Transient tachypnoea: new concepts on the commonest neonatal respiratory disorder

Abstract

Transient tachypnoea of the neonate (TTN) is the commonest neonatal respiratory disorder, but it is quite mild and so has been the subject of relatively little academic and educational work. Recent animal studies and the introduction of new bedside monitoring techniques (e.g. quantitative lung ultrasound and electrical cardiometry) have clarified its pathogenesis. Given its high incidence, TTN is a relevant public health issue and its clinical management should be considered in an era of resource constraints. This review focuses on the latest data on TTN in terms of its pathophysiology, biology, diagnosis, imaging, therapy and cost-effectiveness, so as to optimise clinical care at the bedside. The need for a new pathophysiology-based definition of TTN is also highlighted and the available therapeutics are analysed considering the associated public health issues. This updated knowledge can help to improve the management of TTN and impact positively on its relevant public health consequences. This is particularly important since the mortality of TTN is virtually nil and so cannot be used to evaluate any clinical innovation. We also aim to give some practical guidance for the real-world clinical management of TTN and contribute to the training of neonatologists who care for TTN patients.

Shareable abstract

Understanding of the pathophysiology of transient tachypnoea, the commonest neonatal respiratory disease, can now improve its diagnosis and provide treatments that impact on its relevant public health consequences. https://bit.ly/3YghDyz

Introduction

Transient tachypnoea of the neonate (TTN) is a condition we have been diagnosing for decades. Nonetheless, new concepts about its pathophysiology, imaging and clinical management have recently been advanced, especially thanks to the introduction of novel bedside monitoring. As TTN has a virtually nil mortality rate, it rarely generates significant attention in academic neonatology; however, as many infants are affected, it does impact public health. TTN may become a public health problem when neonatal critical care resources are in short supply, as may happen in certain geographical areas or during outbreaks (e.g. bronchiolitis or COVID-19), with potentially dramatic consequences [1, 2]. Thus, it has become important to provide pathophysiology-driven clinical management and pay particular attention to its financial cost-effectiveness in TTN management.

We focus on the latest data on TTN, including its pathophysiology, biology, diagnosis, imaging, therapy and cost-effectiveness, so as to inform bedside clinical care. Our aim is also to overcome the lack of academic and educational work on the topic and to help train neonatologists who care for TTN patients. This comprehensive review enables us to suggest possible research directions for further improvement of TTN clinical care.

Literature search methods

References were identified through searches of PubMed, without time or language limitations, with the following keywords and/or MeSH (medical subject headings) terms “transient”, “tachypnea” and “neonate”. Articles were also identified through searches of the authors’ own files and the references of the retrieved articles. The final reference list was generated on the basis of originality and relevance to the broad scope of this work.

Epidemiology

TTN, the commonest neonatal respiratory disorder, affects approximately 7–10% of the neonatal population [3], with an incidence of new cases twice that of respiratory distress syndrome (RDS) [4]. While these figures are well known, it is difficult to detail the epidemiology of TTN. In fact, epidemiological data vary between studies and depend on cofactors that are variably expressed in different populations and contexts.

For instance, gestational age, birth weight, maternal diseases, twinning, male sex, operative vaginal delivery and caesarean section are the main risk factors for TTN [5]. The incidence of TTN decreases with gestational age and is halved from 33 to 36 weeks’ gestation [6]. Caesarean section, particularly if performed before the onset of labour, increases the incidence of TTN and the type of delivery and gestational age have an interaction effect [4]. Admission to the neonatal intensive care unit (NICU) after caesarean section is twice as common in early-term as in full-term neonates [7]. This has recently focused attention on two newborn population categories, namely “late preterm” (i.e. between 34⁺⁰ and 36⁺⁶ weeks’ gestation) [8] and “early term” (i.e. between 37⁺⁰ and 38⁺⁶ weeks’ gestation) neonates [9], who may have a higher incidence of perinatal complications, including TTN, compared to more mature infants. Ethnicity may also have an effect: the incidence of late preterm or early term birth is higher in Black and American Indian populations [10, 11], while the caesarean section rate is increasing, particularly in India and Asia [12, 13].

This picture is further complicated by the uncertain clinical distinction between TTN and RDS, which has affected some studies [14], particularly before the introduction of new pathophysiology concepts [15, 16] and point-of-care lung ultrasound [17]. Thus, TTN is likely underdiagnosed in preterm populations [18] and this may be explained in light of current pathophysiology knowledge. These and other factors such as familial socioeconomic status, maternal and obstetrical care, availability of referral NICU beds and admission criteria may impact TTN diagnosis and epidemiology. Nonetheless, irrespective of the definition and setting, TTN is so common it undeniably represents a significant burden of care in terms of NICU admission and associated costs. We review these considerations in the following sections.

