If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Division of Neonatology, Medical University of Graz, Auenbruggerplatz 34/2, 8036 Graz, AustriaResearch Unit of Cerebral Development and Oximetry, Medical University of Graz, Auenbruggerplatz 34/2, 8036 Graz, AustriaNICU “V. Buzzi” Children’s Hospital, ASST-FBF-Sacco, Via Castelvetro 32, 20154 Milan, Italy
Division of Neonatology, Medical University of Graz, Auenbruggerplatz 34/2, 8036 Graz, AustriaResearch Unit of Cerebral Development and Oximetry, Medical University of Graz, Auenbruggerplatz 34/2, 8036 Graz, AustriaCentre for the Studies of Asphyxia and Resuscitation, Neonatal Research Unit, Royal Alexandra Hospital, 10240 Kingsway Avenue NW, T5H 3V9 Edmonton, Alberta, Canada
Division of Neonatology, Medical University of Graz, Auenbruggerplatz 34/2, 8036 Graz, AustriaResearch Unit of Cerebral Development and Oximetry, Medical University of Graz, Auenbruggerplatz 34/2, 8036 Graz, Austria
Division of Neonatology, Medical University of Graz, Auenbruggerplatz 34/2, 8036 Graz, AustriaResearch Unit of Cerebral Development and Oximetry, Medical University of Graz, Auenbruggerplatz 34/2, 8036 Graz, Austria
Division of Neonatology, Medical University of Graz, Auenbruggerplatz 34/2, 8036 Graz, AustriaResearch Unit of Cerebral Development and Oximetry, Medical University of Graz, Auenbruggerplatz 34/2, 8036 Graz, Austria
Research Unit of Cerebral Development and Oximetry, Medical University of Graz, Auenbruggerplatz 34/2, 8036 Graz, AustriaCentre for the Studies of Asphyxia and Resuscitation, Neonatal Research Unit, Royal Alexandra Hospital, 10240 Kingsway Avenue NW, T5H 3V9 Edmonton, Alberta, CanadaDepartment of Pediatrics, University of Alberta, 10240 Kingsway Avenue NW, T5H 3V9 Edmonton, Alberta, Canada
Division of Neonatology, Medical University of Graz, Auenbruggerplatz 34/2, 8036 Graz, AustriaResearch Unit of Cerebral Development and Oximetry, Medical University of Graz, Auenbruggerplatz 34/2, 8036 Graz, Austria
Division of Neonatology, Medical University of Graz, Auenbruggerplatz 34/2, 8036 Graz, AustriaResearch Unit of Cerebral Development and Oximetry, Medical University of Graz, Auenbruggerplatz 34/2, 8036 Graz, Austria
Analysis of the impact of bradycardia and hypoxemia on the course of cerebral and peripheral oxygenation parameters in preterm infants in need for respiratory support during foetal-to-neonatal transition.
Methods
The first 15 min after birth of 150 preterm neonates in need for respiratory support born at the Division of Neonatology, Graz (Austria) were analyzed. Infants were divided into different groups according to duration of bradycardia exposure (no Bradycardia, brief bradycardia <2 min, and prolonged bradycardia ≥2 min) and to systemic oxygen saturation (SpO2) value at 5 min of life (<80% or ≥80%). Analysis was performed considering the degree of bradycardia alone (step 1) and in association with the presence of hypoxemia (step 2).
Results
In step 1, courses of SpO2 differed significantly between bradycardia groups (p = 0.002), while courses of cerebral regional oxygen saturation (crStO2) and cerebral fractional tissue oxygen extraction (cFTOE) were not influenced (p = 0.382 and p = 0.878). In step 2, the additional presence of hypoxemia had a significant impact on the courses of SpO2 (p < 0.001), crStO2 (p < 0.001) and cFTOE (p = 0.045).
Conclusion
Our study shows that the degree of bradycardia has a significant impact on the course of SpO2 only, but when associated with the additional presence of hypoxemia a significant impact on cerebral oxygenation parameters was seen (crStO2, cFTOE). Furthermore, the additional presence of hypoxemia has a significant impact on FiO2 delivered. Our study emphasizes the importance of HR and SpO2 during neonatal resuscitation, underlining the relevance of hypoxemia during the early transitional phase.
