Year XVII n. 3/01

In the neonate the onset of a prompt and effective respiration is of paramount importance for its survival, particularly in cases of preterm birth and of other conditions at risk. In turn, adequate ventilation is dependent upon the stability of the lung and hence its resistance to collapse at the end of expiration, and therefor upon the biochemical and morphologic maturity of the lung. When the maturation of the fetal lung has not reached a full stage, invariably the neonate will develop a syndrome called neonatal respiratory distress syndrome (NRDS). In developed countries it represents the main cause of perinatal mortality and morbidity. Not long ago this disease - otherwise called “hyaline membrane disease” - resulted in the demise of the majority of preterm infants. Eventually, it became apparent that the early pathologic features of hyaline membrane disease include atelectasis and intra-alveolar edema resulting from a delayed resorption of the fetal pulmonary liquid (FPL) and transudation of plasma proteins with the formation of hyaline membranes, which can be lacking in the lungs of the neonates who die within few hours from birth. Consequently, it was concluded that NRDS is a disease characterized by inadequate production of pulmonary surfactant (surface active material) and in part is related to the structural immaturity of the lung and of the cells involved in the synthesis of pulmonary surfactant (Van Golde, 1988). NRDS has a relatively short scientific history. In the early nineties, the so-called hyaline membrane disease was described in association with demise due to respiratory failure. Von Neergard in 1929 first described in the lungs the existence of an air-liquid interface.

There were no substantial progresses in the understanding of lung immaturity until Pattle (1955) and Clements (1957) noted the presence of surfactant in pulmonary edema foam and lung extracts.

In 1959 Avery and Mead correlated respiratory failure with decreased surfactant levels in saline extracts of the lungs of infants with RDS. The quasi-static pressure-volume curves of the lungs of infants succumbed of NRDS and of the lungs of control infants are depicted in Fig. 1.

Once the association between atelectasis with hyaline membranes and surfactant levels was appreciated, a large research effort was focused on the development of the surfactant system of the lung. A significant support to surfactant research was the large financial sustainment granted by the Kennedy family following the death form NRDS of the son Patrick of John Fitzgerald Kennedy. The main pathologic feature of NRDS is atelectasis, as the consequence of the collapse of the alveoli at the end of the expiratory phase, resulting from surfactant deficiency.

This is associated with an increased opening pressure required during insufflation for the recruitment of new peripheral airways; at the end of expiration, the alveoli of the immature lungs tend to collapse, with a pattern resembling the non aerated fetal lungs, in which the alveoli are filled with liquid (FPL). Other forms of neonatal respiratory failure include a transient form of RDS (type II) characterized by tachypnea (otherwise called TTN, transitory tachypnea of the newborn), caused mainly by a delayed resorption of FPL. Another form of respiratory failure is the Adult Respiratory Distress Syndrome (ARDS), described as an independent clinical entity in 1967 (Ausbaugh et al). The syndrome is relatively frequent and often lethal: 150,000 new cases occur yearly in the USA and 60,000 in Europe; despite aggressive management, the mortality rate is about 50%. ARDS occurs in association with a variety of diseases, including trauma, shock, sepsis, viral infections and others. The clinical appearance of ARDS include a reduce pulmonary compliance, associated with bilateral pulmonary infiltrates and a reduced PaO2, resulting from wide-spread pulmonary “shunts” and a resistance to mechanical ventilation, even in the presence of elevated FiO2 (concentration of inspired oxygen) and PEEP (positive pressure at end expiration) (Saitto et al, 1983).

Incidence and Pathophysiology of NRDS

The exact incidence of NRDS is difficult to ascertain, especially in unselected infants, due the difficulty to differentiate RDS from other forms, usually less severe, of neonatal respiratory diseases. The differential diagnosis is particularly difficult in very low birth weight infants, where the typical X-ray appearance (aerial bronchogram) is not always associated with hyaline membranes. Recent studies indicate that the incidence of NRDS ranges from 0.3% in live born in Scandinavian countries, to 0.8% in Switzerland and 1% in USA. Among the various factors conditioning the incidence of NRDS, the first is gestational age. The incidence of NRDS is inversely related to gestational age, with a rate of 30% in infants born before 30 weeks of gestational age and as low as 0.01% in term infants (Reed et al, 1978). Another factor is the sex of the infant. A number of observations have shown that NRDS is more frequent in males, with a ratio males/females of 1.5-2.0.

The difference is likely to be associated with the action of the hormones involved in sexual differentiation.

