Eficacy and safety of restricted fluid therapy in transient tachypnea of the newborn: a randomized controlled study

Abstract

Background

Evidence suggests that increased intravascular and interstitial fluid load in neonates with transient tachypnea of the newborn (TTN) may delay the clearance of fetal alveolar fluid (FAF). Restricted fluid (RF) therapy may accelerated FAF clearance and improve outcomes in these infants.

Methods

Term and late preterm infants with TTN requiring nasal intermittent positive pressure ventilation (NIPPV) were randomized within first 2 h after birth to receive either RF or standard fluid (SF) therapy. Primary outcomes were the duration of NIPPV and the day of discharge. Secondary outcomes included changes in weight, urine output, biochemical parameters, and monitoring of potential adverse effects.

Results

Forty-four infants were analyzed. NIPPV duration (36 ± 25 vs. 57 ± 23 h; p = 0.012) and the time to full enteral feeding (65 ± 25 vs. 94 ± 20 h; p < 0.005) were shorter in the RF group. Although the day of discharge was earlier in the RF group, the difference did not reach statistical significance (6.5 ± 1.3 vs. 7.5 ± 2.3 days; p = 0.084). Mean weight loss on postnatal day 5 was higher in the RF group (p < 0.005). No significant differences were observed in urine output, serum electrolytes, or renal function tests. Hypernatremia (ranging from 146 to 150 mmol/L) developed in seven patients (four in the SF group and three in the RF group), and no hypoglycemia or serious adverse events occurred.

Conclusion

Restricted fluid therapy appears safe and effective in TTN infants requiring NIPPV, reducing respiratory support duration and enabling earlier full enteral feeding.

Trial registration

ClinicalTrials.gov, NCT07549945. Registered 23 April 2026, retrospectively registered. The study was conducted between October 2021 and July 2023. During the final stages of the study, the devastating earthquakes that struck southeastern Türkiye in February 2023 caused major disruptions to clinical and academic activities at our institution, contributing to delays in manuscript preparation and trial registration.

Introduction

Transient tachypnea of the newborn (TTN) is the most common cause of respiratory distress in term and late preterm infants. The condition results from delayed clearance of fetal alveolar fluid during the perinatal period, leading to pulmonary fluid retention, impaired gas exchange, and respiratory distress [1]. The rise in catecholamine and glucocorticoid levels during labor promotes the clearance of fetal lung fluid [2–4]. Therefore, infants delivered by cesarean section before the onset of labor are at increased risk of developing TTN due to delayed fluid absorption [5].

TTN typically presents with respiratory distress shortly after birth and is diagnosed based on clinical and radiographic findings [6]. Most cases resolve within a few days with supportive treatment which includes intravenous fluid support, oxygen therapy, and/or non-invasive ventilation (NIV) support [1]. However, severe cases may be complicated by persistent pulmonary hypertension or air leak syndromes requiring invasive mechanical ventilation [7].

Although diuretics, dopamine, nebulized epinephrine and beta-2 agonists have been investigated in the treatment of TTN, current meta-analyses have not demonstrated strong evidence supporting the efficacy of these agents [8–10]. Recent studies have reported findings suggestive of fluid overload in infants with TTN, including elevated serum NT-proBNP levels and reduced left atrial reservoir strain [11, 12]. Based on this pathophysiological basis, a limited number of studies have evaluated the efficacy of restricted fluid (RF) therapy in TTN. These studies have suggested that restricted fluid therapy is safe and may decrease the duration of respiratory support in infants with TTN [13–15]. Furthermore, Stroustrup et al. reported a potential reduction in hospitalization costs, while Sardar et al. observed that the greatest benefit was seen among late preterm infants who had not received antenatal corticosteroids [13, 15]. However, due to very low certainty about the current evidence, RF therapy is not included in the literature as standard approach and the need for further randomized studies is emphasized [16]. Therefore, in the present study, we aimed to evaluate the efficacy and safety of RF therapy, which is a low-risk and feasible approach that may influence the clinical course of TTN.

