52 patients95 vertebraeThoracic (43,2%) and lumbar (56,8%) spine-1–5 levels/ session63,98 years46% menBreast: 14Lung: 9Prostate: 7 Renal: 5Stomach: 5Others: 12VertebroplastyFluoroscopy guided3 months37. Pflugmacher, R et al [48]2005, GermanyCohort single arm20 patients48 vertebraeThoracolumbar spineT6-L5-62,4 years100% menMultiple Myeloma: 20KyphoplastyFluoroscopy guidedMinimum of 1 year38. Pflugmacher, R et al2007, [69]GermanyCohort single arm31 patients64 vertebraeThoracolumbar spineT6-L5-64,5 years42% menProstate: 7Cervix: 5Rectum: 5Ovary: 5Lung: 3Pancreas: 2Breast: 2Stomach: 2KyphoplastyFluoroscopy guided1 year in 28 patients39. Pflugmacher, R et al2008, [70]GermanyCohort single arm65 patients99 vertebraeThoracolumbar spine-1–4 levels/ session66 years57% menMetastatic diseaseKyphoplastyFluoroscopy guided2 years in 41 patients40. Qian, Z et al.[27]2011, ChinaCohort single arm48 patients97 vertebraeThoracic (64%) and lumbar (36%) spineT5-L52,6 levels/ session68,5 years40% menBreast: 16 Lung: 10 Prostate: 6 Stomach: 6Cervical: 4Rectal: 3Ovarian: 3KyphoplastyFluoroscopy guided2 years in 39 patients41. Rajah, G et al. [71]2015, USAComparative cohort45 patients95 vertebrae in study groupThoracic (55%) and lumbar (45%) spine--38% menMultiple Myeloma: 21Lung: 9Breast: 6Polycythaemia: 2Colon: 1Neck adenocarcinoma: 1Prostate: 1Renal: 1Non-Hodgkin lymphoma: 1Chronic myeloid leukaemia: 1Lymphoma: 1KyphoplastyFluoroscopy guidedno information42. Ramos, L et al [28]2005, SpainCase series12 patients19 vertebraeThoracic (47%) and lumbar (53%) spineT9-L41–3 levels/ session66 years75% menMultiple Myeloma: 12VertebroplastyFluoroscopy guided3,2 years (2–56 months)43. Reyad, RM et al. [29]2017, EgyptCohort single arm63 patients92 vertebraeThoracic (39%) and lumbar (61%) spineT8-L51–4 levels/ session58 years 48% menBreast: 15Lung: 11Hepatocellular: 11Prostate: 9Multiple Myeloma: 5Thyroid: 5Lymphoma: 3Colon: 3Rectal: 1VertebroplastyFluoroscopy guided12 weeks44. Saracen, A et al [72]2016, PolandCohort single arm90 patients137 vertebraeThoracolumbar spine-1–3 levels/ session-30% menProstate: 36Multiple Myeloma: 32Breast: 20Lung: 2VertebroplastyFluoroscopy guidedno information45. Shi X et al. [14]2021, ChinaComparative cohort309 patients309 vertebraeThoracolumbar spine---59% menMetastatic diseaseVertebroplasty-no information46. Shimony, JS et al. [73]2024, USAComparative cohort50 patients129 vertebraeThoracolumbar spineT1-L52,6 (1–7) levels/ session66,7 years50% menMultiple Myeloma: 14Lung: 13Breast: 8Prostate: 4Lymphoma: 3Unknown primary: 3Adenocarcinoma: 1Pancreas: 1Neuroendocrine: 1Bladder: 1Uterine: 1VertebroplastyFluoroscopy guided23,8 weeks47. Tekin, SB et al. [74]2020, TurkeyCohort single arm42 patients76 vertebraeThoracolumbar spineT8-L5-64,2 years36% menMultiple Myeloma: 42VertebroplastyFluoroscopy guided12,6 months48. Tian, Q et al. [30]2016, ChinaCohort single arm39 patients51 vertebraeThoracolumbar spine-1–3 levels/ session60,1 years62% menLung: 15Prostate: 11Breast: 9Liver: 3Colon: 1VertebroplastyFluoroscopy guided14,5 months49. Trumm, CG et al. [31]2008, GermanyCohort single arm53 patients76 vertebraeThoracic (72%) and lumbar (28%) spine-1–3 levels/ session62 years2% menBreast: 53VertebroplastyFluoroscopy guided9,2 months50. Vrionis, FR et al [75]2005, USACohort single arm50 patients128 vertebraeThoracic (65%) and lumbar (35%) spineT3-L52,56 (1–6) levels/ session63 years34% menMultiple Myeloma: 23Lung: 11Breast: 8Prostate: 4Colon: 1Oesophageal: 1Uterine: 1Sarcoma: 1KyphoplastyFluoroscopy guided9 months51. Wang, H et al. [76]2023, ChinaComparative cohort36 patients36 vertebraeThoracic (39%) and lumbar (61%) spineT5-L5-61,1 years44% menBreast: 11Lung: 7Prostate: 6Stomach: 3Liver: 2Kidney: 2Thyroid: 1 Colon: 1Oesophagus: 1 Nasopharynx: 1Cervical: 1Kyphoplasty-Up to 24 months52. Wang, L et al. [32]2022, ChinaComparative retrospective cohort64 patients113 vertebraeThoracic (42%) and lumbar (58%) spine--65 years56% menMetastatic diseaseVertebroplastyFluoroscopy guided3 months53. Wang, Y et al. [33]2016, ChinaComparative cohort41 patients73 vertebraeThoracolumbar spine--56,2 years44% menBreast: 13Lung: 9Prostate: 6Stomach: 5Liver: 4Rectal: 3Cervical: 2Renal: 2Ovarian: 1KyphoplastyFluoroscopy guided1 year54. Weill, A et al. [34]1996, FranceCase series21 patients27 vertebraeThoracic (44,4%) and lumbar (55,6%) spineT6-L41–3 levels/ session57,1 years48% menBreast: 8Lung: 3Kidney: 3Pheochromocytoma: 2Prostate: 2Uterus: 1Oesophagus: 1Bladder: 1VertebroplastyFluoroscopy guided7,1 months55. Wu, W et al. [35]2022, ChinaCohort single arm31 patients58 vertebraeThoracic (32,8%) and lumbar (67,2%) spine-1–6 levels/ session58,5 years39% menLung: 11Breast: 7Kidney: 4Prostate: 2Rectal: 1Oesophagus: 1Stomach: 1Colon: 1Ovarian: 1KyphoplastyFluoroscopy guided1 year in 27 patients56. Wu, W et al.[36]2022, ChinaComparative cohort59 patients59 vertebraeThoracic (36%) and lumbar (64%) spine-1 level/ session60,07 years59% menMetastatic diseaseKyphoplastyFluoroscopy guided30,33 months57. Yamada, K et al. [21]2004, JapanCase series4 patients6 vertebraeThoracolumbar spineT12-L51–2 levels/ session58,8 years75% menMandibular: 1Gastric: 1Liver: 1Colon: 1VertebroplastyCT guided10–18 weeks58. Yan, B C et al. [37]2022, ChinaComparative cohort87 patients111 vertebraeThoracolumbar spineT1-L5-58 years48,3% menLung: 42Liver: 17Breast: 13Thyroid: 10Kidney: 7Others: 7Colon: 6Multiple Myeloma: 3Bladder: 2Oesophageal: 2Haemangioma: 2VertebroplastyFluoroscopy guidedUp to 12 months59. Zhang, H et al. [77]2017, ChinaCase series13 patients13 vertebraeThoracolumbar spine--58,3 years31% menBreast: 4Lung: 4Prostate: 2Bowel: 1Kidney: 1Liver: 1KyphoplastyFluoroscopy guidedno information60. Zhang, J et al. [50]2023, ChinaComparative cohort28 patients37 vertebraeThoracic (49%) and lumbar (51%) spineT7-L5-64,96 years54% menLung: 28KyphoplastyFluoroscopy guidedno information Characteristics of included observational studies. Table 2Characteristics of included case-reports.StudyYearPerformed procedureComplicationTime until symptomsOutcome1. Amoretti, N et al. [78]2007VertebroplastyCement migration to the aortaPrevertebral cement leakage during the procedure (intra-operative CT demonstrated cement migration through the lumbar artery into the abdominal aorta)Intra-operativeAsymptomatic2. Chen, YJ et al. [55]2007VertebroplastyNeedle tract tumour seedingMRI showed epidural and paraspinal metastasis along the trajectory of the needle3 monthsPersistent paraparesis despite decompression surgery3. Chevallier, M et al. [79]2021VertebroplastyCement migration to cardiac chambersIntra-operative paravertebral venous cement leakage (CT demonstrated a cement filament extending into the right heart chambers with pulmonary emboli)1 monthFailed endovascular removalSubsequent pulmonary embolism and death4. Cosar M et al. [56]2009VertebroplastyEpidural cement leakageCement extravasation into the epidural space, causing spinal cord compressionPost-operativeDecompressive laminectomyPersistent paraplegia5. Cruz, JP et al. [80]2014KyphoplastyTumour progressionMRI showed tumour extension along the cannula trajectory2,5 weeksTreated with palliative radiotherapy6. Esmende, SM et al. [81]2013KyphoplastyTumour retropulsion with spinal canal compressionCT/MRI showed a tumour mass displaced into the ventral spinal canalPost-operativeEmergent decompression surgery with full recovery7. Gorospe, L et al. [82]2018VertebroplastyNeedle tract tumour seedingSubcutaneous metastatic nodule along the needle path6 monthsAsymptomatic lesion8. Lee, CH et al. [83]2014VertebroplastyNeedle tract tumour seedingSkin metastasis and epidural tumour extension along the needle tract5 monthsSurgical excision and spinal fusionDeath due to disease progression9. Li, Z