Testosterone, with its inherent anti-estrogenic activity, has been shown to play an important role in inhibiting, in vivo and in vitro, the differentiation of type II pneumocytes, the cells that are responsible for the synthesis and secretion of pulmonary surfactant. Recently, it was demonstrated that male fetuses are exposed in utero to higher levels of a glycoprotein called MIS (Mullerian inhibiting substance); the protein, synthesized by Sertoli cells, essential for normal male sex development, inhibits the production of phosphatydylcholine, the principal component of surfactant. A familial predisposition seems to be associated with the occurrence of NRDS. A recent study has shown that the incidence of NRDS is 90% if the previous low birth infant was affected by RDS and only in 5% in the case the previous child was not affected. Certain pregnancy complications are associated with NRDS. In diabetic pregnancy, fetal lung maturity is related to the different classes of gestational diabetes, with a global risk 6 times greater than the normal population. The delay in fetal lung maturity is associated to Priscilla White’s classes A,B and C, while FLM is accelerated in classes D, F and R (presumable as the results of chronic fetal distress and to increased endogenous production of cortisol). In spite of this, there is a general agreement that in diabetic pregnancies the risk of NRDS is greater, even in the presence of mature amniotic fluid test. The increased risk is the result of the inhibitory action of insulin on the production of pulmonary surfactant. Fetal allo-immunization is also associated with a delayed FLM. It seems that elevated fetal levels of insulin may adversely affect the production of surfactant, in a manner similar to diabetes in pregnancy. Other authors have suggested a causative role played by the degree of fetal anemia responsible for hypoxia thereby resulting in a reduced production of surfactant. Another risk factor for NRDS is cesarean section.

The lack of the surge of catecholamines (particularly epinephrine) characteristic of labor and vaginal delivery without anesthesia seems to play a role, being epinephrine involved in the release of pulmonary surfactant (Olver, 1981) Twin pregnancy seems to play a protective role for the occurrence of NRDS. However, twin pregnancy is not associated per se to an accelerated lung maturity, unless other pregnancy complications, including pregnancy induced hypertension and/or preterm premature rupture of the membranes (PPROM), both associated with accelerated fetal lung maturity, are present. NRDS is seldom a complication after PPROM, if the latency period (interval between PPROM –delivery) is greater than 24 hrs.

The protective effect of PPROM is apparent in neonates with birth weight between 1,500 and 2,500 g and a gestational age between 33 and 36 wk.

A similar protective effect is present in those conditions that are associated with a chronic fetal distress, such as pregnancy-induced hypertension (PIH ) or IUGR (Intrauterine Growth Retardation) (Lee et al, 1976).

In other studies, it was demonstrated that heroin addiction is associated with a lower incidence of NRDS, since heroin seems to promote the synthesis of surfactant by type II cells.

The pathophysiology of NRDS has remained rather mysterious until the discovery of surfactant (Pattle 1955; Clements 1957). As previously mentioned, in 1959 Avery and Mead first demonstrated a deficit of surfactant in the lung extracts of infants succumbed to RDS, that was then considered as the main pathophysiologic feature of the disease. Surfactant reduces surface tension (regulated by the La Place law) of the air-liquid interface of the alveolar wall as low as 0 dyne/cm (0 mN/m), thereby eliminating the tendency of the alveolus to collapse during expiration. Therefore, the alveoli are kept open with lower pressures. Normally, the first breath after birth requires a high pressure to open the airways and to progressively recruit peripheral breathing units. In the presence of normal surfactant levels, which reduce surface tension favoring the initial expansion and the sequential recruitment of small airways, and prevents the collapse of the alveoli, the lung retains more than 20% of the residual air after the first breathi.e., the functional residual capacity.

The next breaths require lower inspiratory pressures. In the presence of a deficit of surfactant, the lungs tend to collapse at the end of expiration and therefore the neonate has to sustain a greater work in the following respiratory efforts, equal to the one of the first respiration.

The progressive atelectasis and reduced lung “compliance” result in a cascade of events characterized by: alteration of ventilation/perfusion ratio (irregular V/Q) associated with hypoventilation, hypoxemia and hypercarbia followed by respiratory acidosis, then pulmonary vasoconstriction with pulmonary hypoperfusion and endothelial damage. This results in transudation of protein and fibrin rich material into the alveolar spaces with the formation of “hyaline membranes”.

These membranes are made of degenerated epithelial cells, blood cells, fibrin and components of the alveolar wall (Laweryns, 1965); they constitute a barrier to gas-exchange, aggravating hypoxemia, hypercarbia, and leading to acidosis.

Hypoxemia by itself seems to further alter the synthesis of surfactant. Over the last years several studies have confirmed that the lungs of the infants affected by NRDS have a high surface tension and tend to collapse. The observation is consistent with the abnormal quasistatic pressure-volume curves in the lungs of the neonates affected by RDS: to open and recruit the peripheral airways high opening pressures are required and at the end of expiration the lung collpases. This is reflected by a reduced hysteresis between the insufflation and the deflation curves (Fig. 2).

