Skip to main content

Table 3 Logistic regression predicting fibrotic HP

From: The new useful high-resolution computed tomography finding for diagnosing fibrotic hypersensitivity pneumonitis: “hexagonal pattern”: a single-center retrospective study

 

Odds ratio

95% CI

P-value

Univariate logistic regression

   

Distribution (zonal)

   

 Upper/middle

0.00

0.00–Inf

0.991

 Lower

0.44

0.10–1.96

0.282

 Diffuse

Reference

  

Distribution (axial)

   

 Peripheral

0.05

0.01–0.51

0.011

 Peripheral with subpleural sparing

0.40

0.03–6.18

0.512

 Peribronchovascular

0.40

0.02–10.00

0.577

 Diffuse

Reference

  

Extensive GGO

7.25

2.13–24.65

0.002

Centrilobular nodules

5.19

1.36–19.82

0.016

Mosaic attenuation

4.17

0.90–19.28

0.068

Air trapping

1.65

0.59–4.63

0.341

Honeycombing

0.36

0.09–1.43

0.147

Consolidation

0.27

0.03–2.31

0.229

Emphysema

0.56

0.20–1.62

0.285

PPFE

0.17

0.02–1.44

0.105

Cyst

0.42

0.15–1.16

0.094

Hexagonal pattern

16.00

4.66–54.91

 < 0.001

Multivariate logistic regression

   

Extensive GGO

4.86

1.20–19.70

0.027

Centrilobular nodules

1.45

0.30–7.08

0.647

Hexagonal pattern

11.00

2.70–44.70

0.001

  1. HP, hypersensitivity pneumonitis; GGO, ground-glass opacity; PPFE, pleuroparenchymal fibroelastosis