Continuous monitoring of heart rate (HR) and peripheral oxygen saturation (SpO2) by using pulse-oximetry (plus ECG optionally) is currently considered standard of care during stabilization of preterm infants in the delivery room (DR). Reaching specific SpO2 values during resuscitation is advocated, and the titration of blended supplemental oxygen accordingly is also recommended by international guidelines.
SpO2 is widely used as a proxy for adequate ventilation and oxygenation stabilisation, whereas monitoring of HR is used as a proxy for cardio-circulatory stability.
There is still uncertainty regarding the optimal initial supplemental oxygen concentration (FiO2) to start resuscitation in preterm infants,
Increased incidence of mortality and adverse outcomes such as intraventricular haemorrhage (IVH) has been reported in those infants not reaching SpO2 80% at 5 min after birth.
Increase in HR is often reflection of adequate respiratory support, but there is still an ongoing debate to define normal ranges. Nevertheless, it has been shown that preterm neonates who experience prolonged bradycardia during DR resuscitation are at increased risk for death and/or IVH.
BRADYPREM study: is heart rate the most vital of all vital signs during preterm resuscitation? in "Selected abstracts of the 2nd Congress of joint European Neonatal Societies (jENS); Venice (Italy); October 31–November 4 2017; session "Neonatal pulmonology, neonatal respiratory support, resuscitation".
To date, the interaction of bradycardia and hypoxaemia and its effect on tissue oxygen delivery and tissue oxygenation during the immediate transition after birth is unclear. More understanding of the physiology of this interaction might explain the interplay between hypoxemia and bradycardia during neonatal resuscitation. With this aim and to detect possible regional differences, we included near infrared spectroscopy (NIRS). NIRS enables the non-invasive measurement of regional cerebral tissue oxygen saturation (crStO2) and the calculation of cerebral fractional tissue oxygen extraction (cFTOE). Thus, NIRS provides information on the balance of cerebral oxygen delivery and oxygen consumption, increasing the spectrum of oxygenation parameters. NIRS has been used in neonatal research setting during postnatal transition with both preterm and term infants.
Typical changes of cerebral oxygenation in the first minutes after birth, as well as differences according to the need of respiratory support, have been described.
NIRS has also been successfully used to guide respiratory support and supplemental oxygen to reduce the burden of cerebral hypoxia during immediate transition and resuscitation after birth.
Cerebral oxygen saturation to guide oxygen delivery in preterm neonates for the immediate transition after birth: a 2-center randomized controlled pilot feasibility trial.
The aim of the present study was to analyze the impact of the degree of bradycardia and the presence of hypoxemia on oxygenation parameters such as SpO2, crStO2 and cFTOE in preterm infants needing respiratory support during early neonatal transition. We hypothesized that the combination of bradycardia and hypoxaemia would be associated with significantly lower oxygenation parameters.
Methods
This study represents a retrospective analysis of four studies, conducted between December 2010 and March 2017 at the Division of Neonatology, Department of Pediatrics and Adolescent Medicine, Medical University of Graz, Austria.
Cerebral oxygen saturation to guide oxygen delivery in preterm neonates for the immediate transition after birth: a 2-center randomized controlled pilot feasibility trial.