et al. [11]2013VertebroplastyCement leakage into the inferior vena cavaVenography showed communication between vertebral veins and IVC, leading to cement migrationIntra-operativeAsymptomaticIVC filter captured cement10. Mao, HQ et al. [84]2011VertebroplastyExtradural aracnoyd cystRadiculopathy caused by cyst formation related to dural injury5 monthsSurgical decompression with full recovery11. Mercer, J et al. [85]2020VertebroplastyPulmonary cement embolismCT demonstrated multiple cement emboli in pulmonary arteries2 monthsAsymptomatic12. Morghen, I et al. [86]2009VertebroplastyTumour extrusion into the spinal canalMRI showed extrusion of tumour tissue into spinal canal without neurological deficit24 hRadiotherapyPain resolution13. Ratliff J et al. [87]2001VertebroplastyCement leakage into the spinal canalCement extravasation through neural foramina, causing neurological deficitIntra-operativeSurgical cement removal with full recovery14. Sabuncuoğlu, H et al. [88]2008VertebroplastyIntradural cement leakageIncorrect cannula trajectory causing dural injury and intradural cement depositionIntra-operativeLaminectomy and cement removal with neurological improvement15. Shi, G et al. [89]2020VertebroplastyTransient paraparesisLikely due to anaesthetic infiltration into the spinal canal rather than cement leakagePost-operativeComplete resolution after 4 h16. Tropeano, MP et al. [90]2017VertebroplastySubdural haematomaIatrogenic dural injury3 weeksSurgical evacuation with full recovery Characteristics of included case-reports. Characteristics of RCTs and observational studies A total of 3234 patients undergoing vertebral augmentation procedures for metastatic thoracolumbar spinal fractures were identified, corresponding to a minimum of 5737 vertebral levels treated, as Shi X et al. do not report the total number of levels treated [14]. Patient characteristics were heterogeneous across studies. The weighted mean age was 64,1 years, with reported study means ranging from 53,5 to 78,9 years. Overall, 49.3% of the patients were male. Follow-up duration was reported in 37 studies, with a weighted mean follow-up of 12.3 months (median 12 months, range 1–118 months). Tumour type was reported in 54 studies and 16 included patients with a single tumour type. Across the overall sample, the most frequent primary tumours were multiple myeloma (28.0%), breast cancer (21.6%), lung cancer (19.8%), and prostate cancer (8.0%). Most of the studies performed percutaneous vertebroplasty, while in 23 the procedure was balloon kyphoplasty, and in three studies both techniques were used. The procedures were guided by fluoroscopy in most cases (82,5% of included studies), with a smaller number guided by CT scan [15], [16], [17], [18], [19], [20], [21]. Twenty-two studies mentioned performing routine CT evaluation in the post-op period [11], [16], [17], [18], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30], [31], [32], [33], [34], [35], [36], [37], and one study if there was cement leakage visualised intra-operatively [38]. Overall, 44,3% of the included studies had a CT evaluation of the spine, either intra-operatively or post-operatively. When reported, the mean injected cement volume ranged from 1.5 to 5.87 mL per treated vertebra, although higher volumes were occasionally described in lumbar vertebrae [37], [39]. Information regarding the number of vertebral levels treated per procedure was reported in 39 studies. Among these, the majority (51.3%) performed vertebral augmentation in a maximum of 3 levels per session [15], [16], [17], [19], [23], [24], [25], [27], [28], [29], [38], [40], [41], [42], [43], [44], [45], [46], [47], [48]. Substantial clinical and methodological heterogeneity was observed across studies, including differences in study designs, populations, procedural techniques, and complication detection methods. Although all studies involved metastatic spinal disease, some included only pathological fractures, while others considered vertebral metastases alone as inclusion criteria. Although none included patients with neurological deficits, some studies included vertebrae with epidural tumour involvement, while others excluded posterior wall destruction or epidural extension. Methodological heterogeneity was also substantial due to variability in study design. In addition, the criteria used to define complications varied between studies. While some authors considered only symptomatic events as complications, others reported all adverse events, including asymptomatic findings detected on imaging. Some studies documented a broad range of medical events during follow-up, whereas others reported only complications directly attributable to vertebral augmentation. Among studies that considered cement leakage as a complication, some reported all types of leakage, whereas others excluded specific patterns such as leakage into paravertebral soft tissues. Moreover, while approximately half of the included studies systematically performed postoperative CT to detect cement leakage, the remaining studies relied primarily on intraoperative fluoroscopy or postoperative radiographs, which may underestimate the true incidence of leakage. Substantial heterogeneity was also observed in the reported complication rates across studies. Given the variability in study populations, methodologies, definitions of complications, and diagnostic approaches, a quantitative meta-analysis was deemed inappropriate. Therefore, the results were synthesised descriptively. Complications of vertebral augmentation procedures Complications of vertebral augmentation procedures are summarised in Table 3.Table 3Treatment characteristics & reported complications(When available, complication rates were extracted according to the denominator reported in each study (per vertebra or per patient). In studies where this was not specified, cement leakage rates were calculated relative to the number of treated vertebrae, whereas other complications were calculated relative to the total number of procedures.)StudyProcedure detailsImaging detailsCement detailsComplicationsSymptomatic complicationsType of cement leakage1. Barr, JD et al.Vertebroplasty* 8 patients* 13 vertebraeFluoroscopy or CT* CT post-op2–3 mL / thoracic vertebra3–5 mL / lumbar vertebraeTotal: 7,7%* Cement Leakage: 7.7% (1/13)0%Venous system: 100% (1/1)2. Berenson, J et al.Kyphoplasty* 68 patients-* No post-op CTNo informationTotal: 12,5%* Cement Leakage: 4,7% (3/68)* Asymptomatic balloon rupture: 1,5% (1/68)* Procedure pain: 1,6% (1/68)* Superficial infection: 1,6% (1/68)* Adjacent level fracture: 3,1% (2/68)Total: 4,7%* Superficial infection: 1,6% (1/68)* Adjacent level fracture: 3,1% (2/68)** one of the fractures occurred 1 day post-op, in a patient with intradiscal cement leakageNo information3. Burton, AW et al.Vertebroplasty* 3 patients* 6 vertebraeFluoroscopy* No post-op CT3–4 mL / vertebra0%0%–4. Caudana, R et al.Vertebroplasty* 39 patients* 62 vertebraeCT- guided* CT post-opMean 5.7 mL/vertebraTotal: 75,2%* Cement Leakage: 72,5% (45/62)* Pulmonary Cement Embolism: 2,6% (1/39)Total: 5,9%* Cement Leakage:3,2% (2/62)** radiculopathy in leakage to the epidural space (medical treatment)* Cement PE: 2,7% (1/39)** mild dyspnoea (medical treatment)Paravertebral soft tissues: 17,8% (8/45)Intracanal: 28,9 (13/45)Intradiscal: 2,2% (1/56)Venous system: 17,8% (8/45)5. Chen, G et al.Kyphoplasty* 6 patients* 6 vertebraeFluoroscopy* CT post-opMean 3.3 mL /vertebra (2–5)Total: 16,7%* Cement Leakage: 16,7% (1/6)0%Venous system: 100% (1/1)6. Chen, L et al.Vertebroplasty* 4 patients* 4 vertebraeFluoroscopy* No post-op CTMean 2,9 mL/vertebra (2,2-3,5)Total: 25%* Cement Leakage: 25% (1/4)0%Venous system: 100% (1/1)7. Chen, KY et al.Vertebroplasty* 