Invasive ventilation at the boundary of viability: A respiratory pathophysiology study of infants born between 22 and 24 weeks of gestation

Background: Invasive ventilation of infants born before 24 weeks of gestation is critical for survival and long-term respiratory outcomes, but currently there is a lack of evidence to guide respiratory management. We aimed to compare respiratory mechanics and gas exchange in ventilated extremely preterm infants born before and after 24 weeks of gestation. Methods: Secondary analysis of two prospective observational cohort studies, comparing respiratory mechanics and indices of gas exchange in ventilated infants born at 22 – 24 weeks of gestation ( N = 14) compared to infants born at 25 – 27 weeks ( N = 37). The ventilation/perfusion ratio (V A /Q), intrapulmonary shunt, alveolar dead space (V Dalv ) and adjusted alveolar surface area (S A ) were measured in infants born at the Neonatal Unit of King ’ s College Hospital NHS Foundation Trust, London, UK. Results: Compared to infants of 25 – 27 weeks, infants of 22 – 24 weeks had higher median (IQR) intrapulmonary shunt [18 (4 - 29) % vs 8 (2 – 12) %, p = 0.044] and higher V Dalv [0.9 (0.6 – 1.4) vs 0.6 (0.5 – 0.7) ml/kg, p = 0.036], but did not differ in V A /Q. Compared to infants of 25 – 27 weeks, the infants of 22 – 24 weeks had a lower adjusted S A [509 (322-687) vs 706 (564 - 800) cm 2 , p = 0.044]. The infants in the two groups did not differ in any of the indices of respiratory mechanics. Conclusion: Ventilated infants born before 24 completed weeks of gestation exhibit abnormal gas exchange, with higher alveolar dead space and intrapulmonary shunt and a decreased alveolar surface area compared to extreme preterms born after 24 weeks of gestation.


Introduction
Neonatal survival following extremely premature birth (less than 28 weeks of gestation) has consistently improved over the past decades, with 81% of extreme preterms surviving to 36 weeks postmenstrual age in a recent period [1].Markedly improved outcomes have been reported in infants born before 24 weeks of gestation, who until recently were thought to be pre-or periviable [2].Survival has been reported even at 21 weeks of gestation [3] calling for the issuing of recommendations on how to actively manage this vulnerable group [4].
The improved survival of extremely preterm infants is partly explained by practices such as the administration of antenatal steroids, the initiation of non-invasive respiratory support from birth, early surfactant administration, caffeine therapy and avoidance of invasive mechanical ventilation when possible [5].Invasive ventilation has been shown to be injurious to the immature developing lung and thus non-invasive support is preferentially provided [6,7].
J o u r n a l P r e -p r o o f birth and remain ventilated for a long period of time [8].We recently reported that the median duration of ventilation in the whole population of extreme preterms in England was ten days [1].Within this population, the median duration of ventilation was five times longer in infants below 24 weeks of gestation (36 days) compared to extreme preterms of 25-27 weeks (7 days).This observation calls for a different approach to ventilation below 24 weeks, aiming towards growth while gently ventilated, rather than early extubation and subsequent growth on non-invasive support [2].
Although centres and networks with favourable outcomes have reported their experience and practice in periviable extreme preterms [2,[9][10][11], there is little data to substantiate these practices, either in the form of randomised trials or physiological clinical studies.It is not certain whether differences in respiratory parameters between term and preterm infants, such as a higher physiological dead space in preterm compared to term [12], would remain present between preterms of 22-24 weeks compared to preterms of 25-27 weeks.Such potential differences might influence the ventilatory approach, which in turn might affect the clinical outcomes of these infants.We have recently undertaken two studies investigating composite aspects of respiratory physiology, which included infants born between 22-24 weeks of gestation.
The aim of this study was to explore differences in respiratory mechanics and gas exchange between extreme preterms of 22-24 weeks of gestation compared to infants of 25-27 weeks.

