Table 3.
Study Outcomes and Adverse Events, According to Study Group, in the Per-Protocol Population.
| Variable | Lower Target (N = 349) | Higher Target (N = 349) | P Value* |
|---|---|---|---|
| No. of ICU-free days through day 28 | 0.86 | ||
|
| |||
| Median | 20.0 | 19.4 | |
| IQR | 1.0–24.2 | 7.1–23.9 | |
|
| |||
| Assigned zero ICU-free days — no. (%) | 87 (24.9) | 70 (20.1) | 0.14 |
| Died by day 28 | 47 (13.5) | 32 (9.2) | |
| Did not meet ICU discharge criteria by day 28 | 33 (9.5) | 37 (10.6) | |
| Transferred to an ICU in a nonparticipating institution by day 28 | 7 (2.0) | 1 (0.3) | |
|
| |||
| No. of ventilator-free days through day 28 | 0.84 | ||
| Median | 21.8 | 20.9 | |
| IQR | 8.4–25.0 | 11.9–24.4 | |
|
| |||
| No. of hospital-free days through day 28 | 0.60 | ||
| Median | 8 | 6 | |
| IQR | 0–17 | 0–16 | |
|
| |||
| Hospital mortality — no. (%) | |||
| At day 28 | 47 (13.5) | 32 (9.2) | 0.09 |
| At day 90 | 52 (14.9) | 40 (11.5) | 0.22 |
|
| |||
| Maximum PELOD score† | 0.38 | ||
| Median | 13 | 13 | |
| IQR | 11–23 | 11–22 | |
|
| |||
| Maximum daily vasoactive-inotrope score33 | 0.55 | ||
| Median | 5 | 4 | |
| IQR | 0–15 | 0–13 | |
|
| |||
| New seizure — no. (%) | 5 (1.4) | 10 (2.9) | 0.20 |
|
| |||
| Cardiopulmonary resuscitation — no. (%) | 10 (2.9) | 11 (3.2) | 0.80 |
|
| |||
| New ECMO initiated after randomization — no. (%) | 8 (2.3) | 5 (1.4) | 0.37 |
|
| |||
| Glucocorticoid therapy after randomization — no. (%) | 264 (75.6) | 269 (77.1) | 0.56 |
|
| |||
| Renal-replacement therapy — no. (%) | 41 (11.7) | 31 (8.9) | 0.19 |
|
| |||
| Red-cell transfusion — no. (%) | 158 (45.3) | 150 (43.0) | 0.62 |
|
| |||
| Empirical or treatment antibiotic agent — no. (%) | 326 (93.4) | 338 (96.8) | 0.04 |
|
| |||
| Health care–associated infection — no. (%) | 12 (3.4) | 4 (1.1) | 0.04 |
|
| |||
| Catheter-associated bloodstream infection | 0.03‡ | ||
| Events — no./no. of central-venous-catheter–days | 5/2577 | 0/2784 | |
| Rate per 1000 central-venous-catheter–days | 1.94 | 0 | |
|
| |||
| Catheter-associated urinary tract infection | 1.0‡ | ||
| Events — no./no. of bladder-catheter–days | 5/2287 | 4/2230 | |
| Rate per 1000 bladder-catheter–days | 2.19 | 1.79 | |
|
| |||
| Ventilator-associated pneumonia | 0.11‡ | ||
| Events — no./no. of ventilator-days | 3/3182 | 0/3371 | |
| Rate per 1000 ventilator-days | 0.94 | 0 | |
|
| |||
| All infections with positive cultures — no. (%)§ | 29 (8.3) | 34 (9.7) | 0.52 |
P values for the comparison between treatment groups were calculated with the use of proportional-hazards, linear, or logistic regression with adjustment for age group and PRISM III-12 score, as appropriate, except where noted.
Pediatric Logistic Organ Dysfunction (PELOD) scores range from 0 to 71, with higher scores indicating more severe multiple organ dysfunction syndrome.
Owing to zero counts or low event frequency, the P value was calculated with the use of exact Poisson regression.
All infections include health care–associated infection as well as upper respiratory tract infection (e.g. tracheitis), lower respiratory tract infection not associated with ventilator (e.g. pneumonia), bloodstream infection not associated with catheter, colitis, urinary tract infection not associated with a urinary catheter, other wound infection, abdominal abscess, empyema, meningitis, and neck abscess.