Methods

This prospective randomized study was conducted in a tertiary neonatal intensive care unit (NICU) between October 2021 and July 2023. The study was approved by Institutional Clinical Trials Ethics Committee (approval number: 2018/76). Neonates born at our hospital with a gestational age of ≥ 34⁰/⁷ weeks who were diagnosed with TTN were included in the study. TTN was diagnosed based on the presence of respiratory distress symptoms beginning within the first hours after birth, together with at least one of the following chest radiographic findings: symmetrical perihilar congestion, increased perihilar vascular markings, widened interlobar fissures, or air trapping, and after exclusion of other causes of respiratory distress, including respiratory distress syndrome, meconium aspiration syndrome, congenital pneumonia, congenital heart disease, polycythemia, and surgical conditions [6].

Infants with severe congenital anomalies, those who required intubation in the delivery room or upon admission to the NICU, infants born through meconium-stained amniotic fluid, perinatally asphyxiated infants, infants with suspected sepsis based on risk factors or clinical findings, and infants diagnosed with TTN who did not require NIV were not included from the study. In addition, infants initially included with the diagnosis of TTN who were diagnosed with congenital pneumonia and/or sepsis in the follow-up or who received surfactant therapy or who were evaluated as “complicated TTN” after developing pneumothorax, pneumomediastinum or persistent pulmonary hypertension were excluded from the study. The final analysis was performed on a per-protocol basis. The parents of all participating infants provided written informed consent. This study was reported in accordance with the CONSORT guidelines for randomized controlled trials.

Clinical and demographic characteristics of the patients were recorded. The degree of respiratory distress was evaluated using the Silverman–Anderson score prior to NIPPV, and clinical status was assessed using the SNAP-PE II score.

Sample size calculation was based on the study by Stroustrup et al. Assuming a median duration of respiratory support of 75 h in the restricted fluid group and 113 h in the standard fluid group, a two-sided alpha level of 0.05, and a statistical power of 80%, a minimum of 33 infants was required in each group. Therefore, a total of 70 infants were planned for randomization to account for potential exclusions and losses during follow-up.

Randomization and follow-up

Infants with TTN requiring NIPPV were randomized within the first two hours of life to either the RF group or the standard fluid (SF) group using the sealed opaque envelope method. To ensure homogeneous distribution of infants with similar gestational ages between the groups, infants born at 34⁰/₇–36⁶/₇ weeks and those born at ≥ 37 weeks of gestation were randomized separately using a stratified randomization approach.

The total fluid volume administered on the first day was initiated at 50 mL/kg/day in the RF group and 70 mL/kg/day in the SF group for infants with a gestational age of 34⁰/₇–36⁶/₇ weeks, whereas it was initiated at 40 mL/kg/day in the RF group and 60 mL/kg/day in the SF group for infants with a gestational age of ≥ 37 weeks. During the subsequent days, the total fluid volume administered per kilogram was either increased by 10–20 mL/kg/day or maintained, depending on whether physiological weight loss occurred and on serum sodium levels. All infants received intravenous fluids containing 10% dextrose and 0.25% NaCl during the first 24 h. At 24 h of postnatal age, 15 mmol/L of KCl was added to the intravenous fluids if serum potassium and creatinine levels were within normal ranges and urine output was adequate. Blood glucose monitoring was performed at least every 6 h. The dextrose concentration of intravenous fluids was increased in infants whose blood glucose levels approached the lower limit (50–60 mg/dL). Blood glucose levels below 50 mg/dL were considered hypoglycemia. Daily body weight, urine output, and fluid intake were recorded. Serum electrolyte levels, blood urea nitrogen (BUN), creatinine, calcium levels, and urine specific gravity were measured at 24 and 72 h of postnatal age and on day 5. In addition, the duration of NIPPV, time to first enteral feeding, time to full enteral feeding, and day of discharge were recorded. Primary outcomes included duration of NIPPV and the day of discharge. Secondary outcomes included changes in weight, urine output, biochemical parameters, and adverse events.

Non-invasive ventilation and weaning

Non-invasive ventilation was administered to all infants NIPPV mode (non-synchronised) and using short binasal prongs (Hudson^® prong) as the nasal interface. Initial NIPPV settings were identical in both groups and were set as follows: positive end-expiratory pressure (PEEP) of 6 cmH₂O, peak inspiratory pressure (PIP) of 20 cmH₂O, inspiratory time of 0.4 s, and a respiratory rate of 20–30 breaths per minute. NIPPV was delivered using a neonatal ventilator (SLE; Specialised Laboratory Equipment, South Croydon, UK).

