Skip to main content

Table 2 Univariate analysis between the different types of pleural effusion

From: Relations between serum and pleural fluid biomarkers: a new look of an old concept

Variable

Malignant (n = 94)

Tuberculous (n = 31)

Para-pneumonic (n = 25)

P1

P2

P3

Age

59.97 (9.25)

38.03 (10.8)

49.16 (13.18)

< 0.001

< 0.001

< 0.001

Pleural fluid glucose

90 (75–109.25)

87 (60–100)

70 (52.5–90.5)

0.5

0.01

0.1

Pleural fluid protein

4.67 (0.9)

4.84 (1.2)

4.96 (0.9)

0.41*

Serum protein

6.45 (5.9–7)

6.5 (5.8–7.1)

6.5 (5.8–7)

0.99*

Pleural fluid/serum protein

0.71 (0.61–0.82)

0.77 (0.64–0.9)

0.76 (0.61–0.88)

0.34*

Pleural fluid LDH

555 (355.75–1011)

600 (299–844)

469 (321–1046)

0.89*

Serum LDH

342.5 (233.75–420)

220 (189–276)

220 (190–259)

0.001

0.001

0.9

Pleural fluid LDH/Serum LDH

1.75 (1.07–3.34)

2.59 (1.49–3.63)

2.13 (1.25–4.95)

0.06

Pleural fluid ADA

19 (11–25)

38 (29–42)

24 (20–30)

< 0.001

0.001

< 0.001

Cancer ratio

18.03 (10.87–29.73)

6.1 (4.64–9.17)

8.7 (6.79–11.79)

< 0.001

< 0.001

< 0.001

Pleural fluid lymphocyte count

0.5 (0.4–0.6)

0.88 (0.8–0.95)

0.2 (0.15–0.3)

< 0.001

< 0.001

< 0.001

Cancer ratio plus

40.13 (24.68–60.88)

8.06 (5.22–11.14)

47.5 (32.15–68.78)

< 0.001

0.5

< 0.001

Serum LDH/pleural fluid lymphocyte count

700 (506.8–950.4)

271.11 (216.49–332.5)

1200 (762.12–1754.54)

< 0.001

< 0.001

< 0.001

  1. Normal distributed data represented as mean (SD) and non-normally distributed data as median (IQR); the test of significant is one-way ANOVA with multiple comparison using Tukey’s methods or Dunnet’s methods
  2. P1 malignant vs tuberculous, P2 malignant vs para-pneumonic, P3 tuberculous vs para-pneumonic, P < 0.05 considered significant
  3. *Not applicable post hoc test, because P > 0.05