Effect of same dose varying concentration poractant alfa on outcomes in preterm infants under 32 weeks of age

Introduction / Введение

Surfactant replacement therapy is based on the pathogenesis of respiratory distress syndrome (RDS), which in part is to stabilize newborn respiratory function, where surfactant therapy is essential. Surfactant therapy is an essential part of the stabilization. Surfactant therapy with poractant alfa (PA) has been studied thoroughly in numerous clinical studies which involved dose finding and timing studies [1][2]. At the same time, the influence of some factors and drug characteristics (concentration and viscosity) on the effectiveness and clinical outcomes of newborn respiratory disorders have not been studied sufficiently. Of particular interest is the hypothesis that a high concentration of phospholipids determining exogenous surfactant viscosity may affect the pattern of drug distribution. A non-homologous distribution may negatively affect disease course and outcome.

According to K.  Cassidy et al. (2001) [3], J. Anderson et al. (2004) [4] and King D.M. et al. (2001) [5], the viscosity of a lung surfactant depends on its molecular composition, microstructure, interaction between components and environmental conditions. Viscosity was the major factor influencing a velocity, degree and uniformity of lung surfactant distribution. A relation between pulmonary surfactant viscosity and phospholipid concentration was non-linear, and increased viscosity was more pronounced at higher vs. lower concentrations. D.M. King et al. (2001) showed that rise in surfactant-related phospholipid concentration causes an intrinsically increased viscosity resulting in newly acquired properties of a non-Newtonian liquid [5], which flow is characterized by viscosity/velocity gradient relation. Usually, such liquids are highly heterogeneous and consist of large molecules [6]. By comparison, a Newtonian fluid may be of any origin that preserves fluid properties no matter what forces act on it, which is summarized in Newton's law of viscosity, exemplified by water. Viscous behavior of surfactant preparations became strongly non-Newtonian upon escalating phospholipid concentration, with larger viscosity level found at low shear rates [5]. Having acquired the properties of a non-Newtonian liquid, an exogenous surfactant, thus, has a lower fluidity, and pattern of its distribution in the respiratory tract depends on a number of physical factors, and above all on the forces of gas flow acting on the liquid during its movement.

Attempts to study the distribution of exogenous surfactant in the lungs coupled to its physical and chemical properties, were made repeatedly. F.F. Espinosa and R.D. Kamm (1998) tried in in vitro studies to simulate an effect of inertia forces, degree of surface tension and gravity on distribution of surfactant "plugs" [6]. It was concluded that injection rate, viscosity of the solution and tilt relative to gravity had the effect on emerging "meniscus". The latter was defined as the curve in the upper surface of a liquid next to the surface of another object caused by surface tension (Fig. 1).The degree of curvature of the "meniscus", in turn, determines a uniform drug distribution [6]. Using bifurcation models of the upper respiratory tract, Y. Zheng et al. (2005, 2006) [7][8], and then A. Copploe et al. (2019) [9] confirmed an impact of gravity, surface tension and inertia on the splitting of the surfactant "plug" in the area of dichotomous division of the respiratory tract (Fig. 1). Treatment effectiveness largely on uniformity distribution for the injected surfactant in the respiratory tract [7–9].

Surfactant distribution at different levels of the respiratory tract occurs due to various mechanisms. Тhe transport zone of the lungs is represented by a large number of dichotomically dividing airways, the diameter of which progressively decreases in the distal direction. The distribution at the tracheal bifurcation level is rapid and depends on surfactant volumetric viscosity, a lower value of which ensures a more uniform and rapid distribution [3][6][10]. The distribution of exogenous surfactant in the small-diameter airways and alveoli depends more on variation in surface tension differences for diverse fluids and components in the pulmonary system (Marangoni effect). Following fairly rapid transients, on the order of seconds, a steady-state transport develops being governed by interplay between Marangoni flow and alveolar kinetics [10]. In addition, the nature of the distribution underlies differences in surface viscosity. Bulk viscosity remains poorly understood.

In 1990s, N. Gilliard et al. found that increasing surfactant concentration approximately fourfold from 14.1 ± 1.89 to 60.0 mg/kg in animal experiments was not associated with improved surfactant distribution in the lungs. Such data on large volume administration can be considered as a positive control validating the method used to analyze the surfactant distribution at a macroscopic level [11].

