2025 年 89 巻 4 号 p. 500-508
Background: Recently, the role of a rapid increase in serum osmolality in the inhibition of postnatal ductal closure has garnered attention. This study evaluated the efficacy of high-humidity care in preventing the onset of patent ductus arteriosus (PDA) in extremely premature infants.
Methods and Results: The high-humidity group (HHG) comprised 28 infants (240 to 276 weeks gestational age) recruited prospectively within 6 h after birth between July 2019 and September 2021; these infants were cared for in 90% humidity for the first 72 h of life. The incidence of PDA within the first 7 days of life and the rate of increase in serum sodium concentrations were compared between the HHG and a conventionally managed historical control group (CG; 29 infants born in 2016–2017). Twelve (43%) infants in the HHG and 22 (76%) in the CG developed PDA (P=0.016). Multivariate logistic regression analysis revealed that high-humidity care was effective in reducing the incidence of PDA onset (odds ratio 0.265; 95% confidence interval 0.078–0.907). The rate of increase in serum sodium concentrations was significantly lower in the HHG than CG (median 0.29 [interquartile range 0.21–0.39] vs. 0.46 [interquartile range 0.32–0.62] mEq/L/h, respectively; P<0.001).
Conclusions: High-humidity care for the first 72 h of life may help reduce the onset of PDA in extremely preterm infants by avoiding rapid increases in serum sodium concentrations.
Closure of the ductus arteriosus is one of the most dynamic and essential changes in the transition from the fetal to neonatal circulation. However, spontaneous ductal closure by 7 days of age occurs in only 30–40% of infants born at 25–28 weeks gestational age (GA) and 10% of infants born at <25 weeks GA.1–3 Preterm infants are at increased risk of patent ductus arteriosus (PDA), which often causes significant morbidity as a result of left-to-right shunting.4,5
Commonly, patients with symptomatic PDA are first treated pharmacologically with the administration of indomethacin, ibuprofen, or acetaminophen, after which surgical ligation is considered in those patients in whom the ductus fails to close.6 However, pharmacological treatments have adverse effects, such as oliguria, renal dysfunction, and hypoglycemia, and the availability of surgical procedures is limited and surgery may cause comorbidities. Therefore, the development of care practices for preterm infants that can prevent the occurrence of PDA is critical in the neonatal clinical setting.
In this context, strategies based on the mechanisms of ductus closure may work to facilitate closure of the ductus.7,8 Serum osmolality decreases transiently after birth and plays an important role in facilitating ductus closure.9,10 In an animal study, low osmolality promoted constriction, whereas high osmolality promoted ductal dilation.10 An observational study in human preterm infants showed that serum osmolality increased more rapidly after birth in infants with than without PDA.10 Because the serum sodium concentration is the most effective factor in determining serum osmolality and several studies have reported that reducing transepidermal water loss (TEWL) by maintaining infants in highly humidified incubators is a better way to prevent hypernatremia than increasing total fluid administration,11,12 we hypothesized that high-humidity care in the early postnatal period may prevent some cases of PDA in extremely preterm infants.
The aim of this study was to evaluate whether high-humidity care could prevent the onset of PDA in extremely preterm infants by reducing TEWL and avoiding a rapid increase in serum sodium concentrations.
This was a single-armed prospective controlled trial evaluating the superiority of high-humidity care in preventing PDA onset compared with conventional management in a historical control group (CG). This study followed the STROBE cohort reporting guideline (https://www.strobe-statement.org).
Study PopulationFor the historical CG, we retrospectively reviewed the medical records of infants born between 240 and 276 weeks gestation at the neonatal intensive care unit (NICU) of Yokohama City University (YCU) Medical Center between January 2016 and December 2017 who received conventional humidity care. Exclusion criteria were congenital anomaly (including congenital heart disease), severe asphyxia with a 5-min Apgar score <4, twin-to-twin transfusion syndrome at birth, prophylactic administration of indomethacin within 6 h after birth, and transfer to other hospitals. Of the 36 infants screened for inclusion in the CG, 29 were deemed eligible for and 7 were excluded (6 because of the prophylactic administration of indomethacin and 1 because of hospital transfer; Figure 1). Parental consent for the use of the data for the CG was obtained by an opt-out method.