TTN may have long-term clinical consequences and is associated with the development of wheezing and even asthma later in life [19–21]. Wheezing may, in turn, increase costs due to hospitalisation and, in severe cases, admission to intensive care units (ICUs). Hospitalisation due to wheezing is known to reduce family quality of life [22]. The association with long-term respiratory morbidity does not seem to be related to prematurity (preterm infants with TTN have a risk of developing wheezing similar to that of TTN patients born at term), but might be influenced by sex (higher risk in males) and environmental factors [19].

What's in a name? The need for a new definition

Avery et al. [23] originally proposed the acronym TTN in 1966, when describing eight neonates in whom other causes of respiratory distress were excluded. These infants presented with what is now considered to be the classical clinical course and imaging of TTN. Avery et al. [23] proposed the delayed reabsorption of lung fluid as the cause of respiratory failure, which is a mechanism similar although not identical to that indicated by current pathophysiology knowledge. They reasoned that periarterial tissue is distended to receive alveolar fluid and allow its drainage into the circulation, as supported by radiographic findings (hilar engorgement and prominently opaque vascularity and fissures) [23]. The transience of the disease seemed consistent with the hypothesis as imaging improved upon fluid elimination.

These characteristics were peculiar to TTN compared to other causes of neonatal respiratory failure, such as RDS, pneumonia or meconium aspiration. At that time, they were not transient and were associated with high mortality [23]. Advances in our knowledge of respiratory pathophysiology and critical care have significantly changed the situation. Mortality due to these other conditions has dramatically decreased, but is still higher than that of TTN [24, 25]. In most cases, these disorders, like TTN, are transient, thanks to prenatal steroid prophylaxis, early use of continuous positive airway pressure (CPAP) or invasive ventilation, and surfactant administration or extracorporeal life support for severe cases. Hospital stays of patients with TTN and other respiratory conditions may be similar, at least in some cases.

Therefore, transience is no longer a feature peculiar to TTN and so this acronym no longer seems entirely appropriate. A more pathophysiology-driven definition would be “persistent postnatal lung oedema” [26]. However, we lack a rigorous consensus methodology exclusively dedicated to the definition of TTN. Meanwhile, the Montreux consensus for the definition of neonatal acute respiratory distress syndrome (NARDS), a severe form of neonatal respiratory failure with a completely different pathobiology, offers criteria to recognise RDS and TTN and distinguish them from NARDS (table 1) [15]. These criteria, and particular ultrasound findings, are based on pathophysiological as well as biological differences between these disorders and are helpful for differential diagnosis [15]. However, they only represent a first step towards more personalised respiratory care, since in some cases RDS and TTN pathobiology may co-exist (see below) [27].

TABLE 1.

Criteria to define transient tachypnoea of the neonate (TTN) and respiratory distress syndrome (RDS) according to the Montreux consensus

	Work of breathing	Typical lung ultrasound	Typical radiography	Clinical course	Need for respiratory support	Additional criteria (surfactant pool)
TTN	Mild (Silverman score ≤3)	Heterogenous interstitial pattern alternated with at least one normal zone (A-pattern); absence of consolidations; pleural line thickness <1 mm LUS at the diagnosis approximately between 4 and 6 [27]	Hilar engorgement and prominently opaque vascularity and fissures No alveolar opacities [23]	Appearing within the first 24 h and resolving within the first 72 h of life	Only supplemental oxygen or nasal CPAP or both	Lamellar body count (if assayed) >30 000 mm−3 [89]
RDS	Moderate–severe (Silverman score >3 but may be reduced by early CPAP), often with expiratory grunting	Homogeneous loss of lung aeration with diffuse alveolar pattern (“white lung”); no normal zones; irregular pleural line with subpleural consolidations (depth <1 cm or <0.5 cm·Kg−1) [90, 91] LUS at the diagnosis between 10 and 12 [27]	Fine, diffuse, reticular-granular loss of aeration with extension of bronchogram proportional to the severity Total white lung without bronchogram in severest cases [92]	Appearing within the first 24 h of life, with variable duration (endogenous surfactant production usually needing 4–6 days) [93]	Completed, sustained and prompt response to surfactant replacement or alveolar recruitment or both	Lamellar body count (if assayed) ≤30 000 mm−3 [89]

These definitions are based on the integration of perinatal history, demographics, laboratory findings, clinical course and imaging data. They were originally issued to exclude TTN and RDS when diagnosing neonatal acute respiratory distress syndrome (NARDS) [15]. Therefore, in addition to the features in the table, none of the criteria of the Montreux definition of NARDS should be fulfilled (i.e. lack of other trigger for respiratory failure such as infection or aspiration, absence of severe oxygenation deficit with oxygenation index ≥4, no cardiogenic lung oedema due to congenital heart defect or fluid overload through the patent ductus arteriosus). References describing the typical imaging findings are cited in the table. Lung ultrasound or radiography can be used alternatively, depending on local expertise, but lung ultrasound is more accurate for the assessment of lung aeration and for the distinction between TTN and RDS [94]. CPAP: continuous positive airway pressure; LUS: lung ultrasound score.

