An approach to define newborns´ sniffing position using an angle based on reproducible facial landmarks

The neutral or sniffing position is advised for mask ventilation in neonates to avoid airway obstruction. As definitions are manifold and often unspecific, we wanted to investigate the reliability and reproducibility of angle measurements based on facial landmarks that may be used in future clinical trials to determine a hypothetical head position with minimal airway obstruction during mask ventilation.

congenital abnormalities like retrognathia, 13 or simply AO due to meconium stained fluid. 14 An incorrect head position during mask ventilation 9 might also contribute to airway obstruction.
The sniffing or neutral position, advised by European Resuscitation Guidelines as being the best position during neonatal mask ventilation, 15 was first described in 1895 16 as a neck flexion with upper cervical extension, 17,18 but without any further detail on exactly what to achieve. Variations of the sniffing position have been used by clinicians for years and are mainly based on expert opinion 19 ; yet a detailed definition of the angle used to unify its application is missing and may be a first step in addressing this omission. 20 Up to now, there is only one study in spontaneously breathing yet sedated neonates receiving magnetic resonance imaging (MRI) that showed that the sniffing position was best to open up the airway, defined by an angle at the back of the head based on landmarks, in particular the head tilt angle, which is the angle between the occipitoophisthion line and the ophisthion-C7-spine process line only visible on sagittal MR scans 21 (Figure 1). MRI, however, is time-consuming and expensive; thus, a simple, easily obtainable and reproducible angle is needed.
The aim of this trial was to objectively define infants head positioning using facial landmarks, since positioning of the head is the most important measure to enable ventilation of the lungs.
For this purpose, we determined facial landmarks and created a simple angle measurement using 2D images that are easily reproducible by different observers in different head positions. 22 This angle can be used in future studies to examine the variation among clinicians in trying to achieve an optimal head position and define the latter while avoiding airway obstruction during mask ventilation. 20 2 | MATERIAL AND ME THODS

| Study design and consent
This single-center, prospective observational study took place at Tuebingen University Children's Hospital, Germany. The institutional Ethics Committee approved the study, and written informed parental consent was obtained (approval number: 704/2017BO1).

| Patients, equipment, and outcome
Infants born between 34 and 41 weeks gestation were eligible within 48 to 72 hours of birth. Exclusion criteria were dysmorphic features or the need for respiratory support. Demographic details including gender, gestational age (GA), head circumference, mode of delivery, and birth weight were obtained from the hospital notes.

What is already known about the topic
• Sniffing position is performed for years and differs widely within clinicians.
• Relationship between head position and airway obstruction remains unclear.

What new information this study adds
• A standardized angle defining infant's head position.
• This study provides a new measuring technique to examine the variation among clinicians and further correlate head position with airway obstruction. Two lateral 2D images were taken using a Nikon D7100 camera (Nikon, Chiyoda, Tokio, Japan). A plastic scale (sliding millimeter caliper, Maped®, Cedex, France) was placed on the side of the infant's head as a measurement reference. ImageJ software (National Institute of Health; USA), a public domain, java-based image processing tool developed at the National Institutes of Health, was used for analysis. 23 We focused on the following two landmarks based on AM Schwarz's concept of facial profiles ( Figure 2 The five observers (physicians) had no experience in using ImageJ software or similar software or regarding special anthropology; they only received a short introduction to the software.
Each infant was photographed in two head positions: the "physiologic" and the "sniffing" position and the angle δ phys , respectively, δ sniff were calculated using the above-mentioned landmarks.  To evaluate whether GA influences angle δ, patients were divided into near-term (34 to 37 weeks' gestation) and term infants (≥ 37 weeks' gestation).

| Statistical analysis
A convenience sample of 24 infants was chosen for this pilot study.
An intraclass correlation coefficient with a one-way model was used to calculate intra-rater and inter-rater reliability, respectively.
A P-value (two-sided) of <.05 was considered to represent statistical significance.
Continuous data are shown as mean ±SD if normally distributed or median (IQR) if skewed. Mean differences are shown with 95% confidence intervals. Categorical data are summarized as counts and percentages. As data were normally distributed, Pearson's correlation coefficient was used. Normally distributed variables were analyzed using t test, otherwise the Mann-Whitney U test was used. For paired samples, the paired t test or Wilcoxon signed-rank test was used.
All statistical analyses were performed using SPSS (version 25, SPSS Inc, Chicago, Ill).

| DISCUSS ION
The correct head positioning might be one of the simplest ways to reduce or even avoid airway obstruction.