31 patients* 42 vertebrae-* -–0%0%–8. Chew, C et al.Vertebroplasty* 128 patients* 158 procedures* 264 vertebraeFluoroscopy* -< 5 mL / vertebraTotal: 3,7%* IVC cement embolism: 0,6% (1/158)* Pulmonary cement embolism: 1,9% (3/158)* Local haematoma: 0,6% (1/158)* Cement leakage: 0,6% (1/158)** “Asymptomatic small para-vertebral leaks were not considered complications or accounted for”Total: 1,2%Local haematoma: 0,6% (1/158)Cement leakage: 0,8% (1/158)* transient loss of sensation in T1 dermatomeIntracanal: 100% (1/1)9. Cui, Y et al.Vertebroplasty* 230 patients* 530 vertebraeFluoroscopy* No post-op CT–Total: 34,9%* Cement Leakage: 34,9% (185/530)0%Intradiscal or paravertebral: 47,5% (97/204)Intracanal: 18,2% (37/204)Intravenous: 34,3% (70/204)10. Da Silva Herrero, CFP et alKyphoplasty* 22 patients* 34 vertebraeFluoroscopy* -–Total: 20,8%* Cement Leakage: 11,8% (4/34)* DVT with PE: 4.5% (1/22)*Adjacent level fracture: 4.5% (1/22)Total: 9%* DVT with PE: 4.5% (1/22)*Adjacent level fracture: 4.5% (1/22)Paravertebral soft tissues: 50% (2/4)Intracanal: 25% (1/4)Intradiscal: 25% (1/4)11. Diamond, TH et al.Vertebroplasty* 7 patients* 14 vertebraeFluoroscopy* No post-op CT1–6 mL / vertebra0%0%–12. Dudeney, BS et alKyphoplasty* 18 patients* 27 procedures* 55 vertebraeFluoroscopic* No post-op CT–Total: 4%* Cement Leakage:4% (2/55)0%Paravertebral soft tissues: 50% (1/2)Intracanal: 50% (1/2)13. Eleraky, M et alKyphoplasty* 14 patients* 30 vertebraeFluoroscopy* -3 mL / vertebra (1,5–4,5 mL)Total: 10%* Cement Leakage:10% (3/30)0%Intradiscal: 100%14. Farrokhi, M et al.Vertebroplasty* 25 patients* 25 vertebraeFluoroscopy* -4,5 mL / vertebra (3–6 mL)Total: 24%* Cement Leakage:24% (6/25)0%Paravertebral soft tissues: 45,5% (5/11)Intracanal: 18,1% (2/11)Intradiscal: 36,4% (4/11)15. Fourney, DR et al.Kyphoplasty &Vertebroplasty* 56 patients* 97 vertebraeFluoroscopy* CT post-op if cement leakage was detected intra-op2–8 mL / vertebraTotal: 6,2%* Cement Leakage:6,2% (6/97)0%Paravertebral soft tissues: 16,7% (1/6)Intradiscal: 83,3% (5/6)16. García-Maroto, R et al.Kyphoplasty* 75 patients* 122 vertebraeFluoroscopy* -–Total: 5,7%* Cement Leakage:5,7% (7/122)0%Paravertebral soft tissues: 28,6% (2/7)Intradiscal: 71,4% (5/7)17. Garland, P et al.Vertebroplasty* 26 patients* 57 vertebraeFluoroscopy* No post-op CT–Total: 28,4%* Cement Leakage:5,3% (5/57)* Post-procedural pain: 23,1% (6/26)Total: 26,6%* Cement Leakage: 3,5% (2/57)** transient pain* Post-procedural pain: 23,1% (6/26)** resolved within 24 h–18. Guzik, GVertebroplasty* 73 patients* 112 vertebraeFluoroscopy* -–Total: 4,1%* Cement Leakage:2,7% (2/73)* Pulmonary cement embolism: 1,4% (1/73)0%Intracanal: 100% (2/2)19.Ha KY et alVertebroplasty & Kyphoplasty* 27 patients* 48 vertebraeFluoroscopy* -4,3 mL / vertebraTotal: 20,8%* Cement Leakage:20,8% (10/48)0%Paravertebral soft tissues: 40% (1/10)Intracanal: 10% (1/10)Intradiscal: 50% (5/10)20. He, S et al.Kyphoplasty* 147 patients* 190 vertebraeFluoroscopy* No post-op CT3,88 mL / vertebraeTotal: 6%* Cement Leakage:2,6% (5/190)* Fever: 3,4% (5/147)0%Paravertebral soft tissue: 80% (4/5)Intradiscal: 20% (1/5)21. Huber FX et al.Kyphoplasty* 76 patients* 190 vertebraeFluoroscopy* CT post-op–Total: 19,7%* Cement Leakage:18,4% (35/190)* Pulmonary cement embolism: 1,3% (1/76)Total: 1,3%* Paraparesis resolved with surgery due to cement intracanal leakage in thoracic vertebra: 1,3% (1/73)–22. Jang JS et al.Vertebroplasty* 27 patients* 72 vertebraeFluoroscopy* -–Total: 11,1%* Pulmonary Cement Embolism: 11,1% (3/27)Total: 7,4%* Pulmonary cement embolism:7,4% (2/27)–23. Julka, A et al.Kyphoplasty* 32 patients* 76 vertebraeFluoroscopy* No post-op CT3 mL / vertebraTotal: 37,5%* Cement Leakage: 37,5% (12/32)0%Paravertebral soft tissue: 16,7% (2/12)Intradiscal: 50% (6/12)Intravenous: 25% (3/12)Unknown: 8,3% (1/12)24. Kirkegaard, AO et al.Vertebroplasty* 30 patients* 81 vertebraeFluoroscopy* No post-op CT–Total: 14,8%* Cement Leakage:14,8% (12/81)0%–25. Kırcelli et al.Kyphoplasty* 72 patients* 82 vertebraeFluoroscopy* No post-op CT5,21 mL / vertebraTotal: 23,5%* Cement Leakage: 8,3% (6/72)*Adjacent level fracture: 15,2% (11/72)0%–26. Kobayashi, T et al.Vertebroplasty* 33 patients* 42 vertebraeFluoroscopy & CT-guided* CT post-op3,5 mL / vertebra (1–8 mL)Total: 3%* Bleeding at puncture site: 3% (1/33)Total: 3%* Bleeding at puncture site: 3% (1/33)** stopped after 5 min compression–27. Kodama, H et al.Vertebroplasty* 6 patients* 8 vertebraeCT guided* CT post-op–0%0%–28. König MA et al.Kyphoplasty* 11 patients* 23 vertebraeFluoroscopy* No post-op CT5,1 mL / vertebra (1–12 mL)Total: 36,4%* Cement Leakage: 18,2% (2/11)* DVT: 9,1% (1/11)* Adjacent level fracture: 9,1% (1/11)Total: 18,2%* DVT: 9,1% (1/11)* Adjacent level fracture: 9,1% (1/11)Intracanal: 100% (2/2)29. Lee, B et alVertebroplasty* 18 patients* 33 vertebraeFluoroscopy & CT-guided* -–Total: 5,6%* Cement Leakage: 5,6% (1/18)Total: 5,6%* Cement Leakage: 5,6% (1/18)** transient pain (<24 h)Puncture site: 100%30.1 Li, Z et al.Kyphoplasty* 42 patients* 83 vertebraeFluoroscopy* CT post-op–Total: 16,9%* Cement Leakage: 16,9% (14/83)0%Paravertebral soft tissue: 28,6% (4/14)Intradiscal: 7,1% (1/14)Intravenous: 64,3% (9/14)30.2 Li, Z et al.Vertebroplasty* 38 patients* 76 vertebraeFluoroscopy* CT post-op–Total: 30,3%* Cement Leakage: 30,3% (23/76)0%Paravertebral soft tissue: 30,4% (7/23)Intradiscal: 13,1% (3/23)Intravenous: 56,5% (13/23)31.Lim, B et al.Vertebroplasty* 102 patients* 185 vertebraeFluoroscopy* No post-op CT3,62 mL / vertebraTotal: 17,3%* Cement Leakage: 17,3% (32/185)0%Intracanal: 40,1% (13/32)Intradiscal: 40,1% (13/32)Intravenous: 18,8% (6/32)32. Liu, H et al.Kyphoplasty* 234 patients* 288 vertebraeFluoroscopy* -–Total: 6,8* Cement Leakage: 6,4% (15/234)* Pulmonary cement embolism: 0,4% (1/234)0%Paravertebral soft tissue: 53,3% (8/15)Intradiscal: 46,7% (7/15)33. Liu, Y et al.Vertebroplasty* 30 patients* 30 vertebraeCT-guided3,47 mL / vertebraTotal: 53,3%* Cement Leakage: 40% (12/30)* Haematoma at puncture site:13,3% (4/30)Total: 13,3%* Haematoma at puncture site: 13,33%–34. Masala, S et al.Vertebroplasty* 64 patients* 198 vertebraeFluoroscopy* No post-op CT–0%0%–35. Moulin, B et al.Vertebroplasty* 50 patients* 397 vertebraeFluoroscopy* CT post-op* Chest-CT performed in 38 patients–Total: 57%* Pulmonary cement embolism: 47% (18/38)* Post-operative pain: 4% (2/50)* Post-operative transient alteration of general status: 2% (1/50)* Retroperitoneal haematoma: 2% (1/50)* Lumbar artery puncture: 2% (1/50)Total: 10%* Pulmonary cement embolism: 2% (1/50)* Post-operative pain: 4% (2/50)* Post-operative transient alteration of general status: 2% (1/50)* Lumbar artery puncture: 2% (1/50)** embolization intra-op–36. Nas, OF et al.Vertebroplasty* 52 patients* 95 vertebraeFluoroscopy* -4,91 mL / vertebra (1,5–9 mL)Total: 45,3%* Cement Leakage: 45,3% (43/95)0%Intradiscal: 46,5% (20/43)Intravenous: 46,5% (20/43)Intradiscal and intravenous: 7% (3/43)37. Pflugmacher, R et alKyphoplasty* 20 patients* 48 vertebraeFluoroscopy* No post-op CT4,6 mL / vertebra (2,6–7,9 mL)Total: 15,4%* Cement Leakage: 10,4% (5/48)* Adjacent level fracture: 5% (1/20)Total: 5%* Adjacent fracture: 5%** need for surgical interventionParavertebral soft tissue: 40% (2/5)Intradiscal: 60% (3/5)38. Pflugmacher, R et alKyphoplasty* 31 patients* 64 vertebraeFluoroscopy* No post-op CT–Total: 12,5%* Cement Leakage: 12,5% (8/64)0%Paravertebral soft tissues: 25% (2/8)Intradiscal: 50% (4/8)Intravenous: 25% (2/8)39. Pflugmacher, R et alKyphoplasty* 65 patients* 99 vertebraeFluoroscopy* No post-op CT–Total: 20,1%* Cement Leakage: 12,1% (12/99)* Adjacent fracture: 8% (5/65)0%–40. Qian, Z et al.Kyphoplasty* 48 patients* 97 vertebrae in study groupFluoroscopy* CT post-op–Total: 