Pathophysiology and biology

Pathophysiology

Avery et al. [23] were right to describe the accumulation of fluid as the factor jeopardising lung function, as this reduces oxygen diffusion through the alveolar–capillary barrier, without influencing CO₂ elimination, which is ≈25 times easier than oxygen diffusion. This phenomenon is essentially similar to that occurring during cardiogenic lung oedema. However, in TTN patients, oedema has an ab extrinseco origin. The liquid does not come from the circulation but rather from accumulation of alveolar fluid, especially in neonates who do not lose it via their upper airways during uterine contractions and squeezing due to vaginal birth. The original hypothesis considered that the accumulation of alveolar fluid was due to its delayed re-absorption, but more recent animal data suggest that an excessively elevated airway liquid volume causes accumulation [16]. Bedside noninvasive electrical cardiometry estimation of extravascular lung water in neonates supports this hypothesis; TTN patients have significantly more extravascular lung water than age-matched control neonates (figure 1). When patients improve and the fluid is finally eliminated (usually ≈48–72 h postnatally), the extravascular lung water becomes similar to that of normal infants [28]. Moreover, at diagnosis, TTN patients have more alveolar fluid than neonates affected by RDS [29] whose lungs are collapsed because of surfactant deficiency and therefore cannot harbour a significant fluid volume. The excessive fluid is eliminated and, consistently, in the first days of life weight loss is greater in TTN patients than in age-matched neonates with RDS [30].

FIGURE 1.

Extravascular lung water (EVLW) in patients with transient tachypnoea of the neonate (TTN) and age-matched control neonates with no respiratory disorder. Extravascular lung water was estimated noninvasively by electrical cardiometry [29] as a) thoracic fluid content or b) weight-indexed fluid content. Data were extracted from the cardiorespiratory monitoring applied in our neonatal intensive care unit (NICU) clinical routine to neonates admitted for TTN (n=10) or for nonrespiratory problems (n=10), as previously defined [54]. Upon NICU admission, parents gave written consent for the anonymous use of monitoring data. Patients were matched for gestational (±1 week) and post-natal age (±2 days). Green and orange dots represent data from individual patients; box plots depict (from top to bottom) 95th, 75th, 50th, 25th and 5th percentiles. The raincloud curves on the right-hand side represent the density (distributions) of datapoints. Data were analysed with the Mann–Whitney test (EVLW=25 (20–29) and 42 (32–54) KOhm⁻¹ in controls and TTN patients, respectively; weight-indexed EVLW=10 (8–15) and 18 (16–21) KOhm⁻¹·kg⁻¹ in controls and TTN patients, respectively).

This feature of lungs affected by RDS is important in understanding TTN pathophysiology and is illustrated in figure 2. The density of lung tissue and the alveolar space (i.e. the number of alveoli) decrease and increase with gestational age, respectively [31]. Thus, in late preterm and term neonates more alveolar fluid should be drained by thinner tissue, whereas in preterm babies a smaller volume of alveolar fluid should be drained in thicker tissue [32], and this partly explains why TTN is more common at greater gestational age. Nonetheless, the occurrence of TTN is also related to the absence of a concomitant relevant surfactant deficiency and the consequent reduced lung expansion typical of RDS. The prevalence of surfactant deficiency is inversely proportional to gestational age [33], so TTN is unlikely to be diagnosed in extremely preterm neonates who frequently have RDS as their lungs will be significantly collapsed with no relevant fluid retention. Conversely, moderately preterm infants are less frequently affected by RDS and their lungs can retain enough fluid to cause TTN. Lung ultrasound findings are not influenced by gestational age and, if the genesis of respiratory failure mainly resides in fluid accumulation, patients show heterogenous interstitial patterns alternating with at least one normal zone. If there is relevant surfactant deficiency, lung ultrasound findings show a homogeneous loss of lung aeration appearing as a diffuse alveolar pattern. This translates into a significantly higher lung ultrasound score (LUS) which can distinguish TTN from RDS (table 1) [27].

FIGURE 2.