We hypothesized that the lower surfactant viscosity at low concentration would improve distribution, but also increase therapy effectiveness. To study this, we developed a study to assess an effect of reducing the concentration (viscosity) for the administered exogenous surfactant (poractant alfa) preparation without changing the recommended dose to determine most optimal strategy for surfactant replacement therapy by comparatively analyzing the results depending on the exogenous surfactant concentration/viscosity relation.

Aim: to assess an effect of PA administered at low (40 mg/mL) vs. standard (80 mg/mL) concentration without changing recommended dosage (200 mg/kg) on outcomes of preterm infants at gestational age (GA) under 32 weeks receiving various respiratory support.

Materials and Methods / Материалы и методы

Study design / Дизайн исследования

А prospective randomized controlled multicenter study was conducted. A total of 325 infants under 32 weeks of GA were randomized in the five perinatal centers. After randomization and birth, the infants received РА at a dose of 200 mg/kg at minute 10 of age if indicated.

Stabilization of patients in the delivery room was carried out according to a unified protocol. Infant condition was assessed immediately after PA administration and during the first hour of life as follows: presence of side effects and their severity after drug administration. Lung damage within 72 hours manifested as air leakage syndrome, pulmonary bleeding, and severe hemodynamic disorders. The third end point was to assess outcomes of patients at the time of discharge from the hospital or death. This study was conducted to investigate an effect of low surfactant concentration during surfactant replacement therapy on lung outcomes. Efficacy was defined as a combination of reducing a risk of lung damage and minimizing pulmonary and extrapulmonary complications of RDS in preterm infants.

Inclusion and exclusion criteria / Критерии включения и исключения

Inclusion criteria: 1) informed parental consent before enrollment into the trial; 2) preterm infants born at one of the five perinatal centers; 3) GA at the time of delivery at least 32 0/7 (inclusive) weeks; 4) clinical need for respiratory therapy – non-invasive respiratory support or continuous positive airway pressure (CPAP), conventional mechanical ventilation (CMV), high frequency oscillatory ventilation (HFOV), oxygen therapy; 5) clinical indications for surfactant administration: for infants with birth weight of more than 1000 g – mean airway pressure (MAP) ≥ 7.0 cm H₂O and fraction of inspired oxygen (FiO₂) ≥ 0.4 to maintain the target oxygen saturation level of 90–94 %; for infants with birth weight less than 1000 g – MAP ≥ 7.0 cm H₂O and FiO₂ ≥ 0.3 to maintain the target oxygen level of 90–94 % in the absence of a pronounced respiratory activity (retracted compliant places of the thoracic cage above clavicles, suprasternal fossa and sternum); 6) exclusion of pneumothorax based on chest X-ray exam performed within the first two hours of life.

Еxclusion criteria: 1) absence of informed parental consent before enrollment into the trial; 2) absence of indications for surfactant treatment; 3) chromosomal and genetic anomalies discovered antenatally or after birth; 4) congenital malformations, discovered antenatally or after birth, which might be involved in developing respiratory or cardiovascular insufficiency; 5) history of using various surfactant preparations during patient’s treatment; 6) gross deviations from the research protocol.

As a result, from 325 preterm infants 61 subjects were excluded from the study for the following reasons: 28 children had no indications for surfactant administration at birth (in delivery room); 11 children had congenital malformations; two patients had birth injuries; two patients had acute renal failure and required peritoneal dialysis; one infant had hereditary metabolic disease (insufficiency of long-chain 3-hydroxyacyl-Co-A-fatty acid dehydrogenase – mutation NM_000182.4: c.1528G>C); 14 patients were excluded due to deviations from the study protocol, and in 3 cases the data were lost. The inclusion criteria were met by 264 children, which were randomized into study groups (Fig. 2).