18,6%Cement Leakage: 18,6% (18/97)0%Paravertebral soft tissue: 22,2% (4/18)Intradiscal: 16,7% (3/18)Intravenous: 61,1% (11/18)41. Rajah, G et al.Kyphoplasty* 45 patients* 75 vertebraeFluoroscopy3,6 mLTotal: 18,6%* Cement Leakage: 13,3% (10/75)* Failure (retropulsion of cement into canal): 5,3% (4/75)Total: 5,3%* Failure: 5,3%** need for surgical decompression and fusion–42. Ramos, L et alVertebroplasty* 12 patients* 19 vertebraeFluoroscopy* CT post-op–Total: 84%Cement Leakage: 84% (16/19)0%Intracanal: 12,5% (2/16)Intradiscal: 18,7% (3/16)Intravenous: 68,8% (11/16)43.Reyad, RM et al.Vertebroplasty* 63 patients* 92 vertebraeFluoroscopy* CT 1 week post-op1,5 mL / vertebraTotal: 82,5%* Transient exacerbation of pain: 61,9% (39/63)* Local haematoma: 9,5% (6/63)* Cement leakage: 7,9% (5/63)* Adjacent level fracture: 3,2% (2/63)Total: 74,6%* Exacerbation of pain: 61,9%*Local haematoma: 9,5%* Adjacent fracture: 3,2%Paravertebral soft tissues: 20% (1/5)Intracanal: 20% (1/5)Intradiscal: 40% (2/5)Intravenous: 40% (2/5)44. Saracen, A et alVertebroplasty* 90 patients* 137 vertebraeFluoroscopy* No post-op CT2–3 mL / vertebraTotal: 35,1%* Cement leakage: 34% (47/137)* Pulmonary embolism: 1,1% (1/90)Total: 1,1%* Monoparesis due to cement intracanal leakage, resolved with surgery: 1,1% (1/90)Paravertebral soft tissue: 25,5% (12/47)Intracanal: 14,9% (7/47)Intradiscal: 19,2% (9/47)Intravenous: 40,4% (19/47)45. Shi X et al.Vertebroplasty* 309 patients-* -–Total: 31,2%* Cement leakage: 31,2% (189/309)Total: 1%* Neurological symptoms due to cement leakage: 1% (3/309)Paravertebral soft tissue: 29,4% (74/252)Intracanal: 13,1% (33/252)Intradiscal: 22,6% (57/252)Intravenous: 34,9% (88/252)* 252 types of leakage due to more than 1 type in the same vertebra46. Shimony, JS et al.Vertebroplasty* 50 patients* 129 vertebraeFluoroscopy* -–Total: 14%* Increased or new sites of pain: 14% (7/50)* Asymptomatic cement leakage was not considered as a complicationTotal: 14%* Increased or new sites of pain: 14% (7/50)–47. Tekin, SB et al.Vertebroplasty & Kyphoplasty* 42 patients* 76 vertebraeFluoroscopy* No post-op CT–Total: 7,1%* Vacuum phenomena: 4,7% (2/42)* Cement leakage: 2,4% (1/42)0%–48. Tian, Q et al.Vertebroplasty* 39 patients* 51 vertebraeFluoroscopy* CT post-op3,3 mL / vertebra (2–5 mL)Total: 38,5%* Cement leakage: 38,5% (15/39)0%Paravertebral soft tissues: 60% (9/15)Intracanal: 6,7% (1/15)Intravenous: 20% (3/15)Puncture site: 13,3% (2/15)49. Trumm, CG et al.Vertebroplasty* 76 vertebraeFluoroscopy* CT post-op3–3,5 mL / vertebraTotal: 75%* Cement leakage: 75% (57/76)0%Paravertebral soft tissue: (42,2% 27/64)Intracanal: 26,6% (17/64)Intradiscal: 31,2% (20/64)50. Vrionis, FR et alKyphoplasty* 50 patients* 128 vertebraeFluoroscopy* No post-op CT–Total: 2%* Asystole: 2%Total: 2%* Asystole: 2%** successfully reversed, no cause found–51. Wang, H et al.Kyphoplasty* 36 patients* 36 vertebrae-* No post-op CT–Total: 11,1%Cement leakage: 11,1% (4/36)0%Paravertebral soft tissue: 50% (2/4)Intradiscal: 25% (1/4)Intravenous: 25% (1/4)52. Wang, L et al.Vertebroplasty* 64 patients* 113 vertebraeFluoroscopy* CT post-op4,84 mL / vertebra (1,2–9 mL)Total: 56,6%* Cement leakage: 56,6% (64/113)0%Paravertebral soft tissue or intradiscal: 54,7% (35/64)Intracanal: 28,1% (18/64)Intravenous: 17,2% (11/64)53. Wang, Y et al.Kyphoplasty* 41 patients* 73 vertebraeFluoroscopy* CT post-op3,6 mL / vertebra (1,5–6,5 mL)Total: 17,1%* Cement leakage: 17,1% (7/41)0%–54. Weill, A et al.Vertebroplasty* 22 sessions* 21 patients* 27 vertebraeFluoroscopy* CT post-op–Total: 45,4%* Cement leakage: 40,9% (9/22)* Cement embolism to IVC: 4,5% (1/22)Total: 9%* Acute raquialgia because of cement leakage into canal: 9% (2/22)** need for surgical treatment in 1/2 casesParavertebral soft tissue: 11,1 (1/9)Intracanal: 22,2 (2/9)Intradiscal: 11,1 (1/9)Intravenous: 55,6 (5/9)55. Wu, W et al.Kyphoplasty* 31 patients* 58 vertebraeFluoroscopy* CT post-op2,6 mL / vertebraTotal: 6,5%* Cement leakage: 6,5% (2/31)0%Intradiscal: 50% (1/2)Puncture Site: 50% (1/2)56. Wu, W et al.Kyphoplasty* 59 patients* 59 vertebraeFluoroscopic* CT post-op3,84 mLTotal: 18,6%* Cement leakage: 18,6% (11/59)0%Intracanal: 18,2% (2/11)Intradiscal: 36,4% (4/11)Intravenous: 18,2% (2/11)Puncture Site: 27,2% (3/11)57. Yamada, K et al.Vertebroplasty* 4 patients* 5 vertebraeCT-guided* CT post-op4 mL /vertebra (1,5–7,5 mL)Total: 50%* Cement leakage: 50% (2/4)0%–58. Yan, B C et al.Vertebroplasty* 87 patients* 111 vertebraeFluoroscopy* CT post-op5,87 mL/vertebraTotal: 114%* Cement leakage: 98% (109/111)* Fever: 12,6% (11/87)* Pain: 3,4% (3/87)Total: 16%* Fever: 12,6% (11/87)* Pain: 3,4% (3/87)Paravertebral soft tissue: 49,1% (77/157)Intracanal: 20,4% (32/157)Intradiscal: 19,7% (31/157)Intravenous: 10,8% (17/157)59. Zhang, H et al.Kyphoplasty* 13 patients* 13 vertebraeFluoroscopy* No post-op CT–Total: 15,4%* Cement leakage: 15,4% (2/13)0%Intracanal: 33,3% (1/3)Intravenous: 66,7% (2/3)60.Zhang, J et al.Kyphoplasty* 28 patients* 37 vertebraeFluoroscopy* No post-op CT3,70 mL / vertebra0%0%– Table 3Treatment characteristics & reported complications(When available, complication rates were extracted according to the denominator reported in each study (per vertebra or per patient). In studies where this was not specified, cement leakage rates were calculated relative to the number of treated vertebrae, whereas other complications were calculated relative to the total number of procedures.)StudyProcedure detailsImaging detailsCement detailsComplicationsSymptomatic complicationsType of cement leakage1. Barr, JD et al.Vertebroplasty* 8 patients* 13 vertebraeFluoroscopy or CT* CT post-op2–3 mL / thoracic vertebra3–5 mL / lumbar vertebraeTotal: 7,7%* Cement Leakage: 7.7% (1/13)0%Venous system: 100% (1/1)2. Berenson, J et al.Kyphoplasty* 68 patients-* No post-op CTNo informationTotal: 12,5%* Cement Leakage: 4,7% (3/68)* Asymptomatic balloon rupture: 1,5% (1/68)* Procedure pain: 1,6% (1/68)* Superficial infection: 1,6% (1/68)* Adjacent level fracture: 3,1% (2/68)Total: 4,7%* Superficial infection: 1,6% (1/68)* Adjacent level fracture: 3,1% (2/68)** one of the fractures occurred 1 day post-op, in a patient with intradiscal cement leakageNo information3. Burton, AW et al.Vertebroplasty* 3 patients* 6 vertebraeFluoroscopy* No post-op CT3–4 mL / vertebra0%0%–4. Caudana, R et al.Vertebroplasty* 39 patients* 62 vertebraeCT- guided* CT post-opMean 5.7 mL/vertebraTotal: 75,2%* Cement Leakage: 72,5% (45/62)* Pulmonary Cement Embolism: 2,6% (1/39)Total: 5,9%* Cement Leakage:3,2% (2/62)** radiculopathy in leakage to the epidural space (medical treatment)* Cement PE: 2,7% (1/39)** mild dyspnoea (medical treatment)Paravertebral soft tissues: 17,8% (8/45)Intracanal: 28,9 (13/45)Intradiscal: 2,2% (1/56)Venous system: 17,8% (8/45)5. Chen, G et al.Kyphoplasty* 6 patients* 6 vertebraeFluoroscopy* CT post-opMean 3.3 mL /vertebra (2–5)Total: 16,7%* Cement Leakage: 16,7% (1/6)0%Venous system: 100% (1/1)6. Chen, L et al.Vertebroplasty* 4 patients* 4 vertebraeFluoroscopy* No post-op CTMean 2,9 mL/vertebra (2,2-3,5)Total: 25%* Cement Leakage: 25% (1/4)0%Venous system: 100% (1/1)7. Chen, KY et al.Vertebroplasty* 31 patients* 42 vertebrae-* -–0%0%–8. Chew, C et al.Vertebroplasty* 128 patients* 158 procedures* 264 vertebraeFluoroscopy* -< 5 mL / vertebraTotal: 3,7%* IVC cement embolism: 0,6% (1/158)* Pulmonary cement embolism: 1,9% (3/158)* Local haematoma: 0,6% (1/158)* Cement leakage: 0,6% (1/158)** “Asymptomatic small para-vertebral leaks were not considered complications or accounted for”Total: 1,2%Local haematoma: 0,6% (1/158)Cement leakage: 0,8% (1/158)* transient loss of sensation in T1 dermatomeIntracanal: 100% (1/1)9. Cui, Y et al.Vertebroplasty* 230 patients* 530 vertebraeFluoroscopy* No post-op