Skip to main content
. 2022 Jul 4;12:61. doi: 10.1186/s13613-022-01037-1

Table 3.

Imbalance of patient characteristics before and after propensity weighting in the assessment of in-ICU mortality

Variables Before weighting After weighting
No exposure to NE
(n = 3448)
Exposure to NE
(n = 1605)
SMD No exposure to NE
(n = 3502)
Exposure to NE
(n = 1582)
SMD
Age, years 68 [60–76] 70 [61–77] 0.120 69 [61–76] 69 [61–76] 0.002
Male gender 2407 (67) 1080 (67) 0.054 2456 (70) 2456 (70) 0.036
BMI, kg m−2 27.48 [24.44–30.72] 26.83 [23.83–30.42] 0.017 27.50 [24.39–30.73] 27.50 [24.39–30.73] 0.005
Hypertension 1925 (56) 890 (56) 0.008 1912 (55) 1919 (55) 0.018
Coronary disease 385 (11.2) 213 (13) 0.064 380 (11) 381 (11) 0.078
Diabetes 690 (20) 294 (18) 0.043 671 (19) 671 (19) 0.003
Dyslipidemia 274 (8) 107 (7) 0.049 259 (7) 258.7 (7) 0.007
Chronic kidney disease 126 (4) 98 (6) 0.114 157 (5) 157 (5) 0.022
Peripheral vascular disease 188 (6) 76 (5) 0.033 187 (5) 187 (5) 0.018

Hemoglobin,

g dl−1

11.52 ± 1.54 11.23 ± 1.64 0.181 11.3 ± 1.56 11.4 ± 1.43 0.004

Platelet,

103 mm−3

157 ± 58 150 ± 66 0.111 146 ± 34 144 ± 35 0.007

Creatinine,

µ mol l−1

77 [64–94] 82 [67–105] 0.187 79 [65–99] 79 [65–99] 0.012
CPB time, min 84 [56–118] 100 [67–136] 0.336 88 [58–125] 88 [58–125] 0.036
Aortic clamp time, min 57 [35–83] 64 [42–94] 0.111 59 [37–87] 59 [37–87] 0.005
Surgery type 0.185
 CABG 962 (28) 468 (29) 1002 (29) 1002 (29) 0.010
 Valve surgery 1254 (36) 620 (39) 1284 (37) 1284 (37)
 Combined surgery 336 (10) 212 (13) 369 (11) 369 (11)
 Others 896 (26) 305 (19) 847 (24) 847 (24)
SAPS II 34 [28–40] 40 [33–48] 0.595 35 [30–43] 35 [30–43] 0.030
Inotropes 0 (0) 128 (8) 0.306 0 (0) 74 (5) 0.223

An absolute MSD < 10% was considered to support the assumption of a balance between the groups. SMD: standardized mean differences. Data are presented as medians [interquartile ranges] or as numbers (percentages). NE: norepinephrine, CABG: coronary bypass graft, CPB: coronary bypass, SAPS II: Simplified Acute Physiology Score. Male gender and coronary disease were not selected for the propensity weighting as the P value for their association with acute kidney injury was over 20%