Modern pathophysiology concepts depicted in two fictitious patients of 32 (left side) and 37 weeks’ (right side) gestation with transient tachypnoea of the neonate (TTN). a) The alveolar space (i.e. the number of alveoli) that can be filled with fluid is greater at 37 than at 32 weeks. b) The density of lung tissue (i.e. the thickness of the lung interstitium), that is the amount of tissue that should drain the fluid, is greater at 32 than at 37 weeks (illustrative sketches modified from publicly available microphotograph of autoptic lung tissues). c) Both neonates with TTN have similar lung ultrasound findings (i.e. interstitial pattern with discrete B-lines; more mature neonates may have a slightly thicker pleural line) depicting a relatively reduced air/fluid ratio (illustrative images obtained from routine lung ultrasound in neonates of 32 and 37 weeks’ gestation with TTN; upon neonatal intensive care unit admission, parents gave written consent for the anonymous use of imaging and clinical data). d) The interdependence of respiratory distress syndrome (RDS) and TTN prevalence (triangles) over gestational age; this panel is independent of the two illustrative cases affected by TTN. Darker grey corresponds to higher prevalence: RDS is more frequent at lower gestational age and prevents the occurrence of TTN given the lung collapse due to surfactant deficiency; TTN is more common at higher gestational age when surfactant deficiency is less frequent and a wider alveolar space can be filled with fluid. Mixed forms of respiratory failure with co-existing mild surfactant deficiency and fluid accumulation may also occur.

The distinction between excessive fluid and delayed re-absorption may seem purely academic, but is important as it explains some forms of respiratory failure resembling TTN that are sometimes observed in preterm infants, particularly in moderately preterm ones (i.e. between 32 and 34 weeks’ gestation). These cases might have been misclassified with consequent underestimation of TTN prevalence in preterm infants.

Despite the lower prevalence of RDS at higher gestational age, the two pathogenetic mechanisms (surfactant deficiency and fluid accumulation) may coexist. In fact, term neonates 1) with clinically diagnosed TTN have a smaller endogenous surfactant pool than normal healthy neonates [34] and 2) those with long-lasting TTN have an amount of endogenous surfactant similar to that observed in preterm infants with RDS [34]. Consistently, surfactant deficiency may eventually translate into poor surfactant function in TTN patients requiring long oxygen supplementation [35]. These data support the co-existence, in some cases, of surfactant insufficiency and fluid accumulation, although they were provided by suboptimal techniques such as lamellar body count and the stable micro-bubble test [36].

Ultrasound evaluation of lung aeration as a global effect of these two mechanisms has proven accurate in guiding surfactant administration in late preterm and term infants [37], and LUS accurately distinguishes clinically defined TTN and RDS [27]. This helps to decide whether a patient needs enhanced respiratory care, although the prevalent mechanism determining this need is unknown. As yet, we lack simultaneous measurement of extravascular lung water and surfactant pool or function. Therefore, the relative contributions of the two mechanisms are unknown. Ultrasound-assessed lung aeration is influenced by both mechanisms and so provides a more comprehensive evaluation of lung disease severity. In fact, lamellar body count used alone failed to guide surfactant replacement even in homogeneous populations of preterm infants with RDS [38].

Biology

Our understanding of lung biology in several neonatal lung diseases has recently improved and new information on the pathobiology of TTN has emerged. Lung fluid elimination is mainly accomplished by two families of molecules. First, epithelial Na⁺ channels (ENaCs), expressed by type-II alveolocytes and airway epithelial cells, are responsible for Na⁺ uptake across the airway epithelium towards the interstitium; ENaC activity reverses the osmotic gradient and leads to airway liquid reabsorption [39]. Second, aquaporin channels, mainly expressed on type-I alveolocytes, enable most water passage [40]. Expression of these channels and the Na⁺-K⁺-ATPase pump increases with gestational age [26]. As surfactant production follows the same incremental path, this supports the co-existence of the two mechanisms; in other words, some neonates may have surfactant deficiency and insufficient fluid accumulation, as both surfactant components and channels dedicated to fluid reabsorption are insufficiently expressed [41].

Overall, at the end of pregnancy there is ≈25–30 mL·Kg⁻¹ of alveolar fluid which approximates the neonatal functional residual capacity and, at birth, at least 100 mL should be cleared in a normal term neonate to avoid significant oedema leading to TTN [39]. Failing this, the alveolar protein concentration can be higher than normal and might contribute to surfactant dysfunction [26]. Mutations of ENaC and several aquaporin genes have been described but none has yet been linked to the risk of TTN. The expression of these proteins has been successfully studied in cells recovered from gastric aspirates obtained at the birth [42] and future similar studies may shed light on the genetic predisposition to TTN and its association with maternal and childhood asthma [19, 43].

