Skip to main content

Table 1 Baseline table of patients with and without cement leakage

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

Type

Yes (N = 150)

NO (N = 224)

P

Type

Yes (N = 150)

NO (N = 224)

P

Age

   

Lumbar vertebra bone density examination

   

Mean(SD)

69.47 (9.47)

71.55 (9.89)

0.42

Normal (n = 21)

11 (52.4%)

10 (47.9%)

0.006

BMI

   

Osteopenia (n = 71)

39 (54.9%)

32 (45.1%)

 

Mean(SD)

22.57 (2.68)

22.32 (2.81)

 

Osteoporosis (n = 282)

100 (35.5%)

182 (64.5%)

 

Bone cement injection volume

  

0.374

Surgical vertebral segment

   

Mean(SD)

4.57 (0.89)

4.61 (0.86)

 

Thoracic vertebra (n = 218)

92 (42.2%)

126 (57.8%)

0.601

Sex

   

Lumbar vertebra (n = 147)

58 (39.5%)

89 (60.5%)

 

Male

37

57

0.657

Puncture path

   

Female

113

167

 

Conventional puncture route (n = 299)

125 (41.8%)

174 (58.2%)

0.181

Associated with old vertebral fractures

   

Lateral puncture path (n = 75)

25 (33.3%)

50 (66.7%)

 

Yes (n = 138)

46 (33.3%)

92 (66.7%)

0.046

VAS score

   

No (n = 235)

103 (43.8%)

132 (56.2%)

 

No

0 (

0

0.903

Previous PVP operation history

   

1–3

0

0

 

Yes (n = 68)

20 (29.4%)

48 (70.6%)

0.047

4–6 (n = 156)

62 (39.7%)

94 (60.3%)

 

No (n = 306)

130 (42.5%)

176 (57.5%)

 

7–10 (n = 218)

88 (40.4%)

130 (59.6%)

 

Combined with thoracolumbar kyphosis

   

VAS score 24 h after surgery

   

Yes (n = 85)

21(24.7%)

64 (75.3%)

0.01

No

0

0

0.243

No (n = 289)

129 (44.6%)

160 (55.4%)

 

1–3 (n = 371)

149 (40.2%)

222 (59.8%)

 

Heart or lung disease

   

4–6

0

2

 

Yes (n = 58)

19 (32.8%)

39 (67.2%)

0.214

7–10

1 (0.7%)

0 (0%)

 

No (n = 316)

131 (41.5%)

185 (58.4%)

 

VAS score at discharge

   

Classification of fracture

   

No (n = 48)

23 (47.9%)

25 (52.1%)

0.237

Simple compression fracture (n = 355)

140 (39.4%)

215 (60.6%)

0.516

1–3 (n = 326)

127 (39%)

199 (61%)

 

Cleavage fracture (n = 15)

8 (53.3%)

7 (46.7%)

 

4–6

0

0

 

Pathological fracture (n = 4)

2 (50%)

2 (50%)

 

7–10

0

0

 

Degree of vertebral compression

   

Pre and post penetration of bone cement

   

 < 25% (n = 218)

78 (35.8%)

140 (64.2%)

0.120

Bone cement did not cross the midline (n = 22)

12 (54.5%)

10 (45.5%)

0.186

26–40% (n = 85)

38 (44.7%)

47 (55.3%)

 

The bone cement did not cross the midline and did not cross the medial projection of the pedicle (n = 100)

44 (44%)

56 (56%)

 

 > 40% (n = 71)

34 (47.9%)

37 (52.1%)

 

The bone cement crosses the midline beyond the medial projection of the pedicle (n = 252)

94 (37.3%)

158 (62.7%)

 

Preoperative MRI showed signal changes in the injured vertebrae

   

Upper and lower penetration of bone cement

   

Middle and upper vertebral body (n = 116)

48 (41.4%)

68 (58.6%)

0.935

The bone cement is not connected to the upper and lower end plates (n = 321)

130 (40.5%)

191 (59.5%)

0.519

Middle and lower vertebral body (n = 17)

7 (41.2%)

10 (58.8%)

 

Bone cement is not connected to upper and lower end plates (n = 3)

1 (33.3%)

2 (66.7%)

 

The upper and lower parts of the vertebral body (n = 241)

95 (39.4%)

146 (60.6%)

 

The bone cement was not attached to the lower endplate (n = 35)

11 (31.4%)

24 (68.6%)

 

History of trauma

   

The bone cement was not attached to the end plate (n = 15)

8 (53.3%)

7 (46.7%)

 

Yes (n = 171)

81 (47.4%)

90 (52.6%)

0.009

    

No (n = 203)

69 (34%)

134 (66%)

     
  1. VAS Visual Analogue Score, P P-value