CT–Total: 34,9%* Cement Leakage: 34,9% (185/530)0%Intradiscal or paravertebral: 47,5% (97/204)Intracanal: 18,2% (37/204)Intravenous: 34,3% (70/204)10. Da Silva Herrero, CFP et alKyphoplasty* 22 patients* 34 vertebraeFluoroscopy* -–Total: 20,8%* Cement Leakage: 11,8% (4/34)* DVT with PE: 4.5% (1/22)*Adjacent level fracture: 4.5% (1/22)Total: 9%* DVT with PE: 4.5% (1/22)*Adjacent level fracture: 4.5% (1/22)Paravertebral soft tissues: 50% (2/4)Intracanal: 25% (1/4)Intradiscal: 25% (1/4)11. Diamond, TH et al.Vertebroplasty* 7 patients* 14 vertebraeFluoroscopy* No post-op CT1–6 mL / vertebra0%0%–12. Dudeney, BS et alKyphoplasty* 18 patients* 27 procedures* 55 vertebraeFluoroscopic* No post-op CT–Total: 4%* Cement Leakage:4% (2/55)0%Paravertebral soft tissues: 50% (1/2)Intracanal: 50% (1/2)13. Eleraky, M et alKyphoplasty* 14 patients* 30 vertebraeFluoroscopy* -3 mL / vertebra (1,5–4,5 mL)Total: 10%* Cement Leakage:10% (3/30)0%Intradiscal: 100%14. Farrokhi, M et al.Vertebroplasty* 25 patients* 25 vertebraeFluoroscopy* -4,5 mL / vertebra (3–6 mL)Total: 24%* Cement Leakage:24% (6/25)0%Paravertebral soft tissues: 45,5% (5/11)Intracanal: 18,1% (2/11)Intradiscal: 36,4% (4/11)15. Fourney, DR et al.Kyphoplasty &Vertebroplasty* 56 patients* 97 vertebraeFluoroscopy* CT post-op if cement leakage was detected intra-op2–8 mL / vertebraTotal: 6,2%* Cement Leakage:6,2% (6/97)0%Paravertebral soft tissues: 16,7% (1/6)Intradiscal: 83,3% (5/6)16. García-Maroto, R et al.Kyphoplasty* 75 patients* 122 vertebraeFluoroscopy* -–Total: 5,7%* Cement Leakage:5,7% (7/122)0%Paravertebral soft tissues: 28,6% (2/7)Intradiscal: 71,4% (5/7)17. Garland, P et al.Vertebroplasty* 26 patients* 57 vertebraeFluoroscopy* No post-op CT–Total: 28,4%* Cement Leakage:5,3% (5/57)* Post-procedural pain: 23,1% (6/26)Total: 26,6%* Cement Leakage: 3,5% (2/57)** transient pain* Post-procedural pain: 23,1% (6/26)** resolved within 24 h–18. Guzik, GVertebroplasty* 73 patients* 112 vertebraeFluoroscopy* -–Total: 4,1%* Cement Leakage:2,7% (2/73)* Pulmonary cement embolism: 1,4% (1/73)0%Intracanal: 100% (2/2)19.Ha KY et alVertebroplasty & Kyphoplasty* 27 patients* 48 vertebraeFluoroscopy* -4,3 mL / vertebraTotal: 20,8%* Cement Leakage:20,8% (10/48)0%Paravertebral soft tissues: 40% (1/10)Intracanal: 10% (1/10)Intradiscal: 50% (5/10)20. He, S et al.Kyphoplasty* 147 patients* 190 vertebraeFluoroscopy* No post-op CT3,88 mL / vertebraeTotal: 6%* Cement Leakage:2,6% (5/190)* Fever: 3,4% (5/147)0%Paravertebral soft tissue: 80% (4/5)Intradiscal: 20% (1/5)21. Huber FX et al.Kyphoplasty* 76 patients* 190 vertebraeFluoroscopy* CT post-op–Total: 19,7%* Cement Leakage:18,4% (35/190)* Pulmonary cement embolism: 1,3% (1/76)Total: 1,3%* Paraparesis resolved with surgery due to cement intracanal leakage in thoracic vertebra: 1,3% (1/73)–22. Jang JS et al.Vertebroplasty* 27 patients* 72 vertebraeFluoroscopy* -–Total: 11,1%* Pulmonary Cement Embolism: 11,1% (3/27)Total: 7,4%* Pulmonary cement embolism:7,4% (2/27)–23. Julka, A et al.Kyphoplasty* 32 patients* 76 vertebraeFluoroscopy* No post-op CT3 mL / vertebraTotal: 37,5%* Cement Leakage: 37,5% (12/32)0%Paravertebral soft tissue: 16,7% (2/12)Intradiscal: 50% (6/12)Intravenous: 25% (3/12)Unknown: 8,3% (1/12)24. Kirkegaard, AO et al.Vertebroplasty* 30 patients* 81 vertebraeFluoroscopy* No post-op CT–Total: 14,8%* Cement Leakage:14,8% (12/81)0%–25. Kırcelli et al.Kyphoplasty* 72 patients* 82 vertebraeFluoroscopy* No post-op CT5,21 mL / vertebraTotal: 23,5%* Cement Leakage: 8,3% (6/72)*Adjacent level fracture: 15,2% (11/72)0%–26. Kobayashi, T et al.Vertebroplasty* 33 patients* 42 vertebraeFluoroscopy & CT-guided* CT post-op3,5 mL / vertebra (1–8 mL)Total: 3%* Bleeding at puncture site: 3% (1/33)Total: 3%* Bleeding at puncture site: 3% (1/33)** stopped after 5 min compression–27. Kodama, H et al.Vertebroplasty* 6 patients* 8 vertebraeCT guided* CT post-op–0%0%–28. König MA et al.Kyphoplasty* 11 patients* 23 vertebraeFluoroscopy* No post-op CT5,1 mL / vertebra (1–12 mL)Total: 36,4%* Cement Leakage: 18,2% (2/11)* DVT: 9,1% (1/11)* Adjacent level fracture: 9,1% (1/11)Total: 18,2%* DVT: 9,1% (1/11)* Adjacent level fracture: 9,1% (1/11)Intracanal: 100% (2/2)29. Lee, B et alVertebroplasty* 18 patients* 33 vertebraeFluoroscopy & CT-guided* -–Total: 5,6%* Cement Leakage: 5,6% (1/18)Total: 5,6%* Cement Leakage: 5,6% (1/18)** transient pain (<24 h)Puncture site: 100%30.1 Li, Z et al.Kyphoplasty* 42 patients* 83 vertebraeFluoroscopy* CT post-op–Total: 16,9%* Cement Leakage: 16,9% (14/83)0%Paravertebral soft tissue: 28,6% (4/14)Intradiscal: 7,1% (1/14)Intravenous: 64,3% (9/14)30.2 Li, Z et al.Vertebroplasty* 38 patients* 76 vertebraeFluoroscopy* CT post-op–Total: 30,3%* Cement Leakage: 30,3% (23/76)0%Paravertebral soft tissue: 30,4% (7/23)Intradiscal: 13,1% (3/23)Intravenous: 56,5% (13/23)31.Lim, B et al.Vertebroplasty* 102 patients* 185 vertebraeFluoroscopy* No post-op CT3,62 mL / vertebraTotal: 17,3%* Cement Leakage: 17,3% (32/185)0%Intracanal: 40,1% (13/32)Intradiscal: 40,1% (13/32)Intravenous: 18,8% (6/32)32. Liu, H et al.Kyphoplasty* 234 patients* 288 vertebraeFluoroscopy* -–Total: 6,8* Cement Leakage: 6,4% (15/234)* Pulmonary cement embolism: 0,4% (1/234)0%Paravertebral soft tissue: 53,3% (8/15)Intradiscal: 46,7% (7/15)33. Liu, Y et al.Vertebroplasty* 30 patients* 30 vertebraeCT-guided3,47 mL / vertebraTotal: 53,3%* Cement Leakage: 40% (12/30)* Haematoma at puncture site:13,3% (4/30)Total: 13,3%* Haematoma at puncture site: 13,33%–34. Masala, S et al.Vertebroplasty* 64 patients* 198 vertebraeFluoroscopy* No post-op CT–0%0%–35. Moulin, B et al.Vertebroplasty* 50 patients* 397 vertebraeFluoroscopy* CT post-op* Chest-CT performed in 38 patients–Total: 57%* Pulmonary cement embolism: 47% (18/38)* Post-operative pain: 4% (2/50)* Post-operative transient alteration of general status: 2% (1/50)* Retroperitoneal haematoma: 2% (1/50)* Lumbar artery puncture: 2% (1/50)Total: 10%* Pulmonary cement embolism: 2% (1/50)* Post-operative pain: 4% (2/50)* Post-operative transient alteration of general status: 2% (1/50)* Lumbar artery puncture: 2% (1/50)** embolization intra-op–36. Nas, OF et al.Vertebroplasty* 52 patients* 95 vertebraeFluoroscopy* -4,91 mL / vertebra (1,5–9 mL)Total: 45,3%* Cement Leakage: 45,3% (43/95)0%Intradiscal: 46,5% (20/43)Intravenous: 46,5% (20/43)Intradiscal and intravenous: 7% (3/43)37. Pflugmacher, R et alKyphoplasty* 20 patients* 48 vertebraeFluoroscopy* No post-op CT4,6 mL / vertebra (2,6–7,9 mL)Total: 15,4%* Cement Leakage: 10,4% (5/48)* Adjacent level fracture: 5% (1/20)Total: 5%* Adjacent fracture: 5%** need for surgical interventionParavertebral soft tissue: 40% (2/5)Intradiscal: 60% (3/5)38. Pflugmacher, R et alKyphoplasty* 31 patients* 64 vertebraeFluoroscopy* No post-op CT–Total: 12,5%* Cement Leakage: 12,5% (8/64)0%Paravertebral soft tissues: 25% (2/8)Intradiscal: 50% (4/8)Intravenous: 25% (2/8)39. Pflugmacher, R et alKyphoplasty* 65 patients* 99 vertebraeFluoroscopy* No post-op CT–Total: 20,1%* Cement Leakage: 12,1% (12/99)* Adjacent fracture: 8% (5/65)0%–40. Qian, Z et al.Kyphoplasty* 48 patients* 97 vertebrae in study groupFluoroscopy* CT post-op–Total: 18,6%Cement Leakage: 18,6% (18/97)0%Paravertebral soft tissue: 22,2% (4/18)Intradiscal: 16,7% (3/18)Intravenous: 61,1% (11/18)41. Rajah, G et al.Kyphoplasty* 45 patients* 75 vertebraeFluoroscopy3,6 mLTotal: 18,6%* Cement Leakage: 13,3% (10/75)* Failure (retropulsion of cement into canal): 5,3% (4/75)Total: 5,3%* Failure: 5,3%** need for surgical decompression and fusion–42. Ramos, L et alVertebroplasty* 12 patients* 19 vertebraeFluoroscopy* CT post-op–Total: 84%Cement Leakage: 84% (16/19)0%Intracanal: 12,5% (2/16)Intradiscal: 18,7% (3/16)Intravenous: 68,8% (11/16)43.Reyad, RM et al.Vertebroplasty* 63 