Mortality After Osteoporotic Vertebral Body Fractures: Cement Augmentation Versus Conservative Treatment

A systematic review and meta-analysis

Background: Osteoporotic vertebral body fractures are associated with a 1-year mortality of 28%, a figure similar to that for hip fractures. Retrospective registry studies from multiple countries have shown a survival advantage in patients treated with an augmentation procedure.

Methods: We carried out a systematic review of publications retrieved from the PubMed, Livivo, Cochrane Central, and Web of Science databases, and from the internet at large via Google Scholar. Studies were selected for analysis if they included a comparison of mortality in patients with osteoporotic vertebral body fractures who were treated conservatively versus those who were treated with cement augmentation. We also carried out a pooled quantitative analysis based on the findings of all studies in which a hazard ratio (HR) comparing the two modes of treatment was reported.

Results: 22 publications met the inclusion criteria, including two randomized, controlled trials (RCTs). The two RCTs and 8 of the remaining studies showed no significant difference in mortality between the two groups; a significant positive effect of augmentation was reported in a further 11 studies, and a significant negative effect in 1. 11 studies were suitable for inclusion in the quantitative analysis. The pooled HR for death was 0.91 (95% confidence interval [0.73; 1.15]; p = 0.39), i.e., mildly in favor of augmentation, but statistically insignificant.

Conclusion: A wide variety of confounders may have led to an overestimation, in retrospective studies, of the survival advantage putatively conferred by cement augmentation of vertebral body fractures compared to conservative treatment. Our systematic review and meta-analysis revealed no significant association of either form of treatment with a higher survival rate.

Cite this as: Wiedl A, Geiger F, Eberl J, Wirries A, Hammad A: Mortality after osteoporotic vertebral body fractures: cement augmentation versus conservative treatment—a systematic review and meta-analysis. Dtsch Arztebl Int 2025; 122: 604–10. DOI: 10.3238/arztebl.m2025.0134

In light of an aging population, the industrial nations are facing a growing burden from osteoporosis and its implications (1). In 2019, the prevalence of osteoporosis in Germany was 6.1% of the total population (2). As a result, the number of osteoporotic thoracolumbar vertebral fractures is also rising (3).

As with proximal femoral fractures (one-year mortality: 27.6% [4]), osteoporotic vertebral body fractures are also associated with increased mortality (one-year mortality 28% [5]). Pre-existing comorbidities and frailty, pain-related immobilization, and progressive kyphosis are just some factors which could be responsible for this connection (6, 7, 8); furthermore, the risk of subsequent fractures secondary to osteoporosis is increased by around three-fold in the following two years (odds ratio [OR]: 3.3; 95% confidence interval: [2.3; 4.7]) (9).

Cement augmentation has proven itself to be a clinically effective means to reduce pain and period of immobilization for the fracture and has become a firm component of therapeutic algorithms (10, 11). The cement injected into the vertebral body has a stabilizing effect on the fracture, thus counteracting further deformity and subsidence of the fracture and reducing pain. The procedure can be performed with (kyphoplasty) or without (vertebroplasty) prior realignment of the vertebral body by inserting a small inflatable balloon.

The positive effect of cement augmentation on the aforementioned clinical issues suggested that the operation also reduces mortality (12). Retrospective studies involving large insurance data sets from the American Medicare system have reported such a survival benefit (12, 13, 14, 15).

A meta-analysis from the year 2020 quantified this survival benefit with a pooled hazard ratio (HR) for death of 0.78 [0.66; 0.92] in comparison with conservative therapy (16). But now, a more recent study criticizes the previously conducted registry studies for significant methodological flaws and was unable to reproduce any survival advantage after making corrections but instead calculated a survival disadvantage of cement augmentation (17).

The aim of the present article is to conduct an updated review of publications and a meta-analysis of all currently available data comparing the mortality risk of patients treated either by augmentation or conservatively after thoracolumbar vertebral body fractures.