We included preterm infants <37 weeks’ gestation, who fulfilled the following criteria: (i) decision to conduct full life support, (ii) written parental consent, (iii) need for respiratory support during resuscitation, and (iv) no severe congenital malformation. The Regional Committee on Biomedical Research Ethics had approved all the studies and allowed post-hoc analysis. For all infants, maternal medical history and neonatal demographic data were documented. The included studies were designed to measure crStO2 during the first 15 min after birth using NIRS. A standardized protocol was followed in all studies. In all infants, the cord was clamped within 30 s after birth. After that, the neonate was placed on the resuscitation table (CosyCot; Fisher& Paykel Healthcare; New Zealand or Giraffe incubator, GE Healthcare; United Kingdom) in supine position. A polyethylene bag was used in neonates <28 weeks’ gestation. Then, a NIRS sensor was attached to the infant’s left forehead. A pulse oximetry sensor (IntelliVue MP50 monitor; Philips; Netherland) was applied on the right palm or wrist to monitor SpO2 and HR. Upper airway suction was performed as needed. Respiratory support as continuous positive airway pressure and/or positive pressure ventilation was provided via a face mask (LSR Silicon mask no. 0/0 or 0/1; Laerdal; Norway) and the ‘Neopuff Infant T- Piece Resuscitator’ (Perivent; Fisher& Paykel Healthcare; New Zealand) with the following starting setting: gas flow 6–8 L/min, positive end-expiratory pressure 5 cmH2O, peak inspiratory pressure 30 cmH2O and FiO2 0.3. HR, SpO2 and crStO2 were recorded every second for the first 15 min after birth and stored in a multichannel system alpha-trace digital MM (BEST Medical Systems; Austria). Cerebral oxygenation was measured with INVOS 5100C (Somanetics, Troy, Michigan) or NIRO 200-NX (Hamamatsu, Japan). cFTOE was calculated for each minute as follows: (SpO2-crStO2)/SpO2.
Use of tissue oxygenation index and fractional tissue oxygen extraction as non-invasive parameters for cerebral oxygenation. A validation study in piglets.
Values of crStO2 higher than correspondent SpO2 were considered artefacts and eliminated from the analysis, as well as HR and/or SpO2 values taking longer than 5 min to be detected and displayed.
Definition of degree of bradycardia and presence of hypoxemia
We created a statistical model to first analyse the impact of the degree of bradycardia alone on oxygenation parameters (step 1), then to explore the interplay between the degree of bradycardia and the presence of hypoxemia (step 2).
Bradycardia was defined as HR < 100 bpm. Degree of bradycardia was calculated considering the sum of episodes the neonate spent with HR < 100 bpm within the first 15 min of life. Hypoxemia was considered as SpO2 <80% at 5 min after birth.
In step 1, we divided our population of infants into three groups, according to duration of exposure to bradycardia during the first 15 min after birth: no Bradycardia (nB), brief bradycardia (<2 min) (bB) and prolonged bradycardia (≥2 min) (pB).
In step 2, we integrated the presence of hypoxemia using a dichotomous criterion. We divided our population into 2 groups: a group with no presence of hypoxemia (H−), infants had SpO2 ≥80% at 5 min after birth; and a group with presence of hypoxemia (H+), infants had SpO2 <80% at 5 min after birth.
Statistical analysis
Observed data are presented as mean ± SD or absolute frequencies and percentages. We investigated the changes in HR, SpO2, FiO2, crStO2 and cFTOE within the first 15 min after birth using a linear mixed model with a first-order ante-dependence covariance structure. In step 1, fixed effects “time”, “degree of bradycardia” (nB vs. bB vs. pB) and the interaction of these two factors were included. In step 2, the fixed effect “presence of hypoxemia” (H− vs. H+), the interaction “time with presence of hypoxemia” and the interaction “time with degree of bradycardia with presence of hypoxemia” was added. Results are presented as estimated means and 95% confidence intervals. Post hoc analysis of differences between groups for each minute were performed. A p-value <0.05 was considered statistically significant. The statistical analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC, USA).
Results
The original four publications included 636 infants, and after selection 150 infants were analysed (Fig. 1, Table 1). Demographic data are outlined in Table 2. The cohort displayed a mean gestational age of 33 weeks and a mean birth weight of 1758 g. Only 11% of infants needed intubation. Two (1%) infants suffered from severe IVH, and the overall mortality was 3%. The mean (SD) duration until HR was displayed on the monitor was 93 (42) seconds.
•
Degree of bradycardia and SpO2at 5 min and differences in the HR course
Fig. 1Process of infants’ selection for post-hoc analysis from four studies.