Weaning was initiated in infants whose respiratory distress regressed. Initially, PEEP was reduced to 4–5 cmH₂O, followed by discontinuation of NIPPV through gradual reduction of the respiratory rate.

Statistics

Continuous variables are presented as mean ± standard deviation or median (and minimum–maximum) and categorical variables are presented as number and percentage. Categorical variables were compared using the chi-square test (2-tailed). Continuous variables were compared with Student’s t-test or Mann-Whitney U test depending on data distribution. A p value < 0.05 was considered as statistically significant. Statistical analyses were performed using SPSS version 17.0 (SPSS Inc., Chicago, IL, USA).

Results

A total of 50 newborns who met the inclusion criteria were randomized. Three patients diagnosed with congenital pneumonia during follow-up, one patient who received surfactant therapy and, two patients who developed pneumothorax (one in SF group and one in RF group) were excluded from the analysis. Therefore, a total of 44 patients (23 in the RF group and 21 in the SF group) were evaluated. Recruitment was discontinued before the planned sample size was reached because of major disruptions to healthcare services caused by the February 2023 earthquakes in southeastern Türkiye.

The mean gestational age of the patients was 36.3 ± 1.4 weeks, and their mean birth weight was 2920 ± 483 g. Demographic and clinical characteristics were similar between groups (Table 1). No significant differences were observed between groups in terms of Silverman-Anderson or SNAP-PE II scores. None of the patients received inotrope or furosemide therapy (Table 1).

Table 1.

Baseline demographic and clinical characteristics of the study population

	All patients (n = 44)	Standard fluid (n = 21)	Restricted fluid, n = 23	p value
Gestational age, weeks, mean ± SD	36.3 ± 1.4	36.2 ± 1.3	36.4 ± 1.4	0.61
Birth weight, g, mean ± SD	2920 ± 483	2950 ± 515	2890 ± 462	0.66
Cesarean delivery, n (%)	42 (95.4)	21 (100)	21 (91.3)	0.26
Male sex, n (%)	31 (70.5)	16 (76.1)	15 (65.2)	0.32
Apgar score at 5 min, median (min-max)	9 (7–10)	9 (8–10)	9 (7–10)	0.69
Multiple gestation, n (%)	7 (15.9)	3 (14.2)	4 (17.3)	0.55
Receive antenatal steroids, n (%)	7 (15.9)	3 (14.2)	4 (17.3)	0.55
Premature rupture of membranes, n (%)	1 (2.3)	0	1 (4.3)	0.52
Small for gestational age, n (%)	3 (6.5)	1 (4.7)	2 (8.6)	0.53
Maternal preeclampsia, n (%)	6 (13.6)	5 (23.8)	1 (4.3)	0.07
Infant of diabetic mother, n (%)	5 (11.4)	4 (19)	1 (4.3)	0.17
Maternal asthma	1 (2.2)	1 (4.7)	0	0.47
SNAP PE-II score, median (min-max)	0 (0–5)	0 (0–5)	0 (0–5)	0.63
Silverman-Anderson score, median (min-max)	5 (4–9)	5 (4–8)	5 (4–9)	0.82
Hypoglycemia (blood glucose < 50 mg/dL), n	0	0	0	-
Hypothyroidism, n	0	0	0	-
Antibiotic therapy initiated, n (%)	7	3	4	0.55
Mechanical ventilation requirement, n (%)	1	1	0	-

SD Standard deviation

Total fluid intake was significantly lower in the RF group at all time points during the first 72 h after birth (Table 2). Although mean urine output was lower in the RF group compared with the SF group, this decrease was not statistically significant, and urine output remained above > 2 ml/kg/hour in all periods (Table 2). No significant differences were observed between the groups in terms of weight measurements. However, on postnatal day 5, mean weight loss relative to birth weight was significantly higher in the RF group (RF: 6.95 ± 2.5% vs. SF: 4.57 ± 2.2%; p < 0.005, Table 2).

Table 2.