Methods

The PRISMA Statement (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guideline was adopted for the present study (18). The study was registered with PROSPERO (CRD42023434497). From September 11 to October 4, 2023, the PubMed, Livivo, Cochrane Central, and Web of Science databases were searched with no restrictions on the publication period. A further internet search was undertaken using Google Scholar. Study selection was conducted using the PICOS framework (eTable 1) (Patient Population, Intervention, Comparison, Outcome, Study Design), retrieving studies which assessed mortality after osteoporotic thoracolumbar vertebral fractures and compared cement augmentations of whatever kind (kyphoplasty or vertebroplasty) with conservative treatment. eTable 2 presents the precise search strategy and the algorithms used. Andreas Wiedl and Ahmed Hammad then reviewed and selected the studies. Particular emphasis was placed on selecting studies which reported an HR as the measure of mortality risk following augmentation of osteoporotic vertebral fractures in comparison with conservative treatment. Both reviewers assessed risk of bias using the Cochrane Collaboration tools RoB 2 and ROBINS-I (19, 20). A qualitative descriptive analysis was performed on all selected studies. A Fisher’s exact test was conducted to assess significance when only a numerical count of deceased patients was available. Studies reporting an HR were included in the quantitative analysis to estimate the pooled HR across all studies. The HR was intended to reflect the risk ratio of augmentation versus conservative treatment. Where the HR was stated as inverse, it was converted using the formula 1/HR for mean HR and the 95% confidence interval. Adjusted values were always extracted when provided, and if multiple adjusted HRs were available, the value assessed by the reviewers as having the highest degree of confounder adjustment was selected. The inverse-variance method of a random-effects model was used to pool the HRs because the preliminary analyses showed a high level of heterogeneity of the results of the individual studies. Heterogeneity variance was estimated using the DerSimonian and Laird method. Cochran’s Q was calculated for this purpose in addition to performing the Higgins and Thompson I² test (21). In order to identify individual studies with high potential for heterogeneity, a sensitivity analysis was performed eliminating one study at a time in a stepwise fashion. A subgroup analysis was performed after pooling studies which either corrected for, or did not correct for, the so-called immortal time bias (22). A second subgroup analysis identified studies with a follow-up of one to two and of five to ten years. The publication bias was assessed using Egger’s test and a funnel plot. Statistical analysis was performed using the RStudio software (Posit Software, Version 2024.04.2+764) and the R programming language (The R Foundation for Statistical Computing, Version 4.4.0).

eTable 1

PICOS framework

eTable 2

Applied search strategy and identified references

Results

Literature review

A total of 1293 studies and citations were identified (eFigure 1). After eliminating 407 duplicates, 886 studies remained. Abstract screening left 67 studies for assessment by full-text screening, after which 22 qualified for the descriptive review (12, 13, 14, 15, 17, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39). Two studies (38, 39) reported an OR for inpatient mortality risk. Fourteen studies reported an HR comparing at least one augmentation procedure and conservative treatment. Three of these studies (12, 14, 15) assessed Medicare data for periods which were also included in another study (13) covering a longer period. Only the last study (13) was included in the quantitative analysis to avoid a double assessment of the findings, leaving a total of eleven studies (13, 17, 23, 24, 26, 27, 28, 32, 33, 34, 35).

eFigure 1

Flow diagram demonstrating study selection process after initial review of publications retrieved from the PubMed, Livivo, Cochrane Central, Web of Science, and Google Scholar databases

General study characteristics

Eighteen studies (12, 13, 14, 15, 17, 23, 24, 25, 27, 28, 30, 32, 33, 34, 35, 36, 38, 39) and, as such, the majority of the study population, assessed the results retrospectively using registry and hospital data. Four studies (26, 29, 31, 37) adopted a prospective design, including two randomized controlled trials (RCTs) (31, 37) (Table 1, Table 2). The follow-up periods reported in the studies appeared highly heterogeneous, ranging from the length of hospital stay to as long as ten years. A total of 4 447 991 patients were included, of which 879 243 (19.8%) were treated with augmentation and 3 568 748 (80.2%) conservatively. Eleven studies (12, 13, 14, 15, 23, 24, 25, 29, 36, 38, 39) reported a statistically significant superior effect, ten studies (26, 27, 28, 30, 31, 32, 33, 34, 35, 37) reported no effect, and one study (17) reported a negative effect of augmentation on survival (Table 1, Table 2).

