TABLE 3

Factors associated with acquisition of CR-GNB among patients during ICU stayc

CharacteristicNo. (%) of patients in the following groups with the indicated characteristic:Univariate analysis P valueMultivariate logistic regression analysis
All patients (n = 206)CR-GNB acquisition (n = 52)Non-CR-GNB acquisition (n = 154)P valueaaOR (95% CI)
Male gender113 (54.9)30 (57.7)83 (53.9)0.540
Age >65 yr84 (40.8)23 (44.2)61 (39.6)0.486
Mechanical devices
    Central venous catheter29 (14.1)8 (15.4)21 (13.6)0.690
    Indwelling urinary catheter168 (81.6)43 (82.7)125 (81.2)0.485
    Enteral feeding tube153 (74.3)46 (88.5)107 (69.5)0.0020.0085.386 (1.563–18.564)
    Endotracheal tube159 (77.2)46 (88.5)113 (73.4)0.0070.7290.763 (0.165–3.527)
Antibiotic usage during ICU stay
    First- and second-generation cephalosporins35 (17.0)6 (11.5)29 (18.8)0.258
    Third-generation cephalosporins121 (58.7)37 (71.2)84 (54.5)0.0170.0322.293 (1.074–4.896)
    Carbapenems42 (20.4)16 (30.8)26 (16.9)0.0230.0452.199 (1.019–4.743)
    Penicillinsb4 (1.9)2 (3.8)2 (1.3)0.254
    β-Lactam–β-lactamase inhibitors36 (17.5)10 (19.2)26 (16.9)0.629
    Aminoglycosidesb2 (1.0)1 (1.9)1 (0.6)0.433
    Fluoroquinolones18 (8.7)7 (13.5)11 (7.1)0.156
    Macrolides26 (12.6)8 (15.4)18 (11.7)0.437
    Colistinb7 (3.4)2 (3.8)5 (3.2)0.682
    Vancomycin11 (5.3)4 (7.7)7 (4.5)0.470
    Clindamycin21 (10.2)4 (7.7)17 (11.0)0.531
    Metronidazole12 (5.8)5 (9.6)7 (4.5)0.174
    Fosfomycinb2 (1.0)1 (1.9)1 (0.6)0.433
Underlying conditions
    Cardiovascular disease97 (47.1)23 (44.2)74 (48.1)0.870
    Respiratory disease25 (12.1)7 (13.5)18 (11.7)0.635
    Renal disease24 (11.7)8 (15.4)16 (10.4)0.264
    Diabetes52 (25.2)14 (26.9)38 (24.7)0.592
    Liver disease14 (6.8)2 (3.8)12 (7.8)0.524
Length of ICU stay (day)
    <8 days129 (62.6)24 (46.2)105 (68.2)0.0080.4980.730 (0.293–1.815)
    8–14 days34 (16.5)10 (19.2)24 (15.6)0.449
    15–21 days22 (10.7)9 (17.3)13 (8.4)0.0550.1312.144 (0.797–5.767)
    >21 days8 (3.9)4 (7.7)4 (2.6)0.1000.1493.227 (0.657–15.860)
30-day mortality16 (7.8)5 (9.6)11 (7.1)0.553
  • a A P value of <0.05 was considered statistically significant.

  • b The rate of use was too low to give an accurate P value.

  • c Data are for a total of 206 patients.