Translating Neonatal Resuscitation Guidelines Into Practice in Brazil

In this prospective cohort study, we used information obtained from the database registry of the Brazilian Network on Neonatal Research, which comprises 20 academic public hospitals. The research ethics committee of the coordinating institution approved the project without requiring individual informed consent (#3691375).

From January 2014 to December 2020, all inborn infants 23^0/7 to 31^6/7 weeks’ gestational age (GA) with a birth weight of 400 to 1499 g and without major congenital anomalies diagnosed before or after birth¹⁵  were included in the study.

The 20 participating hospitals are high-risk pregnancy referral centers located in 12 cities of 7 Brazilian states.¹⁶  All centers had the required equipment and material for neonatal resuscitation in the delivery room, and their neonatal resuscitation teams received BNRP training. Updated BNRP guidelines for care of preterm newborns were published in 2011^10,11  and 2016,^12,13  and both were adapted from ILCOR recommendations.^4,5  In 2011, guidelines for newborns <34 weeks’ GA included the following:

1. Thermal care: The delivery room temperature is ≥26°C, and a radiant warmer, plastic bag, single cotton cap, and prewarmed transport incubator are used.

2. Monitoring: A pulse oximeter sensor is placed as soon as the patient is placed under the radiant warmer.

3. Positive pressure ventilation (PPV): PPV is started with a face mask and T-piece resuscitator followed by specified corrective steps and endotracheal intubation if face mask ventilation fails.

4. Supplemental oxygen: The use of free-flow oxygen was withdrawn from the guidelines. If a blender is available in the delivery room, PPV is started with fraction of inspired oxygen (Fio₂) 0.40 and increased by 0.20 every 15 seconds if the heart rate is <100 beats per minute and/or preductal pulse oxygen saturation (Spo₂) <70%. In the absence of a blender, room air is started and Fio₂ increased to 1.00 if needed.

5. Continuous positive airway pressure (CPAP): CPAP is applied in spontaneously breathing infants if respiratory distress and/or Spo₂ are below the target.

In 2016, the following additions or changes were made:

1. Thermal care: The delivery room temperature should be 23°C to 26°C and a double cap (plastic and cotton) and thermal mattress for infants weighing <1000 g used.

2. PPV: Recommendations stressed that endotracheal intubation should be done only after failure of face mask ventilation with adequate technique.

3. Supplemental oxygen: A blender should be available in the delivery room. Fio₂ 0.30 is started and increased by 0.20 every 30 seconds if the heart rate <100 beats per minute and/or preductal Spo₂ < 70%.

Maternal and neonatal characteristics were analyzed. GA was determined by the best obstetric estimate, and patients were divided into extremely preterm (23^0/7–27^6/7 weeks’ GA) and very preterm (28^0/7–31^6/7 weeks GA) groups. Small for GA was defined by INTERGROWTH-21st.¹⁷  The following clinical data were recorded: 5-minute Apgar score, hypothermia (axillary temperature <36.0°C) at admission to the NICU,¹⁸  death in the delivery room, and early neonatal death (0 to 6 days after birth).

To describe resuscitation practices, we calculated the frequency per year of the following procedures: use of a thermal care bundle (recommended delivery room temperature and use of plastic bag and cap), use of free-flow supplemental oxygen, any PPV during initial resuscitation, face mask PPV without endotracheal intubation, endotracheal intubation without a trial of face mask PPV, PPV using only a T-piece resuscitator, PPV with maximum Fio₂ < 0.40, PPV with maximum Fio₂ 1.00, use of a PPV bundle (face mask PPV with a T-piece resuscitator and maximum Fio₂ < 0.60), advanced resuscitation (PPV by endotracheal tube with chest compressions and/or medications), and use of CPAP in the delivery room.

The χ² test for trend was applied to compare the use of the resuscitation procedures per year. A logistic regression model, adjusted by year and center, was used to estimate the probability of incorporating recommended guidelines into clinical practice. Results are expressed as odds ratios (ORs) and 95% confidence intervals (CIs). Analyses were performed using SPSS for Windows version 21.0 software (IBM Corporation, Armonk, NY).