Table 1

Overview of studies reporting survival advantage after augmentation or conservative treatment

Table 2

Overview of studies reporting no treatment-related survival advantage

Overall, the study population showed a high bias potential (eTable 3). Only three studies demonstrated a moderate risk of bias (33, 34, 37). Nine publications used registry data from the USA (12, 13, 14, 15, 17, 34, 36, 38, 39), three others from France (25), Taiwan (24), and Germany (23). The remaining publications reported single-center or multicenter surveys (26, 27, 28, 29, 30, 31, 32, 33, 35, 37).

eTable 3

Risk of bias analysis using the ROBINS-I and RoB2 tools; ROBINS-I tool applied for non-randomized studies

The most comprehensive study in terms of quantity was by Ong et al. (13) who reported an HR (augmentation versus conservative) of 0.83 [0.82; 0.83]. Many of the retrospective studies corrected for confounding bias using regression methods such as multivariate Cox regression analysis and/or propensity score matching (eTable 4). Apart from the studies by Gold et al. (17), McDonald et al. (35), and McCullough et al. (34), no other article took additional measures to reduce selection bias. Gold et al. (17) was the only publication which calculated a negative effect of augmentation with an HR of 1.81 [1.58; 2.09]. Two RCTs examined patient populations of adequate sizes of 300 (37) and 400 (31) patients, respectively (Table 2). Wardlaw et al. reported 6% (9 of 149) deaths in the augmentation and 4.6% (7 of 151) in the conservative group after one year (37). Leali et al. reported six-month mortality rates of 0.5% (1 out of 200) in the operative and 1.5% (3 out of 200) in the conservative cohort (31).

The present analysis found no statistically significant difference based on Fisher’s exact text for both RCTs (p = 0.618 [37] and p = 0.623 [31]).

Quantitative analysis

A pooled analysis was conducted of eleven studies (13, 17, 23, 24, 26, 27, 28, 32, 33, 34, 35) which were deemed suitable for synthesis (Figure 1). This revealed an HR (augmentation versus conservative management) of 0.91 [0.73; 1.15] and a prediction interval (PI) of [0.44; 1.88], p = 0.39, which indicated no significant effect across studies. Heterogeneity between the studies was marked with an I² value of 93.1% [89.5%; 95.4%]; Cochran’s Q confirmed this with a value of 144.3 (p <0.001).

Figure 1

Forest plot after pooling the HRs of all studies included in the quantitative analysis

In the subgroup analyses, studies that corrected for selection bias (17, 34, 35) showed an HR of 1.19 ([0.48; 2.97]; PI: [0.00; 399.73]; p = 0.49) (Figure 2), whereas studies without such a correction demonstrated an HR of 0.76 ([0.63; 0.91], PI: [0.48; 1.21]; p = 0.009) (eFigure 2) (13, 23, 24, 26, 27, 28, 32, 33). The articles with a follow-up of one to two years (17, 24, 26, 33, 34) showed an average HR of 1.09 ([0.69; 1.72], PI: [0.23; 5.15]; p = 0.65) (eFigure 3) and 0.78 ([0.61; 1.00], PI: [0.44; 1.39] p = 0.048) for studies with a follow-up of five to ten years (12, 23, 27, 28, 32, 35) (eFigure 4). Stepwise elimination of individual studies did not reveal any clear reduction in heterogeneity, while the most striking decrease was evident after exclusion of the study by Gold et al. (17), which reduced heterogeneity to 64% (eFigure 5). The combined evaluation of the funnel plot (eFigure 6) and Egger’s test (p = 0.42, β0 = 1.054) [−1.4; 3.5] did not reveal any clear indication for the presence of a relevant publication bias.

