149 / 208
PEPaNIC trial
147
7
Supplementary Table 6.
P-values for interaction between randomized treatment allocation and the
predefined baseline risk factors on the primary outcomes
P-value
Patients with New Infections
1
PELOD score first 24h in PICU
0.80
Age group
0.55
Diagnostic group
0.25
STRONGkids score
1
0.11
Center
0.72
Time to live discharge from the PICU
2
PELOD score first 24h in PICU
0.80
Age group
0.65
Diagnostic group
0.73
STRONGkids score
1
0.19
Center
0.56
PELOD=PEdiatric Logistic Organ Dysfunction
7
, PICU=Pediatric Intensive Care Unit
1
Multivariable Logistic Regression Analysis censored at 90 days with non-survivors censored at 91 days
2
Multivariable Cox Proportional Hazard Analysis censored at 90 days with non-survivors censored at 91
days
Supplementary Table 7.
Adjusted Odds Ratios and Hazard Ratios further corrected for hypoglycemia (plasma
concentration glucose < 40 mg/dL) and for the average amount of enteral feeding (kcal/kg/day) during
randomisation window
OR or HR adjusted*
(95% CI)
Late PN vs. Early PN
(adjusted
for
hypoglycemia)
P-value OR or HR adjusted*
(95% CI)
Late PN vs. Early PN
(adjusted
for
enteral kcal/kg/d)
P-value
Primary Outcomes
Patients with New Infections
0.45 (0.32-0.62)
< 0.001 0.47 (0.34-0.65)
< 0.001
Likelihood Earlier Live PICU Discharge
1.26 (1.13-1.41)
< 0.001 1.24 (1.12-1.38)
< 0.001
Secondary Efficacy Outcomes
Likelihood Earlier Live Weaning
from Mechanical Ventilatory Support
1.21 (1.09-1.35)
< 0.001 1.19 (1.07-1.32)
0.001
Renal-Replacement Therapy
0.49 (0.24-0.97)
0.03
0.52 (0.25-1.03)
0.06
Likelihood Earlier Live Hospital
Discharge
1.22 (1.09-1.36)
< 0.001 1.19 (1.07-1.33)
0.001
All duration of care outcomes were censored at 90 days with non-survivors censored at 91 days. OR=Odds
Ratio, HR=Hazard Ratio, PN=Parenteral Nutrition, 95% CI= 95% Confidence Intervals, PICU=Pediatric
Intensive Care Unit
* Adjusted for the following risk factors: center, age group, diagnosis group, PELOD score first 24h
7
,
STRONGkids category
1
and hypoglycemia (plasma concentration glucose < 40 mg/dL) or amount of
enterally administered kcal per kg per day.
In both analyses, experiencing hypoglycemia and receiving a higher amount of enterally administered
kcal per kg per day were independent risk factors for infections and for a delayed live discharge from PICU
(all P-values ≤0.004).