Table 1.
Patient characteristics on admission, exposures during the ICU stay and outcomes of the entire population a
| Characteristics on admission | Total (N =850) |
|---|---|
| Age (yr) | 59.8 (17.3) |
| Male sex | 519 (61.1) |
| Pre-ICU stay (days) | 4.6 (12.2) |
| Prior antibiotic (≤1 mo) | 258 (30.4) |
| APACHE II score | 20 (6.5) |
| SOFA score | 6.4 (3.6) |
| Shock on admission | 151 (17.8) |
| Reason for admission | |
| Infection | 486 (57.2) |
| CNS disease | 99 (11.6) |
| Postsurgical | 80 (9.4) |
| Cardiovascular disease | 66 (7.8) |
| Respiratory disease | 28 (3.3) |
| Others | 91 (10.7) |
| Underlying diseases | |
| Diabetes mellitus | 157 (18.5) |
| Haematological malignancy | 114 (13.4) |
| Solid malignancy | 85 (10) |
| COPD | 138 (16.2) |
| Othersb | 193 (19.2) |
| Prior corticosteroids (≤1 mo) | 156 (18.4) |
| Immunosuppressive therapy | 90 (10.6) |
| Exposures during ICU stay | |
| Central venous catheter | 833 (98) |
| Bladder catheter | 800 (94.1) |
| Nasogastric tube | 573 (67.4) |
| Enteral nutrition | 253 (29.8) |
| Parenteral nutrition | 177 (20.8) |
| Orotracheal intubation | 511 (60.1) |
| Tracheostomy | 158 (18.6) |
| Endoscopy | 121 (34.2) |
| Surgery | 189 (22.2) |
| Renal replacement therapies | 77 (9.1) |
| Packed red blood cell transfusion | 291 (34.2) |
| Any antibiotic | 746 (87.8) |
| Any non-antipseudomonal antibiotic | 593 (69.8) |
| Any antipseudomonal antibiotic: | 576 (67.8%) |
| Carbapenem | 281 (33.1) |
| Quinolone | 306 (36) |
| Ceftazidime | 103 (12.1) |
| Piperacillin-tazobactam | 180 (21.2) |
| Amikacin | 52 (6.1) |
| Outcomes | |
| Length of stay (days) | 9.5 (10.3) |
| In-ICU mortality | 117 (13.8) |
| In-hospital mortality | 201 (23.6) |
aAPACHE II, Acute Physiology and Chronic Health Evaluation II; CNS, Central nervous system; COPD, Chronic obstructive pulmonary disease; ICU, Intensive care unit; SOFA, Sequential Organ Failure Assessment. bOthers include patients with HIV infection, hepatic cirrhosis, renal failure and heart failure. Categorical variables are expressed as number of patients (%) and continuous variables as mean (standard deviation).