Figure 2

Forest plot after subgroup analysis of studies that corrected for immortal time bias

eFigure 2

Forest plot after subgroup analysis and pooling of studies which were corrected for immortal time bias

eFigure 3

Forest plot after subgroup analysis of the studies with a follow-up of 1 to 2 years

eFigure 4

Forest plot after subgroup analysis of the studies with a follow-up of 5 to 10 years

eFigure 5

Sensitivity analysis with stepwise elemination of individual studies

eFigure 6

Funnel plot for graphic presemtation of selection bias

Discussion

Current evidence has so far led to the assumption that cement augmentation has a positive effect on survival after osteoporotic vertebral fractures. Similar to the meta-analysis conducted in 2020 (pooled HR of 0.78; significant [16]) the findings of the present article also tend to favor cement augmentation with an HR of 0.91. However, even the inclusion of more recent studies did not result in a significant effect.

Particular consideration should be given to immortal time bias, a subtype of selection bias. If mortality data from the time of fracture diagnosis are compared between the interventions, then it should be considered that patients due for augmentation must survive until the date of the operation and must not suffer a complication which would preclude augmentation. So, there will always be preselection whenever the augmentation group is not assessed in a randomized fashion. For studies which correct for this type of bias (17, 34, 35), subgroup analysis shows a tendency for a higher HR of 1.19 over an HR of 0.76 for studies without such a correction (13, 23, 24, 26, 27, 28, 32, 33).

The majority of the included publications refer to retrospective studies which analyzed very large case numbers (13, 14, 17). Ong et al. (13) conducted the largest study and, like five other publications (12, 14, 15, 17, 34), reported on data from the US Medicare system. Other authors used comparable methods and resorted to other databases from the USA, Taiwan, France, and Germany (23, 24, 25, 36, 38, 39). Altogether, eleven studies postulated a significant survival advantage for patients after cement augmentation. This figure should not be overrated, however, since the data from Chen et. al. and Edidin et al. (12, 14, 15) are fully included in the dataset of Ong et al. (13), thus producing a duplication of the published results. Ten studies did not demonstrate any significant effect direction. The focus of the two available RCTs was not directed towards assessing mortality but rather clinical aspects, such as pain, quality of life, and function (31, 37). None of these two studies showed significant treatment-related differences in mortality. Using Medicare data, only Gold et al. (17) calculated an adverse effect of augmentation on survival with an HR of 1.81. After a primary analysis stage conducted in a manner similar to that of Ong et al. (13) had calculated a comparable HR of 0.84, several matching steps were undertaken to minimize the selection bias until the above result was achieved. The study undertaken by Gold et al. (17) consequently had a stronger impact on the heterogeneity of the results of the quantitative analysis than other studies.

The bias potential of the included studies was largely high. Apart from the prospective randomized study by Wardlaw et al. (37), only the studies by Lotan et al. and McCullough et al. (33, 34) were flawed by a moderate risk of bias.

The non-randomized studies largely used controlling methods for confounders, yet a significant residual risk of bias due to confounding must nevertheless be assumed in comparison with prospective randomized studies. Only the studies by Gold et al., McDonald et al., and McCullough et al. (17, 34, 35) used controlling methods for the immortal time bias mentioned above. In some cases, a respective “pseudo-kyphoplasty” date was created for patients treated conservatively; patients who had died earlier or had relevant complications were either excluded or matched with the respective group receiving the other treatment.

The present article emphasizes the difficulty of drawing consistent conclusions from retrospectively acquired data sets. Observational studies and retrospective analyses are affected by a number of moderating variables in comparison with a prospective randomized study which could influence the choice of treatment arm and act as confounders. Similarly, it is not always clear from the studies what constituted conservative treatment.