In the 20 participating centers, 12 669 very low birth weight infants were born between 2014 and 2020. Of these, 8514 inborn infants without congenital anomalies between 23^0/7 and 31^6/7 weeks’ GA and with a birth weight of 400 to 1499 g comprised the study population (Fig 1). Maternal and neonatal characteristics of both groups are shown in Table 1.

Among both extremely preterm and very preterm newborns, BNRP thermal guidelines were increasingly followed during study period (Table 2). The use of free-flow supplemental oxygen significantly decreased in the very preterm group. Although this use did not change in the extremely preterm group throughout the study period, no extremely preterm newborn received free-flow supplemental oxygen in the final study year (Table 2).

Most extremely preterm newborns (83%) received PPV in the delivery room. During the study period, the use of the T-piece resuscitator as the only device to deliver PPV in this group increased from 63% to 90% (P < .001; Fig 2A). Among the extremely preterm newborns requiring PPV, only 26% received face mask PPV without tracheal intubation across the full study period (Fig 2B). However, initiation of PPV with an endotracheal tube without an initial trial of face mask PPV decreased from 10% in 2014 to 6% in 2020 (P = .002; Fig 2B). As recommended, the use of Fio₂ 1.00 decreased from 77% to 64% (P < .001), and the use of a maximum Fio₂ <0.40 increased from 2% to 12% (P < .001; Fig 2C). The use of the PPV bundle increased from 7% in 2014 to 11% in 2020 (P = .011; Fig 2D). There was no change in the frequency of advanced resuscitation or the use of CPAP in the delivery room (Table 2).

Adjusted by center, there was a significant annual increase in the probability of adopting recommended practices in the extremely preterm group, including the thermal care bundle, attempting face mask ventilation before tracheal intubation, PPV with a T-piece, maximum Fio₂ <0.40, avoidance of Fio₂ 1.00, and use of the PPV bundle (Table 3). During the study period, hypothermia at NICU admission (P < .001) and early neonatal deaths (P = .026) among extremely preterm newborns significantly decreased (Table 4).

Among very preterm newborns, 59% received PPV in the delivery room. Of those receiving PPV, use of the T-piece resuscitator as the only device to deliver PPV increased from 71% in 2014 to 91% in 2020 (P < .001; Fig 2A). Across the full study period, 55% of the ventilated newborns between 28^0/7 and 31^6/7 weeks’ GA received only face mask PPV without tracheal intubation (Fig 2B). When receiving PPV, the use of a maximum Fio₂ of 1.00 decreased from 55% to 38% (P < .001), and the use of a maximum Fio₂ <0.40 increased from 5% to 24% (P < .001; Fig 2C). During the study period, use of the PPV bundle increased from 19% to 34% (P = .001; Fig 2D). In the very preterm group, use of CPAP in the delivery room increased from 62% in 2014 to 73% in 2020 (Table 2).

Adjusted by center, there was a significant annual increase in the probability of adopting recommended practices in the very preterm group, including use of the thermal care bundle, avoidance of supplemental free-flow oxygen, use of the T-piece resuscitator for PPV, avoidance of PPV with Fio₂ 1.00, provision of PPV with a maximum Fio₂ < 0.40, use of the PPV bundle, and use of CPAP in the delivery room (Table 3). During the study period, hypothermia at NICU admission (P < .001) and early neonatal deaths (P = .002) among very preterm newborns also significantly decreased (Table 4).

In this prospective cohort study, we show that among newborns <32 weeks’ GA, there was a gradual implementation of the BNRP guidelines on neonatal resuscitation in the centers affiliated with the Brazilian Network on Neonatal Research. Thermal care, PPV with the T-piece resuscitator, and more judicious use of oxygen during PPV were the main recommendations incorporated into practice. Among extremely preterm newborns (23–27 weeks’ GA), initiating PPV with an endotracheal tube decreased over the study period. In very preterm newborns (28–31 weeks’ GA), the use of CPAP in the delivery room increased. In both GA groups, changes were accompanied by a reduction in hypothermia at NICU admission and in early neonatal deaths.