Comparison groups / Группы сравнения

Parental consent was obtained, and randomization was performed in case of threat of the birth to a preterm infant with GA under 32 weeks. The patient randomization was carried out by pulling closed envelopes immediately before delivery after obtaining informed consent by the parents. In the form of the envelope the number of patient medical history, patient’s surname and doctor's surname, who performed randomization and surfactant administration, and depersonalized patient study identification number, which would subsequently be entered into the database, were recorded. The patient study identifiers were obtained by utilizing a random number generator and distributed among five perinatal centers. In case of medical indications for surfactant administration in the delivery room, a patient was included in the study. A minimally invasive method of surfactant administration was used in infants with spontaneous breathing. In case of no any indications for surfactant therapy, the envelope was drawn up in an identical manner and put in a separate folder.

Thus, two groups were formed and compared: Low concentration (LC) group – PA was administered in concentration 40 mg/mL (n = 111) and Standard concentration (SC) group (control) – PA was administered in concentration 80 mg/mL (n = 153). Additionally, we conducted a comparison in two subgroups with surfactant introduction by minimally invasive methods in spontaneously breathing infants (using LISA – a less invasive method of introducing surfactant through a thin catheter or endotracheal tube): subgroup LC – PA concentration was 40 mg/mL (n = 27) and subgroup SC (control) – PA concentration was 80 mg/mL (n = 34).

Study protocol / Протокол исследования

The study protocol assumed stabilization of the patients' condition using respiratory circuit with a T-piece connector and positive end expiratory pressure (PEEP) valve in delivery room. The minimal PEEP level was set within 6.0–8.0 cm H₂O, FiO₂ = 0.21–0.3 at the start. Blood oxygen saturation (SpO₂) was determined in the delivery room from birth using various monitors with Nellcor^TM pulse oximetry technology (MEDTRONIC, USA). Patient tracheal intubation and transfer to mechanical lung ventilation were performed to reach average MAP > 8.0 cm H₂O and FiO₂ > 0.4–0.5 to maintain the target SpO₂ level or absence of spontaneous respiration. At the age of 10 minutes of life, according to randomization data poractant alpha bolus was injected into the endotracheal tube (ETT) or using minimally invasive method of surfactant treatment (MIST), depending on infant’s condition and capacity to breathe spontaneously. Before the end of stabilization in the delivery room, the following parameters were evaluated: cases of endotracheal tube obstruction, level of SpO₂ decline, the minimum SpO₂ value until recovery, cases of bradycardia and magnitude of maximum bradycardia before recovery. After stabilization, patients were transferred to the neonatal intensive care unit (NICU).

The next stage included assessment and observation of patients at the age of the first 72 hours of life. The following main parameters and events that could indicate severe lung damage during surfactant therapy were monitored: respiratory parameters (maximum FiO₂, MAP, tidal volume), the presence of air leak syndrome, pulmonary hemorrhage as a result of lung injury. The equipment for respiratory therapy (from different manufacturers) was available in perinatal centers. The main goal was to keep control of the tidal volume and maintain it at the minimal levels.

Study scheme is presented in Figure 3.

The outcomes were assessed based on the following indicators: 1) infant’s condition immediately upon PA administration – signs of airways obstruction, bradycardia less than 100 beats per minute and its severity, decreased SpO₂ and intensity of SpO₂ decrease; 2) duration of invasive respiratory support; 3) overall duration of the respiratory therapy; 4) length of hospital stay; 5) the maximal values of following variables in the first 72 hours of life – MAP and FiO₂; 6) degree of manifested pulmonary injury (air leak syndrome, pulmonary hemorrhage); 7) neurological injuries – grade III intraventricular hemorrhage (IVH), periventricular leukomalacia (PVL); 8) bronchopulmonary dysplasia (BPD), defined as a need for supplemental oxygen or respiratory support by 36 weeks of post-conception age; 9) necrotizing enterocolitis (NEC), surgical stage; 10) hemodynamically significant patent ductus arteriosus (HSPDA); 11) retinopathy of prematurity, surgical stage; 12) death (only the deaths in the first hospitalization period were included).

Ethical aspects / Этические аспекты

The study was conducted in accordance with the ethical standards of the 1964 Helsinki Declaration of the World Medical Association, its subsequent amendments and comparable ethical standards. The study was approved by the Ethics Committee of the Yaroslavl State Medical University, Protocol No. 45 dated of April 22, 2021.