F I G U R E 3
Boxplots showing the angle δ in "physiological" and in "sniffing" position grouped in near-term and term infants skull or head) and porion 30 and is considered a reliable basis for clinical cephalometric analysis. 31 It has been used to describe sniffing position in spontaneously breathing adults and is angled at 70-80° to the horizontal plane of the underlying surface. 32 The infraorbital point for generating the FH would be covered by the rim of a facemask, while the porion as an anatomical landmark is constantly visible through a transparent facemask during mask ventilation.
We therefore aimed for a clearly visible anatomical landmark, while using the newly defined angle δ sniff in sniffing position by creating an orthogonale that is easy to estimate.
Our observational study may offer a method to unify the definition of the sniffing position and replacing nonmeasurable terms like "complete atlanto-occipital extension" 33 or "placing the infants with his/her eyes looking directly at the ceiling" or "with the nose pointed upward and forward" 25 by providing a detailed and reliable angle to describe and evaluate an infant's head position. Our new and clearly defined angle δ may be used to validate the "sniffing position" as being the best position during mask ventilation in newborns.
Obtaining the optimal head position for avoiding airway obstruction is fundamental for ventilation of the lungs.
One of the strengths of this study is that all raters had no previous experience in setting landmarks, indicating that the chosen landmarks are reproducible, and the calculated angle δ can be measured with high precision.
Since the aim of this study was only to determine landmarks that can be used to reliably define infants' head position, our study population should represent a rather healthy and normal population. We therefore did not include any patients with head shape anomalies or extreme premature infants. Moreover, there was no difference in angle measurements between near-term and term infants, which confirms once again the quality of the anatomical landmarks to determine angle δ sniff .
A limitation of our study is that there may be difficulties in seeing Limitations of our study include that the sniffing position was always performed by the same investigator, which might have contributed to the small standard deviation.
Nevertheless, we did not yet investigate different clinicians' performance in assuming a so-called "sniffing position," although it unquestionably varies greatly between clinicians. With the current evidence, it remains unclear which angle δ sniff will be the best to avoid airway obstruction during mask ventilation. We are therefore planning a clinical study to determine the correlation between angle δ sniff and the degree of airway obstruction during mask ventilation. At present, we can just speculate that both, head elevation and head extension, can decrease upper airway collapsibility during mask ventilation. Airway obstruction is common during ventilation in the delivery room during neonatal resuscitation. 6 Adverse events like airway obstruction can be identified by a calorimetric CO 2 detector 6 or a respiratory function monitor (RFM), which displays a characteristic AO-associated flow pattern 9 and might provide a guidance for finding the optimal angle δ sniff . A zero inspiratory or expiratory flow, 34 respectively, a zero/very small tidal volume (Vt) on the RFM during inflation, should be considered either as being due to an external (facemask pressure) or internal (head position) airway obstruction. 9 Measuring angle δ sniff is intended to describe an external airway obstruction but will remain a research tool providing valuable information about the optimal head position, which might lead to recommendations that are easy to perform during neonatal resuscitation.

| CON CLUS ION
A standardized angle defining an infant's head position can be used in future studies to examine the variation among clinicians in trying to achieve the optimal head position associated with minimal airway obstructions.

E TH I C S A PPROVA L
The research was conducted ethically in accordance with the World

Medical Association Declaration of Helsinki. The institutional Ethics
Committee of the University of Tuebingen approved the study (approval number: 704/2017BO1). All parents have given their written informed consent. Parental consent for publication of the photographs has been provided.

CO N FLI C T O F I NTE R E S T
The authors have no conflicts of interest to declare.

AUTH O R CO NTR I B UTI O N S
BH was involved in study design, patient recruitment, data collection and analysis and has written the manuscript. AMB was involved in study design, patient recruitment, data collection and analysis and has also reviewed and contributed to each draft of the manuscript. LS was involved in study design and has reviewed and contributed to each draft of the manuscript. CFP was involved in study design and contributed to each draft of the manuscript. BK did the project supervision and has seen and contributed to each draft of the manuscript. All authors participated in critical revision of the manuscript for important intellectual content. All authors approved the final manuscript as submitted and agree to be accountable for all aspects to the work.

DATA AVA I L A B I L I T Y S TAT E M E N T
In accordance with the "DFG Guidelines on the Handling of Research Data," we will make all data available upon request. The data set will be archived for at least 10 years after publication.