A reduced surface tension is therefore of paramount importance to maintain the stability of the alveoli, and this is even more important since the infant with respect to the adult has a less rigid thoracic cage and therefore the forces opposing to alveolar collapse are of reduced entity. Many theories regarding the pathophysiology of NRDS have been put forward, being mostly focussed on the role played by the lack of surfactant as the result of prematurity. Among other factors advocated in the pathophuysiology of NRDS, there are pulmonary vasoconstriction with hypopefusion, cesarean section, cardiac insufficiency, low circulating levels of plasminogen, hypercoagulability, immunologic factors.

Surfactant system of the lung

Pulmonary surfactant is a complex mixture of lipids and proteins with tensio-active properties, lining the surface of the alveoli and of the terminal airways. The alveolar surface is composed of: a) the wall of the alveolus, made up of a mono-layer of cells interrupted by the pores or the channels of Kohn. Of the epithelial cells, 85% are type I cells whereas the remaining 15% are type II cells (pneumocytes) with cytoplasmic granular inclusion bodies (the lamellar bodies) and microvilli on the surface. Type II cells are the factory and the repository of pulmonary surfactant and act as precursor of type I cells; they also have phagocytic activity; b) the film lining the wall and the pores in contact with the air of the alveolus. The liquid phase of the limiting lining is called hypophase and its surface is the surface film or air-liquid interface. The surfactant molecules are concentrated at the air-liquid interface, with the polar moieties (hydrophilic) oriented in the liquid phase and the non polar moieties (hydrophobic) in the air phase.

These molecules reduce the surface tension at the air liquid-interface, and result in alveolar stability of the alveoli at the end of expiration thereby preventing their collapse.

Other functions of the pulmonary surfactant include: anti-edema properties, by influencing the trans-epithelial movements of fluid; stimulation of macrophage-mediated phagocytosis of bacteria, with a definite role in the host defense of the lung. Surfactant contains 90% lipids and about 10% proteins. Among lipids there are surface active phospholipids, i.e., disaturated lecithin, phosphatydylcholine saturated with two molecules of palmitic acid (PC),, lysolecithin, phosphatydylglycerol (PG) sphyngomyelin and phosphatydylethanolamine. The most abundant molecule is disaturated lecithin, whereas PG represents 11% of the surface active molecules. Its importance is highlighted by the observation that the immature neonates whose lung surfactant is lacking PG have a higher occurrence of RDS. The presence of specific surfactant proteins was recognized only in 1973. Over the last few years 4 surfactant-specific proteins have been characterized and their function in part elucidated. According to the nomenclature introduced by Possmayer in 1988, the 4 proteins are called SP-A, SP-B, SP-C and SP-D, respectively. According to their solubility in water or organic solvents, two main groups have been identified. SP-A and SP-D are hydrosoluble, whereas SP-B and SP-C are small hydrophobic proteins, which derive from clivage of pre-proteins of greater dimension.

Surfactant-associated proteins play a number of functions. SP-A and SP-B are essential for the formation of tubular myelin and for assembling the lamellar bodies, that are both important for the storage and secretion of surfactant. SP-A is a metabolic regulator that control the secretion and the re-uptake of other surfactant components through a specific membrane receptor on type II cells and it is a non immune host defense protein. SP-D has been recently isolated and has many structural and functional similarities with SP-A, although its precise role is still to be ascertained. SP-D deprived surfactants conserve their surface active properties; however, a role of SP-D in the host defense properties of the lung has been suggested. The synthesis of surfactant begins at a relatively late stage of pregnancy, beginning between the 20-25 wks; after 33 wks it has the capacity of stabilizing the alveoli. Surfactant synthesis takes place in the smooth and rough endoplasmic reticulum of type II cells, that begin to differentiate between the 20 and the 24c week of gestation (alveolar stage of the lung structural development).

The differentiation of type II cells requires a close contact with lung fibroblasts which in turn produce a peptide known as “ fibroblast pneumocyte factor” promoting the synthesis of surface-active material; its release is positively controlled by glucocorticoids.

The lipid and protein fractions of surfactant are, therefore, synthetized by type II pneumocytes and stored in intracellular organelles called “lamellar bodies”.

The secretion of these structured is mediated by exocytosis, by the fusion of their external membrane with the apical cell membrane and subsequent diffusion of their content in the alveolar space. Surfactant secreted into the alveolar spaces have three different pathways: recycling, degradation or removal.