Fluid balance, renal function, and biochemical parameters of the study groups

	Standard fluid, n = 21	Restricted fluid, n = 23	p value
Gestational age, weeks, mean ± SD	2950 ± 515	2890 ± 462	0.66
Weight at 24 h postnatal, g, mean ± SD	2956 ± 524	2862 ± 479	0.53
Weight at 48 h postnatal, g, mean ± SD	2901 ± 504	2797 ± 487	0.49
Weight at 72 h postnatal, g, mean ± SD	2846 ± 528	2743 ± 478	0.50
Weight on postnatal day 5, g, mean ± SD	2815 ± 513	2689 ± 490	0.42
Mean weight loss on postnatal day 5, %, mean ± SD	4.57 ± 2.2	6.95 ± 2.5	< 0.005
Total fluid intake, 0–12 h, mL/kg/d, mean ± SD	63.5 ± 4.8	46.5 ± 4.8	< 0.005
Total fluid intake, 12–24 h, mL/kg/d, mean ± SD	64.0 ± 5.0	49.1 ± 5.9	< 0.005
Total fluid intake, 24–48 h, mL/kg/d, mean ± SD	67.0 ± 8.0	57.8 ± 9.0	< 0.005
Total fluid intake, 48–72 h, mL/kg/d, mean ± SD	74.0 ± 10.0	65.6 ± 13.1	0.02
Total fluid intake, postnatal days 3–5, mL/kg/d, mean ± SD	86.2 ± 10.3	82.1 ± 14.6	0.29
Urine output, 0–24 h, mL/kg/h, mean ± SD	2.31 ± 0.7	2.12 ± 0.6	0.36
Urine output, 24–48 h, mL/kg/h, mean ± SD	2.72 ± 0.9	2.45 ± 0.7	0.26
Urine output, 48–72 h, mL/kg/h, mean ± SD	2.38 ± 0.6	2.31 ± 0.5	0.68
Urine output, postnatal days 3–5, mL/kg/h, mean ± SD	2.60 ± 0.4	2.55 ± 0.6	0.80
Urine specific gravity at 24 h, mean ± SD	1004 ± 2.8	1006 ± 4.4	0.10
Urine specific gravity at 48 h, mean ± SD	1006 ± 3.7	1006 ± 3.9	0.87
Urine specific gravity at 72 h, mean ± SD	1007 ± 7.1	1007 ± 5.8	0.83
Serum creatinine at 24 h, mg/dL, mean ± SD	0.71 ± 0.10	0.75 ± 0.20	0.56
Serum creatinine at 72 h, mg/dL, mean ± SD	0.57 ± 0.10	0.58 ± 0.10	0.89
Serum creatinine on postnatal day 5, mg/dL, mean ± SD	0.52 ± 0.07	0.51 ± 0.10	0.89
Blood urea nitrogen at 24 h, mg/dL, mean ± SD	10.2 ± 4.6	11.0 ± 5.1	0.60
Blood urea nitrogen at 72 h, mg/dL, mean ± SD	6.5 ± 3.8	7.3 ± 5.5	0.60
Blood urea nitrogen on postnatal day 5, mg/dL, mean ± SD	5.5 ± 3.1	6.4 ± 4.8	0.62
Serum sodium at 24 h, mmol/L, mean ± SD	137.4 ± 3.8	138.3 ± 2.3	0.34
Serum sodium at 72 h, mmol/L, mean ± SD	142.8 ± 3.0	141.7 ± 3.3	0.30
Serum sodium on postnatal day 5, mmol/L, mean ± SD	141.9 ± 2.8	141.6 ± 3.1	0.79
Serum potassium at 24 h, mmol/L, mean ± SD	4.9 ± 0.7	5.1 ± 0.7	0.29
Serum potassium at 72 h, mmol/L, mean ± SD	4.9 ± 0.6	4.9 ± 0.5	0.62
Serum potassium on postnatal day 5, mmol/L, mean ± SD	5.2 ± 0.5	4.9 ± 0.5	0.18
Number of patients with serum Na > 145 mmol/L, n	4	3
Number of patients with Blood urea nitrogen > 20 mg/dL, n	1	2

SD Standard deviation

Serum BUN, creatinine, and electrolyte levels, and mean urine specific gravity, were similar between groups (Table 2). BUN exceeded > 20 mg/dL in three patients (RF, n = 2 vs. SF, n = 1). However, these values were only mildly elevated clinically, with a maximum of 24 mg/dL. Hypernatremia (> 145 mmol/L) occurred in seven patients (RF, n = 3 vs. SF, n = 4), with sodium levels ranging from 146 to 150 mmol/L. Hypernatremia was observed on postnatal day 3 in all affected patients. No episodes of hypoglycemia developed in any of the patients during follow-up (Table 2).