Primary Studies
The data from infants included in two prospective, observational studies of ventilated infants undertaken at the Neonatal Intensive Care Unit of King's College Hospital (KCH) NHS

J o u r n a l P r e -p r o o f
Committee and registered with the Clinical Governance Department at KCH [13,14].
A respiratory function monitor [NM3 respiratory profile monitor (RPM; Philips Respironics, Wallingford, Connecticut, USA] was used in both studies.The monitor was connected to a laptop with customised Spectra software (3.0.2.6; Grove Medical, London, UK).The NM3 RPM had a combined carbon dioxide, pressure and flow sensor, which was placed between the endotracheal tube and the ventilator circuit.

a. Respiratory mechanics
The first study recruited mechanically ventilated preterm infants without congenital anomalies studied when they were clinically stable and ready for extubation.The study was performed between February and August 2016 and assessed respiratory muscle function at the end of a 10-min spontaneous breathing trial to predict the outcome of extubation in ventilated preterm infants [13].The main index was the respiratory muscle time constant of relaxation (tau), which was calculated from the reciprocal of the slope of the decline in airway pressure as a function of time [15].The study reported that tau was significantly greater in infants whose extubation failed and predicted extubation outcome with an area under the receiver operator characteristic curve of 0.937 [13].The peak airway pressure (Peak P aw ) generated at the end of the spontaneous breathing trial was also measured.The time to peak tidal expiratory flow as a proportion of the total expiratory time (T PTEF /T E ) was assessed as a measure of abnormal airway function [12].Compliance of the respiratory system (C RS ) was calculated during assisted ventilator inflations by dividing the expiratory volume by the peak pressure minus the PEEP of the previous inflation and expressed in ml/cmH 2 O. Resistance of the respiratory system (R RS ) was calculated from the applied pressure divided by the peak expiratory flow and expressed in cmH 2 O/l/s [16].

J o u r n a l P r e -p r o o f
The second study was performed between October 2020 and January 2022 and prospectively assessed ventilated extremely preterm infants without congenital anomalies at one week after birth.In brief, paired measurements of the fraction of inspired oxygen (F I O 2 ) and transcutaneous oxygen saturation were used to calculate the ventilation/perfusion ratio (V A /Q) and right to left shunt using the oxyhaemoglobin dissociation curve [17].The shunt was intrapulmonary in origin, as cardiac right-to-left shunting was excluded by echocardiography.
The physiological (V Dphys ), anatomical (V Dana ) and alveolar (V Dalv ) dead spaces were measured using volumetric capnography and expressed in ml/kg of body weight.The above indices were combined to estimate the adjusted alveolar surface area (S A ) which was lower in the infants who required home oxygen compared to those who did not [14].

Respiratory management
Antenatal corticosteroids and active resuscitation was offered to infants above 22 weeks of gestation following risk assessment and multi-professional discussion with parents [4].
Resident neonatal consultants provided cover throughout the day and night, seven days a week during both studies.Endotracheal intubation was performed with Cole's shouldered endotracheal tubes, which minimise leak.All intubated extreme preterms received surfactant during initial stabilisation at birth.Infants were ventilated using volume-targeted ventilation at 5-7 ml/kg [11], with a backup rate of 40-50 inflations/min and inflation times of 0.35-0.40sec with the SLE 6000 neonatal ventilator (SLE, Croydon, UK).Caffeine was started within the first six hours of life [5].

J o u r n a l P r e -p r o o f
Extremely preterm ventilated infants from the above studies were separated in two groups: The ones born between 22-24 completed weeks of gestation and ones born between 25-27 completed weeks of gestation.The following respiratory variables were included from the first study: tau, Peak P aw , T PTEF /T E , C RS , R RS and from the second study: V A /Q, shunt, V Dphys , V Dana , V Dalv , adjusted S A , expiratory tidal volume (V TE ) and physiological dead space to tidal volume ratio (V D /V TE ).The following variables were also collected from both cohorts: full course of antenatal corticosteroids (yes/no) [18], sex (male/female), gestational age (weeks), birth weight (kg), F I O 2 at study, mean airway pressure (MAP) at study, age at study (days), postmenstrual age (PMA) at study, haematocrit at study (%).

