Skip to main content

Table 4 Univariate logistic regression results

From: Application of machine learning in prediction of bone cement leakage during single-level thoracolumbar percutaneous vertebroplasty

Type

OR(95% CI)

P-value

Age

0.9778 (0.9566–0.9991)

0.0425

Sex

1.0424 (0.6485–1.6895)

0.8647

BMI

1.0348 (0.9599–1.1161)

0.3730

Associated with old vertebral fractures

0.6346 (0.4078–0.9794)

0.0416

Previous PVP operation history

0.5641 (0.3136–0.9833)

0.0485

Combined with thoracolumbar kyphosis

0.4070 (0.2316–0.6916)

0.0012

Heart or lung disease

0.6880 (0.3738–1.2294)

0.2159

Classification. of fracture

1.4633 (0.7044–3.1187)

0.3045

Degree of vertebral compression

1.3058 (1.0053–1.6980)

0.0456

Preoperative MRI showed signal changes in the injured vertebrae

0.9526 (0.7623–1.1923)

0.6698

Time of spinal fracture

0.9309 (0.7033–1.2255)

0.6125

History of trauma

1.7478 (1.1526–2.6598)

0.0088

Lumbar vertebra bone density examination

0.5873 (0.4051–0.8433)

0.0043

Surgical vertebral segment

0.8106 (0.5305–1.2337)

0.3288

Puncture path

0.6960 (0.4038–1.1753)

0.1820

VAS score

1.0263 (0.6750–1.5637)

0.9035

VAS score 24 h after surgery

0.8631 (0.1856–2.8007)

0.8107

VAS score at discharge

0.6937 (0.3770–1.2809)

0.2388

Bone cement injection volume

0.9472 (0.7443–1.2015)

0.6556

Pre and post penetration of bone cement

0.7280 (0.5147–1.0275)

0.0709

Upper and lower penetration of bone cement

0.9944(0.7637–1.2836)

0.9658