The duration of NIPPV (36 ± 25 vs. 57 ± 23 h; MD − 20.1 h, 95% CI − 35.65 to − 4.62; p = 0.012), and the time to achieve full enteral feeding (65 ± 25 vs. 94 ± 20 h; MD − 29.4 h, 95% CI − 43.97 to − 14.74; p < 0.005) were significantly shorter in the RF group (Table 3). Although discharge day was shorter in the RF group, this difference was not statistically significant (RF group, 6.5 ± 1.3 days vs. SF group, 7.5 ± 2.3 days; p = 0.084, Table 3). One patient in SF group required intubation. No mortality or serious adverse events occurred in either group.

Table 3.

Clinical outcomes of the study groups

	Standard fluid, n = 21	Restricted fluid, n = 23	Mean difference (95% CI)	Cohen’s d	p value
Duration of NIPPV, hours, mean ± SD	57 ± 23	36 ± 25	−20.1 (− 35.65 to − 4.62)	0.87	0.012
Time to first enteral feeding, hours, mean ± SD	10.7 ± 6.8	10.8 ± 8.1	0.1 (− 4.66 to 4.88)	0.01	0.96
Time to full enteral feeding, hour, mean ± SD	94 ± 20	65 ± 25	−29.4 (− 43.97 to − 14.74)	1.28	< 0.005
Day of discharge, days, mean ± SD	7.5 ± 2.3	6.5 ± 1.3	−1.0 (− 2.20 to 0.14)	0.55	0.084

SD Standard deviation

Discussion

In our study in which we evaluated the reliability and efficacy of restricted fluid therapy in term and late preterm infants with TTN requiring NIPPV, we found that RF treatment (approximately 66–75% of standard maintenance fluid) was clinically safe and significantly reduced the duration of NIPPV and the time to achieve full enteral feeding. In this respect, our study contributes to the limited number of clinical studies supporting the feasibility of a restricted fluid approach in the management of TTN [13–15].

There is evidence of increased fluid load in infants with TTN [11, 12]. Increased intravascular and interstitial fluid may delay the clearance of fetal alveolar fluid by increasing the burden on pulmonary lymphatic drainage. In addition, daily urine output and fractional excretion of sodium (FeNa) have been reported to be lower in infants with TTN compared with the controls [17]. Based on this pathophysiological rationale, several studies have investigated the efficacy of RF therapy in TTN. Stroustrup et al. demonstrated that restricted fluid therapy shortened the duration of respiratory support and reduced hospitalization costs in term and late preterm newborns with uncomplicated severe TTN [13]. Another study reported that restricted fluid therapy in late preterm infants with TTN shortened the duration of NIV [15]. In our study, the duration of NIPPV was significantly shorter in the RF group than in the SF group (36 ± 25 vs. 57 ± 23 h; p = 0.012, Table 3), consistent with previous studies and supporting the potential of restricted fluid therapy to accelerate clinical recovery. The significantly achieve transition to full enteral feeding in the RF group is another indicator of faster clinical recovery. Although TTN is generally a benign condition, it accounts for a substantial proportion of NICU admissions, and prolonged respiratory support increases length of hospitalization and the higher healthcare costs [13]. In our study, the RF group had a shorter length of hospital stay than the SF group (6.5 ± 1.3 vs. 7.5 ± 2.3 days, p = 0.084); however, the difference did not reach statistical significance. This may reflect the limited statistical power resulting from the smaller-than-planned sample size, and a larger adequately powered trial may confirm whether restricted fluid therapy is associated with a shorter hospital stay. Particularly in countries with high rates of cesarean delivery, identifying effective and safe therapeutic interventions is of considerable clinical importance.

