Skip to main content
. 2020 Mar 26;125(1):e130–e139. doi: 10.1016/j.bja.2020.02.018

Table 3.

Subgroup analyses in cases with available intraoperative arterial blood gas analyses (n=5795). Data are expressed as mean (standard deviation), or n (%). Results are reported as odds ratios (ORs). ∗If multiple blood gas analyses were performed during one case, the mean of all measurement was considered for a given case. Logistic regression analysis was adjusted for Paco2 and Pao2/Fio2 ratio in addition to age, sex, BMI, ASA status, emergency surgery, duration of surgery, work relative value units, history of smoking, chronic obstructive pulmonary disease, or heart failure, Charlson comorbidity index, high Score for Prediction of Postoperative Respiratory Complications (SPORC), packed red blood cell units, short and long-acting opioid doses, neuromuscular blocking agent dose, intraoperative hypotension, airway device, median positive end-expiratory pressure (PEEP), median peak inspiratory pressure (PIP), tidal volume, and non-protective ventilation.

Group 1 (n=1723) Group 2 (n=1501) Group 3 (n=1682) Group 4 (n=889)
Arterial blood gas results ∗
 pH 7.40 (0.06) 7.39 (0.06) 7.39 (0.06) 7.38 (0.08)
 Partial pressure of oxygen (Pao2), mm Hg 210.6 (84.8) 210.2 (83.3) 205.3 (80.2) 201.0 (79.0)
 Partial pressure of carbon dioxide (Paco2), mm Hg 40.48 (5.30) 40.50 (5.64) 40.44 (6.53) 41.38 (8.25)
 Pao2/Fio2 ratio 390.9 (146.7) 391.6 (144.7) 378.5 (147.7) 364.3 (141.8)
 Paco2−etCO2 gradient, mm Hg 7.81 (5.09) 7.49 (5.46) 6.84 (6.02) 6.78 (6.73)
Postoperative respiratory complications 365 (21.18) 357 (23.78) 463 (27.53) 325 (36.56)
 Unadjusted OR (95% CI) Reference 1.16 (0.98, 1.37)
P=0.077
1.41 (1.21, 1.65)
P<0.001
2.14 (1.79, 2.56)
P<0.001
 Adjusted OR (95% CI) Reference 1.00 (0.83, 1.21)
P=0.997
1.17 (0.97, 1.42)
P=0.106
1.51 (1.19, 1.91)
P=0.001