Establishment of a risk prediction model for prolonged mechanical ventilation after lung transplantation: a retrospective cohort study

Table 4 Univariate and multivariate logistic regression analyses testing effects of perioperatively assessable variables on predicting PMV in 141 patients after LuTx

Characteristic	Univariable			Multivariable
Characteristic	OR	95% CI	p value	OR	95% CI	p value
Age, y	0.965	0.927–0.991	0.213	NA	NA	NA
BMI	1.213	1.219–1.444	0.030	1.425	1.323–1.767	0.032
Gender	0.957	0.927–0.994	0.445	NA	NA	NA
Smoking history Nonsmoker vs Smoker	1.204	1.162–1.231	0.145	NA	NA	NA
Pulmonary hypertension Normal vs High	2.706	1.278–5.845	0.011	1.894	1.243–3.001	0.034
Primary diagnose as IPF	3.001	1.643–6.153	0.002	1.788	1.245–2.634	0.038
PGDatT₀	2.231	1.601–3.110	< 0.001	1.557	1.331–1.899	0.011
PGDatT₂₄	1.599	1.082–2.361	0.017	0.205	0.044–0.958	0.054
PGDatT₄₈	1.628	1.080–2.454	0.024	0.626	0.099–3.956	0.619
PGDatT₇₂	1.510	1.518–2.262	0.041	2.007	0.427–9.430	0.378
CIT, h	2.068	1.537–2.783	< 0.001	1.777	1.065–2.889	0.012
Ventilation status CV vs AV	2.323	1.621–3.011	0.003	2.007	1.117–3.444	0.138
PIP	1.362	1.203–1.542	< 0.001	1.961	1.211–2.747	< 0.001
Cdyn	1.645	1.200–1.962	0.001	1.266	1.121–1.473	0.001
P/F ratio	0.993	0.989–0.998	0.002	0.991	0.981–0.996	0.015

Bold values indicate statistically significant (P < 0.05)
BMI body mass index; IPF idiopathic pulmonary fibrosis; CIT cold ischemia time; CVcontrolled ventilation; AVassisted ventilation; TV tidal volume; PIP peak airway pressure; PEEPpositive end expiratory pressure; Cdyn, pulmonary dynamic compliance. Confidence interval; P/F ratio, PaO₂/FiO₂ ratio

ISSN: 1471-2466