Statistical analysis
Sample size calculations were performed for both completed prospective studies [13,14].The current study was an exploratory analysis of existing datasets aimed at generating hypotheses, and a formal sample size calculation was not performed.The data were tested for normality using the Shapiro-Wilk test and found to be non-normally distributed.Data were thus presented as median and interquartile range (IQR).The Mann-Whitney U test was performed to determine if differences in respiratory parameters were statistically significant between infants born at 22-24 weeks of gestation compared to the ones born at 25-27 weeks.
Statistical analysis was performed using SPSS software version 28 (SPSS Inc., Chicago, IL).
J o u r n a l P r e -p r o o f

Results
A total of 51 infants were included from both studies, 14 were born below 24  The infants in the two groups did not differ significantly in any of the respiratory parameters: tau, Peak P aw , T PTEF /T E , C RS and R RS (Table 1).

b. Gas exchange
Twenty-five infants were included: six were 22-24 weeks of gestation and 19 were 25-27 weeks of gestation.The infants of 22-24 weeks did not differ significantly from the more mature infant in terms of antenatal steroids, sex, oxygen and MAP requirement, age and haematocrit at study (Table 2).The infants of 22-24 weeks had higher right-to-left shunt [18(4-29) %] compared to infants of 25-27 weeks [8(2-12) %, p=0.044,Table 2, Figure 1].The infants of 22-24 weeks had higher V Dalv (p=0.036) and V D /V TE (p=0.049)compared to infants of 25-27 weeks but did not differ in V Dphys , V Dana and V A /Q (Table 2).The infants of 22-24 weeks had a median (IQR) adjusted S A of 509(322-687) cm 2 , which was significantly smaller compared to the adjusted S A of infants born at 25-27 weeks [706(564-800) cm 2 , p=0.044] J o u r n a l P r e -p r o o f

