Table 2.
Clinical characteristics and outcomes of Subclass 1 and Subclass 2
| Measures | Subclass 1 | Subclass 2 | p Value |
|---|---|---|---|
| N = 21 | N = 25 | ||
| Demographics | |||
| Age-years ‡ | 11.4 ± 5.3 | 7.0 ± 5.1 | 0.004* |
| Male** | 9 (42.9%) | 13 (52.0%) | 0.54 |
| Race** | 0.23 | ||
| White | 15 (71.4%) | 15 (60%) | |
| Black | 2 (9.5%) | 7 (28%) | |
| Asian | 3 (14.3%) | 1 (4%) | |
| American Indian | 1 (4.8%) | 0 (0%) | |
| Mixed | 0 (0%) | 1 (4%) | |
| Declined or Unknown | 0 (0%) | 1 (4%) | |
| Baseline clinical characteristics | |||
| PRISM Score ‡ | 15.1 ± 7.5 | 12.4 ± 9.3 | 0.25 |
| Average blood glucose (mg/dl) ll | 134 (123–170) | 116 (105–125) | 0.021* |
| Clinical Trial Allocation (two groups total, n assigned to group 2)** | 15 (71.4%) | 13 (52.0%) | 0.17 |
| Baseline primary reasons for admission** | 0.12 | ||
| Cardiovascular (including shock) | 5 (23.8%) | 6 (24.0%) | |
| Following elective procedure | 0 (0%) | 2 (8.0%) | |
| Following emergent procedure | 2 (9.5%) | 0 (0%) | |
| Gastrointestinal or Liver | 3 (14.3%) | 0 (0%) | |
| Neurologic | 0 (0%) | 1 (4.0%) | |
| Respiratory (including infections) | 11 (52.4%) | 16 (64.0%) | |
| Baseline primary reasons for mechanical ventilation** | 0.78 | ||
| Acute respiratory failure related to sepsis | 13 (61.9%) | 10 (40.0%) | |
| Aspiration pneumonia | 0 (0%) | 1 (4.0%) | |
| Asthma or reactive airway disease | 1 (4.8%) | 2 (8.0%) | |
| Bronchiolitis | 1 (4.8%) | 1 (8.0%) | |
| Neurologic | 1 (4.8%) | 1 (4.0%) | |
| Pneumonia/hypoxia | 4 (19.0%) | 7 (28.0%) | |
| Procedural | 1 (4.8%) | 3 (12.0%) |
| Clinical outcomes | Adjusted p value† | ||
|---|---|---|---|
| Maximum PELOD score‡ | 18.8 ± 2.2 | 13.0 ± 1.4 | 0.026 * |
| Maximum vasoactive-inotrope score‡ | 21.5 ± 5.0 | 9.8 ± 2.7 | 0.045* |
| ICU length of stay from randomization to discharge (days) ‡ | 24.0 ± 5.0 | 10.4 ± 2.2 | 0.021* |
| Hospital-free days through day 28 (days) ‡ | 5.7 ± 1.5 | 11.1 ± 1.4 | 0.011* |
| Mortality (death at 90 days) § | 1 (4.8%) | 2 (8.0%) | 0.999 |
*Significant p values with FDR < 10% are noted with an asterisk
†Adjusted p value was computed using bootstrap linear regression model to evaluate mean differences between subgroups adjusting for treatment group, PRISM score and baseline use of steroids
‡ Data presented as mean ± SEM by bootstrapping
§ Adjusted p value was computed using the Fisher’s exact test
llData presented as median (interquartile range)
**Data presented as n (percentage)