During recycling surfactant components are re-used, following up-take by type II cells and re-introduction into lamellar bodies. Alternatively, surfactant can be degraded and its components re-utilized for the synthesis of new lipids and proteins inside the type II cell, or removed from the system in form of intact molecules or degradation products such as fatty acids.

Fetal pulmonary liquid (FPL)

Following secretion by type II cells, surfactant enters the liquid (FPL) that fills the alveoli in increasing amounts with advancing gestational age. FPL is a ultra-filtrate of fetal plasma, according to its composition, and constitutes more than 60% of the weight of the fetal lung. At birth FPL is replaced at birth iso-volumetrically by air and its volume corresponds to the lung functional residual capacity (FRC). During intra-uterine life, FPL volume is not subjected to significant variations, as shown by a number of animal studies.

At birth FPL is eliminated by two mechanisms: outflow from the trachea as the result of thoracic compression during the passage of the fetus through the vagina and, more importantly, by resorption through lymphatic drainage (Adams et al, 1971). The resorption rate is strictly related to fetal lung maturity, being slow and rather ineffective in preterm infants. At birth, during the first breath a interface between the liquid, that is subject to resorption and the air that has entered the lung is formed; the surfactant present in FPL begins to localize at the level of the interface, forming the delimiting lining of the alveolus (Fig 2). The complete resorption of FPL coincides with the formation of a mono-layer of surfactant, derived both form accelerated secretion by type II cells promoted by onset of breathing, and by surfactant suspended in the FPL.

The circumstances under which the surfactant system is altered include:

a) immaturity of the metabolic pathway driving the synthesis of surfactant, compounded by acidosis;

b) inadequate secretion of surfactant by damaged type II cells as the consequence of acute fetal asphyxia;

c) deficit of precursors (pulmonary ischemia, low blood levels, etc.)

d) surfactant inhibition (fibrinogen, plasma transudate, etc.)

e) altered resorption of FPL at birth, as result of low blood and/or lymphatic perfusion, with ineffective formation of the lining of the alveolus.

Fetal breathing movements (FB Mss)

Although fetal breathing movements (FBMss) were first observed in 1787, only starting from 1885 they have been considered as a physiologic pattern, with episodes of spontaneous breathing interspersed with periods of apnea. In spite of many research efforts on perinatal respiration and on onset of breathing at birth, little is known on the onset and maintenance of FBMss in utero. Studies on animal models have clearly shown that the fetus has rapid, although episodic, spontaneous respiratory movements starting from early gestational (Dawes, 1974). These movements are intermittent, fast, mostly shallow and irregular, with a wide range of depth and frequency. Usually, they have a rate of 40-70 FBMs/min and are present in 50% of the observation period. In some studies, FBMs are associated with an increase of fetal heart rate (FHR), of beat-toi-beat variability and to an increase of systolic and diastolic pressure. FBMs seem to have a cyclic pattern and are associated with sleep states, being most frequently associated with non-REM (rapid eyes movement) sleep. The link between FBMs and sleep states suggest that FBMs in association with body movements can condition not only the development and the growth of the lung, but even the development of the fetal brain. Fetal breathing movements are controlled by the respiratory muscles (diaphragm, and to a lower extent, inter-costal muscles) and are under the control of the central nervous system, spanning from the bulbar respiratory center up to sub-cortical areas. FBMs in utero have been subjected to intensive research. Another pattern of FBMs, called “gasps” has been observed. “Gasping” is defined as a deep inspiration of low frequency (1-4 mov/min), as opposed to “grunting” as an expiration effort. Noteworthy is the fact that the fetus can cough, may react to fever with high frequency FBMs, and can emit 6-12 times a day great amounts of FPL, suggesting a reduction of FRC. Many techniques have been used for the study of FBMs, including mercury strain transducers around the trachea and on the thoracic cage (Cosmi,1981) or maternal abdomen and more recently, ultrasound (US) techniques. Since 1970 a number of US techniques have been used to study FBMs, including M-mode, B-mode and recently color Doppler techniques. As shown in animals, FPL is normally “poured” into the amniotic cavity in greater amounts during FBMs than during apnea. However, this phenomenon was not shown in human until Doppler techniques have been utilized for the study of upper respiratory airways. Chiba et al (1985) first introduced a US method to study the fetal intra-tracheal flow. In 1991 others have studied the anatomy and respiratory patterns of the upper airways (oro-pharynx, larynx and trachea), by combining bi-dimensional US with color Doppler. The study was completed by Samvel et al (1993), that correlated nasal fluid flow velocity waveforms as determined by color Doppler flow mapping with FBMs. Nasal fluid flow, as shown by color Doppler, shows that FBMs favor the shift of liquid from the airways of the fetus to the amniotic compartment.