Several molecules control the expression and activity of ENaCs and aquaporins and may represent drug targets. For instance, corticosteroids promote fluid re-absorption by upregulating epithelial ENaC gene expression in type-II alveolocytes and boost surfactant production [44, 45]. They may therefore have a dual role and this explains the positive effects of antenatal steroids given to late preterm foetuses. They may prevent respiratory failure by targeting the two pathophysiological mechanisms that might co-exist in this population. The universal administration of antenatal steroids in cases at risk of late preterm delivery is still discussed, given the uncertainties in long-term outcomes and cost-effectiveness, but has been suggested by some scientific societies [46, 47]. ENaC expression and direct activation are also promoted by endogenous cortisone, thyroid hormones, vasopressin and the epinephrine surge during labour [48, 49]. Alveolar epithelial β-adrenergic receptor activation also accelerates active sodium transport and this may explain the clinical effects of salbutamol observed in some TTN trials [50, 51]. β-Agonist drugs may also work by activating the Na⁺-K⁺-ATPase pump, thereby contributing to sodium and water transport [52]. None of these control mechanisms seems prominent or indispensable and their redundancy likely ensures minimal fluid clearance to allow gas exchange at birth [26].

Imaging

The main novelty in imaging has been the introduction of lung ultrasound, which detects the loss of lung aeration more accurately than conventional radiography, besides being a quick, easy, radiation-free and repeatable imaging tool [17]. Due to these characteristics, lung ultrasound has yielded a deeper understanding of TTN pathophysiology and its clinical course [27]. Copetti et al. [53] described the ultrasound appearance of TTN for the first time in 2007 and their preliminary work was followed by several other studies [17]. TTN appears as an inhomogeneous lung disease with normally aerated zones (i.e. with A-lines) alternating with zones with interstitial patterns (i.e. with discrete B-lines, indicating a reduced air/fluid ratio) representing the partially re-absorbed lung fluid. There should be at least one normal zone and no consolidated area (i.e. zones with total loss of lung aeration) (table 1). Thus, TTN is the neonatal lung disorder with the greatest aeration heterogeneity together with bronchopulmonary dysplasia, although with a different distribution of raw aeration patterns [54]. TTN has lung zones with normal and interstitial patterns, whereas these are less present in bronchopulmonary dysplasia patients, who have a significant number of zones with consolidations and alveolar patterns [54].

The interstitial pattern may be more frequent on the right side for anatomical reasons and A- and B- zones might be seen in a unique picture, representing what has been called a “double lung point” [53]. The international, multicentre ATTENTION study has demonstrated that the double lung point can only be seen in approximately 50% of patients, depending on the severity of TTN, the probe/chest size ratio and the distribution of the oedema, so this sign is not an absolute requirement for TTN diagnosis [55].

In TTN patients, lung aeration as assessed by neonatal LUS (i.e. a quantitative score calculated on six chest areas, three per side, ranging from 0 to 18, with a higher score corresponding to worse lung aeration [56]) improves over 72 h [27, 55]. LUS significantly correlates with the degree of dyspnoea and oxygenation impairment [55, 57] and easily distinguishes between TTN and RDS (table 1). In fact, LUS values on the first day of life commonly range between 4 and 6 for TTN, and between 10 and 13 for RDS [27]. During the first 72 h of life, LUS is always lower for TTN than for RDS [27]. Thus, these characteristics predict surfactant need in late preterm and term neonates using LUS as a replacement or triage test. As a replacement test, an LUS value >8 identifies patients needing surfactant with the highest global accuracy; whereas, as a screening test, values <4 make the need for surfactant extremely unlikely [37]. Since LUS can be easily and accurately calculated by transportation teams [58], this may have profound implications for clinical care as LUS may prevent useless transfers of late preterm and term neonates from level I/II perinatal (spoke) centres to level III, referral hospitals (hubs) with optimised availability of NICU beds and potential public health advantages [1].

Therapeutic approach

Therapeutic approaches to TTN over the years have intensified recently, based on improved pathophysiological and biological knowledge. This is not a formal meta-analysis, rather we summarise data from randomised clinical trials of therapeutic interventions supported by recent pathobiological knowledge.

Methodology

The methodology commonly used for systematic reviews in neonatal respiratory critical care was applied [36, 59, 60]. We searched all relevant randomised controlled trials on PubMed (on 19 April 2024) with the following key words and MeSH terms: “transient tachypnoea”, “neonate”, “beta agonist”, “adrenergic”, “diuretics”, “steroids”, “CPAP” and “ventilation”. We limited the search to newborns and randomised clinical trials. No language, geographical or year restrictions were applied. We excluded “grey” literature, unpublished or nonpeer-reviewed reports. The reference list of retrieved articles and authors’ personal archives were also searched. Details of all studies were included in a dedicated database, removing duplicates. Two authors (C.N. and D.D.L.) reviewed abstracts and, when necessary, full texts of the remaining articles, excluding those that were off topic. Data from included trials were extracted independently by these two authors and cross-verified. If further clarifications were needed, authors were contacted (at least two emails 1 week apart). Data collected included study design, year and main trial characteristics, enrolled patients, gestational age, and main results. Outcomes were expressed as mean difference (95% confidence interval) between the trial arms. When, for a given outcome, an intervention was investigated by multiple trials, we applied the inverse variance, restricted maximum likelihood method with DerSimonian–Laird random effects to summarise the effect. Analyses were performed and Forest plots were drawn with JASP (v.0.17.1; JASP Team (2023)); p-values <0.05 were considered significant.

