Postoperative pulmonary complications with adjuvant regional anesthesia versus general anesthesia alone: a sub-analysis of the Perioperative Research Network study

Table 4 Hierarchical multivariable logistic regression models to investigate the association of adjuvant regional anesthesia (RA + GA) versus general anesthesia only (GA) with the occurrence of at least one PPC

Variable	Crude Models (adjusted for site)			Adjusted Model (adjusted for site and other covariates)
Variable	OR	Lower 95%CI	Upper 95%CI	OR	Lower 95%CI	Upper 95%CI	P-value
RA + GA	1.827	1.270	2.629	1.368	0.834	2.246	0.1646
Neurological disease	0.881	0.603	1.287	1.140	0.683	1.902	0.5408
Emergency	2.160	1.069	4.366	3.040	1.091	8.469	0.0394
Abdominal/Pelvic Surgery	2.515	1.837	3.444	2.956	1.854	4.712	0.0019
Non-Depolarizing neuromuscular blockade	1.608	0.997	2.594	1.167	0.526	2.588	0.6398
Neuromuscular blockade monitoring	1.293	0.869	1.924	0.794	0.462	1.362	0.3213
Propofol	0.829	0.548	1.255	1.055	0.581	1.915	0.8168
Colloid	3.291	2.194	4.937	1.999	1.137	3.513	0.0252
Hydromorphone	0.966	0.656	1.423	0.957	0.554	1.651	0.8332
Fentanyl	1.194	0.648	2.198	1.392	0.609	3.183	0.3509
Remifentanil	0.764	0.489	1.193	1.145	0.652	2.010	0.5638
Age, years	1.027	1.018	1.037	1.031	1.018	1.044	< 0.0001
Log of Estimated Blood Loss, mL	1.322	1.217	1.435	1.180	1.068	1.303	0.0012
Preoperative Peripheral Saturation of Oxyhemoglobin (SpO₂), %	0.893	0.840	0.949	0.836	0.774	0.903	< 0.0001
Crystalloid	1.038	1.005	1.073	1.017	0.975	1.060	0.4405
Neostigmine Dose, mcg/kg	1.009	1.003	1.014	1.000	0.992	1.007	0.9524

Regression modeling used site as a random effect, which allows for clustering of patients within hospitals. Crude odds ratios adjusted for site were examined, and important covariates (identified as those with clinical relevance and p-values less than 0.05 in the bivariable analyses) were added to include in the final adjusted model

ISSN: 1471-2253