Flow diagram of study enrollment.
For the high-humidity group (HHG), 28 extremely preterm infants (240 to 276 weeks gestation) born at the YCU Medical Center between July 2019 and September 2021 were prospectively recruited within 6 h after birth. Exclusion criteria were the same as for the CG; in addition, infants deemed ineligible by attending physicians were excluded from the study. Reasons for excluding infants from the HHG included parental consent refused, low Apgar score, congenital anomaly, prophylactic administration of indomethacin within 6 h of birth, and hospital transfer (Figure 1).
There was no transition to conservative management strategies, such as restrictive fluid therapy, high positive end-expiratory pressure ventilation, and there were no other changes in management (except for the high-humidity intervention) implemented between the 2 periods.
InterventionConventional humidity care began with 85% humidity for infants born at 24–25 weeks gestation, and 70–80% humidity for those born at 26–27 weeks gestation, for the first 24 h, followed in both groups by daily decreases of 5 percentage points to 60% humidity. The incidence rate of treated PDAs in this patient population was 76%,13 which was used to calculate the sample size for intervention.
After obtaining written informed consent from parents, the ambient humidity for infants enrolled in the HHG was set at 90% within 6 h after birth, maintained at 90% until 72 h of age, and then gradually decreased by 5 percentage points per day until it reached 60%.
Main Outcome MeasuresWhen an arterial line was available, blood samples were obtained from the artery. If an arterial line was not available, blood samples were obtained by heel stick. Levels of Na+ were measured using the i-STAT 1 Analyzer with an i-STAT CHEM8+ or CG8+ cartridge (Abbott Point of Care Inc., Abbott Park, IL, USA) or ABL800FLEX (Radiometer, Tokyo, Japan) prior to start of high-humidity care and at 8 (±2), 16 (±2), 24 (±4), 48 (±6), 72 (±12), 96 (±12), 168 (±12), and 336 (±12) h of age. For the CG, we collected data that fitted the same time points from the records. We compared the frequency of sample measurements in the CG with that in the HHG to check whether there were any differences between the groups. Doppler echocardiography was performed daily to evaluate the ductus arteriosus in the HHG. In contrast, the frequency of echocardiography in the CG varied because it was conducted when physicians needed to assess the need for treatment.
The primary outcome was the incidence rate of PDA requiring pharmacological treatment (i.e., indomethacin administration) within 168 h (7 days) after birth. The decision to initiate pharmacological treatment for PDA was made by the treating physicians based on both clinical and echocardiographic assessments, including heart rate, blood pressure, urinary output, and echocardiographic findings of significant left-to-right shunting (left atrial : aortic root ratio >1.2, disruption or reverse of diastolic flow at post-ductal main arteries, diastolic blood flow velocity in the left pulmonary artery >0.2 m/s, inner diameter of the ductus arteriosus >1.0 mm).
PDA that was symptomatic or worsening to PDA with echocardiographic findings was considered to be hemodynamically significant PDA (hsPDA), and infants with hsPDA were treated with indomethacin 0.1 mg/kg per dose. If the ductus arteriosus remained open after treatment, the same dosage was repeated up to 3 times at intervals of 24 h.
Predictors and confounders that could be risk factors for the occurrence of PDA were recorded, namely GA, small for gestational age (SGA), respiratory distress syndrome (RDS), mechanical ventilation, and fluid administration. Other factors implicated as being relevant to PDA were also collected from the medical charts.
Secondary outcomes of the study were: (1) the maximum rate of increase in serum sodium concentrations (i.e., the change in serum sodium concentration divided by the time interval between blood analysis) during the first 7 days for non-PDA patients or during the period prior to pharmacological treatment with indomethacin for PDA patients; (2) the incidence of hypernatremia (i.e., Na+ >145 mEq/L); and (3) the total dose administered and the frequency of indomethacin administration during the first 7 days.
