Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Dec 21:17:1304248.
doi: 10.3389/fninf.2023.1304248. eCollection 2023.

Establishing a nomogram to predict refracture after percutaneous kyphoplasty by logistic regression

Aiqi Zhang  1 , Hongye Fu  2 , Junjie Wang  3 , Zhe Chen  3 , Jiajun Fan  3
Affiliations

Establishing a nomogram to predict refracture after percutaneous kyphoplasty by logistic regression

Aiqi Zhang et al. Front Neuroinform. .

Abstract

Introduction: Several studies have examined the risk factors for post-percutaneous kyphoplasty (PKP) refractures and developed many clinical prognostic models. However, no prior research exists using the Random Forest (RF) model, a favored tool for model development, to predict the occurrence of new vertebral compression fractures (NVCFs). Therefore, this study aimed to investigate the risk factors for the occurrence of post-PKP fractures, compare the predictive performance of logistic regression and RF models in forecasting post-PKP fractures, and visualize the logistic regression model.

Methods: We collected clinical data from 349 patients who underwent PKP treatment at our institution from January 2018 to December 2021. Lasso regression was employed to select risk factors associated with the occurrence of NVCFs. Subsequently, logistic regression and RF models were established, and their predictive capabilities were compared. Finally, a nomogram was created.

Results: The variables selected using Lasso regression, including bone density, cement distribution, vertebral fracture location, preoperative vertebral height, and vertebral height restoration rate, were included in both the logistic regression and RF models. The area under the curves of the logistic regression and RF models were 0.868 and 0.786, respectively, in the training set and 0.786 and 0.599, respectively, in the validation set. Furthermore, the calibration curve of the logistic regression model also outperformed that of the RF model.

Conclusion: The logistic regression model provided better predictive capabilities for identifying patients at risk for post-PKP vertebral fractures than the RF model.

Keywords: anterior vertebral height; logistic regression; percutaneous kyphoplasty; random forest model; vertebral compression fractures.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

Figure 1
Figure 1
(A) Bone cement in contact with the upper edge of the vertebral endplate. (B) Bone cement in contact with the lower edge of the vertebral endplate. (C) Bone cement in contact with both the upper and lower edges of the vertebral endplate. (D) Bone cement does not contact either the upper or lower edges of the vertebral endplate.
Figure 2
Figure 2
Variable selection using Lasso regression. (A) Coefficient fluctuations for variables. (B) Lasso regression λ determined through tenfold cross-validation.
Figure 3
Figure 3
(A) Random Forest error rate. (B) Out-of-bag variable importance ranking. AVH, anterior vertebral height; AVHRR, anterior vertebral height recovery ratio; BMD, bone mineral density.
Figure 4
Figure 4
Area under the curves (AUCs) for the logistic regression model in the (A) training and (C) validation sets. AUCs for the Random Forest model in the (B) training and (D) validation sets.
Figure 5
Figure 5
Calibration curves for the Logistic regression model in the (A) training and (C) validation sets. Calibration curves for the Random Forest model in the (B) training and (D) validation sets.
Figure 6
Figure 6
Nomogram for predicting post-PKP refractures. AVH, anterior vertebral height; AVHRR, anterior vertebral height recovery ratio; BMD, bone mineral density; PKP, percutaneous kyphoplasty.
Figure 7
Figure 7
Distribution of vertebral fracture sites. BMD, bone mineral density.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Bian F., Bian G., Zhao L., Huang S., Fang J., An Y. (2022). Risk factors for recollapse of new vertebral compression fractures after percutaneous kyphoplasty in geriatric patients: establishment of a nomogram. BMC Musculoskelet. Disord. 23:458. doi: 10.1186/s12891-022-05409-3, PMID: - DOI - PMC - PubMed
    1. Black D. M., Bauer D. C., Vittinghoff E., Lui L. Y., Grauer A., Marin F., et al. . (2020). Treatment-related changes in bone mineral density as a surrogate biomarker for fracture risk reduction: meta-regression analyses of individual patient data from multiple randomised controlled trials. Lancet Diabetes Endocrinol. 8, 672–682. doi: 10.1016/S2213-8587(20)30159-5, PMID: - DOI - PubMed
    1. Bouxsein M. L., Eastell R., Lui L. Y., Wu L. A., de Papp A. E., Grauer A., et al. . (2019). Change in bone density and reduction in fracture risk: a meta-regression of published trials. J. Bone Miner. Res. 34, 632–642. doi: 10.1002/jbmr.3641 - DOI - PubMed
    1. Cao Z., Wang G., Hui W., Liu B., Liu Z., Sun J. (2020). Percutaneous kyphoplasty for osteoporotic vertebral compression fractures improves spino-pelvic alignment and global sagittal balance maximally in the thoracolumbar region. PLoS One 15:e0228341. doi: 10.1371/journal.pone.0228341, PMID: - DOI - PMC - PubMed
    1. Chen X., Guo W., Li Q., Ou Z., Lao Z., Liu Y., et al. . (2018). Is unilateral percutaneous kyphoplasty superior to bilateral percutaneous kyphoplasty for osteoporotic vertebral compression fractures? Evidence from a systematic review of discordant meta-analyses. Pain Physician 21, 327–336. doi: 10.36076/ppj.2018.4.327, PMID: - DOI - PubMed

LinkOut - more resources