Results

Supplementary table S1 summarises the main characteristics and results of the retrieved randomised trials on TTN. Of 23 trials found, 14 were published in the last 10 years. Most had a single-centre design and were relatively small; 10 did not report a clear blinding method. All trials enrolled late preterm and/or term infants. We focused on three outcomes that were more commonly analysed and reported, namely 1) duration of tachypnoea, 2) time needed to reach full feeding and 3) duration of hospital stay.

Trial results are summarised in figure 3. The duration of tachypnoea was used as outcome for β₂-agonist (two trials), diuretics (two trials), CPAP versus free oxygen supplementation (one trial), noninvasive high-frequency percussive ventilation (NHFPV) versus CPAP (one trial), noninvasive positive pressure ventilation (NIPPV) versus CPAP (one trial) and biphasic positive airway pressure (BiPAP) versus CPAP (one trial). The time needed to reach full enteral feeding was used as outcome for β₂-agonist (three trials), epinephrine (one trial), corticosteroids (two trials), BiPAP versus CPAP (one trial). The duration of hospital stay was used as outcome for β₂-agonist (eight trials), corticosteroids (two trials), diuretics (two trials), fluid restriction (one trial), CPAP versus free oxygen supplementation (one trial), NIPPV versus CPAP (one trial) and BiPAP versus CPAP (one trial).

FIGURE 3.

Effect size of main therapeutic interventions for transient tachypnoea of the neonate (TTN). a) Duration of tachypnoea, b) time to full feeding and c) length of hospital stay are shown. Full squares and horizontal lines represent mean differences and their 95% confidence interval, respectively. The vertical dashed line represents the absence of effect. Salbutamol was the commonest β₂-agonist used in the trials, but analogous molecules were also used. Corticosteroids were given as inhaled budesonide. ^#: Different diuretics at different dosages have been trialled and data were aggregated for this analysis. Noninvasive high-frequency oscillatory ventilation (NHFOV) was investigated in one trial and this is not shown graphically as data were only available as medians [64]. Main data of meta-analysed trials are available in supplementary table S1. BiPAP: biphasic positive airway pressure; CPAP: continuous positive airway pressure; NHFPV: noninvasive high-frequency percussive ventilation; NIPPV: noninvasive positive-pressure ventilation.

The only statistically significant interventions were:

• β₂-Agonist to shorten the duration of tachypnoea, the time to full feeding and the hospital stay (compared to placebo).

• CPAP to reduce the duration of tachypnoea (compared to oxygen supplementation with no distending pressure).

• NHFPV to reduce the duration of tachypnoea (compared to CPAP).

• Fluid restriction to shorten the hospital stay (compared to normal fluid intake).

• BiPAP to shorten the hospital stay (compared to CPAP).

These results should be seen in terms of clinical management and cost-effectiveness (see next section). The first important issue is that the benefits resulting from these interventions have small absolute values. For instance, the duration of tachypnoea is shortened by less than 1 day; full feeding is reached approximately 30 h in advance and hospital stay is shortened by approximately 1 day. From a single patient perspective these are not major advantages. Nonetheless, since TTN is not associated with mortality or with other main outcomes, it is impossible to provide major improvements with any new intervention. Second, there are no data about a possible synergistic effect of therapies with distinct mechanisms; we do not know, for instance, if the joint use of β₂-agonist and fluid restriction shortens hospital stay by 2 or more days. Third, the data on ventilatory modes demonstrate that the increase in alveolar pressure is beneficial as it prevents the air/fluid ratio decrement and facilitates fluid reabsorption, by pushing alveolar liquid towards the interstitium. However, it is unclear if CPAP is sufficient or whether other more complex noninvasive respiratory support techniques should be used, as the data are not consistent. Also, these techniques may be used with different parameters and it is unclear what the best mean airway pressure (P_aw) would be. NIPPV, noninvasive high-frequency oscillatory ventilation (NHFOV) and BiPAP may increase P_aw, but their effect is inconsistent and P_aw can also be safely increased just by raising CPAP [61, 62]. NIPPV and NHFOV may provide true ventilation and improve CO₂ clearance, but the clinical usefulness is unclear, since CO₂ does not accumulate in CPAP-treated neonates with TTN [27]. Interestingly, NHFPV, because of its physical characteristics [63], may attract alveolar liquid and secretions toward the proximal (upper) airways, facilitating their elimination. NHFOV shortens median time to full feeding and hospital stay by 22 and 37 h, respectively [64], but these data are not shown in figure 3 as they were only available as medians and could not be graphically compared with the others. Moreover, both NHFPV and NHFOV were only investigated in one small trial each and these data need to be replicated [64, 65]. Finally, it is questionable whether a relatively mild condition like TTN, where CPAP seems quite effective, needs more complex respiratory support techniques such as NHFPV and NHFOV, which have peculiar physiological features, are not available in every NICU and need specific training [63, 66, 67]. Also, nursing, patient positioning, feeding schedule and sedation may influence both the efficacy and safety of these ventilatory modes and have not been sufficiently investigated. Finally, safety is ensured for some drugs and respiratory support modes long used for neonatal indications other than TTN; nonetheless, other interventions, such as fluid restriction, were tested in few trials and we lack data on their safety.