Daily intake of water and sodium and daily urine output were calculated. The occurrence of the following adverse effects associated with high-humidity care was recorded: hyponatremia (Na+ <130 mEq/L), skin erosions requiring wound dressing or additional care, and skin infections requiring treatment with antibacterial or antifungal agents. Adverse events such as necrotizing enterocolitis and intraventricular hemorrhage (IVH) during the follow-up period were recorded. In addition, data on mid-term complications, such as PDA ligations, chronic lung disease, retinopathy of prematurity, remaining PDA at hospital discharge, and death before hospital discharge, were collected retrospectively.
Statistical AnalysisThe sample size for the intervention group was calculated using the incidence rate of treated PDA as the primary outcome. Using a 2-tailed α of 0.05 and a power of 0.80, it was estimated that 26 infants would be needed for intervention to demonstrate a reduction to 50%. We planned to have 30 infants in the intervention group.
We performed logistic regression analysis to verify the efficacy of high-humidity care in inhibiting the onset of PDA, with baseline adjustment for other relevant variables, such as GA, SGA, and RDS. Adjusted odds ratios (ORs) for independent variables are presented with 95% confidence intervals (CIs).
Continuous variables are presented as median with interquartile range (IQR). Categorical variables are presented as numbers and percentages. Fisher exact tests were used to compare categorical variables between the 2 groups, and the Mann-Whitney rank sum U test was used to compare continuous variables. The log-rank test was used for event-free time analysis. Two-tailed P<0.05 was considered statistically significant. Graphical visualizations were performed in Python (version 3.6), and statistical analysis was performed using JMP Pro15 (JMP Statistical Discovery, Cary, NC, USA).
Ethics ApprovalThis study was approved by the Research Ethics Committee of YCU Medical Center (B190704001, B190100021) and was registered with the Japan Registry of Clinical Trials (ID: jRCT1032190068).
The 28 infants in the HHG were maintained at 90% humidity for the first 72 h of life. Humidity levels were automatically controlled and checked to ensure that they were at the target level in both the HHG and CG eras. Humidity levels were always significantly higher for the HHG than CG throughout the study period (Figure 2). When comparing the HHG and CG, there were no significant differences in demographic characteristics, including RDS, SGA, rates and durations of mechanical ventilation, and antenatal steroid use (Table 1). Other characteristics that may be implicated as risk factors for managing PDA are presented in Supplementary Table 1.
Time course of humidity treatment in the historical control group (CG) and high-humidity group (HHG). Dotted lines connect median humidity values; error bars represent 95% confidence intervals and shaded rectangles represent the interquartile range. Humidity levels were significantly higher in the HHG than CG at all time points. *P<0.05.
Demographic Characteristics
| Characteristic | High-humidity group (n=28) |
Control group (n=29) |
P value |
|---|---|---|---|
| Gestational age (weeks) | 25.8 [25.2–26.8] | 26.1 [25.5–26.9] | 0.34 |
| Birth weight (g) | 727 [638–899] | 826 [672–878] | 0.30 |
| Male sex | 17 (61) | 15 (52) | 0.60 |
| Apgar score | |||
| At 1 min | 4 [3–5] | 4 [3–6] | 0.52 |
| At 5 min | 7 [6–8] | 7 [6.5–8] | 0.16 |
| RDS | 25 (89) | 25 (86) | 1.00 |
| SGA | 7 (25) | 2 (7) | 0.079 |
| Antenatal steroid | 21 (75) | 25 (86) | 0.33 |
| pPROM | 13 (46) | 18 (62) | 0.29 |
| CAM | 16 (57) | 19 (66) | 0.59 |
| Mechanical ventilation | |||
| No. infants (%) | 26 (93) | 26 (90) | 1.00 |
| Duration (days) | 7 [7–7] | 7 [7–7] | 0.33 |
| Hb <12 g/dL at birth | 3 (11) | 0 (0) | 0.11 |
Unless indicated otherwise, data are presented as the median [interquartile range] or n (%). CAM, chorioamnionitis; Hb, hemoglobin; pPROM, preterm premature rupture of membranes; RDS, respiratory distress syndrome; SGA, small for gestational age.