In step 1, courses of HR differed significantly between bradycardia groups (p < 0.001). Neonates in the pB group had significantly lower HR values until minute 8 compared to neonates in the nB group, and until minute 4 compared to neonates in the bB group. Neonates in the bB group had significantly lower HR values until minute 6 compared to neonates with no bradycardia. After minute 8 there were no differences between groups anymore (Fig. 2A).
Fig. 2Course of HR (bpm) during the first 15 min of neonatal transitional according to (A) degree of bradycardia and (B) degree of bradycardia plus integration of presence of hypoxemia. Course of FiO2 (%) according to (C) degree of bradycardia and (D) degree of bradycardia plus integration of presence of hypoxemia.
In step 2, impact of degree of bradycardia on the course of HR was still present. Additionally, the presence of hypoxemia had a significant impact on the course of HR (p = 0.013). In the pB group, neonates with hypoxia (H+ group) had a lower HR within the first minutes (Fig. 2B). Tables of HR, FiO2, CrStO2 and cFTOE values at each minutes and statistical significance are available as Supplementary material.
•
Degree of bradycardia and SpO2at 5 min and differences in the FiO2course
In step 1, courses of FiO2 did not differ significantly between the three bradycardia groups, but there was a trend (p = 0.058). This trend was caused by significantly higher FiO2 from minute 5 to 13 in pB group, and significant higher FiO2 from minute 7 to 10 in bB group, compared to nB group. There were no differences in FiO2 between the bB and the pB groups. (Fig. 2C).
In step 2, there was no significant impact of degree of bradycardia on the course of FiO2. Presence of hypoxemia had a significant impact on the course of FiO2 (p < 0.001). While in the H+ group neonates with no bradycardia had almost no changes in FiO2 values, in the bB and in the pB groups FiO2 values increased after minute 6 and minute 5 respectively (Fig. 2D).
•
Degree of bradycardia and SpO2at 5 min and differences in the SpO2course
In step 1, courses of SpO2 differed significantly between bradycardia groups (p = 0.002). Neonates in the pB group had significantly lower SpO2 values until minute 5 compared to nB group, and until minute 4 compared to bB group. After minute 5 there were no differences between groups (Fig. 3A).
Fig. 3Course of SpO2 (%) during the first 15 min of neonatal transitional phase according to (A) degree of bradycardia and (B) degree of bradycardia plus integration of presence of hypoxemia. Course of crStO2 (%) according to (C) degree of bradycardia and (D) degree of bradycardia plus integration of presence of hypoxemia. Course of cFTOE according to (E) degree of bradycardia and (F) degree of bradycardia plus integration of presence of hypoxemia.
In step 2, impact of degree of bradycardia on the course of SpO2 was still present. Additionally, the presence of hypoxemia had a significant impact on the course of SpO2 (p < 0.001). Regardless of degree of bradycardia, neonates in the H− group had higher SpO2 values in the first minutes compared to infants in the H+ group (Fig. 3B).
•
Degree of bradycardia and SpO2at 5 min and differences in the crStO2course
In step 1, courses of crStO2 were similar between the three bradycardia groups (p = 0.382) (Fig. 3C).
In step 2, there was no significant impact of degree of bradycardia on the course of crStO2. Presence of hypoxemia had a significant impact on the course of crStO2 (p < 0.001). While at the beginning the increase of crStO2 was comparable between both hypoxemia groups, in the H+ group this increase continued until minute 12 and in the H- group flattened at minute 6 (Fig. 3D).
•
Degree of bradycardia and SpO2at 5 min and differences in the cFTOE course
In step 1, courses of cFTOE were similar between the three bradycardia groups (p = 0.878) (Fig. 3E).
In step 2 there was no significant impact of degree of bradycardia on the course of cFTOE. Presence of hypoxemia had a significant impact on the course of cFTOE (p = 0.045). Because of a steeper decrease in the H+ group, this group reached stable values at minute 10, whereas the H− group reached stable values at minute 7 (Fig. 3F).
