Skip to main content
. 2023 Apr 7;15(5):2485–2496. doi: 10.21037/jtd-22-1546

Table 5. Univariable and multivariable logistic regression analysis to identify clinical and radiologic features of lesions associated with post-procedural complication.

Characteristics Univariable logistic regression analysis Multivariable logistic regression analysis (with lesion size) Multivariable logistic regression analysis (with PCL)
OR 95% CI P OR 95% CI P OR 95% CI P
Age 0.98 0.95, 1.02 0.356
Sex 0.144
   Female Reference
   Male 0.55 0.25, 1.27
Number of transthoracic passage 0.66 0.37, 1.12 0.140
Operator’s experience 0.67 0.52, 0.82 <0.001* 0.64 0.49, 0.80 <0.001* 0.66 0.52, 0.82 <0.001*
Lesion size 0.7 0.57, 0.84 <0.001* 0.68 0.54, 0.83 <0.001*
PCL 0.7 0.55, 0.85 <0.001* 0.68 0.52, 0.84 0.001*
Final diagnosis 0.419
   Benign Reference
   Malignancy 0.7 0.31, 1.75
Location
   L Reference
   U 0.29 0.07, 0.85 0.047* Stepwise eliminated Stepwise eliminated
   M 1.66 0.47, 4.66 0.367
Peritumoral emphysema 0.34 0.10, 0.90 0.049* Stepwise eliminated Stepwise eliminated
Peritumoral honeycombing 1.01 0.06, 5.15 0.993
Cavitary change 1.54 0.50, 3.91 0.397
Air-bronchogram 10.84 3.55, 30.08 <0.001* 14.36 4.18, 48.53 <0.001* 17.35 4.99, 55.95 <0.001*

, variables with P<0.1 on univariable analysis were included in multivariable analysis. *, statistically significant. OR, odds ratio; CI, confidence interval; PCL, pleural contact length.