patients* 92 vertebraeFluoroscopy* CT 1 week post-op1,5 mL / vertebraTotal: 82,5%* Transient exacerbation of pain: 61,9% (39/63)* Local haematoma: 9,5% (6/63)* Cement leakage: 7,9% (5/63)* Adjacent level fracture: 3,2% (2/63)Total: 74,6%* Exacerbation of pain: 61,9%*Local haematoma: 9,5%* Adjacent fracture: 3,2%Paravertebral soft tissues: 20% (1/5)Intracanal: 20% (1/5)Intradiscal: 40% (2/5)Intravenous: 40% (2/5)44. Saracen, A et alVertebroplasty* 90 patients* 137 vertebraeFluoroscopy* No post-op CT2–3 mL / vertebraTotal: 35,1%* Cement leakage: 34% (47/137)* Pulmonary embolism: 1,1% (1/90)Total: 1,1%* Monoparesis due to cement intracanal leakage, resolved with surgery: 1,1% (1/90)Paravertebral soft tissue: 25,5% (12/47)Intracanal: 14,9% (7/47)Intradiscal: 19,2% (9/47)Intravenous: 40,4% (19/47)45. Shi X et al.Vertebroplasty* 309 patients-* -–Total: 31,2%* Cement leakage: 31,2% (189/309)Total: 1%* Neurological symptoms due to cement leakage: 1% (3/309)Paravertebral soft tissue: 29,4% (74/252)Intracanal: 13,1% (33/252)Intradiscal: 22,6% (57/252)Intravenous: 34,9% (88/252)* 252 types of leakage due to more than 1 type in the same vertebra46. Shimony, JS et al.Vertebroplasty* 50 patients* 129 vertebraeFluoroscopy* -–Total: 14%* Increased or new sites of pain: 14% (7/50)* Asymptomatic cement leakage was not considered as a complicationTotal: 14%* Increased or new sites of pain: 14% (7/50)–47. Tekin, SB et al.Vertebroplasty & Kyphoplasty* 42 patients* 76 vertebraeFluoroscopy* No post-op CT–Total: 7,1%* Vacuum phenomena: 4,7% (2/42)* Cement leakage: 2,4% (1/42)0%–48. Tian, Q et al.Vertebroplasty* 39 patients* 51 vertebraeFluoroscopy* CT post-op3,3 mL / vertebra (2–5 mL)Total: 38,5%* Cement leakage: 38,5% (15/39)0%Paravertebral soft tissues: 60% (9/15)Intracanal: 6,7% (1/15)Intravenous: 20% (3/15)Puncture site: 13,3% (2/15)49. Trumm, CG et al.Vertebroplasty* 76 vertebraeFluoroscopy* CT post-op3–3,5 mL / vertebraTotal: 75%* Cement leakage: 75% (57/76)0%Paravertebral soft tissue: (42,2% 27/64)Intracanal: 26,6% (17/64)Intradiscal: 31,2% (20/64)50. Vrionis, FR et alKyphoplasty* 50 patients* 128 vertebraeFluoroscopy* No post-op CT–Total: 2%* Asystole: 2%Total: 2%* Asystole: 2%** successfully reversed, no cause found–51. Wang, H et al.Kyphoplasty* 36 patients* 36 vertebrae-* No post-op CT–Total: 11,1%Cement leakage: 11,1% (4/36)0%Paravertebral soft tissue: 50% (2/4)Intradiscal: 25% (1/4)Intravenous: 25% (1/4)52. Wang, L et al.Vertebroplasty* 64 patients* 113 vertebraeFluoroscopy* CT post-op4,84 mL / vertebra (1,2–9 mL)Total: 56,6%* Cement leakage: 56,6% (64/113)0%Paravertebral soft tissue or intradiscal: 54,7% (35/64)Intracanal: 28,1% (18/64)Intravenous: 17,2% (11/64)53. Wang, Y et al.Kyphoplasty* 41 patients* 73 vertebraeFluoroscopy* CT post-op3,6 mL / vertebra (1,5–6,5 mL)Total: 17,1%* Cement leakage: 17,1% (7/41)0%–54. Weill, A et al.Vertebroplasty* 22 sessions* 21 patients* 27 vertebraeFluoroscopy* CT post-op–Total: 45,4%* Cement leakage: 40,9% (9/22)* Cement embolism to IVC: 4,5% (1/22)Total: 9%* Acute raquialgia because of cement leakage into canal: 9% (2/22)** need for surgical treatment in 1/2 casesParavertebral soft tissue: 11,1 (1/9)Intracanal: 22,2 (2/9)Intradiscal: 11,1 (1/9)Intravenous: 55,6 (5/9)55. Wu, W et al.Kyphoplasty* 31 patients* 58 vertebraeFluoroscopy* CT post-op2,6 mL / vertebraTotal: 6,5%* Cement leakage: 6,5% (2/31)0%Intradiscal: 50% (1/2)Puncture Site: 50% (1/2)56. Wu, W et al.Kyphoplasty* 59 patients* 59 vertebraeFluoroscopic* CT post-op3,84 mLTotal: 18,6%* Cement leakage: 18,6% (11/59)0%Intracanal: 18,2% (2/11)Intradiscal: 36,4% (4/11)Intravenous: 18,2% (2/11)Puncture Site: 27,2% (3/11)57. Yamada, K et al.Vertebroplasty* 4 patients* 5 vertebraeCT-guided* CT post-op4 mL /vertebra (1,5–7,5 mL)Total: 50%* Cement leakage: 50% (2/4)0%–58. Yan, B C et al.Vertebroplasty* 87 patients* 111 vertebraeFluoroscopy* CT post-op5,87 mL/vertebraTotal: 114%* Cement leakage: 98% (109/111)* Fever: 12,6% (11/87)* Pain: 3,4% (3/87)Total: 16%* Fever: 12,6% (11/87)* Pain: 3,4% (3/87)Paravertebral soft tissue: 49,1% (77/157)Intracanal: 20,4% (32/157)Intradiscal: 19,7% (31/157)Intravenous: 10,8% (17/157)59. Zhang, H et al.Kyphoplasty* 13 patients* 13 vertebraeFluoroscopy* No post-op CT–Total: 15,4%* Cement leakage: 15,4% (2/13)0%Intracanal: 33,3% (1/3)Intravenous: 66,7% (2/3)60.Zhang, J et al.Kyphoplasty* 28 patients* 37 vertebraeFluoroscopy* No post-op CT3,70 mL / vertebra0%0%– Treatment characteristics & reported complications(When available, complication rates were extracted according to the denominator reported in each study (per vertebra or per patient). In studies where this was not specified, cement leakage rates were calculated relative to the number of treated vertebrae, whereas other complications were calculated relative to the total number of procedures.) The mean reported complication rate was 24.3% across studies, corresponding to a weighted mean of 25.5% when adjusted for study sample size. Six studies reported no complications [18], [40], [41], [47], [49], [50] Fig. 2. Of these, the study population was under 10 patients in three [18], [41], [49] and ranged from 28 to 61 patients in the other three. [40], [47], [50]Fig. 2Incidence of reported complications after verteble augmentation for metastatic spinal fractures. Values represent weighted inceidence(%) across the included studies, adjusted for study sample size. A patient may have experienced more than one complication.Fig. 2 Incidence of reported complications after verteble augmentation for metastatic spinal fractures. Values represent weighted inceidence(%) across the included studies, adjusted for study sample size. A patient may have experienced more than one complication. The most frequently described complication was cement leakage, reported in 50 articles (82%), with rates ranging from 1,3% to 98.7%. The mean leakage rate across studies was 19.3%, with a weighted mean of 20.7% when adjusted for study sample size. Studies using CT either for procedural guidance or postoperative evaluation reported significantly higher leakage rates compared with studies without CT (28.6% vs 11.9%, p = 0.004). No significant association was found between the type of vertebral augmentation procedure and the reported cement leakage rate (p = 0.27). Similarly, the number of vertebral levels treated per session was not significantly associated with cement leakage frequency (p = 0.45). Of the 50 studies reporting cement leakage, only 37 described the pattern of leakage. The most common was venous leakage (mean 27.3%), followed by intradiscal leakage (26.8%), soft tissue leakage (22.0%), intracanal leakage (14.9%) and other types of leakage, such as around the injection site (5.1%). Some studies did not report all types of leaks, which could affect the understanding of the true pattern of cement leakage. Other complications were rare across the included studies. The most frequently reported events were postprocedural pain (mean 1.8%), pulmonary cement embolism (mean 1.1%), and adjacent level fracture (mean 0.8%), while all other complications occurred in less than 0.5% of cases, including local haematoma (0.4%), fever (0.3%), pulmonary embolism or deep vein thrombosis (0.2%), inferior vena cava embolism (0.08%), procedural failure (0.09%), bleeding (0.05%), superficial infection (0.03%), and balloon asymptomatic rupture (0.02%). 