Statistical analysis / Статистический анализ

Data collection was performed by researchers filling out a depersonalized electronic database in order of priority using Microsoft 365 spread sheets. Statistical analysis was carried out using the free Python computing software environment (v.3.8): built-in functions from the Statsmodels.api and Scipy modules for statistical calculations. Quantitative variables were evaluated for compliance with the normal distribution using the Shapiro–Wilk test. Aggregates of quantitative indicators, which distribution differed from normal were described using the median (Me) and the lower and upper quartiles ([Q₁; Q₃]). We used the Mann–Whitney U-test to compare unrelated samples. In the case of describing quantitative variables with normal distribution, the data obtained were combined into variational series in which arithmetic averages (M) and standard deviations (SD) were calculated. While comparing average values in normally distributed sets of quantitative data, the Student's t-test was calculated. The data of quantitative characteristics are expressed as absolute numbers and relevant proportions (%). The comparison of nominal group data was performed using the Pearson χ² test. In the case a number of expected observations in any of the cells in the four-field table was less than 10, Fisher’s exact test was used to assess a significance level. A relationship of features was studied based on correlation analysis (Spearman test). The differences between values were considered statistically significant at p ≤ 0.05.

Results / Результаты

Preparation of surfactant solution / Приготовление раствора сурфактанта

We used an exogenous surfactant – Curosurf^® (poractant alfa licensed through Chiesi Farmaceutici, Parma, Italy). 153 preterm infants were included into SC group receiving PA at dose of 200 mg/kg, concentration 80 mg/mL. Indications for surfactant replacement therapy were in accordance with the 2016 clinical guidelines “Management of newborns with respiratory distress syndrome” at a dose of 200 mg/kg [12]. PA was supplied in ready-to-use, single-use vials containing 1.5 mL (120 mg phospholipid) of surfactant. Derived from minced porcine lung, PA is a creamy white, sterile suspension containing the surfactant in normal saline at a concentration of 80 mg/mL phospholipid. PA is prepared by chloroform-methanol extraction, and liquid-gel affinity chromatography, contains ~ 99 % polar lipids [13]. 111 preterm infants were included into LC group, who met the inclusion criteria and were given the first PA dose of 200 mg/kg (repeat dose 100 mg/kg), concentration of 40 mg/mL in the delivery room. The repeated surfactant administration was carried out in drug concentration given first time.

The surfactant was administered in a standardized manner: invasively (through the endotracheal tube in mechanically ventilated babies), minimally invasive in spontaneously breathing babies by LISA (Less Invasive Surfactant Administration) or using the INSURE strategy (INtubation–SUrfactant – [immediate] Extubation). In our previous study with autopsy material, we measured trachea length in newborns with extremely low birth weight (ELBW) comprising 34.0 ± 5.0 mm from vocal cords to tracheal bifurcation [14]. Thus, we have determined the optimal depth of catheter insertion as 15–20 mm below the vocal cords. Surfactant injection by the LISA method was performed through a thin catheter (4–5 Fr) inserted under direct laryngoscopy control without using Magill forceps, along with non-invasive continuous maintenance of constant CPAP of at least 6.0 cm H₂O via mononasal tube or binasal cannulas using a device with a T-shaped connector [15]. The surfactant administration duration was 60–180 seconds controlled by attending neonatologist. The respiratory support strategy was determined by patient’s clinical needs, because randomization was performed before infant’s birth. In each group we analyzed the data based on mechanical lung ventilation and spontaneous ventilation. A separate analysis was carried out in the spontaneous ventilation subgroups of infants who received surfactant by the LISA and INSURE methods: subgroup LC (n = 27) – surfactant at a concentration of 40 mg/mL, subgroup SC (n = 34) – surfactant at a concentration of 80 mg/mL.

The diluted PA concentration was prepared dissolving surfactant preparation in 0.9 % sodium chloride solution at 1:1 ratio. For instance, to prepare poractant alfa 120 mg or 1.5 mL, there was added 1.5 mL of 0.9 % sodium chloride solution. Surfactant dilution with 0.9 % sodium chloride solution was performed directly in the syringe, by adding surfactant first followed by mixing. As a result, the total volume of a singular dose of 200 mg/kg was doubled in comparison to standard concentration.