The factors influencing FBMs include: 1. gestational age: FBMs are observed in the second trimester of pregnancy, at different periods in relation to the type of FBMs (abdominal, thoracic, and nasal). All type occurs before the 23rd week of pregnancy; the breath-to-breath interval and the length of the inspiratory phase increase in the period between the 22nd and the 35th wk of pregnancy and decrease thereafter.

The same was observed by us in the preterm human fetus (Cosmi 1998) 2. Acid-base balance: hypoxia: a decreased maternal oxygenation that lowers fetal pO2 (from 20-24 to 1-12 mmHg) can result in reduction of FBMs. Interestingly, the responses of the fetus to hypoxia is exactly the contrary of what happens in the adult, probably as the result of immaturity of aortic chemoreceptors.- Maternal hypercarbia: increase FBMs either in the animal an in humans – Asphyxia: a prolonged occlusion of the umbilical cord results in the first place in cessation of FBMs, followed by fetal “gasps”, namely inspiratory efforts that are isolated, deep and prolonged. Acidemia: respiratory and metabolic acidosis all favor FBMs. Glycemia. Maternal administration of glucose induces FBMs, whereas fetal hypoglycemia inhibits them. 3. Labor: during labor a number of observations have shown a decrease of FBMs.

This may result form the release of endogenous prostaglandins associated with labor or by the intermittent episodes of hypoxia associated to uterine activity. 4. Smoke: the decrease of FBMs in response to smoke seems to result from the nicotine induced reduction of uterine blood flow resulting in fetal hypoxia. 5. Alcohol: ethanol ingestion by the mother causes a reduction of FBMs; the reduction is however, direct, mediated by the suppression of REM sleep in the fetus. 6. Hormones: the administration of conjugated estrogens to the mother (10 mg in bolus) was followed by increase of FBMs in association with an increase of umbilical artery blood flow (Cosmi et al, 1998) Betamethasone (4 mg in bolus at 38 wk ) and dexamethasone (4.85 mg at 29 wks) were observed to induce FBMs associated to increased umbilical blood flow, with betamethasone being a more potent inducer (Cosmi, 1998) 7. Aminophylline: the drug belongs to the family of methylxhantines (theophylline ethylendiamine). Methylated xanthines are widely use as an analeptic, in the treatment of cardiac insufficiency and as a smooth muscle relaxant (bronchi, vessels, and uterus). Other therapeutical uses of aminophylline have been recently introduced, such as in the treatment of apnea of prematurity, due to its property to stimulate the respiratory centers. As the result of the inhibitory effect of aminophylline on smooth muscle (through the inhibition of the enzyme phosphodiesterase, that catabolises the increased cAMP), it was shown to inhibit the Fallopian tubes contractions (Coutinho, 1971) and to inhibit uterine contractions (Cosmi, 1981). In another study, Cosmi et al showed that aminophylline administered to the pregnant doe can cross the placenta and increased the levels of cAMP in the fetal lung, which in turn stimulates the synthesis of phosphatydylglyerol (PG). Others have shown that the effect of aminophylline on fetal lung maturation is limited to the stimulation of the CNS respiratory center. Based on these results, Cosmi et al in 1996 have observed that the administration of aminophylline to the mother results in the onset of thoracic and abdominal FBMs, associated with nasal fluid flows that can appear as true vortexes.

The previous findings are the rationale for the in utero administration of supplementary surfactant.

The understanding of the pharmacokinetics and of the pharmacodynamics of aminophylline is essential to its correct use.

Aminophylline is rapidly adsorbed after oral, rectal or parenteral administration. It is distributed in all body fluids, can cross the placenta and is found in milk (Cosmi, 1981) In the therapeutic plasma levels range, 60% is bound to plasma proteins in the adult and 40% in the neonate. Plasma half-life is 8-9 hrs in the adult and between 11 and 57 hrs in the preterm neonate. Rapid infusions of therapeutic dose aminophylline (500 mg) can cause sudden death, probably as consequence of fatal cardiac arrhythmias.

Therefore, aminophylline should be infused slowly, during a period of 10-20 min, in order to avoid acute toxic events (headache, palpitations, vertigo, nausea, hypotension and pre-cordial pain). Other side effects include tachycardia, restlessness, and vomit): these effects are associated with plasma levels greater than 40 microg/ml. With plasma levels ranging from 25 to 40 microg/ml, focal and even generalized convulsion episodes can occur.

These effects are rare if aminophylline is diluted in 5% dextrose solutions.