This analysis has limitations inherent to the design of the original trials (e.g. single-centre, small populations) or the outcome definition. In fact, the time needed to reach full feeding and the duration of hospital stay are relatively weak outcomes that may be influenced by local protocols, logistics, and other medical and social factors. This makes the available results less interesting from a single patient perspective, but does not reduce the potential usefulness on a population scale in terms of public health, which will be treated in the next section.

Practical advice for real-world NICU care

Despite these limitations, our accumulated knowledge yields suggestions to clinicians diagnosing and managing TTN. Quantitative lung ultrasound is surely the easiest and most informative technique to be incorporated into routine care. Its learning curve is steep [68, 69] and approximately 2 weeks is sufficient to learn how to distinguish TTN from RDS, even in low-resource settings [70]. Basically, any probe can be used although linear ones should be preferred, at least in beginners’ hands [71]. In our experience, a few months suffice to integrate lung ultrasound in the daily life of the NICU and, once realised, there is significant benefit in terms of diagnostic accuracy. Electrical cardiometry is valuable for research purposes and hemodynamic monitoring [72], but its accuracy in measuring extravascular lung water has yet to be investigated and, unlike ultrasound, it requires a specific device that may not always be available.

From a therapeutic point of view, CPAP is surely to be preferred as first-line respiratory support, given its wide availability, proven safety and relative efficacy. Nasal masks should be preferred as an interface [73]. The optimal P_aw is still to be determined. Although CPAP levels beyond 6 cmH₂O seem safe [61, 74], they may afford little improvement in terms of oxygenation and aeration [75]. More complex respiratory support techniques should be avoided as they lack supporting data and need specific training and equipment. In addition, interfaces have complex interactions with patients and play an important role in different types of noninvasive ventilation; nasal masks may not always be the optimal interface [76, 77]. Ancillary therapies may have a place in case-by-case evaluation in more severe cases or when public health issues push for shorter hospital stays. In these cases, nebulised β₂-agonist and pronation may be considered. Although no trials of prone positioning in TTN are available, pronation is a safe and inexpensive technique commonly used in the NICU. Pronation improves gas exchange and lung aeration by providing net recruitment in several neonatal lung disorders [78, 79], particularly in inhomogeneous lungs. As TTN is characterised by high aeration heterogeneity [54], prone positioning might be used.

As many TTN babies are cared for in I/II level perinatal centres (spoke hospitals), appropriate telemedicine tools may help paediatricians manage them without transfers to referral NICUs, since transportation may worsen clinical severity [80]. For example, lung ultrasound can be evaluated remotely by sending short videoclips, which may also be useful to evaluate patient dyspnoea. Adequate advice and clinical monitoring can then be provided by expert NICU physicians remotely. Telemedicine is relatively new in neonatology, but was successfully deployed during the recent pandemics [81] and even tested for NICU care of critically ill infants [82].

Depending on local organisation of care (particularly in regions with numerous spoke hospitals and few hub NICUs), the institution of telemedicine networks may reduce NICU utilisation and yield the public health benefits described in the following section. Similarly, these networks may be useful in low- to middle-income countries [83].

Public health issues and cost-effectiveness of TTN treatment

Despite its favourable outcomes, TTN represents a major public health issue. Several thousand neonates with TTN require hospitalisation each year. CPAP is the commonest treatment and its use is increasing over time [84], but it requires admission to an NICU, which is a highly technological environment with high costs. These latter are approximately five times higher than those of gestational age-matched infants without TTN [85]. If we consider that TTN is the commonest neonatal respiratory disorder, its economic impact appears evident. In addition, TTN may have familial, logistic and psychological consequences, which may be difficult to assess but impact perinatal and neonatal well-being.