Overall, the registry studies favoring augmentation used similar to almost identical analytical methods, which may present a risk of systematic bias. Nevertheless, the results of Gold et al. (17) should also still be interpreted with care. Although the analytical approach is plausible in its ability to eliminate bias effectively, this only gradually became evident over the course of the analysis. The two RCTs, which did not demonstrate a significant effect on treatments, also provide too little evidence to allow a clear statement regarding mortality, given that they involved a comparatively small number of patients and events. However, it should be noted that, although the pooled effect did not achieve any statistical significance, it did tend to suggest a reduction in mortality risk after cement augmentation. Nevertheless, the tendency for the effect to reverse should also be acknowledged when the immortal time bias is borne in mind. The question of the direction of effect remains unanswered, so future studies must adequately control for the confounding factors reported above or include larger patient populations in the randomized trials in order to clarify the issue.

So based on current evidence, the present article cannot recommend any definitive treatment to affected patients with the intention of reducing mortality. This does not impact on the potential for cement augmentation to reduce pain, improve mobility, and stabilize osteoporotic vertebral fractures (40). If improvement of these clinical factors is supposed to reduce mortality, it remains unclear why the results did not reflect this. Explanations may be subsequent fractures secondary to osteoporosis (thus neutralizing any positive effects of the augmentation procedure) or perioperative issues.

Box

Conclusions for clinical practice

Limitations

A limiting factor for this article and evaluation of its findings is the high level of heterogeneity of the included studies which was 93.1%. Apart from their different study designs, the studies adopted different bias-controlling methods and follow-up periods, and the use of subgroup analyses did not substantially reduce heterogeneity. The validity of the meta-analysis is also limited by the high bias potential of the included studies. Two of the studies (38, 39) calculated a risk ratio in terms of OR and only took inpatient follow-up into consideration. So, these two studies were not included in the analysis to avoid accentuating heterogeneity even further. Given the strong weighting of the Medicare studies and, with it, the United States as the primary source of data, there may also be a systematic bias in the results in view of the prevailing differences between national healthcare systems. The findings of the assessment of publication bias must be interpreted with caution since only eleven studies were included.

Summary

The present meta-analysis revealed no significant survival advantage or disadvantage resulting from cement augmentation of osteoporotic vertebral fractures in comparison with conservative treatment. Although data from registry studies are available, some of which argue in favor of a survival benefit from augmentation, there is a high potential for bias, and the results must therefore be interpreted with some degree of caution. Although augmentation cannot be recommended for the purpose of reducing mortality as based on the available literature, it remains a valuable option in the symptomatic treatment of osteoporotic vertebral fractures.

Acknowledgments

The authors would like to thank Dr. rer. biol. hum. Dennis Freuer and Prof. Dr. med. Christine Meisinger from the Institute for Epidemiology at the University of Augsburg for their support with the statistical questions.

Conflict of interest statement
The authors declare that there are no conflicts of interest.

Manuscript received on 17 November 2024, revised version accepted on 17 July 2025

Translated from the original German by Dr. Grahame Larkin

Corresponding author:
Dr. med. Andreas Wiedl

References

1. United Nations Department of Economic and Social Affairs. World Population Ageing 2023: Challenges and opportunities of population ageing in the least developed countries. S.l.: United Nations; 2024. www.un.org/development/desa/pd/sites/www.un.org.development.desa.pd/files/undesa_pd_2024_wpa2023-report.pdf.

2. Kanis JA, Norton N, Harvey NC, et al.: SCOPE 2021: A new scorecard for osteoporosis in Europe. Arch Osteoporos 2021; 16: 82.

3. Ballane G, Cauley JA, Luckey MM, El-Hajj Fuleihan G: Worldwide prevalence and incidence of osteoporotic vertebral fractures. Osteoporos Int 2017; 28: 1531–42.

4. Schulz C, König HH, Rapp K, Becker C, Rothenbacher D, Büchele G: Analysis of mortality after hip fracture on patient, hospital, and regional level in Germany. Osteoporos Int 2020; 31: 897–904.

