Table 4.

Univariable and parsimonious multivariable logistic regression analyses for postoperative wound complications

Variable	Univariable OR	95% CI	p-value	Model 1 adjusted OR	95% CI	p-value	Model 2 adjusted OR	95% CI	p-value
Age, per 10 years	1.74	1.10–2.75	0.018	—	—	—	—	—	—
BMI, per 1 kg/m²	1.18	1.05–1.33	0.006	—	—	—	—	—	—
Male sex	1.37	0.52–3.60	0.521	—	—	—	—	—	—
Diabetes mellitus	3.18	1.18–8.58	0.022	2.76	1.01–7.52	0.047	2.89	1.01–8.29	0.048
Current smoking	1.39	0.53–3.63	0.500	—	—	—	—	—	—
ASA III–IV	2.94	1.10–7.85	0.031	—	—	—	—	—	—
Emergency surgery	2.98	1.07–8.28	0.037	—	—	—	—	—	—
Malignant disease	1.90	0.72–5.04	0.194	—	—	—	—	—	—
Open surgical approach	3.15	1.14–8.69	0.027	—	—	—	—	—	—
Major abdominal surgery	3.16	0.99–10.10	0.052	—	—	—	—	—	—
Operative time, per 30 min	1.96	1.35–2.84	0.001	1.84	1.25–2.71	0.002	1.67	1.08–2.58	0.021
Contamination class III–IV	3.12	1.07–9.12	0.037	—	—	—	—	—	—
ΔWINA, per 100 units	2.17	1.55–3.04	0.001	—	—	—	2.12	1.44–3.13	0.001

Model 1 included diabetes mellitus and operative time. Model 2 included diabetes mellitus, operative time, and ΔWINA. Variables included in the multivariable models were prespecified according to clinical relevance, while maintaining a parsimonious events-per-variable ratio. WINA and its individual components were not entered simultaneously into the same model. Multicollinearity was assessed using variance inflation factors, which were below 2.0 for all variables included in Model 2. Abbreviations: ASA American Society of Anesthesiologists, BMI body mass index, CI confidence interval, OR odds ratio, WINA Wound Inflammation–Nutrition Index