In studies assessing breast milk intake in exclusively breastfed newborns using the test-weighing method, the average intake during the first two postnatal days was reported to be 9.6–13 ml/kg/day [18, 19]. These findings strongly support the notion that the physiological fluid requirements of the newborn in the first postnatal days may be quite low. Therefore, in diseases with increased fluid load and reduced diuresis, such as TTN, first day maintenance fluid volumes recommended in current clinical practice may exceed actual neonatal requirements and potentially delay recovery by increasing the pulmonary interstitial fluid load. However, many clinicians avoid restricted fluid therapy in infants due to possible negative effects such as dehydration, hypoglycemia, and impaired kidney functions and do not deviate from long-standing traditional recommendations.

The present study, in which potential adverse effects of restricted fluid therapy were closely monitored, did not identify any adverse effects associated with RF therapy. Although urine output was lower in the RF group compared with the SF group, it remained above 2 ml/kg/hour in all periods (Table 2). Mean weight loss at postnatal day 5 was higher in the RF group (RF, 6.95 ± 2.5% vs. SF, 4.57 ± 2.2%; p < 0.005, Table 2). This degree of weight loss was within physiological limits and may have contributed to a shorter recovery period. Similarly, Sardar et al. reported weight loss to be higher in the restricted fluid group at the end of the third postnatal day [15]. In our study, serum BUN, creatinine and sodium levels, as well as urine specific gravity were similar between groups. During follow-up, BUN values in three patients (two in the RF group and one in the SF group) exceeded 20 mg/dL; however, these elevations were clinically mild, with a maximum value of 24 mg/dL. Hypernatremia occurred in seven patients (four in the SF group and three in the RF group). Sodium levels in these patients ranged from 146 to150 mmol/L. No patients developed hypoglycemia during the study (Table 1). Although no metabolic or renal side effects were observed in infants who received RF therapy in our study, the significantly shorter durations of respiratory support and earlier achieve to full enteral feeding suggest that a restricted fluid approach is a reliable strategy based on physiological principles and is consistent with findings from previous studies [13–15].

Early randomization, standardized NIPPV support, a homogeneous patient population, and the monitoring of both respiratory and fluid-electrolyte parameters with a standard protocol enhance the reliability of the findings. However, several limitations should be acknowledged, including the relatively small sample size, the single-center design, and the exclusion of complicated cases accompanying TTN. In addition, the analyses were performed on a per-protocol basis; therefore, the possibility of attrition bias compared with an intention-to-treat analysis cannot be completely excluded.

In conclusion, restricted fluid therapy may be a safe and potentially beneficial approach for TTN patients requiring NIV, without an apparent increase in adverse effects. Although our findings are consistent with the limited available evidence, they should be interpreted in light of the study’s limitations, including its single-center design and relatively small sample size. Larger prospective multicenter studies are needed to confirm these findings and to evaluate the applicability of restricted fluid therapy across different TTN subgroups.

Abbreviations

BUN

Blood urea nitrogen

FAF

Fetal alveolar fluid

NICU

Neonatal intensive care unit

NIPPV

Nasal intermittent positive pressure ventilation

NIV

Non-invasive ventilation

PEEP

Positive end-expiratory pressure

PIP

Peak inspiratory pressure

RF

Restricted fluid

SF

Standard fluid

TTN

Transient tachypnea of the newborn

Authors’ contributions

İKG and RÖ conceived and designed the study and served as guarantors of the work, taking responsibility for the integrity of the study from conception to publication. İKG, HT, HK, MFD, and MA contributed to patient recruitment and data acquisition. İKG and RÖ performed the statistical analysis and interpreted the data. İKG drafted the initial manuscript. All authors read and approved the final manuscript.

Funding

No funding was received for this study.

Data availability

The datasets generated and/or analyzed during the current study are available in the Figshare repository, https://doi.org/10.6084/m9.figshare.31436674.

Declarations

Ethics approval and consent to participate

This study was approved by the Clinical Trials Ethics Committee of Inonu University Faculty of Medicine (approval number: 2018/76). The study was conducted in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from the parents or legal guardians of all participants.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.