5. Johnell O, Kanis JA, Odén A, et al.: Mortality after osteoporotic fractures. Osteoporos Int 2004; 15: 38–42.

6. Bliuc D, Center JR: Determinants of mortality risk following osteoporotic fractures. Curr Opin Rheumatol 2016; 28: 413–9.

7. Mccarthy J, Davis A: Diagnosis and management of vertebral compression fractures. Am Fam Physician 2016; 94: 44–50.

8. Kado DM, Lui LY, Ensrud KE, et al.: Hyperkyphosis predicts mortality independent of vertebral osteoporosis in older women. Ann Intern Med 2009; 150: 681–7.

9. Wustrack R, Seeman E, Bucci-Rechtweg C, Burch S, Palermo L, Black DM: Predictors of new and severe vertebral fractures: Results from the HORIZON Pivotal Fracture Trial. Osteoporos Int 2012; 23: 53–8.

10. Beall D, Phillips T: Vertebral augmentation: An overview. Skeletal Radiol 2023; 52: 1911–20.

11. Schnake KJ, Bula P, Spiegl UJ, et al.: Thorakolumbale Wirbelsäulenfrakturen beim alten Menschen: Klassifikation und Therapie. Unfallchirurg 2017; 120: 1071–85.

12. Edidin A, Ong K, Lau E, Kurtz S: Mortality risk for operated and nonoperated vertebral fracture patients in the medicare population. J Bone Miner Res 2011; 26: 1617–26.

13. Ong KL, Beall DP, Frohbergh M, Lau E, Hirsch JA: Were VCF patients at higher risk of mortality following the 2009 publication of the vertebroplasty “sham” trials? Osteoporos Int 2018; 29: 375–83.

14. Edidin AA, Ong KL, Lau E, Kurtz SM: Morbidity and mortality after vertebral fractures: Comparison of vertebral augmentation and nonoperative management in the medicare population. Spine (Phila Pa 1976) 2015; 40: 1228–41.

15. Chen AT, Cohen DB, Skolasky RL: Impact of nonoperative treatment, vertebroplasty, and kyphoplasty on survival and morbidity after vertebral compression fracture in the medicare population. J Bone Joint Surg Am 2013; 95: 1729–36.

16. Hinde K, Maingard J, Hirsch JA, Phan K, Asadi H, Chandra RV: Mortality outcomes of vertebral augmentation (vertebroplasty and/or balloon kyphoplasty) for osteoporotic vertebral compression fractures: a systematic review and meta-analysis. Radiology 2020; 295: 96–103.

17. Gold LS, Suri P, O’Reilly MK, Kallmes DF, Heagerty PJ, Jarvik JG: Mortality among older adults with osteoporotic vertebral fracture. Osteoporos Int 2023; 34: 1561–75.

18. Page MJ, McKenzie JE, Bossuyt PM, et al.: The PRISMA 2020 statement: An updated guideline for reporting systematic reviews. BMJ 2021; 372: n71.

19. Schünemann HJ, Cuello C, Akl EA, et al.: GRADE guidelines: 18. How ROBINS-I and other tools to assess risk of bias in nonrandomized studies should be used to rate the certainty of a body of evidence. J Clin Epidemiol 2019; 111: 105–14.

20. Sterne JAC, Savović J, Page MJ, et al.: RoB 2: A revised tool for assessing risk of bias in randomised trials. BMJ 2019; 366: l4898.

21. Higgins JPT, Thompson SG: Quantifying heterogeneity in a meta-analysis. Stat Med 2002; 21: 1539–58.

22. Yadav K, Lewis RJ: Immortal time bias in observational studies. JAMA 2021; 325: 686.

23. Lange A, Kasperk C, Alvares L, Sauermann S, Braun S: Survival and cost comparison of kyphoplasty and percutaneous vertebroplasty using German claims data. Spine (Phila Pa 1976) 2014; 39: 318–26.