Discussion
To our knowledge, this is the first study analysing the impact of the degree of bradycardia exposure alone and in combination with the presence of hypoxemia on circulatory and oxygenation parameters in preterm infants needing respiratory support during stabilisation at birth. Analysing the degree of bradycardia alone, our results showed that it had impact on systemic oxygenation only (SpO2). Including the presence of hypoxemia, we showed that the combination had impact on all three oxygenation parameters, confirming our hypothesis. The additional presence of hypoxaemia had a significant impact on FiO2 delivery as well.
Degree of bradycardia
Traditionally, HR < 100 bpm during neonatal transition is considered bradycardia. However, this definition still rises concern. There is lack of evidence that this threshold value is clinically relevant. In this regard, Smit et al. have shown that healthy infants after uncomplicated birth had significantly lower HR values than defined referenced ranges, and the 10th percentile was even <100 bpm until 5 min after birth.
In our analysis, we considered bradycardia HR < 100 bpm using the same criteria as Kapadia and co-workers. They showed that preterm neonates who experience prolonged bradycardia during DR resuscitation have increased risk for death and for IVH.
BRADYPREM study: is heart rate the most vital of all vital signs during preterm resuscitation? in "Selected abstracts of the 2nd Congress of joint European Neonatal Societies (jENS); Venice (Italy); October 31–November 4 2017; session "Neonatal pulmonology, neonatal respiratory support, resuscitation".
However, our population is substantially different, as they analyzed only infants less than 32 weeks’ gestation. In our study, most of the time spent on bradycardia was within the first minutes after birth. This finding might be explained with the slowly increasing oxygen tension during the initial aeration of the lungs, which are both known to be strong impulses for post-natal increment in HR.
We also found that only courses of SpO2 differed significantly between the bradycardia groups. On the contrary, courses of cerebral oxygenation parameters did not show significant differences. This finding further emphasizes that cerebral oxygenation was still preserved during periods of bradycardia. Moreover, such a situation potentially even allows preservation of oxygenation capacity for the heart to be able to increase HR and overcome bradycardia. However, our findings might take into account a slight compensation by increase in FiO2 delivered. Although courses of FiO2 did not differ significantly between the three bradycardia groups, there was a trend to higher FiO2 in the groups with bradycardia. Certainly, both degree of bradycardia and presence of hypoxemia trigger clinical decision to change FiO2, then course of FiO2 is biased by the clinical approach of the neonatologists.
Additional presence of hypoxaemia
In an individual patient analysis of 8 RCTs, Oei and co-workers found that if SpO2 80% was not reached within 5 min after birth, there was a 2-fold risk of death and increased morbidity, such severe IVH.
The authors concluded that whether these findings are due to the infants’ illness or to the amount of oxygen administered during stabilization remains unclear. In view of these findings, we integrated SpO2 at 5 min into our analysis, and we used it to define presence of hypoxemia. However, again our population is substantially different, as they analyzed only infants less than 32 weeks’ gestation. We found a significant impact of the combination of bradycardia and hypoxemia on all three oxygenation parameters and on FiO2 in our cohort. Particularly, only if there was a combination of bradycardia and hypoxemia, cerebral oxygenation dropped. We previously reported that the brain had the highest saturation levels in infants during uncomplicated fetal-to-neonatal transition, indicating a preference for oxygen delivery to the brain.
The underlying mechanisms are unknown. Further we showed that in preterm infants reduced oxygen delivery to the brain was not only associated with lower crStO2 values, but with changes in cerebral perfusion as well.
In healthy neonates, cerebral blood volume decreases significantly during fetal-to-neonatal transition, most probably due to cerebral vasoconstriction associated with the steep increase in blood oxygen tension.
In infants with need of respiratory support with a diminished increase in oxygen tension in the first minutes after birth, there is consecutively less change in cerebral blood volume, which may improve cerebral oxygenation by maintaining cerebral blood volume including oxygenated hemoglobin. Nevertheless, cerebral oxygenation was less challenged by degree of bradycardia as compared to additional presence of hypoxemia. No differences in course of cFTOE was seen when the neonate experience various degree of bradycardia alone. This implies that there was no increase in cerebral oxygen extraction during bradycardia to compensate for a significant change in oxygen delivery. On the contrary, additional presence of hypoxemia resulted in a significant increase of FTOE. Recently, it has been shown that during resuscitation cFTOE was a sensible marker displaying changes in cerebral oxygen delivery during return of spontaneous circulation in a newborn asphyxiated lamb model.