Table 2 summarises the estimated financial costs associated with NICU care of TTN in five developed countries as illustrative examples. This estimation does not include patient transfer costs (when delivery occurs in a low-level perinatal centre), familial and societal problems (e.g. loss of working days, logistic costs, parental anxiety and breastfeeding problems) and the risk of NICU beds being unavailable for other patients, including those with life-threatening conditions. This may occur because neonatal emergencies are unpredictable and NICU beds are generally insufficient, particularly during winter when seasonal outbreaks of respiratory infections occur [2].

TABLE 2.

Estimated costs associated with critical care of transient tachypnoea of the neonate (TTN) patients in five developed countries

	NICU hospitalisation cost EUR per day	TTN cases per year	Min–max estimated cost (EUR)	Estimated yearly gain EUR per NICU day
Canada	2000	3520	35–42 M	−7 M
France	1500	7000	53–63 M	−11 M
Italy	1000	3930	20–24 M	−4 M
Spain	2500	3200	40–48 M	−8 M
UK	900	6000	28–33 M	−5 M

Calculations were performed considering the 2002 publicly available data on the number of births per country, a mean TTN prevalence of 10‰ [3] and daily cost of neonatal intensive care unit (NICU) care as reported by local studies [95–99]. A mean hospitalisation of 5–6 days has been considered to generate a minimum–maximum estimated cost. The estimated yearly gain is calculated per spared day of all NICU stays. Costs were in Euros as per the exchange rate on 20 April 2024. Numbers are rounded to the nearest integer.

Additionally, some rare cases of TTN may actually become life-threatening (so-called “malignant TTN”) [86]. In fact, infants with TTN may have subclinical right ventricular dysfunction owing to low alveolar oxygen concentration due to fluid accumulation [87]. Occasionally, this may result in severe pulmonary hypertension due to absorption atelectasis and oxidative stress induced by high-concentration oxygen supplementation, particularly if CPAP or other respiratory support is not provided [39]. The care of these patients may be much more expensive as they may die or require highly technological monitoring, extracorporeal life support and longer NICU hospitalisation [86].

In this context, sparing just 1 day of NICU care might translate into a relevant financial gain; reducing NICU stay by 1 day would spare between EUR 4 and 11 M from a countrywide perspective. With synergy between treatments, healthcare systems may save at least ≈EUR 10 M per year in any of these countries (table 2). Considering the small budgets usually dedicated to neonatology compared to adult medicine, this is a relevant result, particularly for countries with publicly funded healthcare systems. Thus, the aforementioned treatments may become valuable even if their absolute effect is not clinically relevant for a single patient. This obviously cannot be considered a formal cost-effective analysis, as results may be influenced by local admission criteria and other factors. More precise and granular data are needed for accurate analysis, including hospitalisations and treatments performed out of the NICU as well as the other costs mentioned above. Nonetheless, these data depict the public health burden of TTN and the value of interventions that influence its clinical course. Moreover, it is known that ICU admission is the main determinant of hospital costs [88]; therefore, if an intervention reduces NICU stay, this may translate into improved NICU bed availability, decreased inter-hospital transfers and fewer lost workdays and other negative consequences for families.

Finally, there are no data regarding the burden of care associated with TTN in low- to middle-income countries, but it is logical to believe that outcomes may be worse and that optimisation of TTN care may achieve relevant benefits in this context too.

Future steps

Many issues still need to be addressed. Larger, high-quality, multicentre trials with shared co-interventions and integrated economic evaluations are needed to optimise the benefit of possible therapeutics and increase awareness of TTN's importance in public health value. Widespread dissemination of diagnostic criteria is needed, although this is surely an easier task than the conduct of large trials. With this in mind, honest discussion with healthcare authorities is warranted to improve perinatal care organisation. This should include dissemination of quantitative lung ultrasound and telemedicine tools, to prevent unnecessary newborn transfers and keep NICU beds available for more critically ill neonates. A modern rethinking of the commonest neonatal respiratory disorder will enable better care and patient and public health benefits.

Points for clinical practice

Recent advances in understanding of the pathophysiology and biology of TTN have clarified the mechanism of action of any possible therapeutics. We review the available therapies and suggest their use from a public health perspective to optimise NICU resources.

Questions for future research

We suggest several important lines of research for the future, including the outline of a new, pathophysiology-based definition of TTN. Since TTN has a virtually nil mortality, large, multicentre trials with shared co-interventions and integrated economic evaluations are needed to optimise the benefit of possible therapeutics. The implementation of telemedicine tools should also be studied to prevent unnecessary patient transfers and optimise NICU resources.

Supplementary material

Please note: supplementary material is not edited by the Editorial Office, and is uploaded as it has been supplied by the author.

Supplementary material ERR-0112-2024.SUPPLEMENT ^{(309.3KB, pdf)}