Prevention and treatment of NRDS

Various pharmacological agents and hormones have been experimented to induce fetal lung maturity (FLM) thereby tending to prevent NRDS (table I). Among all drugs tested for accelerating FLM, glucocorticoids administered prenatally to the mother have proven effective in reducing the incidence of NRDS and of other complications such as intraventricular hemorrhage (IVH) and retinopathy of prematurity (ROP). The clinical efficacy of glucocorticoids prophylaxis has been shown in several clinical trials (Robertson B, 1993) and further evaluated by meta-analysis (Crowley P, 1990). Although a 20-year follow-up studies indicate that one course of antenatally administered corticosteroids to enhance FLM does not exert side effects on the adult (Dessens AB, 2000), there are several lines of evidence suggesting that antenatal corticosteroids may be detrimental to the fetus and newborn infants particularly when multiple doses are given. In the animal model, Ikegami et al. (1997) have shown that following 3 exposures to betamethasone there was a progressive enhancement of postnatal lung function in preterm lambs at the expenses of birth weights, which decreased up to 27%. In a similar study Sloboda et al. (2000) have demonstrated that multiple doses of betamethasone in the preterm lamb resulted in reduced weight at birth at 125 and 146 days, and altered basal cord levels of plasma ACTH and corticosteroid binding capacity. Recently, in the human the effect of multiple exposure of the fetus to antenatal corticosteroids has been evaluated. In a retrospective analysis the outcome for premature neonates after multiple courses of antenatal corticosteroids was compared with a single course. There was no detectable clinical difference in the incidence of NRDS, chronic lung disease, and IVH related to the courses of antenatal corticosteroids; the outcome was similar for infants delivered at 7-13 days compared with those delivered at 1-6 days after receiving antenatal corticosteroids. Compared with those who received a single course, neonates who received ->2 courses had lower birth weights (<39 g, p = .02), and those receiving ->3 courses had increased risk of death (odds ratio 2.8) and lower levels of plasma cortisol at 2 hours from birth. The results of this study indicate that multiple courses of antenatal corticosteroids did not improve the outcome of newborn babies but were associated with increased mortality, decreased fetal growth, and prolonged adrenal suppression (Banks BA, 1999). In a cohort study French et al. (1999) studied the effects of repeated antenatal corticosteroids on birth size, growth, and development in preterm infants. Birth weight ratio decreased with increasing number of corticosteroid courses (p = .001); multivariate analyses confirmed a reduction of birth weight as much as 9% (p = .014) and a reduction in head circumference of as much as 4% (p = .0024). There were no additional benefits in mortality or respiratory outcome, and there was a trend toward more severe chronic lung disease. Repeated corticosteroid courses were associated with adverse effects on size at birth without apparent benefits. In a retrospective study performed in 609 mothers and their 713 infants who were treated with 1 to 12 courses of antenatal corticosteroids, Abbasi et al (2000) observed that exposure to multiple courses of antenatal corticosteroids, as compared with a single course, resulted in a significant reduction in the incidence of RDS in singleton preterm infants delivered within a week of the last corticosteroid dose. This was associated with a reduction of head circumference at birth with an increased incidence of maternal endometritis. Among the side effects of antenatal corticosteroids there is a report from the literature in which three newborns, whose mothers were treated with betamethasone prenatally in different doses, developed various degrees of transient hypertrophic cardiomyopathy as diagnosed by echocardiography (Yunis et al, 1999). In table II are depicted the side effects of corticosteoids in the mother, fetus, and newborn. It is well established that post-natal supplementary surfactant (SS) is effective for the treatment of NRDS, although it is not devoid of complications, e.g., broncho-pulmonary dysplasia (BPD); in addition, repeated doses may be needed. This is why early or prophylactic treatment of NRDS is continuously sought. Intra-amniotic surfactant (IAS) We have indicated that the most rational approach would be the prevention rather than treatment of NRDS and that the most natural way would be to instil SS into the fetal pulmonary liquid (FPL) compartment, i.e., either 1) at birth before the first breath after endotracheal intubation of the extracted head from the vagina or from the incised uterine muscle during cesarean section (CS); or 2) in utero by direct injection of SS into amniotic fluid (IAF) close to the fetal mouth and nostrils so that it will undergo uniform distribution within the FPL. The scientific basis for this “prophylactic” IAF administration of SS spans three decades of research on fluid dynamics and related biochemistry of FPL and amniotic fluid (AF) compartments. Three lines of research have paved the way: a) Adams et al (1963) provided the first detailed and systematic studies of FPL. They defined organic and inorganic composition, surface activity, fluid movement and comparison with other liquid compartments of the maternal-fetal complex; b) Gluck et al (1967), presented meticulous studies of the chemical development of FPL and fetal pulmonary tissue; and c) Scarpelli (1968) proved the metabolic origin of FPL surfactant from pulmonary tissue and defined the lipid and protein gradients between FPL and AF. From these data, Scarpelli suggested that AF phospholipids might be used to diagnose fetal lung maturity (FLM); Gluck et al (1971) later proved this to be correct. These studies had established the biochemical interrelationship between FPL and AF. Two additional findings have reinforced our appreciation of the importance of FPL surfactants to successful transition to air-breathing at birth: 1) FPL surfactants are the main substrate for formation of intra-alveolar bubbles from FPL at the onset of air-breathing at birth (Scarpelli EM, 1978); surfactants form the ultra-thin films of bubbles that carry