Discussion
We have demonstrated that infants born at 22-24 weeks of gestation exhibit significantly impaired gas exchange compared to their counterparts of 25-27 weeks gestation, as evidenced by a smaller alveolar surface area and a higher intrapulmonary shunt and alveolar dead space.We also demonstrated that in our population, which was measured preextubation, respiratory mechanics were not significantly different in infants of 22-24 weeks of gestation compared to the ones of 25-27 weeks.
The evolution of the relationship of respiratory dead space with age can only be suspected by previous studies which did not involve periviable infants.We have previously described that the anatomical dead space (corrected for body weight) was higher in preterm compared to term infants [19] while respiratory mechanics have long been known to be abnormal in preterm infants [20].During childhood, the total anatomic dead space (in ml/kg) has been reported to decrease with age and was more than 3 ml/kg in early infancy [21].We have also described that the alveolar surface area was smaller in premature infants compared to term ones, using functional methods as employed in the current paper [22].It is difficult to compare our V A /Q results with previously published data, as there is no effective ventilation before birth and perfusion refers to the fetal lung which is a non-gas-exchanging organ.
We should note that although in our study some indices did not differ between the groups, they were markedly abnormal in both groups.For example, the median physiological dead space was between 5.5-6 ml/kg in both groups, which is substantially higher than the reported dead space in term ventilated infants of 1.5 ml/kg [12].Similarly, the V A /Q was 0.52 and 0.53 in the two groups, which is markedly decreased compared to 0.95 in healthy term infants, measured using the same methodology [23,24].In the respective primary studies we reported tau values below 10 sec/cmH 2 O in term infants [25] compared to our current median values of 13-17 sec/cmH 2 O in all extreme preterms.We had also reported a median T PTEF /T E of 0.40 in term infants [12] compared to 0.21-0.28 in this study.The above results describe severely J o u r n a l P r e -p r o o f abnormal respiratory muscle and airway function in all extreme preterms compared to term infants.
Our methodology treated the lungs as a single unit and could not account for regional differences and heterogeneity of lung disease, which is common in extreme preterms with areas of hyperinflation and atelectasis.Lung disease heterogeneity could possibly be further elucidated by imaging studies such as lung MRI, which, however, have significant practical and methodological limitations in extremely preterm infants.Assessment of respiratory mechanics was undertaken by measurement of C RS and R RS , newer techniques however, such as forced oscillometry, might also capture such impairment and have recently been shown to predict the development of bronchopulmonary dysplasia [26].
The main area we reported significant differences in 22-24 weeks compared to 25-27 weeks was in gas exchange, with the smaller infants exhibiting higher intrapulmonary shunt and alveolar dead space and a smaller alveolar surface area.The observation that both ventilation (alveolar dead space) and perfusion (intrapulmonary shunting) were affected, may explain why the V A /Q was not significantly different between the two groups, but refers to an overall smaller total alveolar surface area for gas exchange (Figure 1).On the basis of the above, we can speculate why high-frequency or volume-targeted ventilation is beneficial in these infants, as a strategy which limits volutrauma, but might also assist in lung recruitment while avoiding perfusion abnormalities which could occur with overdistention.Also, the fraction of wasted ventilation (V D /V TE ) was higher in 22-24 weeks compared to 25-27 weeks, which might also explain the success of high frequency ventilation where tidal ventilation is less important for gas exchange [3].
Future work could aim to identify optimal ventilation settings which could optimise ventilation to perfusion matching while maximising both ventilation and perfusion.Optimal ventilation could be achieved by recruitment manoeuvres assisted by newer technologies to quantify lung aeration such as electrical impedance tomography [27].Optimal perfusion could be assessed J o u r n a l P r e -p r o o f by the effect of different distending pressures on artery velocity profiles assessed by Doppler echocardiography or non-invasive monitoring such as bioelectrical impedance [28].It is plausible that an enhanced level of ventilation fine-tuning, as we describe in this paper, could be a deciding factor for the survival of some of these infants, as the functional respiratory reserves of periviable infants are limited.It is also possible that optimisation of care through knowledge gained from these extreme cases could improve neonatal respiratory care across all gestations and maybe in different pathologies.
This study has strengths and some limitations.To our knowledge, this was the first study to explore differences of respiratory pathophysiology between extreme preterms before and after 24 weeks of gestation.We used a modest cohort of 51 infants but performed an extensive assessment, rarely undertaken even above periviable gestations.As a retrospective analysis, this study has limitations relating to the design of the primary studies.For example, our infants from the respiratory mechanics study were measured before extubation when deemed ready for extubation.As such, these findings might not translate to the earlier phases of the respiratory distress syndrome.We believe however, that the paucity of published evidence makes our results useful as a possible starting point in understanding the respiratory pathophysiological differences in these very low gestations.Given our modest population size and the multifactorial origin of respiratory disease following extremely preterm birth, we selected to undertake an uncomplicated comparison analysis and did not embark in multivariable models to account for the possible interaction of some interrelated parameters.
In conclusion, we reported that ventilated infants below 24 weeks of gestation exhibited abnormal gas exchange, with higher alveolar dead space and intrapulmonary shunt and decreased alveolar surface area compared to extreme preterms above 24 weeks of gestation.These findings could form the basis for future studies aimed to optimise the respiratory care of these infants.b: In preterms of 25-27 weeks of gestation, increased shunt and V Dalv cause a mismatch of V A /Q and a smaller overall S A , c: In preterms of 22-24 of weeks, an even more increased shunt and V Dalv correspond to a more decreased S A , the V A /Q ratio however is preserved.
J o u r n a l P r e -p r o o f J o u r n a l P r e -p r o o f completed weeks of gestation.a. Respiratory mechanics Twenty-six infants were included: eight were 22-24 weeks of gestation and 18 were 25-27 weeks of gestation.The infants of 22-24 weeks were studied at a median (IQR) age of 59 (40-68) days, significantly later compared to infants of 25-27 infants [12 (5-42) days, p=0.016].The infants of 22-24 weeks had a higher MAP at study[9 (8-10)  cmH 2 O] compared to infants of 25-27 weeks [8 (7-9) cmH 2 O, p=0.029].

Figure 1 :
Figure 1: Schematic depiction of the ventilation to perfusion (V A /Q) relationships and the alveolar surface area (S A ) in ventilated infants.a: In healthy term infants, the ventilation is matched to perfusion and there is minimal right-to-left shunt and alveolar dead space (V Dalv ).

Table 1 :
Respiratory mechanics -Demographic and respiratory function parameters in infants of 22-24 weeks gestation compared to infants of 25-27 weeks.Data are presented as median (IQR) or N (%).Fraction of inspired oxygen, MAP: mean airway pressure, PMA: postmenstrual age, Peak P aw : peak airway pressure, T PTEF /T E : The time to peak tidal expiratory flow as a proportion of the total expiratory time.