Analysis of physical properties of surfactant preparations / Анализ физических свойств препаратов сурфактанта

PA drop at standard concentration and drop diluted two times with 0.9 % sodium chloride solution were placed on a glass slide from a height of 1.0 cm using a 1.0 mL syringe. The glass slide was pre-cleaned and degreased. The photo was taken under side illumination on a Nikon D750 camera with a Nikkor AF-S ED 18-300 mm VR lens with a focal length of 300 mm and a minimum distance to the object. The contact angles of wetting of surfactant solutions were measured using graphic computer software while processing photographs. Micrographs were taken with an iPhone 8 camera with adapter and a Nikon Eclipse Ci-S direct light microscope. The prepared standard and diluted surfactant drops were placed on a glass slide to take a photo 1.0 minute later at room temperature (Fig. 4).

Comparison of surfactant dynamic viscosity / Сравнение динамической вязкости сурфактантов

Figure 4 shows the spreading effect of drops at standard (80 mg/mL) and reduced (40 mg/mL) РА concentration. In the latter, the angle between the diluted surfactant and the surface was much smaller.

In experimental work, the shear viscosity of Survanta, Curosurf and diluted Curosurf shows a wide range of values and dependence on shear rate at 22 °C (Fig. 5).

Baseline clinical and anamnestic data of preterm infants / Основные клинико-анамнестические данные недоношенных детей

The analysis of baseline clinical and anamnestic data of preterm infants revealed no differences in all examined characteristics, excepting solely operative delivery frequency, which in LC vs. SC group turned out to be significantly higher (Table 1).

The full course of steroid prophylaxis of RDS was carried out in 60–64 % of all cases in both groups. No differences in birth weight and GA at birth as well as Apgar score and gender distribution were found in both groups.

Surfactant use in preterm infants / Результаты применения сурфактанта у недоношенных детей

The first РА administration was performed applying different methods and, as a standard operating procedure, in the delivery room at the first 10 minutes of life (Table 2).

As shown from Table 2, in both groups, the invasive method of surfactant administration was prevalent, which was based on the severity of the condition of the children and by the presence of indication for tracheal intubation and subsequent invasive respiratory therapy in the delivery room. The least invasive РА administration in each of the groups was carried out to almost every fourth child who was also spontaneously breathing. At the same time, the analysis of the development of the signs of airway obstruction during the surfactant administration (bradycardia less than 100 beats per minute and its severity, a decrease in SpO₂ below 85 % and a minimum value of SpO₂ decrease) did not reveal any statistically significant differences between the groups of mechanically ventilated and spontaneously breathing babies (Table 3).

Table 3 shows the short-term effects of standard and diluted surfactant concentration during the first minutes after administration. No significant differences were found in both groups. The increase in volume of the administered surfactant also had no effect on the oxygen demand as well as MAP and tidal volumes at the first 72 hours of life (Table 4).

The level requiring CMV and/or HFOV repeated surfactant administrations as well as duration of invasive respiratory support are presented in Table 5.

As shown in Table 5, the vast majority of children in both groups required invasive respiratory therapy, whereas every 4^th–5^th child required HFOV use. The need for repeated surfactant administration and the age of repeated administration did not differ between the groups with different surfactant concentration and volume.

To avoid potential effect of a congenital infection on the study data, we comparatively analyzed inflammation markers in preterm children at the first 72 hours of life (Table 6).

Table 6 shows the markers of inflammation primarily C-reactive protein (CRP) finding no significant differences between the LC and SC groups (p ≥ 0.05). However, it tended to decrease based on the approach assuming that infectious markers are more pronounced in the LC group. Apparently, the number of infants with thrombocytopenia was significantly higher in the group with the surfactant concentration of 40 mg/mL allowing to consider it as a risk factor of increased hemorrhagic complications. However, according to the data presented in Table 7, pulmonary hemorrhages developed at significantly higher level after receiving surfactant at concentration of 80 mg/mL.