air to the alveoli (“saccules”) and establish normal gas exchange and alveolar stability; 2) even after therapeutic intratracheal instillation of surfactant to the postnatal infant - first reported in the pioneering study of Fujiwara et al. (1980) - significant clinical intervention is required. For example, multiple doses of SS may be needed and the position of the neonate must be changed regularly to facilitate an even distribution of SS within the lungs. In addition, the possible untoward effects of endotracheal intubation and postnatal instillation of SS must be considered, including the associated hypoxia, bradycardia and barotrauma. It must be recalled that during the first breaths preterm babies may generate intrathoracic pressure swings up to –70 cm H2O, thereby causing barotrauma; and 3) recent studies have shown that the intraamniotic injection of natural surfactant is capable of preventing pulmonary hypoplasia in rabbit fetuses with surgically induced congenital diaphragmatic hernia and can be as effective as tracheal ligation (Tannuri U et al, 1998). Whether fetal breathing movements (FBMss) result in the movement of amniotic fluid into the distal airways is critical to the therapeutic efficacy of IAS. By studying the effect of paralysis of the preterm rabbit fetus on the pulmonary distribution of iron dextran injected intra-amniotically, Galan et al. (1997) demonstrated that paralysis – by abolishing FBMss with a muscle relaxant pancuronium – prevent the uptake of iron dextran into the main and distal airways of the rabbit fetus whereas in the control untreated rabbit fetuses there was an uniform distribution of iron dextrane into the peripheral airways. Although FLF production results in a net efflux of fluid, FBMss particularly when induced by analeptic drugs, such as aminophylline (A) given to the mother, do actually result in the movement of fluid into distal airways. Another evidence of the entry of IAS into peripheral airways is the study by Hallman et al. (1997). In a rabbit model aimed to investigate the metabolism of intra-amniotic surfactant, surfactant containing double-labelled dipalmitoyl-phosphatidylcholine (DPPC) was injected into amniotic fluid on days 23-27 of gestation. Within 44 h, DPPC was distributed to the gastrointestinal tract (45.9%), fetal membranes and placenta (8.2%), fetal lung (6.6%), and liver (1.9%). DPPC uptake was higher in the upper than in the lower lung lobes. There was no detectable metabolism of DPPC taken up by fetal lung. Surfactant protein A, originating from intra-amniotic heterologous surfactant, was detected immunohistochemically in alveolar epithelium. Intra-amniotic surfactant (1,500-2,000 mg/kg on day 25.3) improved lung compliance of ventilated 27.0-day premature rabbits. It should be noted, however, that in this study FBMss were absent and were not induced. Solana et al (1998) have shown in rabbits that the intra-amnitoc administration of surfactant labeled with Tc99 is uniformly distributed into the lungs of treated fetuses. Tc99 is also used for the identification and eventual biopsy of the sentinel node in the diagnostic work-out of breast cancer. Its application can avoid the axillary lymphadenectomy in the case that the sentinel node is not found to be involved by metastatic cells. Cosmi et al (1996 a, b, c) were the first to introduce the use of IAS in the management of high risk pregnacies, i.e., fetuses in severe distress and/or imminent delivery, when IAS was the only therapeutic options. The protocol for IAS is the following: after obtaining informed consent, amniocentesis is performed under ultrasound (US) guidance and the needle is placed in proximity of the fetal nares to collect amniotic fluid (AF) for testing fetal lung maturity (FLM); the needle is kept in place. FLM is assessed by a rapid test, the shake test and/or the lamellar bodies count, and later on by L/S determination and phosphatidylglycerol (PG) measurement. If the shake test and/or the lamellar bodies counts indicates fetal lung immaturity, a loading IV dose of 240 mg aminophylline is administered over 10 minutes to the mother followed by IV infusion at the rate of 0.02-0.1 mg/kg/min. The fetus is continuously monitored by Doppler velocimetry; 5 to 15 minutes after the loading dose of aminophylline, FBMss usually begin at a rate of 10 to 12/min as documented by chest wall movements and inspiratory and expiratory flows of AF liquid through the nostrils which are recorded continuously with a color Doppler equipment. Natural Surfactant (80-120 mg in 1 ml,)1 is then instilled through the amniocentesis needle directed toward the fetal mouth. C.S. was performed under epidural anesthesia 60 to 150 minutes after the administration of surfactant. Before the incision of amniotic membranes a sample of AF was collected with a 10 ml syringe. All the neonates followed an uneventful clinical course to the time of discharge from the hospital. These studies showed a successful outcome following prenatal administration of natural surfactant to the human fetus. Several lines of evidence suggest that effective intrapulmonary distribution of SS was achieved: 1) At the beginning nasal fluid waveforms appeared as vortexes, then as regular FBMss induced by the i.v. administration of A to the mother; they were sustained and deep and increased in frequency up to 88/min.; they were synchronous with chest wall FBMss as documented by ultrasound and colour Doppler. Entry of surfactant into and distribution by diffusion throughout all potential airspaces are promoted by the agitation produced by FBMss; 2) Continued FBMss favour rapid dispersion and uniform distribution of the surfactant into the smallest airways and saccules as suggested by our studies in the sheep fetus and also by Adams et al (1963); 3) It is also possible that smooth muscle relaxation induced by A may lower the resistance to the movement of SS through the airways; 4) The uneventful clinical course of most newborn infants in our study, is consistent with the known role of FPL as the first substrate for normal surfactant function at birth (Scarpelli EM, 1978). In conclusion, it is worth to note that IAS is particularly useful in pregnancies where delivery is expected to be imminent and there is not time for corticosteroids to elicit their effect, and/or in cases of severe fetal distress.