Outcomes
The proportion of infants who received pharmacological treatment for PDA during the first 7 days of life was significantly lower in the HHG than CG (43% vs. 76%; P=0.016; Table 2). In addition, multivariate logistic regression analysis of 57 infants (28 in the HHG and 29 in the CG) showed that the variable of “high-humidity care” was effective in reducing the incidence rate of PDA during the first 7 days (adjusted OR 0.265; 95% CI 0.078–0.907; P=0.034; Table 3). The cumulative event-free rate, derived from the timing of treatment for PDA in both groups, showed that PDA occurred later and that there were fewer infants with PDA in the HHG than CG by Day 7 (P=0.0037; Supplementary Figure 1). There were no significant differences in the most recent echocardiographic findings, namely left pulmonary artery end-diastolic velocity and the left atrial-to-aortic ratio, between the 2 groups prior to the pharmacological treatment of PDA (Supplementary Table 2).
Outcomes in the High-Humidity vs. Control Group
| Outcome | High-humidity group (n=28) |
Control group (n=29) |
P value |
|---|---|---|---|
| hsPDA onset | 12 (43) | 22 (76) | 0.016 |
| Hypernatremia (Na+ >145 mEq/L) | |||
| First 3 days | 9 (32) | 16 (55) | 0.111 |
| First 7 days | 13 (46) | 16 (55) | 0.60 |
| Maximum rate of increase in serum sodium concentration (mEq/L/h) |
0.29 [0.21–0.39] | 0.46 [0.32–0.62] | <0.001 |
Unless indicated otherwise, data are presented as the median [interquartile range] or n (%). hsPDA, hemodynamically significant patent ductus arteriosus.
Logistic Regression Analysis for the Onset of hsPDA
| Factor | Adjusted OR | 95% CI | P value |
|---|---|---|---|
| High-humidity care | 0.265 | 0.078–0.907 | 0.034 |
| SGA | 0.182 | 0.031–1.083 | 0.061 |
| RDS | 2.835 | 0.397–20.246 | 0.30 |
| Gestational age (+1 week) | 0.946 | 0.471–1.898 | 0.88 |
CI, confidence interval; OR, odds ratio. Other abbreviations as in Tables 1,2.
Figure 3A shows changes in sodium concentrations in the CG and HHG, whereas Figure 3B,C shows changes in sodium concentrations in patients with and without hsPDA in the CG and HHG, respectively. By 7 days of age, the median number of blood tests in the CG and HHG was 22 (IQR 19.5–24.5) and 20.5 (IQR 19–24), respectively, with blood tests occurring at about the same time. The number of blood tests depended on the severity of illness in each case, but there were no significant differences between the 2 groups (P=0.56). There was no difference in the rate of hypernatremia (Na+ >145 mEq/L) between the CG and HHG during the first 7 days of life. Of note, the maximum rate of increase in serum sodium concentrations was significantly lower in the HHG than CG (0.29 vs. 0.46 mEq/L/h; P<0.001; Table 2). In the HHG, the maximum rate of increase in sodium concentrations was lower in infants with than without hsPDA (0.23 vs. 0.36 mEq/L/h, respectively). Additional analysis of the whole patient group revealed that sodium concentrations in infants with hsPDA increased significantly during the first 24 h, whereas sodium levels remained unchanged in infants without hsPDA (Figure 3D).