39 of the 60 included studies reported no symptomatic complications. The mean rate of symptomatic complications across studies was 3,8%, with a weighted mean of 3,4% when adjusted for study sample size, ranging from 1% to 74,6%. The latter corresponds to the study by Reyad, RM et al., where 61,9% of patients presented with post-procedural pain, which may simply reflect the use of local anaesthesia. No significant association was observed between symptomatic complication rates and the type of vertebral augmentation procedure (p = 0.274) or the number of vertebral levels treated per session (p = 0.375). Studies using CT for intraoperative guidance or postoperative evaluation reported higher mean rates of symptomatic complications (5.1% vs 2.8%), although this difference was not statistically significant (p = 0.918). Among symptomatic complications, the most frequently reported were post-procedural pain (21.3%), fever (12.6%), local haematoma (7.8%), and deep vein thrombosis (6.8%). Less frequently, adjacent-level fracture (5.0%), pulmonary cement embolism (4.0%), and symptomatic cement leakage. Complications described in case reports Case reports identified several rare complications associated with vertebral augmentation procedures. Case reports included descriptions of cement embolism, symptomatic cement leakage with neurological deficits, needle-tract tumour seeding, tumour extrusion beyond the vertebral body, cement migration to cardiac structures, subdural haematoma, extradural cyst, and transient paraparesis. Several of these complications, particularly symptomatic cement leakage and embolic events, were also reported in larger observational studies included in this review. In contrast, complications such as needle tract tumour seeding, tumour extrusion, and cement migration to cardiac structures were described almost exclusively in case reports. Quality assessment The methodological quality of the included analytical studies was assessed using the JBI critical appraisal tools according to study design (see Appendix B). Overall, the risk of bias among cohort and comparative studies was moderate, with the proportion of positive responses across JBI items ranging from approximately 44% to 100%. Consequently, the findings of the present review should be interpreted in the context of the observational nature of the available evidence. Discussion This systematic review aimed to synthesise complications associated with vertebral augmentation in metastatic spinal disease. Overall, vertebroplasty and kyphoplasty appear safe for pain relief and spinal stabilisation in oncologic patients. After adjustment for sample size, the overall complication rate was 25.5%, while symptomatic complications occurred in only 3,4%. Cement leakage Several previous systematic reviews have attempted to identify risk factors for complications after vertebral augmentation. A systematic review published in 2024 suggested that posterior cortical disruption, vertebral body fracture, and the number of treated vertebrae may increase the risk of cement leakage. However, that review included both percutaneous vertebral augmentation and open surgical techniques, which may limit direct comparison with the present review [51]. Similarly, previous systematic reviews evaluating vertebral augmentation for malignant vertebral compression fractures have reported cement leakage as the most frequently observed complication, although symptomatic events remain uncommon [7], [52]. Patients with metastatic spinal disease often exhibit extensive vertebral destruction, including cortical posterior wall involvement. Coupled with the rich vascular supply of the vertebral column, this may increase the risk of cement leakage. Reduced bone quality related to age and systemic anticancer treatments further increases their risk and susceptibility to complications [53] [54]. In our review, cement leakage was the most frequently reported complication, consistent with prior literature. However, its clinical impact appears limited, as most cases were asymptomatic (20.7% of overall complications vs 3,2% of symptomatic ones). This finding suggests that radiological cement leakage should not automatically be interpreted as a clinically relevant adverse event. In some studies, small asymptomatic cement leaks were not considered complications and were therefore excluded from complication rates, increasing uncertainty regarding the true clinical relevance of cement leakage. The heterogeneity among the included studies in the definition of included patients did not allow for a statistical evaluation of the influence of posterior vertebral wall compromise in the occurrence of leakage. Cement leakage can occur through intradiscal, intravascular, or paravertebral pathways, and was associated with complications such as nerve compression, pulmonary and intracardiac embolism, adjacent vertebral fractures, and local pain, which is consistent with previous reports in the literature [53]. Neurological complications are rare but particularly serious, as described by Chen, YJ et al. and Cosar M et al. [55], [56]. However, it remains unclear how many patients develop symptoms after cement leakage, and the factors that increase the risk of clinically significant complications are not fully understood. Future studies should distinguish between radiographic and clinically significant leakage when reporting outcomes. Experience with vertebral augmentation for osteoporotic and traumatic fractures may inform the interpretation of findings in metastatic disease. In osteoporotic fractures, systematic reviews report cement leakage rates of 18.4% to 54.7%, depending on the augmentation technique, with most events radiographic rather than clinically significant [91], [92]. These rates are comparable to, or even higher than, the weighted leakage rate of 20.7% observed in our metastatic population. Importantly, cortical disruption has been identified as a major risk factor for leakage in osteoporotic fractures, together with intravertebral clefts, lower cement viscosity, and greater cement volume [92]. These factors may also be relevant in metastatic lesions, particularly when cortical destruction is present; however, the heterogeneity of the included studies precluded their assessment in our review. Evidence for traumatic fractures is more limited but shows a similar predominance of radiographic over symptomatic leakage: a systematic review identified cement leakage as the most reported complication, without associated clinical consequences [93], whereas a CT-based series detected leakage in 50% of cases, all asymptomatic [94]. Although these findings may suggest a comparable pattern of procedural safety, the risk and spectrum of complications cannot be fully extrapolated across indications, given their distinct biological settings. Metastatic disease may be associated with complications not described in the osteoporotic or traumatic literature, such as tumour extrusion and needle-tract tumour seeding, as described in this review. Pulmonary cement embolism Pulmonary cement embolism following venous leakage has been linked to treatment of multiple vertebral levels, involvement of the thoracic vertebrae, and use of larger cement volumes or lower cement viscosity [11], [29], [45]. Similar factors have been identified in a systematic review and meta-analysis of predominantly osteoporotic vertebral fractures, which reported an overall PCE incidence of 2.3% [95]. In our review, we identified an overall incidence of 1.08%. This overlap in reported risk factors suggests that cement embolisation may be substantially influenced by procedural factors rather than by the underlying vertebral pathology alone. Nevertheless, the influence of the underlying disease on the clinical significance of PCE remains unclear. Although some authors describe it as an incidental finding on CT scans, several symptomatic cases have been reported, not only in case reports but also in observational studies (accounting for 4% of all symptomatic complications). Therefore, although the true incidence of cement embolism after venous