Variable | Statistics | Treatment modality | CLAC_{fast} versus RMC_{only} | CLAC_{fast} versus CLAC_{slow} | ||
RMC_{only} | CLAC_{fast} | CLAC_{slow} |
Adjusted mean difference*
(90 % CI) [95% CI] One sided P value Superiority |
Adjusted mean difference*
(90 % CI) [95% CI] One sided P value Non-inferiority | ||
Time within target range | %, mean±SD | 58±11 | 68±11 | 65±11 | 9.5 (5.6 to 13.5) [4.8 to 14.3] 0.0001 | 3.4 (−0.6 to 7.3) [−1.4 to 8.1] 0.0005 |
Adjusted mean difference*
[95% CI] Two sided P value |
Adjusted mean difference*
[95% CI] Two sided P value | |||||
Time above target range | %, mean±SD | 13±10 | 13±7 | 12±7 | 0.0 [−4.2 to 4.2] 0.99 | 1.2 [−3.0 to 5.5] 0.55 |
Time below target range | %, mean±SD | 22±11 | 14±5 | 17±8 | −9.1 [−13.5 to –4.6] 0.0002 | −3.1 [−7.6 to 1.3] 0.16 |
Time with SpO_{2} <80% | %, mean±SD | 3.5±2.7 | 2.8±2.7 | 3.3±3.0 | −0.6 [−1.7 to 0.4] 0.24 | −0.5 [−1.6 to 0.6] 0.35 |
Severe hypoxaemic episodes† | Patients with episodes (n) | 12 | 11 | 13 | N.A. | N.A. |
Rate (counts per hour) Median (IQR) | 0.37 (0.00–0.63) | 0.12 (0.00–0.57) | 0.23 (0.00–0.44) | Two sided P value 0.42‡ | Two sided P value 0.89‡ | |
Duration(s) Median† (IQR) | 83 (70–104) | 84 (73–99) | 79 (75–94) | Two sided P value 0.59‡ (n=9) | Two sided P value 0.56‡ (n=11) | |
Manual FiO_{2} adjustments | Rate (counts per hour) Median (IQR) | 3.8 (2.1–5.4) | 2.7 (2.0–3.8) | 3.6 (1.0–6.1) | Two sided P value 0.08‡ | Two sided P value 0.44‡ |
Non-inferiority margin 5% points.
*Effect estimates resulting from fitting a linear mixed model with random subject effect and fixed effects for period and intervention.
†Severe hypoxaemic episodes had only occurred in 15 out of 19 infants. Results are given only for patients with at least one episode during the study.
‡Wilcoxon signed-rank test.
CLAC, closed-loop automatic control; FiO_{2}, fractional inspired oxygen; RMC, routine manual control; SpO_{2}, pulse oximeter saturation.