Serial changes in serum sodium concentrations in (A) infants in the historical control group (CG) and high-humidity group (HHG) and (B,C) infants with and without hemodynamically significant PDA (hsPDA) in the CG (B) and HHG (C). (D) Changes in serum sodium concentrations within 24 h after birth in the entire study cohort according to the presence of hsPDA. Serum sodium concentrations increased significantly in infants with hsPDA. *P<0.05. Dotted lines connect median serum sodium concentration values; error bars represent 95% confidence intervals and shaded rectangles represent the interquartile range.
The median total dose of indomethacin during the first 7 days of life was 0 mg/kg (IQR 0–0.2 mg/kg) in the HHG and 0.1 mg/kg (IQR 0.05–0.2 mg/kg) in the CG (P=0.054). The median number of treatments during the first 7 days of life was 0.5 (IQR 0–2) in the HHG and 2 (IQR 1–3) in the CG (P=0.047; Table 4). When we focused on infants with hsPDA, the total dose of indomethacin and the number of treatments were similar between the HHG and CG (Supplementary Table 3). The number of surgical ligations was also similar between the 2 groups.
Additional Analysis for Clinical Outcomes
| Outcome | High-humidity group (n=28) |
Control group (n=29) |
P value |
|---|---|---|---|
| Indomethacin | |||
| Total dose (mg/kg) | 0 [0–0.2] | 0.1 [0.05–0.2] | 0.054 |
| No. doses | 0.5 [0–2] | 2 [1–3] | 0.047 |
| Hyponatremia (Na+ <130 mEq/L) | 4 (14) | 3 (10) | 0.71 |
| Skin erosion | 5 (18) | 5 (17) | 1.00 |
| Skin infection | 2 (7) | 2 (7) | 1.00 |
| IVH | |||
| ≤Grade 2A | 2 (7) | 6 (21) | 0.25 |
| >Grade 2B | 5 (18) | 1 (4) | 0.102 |
| NEC | 0 (0) | 0 (0) | 1.0 |
| Surgical closure of PDA | |||
| ≤14 days | 0 (0) | 0 (0) | 1.0 |
| >14 days | 3 (11) | 5 (17) | 0.71 |
Unless indicated otherwise, data are presented as the median [interquartile range] or n (%). AThe high-humidity group included 1 infant with intraventricular hemorrhage (IVH) detected at birth; the control group had 3 infants detected with IVH at birth. BThe high-humidity group had 2 infants with IVH following an idiopathic pulmonary hemorrhage, 1 infant with IVH following a pneumothorax, and 1 infant with IVH detected at birth. NEC, necrotizing enterocolitis; PDA, patent ductus arteriosus.
Daily water and sodium intake was greater in the HHG than CG on Day 0, reflecting a larger ratio of patients with SGA who initially received a greater fluid infusion volume, but there was no difference between the 2 groups from Day 1 to Day 7 (Figure 4A,B). Daily urine output was significantly greater in the HHG than CG on Days 3, 4, 5, and 7 (Figure 4C). There were no differences in serum potassium and blood glucose levels between the 2 groups (Supplementary Figures 2,3). Blood urea nitrogen data for the HHG are shown in Supplementary Figure 4, but were not available for the CG.
(A) Total water intake and (B) total sodium intake in the historical control group (CG) and high-humidity group (HHG). There were no significant differences between the 2 groups except on Day 0. (C) Daily urine output. Urinary output was significantly greater in the HHG than CG on Days 3, 4, 5, and 7. Dotted lines connect median values; error bars represent 95% confidence intervals and shaded rectangles represent the interquartile range. *P<0.05.