In order to assess a relationship between pulmonary hemorrhage and other clinical and laboratory characteristics, we performed a statistical analysis using Spearman’s rank correlation coefficient that uncovered a direct correlation between pulmonary hemorrhage and HSPDA (r = 0.318; p < 0.001). Not only did pulmonary hemorrhage prolongs the length of stay in the NICU (r = 0.488; p < 0.001), but it also increased a risk of death (r = 0.507; p < 0.001). At the same time, the incidence of IVH grade III, PVL, BPD, pneumothorax and lethal outcomes was comparable in both groups.

An analysis of the clinical data from spontaneously breathing preterm infants in LC and SC subgroups demonstrated no significant inter-group differences for all signs presented in Table 8, which allowed the comparison of the discussed subgroups.

The response to surfactant administration and outcomes in both subgroups are presented in Table 9.

Comparing the two subgroups from all spontaneously breathing patients in our study who received noninvasive respiratory support we obtained a significant difference in ventilatory support duration and BPD development: 142.0 [ 70.0; 219.0] hours vs. 250.5 [ 141.0; 690.0] hours (p = 0.008) and 1 (4.0 %) vs. 10 (29 %) (p = 0.009) respectively.

In the two subgroups with surfactant treatment without mechanical lung ventilation there were revealed significant differences in outcomes depending on PA concentration. In patients with PA concentration of 40 mg/mL, the total duration of respiratory support was significantly shorter, while the incidence of BPD was significantly lower, which probably influenced overall duration of hospitalization both in the NICU and in the hospital, because it was significantly shorter in the LC subgroup. No difference in the bradycardia incidence and drop-in blood oxygen saturation in response to surfactant were found between the groups, i. e., an increase in PA volume did not lead to a reduced tolerance.

Discussion / Обсуждение

We studied the effects on respiratory support, (serious) adverse events and short-term outcomes in two groups of preterm infants after receiving similar total surfactant dose but split into varying concentrations and thus different volumes. Surfactant volume was doubled while lowering the concentration by 50 %. We did not encounter technical problems in administering the larger volume, even to spontaneously breathing babies. We did not find a negative effect of diluted surfactant on outcomes in preterm infants with GA under 32 weeks. Moreover, a significantly increased risk of developing pulmonary hemorrhage was found in the case of surfactant used at concentration of 80 mg/mL (SC group) vs. LC group.

Each drug treatment depends on the five essential questions of what drug at which dose at which time to which patient via which route. Surfactant replacement therapy was the first therapy deliberately developed for preterm babies. The introduction of the surfactant replacement therapy is an example of how dose and timing were prospectively tested [1]. We were however unable to find publications comparing the effect of different surfactant concentrations within same total dose on infants’ outcomes. We now studied a reduced surfactant concentration. The decrease in concentration was achieved by increasing the surfactant volume however maintaining the dose. The results of our study demonstrated that there was no increase in unfavorable outcome incidence regarding the volume of surfactant administered when it was doubled. To our knowledge, this is the first study to prospectively compare the same surfactant dose given with different concentrations in preterm babies.

We found a decline in pulmonary hemorrhages in mechanically ventilated babies treated with a lower concentration. The increased risk of pulmonary hemorrhage with patent HSPDA has been described in numerous studies [19–21]. We noted no HSPDA difference between the groups. Pulmonary hemorrhages could have been triggered by congenital infectious processes, however no significant differences between such risk factors in the groups were found. Moreover, the LC group tended to have a higher incidence of CRP concentrations exceeding 6.0 mg/L, therefore we suspect more congenital infectious processes, which could predispose, among other things considered, to developing pulmonary hemorrhage. Thus, we have no obvious explanation for the increased frequency of pulmonary hemorrhages in our study other than the difference in effect of PA doses 80 mg/mL vs. 40 mg/mL. We speculate that perhaps the more viscous surfactant, being distributed less evenly in the lungs, contributed to a more mosaic distribution in the lungs, their opening and the involvement of the alveoli in the gas exchange process. This perfusion-ventilation mismatch can lead to a higher risk of lung tissue damage in the most extended and possibly even compensatory over-inflated areas.