leakage remains unclear, it is important to be aware of its potential in symptomatic patients. Influence of multilevel vertebral augmentation Multilevel vertebral augmentation has been associated with greater complexity, leading to increased procedural difficulty and risks, including cement leakage and embolic complications [34]. Although treatment of more than 3 vertebrae in a single session occurred in 47,3% of studies, these cases represented a minority of the patients within each study, and there was no significant correlation with increased complications (p = 0.45). However, the absence of subgroup analysis in studies involving multilevel procedures affects the ability to draw definitive conclusions, and further research is necessary to better understand the safety of such procedures. Influence of type of procedure Some studies have suggested that the type of augmentation procedure may influence complication rates, with lower rates of cement leakage with kyphoplasty [11], [46]. However, in this review, there was no statistically significant association between procedure type and complication occurrence (p = 0.27). Influence of CT evaluation Postoperative imaging protocols may also affect the reported rates of complications. In studies in which patients underwent intraoperative or postoperative CT evaluation, the rate of detected cement leakage was significantly higher, but not the rate of symptomatic complications. This reflects CT's higher sensitivity for detecting small cement leaks [16], but its clinical relevance is unknown. Further studies are needed to determine whether the increased sensitivity of CT justifies its cost and routine use, given that most detected cement leaks are asymptomatic. Rare complications The case reports provided valuable information about rare complications that are not commonly described in larger observational studies. Some of these complications, such as neurological compromise from cement leakage and symptomatic pulmonary cement embolisms, are described in the included observational studies, but extremely rare complications, including needle-tract seeding, tumour extrusion, and cement migration into cardiac structures, were reported almost exclusively in case reports. These findings emphasise that although most complications linked to vertebral augmentation procedures are rare, infrequent yet potentially serious events can still occur and should always be considered if such symptoms arise. Beyond documented clinical complications, Mohme et al. reported a transient increase in circulating tumour cells after cement augmentation of vertebral metastases, raising the possibility of procedure-related mechanical dissemination of tumour cells. Although the clinical significance of this finding remains uncertain, it highlights a potential biological effect of vertebral augmentation that warrants further investigation [96]. DVT and non-cement pulmonary embolism were also reported as rare postoperative events. However, their direct relationship with vertebral augmentation remains uncertain, as patients with metastatic disease have multiple concurrent risk factors for venous thromboembolism, including malignancy and reduced mobility. This further supports the need for a clear definition of postoperative complications, as these events may be related to the underlying disease or the peri-procedural setting rather than to cement injection itself. Strengths and limitations This review also has some limitations. First, the heterogeneity of study designs and reporting methods limited comparison between studies and the ability to draw firm conclusions about complication rates and risk factors. In addition, many of the included studies were case series or retrospective observational studies, which may be affected by selection bias and incomplete reporting of complications. Furthermore, the definition and reporting of complications varied widely across studies, particularly for asymptomatic cement leakage, which was not always considered a complication. Another factor that was not consistently reported is follow-up duration; when available, it varied considerably. This variability may influence the detection of complications, particularly events such as adjacent-level fractures that can occur only months after the procedure. Furthermore, the radiological pattern of metastatic bone involvement (lytic, sclerotic, or mixed), patient-selection criteria, and prior oncological treatments were inconsistently reported, preventing assessment of their potential influence on complication rates. Thus, despite the large number of available studies, the overall quality of evidence remains limited, highlighting the need for higher-quality prospective studies rather than simply a greater volume of literature. Additionally, studies combining vertebral augmentation with other spinal interventions, such as surgical fixation or tumour ablation, were excluded to minimise confounding, as reported complications could not be reliably attributed to vertebral augmentation alone. Despite these limitations, this review also has several strengths. First, it includes many patients and treated vertebral levels across the analysed studies, with more than 3000 patients and over 5700 treated vertebrae, providing a broad overview of complications associated with vertebral augmentation procedures in metastatic spinal fractures. In addition, by including different study designs, it was possible to identify both common and rare complications. To our knowledge, few previous reviews have attempted to systematically describe the full spectrum of complications, including uncommon but clinically relevant events such as needle tract tumour seeding, tumour extrusion, and intracardiac cement embolism. This broader perspective may help provide a more complete understanding of the safety profile of vertebral augmentation in oncologic patients. Conclusion The available evidence suggests that vertebral augmentation is a safe treatment option for painful metastatic vertebral fractures, with a low rate of symptomatic complications (< 4%). However, important uncertainties remain, particularly the role of multilevel procedures, the clinical significance of cement leakage, and the relevance of CT evaluation. The apparent safety profile of vertebral augmentation depends strongly on how complications are defined and detected. A clearer understanding of procedural safety is essential to support informed clinical decision-making, optimise procedural planning, and appropriately balance potential benefits and risks in this vulnerable patient population. CRediT authorship contribution statement Maria Clara Correia: Writing – review & editing, Writing – original draft, Visualization, Validation, Software, Resources, Project administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization. Ana Margarida Vieira: Writing – original draft, Methodology, Investigation, Formal analysis, Data curation. Filipe Duarte: Writing – review & editing, Validation, Supervision, Resources, Formal analysis, Data curation, Conceptualization. André Rodrigues Pinho: Writing – review & editing, Project administration, Methodology, Investigation, Conceptualization. Francisco Serdoura: Supervision. Maria João Leite: Writing – review & editing, Validation, Supervision, Project administration, Methodology, Investigation, Formal analysis, Conceptualization. Ethics approval Not applicable. This study is a systematic review of previously published data. Consent to participate. Not applicable. Consent for publication. Not applicable. Funding The authors declare that no funds, grants, or other support were received during the preparation of this manuscript. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.