Adverse Events
There was no significant difference in the incidence rate of hyponatremia (Na+ <130 mEq/L) for the first 14 days of life between the HHG and CG (14% vs. 10%, respectively; P=0.71). The incidence of skin erosion and skin infections requiring treatment was also similar between the HHG and CG (skin erosion: 18% vs. 17%, respectively; skin infections requiring treatment: 7% vs. 7%, respectively). No necrotizing enterocolitis was observed, and equal numbers of cases of IVH were observed in the 2 groups. Five cases of Grade 3 IVH occurred in the HHG, 2 following idiopathic pulmonary hemorrhage and 1 following pneumothorax (Table 4). There was 1 infant with focal intestinal perforation on Day 2 in the HHG, for whom the high-humidity care intervention was terminated; 3 cases of meconium ileus in the CG (2 of which were previously excluded because of the administration of prophylactic indomethacin) were excluded from analysis because of hospital transfer. No differences were observed in the rates of adverse effects and mid-term complications between the 2 groups (Supplementary Table 4). With regard to remaining PDA at discharge, no infant in the HHG had a remaining ductus arteriosus at discharge, whereas in the CG 1 infant was discharged home with a remaining PDA and 2 infants were transferred back to local hospitals with remaining PDA at 31 weeks and 36 weeks of postmenstrual age.
The aim of this study was to evaluate the efficacy of high-humidity care in preventing the onset of PDA. In this single-arm prospective trial, the incidence rate of hsPDA was significantly lower in the HHG than CG. Moreover, the maximum rate of increase in serum sodium concentrations was significantly lower in the HHG than CG. Multivariate logistic regression analysis revealed that high-humidity care was associated with reduced risk of hsPDA after adjustment for confounders. As far as we could see, there was no increase in adverse events with the intervention (high-humidity care).
This study was based on the rationale that a rapid increase in serum osmolality interrupts ductus closure.10 It is well known that extremely preterm infants are at high risk of hypernatremia because of dehydration.14,15 Because the epidermal structure of premature infants is immature, TEWL in infants born at 24–25 weeks gestation and maintained in 50% humidity is estimated to be 150 mL/kg/day on Day 0 and Day 1,16 gradually decreasing as the epidermal structures mature.17 Since the 1970s, the effects of humidification for extremely preterm infants have been focused on improving body temperature control, reducing skin water loss, and improving electrolyte balance.18 Increasing fluid infusion is one way to compensate for the volume depletion in these infants. However, methods that can reduce water evaporation from the skin (e.g., a high-humidity environment) are expected to be more effective in preventing hypernatremia and beneficial in maintaining blood glucose stability.11,15
In this study we showed that high-humidity care during the first 72 h of life was effective in reducing the rate of increase in serum sodium concentrations by retaining endogenous water, and that the increased urinary output could be attributed to this endogenous water.
Serum osmolality has been reported to decrease transiently after birth in both humans and rats,9,10 and the hypo-osmolality sensor transient receptor potential melastatin 3 (TRPM3) is involved in promoting ductal constriction after birth.10 In contrast, increasing serum osmolality by NaCl administration dilates the constricted ductus arteriosus in rats.10 In a human cohort study, Cakir et al. recently reported that serum osmolality was higher in the hsPDA than non-hsPDA group at 48 h of postnatal age.19 In a retrospective cohort study, Toya et al. also reported that serum osmolality was higher in the PDA than non-PDA group from Day 1 to 3, and that the increased serum osmolality was significantly correlated with serum sodium and correlated with PDA onset.20 These observations are consistent with our findings that preventing a rapid increase in sodium concentrations contributes to the prevention of hsPDA.
In our study, water and sodium intake was the same between the 2 groups, except on the first day of life. In contrast, urinary output was greater in the HHG than CG from Day 3 to Day 7. This may be attributed to preventing dehydration, especially by reducing TEWL during the first 72 h of life. In fact, the rate of increase in serum sodium concentrations was significantly lower in the HHG than CG. These results are consistent with human and animal studies,10,19,20 showing our study to be a proof-of-concept study. Unfortunately, body weight changes were not evaluated in our study because preterm care warranted minimal handling of the infants.