In 2010, emphasis on thermal care for newborns <28 weeks’ GA during transition after birth was highlighted when ILCOR recommended covering their bodies with a polythene bag or wrap, without drying, and placing them under a radiant heater in a delivery room with temperature ≥26°C.⁴  In 2011, the BNRP adopted these recommendations and introduced the use of a cotton cap for very low birth weight newborns. The importance of adopting these measures was demonstrated by a Brazilian multicenter cohort study showing that hypothermia <36.0°C was present upon NICU admission in 55% of infants born at 23 to 33 weeks’ GA and increased the chance of early neonatal death 1.64-fold (95% CI, 1.03–2.61).¹⁹  In 2014, data forms for the Brazilian Network on Neonatal Research were revised to include information on the use of plastic bag, cap, and delivery room temperature, encouraging centers to pay attention to thermal care after birth. These actions likely had a role in the significant increase in the use of recommended thermal care measures noted in 2015 and maintained through 2020. Despite widespread adoption of these guidelines, hypothermia at NICU admission continues to be an important problem, occurring in 43% of all included preterm newborns at the study conclusion. Additional collaborative quality improvement initiatives are needed to further decrease hypothermia at admission.²⁰ 

In 2010, ILCOR recommended that ventilation of the newborn could be performed effectively with a self-inflating bag or a pressure-limited T-piece resuscitator but highlighted that in mechanical models, delivery of target inflation pressures and maintenance of positive end-expiratory pressure (PEEP) were more consistent when using a T-piece resuscitator.⁴  In 2015, ILCOR suggested the use of PEEP if the facilities and equipment were available and cautioned that PEEP could not be reliably administered with a self-inflating bag.⁵  In this context, in 2012, the BNRP started teaching the use of the T-piece resuscitator as an option for PPV among infants <34 weeks’ GA, and in 2016, the T-piece resuscitator was recommended as the first choice for PPV in this group of infants,^10,12  explaining the high frequency of T-piece use in 2014 and its sustained use throughout the study period. It is encouraging to note that the adoption of the T-piece for PPV at the participating centers was associated with an increased chance of survival to discharge from the hospital without either bronchopulmonary dysplasia or intraventricular hemorrhage.²¹ 

Controversy has persisted over the initial concentration of oxygen during PPV in the delivery room. In 2010, ILCOR suggested that blended oxygen and air should be used judiciously among newborns <32 weeks’ GA and ideally guided by pulse oximetry.⁴  In 2015, ILCOR recommended against initiating resuscitation of preterm newborns with high Fio₂ (0.65–1.00).⁵  On the basis of these recommendations, in 2010, the BNRP first suggested the use of gas blenders in the delivery room,¹⁰  which was strongly reinforced in 2015.¹²  Indeed, since 2015, all 20 participating Brazilian centers reported the presence of gas blenders, pulse oximeters, and Spo₂ targets clearly posted in the delivery room. Despite the widespread presence of adequate equipment and increase in the use of Fio₂ 0.40 during the study period, most extremely preterm (88%) and very preterm (76%) infants receiving PPV were still exposed to a maximum Fio₂ 1.00. Increasing access to appropriate interfaces for providing noninvasive PPV to the smallest preterm newborns, additional training to improve PPV technique, and guidance on adjusting Fio₂ during the first minutes of life may help Brazilian neonatologists to achieve a more judicious use of supplementary oxygen.

Among neonatologists, there is a consensus that tracheal intubation is difficult and associated with potential risks, including bradycardia and severe oxygen desaturations.^22,23  Therefore, at birth, emphasis has been placed on improving face mask PPV technique and optimizing the use of CPAP. In 2010, ILCOR stated that spontaneously breathing preterm infants who have respiratory distress may be supported with CPAP or intubation and mechanical ventilation.⁴  In 2015, ILCOR suggested the initial use of CPAP for spontaneously breathing preterm newborns with respiratory distress in the delivery room.⁵  Beginning in 2013, the BNRP stated that use of CPAP was feasible and might be beneficial in preterm newborns with respiratory distress and/or Spo₂ lower than expected at birth.¹⁰  In 2016, the wording was strengthened with recommendations for initial use of CPAP in newborns <34 week’s GA with respiratory distress.¹²  During the study period, tracheal intubation as the first option to ventilate extremely preterm newborns decreased 13% per year, and CPAP use in very preterm newborns at birth increased 11% per year. However, there are still opportunities for improvement. Research on new interfaces and modes to deliver CPAP to extremely and very preterm newly born infants may decrease the intubation rate and increase the success of CPAP in these newborns.²⁴ 