Other uses for intra amniotic surfactant (IAS) Recently, several research studies have shown that pulmonary surfactant could be useful also for the treatment of other diseases (table III). Lung surfactant deficiency contributes to the pathophysiology of congenital diaphragmatic hernia (CDH) and it is associated with high neonatal mortality rate. Acceleration of FLM by prenatal administration of corticosteroids has been described in animal models of CDH. However, in utero tracheal ligation (TL) is the best method to accelerate lung growth and to reverse pulmonary hypoplasia associated with CDH. Although this method seems to be promising, its application in humans is limited. In an animal model with CDH the prenatal intra-amniotic administration of exogenous porcine surfactant was capable of preventing pulmonary hypoplasia in fetuses. IAS may become an option to tracheal ligation for the treatment of lung hypoplasia and be used to prevent the functional and structural immaturity of lungs associated with congenital diaphragmatic hernia (Tannuri U, 1998; Tannuri U, 1998). Because now it is possible to make prenatal diagnosis of cystic fibrosis, IAS could be a proper treatment in certain cases.

The use of exogenous surfactant for the treatment of meconium aspiration syndrome (MAS) is another well-defined approach.

MAS is a serious pathology occurring mainly in full-term and post-term pregnancies. About 2% of deliveries can be complicated by MAS with a 40% mortality rate. Several studies have suggested a potential dysfunction and inactivation of the surfactant by meconium. In vivo and in vitro studies have demonstrated the action of meconium on the inhibition of the pulmonary surfactant, probably due to lipids, proteins, and bilirubin present in the meconium. Recently, it has been suggested that the noxious effects of the inhibitory factors of the surfactant present in the meconium could be compensated by the concomitant instillation of large amounts of exogenous surfactant into the airways (Diniz EM, 2000). It is known that surfactant has antibacterial properties both in vitro and in vivo experiments. Such activity seems to be highest with porcine surfactant and the one derived from human AF (Sherman MP, 1994).

Surfactant protein A and D have been found to be expressed in the epithelial cells of the Eustachian tube and therefore may be important in the antibody-independent protection of the middle ear (Paananen R, 1999). These findings pose the basis for possible use of IAS in the prevention of infection of the middle ear in the neonate. In addition they may play a role in the pathogenesis of sudden infant death syndrome (SIDS). Another effect of surfactant is related to its lubricating properties (Hills BA, 1994). The surface-active phospholipids of surfactant are able to form planar structures running parallel to the amniotic membrane: these sheets of surface-active phospholipid could be very cohesive within the planes, thus improving tensile strength of the membrane, thereby playing a possible major role in maintaining the mechanical integrity of the chorioamniotic sac. We speculate that IAS may play an important in the prevention of premature rupture of the fetal membranes (PPROM) and in other diseases of the genital tract (Hills BA, 1995).

(traduzione dell'Autore)

Ermelando V. Cosmi

Direttore del II Istituto di Ginecologia ed

Ostetricia Università degli Studi di Roma “La Sapienza”