The present study emphasizes that the combination of both parameters is important and should be followed during neonatal resuscitation. It seems reassuring that cerebral tissue may be less challenged by bradycardia alone as compared to additional presence of hypoxaemia. This aspect strongly underlines the importance of a quick titration of FiO2 to ensure adequate oxygen delivery within the first minutes, although often challenging for clinicians.
Our analysis has some limitations. First, the uncertainty in the definition of bradycardia in contrast to the more structured definition of hypoxemia imposes caution in the interpretation of the results. Then, the present study is retrospective, and our findings should be confirmed by future prospective data. Our study included mainly low birth weight infants, then comparison with other studies is not appropriate especially in regard to clinical outcomes.
BRADYPREM study: is heart rate the most vital of all vital signs during preterm resuscitation? in "Selected abstracts of the 2nd Congress of joint European Neonatal Societies (jENS); Venice (Italy); October 31–November 4 2017; session "Neonatal pulmonology, neonatal respiratory support, resuscitation".
However, it is innovative in showing that the combination of bradycardia and hypoxemia is not only a threat for the most immature infants.
With the use of SpO2 and FiO2 we were only able to integrate supplemental oxygen and arterial oxygenation into our analysis, but certainly pCO2 might have influenced cerebral perfusion too. In this regard, we showed that in healthy preterm and term infants pCO2 levels were within normal ranges during neonatal transition.
In addition, we did not differentiate between different respiratory support modalities, which could have influenced our data. Furthermore, blood glucose levels are associated with crStO2 values,
but we did not integrate blood glucose levels into our model. In addition, crStO2 was measured with two different devices (INVOS 5100C and NIRO 200-NX), which provide systematically different values. However, in separate analysis these tools showed very similar results.
Our study shows that the degree of bradycardia has a significant impact on the course of SpO2 only, but when associated with the additional presence of hypoxemia a significant impact on cerebral oxygenation parameters is observed. Furthermore, the additional presence of hypoxemia has a significant impact on FiO2 delivered. Our study emphasizes the importance of HR and SpO2 during neonatal resuscitation, underlining the relevance of hypoxemia during the neonatal transitional phase.
Funding
None.
Conflict of interest
None.
CRediT authorship contribution statement
Ilia Bresesti: Conceptualization, Investigation, Data curation, Writing - original draft, Writing - review & editing. Alexander Avian: Methodology, Formal analysis, Data curation, Writing - review & editing. Marlies Bruckner: Investigation, Writing - review & editing. Corinna Binder-Heschl: Investigation, Data curation, Writing - review & editing. Bernhard Schwaberger: Investigation, Data curation, Writing - review & editing. Nariae Baik-Schneditz: Investigation, Data curation, Writing - review & editing. Georg Schmölzer: Conceptualization, Writing - review & editing. Gerhard Pichler: Conceptualization, Writing - review & editing, Supervision. Berndt Urlesberger: Conceptualization, Writing - original draft, Writing - review & editing, Supervision.
Appendix A. Supplementary data
The following is Supplementary data to this article:
BRADYPREM study: is heart rate the most vital of all vital signs during preterm resuscitation? in "Selected abstracts of the 2nd Congress of joint European Neonatal Societies (jENS); Venice (Italy); October 31–November 4 2017; session "Neonatal pulmonology, neonatal respiratory support, resuscitation".
Cerebral oxygen saturation to guide oxygen delivery in preterm neonates for the immediate transition after birth: a 2-center randomized controlled pilot feasibility trial.
Use of tissue oxygenation index and fractional tissue oxygen extraction as non-invasive parameters for cerebral oxygenation. A validation study in piglets.
00184-2&id=gr1.jpg){kind=link}
00184-2&id=gr2.jpg){kind=link}
00184-2&id=gr3.jpg){kind=link}