It is interesting to consider whether some interventions could compete with others in our study. Appropriate thermal care (use of plastic bag and cap and delivery room temperature ≥23°C) and appropriate respiratory care (PPV only by face mask with T-piece resuscitator and Fio₂ <0.60) were provided simultaneously to 847 (15.3%) infants (53 [5.6%] in 2014 and 133 [19.9%] in 2020; P < .001). This finding reveals that improvements were across the board, and that one group of interventions did not compete with the others.

The adoption of the practices recommended by the BNRP were not associated with changes in frequency of advanced resuscitation. Throughout the study period, 10% to 12% of extremely preterm and 2% to 4% of very preterm newborns received chest compressions and/or medications. Frequency of advanced resuscitation at birth in preterm infants <32 weeks’ GA varied from 6.2% to 11.8% in cohort studies in California,²⁵  Italy,²⁶  and China.²⁷  The fact that the 20 study centers are referral units for high-risk pregnancies may be an explanation for the results. However, it is possible that improvements in PPV technique, first-attempt success with tracheal intubation, and use of electronic cardiac monitoring may lead to decreases in the rate of advanced resuscitation.

The reduction of early neonatal mortality during the study period is likely multifactorial. During the past 20 years, Brazil has focused special attention on maternal and neonatal care, introducing a sequence of initiatives, such as the National Program for the Humanization of Pregnancy and Childbirth (2000), the Pact for the Reduction of Maternal and Newborn Mortality (2004), the creation of local committees for the prevention of infant mortality (2005), the Pact for Life (2006), the Stork Network (2011),²⁸  and the Qualification of Care to Newborns at Risk–Qualineo initiative (2017).²⁹  These strategies included training in resuscitation at birth and may have contributed to improving the quality of neonatal care and decreasing early mortality.

The study is limited by using a database that was not specifically designed to answer questions related to the indication, time of initiation after birth, duration, and sequence of performed resuscitation procedures. In addition, some variables of interest were not available, such as cord-clamping strategies, initial Fio₂ for PPV, reasons why providers decided to use a self-inflating bag instead of a T-piece resuscitator, and the indication for tracheal intubation after face mask ventilation. Patient/case mix profile change over time was not assessed. Because the main outcomes of the study were related to the technical aspects of the resuscitation procedures, this limitation does not compromise the internal validity of the results. There are also some limitations on the external validity of the study that enrolled a selected group of Brazilian public university hospitals.

Despite these limitations, this is one of the few studies that evaluates actual delivery room practices in a large population of extremely and very preterm infants instead of surveying centers about their policies and intended practices. Moreover, the data were collected in a prospective and systematic way with strict definitions for each event and continuous input into an electronic database wherein consistency is verified by an independent data committee. Therefore, information on all outcomes was available for >99% of the studied infants, except for delivery room temperature, which was available for 93%. The fact that the study was completed in a middle-income country adds information about the translation of ILCOR recommendations outside the highest income countries and may inform future studies in less developed settings.

In conclusion, we show in this study that ILCOR recommendations on resuscitation of preterm newborns were successfully translated through BNRP guidelines and training into clinical practice in public hospitals in Brazil from 2014 to 2020. We suggest that the reproducibility of the Brazilian experience in other contexts depends on the presence of national programs capable of discussing and adapting ILCOR treatment recommendations into their local realities.

The authors acknowledge the Brazilian Pediatric Society for the support to the Brazilian Neonatal Resuscitation Program and thank Olga Bomfim, MD, and Fabio Carmona, MD, for their important role in helping with the Brazilian Network on Neonatal Research database.

BNRP

Neonatal Resuscitation Program of the Brazilian Society of Pediatrics

CI

confidence interval

CPAP

continuous positive airway pressure

Fio₂

fraction of inspired oxygen

GA

gestational age

ILCOR

International Liaison Committee on Resuscitation

OR

odds ratio

PEEP

positive end-expiratory pressure

PPV

positive pressure ventilation

Spo₂

pulse oxygen saturation