Population
Our tertiary level hospital covers the Saint-Denis area in the north of Paris, France, with a rate of ~ 4500 deliveries/year. Inborn VLGAI requiring pulmonary resuscitation born from July 2014 to June 2015 (n=91) and for all of 2016 (n=72) were included in this retrospective cohort study. A 6-month gap between both periods was needed to introduce LR maneuvers to the NICU team. Data were collected from patients’ medical records and reports.
A total of 242 VLGAI were managed during these periods (119 during 2014/2015 and 123 during 2016). We excluded infants transferred from other hospitals (n= 12) and (n=14), infants suffering from major congenital anomalies (n=3) and (n=3), and newborns requiring no respiratory support beyond a CPAP value of 5 cmH2O (n=13) and (n=10) in 2014/2015 and 2016, respectively. Respiratory support was started based on currently used international guidelines [8,9]. The adherence to the new protocol was judged on the written words describing the resuscitation in the DR. During the second period, 24/96 (25%) pulmonary resuscitations did not follow the LR algorithm and were excluded from the analysis.
Data collection was approved by the local area ethics committee on human research which allowed a waiver of informed consent for this retrospective study (Protocol No. CLEA-2017-031).
Intervention (Figure 1)
The intervention was the introduction of the LR strategy. In the 2014/2015 period, the respiratory assistance of VLGAI was conducted according to international recommendations (PEEP of 5 cm H2O, inflation duration of 2-3 s, peak inspiratory pressure of 20-25 cmH2O) using a face mask with a T-piece resuscitator (Neopuff, Fisher & Paykel, Auckland, New-Zealand), followed by intubation if the baby remained apneic or cyanotic with a pulse oxygen saturation (SpO2) < 75% at 5 min of life whatever the fraction of inspired oxygen (FiO2), or with a heart rate (HR) < 100/min or an FiO2 > 0.4 at 10 min of life in order to achieve an SpO2 > 85%. During 2016, respiratory support was changed to gradual increase in inflations duration (from 3 to 5, to 10, then 15 seconds), as needed, and performed with a simultaneous increase in PEEP every 1 to 2 SI from 5 to 8, 10, 12, then 15 cmH2O, according to the clinical response, which included persistent bradycardia, apnea or gasping. A gradual decrease in PEEP was performed when FiO2 reached 0.4, while SIs were stopped when regular spontaneous breathing was observed. In case of severe asphyxia, emergency intubation with or without chest compressions did not cancel the LR strategy. During both periods, the resuscitation team included at least a neonatologist, a pediatric resident, and a midwife. The team member at the head of each newborn kept track of the titrating of SIs by counting aloud each duration, and announcing the PEEP level set at the PEEP cap located on the T-piece. The PIP level which was not modified with the PEEP level, was eventually increased by turning the ad-hoc knob by the team member closest to it.
Monitoring in the DR
Postnatal hypothermia was prevented by using radiant heaters and warmed blankets. After drying, extremely low gestational age infants (ELGAIs) < 29 weeks were immediately wrapped in a sterile, transparent plastic bag. The head was covered by a warmed cap. Temperature was monitored via a cutaneous sensor. HR was measured by stethoscope and then monitored by an SpO2 sensor (MassimoTM) placed at the right hand.
Transport to the neonatal intensive care unit (NICU)
A preheated neonatal transport incubator was used for the transfer to the NICU. During both periods, infants were ventilated (invasively or not) using a Babylog 8000 (Dräger, Lübeck, Germany). Continuous positive airway pressure (CPAP) or noninvasive ventilation (NIV) was provided during the transfer by using nasal cannula (Neotech RAM CannulaTM, Valencia, CA, USA).
Co-Interventions
Obstetric Management
Some changes in obstetric treatment were introduced in October 2016: antibiotic therapy duration after preterm premature rupture of membrane was shortened from 5 to 3 days [12]; and magnesium sulfate was infused to pregnant women prior to the preterm birth of infants <33 weeks’ gestation [13].
NICU Management
Neonatal management was as follows:
- Intubation criteria did not change: (i) in the DR, they remained as stated above; (ii) in the NICU, the criteria were an FiO2 > 0.4, frequent (> 6 occurrences during 6 hours) or severe apneas (> 1 per 6 hours requiring bag-mask ventilation).
- CPAP or NIV was maintained using the InfantFlow® biphasic nCPAP device (SEBAC, Gennevilliers, France). The synchronized mode was used in case of hypercapnia or moderate apnea despite caffeine treatment.
- During 2016, permissive hypercapnia was introduced with a maximal tolerated pCO2 limit of 65 mmHg (vs 55 mmHg during 2014/2015).
- Oxygen level was adjusted to obtain an SpO2 target range of 87-93% during 2016 instead of 85-95% during 2014/2015.
- During 2016, the maintenance dose of caffeine (5 mg/kg/day) was doubled in case of moderate apnea.
- During the second period, the first surfactant administration (200 mg/kg of Curosurf, Chiesi, Parma, Italy) was indicated when FiO2 exceeded 0.25 and 0.30 in ELGAIs and VLGAIs respectively, instead of 0.4 during the first period, whatever the gestational age (GA). During both periods, a second dose (100 mg/kg) was administered when FiO2 exceeded 0.4. If the infant was not intubated, the intubation-surfactant-extubation (INSURE) procedure was performed, and considered successful (without MV) if the extubation was carried out less than one hour later.
- Postnatal corticosteroid therapy criteria remained the same for premature infants more than 21 days old: persistent need for MV with FiO2 exceeding 0.4, or after failure of extubation.
- During 2016, early-rescue high frequency oscillation (HFO) ventilation was introduced in case of INSURE failure or an early (< 72h of life) need for invasive PPV, while volume guarantee ventilation was introduced to accelerate the weaning process. A change from a systematic sedation-analgesia toward an individualized approach based on the EDIN scale [14] was also performed in invasively ventilated infants.
- No major changes in the nutrition protocol occurred, except for probiotics (Lactobacillus casei and Lactobacillus rhamnosus) introduced in April 2016.
Primary and Secondary Outcomes
The primary outcome was the rate of MV in the first 72 hours. Secondary outcomes were the rate of intubation in the DR; Apgar score at 5 min; surfactant use; ventilator days; postnatal corticosteroids; mortality and morbidity including physiologic BPD [12]; pneumothorax; intraventricular hemorrhage (IVH) > grade 2 [13]; periventricular leukomalacia [14]; treated patent ductus arteriosus (PDA); necrotizing enterocolitis (NEC) ≥ stage 2a [15]; and retinopathy of prematurity (ROP) > stage 2 [16].
Statistical Analysis
Data are summarized as median and interquartile range for quantitative data and as count and percentage for categorical data. Statistical analyses were performed using chi-squared test or Fisher’s exact test for categorical variables and the t-test or Mann-Whitney U-test for continuous variables. All outcomes with p < 0.10 in univariate analysis were analyzed with multiple logistic regression. Adjustments were made for confounding, such as sex, birth weight, GA, mode of delivery and low Apgar score (< 5) at 1 minute. Similar analyses were performed for the infants of GA between 29 and 31 weeks and ELGAI subgroup. No adjustment of P values was performed to account for multiple comparisons because subgroups analyses are considered exploratory. All tests were two-sided at a 0.05 significance level. Analyses were carried out using R statistical software version 3.3.2.