24. Lin J, Chien L, Tsai W, Chen L, Chiang Y, Hsieh Y: Early vertebroplasty associated with a lower risk of mortality and respiratory failure in aged patients with painful vertebral compression fractures: A population-based cohort study in Taiwan. Spine J 2017; 17: 1310–8.

25. Maravic M, Taupin P, Roux C: Hospital burden of vertebral fractures in France: Influence of vertebroplasty. Osteoporos Int 2013; 24: 2001–6.

26. Diamond TH, Bryant C, Browne L, Clark WA: Clinical outcomes after acute osteoporotic vertebral fractures: A 2-year non-randomised trial comparing percutaneous vertebroplasty with conservative therapy. Med J Aust 2006; 184: 113–7.

27. Gerling MC, Eubanks JD, Patel R, Whang PG, Bohlman HH, Ahn NU: Cement augmentation of refractory osteoporotic vertebral compression fractures: Survivorship analysis. Spine (Phila Pa 1976) 2011; 36: E1266–9.

28. Gutierrez-Gonzalez R, Royuela A, Zamarron A: Survival following vertebral compression fractures in population over 65 years old. Aging Clin Exp Res 2023; 35: 1609–17.

29. Klezl Z, Bhangoo N, Phillips J, Swamy G, Calthorpe D, Bommireddy R: Social implications of balloon kyphoplasty: Prospective study from a single UK centre. Eur Spine J 2012; 21: 1880–6.

30. Lavelle WF, Khaleel MA, Cheney R, Demers E, Carl AL: Effect of kyphoplasty on survival after vertebral compression fractures. Spine J 2008; 8: 763–9.

31. Leali PT, Solla F, Maestretti G, Balsano M, Doria C: Safety and efficacy of vertebroplasty in the treatment of osteoporotic vertebral compression fractures: A prospective multicenter international randomized controlled study. Clin Cases Miner Bone Metab 2016; 13: 234–6.

32. Levy H, Seydafkan S, Rice JD, Easley KA, Tangpricha V: Comparative efficacy of vertebroplasty, kyphoplasty, and medical therapy for vertebral fractures on survival and prevention of recurrent fractures. Endocr Pract 2012; 18: 499–507.

33. Lotan R, Smorgick Y, Anekstein Y, Rudik O, Prosso I, Hershkovich O: Kyphoplasty for elderly patients with vertebral compression fractures-do we save lives? Mortality rates analysis comparison in a long-term follow-up cohort. Global Spine J 2022; 12: 1443–8.

34. McCullough BJ, Comstock BA, Deyo RA, Kreuter W, Jarvik JG: Major medical outcomes with spinal augmentation vs conservative therapy. JAMA Intern Med 2013; 173: 1514–21.

35. McDonald RJ, Achenbach SJ, Atkinson EJ, et al.: Mortality in the vertebroplasty population. AJNR Am J Neuroradiol 2011; 32: 1818–23.

36. Mills E, Ton A, Bouz G, Alluri R, Hah R: Acute operative management of osteoporotic vertebral compression fractures is associated with decreased morbidity. Asian Spine J 2022; 16: 634–42.

37. Wardlaw D, Cummings SR, Meirhaeghe JV, et al.: Efficacy and safety of balloon kyphoplasty compared with non-surgical care for vertebral compression fracture (FREE): A randomised controlled trial. Lancet 2009; 373 (9668): 1016–24.

38. Winkler E, Yue J, Birk H, et al.: Perioperative morbidity and mortality after lumbar trauma in the elderly. Neurosurg Focus 2015; 39: E2.

39. Zampini JM, White AP, McGuire KJ: Comparison of 5766 vertebral compression fractures treated with or without kyphoplasty. Clin Orthop Relat Res 2010; 468: 1773–80.

40. Hoffmann J, Preston G, Whaley J, Khalil JG: Vertebral augmentation in spine surgery. J Am Acad Orthop Surg 2023; 31: 477–89.