Our findings that infants with hsPDA in the HHG had a lower rate of change in serum sodium concentrations appears incongruent with our hypothesis. This could be explained by a different mechanism of PDA onset. From the overall change in serum sodium concentrations in the first 24 h shown in Figure 3D, sodium levels initially rose only in patients with PDA. In infants with PDA, the initial decrease and following recovery in serum sodium concentrations were lacking, consequently offsetting the initial rate of increase. Because the transient decrease of serum osmolality after birth is initiated by regular postnatal adaptation (i.e., a shift of lung fluid to the circulating compartment followed by extracellular fluid loss and diuresis21), we attribute these distinct changes in sodium levels observed in infants with hsPDA in both groups to their insufficient primary postnatal adaptation, which may be associated with RDS. Thus, postnatal serum sodium changes may be epiphenomena, but causal in the etiology of PDA development. The effect of Na+ fluctuations is just one part of the development of PDA, but may be an important consideration to prevent the occurrence of PDA.
Because humidity practices are nursing practices and management standards are determined at each NICU, there may be variations from facility to facility.22–25 The incidence of PDA also varies between facilities, and differences in humidity management during the early postnatal period may account for these differences. Previous studies have evaluated the efficacy and safety of humidity regimens from the perspective of skin barrier maturation and the potential for skin infection,16,26 but to date no such trials have evaluated the effects on PDA onset. Because current pharmacological and surgical therapies for PDA are not always ideal, the prevention of PDA will be of great benefit to extremely preterm infants who are at risk of PDA. High-humidity care seems feasible and can be practically applied in facilities where the incidence rate of PDA is higher and the standard humidity level in the early postnatal period is relatively low compared with other facilities. Based on our observations, 90% humidity was safe for the first 72 h of life. No significant differences in complication rates were observed between the 2 groups in the present study. However, there is a concern that high-humidity care may prolong skin barrier development and increase the risk of skin erosions or infections. Moreover, the higher incidence of cases of severe IVH in the intervention group should be noted. Although these cases followed pneumothorax or lung hemorrhage, it may be possible that the high-humidity intervention group was at risk of IVH due to extreme suppression of TEWL, which led to higher intravascular volume. The potential disadvantages of high-humidity care have not been fully investigated, so precautions for its use in clinical practice are needed. In particular, the advantages and disadvantages of high-humidity care should be carefully evaluated in each NICU according to settings such as fluid intake, incubator type, cleaning cycles, skin infection risks, and body temperature fluctuations.
This study has some limitations. First, this was a single-arm controlled trial using the past incidence rate of hsPDA as a historical control. This resulted in a difference of up to 5 years between the intervention and control groups. We neither shifted to conservative management nor changed the pharmacological agents used during the 2 study periods. Because temporal changes in echocardiographic indicators were not fully available in the CG, we cannot entirely exclude the possibility of treatment bias when comparing the groups. Second, although treatment plans were discussed to ensure the provision of relatively uniform medical care, the attending physicians determined the need for indomethacin treatment based on the results of echocardiography, vital signs, and GA. The attending physicians were not blinded to group assignment, because most infants were included during the study period and humidity care could not be concealed at the bedside. Third, a precise comparison of adverse events was not possible because the allocation method was not identical between the 2 groups. These factors may have introduced potential bias. Further validation through well-designed multicenter studies comparing 2 groups concurrently is needed to confirm our results and may offer a clearer understanding of the risk factors associated with the onset of PDA.
High-humidity care for the first 72 h of life may be a useful method to reduce the incidence rate of hsPDA in extremely preterm infants by helping maintain a stable serum sodium concentration. High-humidity care may play a novel and important role in the postnatal adaptation of premature infants and may be of benefit for those at risk of PDA.
The authors thank Yoshihiro Ishikawa from the Yokohama City University Cardiovascular Research Institute for his support and encouragement in planning this translational research. The authors also thank all the parents and infants who participated in the study and the staff of the neonatal intensive care unit at the Yokohama City University Medical Center.
There are no funders in our study.
The authors declare that there are no conflicts of interest related to this work.
This study was approved by the Research Ethics Committee of Yokohama City University Medical Center (Approval no. B190704001, B190100021).
The deidentified participant data will not be shared.
Please find supplementary file(s);
https://doi.org/10.1253/circj.CJ-24-0705