In our study HSPDA incidence in preterm infants with GA under 32 weeks did not exceed that in other studies. Thus, S. Ngo et al. (2017) [22] showed that the 2014 incidence of patent ductus arteriosus (PDA) in 134 hospitals in California (USA) was 38.5 %, whereas only 15.7 % cases required pharmacological treatment, which is comparable with our data – 16.3–17.1 %, respectively. In another observational study, J.I. Hagadorn et al. (2016) showed that over a 10-year period there were recorded 42 % of patients weighing less than 1500 g with PDA, of whom 74 % required pharmacological or surgical treatment [23]. The same was found with the incidence of IVH grade III (ICD-10 – P52.2), which is comparable with the data by P. Härkin et al. (2021) [25] for infants with GA under 28 weeks and is 15.7 % (in our study 13.7–16.2%) as well as in study by X. Kong et al. (2016) for patients up to 31 full weeks – 15.4% [25]. BPD incidence when defined as the need for oxygen and/or respiratory support by 36 weeks of post-conceptual age in our study is virtually equal to BPD incidence presented in the review by D. Hines et al. (2017), and ranges from 15 to 55 % (in our study – 35.1–39.2 %) [26].

Another cause of pulmonary hemorrhage may be the plug formation in the airways occurring when externally instilled liquid obstruction form and are transported through the flexible airway networks. Plug propagation in airway branches, distribution in the lung, and resulting wall stresses may affect therapeutic efficiency and even cause additional lung injury [27]. It is thus of great importance to study the mechanisms of liquid plug propagation in theory and practice. This will be a challenge because a 2021 neonatologist’s survey in the Russian Federation described that most of them were afraid to use «large» surfactant volumes in ELBW patients [28].

The use of less invasive respiratory support methods in our study reduced BPD risk by 50 %. However, it was a surprise for us that this low PA concentration (40 mg/mL) significantly reduced BPD incidence in preterm infants with GA under 32 weeks in case PA was administered in a minimally invasive manner. Perhaps, in our study, this is due to a decrease in the duration of respiratory support allowing to reduce a risk of lung damage during respiratory therapy. As a result, such an improvement led to a faster discharge from hospital. Unfortunately, it is not possible to compare these findings with other similar clinical studies.

Considering the positive effect of a diluted surfactant, we can highlight several directions for studying this phenomenon. First, a change in surfactant physical properties (viscosity, fluidity) can contribute to a more even drug distribution in large-caliber bronchi, as well as in small bronchioles (Marangoni effect) due to a less prominent properties of a non-Newtonian fluid under gravity conditions. Secondly, the reduced surfactant concentration may promote better spreading, which leads to a more even distribution. The low viscosity in the diluted surfactant contributed to a greater drug penetration into the small airways and better filling of the bronchioles. The increased aliquot volume also helps to ensure good surfactant distribution.

Third, D. Luca et al. (2014) recommended adding 10–15 % to the required surfactant dose due to the inevitable reflux-related surfactant loss during minimally invasive administration [29]. In our study, we did not change the dose and kept it at 200 mg/kg. If there was a recommendation or need to increase a dose due to preparation losses, then dilution would allow to moderately compensate for it.

Summarizing, low PA concentration at 40 mg/mL in a total dose of 200 mg/kg did not negatively affect the outcome results in preterm infants with GA under 32 weeks. A twofold increase in administered PA volume did not affect PA tolerance, did not increase a risk of developing hypoxemia and bradycardia and relevant severity immediately upon drug administration, and did not aggravate respiratory and neurological outcomes. Moreover, minimally invasive administration of a surfactant with a PA concentration of 40 mg/mL reduced the risk of BPD probably also lowering a need for respiratory therapy. Poractant alfa at a concentration of 80 mg/mL may predispose to an increased risk of pulmonary hemorrhage during mechanical lung ventilation.

Conclusion / Заключение

PA use at a concentration of 40 mg/mL without changing the recommended dose did not aggravate nursing preterm infants with GA under 32 weeks. With minimally invasive PA administration at concentration of 40 mg/mL, risk of bronchopulmonary dysplasia decreased, and when used in infants on mechanical lung ventilation, it lowered a risk of pulmonary hemorrhage. Hence, the findings discussed require to further conduct large prospective, multicenter, randomized studies enrolling large patient cohorts.